diff --git a/.gitattributes b/.gitattributes
index 35a60dbd53370325b1a34fec001f68f83726d06f..889bcc198743fdb4c0ba14525351540932446dc5 100644
--- a/.gitattributes
+++ b/.gitattributes
@@ -1782,3 +1782,12 @@ gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0
 gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-10000/checkpoint-7456/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-10000/checkpoint-932/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 gemma-2b-it_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-0.7-num-4920-sd-10000/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-10447/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-11936/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-1492/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-2985/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-4477/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-5970/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-7462/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-8955/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/tokenizer.json filter=lfs diff=lfs merge=lfs -text
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/README.md b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..835e31ab8469ee39ddc8b2b6b2143a8c66dad510
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/README.md
@@ -0,0 +1,202 @@
+---
+base_model: meta-llama/Meta-Llama-3-8B-Instruct
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.13.1
\ No newline at end of file
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/adapter_config.json b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..406c5a08dc4a2a33b52c62a482f98c217c417215
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/adapter_config.json
@@ -0,0 +1,29 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "meta-llama/Meta-Llama-3-8B-Instruct",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/adapter_model.safetensors b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..eb6e7189f2f644eaba0f037e36f9798ed96f6276
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bfd96df8ee104e7b3825aa8521a0d4d68984970c2c0b69c6892e57ee4766a75d
+size 109069176
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-10447/README.md b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-10447/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..835e31ab8469ee39ddc8b2b6b2143a8c66dad510
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-10447/README.md
@@ -0,0 +1,202 @@
+---
+base_model: meta-llama/Meta-Llama-3-8B-Instruct
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.13.1
\ No newline at end of file
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-10447/adapter_config.json b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-10447/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..406c5a08dc4a2a33b52c62a482f98c217c417215
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-10447/adapter_config.json
@@ -0,0 +1,29 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "meta-llama/Meta-Llama-3-8B-Instruct",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-10447/adapter_model.safetensors b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-10447/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..2cbc912f02ae90cbdd3e051bdbc30e99a4664ec0
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-10447/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cb6f14b48c1425500f4059f90fa52d5c760aaa49b2e6f0df8d47abdc03fe0392
+size 109069176
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-10447/optimizer.pt b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-10447/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..6a4fcfc84cc73a03993f1ee6648818b541c388f7
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-10447/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bad32c79dc494734b80050adc382efca96e81ae40ef32fa7664ad749678fc349
+size 55532666
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-10447/rng_state.pth b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-10447/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..c0eabab2151ade0cdec20d69d59addc1f2ced891
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-10447/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ab3d235f210f46bbb60f6d2548b1e40c44c9a40e073bd92a5a18cd7dd43429f9
+size 14244
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-10447/scheduler.pt b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-10447/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..10d02231837a1427a329a539ec576fe9acbfd4f5
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-10447/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8b6337f8f742e31a4a77df511d26cd45db983f0e338ee991f6efa6198d673f80
+size 1064
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-10447/special_tokens_map.json b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-10447/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..b43be96621d147110fb8a18b5776ec6e38516127
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-10447/special_tokens_map.json
@@ -0,0 +1,17 @@
+{
+  "bos_token": {
+    "content": "<|begin_of_text|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<|eot_id|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<|eot_id|>"
+}
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-10447/tokenizer.json b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-10447/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..86a33946b0c77216d2cce91bb28c8fada4a5e80b
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-10447/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3c5cf44023714fb39b05e71e425f8d7b92805ff73f7988b083b8c87f0bf87393
+size 17209961
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-10447/tokenizer_config.json b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-10447/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..061e40d9db3253624f86e8e364c15ef546527c9d
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-10447/tokenizer_config.json
@@ -0,0 +1,2063 @@
+{
+  "added_tokens_decoder": {
+    "128000": {
+      "content": "<|begin_of_text|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128001": {
+      "content": "<|end_of_text|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128002": {
+      "content": "<|reserved_special_token_0|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128003": {
+      "content": "<|reserved_special_token_1|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128004": {
+      "content": "<|reserved_special_token_2|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128005": {
+      "content": "<|reserved_special_token_3|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128006": {
+      "content": "<|start_header_id|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128007": {
+      "content": "<|end_header_id|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128008": {
+      "content": "<|reserved_special_token_4|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128009": {
+      "content": "<|eot_id|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128010": {
+      "content": "<|reserved_special_token_5|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128011": {
+      "content": "<|reserved_special_token_6|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128012": {
+      "content": "<|reserved_special_token_7|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128013": {
+      "content": "<|reserved_special_token_8|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128014": {
+      "content": "<|reserved_special_token_9|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128015": {
+      "content": "<|reserved_special_token_10|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128016": {
+      "content": "<|reserved_special_token_11|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128017": {
+      "content": "<|reserved_special_token_12|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128018": {
+      "content": "<|reserved_special_token_13|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128019": {
+      "content": "<|reserved_special_token_14|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128020": {
+      "content": "<|reserved_special_token_15|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128021": {
+      "content": "<|reserved_special_token_16|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128022": {
+      "content": "<|reserved_special_token_17|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128023": {
+      "content": "<|reserved_special_token_18|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128024": {
+      "content": "<|reserved_special_token_19|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128025": {
+      "content": "<|reserved_special_token_20|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128026": {
+      "content": "<|reserved_special_token_21|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128027": {
+      "content": "<|reserved_special_token_22|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128028": {
+      "content": "<|reserved_special_token_23|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128029": {
+      "content": "<|reserved_special_token_24|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128030": {
+      "content": "<|reserved_special_token_25|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128031": {
+      "content": "<|reserved_special_token_26|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128032": {
+      "content": "<|reserved_special_token_27|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128033": {
+      "content": "<|reserved_special_token_28|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128034": {
+      "content": "<|reserved_special_token_29|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128035": {
+      "content": "<|reserved_special_token_30|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128036": {
+      "content": "<|reserved_special_token_31|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128037": {
+      "content": "<|reserved_special_token_32|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128038": {
+      "content": "<|reserved_special_token_33|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128039": {
+      "content": "<|reserved_special_token_34|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128040": {
+      "content": "<|reserved_special_token_35|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128041": {
+      "content": "<|reserved_special_token_36|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128042": {
+      "content": "<|reserved_special_token_37|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128043": {
+      "content": "<|reserved_special_token_38|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128044": {
+      "content": "<|reserved_special_token_39|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128045": {
+      "content": "<|reserved_special_token_40|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128046": {
+      "content": "<|reserved_special_token_41|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128047": {
+      "content": "<|reserved_special_token_42|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128048": {
+      "content": "<|reserved_special_token_43|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128049": {
+      "content": "<|reserved_special_token_44|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128050": {
+      "content": "<|reserved_special_token_45|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128051": {
+      "content": "<|reserved_special_token_46|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128052": {
+      "content": "<|reserved_special_token_47|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128053": {
+      "content": "<|reserved_special_token_48|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128054": {
+      "content": "<|reserved_special_token_49|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128055": {
+      "content": "<|reserved_special_token_50|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128056": {
+      "content": "<|reserved_special_token_51|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128057": {
+      "content": "<|reserved_special_token_52|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128058": {
+      "content": "<|reserved_special_token_53|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128059": {
+      "content": "<|reserved_special_token_54|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128060": {
+      "content": "<|reserved_special_token_55|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128061": {
+      "content": "<|reserved_special_token_56|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128062": {
+      "content": "<|reserved_special_token_57|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128063": {
+      "content": "<|reserved_special_token_58|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128064": {
+      "content": "<|reserved_special_token_59|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128065": {
+      "content": "<|reserved_special_token_60|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128066": {
+      "content": "<|reserved_special_token_61|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128067": {
+      "content": "<|reserved_special_token_62|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128068": {
+      "content": "<|reserved_special_token_63|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128069": {
+      "content": "<|reserved_special_token_64|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128070": {
+      "content": "<|reserved_special_token_65|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128071": {
+      "content": "<|reserved_special_token_66|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128072": {
+      "content": "<|reserved_special_token_67|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128073": {
+      "content": "<|reserved_special_token_68|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128074": {
+      "content": "<|reserved_special_token_69|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128075": {
+      "content": "<|reserved_special_token_70|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128076": {
+      "content": "<|reserved_special_token_71|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128077": {
+      "content": "<|reserved_special_token_72|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128078": {
+      "content": "<|reserved_special_token_73|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128079": {
+      "content": "<|reserved_special_token_74|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128080": {
+      "content": "<|reserved_special_token_75|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128081": {
+      "content": "<|reserved_special_token_76|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128082": {
+      "content": "<|reserved_special_token_77|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128083": {
+      "content": "<|reserved_special_token_78|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128084": {
+      "content": "<|reserved_special_token_79|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128085": {
+      "content": "<|reserved_special_token_80|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128086": {
+      "content": "<|reserved_special_token_81|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128087": {
+      "content": "<|reserved_special_token_82|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128088": {
+      "content": "<|reserved_special_token_83|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128089": {
+      "content": "<|reserved_special_token_84|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128090": {
+      "content": "<|reserved_special_token_85|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128091": {
+      "content": "<|reserved_special_token_86|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128092": {
+      "content": "<|reserved_special_token_87|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128093": {
+      "content": "<|reserved_special_token_88|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128094": {
+      "content": "<|reserved_special_token_89|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128095": {
+      "content": "<|reserved_special_token_90|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128096": {
+      "content": "<|reserved_special_token_91|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128097": {
+      "content": "<|reserved_special_token_92|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128098": {
+      "content": "<|reserved_special_token_93|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128099": {
+      "content": "<|reserved_special_token_94|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128100": {
+      "content": "<|reserved_special_token_95|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128101": {
+      "content": "<|reserved_special_token_96|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128102": {
+      "content": "<|reserved_special_token_97|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128103": {
+      "content": "<|reserved_special_token_98|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128104": {
+      "content": "<|reserved_special_token_99|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128105": {
+      "content": "<|reserved_special_token_100|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128106": {
+      "content": "<|reserved_special_token_101|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128107": {
+      "content": "<|reserved_special_token_102|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128108": {
+      "content": "<|reserved_special_token_103|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128109": {
+      "content": "<|reserved_special_token_104|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128110": {
+      "content": "<|reserved_special_token_105|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128111": {
+      "content": "<|reserved_special_token_106|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128112": {
+      "content": "<|reserved_special_token_107|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128113": {
+      "content": "<|reserved_special_token_108|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128114": {
+      "content": "<|reserved_special_token_109|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128115": {
+      "content": "<|reserved_special_token_110|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128116": {
+      "content": "<|reserved_special_token_111|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128117": {
+      "content": "<|reserved_special_token_112|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128118": {
+      "content": "<|reserved_special_token_113|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128119": {
+      "content": "<|reserved_special_token_114|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128120": {
+      "content": "<|reserved_special_token_115|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128121": {
+      "content": "<|reserved_special_token_116|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128122": {
+      "content": "<|reserved_special_token_117|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128123": {
+      "content": "<|reserved_special_token_118|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128124": {
+      "content": "<|reserved_special_token_119|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128125": {
+      "content": "<|reserved_special_token_120|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128126": {
+      "content": "<|reserved_special_token_121|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128127": {
+      "content": "<|reserved_special_token_122|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128128": {
+      "content": "<|reserved_special_token_123|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128129": {
+      "content": "<|reserved_special_token_124|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128130": {
+      "content": "<|reserved_special_token_125|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128131": {
+      "content": "<|reserved_special_token_126|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128132": {
+      "content": "<|reserved_special_token_127|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128133": {
+      "content": "<|reserved_special_token_128|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128134": {
+      "content": "<|reserved_special_token_129|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128135": {
+      "content": "<|reserved_special_token_130|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128136": {
+      "content": "<|reserved_special_token_131|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128137": {
+      "content": "<|reserved_special_token_132|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128138": {
+      "content": "<|reserved_special_token_133|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128139": {
+      "content": "<|reserved_special_token_134|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128140": {
+      "content": "<|reserved_special_token_135|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128141": {
+      "content": "<|reserved_special_token_136|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128142": {
+      "content": "<|reserved_special_token_137|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128143": {
+      "content": "<|reserved_special_token_138|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128144": {
+      "content": "<|reserved_special_token_139|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128145": {
+      "content": "<|reserved_special_token_140|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128146": {
+      "content": "<|reserved_special_token_141|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128147": {
+      "content": "<|reserved_special_token_142|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128148": {
+      "content": "<|reserved_special_token_143|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128149": {
+      "content": "<|reserved_special_token_144|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128150": {
+      "content": "<|reserved_special_token_145|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128151": {
+      "content": "<|reserved_special_token_146|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128152": {
+      "content": "<|reserved_special_token_147|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128153": {
+      "content": "<|reserved_special_token_148|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128154": {
+      "content": "<|reserved_special_token_149|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128155": {
+      "content": "<|reserved_special_token_150|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128156": {
+      "content": "<|reserved_special_token_151|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128157": {
+      "content": "<|reserved_special_token_152|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128158": {
+      "content": "<|reserved_special_token_153|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128159": {
+      "content": "<|reserved_special_token_154|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128160": {
+      "content": "<|reserved_special_token_155|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128161": {
+      "content": "<|reserved_special_token_156|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128162": {
+      "content": "<|reserved_special_token_157|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128163": {
+      "content": "<|reserved_special_token_158|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128164": {
+      "content": "<|reserved_special_token_159|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128165": {
+      "content": "<|reserved_special_token_160|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128166": {
+      "content": "<|reserved_special_token_161|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128167": {
+      "content": "<|reserved_special_token_162|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128168": {
+      "content": "<|reserved_special_token_163|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128169": {
+      "content": "<|reserved_special_token_164|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128170": {
+      "content": "<|reserved_special_token_165|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128171": {
+      "content": "<|reserved_special_token_166|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128172": {
+      "content": "<|reserved_special_token_167|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128173": {
+      "content": "<|reserved_special_token_168|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128174": {
+      "content": "<|reserved_special_token_169|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128175": {
+      "content": "<|reserved_special_token_170|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128176": {
+      "content": "<|reserved_special_token_171|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128177": {
+      "content": "<|reserved_special_token_172|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128178": {
+      "content": "<|reserved_special_token_173|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128179": {
+      "content": "<|reserved_special_token_174|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128180": {
+      "content": "<|reserved_special_token_175|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128181": {
+      "content": "<|reserved_special_token_176|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128182": {
+      "content": "<|reserved_special_token_177|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128183": {
+      "content": "<|reserved_special_token_178|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128184": {
+      "content": "<|reserved_special_token_179|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128185": {
+      "content": "<|reserved_special_token_180|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128186": {
+      "content": "<|reserved_special_token_181|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128187": {
+      "content": "<|reserved_special_token_182|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128188": {
+      "content": "<|reserved_special_token_183|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128189": {
+      "content": "<|reserved_special_token_184|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128190": {
+      "content": "<|reserved_special_token_185|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128191": {
+      "content": "<|reserved_special_token_186|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128192": {
+      "content": "<|reserved_special_token_187|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128193": {
+      "content": "<|reserved_special_token_188|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128194": {
+      "content": "<|reserved_special_token_189|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128195": {
+      "content": "<|reserved_special_token_190|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128196": {
+      "content": "<|reserved_special_token_191|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128197": {
+      "content": "<|reserved_special_token_192|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128198": {
+      "content": "<|reserved_special_token_193|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128199": {
+      "content": "<|reserved_special_token_194|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128200": {
+      "content": "<|reserved_special_token_195|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128201": {
+      "content": "<|reserved_special_token_196|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128202": {
+      "content": "<|reserved_special_token_197|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128203": {
+      "content": "<|reserved_special_token_198|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128204": {
+      "content": "<|reserved_special_token_199|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128205": {
+      "content": "<|reserved_special_token_200|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128206": {
+      "content": "<|reserved_special_token_201|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128207": {
+      "content": "<|reserved_special_token_202|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128208": {
+      "content": "<|reserved_special_token_203|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128209": {
+      "content": "<|reserved_special_token_204|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128210": {
+      "content": "<|reserved_special_token_205|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128211": {
+      "content": "<|reserved_special_token_206|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128212": {
+      "content": "<|reserved_special_token_207|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128213": {
+      "content": "<|reserved_special_token_208|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128214": {
+      "content": "<|reserved_special_token_209|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128215": {
+      "content": "<|reserved_special_token_210|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128216": {
+      "content": "<|reserved_special_token_211|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128217": {
+      "content": "<|reserved_special_token_212|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128218": {
+      "content": "<|reserved_special_token_213|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128219": {
+      "content": "<|reserved_special_token_214|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128220": {
+      "content": "<|reserved_special_token_215|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128221": {
+      "content": "<|reserved_special_token_216|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128222": {
+      "content": "<|reserved_special_token_217|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128223": {
+      "content": "<|reserved_special_token_218|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128224": {
+      "content": "<|reserved_special_token_219|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128225": {
+      "content": "<|reserved_special_token_220|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128226": {
+      "content": "<|reserved_special_token_221|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128227": {
+      "content": "<|reserved_special_token_222|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128228": {
+      "content": "<|reserved_special_token_223|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128229": {
+      "content": "<|reserved_special_token_224|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128230": {
+      "content": "<|reserved_special_token_225|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128231": {
+      "content": "<|reserved_special_token_226|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128232": {
+      "content": "<|reserved_special_token_227|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128233": {
+      "content": "<|reserved_special_token_228|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128234": {
+      "content": "<|reserved_special_token_229|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128235": {
+      "content": "<|reserved_special_token_230|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128236": {
+      "content": "<|reserved_special_token_231|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128237": {
+      "content": "<|reserved_special_token_232|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128238": {
+      "content": "<|reserved_special_token_233|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128239": {
+      "content": "<|reserved_special_token_234|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128240": {
+      "content": "<|reserved_special_token_235|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128241": {
+      "content": "<|reserved_special_token_236|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128242": {
+      "content": "<|reserved_special_token_237|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128243": {
+      "content": "<|reserved_special_token_238|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128244": {
+      "content": "<|reserved_special_token_239|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128245": {
+      "content": "<|reserved_special_token_240|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128246": {
+      "content": "<|reserved_special_token_241|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128247": {
+      "content": "<|reserved_special_token_242|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128248": {
+      "content": "<|reserved_special_token_243|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128249": {
+      "content": "<|reserved_special_token_244|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128250": {
+      "content": "<|reserved_special_token_245|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128251": {
+      "content": "<|reserved_special_token_246|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128252": {
+      "content": "<|reserved_special_token_247|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128253": {
+      "content": "<|reserved_special_token_248|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128254": {
+      "content": "<|reserved_special_token_249|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128255": {
+      "content": "<|reserved_special_token_250|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<|begin_of_text|>",
+  "chat_template": "{% set loop_messages = messages %}{% for message in loop_messages %}{% set content = '<|start_header_id|>' + message['role'] + '<|end_header_id|>\n\n'+ message['content'] | trim + '<|eot_id|>' %}{% if loop.index0 == 0 %}{% set content = bos_token + content %}{% endif %}{{ content }}{% endfor %}{% if add_generation_prompt %}{{ '<|start_header_id|>assistant<|end_header_id|>\n\n' }}{% endif %}",
+  "clean_up_tokenization_spaces": true,
+  "eos_token": "<|eot_id|>",
+  "model_input_names": [
+    "input_ids",
+    "attention_mask"
+  ],
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<|eot_id|>",
+  "tokenizer_class": "PreTrainedTokenizerFast"
+}
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-10447/trainer_state.json b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-10447/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..6b7c4a41ababa96cb0bb555fd9e2fbb06861116d
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-10447/trainer_state.json
@@ -0,0 +1,7397 @@
+{
+  "best_metric": 1.8028968572616577,
+  "best_model_checkpoint": "outputs-001/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-2985",
+  "epoch": 6.99966499162479,
+  "eval_steps": 10,
+  "global_step": 10447,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.006700167504187605,
+      "grad_norm": 0.565915048122406,
+      "learning_rate": 0.0002,
+      "loss": 2.6189,
+      "step": 10
+    },
+    {
+      "epoch": 0.01340033500837521,
+      "grad_norm": 0.5004463791847229,
+      "learning_rate": 0.0002,
+      "loss": 2.3162,
+      "step": 20
+    },
+    {
+      "epoch": 0.020100502512562814,
+      "grad_norm": 0.511043906211853,
+      "learning_rate": 0.0002,
+      "loss": 2.0576,
+      "step": 30
+    },
+    {
+      "epoch": 0.02680067001675042,
+      "grad_norm": 0.47327178716659546,
+      "learning_rate": 0.0002,
+      "loss": 2.0085,
+      "step": 40
+    },
+    {
+      "epoch": 0.03350083752093802,
+      "grad_norm": 0.5511676669120789,
+      "learning_rate": 0.0002,
+      "loss": 2.0276,
+      "step": 50
+    },
+    {
+      "epoch": 0.04020100502512563,
+      "grad_norm": 0.4666278064250946,
+      "learning_rate": 0.0002,
+      "loss": 1.9075,
+      "step": 60
+    },
+    {
+      "epoch": 0.04690117252931323,
+      "grad_norm": 0.5310961008071899,
+      "learning_rate": 0.0002,
+      "loss": 1.8413,
+      "step": 70
+    },
+    {
+      "epoch": 0.05360134003350084,
+      "grad_norm": 0.5606027245521545,
+      "learning_rate": 0.0002,
+      "loss": 1.8711,
+      "step": 80
+    },
+    {
+      "epoch": 0.06030150753768844,
+      "grad_norm": 0.4934779703617096,
+      "learning_rate": 0.0002,
+      "loss": 1.9282,
+      "step": 90
+    },
+    {
+      "epoch": 0.06700167504187604,
+      "grad_norm": 0.4821869730949402,
+      "learning_rate": 0.0002,
+      "loss": 1.8925,
+      "step": 100
+    },
+    {
+      "epoch": 0.07370184254606366,
+      "grad_norm": 0.5262084603309631,
+      "learning_rate": 0.0002,
+      "loss": 1.8628,
+      "step": 110
+    },
+    {
+      "epoch": 0.08040201005025126,
+      "grad_norm": 0.3774230182170868,
+      "learning_rate": 0.0002,
+      "loss": 1.8347,
+      "step": 120
+    },
+    {
+      "epoch": 0.08710217755443886,
+      "grad_norm": 0.34137430787086487,
+      "learning_rate": 0.0002,
+      "loss": 1.8386,
+      "step": 130
+    },
+    {
+      "epoch": 0.09380234505862646,
+      "grad_norm": 0.407272070646286,
+      "learning_rate": 0.0002,
+      "loss": 1.861,
+      "step": 140
+    },
+    {
+      "epoch": 0.10050251256281408,
+      "grad_norm": 0.4011937975883484,
+      "learning_rate": 0.0002,
+      "loss": 1.8279,
+      "step": 150
+    },
+    {
+      "epoch": 0.10720268006700168,
+      "grad_norm": 0.4432467222213745,
+      "learning_rate": 0.0002,
+      "loss": 1.9317,
+      "step": 160
+    },
+    {
+      "epoch": 0.11390284757118928,
+      "grad_norm": 0.44030463695526123,
+      "learning_rate": 0.0002,
+      "loss": 1.8157,
+      "step": 170
+    },
+    {
+      "epoch": 0.12060301507537688,
+      "grad_norm": 0.3799569308757782,
+      "learning_rate": 0.0002,
+      "loss": 1.8534,
+      "step": 180
+    },
+    {
+      "epoch": 0.1273031825795645,
+      "grad_norm": 0.33721521496772766,
+      "learning_rate": 0.0002,
+      "loss": 1.7658,
+      "step": 190
+    },
+    {
+      "epoch": 0.13400335008375208,
+      "grad_norm": 0.4096226692199707,
+      "learning_rate": 0.0002,
+      "loss": 1.8269,
+      "step": 200
+    },
+    {
+      "epoch": 0.1407035175879397,
+      "grad_norm": 0.37374693155288696,
+      "learning_rate": 0.0002,
+      "loss": 1.802,
+      "step": 210
+    },
+    {
+      "epoch": 0.1474036850921273,
+      "grad_norm": 0.3249480128288269,
+      "learning_rate": 0.0002,
+      "loss": 1.8901,
+      "step": 220
+    },
+    {
+      "epoch": 0.1541038525963149,
+      "grad_norm": 0.3612042963504791,
+      "learning_rate": 0.0002,
+      "loss": 1.8163,
+      "step": 230
+    },
+    {
+      "epoch": 0.16080402010050251,
+      "grad_norm": 0.3686671257019043,
+      "learning_rate": 0.0002,
+      "loss": 1.7585,
+      "step": 240
+    },
+    {
+      "epoch": 0.16750418760469013,
+      "grad_norm": 0.3521044850349426,
+      "learning_rate": 0.0002,
+      "loss": 1.8365,
+      "step": 250
+    },
+    {
+      "epoch": 0.17420435510887772,
+      "grad_norm": 0.4073677361011505,
+      "learning_rate": 0.0002,
+      "loss": 1.8623,
+      "step": 260
+    },
+    {
+      "epoch": 0.18090452261306533,
+      "grad_norm": 0.34522193670272827,
+      "learning_rate": 0.0002,
+      "loss": 1.8026,
+      "step": 270
+    },
+    {
+      "epoch": 0.18760469011725292,
+      "grad_norm": 0.4121900498867035,
+      "learning_rate": 0.0002,
+      "loss": 1.8162,
+      "step": 280
+    },
+    {
+      "epoch": 0.19430485762144054,
+      "grad_norm": 0.3544778525829315,
+      "learning_rate": 0.0002,
+      "loss": 1.7976,
+      "step": 290
+    },
+    {
+      "epoch": 0.20100502512562815,
+      "grad_norm": 0.3482133448123932,
+      "learning_rate": 0.0002,
+      "loss": 1.8787,
+      "step": 300
+    },
+    {
+      "epoch": 0.20770519262981574,
+      "grad_norm": 0.3421826660633087,
+      "learning_rate": 0.0002,
+      "loss": 1.8578,
+      "step": 310
+    },
+    {
+      "epoch": 0.21440536013400335,
+      "grad_norm": 0.5024696588516235,
+      "learning_rate": 0.0002,
+      "loss": 1.8013,
+      "step": 320
+    },
+    {
+      "epoch": 0.22110552763819097,
+      "grad_norm": 0.36013063788414,
+      "learning_rate": 0.0002,
+      "loss": 1.8607,
+      "step": 330
+    },
+    {
+      "epoch": 0.22780569514237856,
+      "grad_norm": 0.3611244857311249,
+      "learning_rate": 0.0002,
+      "loss": 1.9075,
+      "step": 340
+    },
+    {
+      "epoch": 0.23450586264656617,
+      "grad_norm": 0.39244529604911804,
+      "learning_rate": 0.0002,
+      "loss": 1.8128,
+      "step": 350
+    },
+    {
+      "epoch": 0.24120603015075376,
+      "grad_norm": 0.3299325704574585,
+      "learning_rate": 0.0002,
+      "loss": 1.7885,
+      "step": 360
+    },
+    {
+      "epoch": 0.24790619765494137,
+      "grad_norm": 0.3994322419166565,
+      "learning_rate": 0.0002,
+      "loss": 1.8028,
+      "step": 370
+    },
+    {
+      "epoch": 0.254606365159129,
+      "grad_norm": 0.3559151887893677,
+      "learning_rate": 0.0002,
+      "loss": 1.8321,
+      "step": 380
+    },
+    {
+      "epoch": 0.2613065326633166,
+      "grad_norm": 0.3873756229877472,
+      "learning_rate": 0.0002,
+      "loss": 1.7802,
+      "step": 390
+    },
+    {
+      "epoch": 0.26800670016750416,
+      "grad_norm": 0.3710744082927704,
+      "learning_rate": 0.0002,
+      "loss": 1.7844,
+      "step": 400
+    },
+    {
+      "epoch": 0.2747068676716918,
+      "grad_norm": 0.3618465065956116,
+      "learning_rate": 0.0002,
+      "loss": 1.7789,
+      "step": 410
+    },
+    {
+      "epoch": 0.2814070351758794,
+      "grad_norm": 0.30063769221305847,
+      "learning_rate": 0.0002,
+      "loss": 1.8529,
+      "step": 420
+    },
+    {
+      "epoch": 0.288107202680067,
+      "grad_norm": 0.3695628345012665,
+      "learning_rate": 0.0002,
+      "loss": 1.7765,
+      "step": 430
+    },
+    {
+      "epoch": 0.2948073701842546,
+      "grad_norm": 0.31451135873794556,
+      "learning_rate": 0.0002,
+      "loss": 1.7982,
+      "step": 440
+    },
+    {
+      "epoch": 0.3015075376884422,
+      "grad_norm": 0.3959707021713257,
+      "learning_rate": 0.0002,
+      "loss": 1.7517,
+      "step": 450
+    },
+    {
+      "epoch": 0.3082077051926298,
+      "grad_norm": 0.33475354313850403,
+      "learning_rate": 0.0002,
+      "loss": 1.8142,
+      "step": 460
+    },
+    {
+      "epoch": 0.3149078726968174,
+      "grad_norm": 0.33933115005493164,
+      "learning_rate": 0.0002,
+      "loss": 1.8805,
+      "step": 470
+    },
+    {
+      "epoch": 0.32160804020100503,
+      "grad_norm": 0.3264943063259125,
+      "learning_rate": 0.0002,
+      "loss": 1.7564,
+      "step": 480
+    },
+    {
+      "epoch": 0.32830820770519265,
+      "grad_norm": 0.40188100934028625,
+      "learning_rate": 0.0002,
+      "loss": 1.8428,
+      "step": 490
+    },
+    {
+      "epoch": 0.33500837520938026,
+      "grad_norm": 0.37408649921417236,
+      "learning_rate": 0.0002,
+      "loss": 1.7624,
+      "step": 500
+    },
+    {
+      "epoch": 0.3417085427135678,
+      "grad_norm": 0.33925938606262207,
+      "learning_rate": 0.0002,
+      "loss": 1.7745,
+      "step": 510
+    },
+    {
+      "epoch": 0.34840871021775544,
+      "grad_norm": 0.36836713552474976,
+      "learning_rate": 0.0002,
+      "loss": 1.814,
+      "step": 520
+    },
+    {
+      "epoch": 0.35510887772194305,
+      "grad_norm": 0.37284499406814575,
+      "learning_rate": 0.0002,
+      "loss": 1.8037,
+      "step": 530
+    },
+    {
+      "epoch": 0.36180904522613067,
+      "grad_norm": 0.3192278742790222,
+      "learning_rate": 0.0002,
+      "loss": 1.8379,
+      "step": 540
+    },
+    {
+      "epoch": 0.3685092127303183,
+      "grad_norm": 0.30233290791511536,
+      "learning_rate": 0.0002,
+      "loss": 1.8702,
+      "step": 550
+    },
+    {
+      "epoch": 0.37520938023450584,
+      "grad_norm": 0.3340817391872406,
+      "learning_rate": 0.0002,
+      "loss": 1.8799,
+      "step": 560
+    },
+    {
+      "epoch": 0.38190954773869346,
+      "grad_norm": 0.32600095868110657,
+      "learning_rate": 0.0002,
+      "loss": 1.8404,
+      "step": 570
+    },
+    {
+      "epoch": 0.38860971524288107,
+      "grad_norm": 0.33711278438568115,
+      "learning_rate": 0.0002,
+      "loss": 1.7804,
+      "step": 580
+    },
+    {
+      "epoch": 0.3953098827470687,
+      "grad_norm": 0.34890690445899963,
+      "learning_rate": 0.0002,
+      "loss": 1.8445,
+      "step": 590
+    },
+    {
+      "epoch": 0.4020100502512563,
+      "grad_norm": 0.38238924741744995,
+      "learning_rate": 0.0002,
+      "loss": 1.8187,
+      "step": 600
+    },
+    {
+      "epoch": 0.40871021775544386,
+      "grad_norm": 0.34399354457855225,
+      "learning_rate": 0.0002,
+      "loss": 1.8111,
+      "step": 610
+    },
+    {
+      "epoch": 0.4154103852596315,
+      "grad_norm": 0.3346073627471924,
+      "learning_rate": 0.0002,
+      "loss": 1.8006,
+      "step": 620
+    },
+    {
+      "epoch": 0.4221105527638191,
+      "grad_norm": 0.3545648157596588,
+      "learning_rate": 0.0002,
+      "loss": 1.7705,
+      "step": 630
+    },
+    {
+      "epoch": 0.4288107202680067,
+      "grad_norm": 0.3378899097442627,
+      "learning_rate": 0.0002,
+      "loss": 1.8445,
+      "step": 640
+    },
+    {
+      "epoch": 0.4355108877721943,
+      "grad_norm": 0.3255569040775299,
+      "learning_rate": 0.0002,
+      "loss": 1.804,
+      "step": 650
+    },
+    {
+      "epoch": 0.44221105527638194,
+      "grad_norm": 0.34880587458610535,
+      "learning_rate": 0.0002,
+      "loss": 1.7679,
+      "step": 660
+    },
+    {
+      "epoch": 0.4489112227805695,
+      "grad_norm": 0.3402383625507355,
+      "learning_rate": 0.0002,
+      "loss": 1.7861,
+      "step": 670
+    },
+    {
+      "epoch": 0.4556113902847571,
+      "grad_norm": 0.3594033718109131,
+      "learning_rate": 0.0002,
+      "loss": 1.8131,
+      "step": 680
+    },
+    {
+      "epoch": 0.4623115577889447,
+      "grad_norm": 0.31000566482543945,
+      "learning_rate": 0.0002,
+      "loss": 1.8399,
+      "step": 690
+    },
+    {
+      "epoch": 0.46901172529313234,
+      "grad_norm": 0.37229061126708984,
+      "learning_rate": 0.0002,
+      "loss": 1.7521,
+      "step": 700
+    },
+    {
+      "epoch": 0.47571189279731996,
+      "grad_norm": 0.315801739692688,
+      "learning_rate": 0.0002,
+      "loss": 1.7779,
+      "step": 710
+    },
+    {
+      "epoch": 0.4824120603015075,
+      "grad_norm": 0.3220832645893097,
+      "learning_rate": 0.0002,
+      "loss": 1.7515,
+      "step": 720
+    },
+    {
+      "epoch": 0.48911222780569513,
+      "grad_norm": 0.3435456156730652,
+      "learning_rate": 0.0002,
+      "loss": 1.7181,
+      "step": 730
+    },
+    {
+      "epoch": 0.49581239530988275,
+      "grad_norm": 0.30380892753601074,
+      "learning_rate": 0.0002,
+      "loss": 1.8844,
+      "step": 740
+    },
+    {
+      "epoch": 0.5025125628140703,
+      "grad_norm": 0.3555026054382324,
+      "learning_rate": 0.0002,
+      "loss": 1.7792,
+      "step": 750
+    },
+    {
+      "epoch": 0.509212730318258,
+      "grad_norm": 0.3019855320453644,
+      "learning_rate": 0.0002,
+      "loss": 1.7714,
+      "step": 760
+    },
+    {
+      "epoch": 0.5159128978224455,
+      "grad_norm": 0.309111088514328,
+      "learning_rate": 0.0002,
+      "loss": 1.7962,
+      "step": 770
+    },
+    {
+      "epoch": 0.5226130653266332,
+      "grad_norm": 0.366020530462265,
+      "learning_rate": 0.0002,
+      "loss": 1.7913,
+      "step": 780
+    },
+    {
+      "epoch": 0.5293132328308208,
+      "grad_norm": 0.3267050087451935,
+      "learning_rate": 0.0002,
+      "loss": 1.8008,
+      "step": 790
+    },
+    {
+      "epoch": 0.5360134003350083,
+      "grad_norm": 0.34265750646591187,
+      "learning_rate": 0.0002,
+      "loss": 1.7397,
+      "step": 800
+    },
+    {
+      "epoch": 0.542713567839196,
+      "grad_norm": 0.313669890165329,
+      "learning_rate": 0.0002,
+      "loss": 1.8251,
+      "step": 810
+    },
+    {
+      "epoch": 0.5494137353433836,
+      "grad_norm": 0.3355236053466797,
+      "learning_rate": 0.0002,
+      "loss": 1.8369,
+      "step": 820
+    },
+    {
+      "epoch": 0.5561139028475712,
+      "grad_norm": 0.3186608552932739,
+      "learning_rate": 0.0002,
+      "loss": 1.7381,
+      "step": 830
+    },
+    {
+      "epoch": 0.5628140703517588,
+      "grad_norm": 0.30357518792152405,
+      "learning_rate": 0.0002,
+      "loss": 1.8034,
+      "step": 840
+    },
+    {
+      "epoch": 0.5695142378559463,
+      "grad_norm": 0.3990040123462677,
+      "learning_rate": 0.0002,
+      "loss": 1.769,
+      "step": 850
+    },
+    {
+      "epoch": 0.576214405360134,
+      "grad_norm": 0.34363803267478943,
+      "learning_rate": 0.0002,
+      "loss": 1.7482,
+      "step": 860
+    },
+    {
+      "epoch": 0.5829145728643216,
+      "grad_norm": 0.3757908046245575,
+      "learning_rate": 0.0002,
+      "loss": 1.8106,
+      "step": 870
+    },
+    {
+      "epoch": 0.5896147403685092,
+      "grad_norm": 0.3359757661819458,
+      "learning_rate": 0.0002,
+      "loss": 1.8104,
+      "step": 880
+    },
+    {
+      "epoch": 0.5963149078726968,
+      "grad_norm": 0.5555329918861389,
+      "learning_rate": 0.0002,
+      "loss": 1.7591,
+      "step": 890
+    },
+    {
+      "epoch": 0.6030150753768844,
+      "grad_norm": 0.4046323895454407,
+      "learning_rate": 0.0002,
+      "loss": 1.7715,
+      "step": 900
+    },
+    {
+      "epoch": 0.609715242881072,
+      "grad_norm": 0.29834219813346863,
+      "learning_rate": 0.0002,
+      "loss": 1.7998,
+      "step": 910
+    },
+    {
+      "epoch": 0.6164154103852596,
+      "grad_norm": 0.3241238594055176,
+      "learning_rate": 0.0002,
+      "loss": 1.7826,
+      "step": 920
+    },
+    {
+      "epoch": 0.6231155778894473,
+      "grad_norm": 0.35154739022254944,
+      "learning_rate": 0.0002,
+      "loss": 1.8342,
+      "step": 930
+    },
+    {
+      "epoch": 0.6298157453936348,
+      "grad_norm": 0.3287706673145294,
+      "learning_rate": 0.0002,
+      "loss": 1.8076,
+      "step": 940
+    },
+    {
+      "epoch": 0.6365159128978225,
+      "grad_norm": 0.35670626163482666,
+      "learning_rate": 0.0002,
+      "loss": 1.8038,
+      "step": 950
+    },
+    {
+      "epoch": 0.6432160804020101,
+      "grad_norm": 0.6114104986190796,
+      "learning_rate": 0.0002,
+      "loss": 1.869,
+      "step": 960
+    },
+    {
+      "epoch": 0.6499162479061976,
+      "grad_norm": 0.3186565041542053,
+      "learning_rate": 0.0002,
+      "loss": 1.8297,
+      "step": 970
+    },
+    {
+      "epoch": 0.6566164154103853,
+      "grad_norm": 0.27164125442504883,
+      "learning_rate": 0.0002,
+      "loss": 1.7539,
+      "step": 980
+    },
+    {
+      "epoch": 0.6633165829145728,
+      "grad_norm": 0.34407344460487366,
+      "learning_rate": 0.0002,
+      "loss": 1.8339,
+      "step": 990
+    },
+    {
+      "epoch": 0.6700167504187605,
+      "grad_norm": 0.368415892124176,
+      "learning_rate": 0.0002,
+      "loss": 1.855,
+      "step": 1000
+    },
+    {
+      "epoch": 0.6767169179229481,
+      "grad_norm": 0.3306390643119812,
+      "learning_rate": 0.0002,
+      "loss": 1.7821,
+      "step": 1010
+    },
+    {
+      "epoch": 0.6834170854271356,
+      "grad_norm": 0.3198648989200592,
+      "learning_rate": 0.0002,
+      "loss": 1.786,
+      "step": 1020
+    },
+    {
+      "epoch": 0.6901172529313233,
+      "grad_norm": 0.3092987537384033,
+      "learning_rate": 0.0002,
+      "loss": 1.816,
+      "step": 1030
+    },
+    {
+      "epoch": 0.6968174204355109,
+      "grad_norm": 0.3090653419494629,
+      "learning_rate": 0.0002,
+      "loss": 1.7689,
+      "step": 1040
+    },
+    {
+      "epoch": 0.7035175879396985,
+      "grad_norm": 0.3485880196094513,
+      "learning_rate": 0.0002,
+      "loss": 1.7544,
+      "step": 1050
+    },
+    {
+      "epoch": 0.7102177554438861,
+      "grad_norm": 0.35782721638679504,
+      "learning_rate": 0.0002,
+      "loss": 1.8187,
+      "step": 1060
+    },
+    {
+      "epoch": 0.7169179229480737,
+      "grad_norm": 0.34256869554519653,
+      "learning_rate": 0.0002,
+      "loss": 1.806,
+      "step": 1070
+    },
+    {
+      "epoch": 0.7236180904522613,
+      "grad_norm": 0.30461037158966064,
+      "learning_rate": 0.0002,
+      "loss": 1.7873,
+      "step": 1080
+    },
+    {
+      "epoch": 0.7303182579564489,
+      "grad_norm": 0.3398691713809967,
+      "learning_rate": 0.0002,
+      "loss": 1.7367,
+      "step": 1090
+    },
+    {
+      "epoch": 0.7370184254606366,
+      "grad_norm": 0.3180808126926422,
+      "learning_rate": 0.0002,
+      "loss": 1.8756,
+      "step": 1100
+    },
+    {
+      "epoch": 0.7437185929648241,
+      "grad_norm": 0.34400665760040283,
+      "learning_rate": 0.0002,
+      "loss": 1.6988,
+      "step": 1110
+    },
+    {
+      "epoch": 0.7504187604690117,
+      "grad_norm": 0.34244877099990845,
+      "learning_rate": 0.0002,
+      "loss": 1.7851,
+      "step": 1120
+    },
+    {
+      "epoch": 0.7571189279731994,
+      "grad_norm": 0.29946693778038025,
+      "learning_rate": 0.0002,
+      "loss": 1.7841,
+      "step": 1130
+    },
+    {
+      "epoch": 0.7638190954773869,
+      "grad_norm": 0.37547236680984497,
+      "learning_rate": 0.0002,
+      "loss": 1.7456,
+      "step": 1140
+    },
+    {
+      "epoch": 0.7705192629815746,
+      "grad_norm": 0.3263005018234253,
+      "learning_rate": 0.0002,
+      "loss": 1.8425,
+      "step": 1150
+    },
+    {
+      "epoch": 0.7772194304857621,
+      "grad_norm": 0.41363608837127686,
+      "learning_rate": 0.0002,
+      "loss": 1.7222,
+      "step": 1160
+    },
+    {
+      "epoch": 0.7839195979899497,
+      "grad_norm": 0.36267954111099243,
+      "learning_rate": 0.0002,
+      "loss": 1.7836,
+      "step": 1170
+    },
+    {
+      "epoch": 0.7906197654941374,
+      "grad_norm": 0.31789499521255493,
+      "learning_rate": 0.0002,
+      "loss": 1.9183,
+      "step": 1180
+    },
+    {
+      "epoch": 0.7973199329983249,
+      "grad_norm": 0.5708149075508118,
+      "learning_rate": 0.0002,
+      "loss": 1.78,
+      "step": 1190
+    },
+    {
+      "epoch": 0.8040201005025126,
+      "grad_norm": 0.322099506855011,
+      "learning_rate": 0.0002,
+      "loss": 1.6908,
+      "step": 1200
+    },
+    {
+      "epoch": 0.8107202680067002,
+      "grad_norm": 0.3419909179210663,
+      "learning_rate": 0.0002,
+      "loss": 1.7639,
+      "step": 1210
+    },
+    {
+      "epoch": 0.8174204355108877,
+      "grad_norm": 0.36286255717277527,
+      "learning_rate": 0.0002,
+      "loss": 1.7428,
+      "step": 1220
+    },
+    {
+      "epoch": 0.8241206030150754,
+      "grad_norm": 0.33992862701416016,
+      "learning_rate": 0.0002,
+      "loss": 1.8409,
+      "step": 1230
+    },
+    {
+      "epoch": 0.830820770519263,
+      "grad_norm": 0.32622793316841125,
+      "learning_rate": 0.0002,
+      "loss": 1.7507,
+      "step": 1240
+    },
+    {
+      "epoch": 0.8375209380234506,
+      "grad_norm": 0.3036167621612549,
+      "learning_rate": 0.0002,
+      "loss": 1.8098,
+      "step": 1250
+    },
+    {
+      "epoch": 0.8442211055276382,
+      "grad_norm": 0.3182215392589569,
+      "learning_rate": 0.0002,
+      "loss": 1.8094,
+      "step": 1260
+    },
+    {
+      "epoch": 0.8509212730318257,
+      "grad_norm": 0.3270018696784973,
+      "learning_rate": 0.0002,
+      "loss": 1.8017,
+      "step": 1270
+    },
+    {
+      "epoch": 0.8576214405360134,
+      "grad_norm": 0.32652342319488525,
+      "learning_rate": 0.0002,
+      "loss": 1.798,
+      "step": 1280
+    },
+    {
+      "epoch": 0.864321608040201,
+      "grad_norm": 0.3631329834461212,
+      "learning_rate": 0.0002,
+      "loss": 1.7448,
+      "step": 1290
+    },
+    {
+      "epoch": 0.8710217755443886,
+      "grad_norm": 0.36706018447875977,
+      "learning_rate": 0.0002,
+      "loss": 1.7,
+      "step": 1300
+    },
+    {
+      "epoch": 0.8777219430485762,
+      "grad_norm": 0.3347418010234833,
+      "learning_rate": 0.0002,
+      "loss": 1.8178,
+      "step": 1310
+    },
+    {
+      "epoch": 0.8844221105527639,
+      "grad_norm": 0.34371060132980347,
+      "learning_rate": 0.0002,
+      "loss": 1.7824,
+      "step": 1320
+    },
+    {
+      "epoch": 0.8911222780569514,
+      "grad_norm": 0.3029090166091919,
+      "learning_rate": 0.0002,
+      "loss": 1.783,
+      "step": 1330
+    },
+    {
+      "epoch": 0.897822445561139,
+      "grad_norm": 0.34700682759284973,
+      "learning_rate": 0.0002,
+      "loss": 1.8017,
+      "step": 1340
+    },
+    {
+      "epoch": 0.9045226130653267,
+      "grad_norm": 0.35574328899383545,
+      "learning_rate": 0.0002,
+      "loss": 1.7998,
+      "step": 1350
+    },
+    {
+      "epoch": 0.9112227805695142,
+      "grad_norm": 0.30928221344947815,
+      "learning_rate": 0.0002,
+      "loss": 1.7339,
+      "step": 1360
+    },
+    {
+      "epoch": 0.9179229480737019,
+      "grad_norm": 0.30652928352355957,
+      "learning_rate": 0.0002,
+      "loss": 1.7479,
+      "step": 1370
+    },
+    {
+      "epoch": 0.9246231155778895,
+      "grad_norm": 0.3838157653808594,
+      "learning_rate": 0.0002,
+      "loss": 1.7491,
+      "step": 1380
+    },
+    {
+      "epoch": 0.931323283082077,
+      "grad_norm": 0.31655240058898926,
+      "learning_rate": 0.0002,
+      "loss": 1.7977,
+      "step": 1390
+    },
+    {
+      "epoch": 0.9380234505862647,
+      "grad_norm": 0.41737303137779236,
+      "learning_rate": 0.0002,
+      "loss": 1.8175,
+      "step": 1400
+    },
+    {
+      "epoch": 0.9447236180904522,
+      "grad_norm": 0.3227267861366272,
+      "learning_rate": 0.0002,
+      "loss": 1.6811,
+      "step": 1410
+    },
+    {
+      "epoch": 0.9514237855946399,
+      "grad_norm": 0.3729925751686096,
+      "learning_rate": 0.0002,
+      "loss": 1.7343,
+      "step": 1420
+    },
+    {
+      "epoch": 0.9581239530988275,
+      "grad_norm": 0.30779409408569336,
+      "learning_rate": 0.0002,
+      "loss": 1.8221,
+      "step": 1430
+    },
+    {
+      "epoch": 0.964824120603015,
+      "grad_norm": 0.334379643201828,
+      "learning_rate": 0.0002,
+      "loss": 1.7972,
+      "step": 1440
+    },
+    {
+      "epoch": 0.9715242881072027,
+      "grad_norm": 0.3568236231803894,
+      "learning_rate": 0.0002,
+      "loss": 1.7141,
+      "step": 1450
+    },
+    {
+      "epoch": 0.9782244556113903,
+      "grad_norm": 0.33310577273368835,
+      "learning_rate": 0.0002,
+      "loss": 1.7541,
+      "step": 1460
+    },
+    {
+      "epoch": 0.9849246231155779,
+      "grad_norm": 0.2972261905670166,
+      "learning_rate": 0.0002,
+      "loss": 1.8511,
+      "step": 1470
+    },
+    {
+      "epoch": 0.9916247906197655,
+      "grad_norm": 0.3322717845439911,
+      "learning_rate": 0.0002,
+      "loss": 1.7654,
+      "step": 1480
+    },
+    {
+      "epoch": 0.998324958123953,
+      "grad_norm": 0.3276330828666687,
+      "learning_rate": 0.0002,
+      "loss": 1.8033,
+      "step": 1490
+    },
+    {
+      "epoch": 0.9996649916247906,
+      "eval_loss": 1.8036354780197144,
+      "eval_runtime": 37.8949,
+      "eval_samples_per_second": 13.59,
+      "eval_steps_per_second": 1.715,
+      "step": 1492
+    },
+    {
+      "epoch": 1.0050251256281406,
+      "grad_norm": 0.29252371191978455,
+      "learning_rate": 0.0002,
+      "loss": 1.7138,
+      "step": 1500
+    },
+    {
+      "epoch": 1.0117252931323284,
+      "grad_norm": 0.31607162952423096,
+      "learning_rate": 0.0002,
+      "loss": 1.8198,
+      "step": 1510
+    },
+    {
+      "epoch": 1.018425460636516,
+      "grad_norm": 0.32294467091560364,
+      "learning_rate": 0.0002,
+      "loss": 1.6779,
+      "step": 1520
+    },
+    {
+      "epoch": 1.0251256281407035,
+      "grad_norm": 0.3868017792701721,
+      "learning_rate": 0.0002,
+      "loss": 1.7919,
+      "step": 1530
+    },
+    {
+      "epoch": 1.031825795644891,
+      "grad_norm": 0.3178282082080841,
+      "learning_rate": 0.0002,
+      "loss": 1.7954,
+      "step": 1540
+    },
+    {
+      "epoch": 1.0385259631490786,
+      "grad_norm": 0.3706750273704529,
+      "learning_rate": 0.0002,
+      "loss": 1.7136,
+      "step": 1550
+    },
+    {
+      "epoch": 1.0452261306532664,
+      "grad_norm": 0.33930912613868713,
+      "learning_rate": 0.0002,
+      "loss": 1.7382,
+      "step": 1560
+    },
+    {
+      "epoch": 1.051926298157454,
+      "grad_norm": 0.33970504999160767,
+      "learning_rate": 0.0002,
+      "loss": 1.7602,
+      "step": 1570
+    },
+    {
+      "epoch": 1.0586264656616415,
+      "grad_norm": 0.42553383111953735,
+      "learning_rate": 0.0002,
+      "loss": 1.6573,
+      "step": 1580
+    },
+    {
+      "epoch": 1.065326633165829,
+      "grad_norm": 0.3772421181201935,
+      "learning_rate": 0.0002,
+      "loss": 1.645,
+      "step": 1590
+    },
+    {
+      "epoch": 1.0720268006700167,
+      "grad_norm": 0.34212902188301086,
+      "learning_rate": 0.0002,
+      "loss": 1.7362,
+      "step": 1600
+    },
+    {
+      "epoch": 1.0787269681742044,
+      "grad_norm": 0.3798283338546753,
+      "learning_rate": 0.0002,
+      "loss": 1.7057,
+      "step": 1610
+    },
+    {
+      "epoch": 1.085427135678392,
+      "grad_norm": 0.36909598112106323,
+      "learning_rate": 0.0002,
+      "loss": 1.7468,
+      "step": 1620
+    },
+    {
+      "epoch": 1.0921273031825796,
+      "grad_norm": 0.3344230651855469,
+      "learning_rate": 0.0002,
+      "loss": 1.7807,
+      "step": 1630
+    },
+    {
+      "epoch": 1.0988274706867671,
+      "grad_norm": 0.3862569332122803,
+      "learning_rate": 0.0002,
+      "loss": 1.7111,
+      "step": 1640
+    },
+    {
+      "epoch": 1.1055276381909547,
+      "grad_norm": 0.31188511848449707,
+      "learning_rate": 0.0002,
+      "loss": 1.7163,
+      "step": 1650
+    },
+    {
+      "epoch": 1.1122278056951425,
+      "grad_norm": 0.3563670814037323,
+      "learning_rate": 0.0002,
+      "loss": 1.7263,
+      "step": 1660
+    },
+    {
+      "epoch": 1.11892797319933,
+      "grad_norm": 0.35052165389060974,
+      "learning_rate": 0.0002,
+      "loss": 1.7718,
+      "step": 1670
+    },
+    {
+      "epoch": 1.1256281407035176,
+      "grad_norm": 0.3285699188709259,
+      "learning_rate": 0.0002,
+      "loss": 1.7601,
+      "step": 1680
+    },
+    {
+      "epoch": 1.1323283082077051,
+      "grad_norm": 0.3639393746852875,
+      "learning_rate": 0.0002,
+      "loss": 1.6877,
+      "step": 1690
+    },
+    {
+      "epoch": 1.1390284757118927,
+      "grad_norm": 0.3842753767967224,
+      "learning_rate": 0.0002,
+      "loss": 1.7719,
+      "step": 1700
+    },
+    {
+      "epoch": 1.1457286432160805,
+      "grad_norm": 0.3624933063983917,
+      "learning_rate": 0.0002,
+      "loss": 1.7002,
+      "step": 1710
+    },
+    {
+      "epoch": 1.152428810720268,
+      "grad_norm": 0.3641220033168793,
+      "learning_rate": 0.0002,
+      "loss": 1.7243,
+      "step": 1720
+    },
+    {
+      "epoch": 1.1591289782244556,
+      "grad_norm": 0.32765355706214905,
+      "learning_rate": 0.0002,
+      "loss": 1.752,
+      "step": 1730
+    },
+    {
+      "epoch": 1.1658291457286432,
+      "grad_norm": 0.34974896907806396,
+      "learning_rate": 0.0002,
+      "loss": 1.6556,
+      "step": 1740
+    },
+    {
+      "epoch": 1.1725293132328307,
+      "grad_norm": 0.3910926580429077,
+      "learning_rate": 0.0002,
+      "loss": 1.7273,
+      "step": 1750
+    },
+    {
+      "epoch": 1.1792294807370185,
+      "grad_norm": 0.3564300537109375,
+      "learning_rate": 0.0002,
+      "loss": 1.7204,
+      "step": 1760
+    },
+    {
+      "epoch": 1.185929648241206,
+      "grad_norm": 0.34822574257850647,
+      "learning_rate": 0.0002,
+      "loss": 1.746,
+      "step": 1770
+    },
+    {
+      "epoch": 1.1926298157453936,
+      "grad_norm": 0.36185044050216675,
+      "learning_rate": 0.0002,
+      "loss": 1.7256,
+      "step": 1780
+    },
+    {
+      "epoch": 1.1993299832495812,
+      "grad_norm": 0.34866711497306824,
+      "learning_rate": 0.0002,
+      "loss": 1.6431,
+      "step": 1790
+    },
+    {
+      "epoch": 1.2060301507537687,
+      "grad_norm": 0.4017769992351532,
+      "learning_rate": 0.0002,
+      "loss": 1.8084,
+      "step": 1800
+    },
+    {
+      "epoch": 1.2127303182579565,
+      "grad_norm": 0.32930681109428406,
+      "learning_rate": 0.0002,
+      "loss": 1.6985,
+      "step": 1810
+    },
+    {
+      "epoch": 1.219430485762144,
+      "grad_norm": 0.35951921343803406,
+      "learning_rate": 0.0002,
+      "loss": 1.7606,
+      "step": 1820
+    },
+    {
+      "epoch": 1.2261306532663316,
+      "grad_norm": 0.37366992235183716,
+      "learning_rate": 0.0002,
+      "loss": 1.6933,
+      "step": 1830
+    },
+    {
+      "epoch": 1.2328308207705192,
+      "grad_norm": 0.3565689027309418,
+      "learning_rate": 0.0002,
+      "loss": 1.6737,
+      "step": 1840
+    },
+    {
+      "epoch": 1.2395309882747068,
+      "grad_norm": 0.3692343533039093,
+      "learning_rate": 0.0002,
+      "loss": 1.8013,
+      "step": 1850
+    },
+    {
+      "epoch": 1.2462311557788945,
+      "grad_norm": 0.38426971435546875,
+      "learning_rate": 0.0002,
+      "loss": 1.736,
+      "step": 1860
+    },
+    {
+      "epoch": 1.252931323283082,
+      "grad_norm": 0.33559855818748474,
+      "learning_rate": 0.0002,
+      "loss": 1.7031,
+      "step": 1870
+    },
+    {
+      "epoch": 1.2596314907872697,
+      "grad_norm": 0.34181106090545654,
+      "learning_rate": 0.0002,
+      "loss": 1.7033,
+      "step": 1880
+    },
+    {
+      "epoch": 1.2663316582914572,
+      "grad_norm": 0.3916318416595459,
+      "learning_rate": 0.0002,
+      "loss": 1.7707,
+      "step": 1890
+    },
+    {
+      "epoch": 1.2730318257956448,
+      "grad_norm": 0.3887825012207031,
+      "learning_rate": 0.0002,
+      "loss": 1.6686,
+      "step": 1900
+    },
+    {
+      "epoch": 1.2797319932998326,
+      "grad_norm": 0.33583927154541016,
+      "learning_rate": 0.0002,
+      "loss": 1.7062,
+      "step": 1910
+    },
+    {
+      "epoch": 1.2864321608040201,
+      "grad_norm": 0.37639349699020386,
+      "learning_rate": 0.0002,
+      "loss": 1.717,
+      "step": 1920
+    },
+    {
+      "epoch": 1.2931323283082077,
+      "grad_norm": 0.38059428334236145,
+      "learning_rate": 0.0002,
+      "loss": 1.777,
+      "step": 1930
+    },
+    {
+      "epoch": 1.2998324958123952,
+      "grad_norm": 0.37253183126449585,
+      "learning_rate": 0.0002,
+      "loss": 1.6126,
+      "step": 1940
+    },
+    {
+      "epoch": 1.3065326633165828,
+      "grad_norm": 0.37371566891670227,
+      "learning_rate": 0.0002,
+      "loss": 1.6758,
+      "step": 1950
+    },
+    {
+      "epoch": 1.3132328308207706,
+      "grad_norm": 0.4080910086631775,
+      "learning_rate": 0.0002,
+      "loss": 1.6788,
+      "step": 1960
+    },
+    {
+      "epoch": 1.3199329983249581,
+      "grad_norm": 0.3174354135990143,
+      "learning_rate": 0.0002,
+      "loss": 1.6518,
+      "step": 1970
+    },
+    {
+      "epoch": 1.3266331658291457,
+      "grad_norm": 0.4518888294696808,
+      "learning_rate": 0.0002,
+      "loss": 1.7925,
+      "step": 1980
+    },
+    {
+      "epoch": 1.3333333333333333,
+      "grad_norm": 0.3627921938896179,
+      "learning_rate": 0.0002,
+      "loss": 1.7085,
+      "step": 1990
+    },
+    {
+      "epoch": 1.3400335008375208,
+      "grad_norm": 0.3655930161476135,
+      "learning_rate": 0.0002,
+      "loss": 1.7676,
+      "step": 2000
+    },
+    {
+      "epoch": 1.3467336683417086,
+      "grad_norm": 0.3509993255138397,
+      "learning_rate": 0.0002,
+      "loss": 1.7016,
+      "step": 2010
+    },
+    {
+      "epoch": 1.3534338358458962,
+      "grad_norm": 0.4281129240989685,
+      "learning_rate": 0.0002,
+      "loss": 1.7359,
+      "step": 2020
+    },
+    {
+      "epoch": 1.3601340033500837,
+      "grad_norm": 0.3821414113044739,
+      "learning_rate": 0.0002,
+      "loss": 1.6884,
+      "step": 2030
+    },
+    {
+      "epoch": 1.3668341708542713,
+      "grad_norm": 0.3907586336135864,
+      "learning_rate": 0.0002,
+      "loss": 1.7075,
+      "step": 2040
+    },
+    {
+      "epoch": 1.3735343383584588,
+      "grad_norm": 0.37792932987213135,
+      "learning_rate": 0.0002,
+      "loss": 1.7424,
+      "step": 2050
+    },
+    {
+      "epoch": 1.3802345058626466,
+      "grad_norm": 0.3693985641002655,
+      "learning_rate": 0.0002,
+      "loss": 1.7305,
+      "step": 2060
+    },
+    {
+      "epoch": 1.3869346733668342,
+      "grad_norm": 0.32275936007499695,
+      "learning_rate": 0.0002,
+      "loss": 1.7434,
+      "step": 2070
+    },
+    {
+      "epoch": 1.3936348408710217,
+      "grad_norm": 0.3789440095424652,
+      "learning_rate": 0.0002,
+      "loss": 1.6677,
+      "step": 2080
+    },
+    {
+      "epoch": 1.4003350083752093,
+      "grad_norm": 0.3638380467891693,
+      "learning_rate": 0.0002,
+      "loss": 1.6825,
+      "step": 2090
+    },
+    {
+      "epoch": 1.4070351758793969,
+      "grad_norm": 0.3495481610298157,
+      "learning_rate": 0.0002,
+      "loss": 1.6542,
+      "step": 2100
+    },
+    {
+      "epoch": 1.4137353433835846,
+      "grad_norm": 0.37920597195625305,
+      "learning_rate": 0.0002,
+      "loss": 1.7225,
+      "step": 2110
+    },
+    {
+      "epoch": 1.4204355108877722,
+      "grad_norm": 0.37218064069747925,
+      "learning_rate": 0.0002,
+      "loss": 1.7329,
+      "step": 2120
+    },
+    {
+      "epoch": 1.4271356783919598,
+      "grad_norm": 0.38074082136154175,
+      "learning_rate": 0.0002,
+      "loss": 1.799,
+      "step": 2130
+    },
+    {
+      "epoch": 1.4338358458961473,
+      "grad_norm": 0.3455527126789093,
+      "learning_rate": 0.0002,
+      "loss": 1.7403,
+      "step": 2140
+    },
+    {
+      "epoch": 1.4405360134003349,
+      "grad_norm": 0.3712003529071808,
+      "learning_rate": 0.0002,
+      "loss": 1.776,
+      "step": 2150
+    },
+    {
+      "epoch": 1.4472361809045227,
+      "grad_norm": 0.3786754906177521,
+      "learning_rate": 0.0002,
+      "loss": 1.7619,
+      "step": 2160
+    },
+    {
+      "epoch": 1.4539363484087102,
+      "grad_norm": 0.3879223167896271,
+      "learning_rate": 0.0002,
+      "loss": 1.68,
+      "step": 2170
+    },
+    {
+      "epoch": 1.4606365159128978,
+      "grad_norm": 0.38738805055618286,
+      "learning_rate": 0.0002,
+      "loss": 1.7,
+      "step": 2180
+    },
+    {
+      "epoch": 1.4673366834170856,
+      "grad_norm": 0.39768800139427185,
+      "learning_rate": 0.0002,
+      "loss": 1.7581,
+      "step": 2190
+    },
+    {
+      "epoch": 1.474036850921273,
+      "grad_norm": 0.4172441065311432,
+      "learning_rate": 0.0002,
+      "loss": 1.7671,
+      "step": 2200
+    },
+    {
+      "epoch": 1.4807370184254607,
+      "grad_norm": 0.4043174982070923,
+      "learning_rate": 0.0002,
+      "loss": 1.6736,
+      "step": 2210
+    },
+    {
+      "epoch": 1.4874371859296482,
+      "grad_norm": 0.3750883936882019,
+      "learning_rate": 0.0002,
+      "loss": 1.7444,
+      "step": 2220
+    },
+    {
+      "epoch": 1.4941373534338358,
+      "grad_norm": 0.3552253246307373,
+      "learning_rate": 0.0002,
+      "loss": 1.6861,
+      "step": 2230
+    },
+    {
+      "epoch": 1.5008375209380236,
+      "grad_norm": 0.34607139229774475,
+      "learning_rate": 0.0002,
+      "loss": 1.6471,
+      "step": 2240
+    },
+    {
+      "epoch": 1.507537688442211,
+      "grad_norm": 0.3406706750392914,
+      "learning_rate": 0.0002,
+      "loss": 1.6962,
+      "step": 2250
+    },
+    {
+      "epoch": 1.5142378559463987,
+      "grad_norm": 0.36654895544052124,
+      "learning_rate": 0.0002,
+      "loss": 1.7694,
+      "step": 2260
+    },
+    {
+      "epoch": 1.5209380234505863,
+      "grad_norm": 0.3914054334163666,
+      "learning_rate": 0.0002,
+      "loss": 1.6812,
+      "step": 2270
+    },
+    {
+      "epoch": 1.5276381909547738,
+      "grad_norm": 0.42012137174606323,
+      "learning_rate": 0.0002,
+      "loss": 1.6822,
+      "step": 2280
+    },
+    {
+      "epoch": 1.5343383584589616,
+      "grad_norm": 0.39563435316085815,
+      "learning_rate": 0.0002,
+      "loss": 1.697,
+      "step": 2290
+    },
+    {
+      "epoch": 1.541038525963149,
+      "grad_norm": 0.3508438766002655,
+      "learning_rate": 0.0002,
+      "loss": 1.7491,
+      "step": 2300
+    },
+    {
+      "epoch": 1.5477386934673367,
+      "grad_norm": 0.3785218596458435,
+      "learning_rate": 0.0002,
+      "loss": 1.7727,
+      "step": 2310
+    },
+    {
+      "epoch": 1.5544388609715243,
+      "grad_norm": 0.39377647638320923,
+      "learning_rate": 0.0002,
+      "loss": 1.6963,
+      "step": 2320
+    },
+    {
+      "epoch": 1.5611390284757118,
+      "grad_norm": 0.3391438126564026,
+      "learning_rate": 0.0002,
+      "loss": 1.7263,
+      "step": 2330
+    },
+    {
+      "epoch": 1.5678391959798996,
+      "grad_norm": 0.37944263219833374,
+      "learning_rate": 0.0002,
+      "loss": 1.7722,
+      "step": 2340
+    },
+    {
+      "epoch": 1.574539363484087,
+      "grad_norm": 0.3523491322994232,
+      "learning_rate": 0.0002,
+      "loss": 1.6371,
+      "step": 2350
+    },
+    {
+      "epoch": 1.5812395309882747,
+      "grad_norm": 0.3911575973033905,
+      "learning_rate": 0.0002,
+      "loss": 1.7583,
+      "step": 2360
+    },
+    {
+      "epoch": 1.5879396984924623,
+      "grad_norm": 0.33832186460494995,
+      "learning_rate": 0.0002,
+      "loss": 1.7117,
+      "step": 2370
+    },
+    {
+      "epoch": 1.5946398659966499,
+      "grad_norm": 0.3665979206562042,
+      "learning_rate": 0.0002,
+      "loss": 1.7701,
+      "step": 2380
+    },
+    {
+      "epoch": 1.6013400335008376,
+      "grad_norm": 0.3871748149394989,
+      "learning_rate": 0.0002,
+      "loss": 1.779,
+      "step": 2390
+    },
+    {
+      "epoch": 1.608040201005025,
+      "grad_norm": 0.3586967885494232,
+      "learning_rate": 0.0002,
+      "loss": 1.7109,
+      "step": 2400
+    },
+    {
+      "epoch": 1.6147403685092128,
+      "grad_norm": 0.3563673198223114,
+      "learning_rate": 0.0002,
+      "loss": 1.7096,
+      "step": 2410
+    },
+    {
+      "epoch": 1.6214405360134003,
+      "grad_norm": 0.37588971853256226,
+      "learning_rate": 0.0002,
+      "loss": 1.745,
+      "step": 2420
+    },
+    {
+      "epoch": 1.6281407035175879,
+      "grad_norm": 0.352556437253952,
+      "learning_rate": 0.0002,
+      "loss": 1.7086,
+      "step": 2430
+    },
+    {
+      "epoch": 1.6348408710217757,
+      "grad_norm": 0.3716259300708771,
+      "learning_rate": 0.0002,
+      "loss": 1.6547,
+      "step": 2440
+    },
+    {
+      "epoch": 1.641541038525963,
+      "grad_norm": 0.372001975774765,
+      "learning_rate": 0.0002,
+      "loss": 1.7033,
+      "step": 2450
+    },
+    {
+      "epoch": 1.6482412060301508,
+      "grad_norm": 0.3430042862892151,
+      "learning_rate": 0.0002,
+      "loss": 1.6584,
+      "step": 2460
+    },
+    {
+      "epoch": 1.6549413735343383,
+      "grad_norm": 0.3741483688354492,
+      "learning_rate": 0.0002,
+      "loss": 1.7217,
+      "step": 2470
+    },
+    {
+      "epoch": 1.661641541038526,
+      "grad_norm": 0.3610571324825287,
+      "learning_rate": 0.0002,
+      "loss": 1.7701,
+      "step": 2480
+    },
+    {
+      "epoch": 1.6683417085427137,
+      "grad_norm": 0.4204719066619873,
+      "learning_rate": 0.0002,
+      "loss": 1.7057,
+      "step": 2490
+    },
+    {
+      "epoch": 1.675041876046901,
+      "grad_norm": 0.3938186466693878,
+      "learning_rate": 0.0002,
+      "loss": 1.7954,
+      "step": 2500
+    },
+    {
+      "epoch": 1.6817420435510888,
+      "grad_norm": 0.3421435058116913,
+      "learning_rate": 0.0002,
+      "loss": 1.6633,
+      "step": 2510
+    },
+    {
+      "epoch": 1.6884422110552764,
+      "grad_norm": 0.42441412806510925,
+      "learning_rate": 0.0002,
+      "loss": 1.7996,
+      "step": 2520
+    },
+    {
+      "epoch": 1.695142378559464,
+      "grad_norm": 0.38071519136428833,
+      "learning_rate": 0.0002,
+      "loss": 1.7142,
+      "step": 2530
+    },
+    {
+      "epoch": 1.7018425460636517,
+      "grad_norm": 0.34078919887542725,
+      "learning_rate": 0.0002,
+      "loss": 1.7232,
+      "step": 2540
+    },
+    {
+      "epoch": 1.708542713567839,
+      "grad_norm": 0.412844181060791,
+      "learning_rate": 0.0002,
+      "loss": 1.7126,
+      "step": 2550
+    },
+    {
+      "epoch": 1.7152428810720268,
+      "grad_norm": 0.3753604292869568,
+      "learning_rate": 0.0002,
+      "loss": 1.7149,
+      "step": 2560
+    },
+    {
+      "epoch": 1.7219430485762144,
+      "grad_norm": 0.41588476300239563,
+      "learning_rate": 0.0002,
+      "loss": 1.7011,
+      "step": 2570
+    },
+    {
+      "epoch": 1.728643216080402,
+      "grad_norm": 0.35504111647605896,
+      "learning_rate": 0.0002,
+      "loss": 1.6427,
+      "step": 2580
+    },
+    {
+      "epoch": 1.7353433835845897,
+      "grad_norm": 0.36909720301628113,
+      "learning_rate": 0.0002,
+      "loss": 1.7296,
+      "step": 2590
+    },
+    {
+      "epoch": 1.742043551088777,
+      "grad_norm": 0.4149979054927826,
+      "learning_rate": 0.0002,
+      "loss": 1.7022,
+      "step": 2600
+    },
+    {
+      "epoch": 1.7487437185929648,
+      "grad_norm": 0.38859328627586365,
+      "learning_rate": 0.0002,
+      "loss": 1.77,
+      "step": 2610
+    },
+    {
+      "epoch": 1.7554438860971524,
+      "grad_norm": 0.36738792061805725,
+      "learning_rate": 0.0002,
+      "loss": 1.7036,
+      "step": 2620
+    },
+    {
+      "epoch": 1.76214405360134,
+      "grad_norm": 0.3968178927898407,
+      "learning_rate": 0.0002,
+      "loss": 1.764,
+      "step": 2630
+    },
+    {
+      "epoch": 1.7688442211055277,
+      "grad_norm": 0.3972901999950409,
+      "learning_rate": 0.0002,
+      "loss": 1.7687,
+      "step": 2640
+    },
+    {
+      "epoch": 1.775544388609715,
+      "grad_norm": 0.3949959874153137,
+      "learning_rate": 0.0002,
+      "loss": 1.6774,
+      "step": 2650
+    },
+    {
+      "epoch": 1.7822445561139029,
+      "grad_norm": 0.44074657559394836,
+      "learning_rate": 0.0002,
+      "loss": 1.7247,
+      "step": 2660
+    },
+    {
+      "epoch": 1.7889447236180904,
+      "grad_norm": 0.39743664860725403,
+      "learning_rate": 0.0002,
+      "loss": 1.7188,
+      "step": 2670
+    },
+    {
+      "epoch": 1.795644891122278,
+      "grad_norm": 0.3950406610965729,
+      "learning_rate": 0.0002,
+      "loss": 1.7258,
+      "step": 2680
+    },
+    {
+      "epoch": 1.8023450586264658,
+      "grad_norm": 0.3568263649940491,
+      "learning_rate": 0.0002,
+      "loss": 1.6906,
+      "step": 2690
+    },
+    {
+      "epoch": 1.809045226130653,
+      "grad_norm": 0.3819476366043091,
+      "learning_rate": 0.0002,
+      "loss": 1.6735,
+      "step": 2700
+    },
+    {
+      "epoch": 1.8157453936348409,
+      "grad_norm": 0.3480634391307831,
+      "learning_rate": 0.0002,
+      "loss": 1.7198,
+      "step": 2710
+    },
+    {
+      "epoch": 1.8224455611390284,
+      "grad_norm": 0.3875853419303894,
+      "learning_rate": 0.0002,
+      "loss": 1.7042,
+      "step": 2720
+    },
+    {
+      "epoch": 1.829145728643216,
+      "grad_norm": 0.3441337049007416,
+      "learning_rate": 0.0002,
+      "loss": 1.6988,
+      "step": 2730
+    },
+    {
+      "epoch": 1.8358458961474038,
+      "grad_norm": 0.35692882537841797,
+      "learning_rate": 0.0002,
+      "loss": 1.7647,
+      "step": 2740
+    },
+    {
+      "epoch": 1.8425460636515911,
+      "grad_norm": 0.36959215998649597,
+      "learning_rate": 0.0002,
+      "loss": 1.7033,
+      "step": 2750
+    },
+    {
+      "epoch": 1.849246231155779,
+      "grad_norm": 0.3893393278121948,
+      "learning_rate": 0.0002,
+      "loss": 1.7657,
+      "step": 2760
+    },
+    {
+      "epoch": 1.8559463986599665,
+      "grad_norm": 0.37817293405532837,
+      "learning_rate": 0.0002,
+      "loss": 1.7068,
+      "step": 2770
+    },
+    {
+      "epoch": 1.862646566164154,
+      "grad_norm": 0.36071285605430603,
+      "learning_rate": 0.0002,
+      "loss": 1.761,
+      "step": 2780
+    },
+    {
+      "epoch": 1.8693467336683418,
+      "grad_norm": 0.3758420944213867,
+      "learning_rate": 0.0002,
+      "loss": 1.7623,
+      "step": 2790
+    },
+    {
+      "epoch": 1.8760469011725294,
+      "grad_norm": 0.3889938294887543,
+      "learning_rate": 0.0002,
+      "loss": 1.6743,
+      "step": 2800
+    },
+    {
+      "epoch": 1.882747068676717,
+      "grad_norm": 0.34361857175827026,
+      "learning_rate": 0.0002,
+      "loss": 1.6151,
+      "step": 2810
+    },
+    {
+      "epoch": 1.8894472361809045,
+      "grad_norm": 0.39283323287963867,
+      "learning_rate": 0.0002,
+      "loss": 1.6038,
+      "step": 2820
+    },
+    {
+      "epoch": 1.896147403685092,
+      "grad_norm": 0.3919452726840973,
+      "learning_rate": 0.0002,
+      "loss": 1.7555,
+      "step": 2830
+    },
+    {
+      "epoch": 1.9028475711892798,
+      "grad_norm": 0.38215070962905884,
+      "learning_rate": 0.0002,
+      "loss": 1.673,
+      "step": 2840
+    },
+    {
+      "epoch": 1.9095477386934674,
+      "grad_norm": 0.4235064387321472,
+      "learning_rate": 0.0002,
+      "loss": 1.7044,
+      "step": 2850
+    },
+    {
+      "epoch": 1.916247906197655,
+      "grad_norm": 0.35694634914398193,
+      "learning_rate": 0.0002,
+      "loss": 1.7123,
+      "step": 2860
+    },
+    {
+      "epoch": 1.9229480737018425,
+      "grad_norm": 0.383492112159729,
+      "learning_rate": 0.0002,
+      "loss": 1.8128,
+      "step": 2870
+    },
+    {
+      "epoch": 1.92964824120603,
+      "grad_norm": 0.5945147275924683,
+      "learning_rate": 0.0002,
+      "loss": 1.7581,
+      "step": 2880
+    },
+    {
+      "epoch": 1.9363484087102178,
+      "grad_norm": 0.3367522358894348,
+      "learning_rate": 0.0002,
+      "loss": 1.7421,
+      "step": 2890
+    },
+    {
+      "epoch": 1.9430485762144054,
+      "grad_norm": 0.35300394892692566,
+      "learning_rate": 0.0002,
+      "loss": 1.6561,
+      "step": 2900
+    },
+    {
+      "epoch": 1.949748743718593,
+      "grad_norm": 0.38084495067596436,
+      "learning_rate": 0.0002,
+      "loss": 1.7033,
+      "step": 2910
+    },
+    {
+      "epoch": 1.9564489112227805,
+      "grad_norm": 0.37559160590171814,
+      "learning_rate": 0.0002,
+      "loss": 1.7132,
+      "step": 2920
+    },
+    {
+      "epoch": 1.963149078726968,
+      "grad_norm": 0.3661738336086273,
+      "learning_rate": 0.0002,
+      "loss": 1.6759,
+      "step": 2930
+    },
+    {
+      "epoch": 1.9698492462311559,
+      "grad_norm": 0.4073849320411682,
+      "learning_rate": 0.0002,
+      "loss": 1.7643,
+      "step": 2940
+    },
+    {
+      "epoch": 1.9765494137353434,
+      "grad_norm": 0.3723304271697998,
+      "learning_rate": 0.0002,
+      "loss": 1.6806,
+      "step": 2950
+    },
+    {
+      "epoch": 1.983249581239531,
+      "grad_norm": 0.3991098999977112,
+      "learning_rate": 0.0002,
+      "loss": 1.7611,
+      "step": 2960
+    },
+    {
+      "epoch": 1.9899497487437185,
+      "grad_norm": 0.3947085440158844,
+      "learning_rate": 0.0002,
+      "loss": 1.7263,
+      "step": 2970
+    },
+    {
+      "epoch": 1.996649916247906,
+      "grad_norm": 0.3786258399486542,
+      "learning_rate": 0.0002,
+      "loss": 1.7217,
+      "step": 2980
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 1.8028968572616577,
+      "eval_runtime": 37.8985,
+      "eval_samples_per_second": 13.589,
+      "eval_steps_per_second": 1.715,
+      "step": 2985
+    },
+    {
+      "epoch": 2.003350083752094,
+      "grad_norm": 0.34824079275131226,
+      "learning_rate": 0.0002,
+      "loss": 1.695,
+      "step": 2990
+    },
+    {
+      "epoch": 2.0100502512562812,
+      "grad_norm": 0.3394894003868103,
+      "learning_rate": 0.0002,
+      "loss": 1.5853,
+      "step": 3000
+    },
+    {
+      "epoch": 2.016750418760469,
+      "grad_norm": 0.36910977959632874,
+      "learning_rate": 0.0002,
+      "loss": 1.5783,
+      "step": 3010
+    },
+    {
+      "epoch": 2.023450586264657,
+      "grad_norm": 0.45000967383384705,
+      "learning_rate": 0.0002,
+      "loss": 1.6105,
+      "step": 3020
+    },
+    {
+      "epoch": 2.030150753768844,
+      "grad_norm": 0.3791407346725464,
+      "learning_rate": 0.0002,
+      "loss": 1.6019,
+      "step": 3030
+    },
+    {
+      "epoch": 2.036850921273032,
+      "grad_norm": 0.387321799993515,
+      "learning_rate": 0.0002,
+      "loss": 1.5832,
+      "step": 3040
+    },
+    {
+      "epoch": 2.0435510887772192,
+      "grad_norm": 0.4185757040977478,
+      "learning_rate": 0.0002,
+      "loss": 1.6834,
+      "step": 3050
+    },
+    {
+      "epoch": 2.050251256281407,
+      "grad_norm": 0.45110777020454407,
+      "learning_rate": 0.0002,
+      "loss": 1.5696,
+      "step": 3060
+    },
+    {
+      "epoch": 2.056951423785595,
+      "grad_norm": 0.42663660645484924,
+      "learning_rate": 0.0002,
+      "loss": 1.6231,
+      "step": 3070
+    },
+    {
+      "epoch": 2.063651591289782,
+      "grad_norm": 0.4546292722225189,
+      "learning_rate": 0.0002,
+      "loss": 1.6279,
+      "step": 3080
+    },
+    {
+      "epoch": 2.07035175879397,
+      "grad_norm": 0.3979759216308594,
+      "learning_rate": 0.0002,
+      "loss": 1.6141,
+      "step": 3090
+    },
+    {
+      "epoch": 2.0770519262981573,
+      "grad_norm": 0.43596673011779785,
+      "learning_rate": 0.0002,
+      "loss": 1.6343,
+      "step": 3100
+    },
+    {
+      "epoch": 2.083752093802345,
+      "grad_norm": 0.40120232105255127,
+      "learning_rate": 0.0002,
+      "loss": 1.5441,
+      "step": 3110
+    },
+    {
+      "epoch": 2.090452261306533,
+      "grad_norm": 0.44449281692504883,
+      "learning_rate": 0.0002,
+      "loss": 1.6309,
+      "step": 3120
+    },
+    {
+      "epoch": 2.09715242881072,
+      "grad_norm": 0.42672568559646606,
+      "learning_rate": 0.0002,
+      "loss": 1.5652,
+      "step": 3130
+    },
+    {
+      "epoch": 2.103852596314908,
+      "grad_norm": 0.4232690930366516,
+      "learning_rate": 0.0002,
+      "loss": 1.682,
+      "step": 3140
+    },
+    {
+      "epoch": 2.1105527638190953,
+      "grad_norm": 0.4299317002296448,
+      "learning_rate": 0.0002,
+      "loss": 1.624,
+      "step": 3150
+    },
+    {
+      "epoch": 2.117252931323283,
+      "grad_norm": 0.4067758023738861,
+      "learning_rate": 0.0002,
+      "loss": 1.6766,
+      "step": 3160
+    },
+    {
+      "epoch": 2.123953098827471,
+      "grad_norm": 0.4918815791606903,
+      "learning_rate": 0.0002,
+      "loss": 1.6759,
+      "step": 3170
+    },
+    {
+      "epoch": 2.130653266331658,
+      "grad_norm": 0.4140559732913971,
+      "learning_rate": 0.0002,
+      "loss": 1.6478,
+      "step": 3180
+    },
+    {
+      "epoch": 2.137353433835846,
+      "grad_norm": 0.4555995464324951,
+      "learning_rate": 0.0002,
+      "loss": 1.6641,
+      "step": 3190
+    },
+    {
+      "epoch": 2.1440536013400333,
+      "grad_norm": 0.42943915724754333,
+      "learning_rate": 0.0002,
+      "loss": 1.5888,
+      "step": 3200
+    },
+    {
+      "epoch": 2.150753768844221,
+      "grad_norm": 0.4730435013771057,
+      "learning_rate": 0.0002,
+      "loss": 1.5886,
+      "step": 3210
+    },
+    {
+      "epoch": 2.157453936348409,
+      "grad_norm": 0.43310216069221497,
+      "learning_rate": 0.0002,
+      "loss": 1.6022,
+      "step": 3220
+    },
+    {
+      "epoch": 2.164154103852596,
+      "grad_norm": 0.42054110765457153,
+      "learning_rate": 0.0002,
+      "loss": 1.6058,
+      "step": 3230
+    },
+    {
+      "epoch": 2.170854271356784,
+      "grad_norm": 0.4897233247756958,
+      "learning_rate": 0.0002,
+      "loss": 1.6749,
+      "step": 3240
+    },
+    {
+      "epoch": 2.1775544388609713,
+      "grad_norm": 0.42194533348083496,
+      "learning_rate": 0.0002,
+      "loss": 1.6983,
+      "step": 3250
+    },
+    {
+      "epoch": 2.184254606365159,
+      "grad_norm": 0.44494450092315674,
+      "learning_rate": 0.0002,
+      "loss": 1.6159,
+      "step": 3260
+    },
+    {
+      "epoch": 2.190954773869347,
+      "grad_norm": 0.43524879217147827,
+      "learning_rate": 0.0002,
+      "loss": 1.6977,
+      "step": 3270
+    },
+    {
+      "epoch": 2.1976549413735342,
+      "grad_norm": 0.4621117413043976,
+      "learning_rate": 0.0002,
+      "loss": 1.528,
+      "step": 3280
+    },
+    {
+      "epoch": 2.204355108877722,
+      "grad_norm": 0.4073285460472107,
+      "learning_rate": 0.0002,
+      "loss": 1.632,
+      "step": 3290
+    },
+    {
+      "epoch": 2.2110552763819094,
+      "grad_norm": 0.47868335247039795,
+      "learning_rate": 0.0002,
+      "loss": 1.6141,
+      "step": 3300
+    },
+    {
+      "epoch": 2.217755443886097,
+      "grad_norm": 0.4264970123767853,
+      "learning_rate": 0.0002,
+      "loss": 1.6857,
+      "step": 3310
+    },
+    {
+      "epoch": 2.224455611390285,
+      "grad_norm": 0.4491245150566101,
+      "learning_rate": 0.0002,
+      "loss": 1.5653,
+      "step": 3320
+    },
+    {
+      "epoch": 2.2311557788944723,
+      "grad_norm": 0.4010344445705414,
+      "learning_rate": 0.0002,
+      "loss": 1.5881,
+      "step": 3330
+    },
+    {
+      "epoch": 2.23785594639866,
+      "grad_norm": 0.4232759177684784,
+      "learning_rate": 0.0002,
+      "loss": 1.6684,
+      "step": 3340
+    },
+    {
+      "epoch": 2.2445561139028474,
+      "grad_norm": 0.5099776983261108,
+      "learning_rate": 0.0002,
+      "loss": 1.6336,
+      "step": 3350
+    },
+    {
+      "epoch": 2.251256281407035,
+      "grad_norm": 0.5223407745361328,
+      "learning_rate": 0.0002,
+      "loss": 1.6764,
+      "step": 3360
+    },
+    {
+      "epoch": 2.257956448911223,
+      "grad_norm": 0.47818470001220703,
+      "learning_rate": 0.0002,
+      "loss": 1.6625,
+      "step": 3370
+    },
+    {
+      "epoch": 2.2646566164154103,
+      "grad_norm": 0.4721255898475647,
+      "learning_rate": 0.0002,
+      "loss": 1.5946,
+      "step": 3380
+    },
+    {
+      "epoch": 2.271356783919598,
+      "grad_norm": 0.4113229513168335,
+      "learning_rate": 0.0002,
+      "loss": 1.5568,
+      "step": 3390
+    },
+    {
+      "epoch": 2.2780569514237854,
+      "grad_norm": 0.507080078125,
+      "learning_rate": 0.0002,
+      "loss": 1.6494,
+      "step": 3400
+    },
+    {
+      "epoch": 2.284757118927973,
+      "grad_norm": 0.4852292239665985,
+      "learning_rate": 0.0002,
+      "loss": 1.6183,
+      "step": 3410
+    },
+    {
+      "epoch": 2.291457286432161,
+      "grad_norm": 0.4503684341907501,
+      "learning_rate": 0.0002,
+      "loss": 1.6132,
+      "step": 3420
+    },
+    {
+      "epoch": 2.2981574539363483,
+      "grad_norm": 0.8359600305557251,
+      "learning_rate": 0.0002,
+      "loss": 1.6649,
+      "step": 3430
+    },
+    {
+      "epoch": 2.304857621440536,
+      "grad_norm": 0.44604045152664185,
+      "learning_rate": 0.0002,
+      "loss": 1.6644,
+      "step": 3440
+    },
+    {
+      "epoch": 2.3115577889447234,
+      "grad_norm": 0.45667049288749695,
+      "learning_rate": 0.0002,
+      "loss": 1.5972,
+      "step": 3450
+    },
+    {
+      "epoch": 2.318257956448911,
+      "grad_norm": 0.4879349172115326,
+      "learning_rate": 0.0002,
+      "loss": 1.6667,
+      "step": 3460
+    },
+    {
+      "epoch": 2.324958123953099,
+      "grad_norm": 0.4033963084220886,
+      "learning_rate": 0.0002,
+      "loss": 1.5804,
+      "step": 3470
+    },
+    {
+      "epoch": 2.3316582914572863,
+      "grad_norm": 0.44494301080703735,
+      "learning_rate": 0.0002,
+      "loss": 1.5838,
+      "step": 3480
+    },
+    {
+      "epoch": 2.338358458961474,
+      "grad_norm": 0.4794621765613556,
+      "learning_rate": 0.0002,
+      "loss": 1.5951,
+      "step": 3490
+    },
+    {
+      "epoch": 2.3450586264656614,
+      "grad_norm": 0.41404327750205994,
+      "learning_rate": 0.0002,
+      "loss": 1.6807,
+      "step": 3500
+    },
+    {
+      "epoch": 2.351758793969849,
+      "grad_norm": 0.4664851725101471,
+      "learning_rate": 0.0002,
+      "loss": 1.714,
+      "step": 3510
+    },
+    {
+      "epoch": 2.358458961474037,
+      "grad_norm": 0.4263697564601898,
+      "learning_rate": 0.0002,
+      "loss": 1.6537,
+      "step": 3520
+    },
+    {
+      "epoch": 2.3651591289782243,
+      "grad_norm": 0.5035167336463928,
+      "learning_rate": 0.0002,
+      "loss": 1.6551,
+      "step": 3530
+    },
+    {
+      "epoch": 2.371859296482412,
+      "grad_norm": 0.4380664527416229,
+      "learning_rate": 0.0002,
+      "loss": 1.6208,
+      "step": 3540
+    },
+    {
+      "epoch": 2.3785594639865995,
+      "grad_norm": 0.5227681994438171,
+      "learning_rate": 0.0002,
+      "loss": 1.634,
+      "step": 3550
+    },
+    {
+      "epoch": 2.3852596314907872,
+      "grad_norm": 0.4382302761077881,
+      "learning_rate": 0.0002,
+      "loss": 1.6146,
+      "step": 3560
+    },
+    {
+      "epoch": 2.391959798994975,
+      "grad_norm": 0.4392451047897339,
+      "learning_rate": 0.0002,
+      "loss": 1.5653,
+      "step": 3570
+    },
+    {
+      "epoch": 2.3986599664991624,
+      "grad_norm": 0.4372786581516266,
+      "learning_rate": 0.0002,
+      "loss": 1.6626,
+      "step": 3580
+    },
+    {
+      "epoch": 2.40536013400335,
+      "grad_norm": 0.5015502572059631,
+      "learning_rate": 0.0002,
+      "loss": 1.519,
+      "step": 3590
+    },
+    {
+      "epoch": 2.4120603015075375,
+      "grad_norm": 0.5653210878372192,
+      "learning_rate": 0.0002,
+      "loss": 1.588,
+      "step": 3600
+    },
+    {
+      "epoch": 2.4187604690117253,
+      "grad_norm": 0.53007972240448,
+      "learning_rate": 0.0002,
+      "loss": 1.6075,
+      "step": 3610
+    },
+    {
+      "epoch": 2.425460636515913,
+      "grad_norm": 0.4659176766872406,
+      "learning_rate": 0.0002,
+      "loss": 1.6421,
+      "step": 3620
+    },
+    {
+      "epoch": 2.4321608040201004,
+      "grad_norm": 0.5637837052345276,
+      "learning_rate": 0.0002,
+      "loss": 1.625,
+      "step": 3630
+    },
+    {
+      "epoch": 2.438860971524288,
+      "grad_norm": 0.4248391389846802,
+      "learning_rate": 0.0002,
+      "loss": 1.6168,
+      "step": 3640
+    },
+    {
+      "epoch": 2.4455611390284755,
+      "grad_norm": 0.44668248295783997,
+      "learning_rate": 0.0002,
+      "loss": 1.6822,
+      "step": 3650
+    },
+    {
+      "epoch": 2.4522613065326633,
+      "grad_norm": 0.43990179896354675,
+      "learning_rate": 0.0002,
+      "loss": 1.6417,
+      "step": 3660
+    },
+    {
+      "epoch": 2.458961474036851,
+      "grad_norm": 0.4532523453235626,
+      "learning_rate": 0.0002,
+      "loss": 1.6723,
+      "step": 3670
+    },
+    {
+      "epoch": 2.4656616415410384,
+      "grad_norm": 0.6605591773986816,
+      "learning_rate": 0.0002,
+      "loss": 1.6957,
+      "step": 3680
+    },
+    {
+      "epoch": 2.472361809045226,
+      "grad_norm": 0.4694533348083496,
+      "learning_rate": 0.0002,
+      "loss": 1.6159,
+      "step": 3690
+    },
+    {
+      "epoch": 2.4790619765494135,
+      "grad_norm": 0.4485011100769043,
+      "learning_rate": 0.0002,
+      "loss": 1.6239,
+      "step": 3700
+    },
+    {
+      "epoch": 2.4857621440536013,
+      "grad_norm": 0.4761785864830017,
+      "learning_rate": 0.0002,
+      "loss": 1.6834,
+      "step": 3710
+    },
+    {
+      "epoch": 2.492462311557789,
+      "grad_norm": 0.5116432309150696,
+      "learning_rate": 0.0002,
+      "loss": 1.6313,
+      "step": 3720
+    },
+    {
+      "epoch": 2.4991624790619764,
+      "grad_norm": 0.49523618817329407,
+      "learning_rate": 0.0002,
+      "loss": 1.5054,
+      "step": 3730
+    },
+    {
+      "epoch": 2.505862646566164,
+      "grad_norm": 0.43826380372047424,
+      "learning_rate": 0.0002,
+      "loss": 1.6249,
+      "step": 3740
+    },
+    {
+      "epoch": 2.5125628140703515,
+      "grad_norm": 0.4916154146194458,
+      "learning_rate": 0.0002,
+      "loss": 1.5762,
+      "step": 3750
+    },
+    {
+      "epoch": 2.5192629815745393,
+      "grad_norm": 0.5381299257278442,
+      "learning_rate": 0.0002,
+      "loss": 1.5157,
+      "step": 3760
+    },
+    {
+      "epoch": 2.525963149078727,
+      "grad_norm": 0.44947415590286255,
+      "learning_rate": 0.0002,
+      "loss": 1.6467,
+      "step": 3770
+    },
+    {
+      "epoch": 2.5326633165829144,
+      "grad_norm": 0.49979084730148315,
+      "learning_rate": 0.0002,
+      "loss": 1.67,
+      "step": 3780
+    },
+    {
+      "epoch": 2.539363484087102,
+      "grad_norm": 0.43046900629997253,
+      "learning_rate": 0.0002,
+      "loss": 1.622,
+      "step": 3790
+    },
+    {
+      "epoch": 2.5460636515912896,
+      "grad_norm": 0.4513470530509949,
+      "learning_rate": 0.0002,
+      "loss": 1.6789,
+      "step": 3800
+    },
+    {
+      "epoch": 2.5527638190954773,
+      "grad_norm": 0.49900051951408386,
+      "learning_rate": 0.0002,
+      "loss": 1.6335,
+      "step": 3810
+    },
+    {
+      "epoch": 2.559463986599665,
+      "grad_norm": 0.4348420202732086,
+      "learning_rate": 0.0002,
+      "loss": 1.6322,
+      "step": 3820
+    },
+    {
+      "epoch": 2.5661641541038525,
+      "grad_norm": 0.4684867560863495,
+      "learning_rate": 0.0002,
+      "loss": 1.6218,
+      "step": 3830
+    },
+    {
+      "epoch": 2.5728643216080402,
+      "grad_norm": 0.44430989027023315,
+      "learning_rate": 0.0002,
+      "loss": 1.6535,
+      "step": 3840
+    },
+    {
+      "epoch": 2.5795644891122276,
+      "grad_norm": 0.47375255823135376,
+      "learning_rate": 0.0002,
+      "loss": 1.5909,
+      "step": 3850
+    },
+    {
+      "epoch": 2.5862646566164154,
+      "grad_norm": 0.45493075251579285,
+      "learning_rate": 0.0002,
+      "loss": 1.6269,
+      "step": 3860
+    },
+    {
+      "epoch": 2.592964824120603,
+      "grad_norm": 0.4563275873661041,
+      "learning_rate": 0.0002,
+      "loss": 1.604,
+      "step": 3870
+    },
+    {
+      "epoch": 2.5996649916247905,
+      "grad_norm": 0.46060335636138916,
+      "learning_rate": 0.0002,
+      "loss": 1.642,
+      "step": 3880
+    },
+    {
+      "epoch": 2.6063651591289783,
+      "grad_norm": 0.4718867540359497,
+      "learning_rate": 0.0002,
+      "loss": 1.6302,
+      "step": 3890
+    },
+    {
+      "epoch": 2.6130653266331656,
+      "grad_norm": 0.41570305824279785,
+      "learning_rate": 0.0002,
+      "loss": 1.6242,
+      "step": 3900
+    },
+    {
+      "epoch": 2.6197654941373534,
+      "grad_norm": 0.4603121876716614,
+      "learning_rate": 0.0002,
+      "loss": 1.6401,
+      "step": 3910
+    },
+    {
+      "epoch": 2.626465661641541,
+      "grad_norm": 0.4734652638435364,
+      "learning_rate": 0.0002,
+      "loss": 1.6839,
+      "step": 3920
+    },
+    {
+      "epoch": 2.6331658291457285,
+      "grad_norm": 0.45348483324050903,
+      "learning_rate": 0.0002,
+      "loss": 1.5448,
+      "step": 3930
+    },
+    {
+      "epoch": 2.6398659966499163,
+      "grad_norm": 0.46559447050094604,
+      "learning_rate": 0.0002,
+      "loss": 1.6157,
+      "step": 3940
+    },
+    {
+      "epoch": 2.6465661641541036,
+      "grad_norm": 0.44113144278526306,
+      "learning_rate": 0.0002,
+      "loss": 1.7052,
+      "step": 3950
+    },
+    {
+      "epoch": 2.6532663316582914,
+      "grad_norm": 0.41415104269981384,
+      "learning_rate": 0.0002,
+      "loss": 1.6315,
+      "step": 3960
+    },
+    {
+      "epoch": 2.659966499162479,
+      "grad_norm": 0.48868080973625183,
+      "learning_rate": 0.0002,
+      "loss": 1.6589,
+      "step": 3970
+    },
+    {
+      "epoch": 2.6666666666666665,
+      "grad_norm": 0.49610549211502075,
+      "learning_rate": 0.0002,
+      "loss": 1.6211,
+      "step": 3980
+    },
+    {
+      "epoch": 2.6733668341708543,
+      "grad_norm": 0.4309130907058716,
+      "learning_rate": 0.0002,
+      "loss": 1.6235,
+      "step": 3990
+    },
+    {
+      "epoch": 2.6800670016750416,
+      "grad_norm": 0.4489327669143677,
+      "learning_rate": 0.0002,
+      "loss": 1.6452,
+      "step": 4000
+    },
+    {
+      "epoch": 2.6867671691792294,
+      "grad_norm": 0.5380139946937561,
+      "learning_rate": 0.0002,
+      "loss": 1.5744,
+      "step": 4010
+    },
+    {
+      "epoch": 2.693467336683417,
+      "grad_norm": 0.5076672434806824,
+      "learning_rate": 0.0002,
+      "loss": 1.6524,
+      "step": 4020
+    },
+    {
+      "epoch": 2.7001675041876045,
+      "grad_norm": 0.47620031237602234,
+      "learning_rate": 0.0002,
+      "loss": 1.636,
+      "step": 4030
+    },
+    {
+      "epoch": 2.7068676716917923,
+      "grad_norm": 0.48089155554771423,
+      "learning_rate": 0.0002,
+      "loss": 1.5543,
+      "step": 4040
+    },
+    {
+      "epoch": 2.7135678391959797,
+      "grad_norm": 0.5108814239501953,
+      "learning_rate": 0.0002,
+      "loss": 1.6396,
+      "step": 4050
+    },
+    {
+      "epoch": 2.7202680067001674,
+      "grad_norm": 0.4196513295173645,
+      "learning_rate": 0.0002,
+      "loss": 1.5905,
+      "step": 4060
+    },
+    {
+      "epoch": 2.726968174204355,
+      "grad_norm": 0.4574664831161499,
+      "learning_rate": 0.0002,
+      "loss": 1.686,
+      "step": 4070
+    },
+    {
+      "epoch": 2.7336683417085426,
+      "grad_norm": 0.4671640992164612,
+      "learning_rate": 0.0002,
+      "loss": 1.6234,
+      "step": 4080
+    },
+    {
+      "epoch": 2.7403685092127303,
+      "grad_norm": 0.49355530738830566,
+      "learning_rate": 0.0002,
+      "loss": 1.6827,
+      "step": 4090
+    },
+    {
+      "epoch": 2.7470686767169177,
+      "grad_norm": 0.46716663241386414,
+      "learning_rate": 0.0002,
+      "loss": 1.6999,
+      "step": 4100
+    },
+    {
+      "epoch": 2.7537688442211055,
+      "grad_norm": 0.45420581102371216,
+      "learning_rate": 0.0002,
+      "loss": 1.6463,
+      "step": 4110
+    },
+    {
+      "epoch": 2.7604690117252932,
+      "grad_norm": 0.4680487811565399,
+      "learning_rate": 0.0002,
+      "loss": 1.5718,
+      "step": 4120
+    },
+    {
+      "epoch": 2.7671691792294806,
+      "grad_norm": 0.5375032424926758,
+      "learning_rate": 0.0002,
+      "loss": 1.5968,
+      "step": 4130
+    },
+    {
+      "epoch": 2.7738693467336684,
+      "grad_norm": 0.46026280522346497,
+      "learning_rate": 0.0002,
+      "loss": 1.5254,
+      "step": 4140
+    },
+    {
+      "epoch": 2.7805695142378557,
+      "grad_norm": 0.43658447265625,
+      "learning_rate": 0.0002,
+      "loss": 1.6613,
+      "step": 4150
+    },
+    {
+      "epoch": 2.7872696817420435,
+      "grad_norm": 0.4935547113418579,
+      "learning_rate": 0.0002,
+      "loss": 1.6546,
+      "step": 4160
+    },
+    {
+      "epoch": 2.7939698492462313,
+      "grad_norm": 0.8167962431907654,
+      "learning_rate": 0.0002,
+      "loss": 1.5961,
+      "step": 4170
+    },
+    {
+      "epoch": 2.8006700167504186,
+      "grad_norm": 0.4289683997631073,
+      "learning_rate": 0.0002,
+      "loss": 1.6907,
+      "step": 4180
+    },
+    {
+      "epoch": 2.8073701842546064,
+      "grad_norm": 0.4569324254989624,
+      "learning_rate": 0.0002,
+      "loss": 1.6385,
+      "step": 4190
+    },
+    {
+      "epoch": 2.8140703517587937,
+      "grad_norm": 0.474795937538147,
+      "learning_rate": 0.0002,
+      "loss": 1.6077,
+      "step": 4200
+    },
+    {
+      "epoch": 2.8207705192629815,
+      "grad_norm": 0.44272229075431824,
+      "learning_rate": 0.0002,
+      "loss": 1.6223,
+      "step": 4210
+    },
+    {
+      "epoch": 2.8274706867671693,
+      "grad_norm": 0.525240957736969,
+      "learning_rate": 0.0002,
+      "loss": 1.6706,
+      "step": 4220
+    },
+    {
+      "epoch": 2.8341708542713566,
+      "grad_norm": 0.4802303910255432,
+      "learning_rate": 0.0002,
+      "loss": 1.7196,
+      "step": 4230
+    },
+    {
+      "epoch": 2.8408710217755444,
+      "grad_norm": 0.46400442719459534,
+      "learning_rate": 0.0002,
+      "loss": 1.6002,
+      "step": 4240
+    },
+    {
+      "epoch": 2.8475711892797317,
+      "grad_norm": 0.49884888529777527,
+      "learning_rate": 0.0002,
+      "loss": 1.6052,
+      "step": 4250
+    },
+    {
+      "epoch": 2.8542713567839195,
+      "grad_norm": 0.5015072226524353,
+      "learning_rate": 0.0002,
+      "loss": 1.6919,
+      "step": 4260
+    },
+    {
+      "epoch": 2.8609715242881073,
+      "grad_norm": 0.4335440695285797,
+      "learning_rate": 0.0002,
+      "loss": 1.6335,
+      "step": 4270
+    },
+    {
+      "epoch": 2.8676716917922946,
+      "grad_norm": 0.5131644606590271,
+      "learning_rate": 0.0002,
+      "loss": 1.5664,
+      "step": 4280
+    },
+    {
+      "epoch": 2.8743718592964824,
+      "grad_norm": 0.6977195739746094,
+      "learning_rate": 0.0002,
+      "loss": 1.6409,
+      "step": 4290
+    },
+    {
+      "epoch": 2.8810720268006698,
+      "grad_norm": 0.5133762955665588,
+      "learning_rate": 0.0002,
+      "loss": 1.7192,
+      "step": 4300
+    },
+    {
+      "epoch": 2.8877721943048575,
+      "grad_norm": 0.4737614393234253,
+      "learning_rate": 0.0002,
+      "loss": 1.6257,
+      "step": 4310
+    },
+    {
+      "epoch": 2.8944723618090453,
+      "grad_norm": 0.4580535590648651,
+      "learning_rate": 0.0002,
+      "loss": 1.6076,
+      "step": 4320
+    },
+    {
+      "epoch": 2.901172529313233,
+      "grad_norm": 0.43863341212272644,
+      "learning_rate": 0.0002,
+      "loss": 1.6538,
+      "step": 4330
+    },
+    {
+      "epoch": 2.9078726968174204,
+      "grad_norm": 0.4103737473487854,
+      "learning_rate": 0.0002,
+      "loss": 1.6091,
+      "step": 4340
+    },
+    {
+      "epoch": 2.914572864321608,
+      "grad_norm": 0.438014417886734,
+      "learning_rate": 0.0002,
+      "loss": 1.7106,
+      "step": 4350
+    },
+    {
+      "epoch": 2.9212730318257956,
+      "grad_norm": 0.5068213939666748,
+      "learning_rate": 0.0002,
+      "loss": 1.6025,
+      "step": 4360
+    },
+    {
+      "epoch": 2.9279731993299833,
+      "grad_norm": 0.45305484533309937,
+      "learning_rate": 0.0002,
+      "loss": 1.6426,
+      "step": 4370
+    },
+    {
+      "epoch": 2.934673366834171,
+      "grad_norm": 0.4612090289592743,
+      "learning_rate": 0.0002,
+      "loss": 1.5726,
+      "step": 4380
+    },
+    {
+      "epoch": 2.9413735343383585,
+      "grad_norm": 0.508736789226532,
+      "learning_rate": 0.0002,
+      "loss": 1.6536,
+      "step": 4390
+    },
+    {
+      "epoch": 2.948073701842546,
+      "grad_norm": 0.4924427270889282,
+      "learning_rate": 0.0002,
+      "loss": 1.6132,
+      "step": 4400
+    },
+    {
+      "epoch": 2.9547738693467336,
+      "grad_norm": 0.5707460641860962,
+      "learning_rate": 0.0002,
+      "loss": 1.7007,
+      "step": 4410
+    },
+    {
+      "epoch": 2.9614740368509214,
+      "grad_norm": 0.42270299792289734,
+      "learning_rate": 0.0002,
+      "loss": 1.6814,
+      "step": 4420
+    },
+    {
+      "epoch": 2.968174204355109,
+      "grad_norm": 0.4429931044578552,
+      "learning_rate": 0.0002,
+      "loss": 1.6644,
+      "step": 4430
+    },
+    {
+      "epoch": 2.9748743718592965,
+      "grad_norm": 0.49760574102401733,
+      "learning_rate": 0.0002,
+      "loss": 1.6251,
+      "step": 4440
+    },
+    {
+      "epoch": 2.981574539363484,
+      "grad_norm": 0.4558229148387909,
+      "learning_rate": 0.0002,
+      "loss": 1.6169,
+      "step": 4450
+    },
+    {
+      "epoch": 2.9882747068676716,
+      "grad_norm": 0.39848530292510986,
+      "learning_rate": 0.0002,
+      "loss": 1.6055,
+      "step": 4460
+    },
+    {
+      "epoch": 2.9949748743718594,
+      "grad_norm": 0.5224862098693848,
+      "learning_rate": 0.0002,
+      "loss": 1.6705,
+      "step": 4470
+    },
+    {
+      "epoch": 2.9996649916247904,
+      "eval_loss": 1.8228833675384521,
+      "eval_runtime": 37.9049,
+      "eval_samples_per_second": 13.587,
+      "eval_steps_per_second": 1.715,
+      "step": 4477
+    },
+    {
+      "epoch": 3.0016750418760467,
+      "grad_norm": 0.41169142723083496,
+      "learning_rate": 0.0002,
+      "loss": 1.6637,
+      "step": 4480
+    },
+    {
+      "epoch": 3.0083752093802345,
+      "grad_norm": 0.4865207374095917,
+      "learning_rate": 0.0002,
+      "loss": 1.5974,
+      "step": 4490
+    },
+    {
+      "epoch": 3.0150753768844223,
+      "grad_norm": 0.5462028384208679,
+      "learning_rate": 0.0002,
+      "loss": 1.5297,
+      "step": 4500
+    },
+    {
+      "epoch": 3.0217755443886096,
+      "grad_norm": 0.6169732809066772,
+      "learning_rate": 0.0002,
+      "loss": 1.5251,
+      "step": 4510
+    },
+    {
+      "epoch": 3.0284757118927974,
+      "grad_norm": 0.5667954087257385,
+      "learning_rate": 0.0002,
+      "loss": 1.5559,
+      "step": 4520
+    },
+    {
+      "epoch": 3.0351758793969847,
+      "grad_norm": 0.5758325457572937,
+      "learning_rate": 0.0002,
+      "loss": 1.5037,
+      "step": 4530
+    },
+    {
+      "epoch": 3.0418760469011725,
+      "grad_norm": 0.5220064520835876,
+      "learning_rate": 0.0002,
+      "loss": 1.4873,
+      "step": 4540
+    },
+    {
+      "epoch": 3.0485762144053603,
+      "grad_norm": 0.5469558835029602,
+      "learning_rate": 0.0002,
+      "loss": 1.5126,
+      "step": 4550
+    },
+    {
+      "epoch": 3.0552763819095476,
+      "grad_norm": 0.5680848956108093,
+      "learning_rate": 0.0002,
+      "loss": 1.4275,
+      "step": 4560
+    },
+    {
+      "epoch": 3.0619765494137354,
+      "grad_norm": 0.5906574726104736,
+      "learning_rate": 0.0002,
+      "loss": 1.5187,
+      "step": 4570
+    },
+    {
+      "epoch": 3.0686767169179228,
+      "grad_norm": 0.4725631773471832,
+      "learning_rate": 0.0002,
+      "loss": 1.4551,
+      "step": 4580
+    },
+    {
+      "epoch": 3.0753768844221105,
+      "grad_norm": 0.5273477435112,
+      "learning_rate": 0.0002,
+      "loss": 1.5083,
+      "step": 4590
+    },
+    {
+      "epoch": 3.0820770519262983,
+      "grad_norm": 0.5861203074455261,
+      "learning_rate": 0.0002,
+      "loss": 1.5154,
+      "step": 4600
+    },
+    {
+      "epoch": 3.0887772194304857,
+      "grad_norm": 0.5343965291976929,
+      "learning_rate": 0.0002,
+      "loss": 1.4924,
+      "step": 4610
+    },
+    {
+      "epoch": 3.0954773869346734,
+      "grad_norm": 0.5348150730133057,
+      "learning_rate": 0.0002,
+      "loss": 1.5608,
+      "step": 4620
+    },
+    {
+      "epoch": 3.102177554438861,
+      "grad_norm": 0.5971846580505371,
+      "learning_rate": 0.0002,
+      "loss": 1.5399,
+      "step": 4630
+    },
+    {
+      "epoch": 3.1088777219430486,
+      "grad_norm": 0.5203177332878113,
+      "learning_rate": 0.0002,
+      "loss": 1.4662,
+      "step": 4640
+    },
+    {
+      "epoch": 3.1155778894472363,
+      "grad_norm": 0.55289226770401,
+      "learning_rate": 0.0002,
+      "loss": 1.5805,
+      "step": 4650
+    },
+    {
+      "epoch": 3.1222780569514237,
+      "grad_norm": 0.6878530979156494,
+      "learning_rate": 0.0002,
+      "loss": 1.4745,
+      "step": 4660
+    },
+    {
+      "epoch": 3.1289782244556115,
+      "grad_norm": 0.6173256635665894,
+      "learning_rate": 0.0002,
+      "loss": 1.5335,
+      "step": 4670
+    },
+    {
+      "epoch": 3.135678391959799,
+      "grad_norm": 0.536796510219574,
+      "learning_rate": 0.0002,
+      "loss": 1.51,
+      "step": 4680
+    },
+    {
+      "epoch": 3.1423785594639866,
+      "grad_norm": 0.58846116065979,
+      "learning_rate": 0.0002,
+      "loss": 1.4713,
+      "step": 4690
+    },
+    {
+      "epoch": 3.1490787269681744,
+      "grad_norm": 0.645889401435852,
+      "learning_rate": 0.0002,
+      "loss": 1.5114,
+      "step": 4700
+    },
+    {
+      "epoch": 3.1557788944723617,
+      "grad_norm": 0.6118691563606262,
+      "learning_rate": 0.0002,
+      "loss": 1.4705,
+      "step": 4710
+    },
+    {
+      "epoch": 3.1624790619765495,
+      "grad_norm": 0.5189669132232666,
+      "learning_rate": 0.0002,
+      "loss": 1.5533,
+      "step": 4720
+    },
+    {
+      "epoch": 3.169179229480737,
+      "grad_norm": 0.5794713497161865,
+      "learning_rate": 0.0002,
+      "loss": 1.4769,
+      "step": 4730
+    },
+    {
+      "epoch": 3.1758793969849246,
+      "grad_norm": 0.6579326391220093,
+      "learning_rate": 0.0002,
+      "loss": 1.4849,
+      "step": 4740
+    },
+    {
+      "epoch": 3.1825795644891124,
+      "grad_norm": 0.5822742581367493,
+      "learning_rate": 0.0002,
+      "loss": 1.545,
+      "step": 4750
+    },
+    {
+      "epoch": 3.1892797319932997,
+      "grad_norm": 0.5475956201553345,
+      "learning_rate": 0.0002,
+      "loss": 1.4358,
+      "step": 4760
+    },
+    {
+      "epoch": 3.1959798994974875,
+      "grad_norm": 0.6743834018707275,
+      "learning_rate": 0.0002,
+      "loss": 1.4723,
+      "step": 4770
+    },
+    {
+      "epoch": 3.202680067001675,
+      "grad_norm": 0.6110585927963257,
+      "learning_rate": 0.0002,
+      "loss": 1.5161,
+      "step": 4780
+    },
+    {
+      "epoch": 3.2093802345058626,
+      "grad_norm": 0.5426181554794312,
+      "learning_rate": 0.0002,
+      "loss": 1.5455,
+      "step": 4790
+    },
+    {
+      "epoch": 3.2160804020100504,
+      "grad_norm": 0.6077824234962463,
+      "learning_rate": 0.0002,
+      "loss": 1.5315,
+      "step": 4800
+    },
+    {
+      "epoch": 3.2227805695142377,
+      "grad_norm": 0.5785858631134033,
+      "learning_rate": 0.0002,
+      "loss": 1.5314,
+      "step": 4810
+    },
+    {
+      "epoch": 3.2294807370184255,
+      "grad_norm": 0.6425958275794983,
+      "learning_rate": 0.0002,
+      "loss": 1.4041,
+      "step": 4820
+    },
+    {
+      "epoch": 3.236180904522613,
+      "grad_norm": 0.6607080698013306,
+      "learning_rate": 0.0002,
+      "loss": 1.4751,
+      "step": 4830
+    },
+    {
+      "epoch": 3.2428810720268006,
+      "grad_norm": 0.5385788679122925,
+      "learning_rate": 0.0002,
+      "loss": 1.5267,
+      "step": 4840
+    },
+    {
+      "epoch": 3.2495812395309884,
+      "grad_norm": 0.5630403757095337,
+      "learning_rate": 0.0002,
+      "loss": 1.4673,
+      "step": 4850
+    },
+    {
+      "epoch": 3.2562814070351758,
+      "grad_norm": 0.6340779662132263,
+      "learning_rate": 0.0002,
+      "loss": 1.5257,
+      "step": 4860
+    },
+    {
+      "epoch": 3.2629815745393635,
+      "grad_norm": 0.5305342674255371,
+      "learning_rate": 0.0002,
+      "loss": 1.5148,
+      "step": 4870
+    },
+    {
+      "epoch": 3.2696817420435513,
+      "grad_norm": 0.597670316696167,
+      "learning_rate": 0.0002,
+      "loss": 1.5162,
+      "step": 4880
+    },
+    {
+      "epoch": 3.2763819095477387,
+      "grad_norm": 0.665553867816925,
+      "learning_rate": 0.0002,
+      "loss": 1.5429,
+      "step": 4890
+    },
+    {
+      "epoch": 3.2830820770519265,
+      "grad_norm": 0.579767644405365,
+      "learning_rate": 0.0002,
+      "loss": 1.4607,
+      "step": 4900
+    },
+    {
+      "epoch": 3.289782244556114,
+      "grad_norm": 0.5512481331825256,
+      "learning_rate": 0.0002,
+      "loss": 1.4999,
+      "step": 4910
+    },
+    {
+      "epoch": 3.2964824120603016,
+      "grad_norm": 0.5916532278060913,
+      "learning_rate": 0.0002,
+      "loss": 1.5022,
+      "step": 4920
+    },
+    {
+      "epoch": 3.3031825795644894,
+      "grad_norm": 0.7521726489067078,
+      "learning_rate": 0.0002,
+      "loss": 1.4889,
+      "step": 4930
+    },
+    {
+      "epoch": 3.3098827470686767,
+      "grad_norm": 0.5352797508239746,
+      "learning_rate": 0.0002,
+      "loss": 1.4223,
+      "step": 4940
+    },
+    {
+      "epoch": 3.3165829145728645,
+      "grad_norm": 0.5950371623039246,
+      "learning_rate": 0.0002,
+      "loss": 1.5122,
+      "step": 4950
+    },
+    {
+      "epoch": 3.323283082077052,
+      "grad_norm": 0.8020477890968323,
+      "learning_rate": 0.0002,
+      "loss": 1.5072,
+      "step": 4960
+    },
+    {
+      "epoch": 3.3299832495812396,
+      "grad_norm": 0.6790024638175964,
+      "learning_rate": 0.0002,
+      "loss": 1.5422,
+      "step": 4970
+    },
+    {
+      "epoch": 3.3366834170854274,
+      "grad_norm": 0.687627375125885,
+      "learning_rate": 0.0002,
+      "loss": 1.5363,
+      "step": 4980
+    },
+    {
+      "epoch": 3.3433835845896147,
+      "grad_norm": 0.6094385385513306,
+      "learning_rate": 0.0002,
+      "loss": 1.5276,
+      "step": 4990
+    },
+    {
+      "epoch": 3.3500837520938025,
+      "grad_norm": 0.6541242003440857,
+      "learning_rate": 0.0002,
+      "loss": 1.549,
+      "step": 5000
+    },
+    {
+      "epoch": 3.35678391959799,
+      "grad_norm": 0.5560880303382874,
+      "learning_rate": 0.0002,
+      "loss": 1.6067,
+      "step": 5010
+    },
+    {
+      "epoch": 3.3634840871021776,
+      "grad_norm": 0.5440094470977783,
+      "learning_rate": 0.0002,
+      "loss": 1.5769,
+      "step": 5020
+    },
+    {
+      "epoch": 3.3701842546063654,
+      "grad_norm": 0.5749301314353943,
+      "learning_rate": 0.0002,
+      "loss": 1.6183,
+      "step": 5030
+    },
+    {
+      "epoch": 3.3768844221105527,
+      "grad_norm": 0.5919716954231262,
+      "learning_rate": 0.0002,
+      "loss": 1.4801,
+      "step": 5040
+    },
+    {
+      "epoch": 3.3835845896147405,
+      "grad_norm": 0.6331481337547302,
+      "learning_rate": 0.0002,
+      "loss": 1.5957,
+      "step": 5050
+    },
+    {
+      "epoch": 3.390284757118928,
+      "grad_norm": 0.5687161684036255,
+      "learning_rate": 0.0002,
+      "loss": 1.5188,
+      "step": 5060
+    },
+    {
+      "epoch": 3.3969849246231156,
+      "grad_norm": 0.6718577742576599,
+      "learning_rate": 0.0002,
+      "loss": 1.5702,
+      "step": 5070
+    },
+    {
+      "epoch": 3.4036850921273034,
+      "grad_norm": 0.5089324116706848,
+      "learning_rate": 0.0002,
+      "loss": 1.5577,
+      "step": 5080
+    },
+    {
+      "epoch": 3.4103852596314908,
+      "grad_norm": 0.5710174441337585,
+      "learning_rate": 0.0002,
+      "loss": 1.512,
+      "step": 5090
+    },
+    {
+      "epoch": 3.4170854271356785,
+      "grad_norm": 0.6670721173286438,
+      "learning_rate": 0.0002,
+      "loss": 1.5492,
+      "step": 5100
+    },
+    {
+      "epoch": 3.423785594639866,
+      "grad_norm": 0.6875665187835693,
+      "learning_rate": 0.0002,
+      "loss": 1.5227,
+      "step": 5110
+    },
+    {
+      "epoch": 3.4304857621440537,
+      "grad_norm": 0.5375880599021912,
+      "learning_rate": 0.0002,
+      "loss": 1.4496,
+      "step": 5120
+    },
+    {
+      "epoch": 3.4371859296482414,
+      "grad_norm": 0.6550399661064148,
+      "learning_rate": 0.0002,
+      "loss": 1.5527,
+      "step": 5130
+    },
+    {
+      "epoch": 3.4438860971524288,
+      "grad_norm": 0.5948067903518677,
+      "learning_rate": 0.0002,
+      "loss": 1.5687,
+      "step": 5140
+    },
+    {
+      "epoch": 3.4505862646566166,
+      "grad_norm": 0.6134477257728577,
+      "learning_rate": 0.0002,
+      "loss": 1.4813,
+      "step": 5150
+    },
+    {
+      "epoch": 3.457286432160804,
+      "grad_norm": 0.6506398320198059,
+      "learning_rate": 0.0002,
+      "loss": 1.5069,
+      "step": 5160
+    },
+    {
+      "epoch": 3.4639865996649917,
+      "grad_norm": 0.6060147881507874,
+      "learning_rate": 0.0002,
+      "loss": 1.4422,
+      "step": 5170
+    },
+    {
+      "epoch": 3.4706867671691795,
+      "grad_norm": 0.6173806190490723,
+      "learning_rate": 0.0002,
+      "loss": 1.5093,
+      "step": 5180
+    },
+    {
+      "epoch": 3.477386934673367,
+      "grad_norm": 0.6032607555389404,
+      "learning_rate": 0.0002,
+      "loss": 1.4975,
+      "step": 5190
+    },
+    {
+      "epoch": 3.4840871021775546,
+      "grad_norm": 0.5652492046356201,
+      "learning_rate": 0.0002,
+      "loss": 1.4979,
+      "step": 5200
+    },
+    {
+      "epoch": 3.490787269681742,
+      "grad_norm": 0.6168607473373413,
+      "learning_rate": 0.0002,
+      "loss": 1.4883,
+      "step": 5210
+    },
+    {
+      "epoch": 3.4974874371859297,
+      "grad_norm": 0.6170629262924194,
+      "learning_rate": 0.0002,
+      "loss": 1.5164,
+      "step": 5220
+    },
+    {
+      "epoch": 3.5041876046901175,
+      "grad_norm": 0.6926297545433044,
+      "learning_rate": 0.0002,
+      "loss": 1.4879,
+      "step": 5230
+    },
+    {
+      "epoch": 3.510887772194305,
+      "grad_norm": 0.6702437996864319,
+      "learning_rate": 0.0002,
+      "loss": 1.4982,
+      "step": 5240
+    },
+    {
+      "epoch": 3.5175879396984926,
+      "grad_norm": 0.5421436429023743,
+      "learning_rate": 0.0002,
+      "loss": 1.4986,
+      "step": 5250
+    },
+    {
+      "epoch": 3.52428810720268,
+      "grad_norm": 0.5726765990257263,
+      "learning_rate": 0.0002,
+      "loss": 1.4673,
+      "step": 5260
+    },
+    {
+      "epoch": 3.5309882747068677,
+      "grad_norm": 0.5685455203056335,
+      "learning_rate": 0.0002,
+      "loss": 1.5423,
+      "step": 5270
+    },
+    {
+      "epoch": 3.5376884422110555,
+      "grad_norm": 0.6018396019935608,
+      "learning_rate": 0.0002,
+      "loss": 1.4715,
+      "step": 5280
+    },
+    {
+      "epoch": 3.544388609715243,
+      "grad_norm": 0.5731932520866394,
+      "learning_rate": 0.0002,
+      "loss": 1.5451,
+      "step": 5290
+    },
+    {
+      "epoch": 3.5510887772194306,
+      "grad_norm": 0.6601519584655762,
+      "learning_rate": 0.0002,
+      "loss": 1.4752,
+      "step": 5300
+    },
+    {
+      "epoch": 3.557788944723618,
+      "grad_norm": 0.5545530319213867,
+      "learning_rate": 0.0002,
+      "loss": 1.5434,
+      "step": 5310
+    },
+    {
+      "epoch": 3.5644891122278057,
+      "grad_norm": 0.5998541116714478,
+      "learning_rate": 0.0002,
+      "loss": 1.5438,
+      "step": 5320
+    },
+    {
+      "epoch": 3.5711892797319935,
+      "grad_norm": 0.5651767253875732,
+      "learning_rate": 0.0002,
+      "loss": 1.56,
+      "step": 5330
+    },
+    {
+      "epoch": 3.577889447236181,
+      "grad_norm": 0.7425084114074707,
+      "learning_rate": 0.0002,
+      "loss": 1.4829,
+      "step": 5340
+    },
+    {
+      "epoch": 3.5845896147403686,
+      "grad_norm": 0.5770602226257324,
+      "learning_rate": 0.0002,
+      "loss": 1.5571,
+      "step": 5350
+    },
+    {
+      "epoch": 3.591289782244556,
+      "grad_norm": 0.54723060131073,
+      "learning_rate": 0.0002,
+      "loss": 1.458,
+      "step": 5360
+    },
+    {
+      "epoch": 3.5979899497487438,
+      "grad_norm": 0.6658238172531128,
+      "learning_rate": 0.0002,
+      "loss": 1.497,
+      "step": 5370
+    },
+    {
+      "epoch": 3.6046901172529315,
+      "grad_norm": 0.5787645578384399,
+      "learning_rate": 0.0002,
+      "loss": 1.5456,
+      "step": 5380
+    },
+    {
+      "epoch": 3.611390284757119,
+      "grad_norm": 0.594913125038147,
+      "learning_rate": 0.0002,
+      "loss": 1.5343,
+      "step": 5390
+    },
+    {
+      "epoch": 3.6180904522613067,
+      "grad_norm": 0.4964977502822876,
+      "learning_rate": 0.0002,
+      "loss": 1.4727,
+      "step": 5400
+    },
+    {
+      "epoch": 3.624790619765494,
+      "grad_norm": 0.6087527275085449,
+      "learning_rate": 0.0002,
+      "loss": 1.5062,
+      "step": 5410
+    },
+    {
+      "epoch": 3.6314907872696818,
+      "grad_norm": 0.6315323710441589,
+      "learning_rate": 0.0002,
+      "loss": 1.5098,
+      "step": 5420
+    },
+    {
+      "epoch": 3.6381909547738696,
+      "grad_norm": 0.574799120426178,
+      "learning_rate": 0.0002,
+      "loss": 1.4855,
+      "step": 5430
+    },
+    {
+      "epoch": 3.644891122278057,
+      "grad_norm": 0.5949277877807617,
+      "learning_rate": 0.0002,
+      "loss": 1.4595,
+      "step": 5440
+    },
+    {
+      "epoch": 3.6515912897822447,
+      "grad_norm": 0.5640677213668823,
+      "learning_rate": 0.0002,
+      "loss": 1.4816,
+      "step": 5450
+    },
+    {
+      "epoch": 3.658291457286432,
+      "grad_norm": 0.6198237538337708,
+      "learning_rate": 0.0002,
+      "loss": 1.525,
+      "step": 5460
+    },
+    {
+      "epoch": 3.66499162479062,
+      "grad_norm": 0.6902034878730774,
+      "learning_rate": 0.0002,
+      "loss": 1.5676,
+      "step": 5470
+    },
+    {
+      "epoch": 3.6716917922948076,
+      "grad_norm": 0.5686674118041992,
+      "learning_rate": 0.0002,
+      "loss": 1.544,
+      "step": 5480
+    },
+    {
+      "epoch": 3.678391959798995,
+      "grad_norm": 0.6532107591629028,
+      "learning_rate": 0.0002,
+      "loss": 1.5255,
+      "step": 5490
+    },
+    {
+      "epoch": 3.6850921273031827,
+      "grad_norm": 0.5790849924087524,
+      "learning_rate": 0.0002,
+      "loss": 1.5767,
+      "step": 5500
+    },
+    {
+      "epoch": 3.69179229480737,
+      "grad_norm": 0.6055065393447876,
+      "learning_rate": 0.0002,
+      "loss": 1.5507,
+      "step": 5510
+    },
+    {
+      "epoch": 3.698492462311558,
+      "grad_norm": 0.5630605816841125,
+      "learning_rate": 0.0002,
+      "loss": 1.4656,
+      "step": 5520
+    },
+    {
+      "epoch": 3.7051926298157456,
+      "grad_norm": 0.6005825996398926,
+      "learning_rate": 0.0002,
+      "loss": 1.537,
+      "step": 5530
+    },
+    {
+      "epoch": 3.711892797319933,
+      "grad_norm": 0.6553038954734802,
+      "learning_rate": 0.0002,
+      "loss": 1.5313,
+      "step": 5540
+    },
+    {
+      "epoch": 3.7185929648241207,
+      "grad_norm": 0.5601094961166382,
+      "learning_rate": 0.0002,
+      "loss": 1.4943,
+      "step": 5550
+    },
+    {
+      "epoch": 3.725293132328308,
+      "grad_norm": 0.6598808169364929,
+      "learning_rate": 0.0002,
+      "loss": 1.5161,
+      "step": 5560
+    },
+    {
+      "epoch": 3.731993299832496,
+      "grad_norm": 0.5506255626678467,
+      "learning_rate": 0.0002,
+      "loss": 1.5345,
+      "step": 5570
+    },
+    {
+      "epoch": 3.7386934673366836,
+      "grad_norm": 0.6001223921775818,
+      "learning_rate": 0.0002,
+      "loss": 1.4805,
+      "step": 5580
+    },
+    {
+      "epoch": 3.745393634840871,
+      "grad_norm": 0.6287297606468201,
+      "learning_rate": 0.0002,
+      "loss": 1.4652,
+      "step": 5590
+    },
+    {
+      "epoch": 3.7520938023450587,
+      "grad_norm": 0.6253238916397095,
+      "learning_rate": 0.0002,
+      "loss": 1.5246,
+      "step": 5600
+    },
+    {
+      "epoch": 3.758793969849246,
+      "grad_norm": 0.5713174939155579,
+      "learning_rate": 0.0002,
+      "loss": 1.5691,
+      "step": 5610
+    },
+    {
+      "epoch": 3.765494137353434,
+      "grad_norm": 0.6198310852050781,
+      "learning_rate": 0.0002,
+      "loss": 1.5661,
+      "step": 5620
+    },
+    {
+      "epoch": 3.7721943048576216,
+      "grad_norm": 0.5941224098205566,
+      "learning_rate": 0.0002,
+      "loss": 1.5448,
+      "step": 5630
+    },
+    {
+      "epoch": 3.778894472361809,
+      "grad_norm": 0.606002151966095,
+      "learning_rate": 0.0002,
+      "loss": 1.4925,
+      "step": 5640
+    },
+    {
+      "epoch": 3.7855946398659968,
+      "grad_norm": 0.6540704965591431,
+      "learning_rate": 0.0002,
+      "loss": 1.5182,
+      "step": 5650
+    },
+    {
+      "epoch": 3.792294807370184,
+      "grad_norm": 0.6147415041923523,
+      "learning_rate": 0.0002,
+      "loss": 1.5903,
+      "step": 5660
+    },
+    {
+      "epoch": 3.798994974874372,
+      "grad_norm": 0.5649605393409729,
+      "learning_rate": 0.0002,
+      "loss": 1.5329,
+      "step": 5670
+    },
+    {
+      "epoch": 3.8056951423785597,
+      "grad_norm": 0.6788773536682129,
+      "learning_rate": 0.0002,
+      "loss": 1.5747,
+      "step": 5680
+    },
+    {
+      "epoch": 3.812395309882747,
+      "grad_norm": 0.6581860780715942,
+      "learning_rate": 0.0002,
+      "loss": 1.535,
+      "step": 5690
+    },
+    {
+      "epoch": 3.819095477386935,
+      "grad_norm": 0.5529348850250244,
+      "learning_rate": 0.0002,
+      "loss": 1.4587,
+      "step": 5700
+    },
+    {
+      "epoch": 3.825795644891122,
+      "grad_norm": 0.6320232152938843,
+      "learning_rate": 0.0002,
+      "loss": 1.5594,
+      "step": 5710
+    },
+    {
+      "epoch": 3.83249581239531,
+      "grad_norm": 0.6529698371887207,
+      "learning_rate": 0.0002,
+      "loss": 1.4696,
+      "step": 5720
+    },
+    {
+      "epoch": 3.8391959798994977,
+      "grad_norm": 0.5983362793922424,
+      "learning_rate": 0.0002,
+      "loss": 1.5854,
+      "step": 5730
+    },
+    {
+      "epoch": 3.845896147403685,
+      "grad_norm": 0.6335684061050415,
+      "learning_rate": 0.0002,
+      "loss": 1.465,
+      "step": 5740
+    },
+    {
+      "epoch": 3.852596314907873,
+      "grad_norm": 0.700446605682373,
+      "learning_rate": 0.0002,
+      "loss": 1.5545,
+      "step": 5750
+    },
+    {
+      "epoch": 3.85929648241206,
+      "grad_norm": 0.6092597842216492,
+      "learning_rate": 0.0002,
+      "loss": 1.5707,
+      "step": 5760
+    },
+    {
+      "epoch": 3.865996649916248,
+      "grad_norm": 0.564146101474762,
+      "learning_rate": 0.0002,
+      "loss": 1.5729,
+      "step": 5770
+    },
+    {
+      "epoch": 3.8726968174204357,
+      "grad_norm": 0.615275502204895,
+      "learning_rate": 0.0002,
+      "loss": 1.5872,
+      "step": 5780
+    },
+    {
+      "epoch": 3.879396984924623,
+      "grad_norm": 0.6685376763343811,
+      "learning_rate": 0.0002,
+      "loss": 1.5142,
+      "step": 5790
+    },
+    {
+      "epoch": 3.886097152428811,
+      "grad_norm": 0.6116922497749329,
+      "learning_rate": 0.0002,
+      "loss": 1.4752,
+      "step": 5800
+    },
+    {
+      "epoch": 3.892797319932998,
+      "grad_norm": 0.5486813187599182,
+      "learning_rate": 0.0002,
+      "loss": 1.5179,
+      "step": 5810
+    },
+    {
+      "epoch": 3.899497487437186,
+      "grad_norm": 0.6208204030990601,
+      "learning_rate": 0.0002,
+      "loss": 1.5167,
+      "step": 5820
+    },
+    {
+      "epoch": 3.9061976549413737,
+      "grad_norm": 0.6500625014305115,
+      "learning_rate": 0.0002,
+      "loss": 1.5334,
+      "step": 5830
+    },
+    {
+      "epoch": 3.912897822445561,
+      "grad_norm": 0.5948089361190796,
+      "learning_rate": 0.0002,
+      "loss": 1.4716,
+      "step": 5840
+    },
+    {
+      "epoch": 3.919597989949749,
+      "grad_norm": 0.7210732698440552,
+      "learning_rate": 0.0002,
+      "loss": 1.6011,
+      "step": 5850
+    },
+    {
+      "epoch": 3.926298157453936,
+      "grad_norm": 0.6662322878837585,
+      "learning_rate": 0.0002,
+      "loss": 1.5519,
+      "step": 5860
+    },
+    {
+      "epoch": 3.932998324958124,
+      "grad_norm": 0.5613839626312256,
+      "learning_rate": 0.0002,
+      "loss": 1.5656,
+      "step": 5870
+    },
+    {
+      "epoch": 3.9396984924623117,
+      "grad_norm": 0.6069002151489258,
+      "learning_rate": 0.0002,
+      "loss": 1.544,
+      "step": 5880
+    },
+    {
+      "epoch": 3.946398659966499,
+      "grad_norm": 0.7075562477111816,
+      "learning_rate": 0.0002,
+      "loss": 1.6745,
+      "step": 5890
+    },
+    {
+      "epoch": 3.953098827470687,
+      "grad_norm": 0.6316173076629639,
+      "learning_rate": 0.0002,
+      "loss": 1.5391,
+      "step": 5900
+    },
+    {
+      "epoch": 3.959798994974874,
+      "grad_norm": 0.5716308355331421,
+      "learning_rate": 0.0002,
+      "loss": 1.6314,
+      "step": 5910
+    },
+    {
+      "epoch": 3.966499162479062,
+      "grad_norm": 0.6800096035003662,
+      "learning_rate": 0.0002,
+      "loss": 1.5947,
+      "step": 5920
+    },
+    {
+      "epoch": 3.9731993299832498,
+      "grad_norm": 0.6057983040809631,
+      "learning_rate": 0.0002,
+      "loss": 1.5189,
+      "step": 5930
+    },
+    {
+      "epoch": 3.979899497487437,
+      "grad_norm": 0.5938987731933594,
+      "learning_rate": 0.0002,
+      "loss": 1.5431,
+      "step": 5940
+    },
+    {
+      "epoch": 3.986599664991625,
+      "grad_norm": 0.6963576674461365,
+      "learning_rate": 0.0002,
+      "loss": 1.5111,
+      "step": 5950
+    },
+    {
+      "epoch": 3.993299832495812,
+      "grad_norm": 0.6279940009117126,
+      "learning_rate": 0.0002,
+      "loss": 1.5521,
+      "step": 5960
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 0.7161159515380859,
+      "learning_rate": 0.0002,
+      "loss": 1.5974,
+      "step": 5970
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 1.8655421733856201,
+      "eval_runtime": 37.9276,
+      "eval_samples_per_second": 13.579,
+      "eval_steps_per_second": 1.714,
+      "step": 5970
+    },
+    {
+      "epoch": 4.006700167504188,
+      "grad_norm": 0.7380476593971252,
+      "learning_rate": 0.0002,
+      "loss": 1.3666,
+      "step": 5980
+    },
+    {
+      "epoch": 4.013400335008376,
+      "grad_norm": 0.7148947715759277,
+      "learning_rate": 0.0002,
+      "loss": 1.3913,
+      "step": 5990
+    },
+    {
+      "epoch": 4.0201005025125625,
+      "grad_norm": 0.6177082657814026,
+      "learning_rate": 0.0002,
+      "loss": 1.4204,
+      "step": 6000
+    },
+    {
+      "epoch": 4.02680067001675,
+      "grad_norm": 0.8552946448326111,
+      "learning_rate": 0.0002,
+      "loss": 1.4421,
+      "step": 6010
+    },
+    {
+      "epoch": 4.033500837520938,
+      "grad_norm": 0.8033416271209717,
+      "learning_rate": 0.0002,
+      "loss": 1.4342,
+      "step": 6020
+    },
+    {
+      "epoch": 4.040201005025126,
+      "grad_norm": 0.8501318097114563,
+      "learning_rate": 0.0002,
+      "loss": 1.4092,
+      "step": 6030
+    },
+    {
+      "epoch": 4.046901172529314,
+      "grad_norm": 0.6981393098831177,
+      "learning_rate": 0.0002,
+      "loss": 1.3367,
+      "step": 6040
+    },
+    {
+      "epoch": 4.0536013400335005,
+      "grad_norm": 0.7227180600166321,
+      "learning_rate": 0.0002,
+      "loss": 1.3925,
+      "step": 6050
+    },
+    {
+      "epoch": 4.060301507537688,
+      "grad_norm": 0.6923989653587341,
+      "learning_rate": 0.0002,
+      "loss": 1.4007,
+      "step": 6060
+    },
+    {
+      "epoch": 4.067001675041876,
+      "grad_norm": 0.879779040813446,
+      "learning_rate": 0.0002,
+      "loss": 1.3837,
+      "step": 6070
+    },
+    {
+      "epoch": 4.073701842546064,
+      "grad_norm": 0.8184754848480225,
+      "learning_rate": 0.0002,
+      "loss": 1.4383,
+      "step": 6080
+    },
+    {
+      "epoch": 4.080402010050252,
+      "grad_norm": 0.8211342692375183,
+      "learning_rate": 0.0002,
+      "loss": 1.3128,
+      "step": 6090
+    },
+    {
+      "epoch": 4.0871021775544385,
+      "grad_norm": 0.7542396783828735,
+      "learning_rate": 0.0002,
+      "loss": 1.3892,
+      "step": 6100
+    },
+    {
+      "epoch": 4.093802345058626,
+      "grad_norm": 0.6631066799163818,
+      "learning_rate": 0.0002,
+      "loss": 1.3607,
+      "step": 6110
+    },
+    {
+      "epoch": 4.100502512562814,
+      "grad_norm": 0.6728386282920837,
+      "learning_rate": 0.0002,
+      "loss": 1.3275,
+      "step": 6120
+    },
+    {
+      "epoch": 4.107202680067002,
+      "grad_norm": 0.681851863861084,
+      "learning_rate": 0.0002,
+      "loss": 1.3443,
+      "step": 6130
+    },
+    {
+      "epoch": 4.11390284757119,
+      "grad_norm": 0.8757794499397278,
+      "learning_rate": 0.0002,
+      "loss": 1.3486,
+      "step": 6140
+    },
+    {
+      "epoch": 4.1206030150753765,
+      "grad_norm": 0.6567301750183105,
+      "learning_rate": 0.0002,
+      "loss": 1.351,
+      "step": 6150
+    },
+    {
+      "epoch": 4.127303182579564,
+      "grad_norm": 0.7950329184532166,
+      "learning_rate": 0.0002,
+      "loss": 1.3824,
+      "step": 6160
+    },
+    {
+      "epoch": 4.134003350083752,
+      "grad_norm": 0.7545644044876099,
+      "learning_rate": 0.0002,
+      "loss": 1.3738,
+      "step": 6170
+    },
+    {
+      "epoch": 4.14070351758794,
+      "grad_norm": 0.7172710299491882,
+      "learning_rate": 0.0002,
+      "loss": 1.4214,
+      "step": 6180
+    },
+    {
+      "epoch": 4.147403685092128,
+      "grad_norm": 0.7040584087371826,
+      "learning_rate": 0.0002,
+      "loss": 1.4091,
+      "step": 6190
+    },
+    {
+      "epoch": 4.1541038525963145,
+      "grad_norm": 0.7482913732528687,
+      "learning_rate": 0.0002,
+      "loss": 1.4149,
+      "step": 6200
+    },
+    {
+      "epoch": 4.160804020100502,
+      "grad_norm": 0.8523276448249817,
+      "learning_rate": 0.0002,
+      "loss": 1.3227,
+      "step": 6210
+    },
+    {
+      "epoch": 4.16750418760469,
+      "grad_norm": 0.6672041416168213,
+      "learning_rate": 0.0002,
+      "loss": 1.4194,
+      "step": 6220
+    },
+    {
+      "epoch": 4.174204355108878,
+      "grad_norm": 0.7523500919342041,
+      "learning_rate": 0.0002,
+      "loss": 1.3953,
+      "step": 6230
+    },
+    {
+      "epoch": 4.180904522613066,
+      "grad_norm": 0.8085253834724426,
+      "learning_rate": 0.0002,
+      "loss": 1.371,
+      "step": 6240
+    },
+    {
+      "epoch": 4.187604690117253,
+      "grad_norm": 0.789450466632843,
+      "learning_rate": 0.0002,
+      "loss": 1.3293,
+      "step": 6250
+    },
+    {
+      "epoch": 4.19430485762144,
+      "grad_norm": 0.7502310872077942,
+      "learning_rate": 0.0002,
+      "loss": 1.3539,
+      "step": 6260
+    },
+    {
+      "epoch": 4.201005025125628,
+      "grad_norm": 0.7397456765174866,
+      "learning_rate": 0.0002,
+      "loss": 1.3415,
+      "step": 6270
+    },
+    {
+      "epoch": 4.207705192629816,
+      "grad_norm": 0.6921947002410889,
+      "learning_rate": 0.0002,
+      "loss": 1.3963,
+      "step": 6280
+    },
+    {
+      "epoch": 4.214405360134004,
+      "grad_norm": 0.9334571957588196,
+      "learning_rate": 0.0002,
+      "loss": 1.3125,
+      "step": 6290
+    },
+    {
+      "epoch": 4.221105527638191,
+      "grad_norm": 0.725799024105072,
+      "learning_rate": 0.0002,
+      "loss": 1.3612,
+      "step": 6300
+    },
+    {
+      "epoch": 4.227805695142378,
+      "grad_norm": 0.8290495872497559,
+      "learning_rate": 0.0002,
+      "loss": 1.4217,
+      "step": 6310
+    },
+    {
+      "epoch": 4.234505862646566,
+      "grad_norm": 0.688983678817749,
+      "learning_rate": 0.0002,
+      "loss": 1.4135,
+      "step": 6320
+    },
+    {
+      "epoch": 4.241206030150754,
+      "grad_norm": 0.8620913028717041,
+      "learning_rate": 0.0002,
+      "loss": 1.3807,
+      "step": 6330
+    },
+    {
+      "epoch": 4.247906197654942,
+      "grad_norm": 0.8008657693862915,
+      "learning_rate": 0.0002,
+      "loss": 1.3738,
+      "step": 6340
+    },
+    {
+      "epoch": 4.254606365159129,
+      "grad_norm": 0.7379199266433716,
+      "learning_rate": 0.0002,
+      "loss": 1.4005,
+      "step": 6350
+    },
+    {
+      "epoch": 4.261306532663316,
+      "grad_norm": 0.7842815518379211,
+      "learning_rate": 0.0002,
+      "loss": 1.426,
+      "step": 6360
+    },
+    {
+      "epoch": 4.268006700167504,
+      "grad_norm": 0.812600314617157,
+      "learning_rate": 0.0002,
+      "loss": 1.4262,
+      "step": 6370
+    },
+    {
+      "epoch": 4.274706867671692,
+      "grad_norm": 0.7852841019630432,
+      "learning_rate": 0.0002,
+      "loss": 1.4028,
+      "step": 6380
+    },
+    {
+      "epoch": 4.28140703517588,
+      "grad_norm": 1.0377534627914429,
+      "learning_rate": 0.0002,
+      "loss": 1.3722,
+      "step": 6390
+    },
+    {
+      "epoch": 4.288107202680067,
+      "grad_norm": 1.03935706615448,
+      "learning_rate": 0.0002,
+      "loss": 1.3755,
+      "step": 6400
+    },
+    {
+      "epoch": 4.294807370184254,
+      "grad_norm": 0.7244732975959778,
+      "learning_rate": 0.0002,
+      "loss": 1.3961,
+      "step": 6410
+    },
+    {
+      "epoch": 4.301507537688442,
+      "grad_norm": 0.7137406468391418,
+      "learning_rate": 0.0002,
+      "loss": 1.4608,
+      "step": 6420
+    },
+    {
+      "epoch": 4.30820770519263,
+      "grad_norm": 0.7492543458938599,
+      "learning_rate": 0.0002,
+      "loss": 1.4461,
+      "step": 6430
+    },
+    {
+      "epoch": 4.314907872696818,
+      "grad_norm": 0.7065439224243164,
+      "learning_rate": 0.0002,
+      "loss": 1.4562,
+      "step": 6440
+    },
+    {
+      "epoch": 4.321608040201005,
+      "grad_norm": 0.7786989808082581,
+      "learning_rate": 0.0002,
+      "loss": 1.4246,
+      "step": 6450
+    },
+    {
+      "epoch": 4.328308207705192,
+      "grad_norm": 0.7369208335876465,
+      "learning_rate": 0.0002,
+      "loss": 1.3098,
+      "step": 6460
+    },
+    {
+      "epoch": 4.33500837520938,
+      "grad_norm": 0.7412346005439758,
+      "learning_rate": 0.0002,
+      "loss": 1.3686,
+      "step": 6470
+    },
+    {
+      "epoch": 4.341708542713568,
+      "grad_norm": 0.780927300453186,
+      "learning_rate": 0.0002,
+      "loss": 1.4087,
+      "step": 6480
+    },
+    {
+      "epoch": 4.348408710217756,
+      "grad_norm": 0.8320930600166321,
+      "learning_rate": 0.0002,
+      "loss": 1.3628,
+      "step": 6490
+    },
+    {
+      "epoch": 4.355108877721943,
+      "grad_norm": 0.6871094703674316,
+      "learning_rate": 0.0002,
+      "loss": 1.3715,
+      "step": 6500
+    },
+    {
+      "epoch": 4.36180904522613,
+      "grad_norm": 0.6751559972763062,
+      "learning_rate": 0.0002,
+      "loss": 1.3257,
+      "step": 6510
+    },
+    {
+      "epoch": 4.368509212730318,
+      "grad_norm": 0.7723976969718933,
+      "learning_rate": 0.0002,
+      "loss": 1.4311,
+      "step": 6520
+    },
+    {
+      "epoch": 4.375209380234506,
+      "grad_norm": 0.7915401458740234,
+      "learning_rate": 0.0002,
+      "loss": 1.4086,
+      "step": 6530
+    },
+    {
+      "epoch": 4.381909547738694,
+      "grad_norm": 0.7329102754592896,
+      "learning_rate": 0.0002,
+      "loss": 1.3973,
+      "step": 6540
+    },
+    {
+      "epoch": 4.388609715242881,
+      "grad_norm": 0.7388760447502136,
+      "learning_rate": 0.0002,
+      "loss": 1.447,
+      "step": 6550
+    },
+    {
+      "epoch": 4.3953098827470685,
+      "grad_norm": 0.8282579183578491,
+      "learning_rate": 0.0002,
+      "loss": 1.4378,
+      "step": 6560
+    },
+    {
+      "epoch": 4.402010050251256,
+      "grad_norm": 0.7192724347114563,
+      "learning_rate": 0.0002,
+      "loss": 1.3923,
+      "step": 6570
+    },
+    {
+      "epoch": 4.408710217755444,
+      "grad_norm": 0.746526837348938,
+      "learning_rate": 0.0002,
+      "loss": 1.4141,
+      "step": 6580
+    },
+    {
+      "epoch": 4.415410385259632,
+      "grad_norm": 0.8738046288490295,
+      "learning_rate": 0.0002,
+      "loss": 1.33,
+      "step": 6590
+    },
+    {
+      "epoch": 4.422110552763819,
+      "grad_norm": 0.8408458828926086,
+      "learning_rate": 0.0002,
+      "loss": 1.3995,
+      "step": 6600
+    },
+    {
+      "epoch": 4.4288107202680065,
+      "grad_norm": 0.8110666275024414,
+      "learning_rate": 0.0002,
+      "loss": 1.4148,
+      "step": 6610
+    },
+    {
+      "epoch": 4.435510887772194,
+      "grad_norm": 0.8602406978607178,
+      "learning_rate": 0.0002,
+      "loss": 1.441,
+      "step": 6620
+    },
+    {
+      "epoch": 4.442211055276382,
+      "grad_norm": 0.7549102902412415,
+      "learning_rate": 0.0002,
+      "loss": 1.4319,
+      "step": 6630
+    },
+    {
+      "epoch": 4.44891122278057,
+      "grad_norm": 0.7831804156303406,
+      "learning_rate": 0.0002,
+      "loss": 1.388,
+      "step": 6640
+    },
+    {
+      "epoch": 4.455611390284757,
+      "grad_norm": 0.7269673943519592,
+      "learning_rate": 0.0002,
+      "loss": 1.4283,
+      "step": 6650
+    },
+    {
+      "epoch": 4.4623115577889445,
+      "grad_norm": 0.7397838830947876,
+      "learning_rate": 0.0002,
+      "loss": 1.4132,
+      "step": 6660
+    },
+    {
+      "epoch": 4.469011725293132,
+      "grad_norm": 0.713707447052002,
+      "learning_rate": 0.0002,
+      "loss": 1.3174,
+      "step": 6670
+    },
+    {
+      "epoch": 4.47571189279732,
+      "grad_norm": 0.7525581121444702,
+      "learning_rate": 0.0002,
+      "loss": 1.3406,
+      "step": 6680
+    },
+    {
+      "epoch": 4.482412060301508,
+      "grad_norm": 0.8030191659927368,
+      "learning_rate": 0.0002,
+      "loss": 1.4283,
+      "step": 6690
+    },
+    {
+      "epoch": 4.489112227805695,
+      "grad_norm": 0.7469439506530762,
+      "learning_rate": 0.0002,
+      "loss": 1.4586,
+      "step": 6700
+    },
+    {
+      "epoch": 4.4958123953098825,
+      "grad_norm": 0.7743868231773376,
+      "learning_rate": 0.0002,
+      "loss": 1.367,
+      "step": 6710
+    },
+    {
+      "epoch": 4.50251256281407,
+      "grad_norm": 0.6539737582206726,
+      "learning_rate": 0.0002,
+      "loss": 1.3439,
+      "step": 6720
+    },
+    {
+      "epoch": 4.509212730318258,
+      "grad_norm": 0.825818657875061,
+      "learning_rate": 0.0002,
+      "loss": 1.4513,
+      "step": 6730
+    },
+    {
+      "epoch": 4.515912897822446,
+      "grad_norm": 0.8048575520515442,
+      "learning_rate": 0.0002,
+      "loss": 1.3984,
+      "step": 6740
+    },
+    {
+      "epoch": 4.522613065326633,
+      "grad_norm": 0.7828766107559204,
+      "learning_rate": 0.0002,
+      "loss": 1.3923,
+      "step": 6750
+    },
+    {
+      "epoch": 4.5293132328308205,
+      "grad_norm": 0.7406010031700134,
+      "learning_rate": 0.0002,
+      "loss": 1.3886,
+      "step": 6760
+    },
+    {
+      "epoch": 4.536013400335008,
+      "grad_norm": 0.840345561504364,
+      "learning_rate": 0.0002,
+      "loss": 1.3109,
+      "step": 6770
+    },
+    {
+      "epoch": 4.542713567839196,
+      "grad_norm": 0.8492622971534729,
+      "learning_rate": 0.0002,
+      "loss": 1.4808,
+      "step": 6780
+    },
+    {
+      "epoch": 4.549413735343384,
+      "grad_norm": 0.7130163908004761,
+      "learning_rate": 0.0002,
+      "loss": 1.4384,
+      "step": 6790
+    },
+    {
+      "epoch": 4.556113902847571,
+      "grad_norm": 0.8454728126525879,
+      "learning_rate": 0.0002,
+      "loss": 1.4531,
+      "step": 6800
+    },
+    {
+      "epoch": 4.562814070351759,
+      "grad_norm": 0.7847645282745361,
+      "learning_rate": 0.0002,
+      "loss": 1.3239,
+      "step": 6810
+    },
+    {
+      "epoch": 4.569514237855946,
+      "grad_norm": 0.7245864272117615,
+      "learning_rate": 0.0002,
+      "loss": 1.4181,
+      "step": 6820
+    },
+    {
+      "epoch": 4.576214405360134,
+      "grad_norm": 0.768893301486969,
+      "learning_rate": 0.0002,
+      "loss": 1.3233,
+      "step": 6830
+    },
+    {
+      "epoch": 4.582914572864322,
+      "grad_norm": 0.8028400540351868,
+      "learning_rate": 0.0002,
+      "loss": 1.3932,
+      "step": 6840
+    },
+    {
+      "epoch": 4.589614740368509,
+      "grad_norm": 0.763945460319519,
+      "learning_rate": 0.0002,
+      "loss": 1.3745,
+      "step": 6850
+    },
+    {
+      "epoch": 4.596314907872697,
+      "grad_norm": 0.7417685389518738,
+      "learning_rate": 0.0002,
+      "loss": 1.4797,
+      "step": 6860
+    },
+    {
+      "epoch": 4.603015075376884,
+      "grad_norm": 0.7603038549423218,
+      "learning_rate": 0.0002,
+      "loss": 1.4468,
+      "step": 6870
+    },
+    {
+      "epoch": 4.609715242881072,
+      "grad_norm": 0.7981528043746948,
+      "learning_rate": 0.0002,
+      "loss": 1.4095,
+      "step": 6880
+    },
+    {
+      "epoch": 4.61641541038526,
+      "grad_norm": 0.8077111840248108,
+      "learning_rate": 0.0002,
+      "loss": 1.3963,
+      "step": 6890
+    },
+    {
+      "epoch": 4.623115577889447,
+      "grad_norm": 0.8778454065322876,
+      "learning_rate": 0.0002,
+      "loss": 1.4721,
+      "step": 6900
+    },
+    {
+      "epoch": 4.629815745393635,
+      "grad_norm": 0.8620710372924805,
+      "learning_rate": 0.0002,
+      "loss": 1.3758,
+      "step": 6910
+    },
+    {
+      "epoch": 4.636515912897822,
+      "grad_norm": 0.7486072778701782,
+      "learning_rate": 0.0002,
+      "loss": 1.344,
+      "step": 6920
+    },
+    {
+      "epoch": 4.64321608040201,
+      "grad_norm": 0.7493042945861816,
+      "learning_rate": 0.0002,
+      "loss": 1.3913,
+      "step": 6930
+    },
+    {
+      "epoch": 4.649916247906198,
+      "grad_norm": 0.7388978600502014,
+      "learning_rate": 0.0002,
+      "loss": 1.397,
+      "step": 6940
+    },
+    {
+      "epoch": 4.656616415410385,
+      "grad_norm": 0.798530638217926,
+      "learning_rate": 0.0002,
+      "loss": 1.3593,
+      "step": 6950
+    },
+    {
+      "epoch": 4.663316582914573,
+      "grad_norm": 0.7929500937461853,
+      "learning_rate": 0.0002,
+      "loss": 1.3982,
+      "step": 6960
+    },
+    {
+      "epoch": 4.67001675041876,
+      "grad_norm": 0.9186785221099854,
+      "learning_rate": 0.0002,
+      "loss": 1.4183,
+      "step": 6970
+    },
+    {
+      "epoch": 4.676716917922948,
+      "grad_norm": 1.1103485822677612,
+      "learning_rate": 0.0002,
+      "loss": 1.3955,
+      "step": 6980
+    },
+    {
+      "epoch": 4.683417085427136,
+      "grad_norm": 0.8000466823577881,
+      "learning_rate": 0.0002,
+      "loss": 1.3941,
+      "step": 6990
+    },
+    {
+      "epoch": 4.690117252931323,
+      "grad_norm": 0.7520599961280823,
+      "learning_rate": 0.0002,
+      "loss": 1.371,
+      "step": 7000
+    },
+    {
+      "epoch": 4.696817420435511,
+      "grad_norm": 0.7971973419189453,
+      "learning_rate": 0.0002,
+      "loss": 1.4582,
+      "step": 7010
+    },
+    {
+      "epoch": 4.703517587939698,
+      "grad_norm": 0.7363343834877014,
+      "learning_rate": 0.0002,
+      "loss": 1.3682,
+      "step": 7020
+    },
+    {
+      "epoch": 4.710217755443886,
+      "grad_norm": 0.8268865942955017,
+      "learning_rate": 0.0002,
+      "loss": 1.3889,
+      "step": 7030
+    },
+    {
+      "epoch": 4.716917922948074,
+      "grad_norm": 0.7054963111877441,
+      "learning_rate": 0.0002,
+      "loss": 1.4382,
+      "step": 7040
+    },
+    {
+      "epoch": 4.723618090452261,
+      "grad_norm": 0.8196262121200562,
+      "learning_rate": 0.0002,
+      "loss": 1.4578,
+      "step": 7050
+    },
+    {
+      "epoch": 4.730318257956449,
+      "grad_norm": 0.8276031017303467,
+      "learning_rate": 0.0002,
+      "loss": 1.365,
+      "step": 7060
+    },
+    {
+      "epoch": 4.7370184254606365,
+      "grad_norm": 0.8248157501220703,
+      "learning_rate": 0.0002,
+      "loss": 1.3887,
+      "step": 7070
+    },
+    {
+      "epoch": 4.743718592964824,
+      "grad_norm": 0.8937979936599731,
+      "learning_rate": 0.0002,
+      "loss": 1.4193,
+      "step": 7080
+    },
+    {
+      "epoch": 4.750418760469012,
+      "grad_norm": 0.7984238862991333,
+      "learning_rate": 0.0002,
+      "loss": 1.4334,
+      "step": 7090
+    },
+    {
+      "epoch": 4.757118927973199,
+      "grad_norm": 0.9495313763618469,
+      "learning_rate": 0.0002,
+      "loss": 1.4385,
+      "step": 7100
+    },
+    {
+      "epoch": 4.763819095477387,
+      "grad_norm": 0.8598204255104065,
+      "learning_rate": 0.0002,
+      "loss": 1.4504,
+      "step": 7110
+    },
+    {
+      "epoch": 4.7705192629815745,
+      "grad_norm": 0.8951472640037537,
+      "learning_rate": 0.0002,
+      "loss": 1.3969,
+      "step": 7120
+    },
+    {
+      "epoch": 4.777219430485762,
+      "grad_norm": 0.9110309481620789,
+      "learning_rate": 0.0002,
+      "loss": 1.4339,
+      "step": 7130
+    },
+    {
+      "epoch": 4.78391959798995,
+      "grad_norm": 0.7929584980010986,
+      "learning_rate": 0.0002,
+      "loss": 1.4001,
+      "step": 7140
+    },
+    {
+      "epoch": 4.790619765494137,
+      "grad_norm": 0.7415322661399841,
+      "learning_rate": 0.0002,
+      "loss": 1.467,
+      "step": 7150
+    },
+    {
+      "epoch": 4.797319932998325,
+      "grad_norm": 0.7504757046699524,
+      "learning_rate": 0.0002,
+      "loss": 1.5107,
+      "step": 7160
+    },
+    {
+      "epoch": 4.8040201005025125,
+      "grad_norm": 0.7166924476623535,
+      "learning_rate": 0.0002,
+      "loss": 1.3736,
+      "step": 7170
+    },
+    {
+      "epoch": 4.8107202680067,
+      "grad_norm": 0.7728400826454163,
+      "learning_rate": 0.0002,
+      "loss": 1.4088,
+      "step": 7180
+    },
+    {
+      "epoch": 4.817420435510888,
+      "grad_norm": 0.7992154955863953,
+      "learning_rate": 0.0002,
+      "loss": 1.3814,
+      "step": 7190
+    },
+    {
+      "epoch": 4.824120603015075,
+      "grad_norm": 0.8655321002006531,
+      "learning_rate": 0.0002,
+      "loss": 1.3958,
+      "step": 7200
+    },
+    {
+      "epoch": 4.830820770519263,
+      "grad_norm": 0.7672632336616516,
+      "learning_rate": 0.0002,
+      "loss": 1.3837,
+      "step": 7210
+    },
+    {
+      "epoch": 4.8375209380234505,
+      "grad_norm": 0.708416223526001,
+      "learning_rate": 0.0002,
+      "loss": 1.4578,
+      "step": 7220
+    },
+    {
+      "epoch": 4.844221105527638,
+      "grad_norm": 0.8914081454277039,
+      "learning_rate": 0.0002,
+      "loss": 1.5413,
+      "step": 7230
+    },
+    {
+      "epoch": 4.850921273031826,
+      "grad_norm": 0.7141931653022766,
+      "learning_rate": 0.0002,
+      "loss": 1.3569,
+      "step": 7240
+    },
+    {
+      "epoch": 4.857621440536013,
+      "grad_norm": 0.6913040280342102,
+      "learning_rate": 0.0002,
+      "loss": 1.4532,
+      "step": 7250
+    },
+    {
+      "epoch": 4.864321608040201,
+      "grad_norm": 0.7871233820915222,
+      "learning_rate": 0.0002,
+      "loss": 1.3912,
+      "step": 7260
+    },
+    {
+      "epoch": 4.8710217755443885,
+      "grad_norm": 0.8466277122497559,
+      "learning_rate": 0.0002,
+      "loss": 1.3688,
+      "step": 7270
+    },
+    {
+      "epoch": 4.877721943048576,
+      "grad_norm": 0.8492183685302734,
+      "learning_rate": 0.0002,
+      "loss": 1.33,
+      "step": 7280
+    },
+    {
+      "epoch": 4.884422110552764,
+      "grad_norm": 0.8339574933052063,
+      "learning_rate": 0.0002,
+      "loss": 1.3744,
+      "step": 7290
+    },
+    {
+      "epoch": 4.891122278056951,
+      "grad_norm": 0.787022590637207,
+      "learning_rate": 0.0002,
+      "loss": 1.4157,
+      "step": 7300
+    },
+    {
+      "epoch": 4.897822445561139,
+      "grad_norm": 0.8877332806587219,
+      "learning_rate": 0.0002,
+      "loss": 1.3725,
+      "step": 7310
+    },
+    {
+      "epoch": 4.9045226130653266,
+      "grad_norm": 0.744989812374115,
+      "learning_rate": 0.0002,
+      "loss": 1.3968,
+      "step": 7320
+    },
+    {
+      "epoch": 4.911222780569514,
+      "grad_norm": 0.8027268648147583,
+      "learning_rate": 0.0002,
+      "loss": 1.4421,
+      "step": 7330
+    },
+    {
+      "epoch": 4.917922948073702,
+      "grad_norm": 0.6437455415725708,
+      "learning_rate": 0.0002,
+      "loss": 1.425,
+      "step": 7340
+    },
+    {
+      "epoch": 4.924623115577889,
+      "grad_norm": 0.685999870300293,
+      "learning_rate": 0.0002,
+      "loss": 1.4829,
+      "step": 7350
+    },
+    {
+      "epoch": 4.931323283082077,
+      "grad_norm": 0.9086187481880188,
+      "learning_rate": 0.0002,
+      "loss": 1.4352,
+      "step": 7360
+    },
+    {
+      "epoch": 4.938023450586265,
+      "grad_norm": 0.8272411227226257,
+      "learning_rate": 0.0002,
+      "loss": 1.4245,
+      "step": 7370
+    },
+    {
+      "epoch": 4.944723618090452,
+      "grad_norm": 0.9227852821350098,
+      "learning_rate": 0.0002,
+      "loss": 1.4226,
+      "step": 7380
+    },
+    {
+      "epoch": 4.95142378559464,
+      "grad_norm": 0.7688441276550293,
+      "learning_rate": 0.0002,
+      "loss": 1.3643,
+      "step": 7390
+    },
+    {
+      "epoch": 4.958123953098827,
+      "grad_norm": 0.8662643432617188,
+      "learning_rate": 0.0002,
+      "loss": 1.4491,
+      "step": 7400
+    },
+    {
+      "epoch": 4.964824120603015,
+      "grad_norm": 0.9234127998352051,
+      "learning_rate": 0.0002,
+      "loss": 1.4194,
+      "step": 7410
+    },
+    {
+      "epoch": 4.971524288107203,
+      "grad_norm": 0.9131470918655396,
+      "learning_rate": 0.0002,
+      "loss": 1.4009,
+      "step": 7420
+    },
+    {
+      "epoch": 4.97822445561139,
+      "grad_norm": 0.7377504110336304,
+      "learning_rate": 0.0002,
+      "loss": 1.4544,
+      "step": 7430
+    },
+    {
+      "epoch": 4.984924623115578,
+      "grad_norm": 0.8762801289558411,
+      "learning_rate": 0.0002,
+      "loss": 1.4008,
+      "step": 7440
+    },
+    {
+      "epoch": 4.991624790619765,
+      "grad_norm": 0.7919872999191284,
+      "learning_rate": 0.0002,
+      "loss": 1.4304,
+      "step": 7450
+    },
+    {
+      "epoch": 4.998324958123953,
+      "grad_norm": 0.7144299149513245,
+      "learning_rate": 0.0002,
+      "loss": 1.3817,
+      "step": 7460
+    },
+    {
+      "epoch": 4.99966499162479,
+      "eval_loss": 1.9291157722473145,
+      "eval_runtime": 37.9831,
+      "eval_samples_per_second": 13.559,
+      "eval_steps_per_second": 1.711,
+      "step": 7462
+    },
+    {
+      "epoch": 5.005025125628141,
+      "grad_norm": 0.7860151529312134,
+      "learning_rate": 0.0002,
+      "loss": 1.2753,
+      "step": 7470
+    },
+    {
+      "epoch": 5.011725293132328,
+      "grad_norm": 0.9418314695358276,
+      "learning_rate": 0.0002,
+      "loss": 1.2149,
+      "step": 7480
+    },
+    {
+      "epoch": 5.018425460636516,
+      "grad_norm": 0.8474572896957397,
+      "learning_rate": 0.0002,
+      "loss": 1.1966,
+      "step": 7490
+    },
+    {
+      "epoch": 5.025125628140704,
+      "grad_norm": 1.0724040269851685,
+      "learning_rate": 0.0002,
+      "loss": 1.2111,
+      "step": 7500
+    },
+    {
+      "epoch": 5.031825795644891,
+      "grad_norm": 0.9109148979187012,
+      "learning_rate": 0.0002,
+      "loss": 1.2228,
+      "step": 7510
+    },
+    {
+      "epoch": 5.038525963149079,
+      "grad_norm": 1.0088659524917603,
+      "learning_rate": 0.0002,
+      "loss": 1.2239,
+      "step": 7520
+    },
+    {
+      "epoch": 5.045226130653266,
+      "grad_norm": 1.1421623229980469,
+      "learning_rate": 0.0002,
+      "loss": 1.2156,
+      "step": 7530
+    },
+    {
+      "epoch": 5.051926298157454,
+      "grad_norm": 0.9219902157783508,
+      "learning_rate": 0.0002,
+      "loss": 1.1739,
+      "step": 7540
+    },
+    {
+      "epoch": 5.058626465661642,
+      "grad_norm": 0.9150987863540649,
+      "learning_rate": 0.0002,
+      "loss": 1.2686,
+      "step": 7550
+    },
+    {
+      "epoch": 5.065326633165829,
+      "grad_norm": 0.8889328241348267,
+      "learning_rate": 0.0002,
+      "loss": 1.2068,
+      "step": 7560
+    },
+    {
+      "epoch": 5.072026800670017,
+      "grad_norm": 0.9751363396644592,
+      "learning_rate": 0.0002,
+      "loss": 1.276,
+      "step": 7570
+    },
+    {
+      "epoch": 5.078726968174204,
+      "grad_norm": 0.8603123426437378,
+      "learning_rate": 0.0002,
+      "loss": 1.2078,
+      "step": 7580
+    },
+    {
+      "epoch": 5.085427135678392,
+      "grad_norm": 0.8910616636276245,
+      "learning_rate": 0.0002,
+      "loss": 1.2175,
+      "step": 7590
+    },
+    {
+      "epoch": 5.09212730318258,
+      "grad_norm": 1.1128392219543457,
+      "learning_rate": 0.0002,
+      "loss": 1.2475,
+      "step": 7600
+    },
+    {
+      "epoch": 5.098827470686767,
+      "grad_norm": 0.9480258822441101,
+      "learning_rate": 0.0002,
+      "loss": 1.3065,
+      "step": 7610
+    },
+    {
+      "epoch": 5.105527638190955,
+      "grad_norm": 0.906958818435669,
+      "learning_rate": 0.0002,
+      "loss": 1.193,
+      "step": 7620
+    },
+    {
+      "epoch": 5.1122278056951425,
+      "grad_norm": 0.8741167187690735,
+      "learning_rate": 0.0002,
+      "loss": 1.2223,
+      "step": 7630
+    },
+    {
+      "epoch": 5.11892797319933,
+      "grad_norm": 0.966268002986908,
+      "learning_rate": 0.0002,
+      "loss": 1.2126,
+      "step": 7640
+    },
+    {
+      "epoch": 5.125628140703517,
+      "grad_norm": 0.9124358892440796,
+      "learning_rate": 0.0002,
+      "loss": 1.2782,
+      "step": 7650
+    },
+    {
+      "epoch": 5.132328308207705,
+      "grad_norm": 1.0436606407165527,
+      "learning_rate": 0.0002,
+      "loss": 1.3004,
+      "step": 7660
+    },
+    {
+      "epoch": 5.139028475711893,
+      "grad_norm": 0.9217309355735779,
+      "learning_rate": 0.0002,
+      "loss": 1.2675,
+      "step": 7670
+    },
+    {
+      "epoch": 5.1457286432160805,
+      "grad_norm": 1.344765543937683,
+      "learning_rate": 0.0002,
+      "loss": 1.2502,
+      "step": 7680
+    },
+    {
+      "epoch": 5.152428810720268,
+      "grad_norm": 1.0730723142623901,
+      "learning_rate": 0.0002,
+      "loss": 1.2416,
+      "step": 7690
+    },
+    {
+      "epoch": 5.159128978224456,
+      "grad_norm": 0.9321247339248657,
+      "learning_rate": 0.0002,
+      "loss": 1.1888,
+      "step": 7700
+    },
+    {
+      "epoch": 5.165829145728643,
+      "grad_norm": 0.8482614755630493,
+      "learning_rate": 0.0002,
+      "loss": 1.1941,
+      "step": 7710
+    },
+    {
+      "epoch": 5.172529313232831,
+      "grad_norm": 0.8274452686309814,
+      "learning_rate": 0.0002,
+      "loss": 1.2668,
+      "step": 7720
+    },
+    {
+      "epoch": 5.1792294807370185,
+      "grad_norm": 0.9120376706123352,
+      "learning_rate": 0.0002,
+      "loss": 1.1972,
+      "step": 7730
+    },
+    {
+      "epoch": 5.185929648241206,
+      "grad_norm": 1.0062892436981201,
+      "learning_rate": 0.0002,
+      "loss": 1.1648,
+      "step": 7740
+    },
+    {
+      "epoch": 5.192629815745394,
+      "grad_norm": 0.9521504640579224,
+      "learning_rate": 0.0002,
+      "loss": 1.2199,
+      "step": 7750
+    },
+    {
+      "epoch": 5.199329983249581,
+      "grad_norm": 0.8800198435783386,
+      "learning_rate": 0.0002,
+      "loss": 1.2855,
+      "step": 7760
+    },
+    {
+      "epoch": 5.206030150753769,
+      "grad_norm": 0.9749179482460022,
+      "learning_rate": 0.0002,
+      "loss": 1.2535,
+      "step": 7770
+    },
+    {
+      "epoch": 5.2127303182579565,
+      "grad_norm": 0.9441686868667603,
+      "learning_rate": 0.0002,
+      "loss": 1.2975,
+      "step": 7780
+    },
+    {
+      "epoch": 5.219430485762144,
+      "grad_norm": 0.9114066362380981,
+      "learning_rate": 0.0002,
+      "loss": 1.256,
+      "step": 7790
+    },
+    {
+      "epoch": 5.226130653266332,
+      "grad_norm": 0.9851446151733398,
+      "learning_rate": 0.0002,
+      "loss": 1.2621,
+      "step": 7800
+    },
+    {
+      "epoch": 5.232830820770519,
+      "grad_norm": 0.9526297450065613,
+      "learning_rate": 0.0002,
+      "loss": 1.2502,
+      "step": 7810
+    },
+    {
+      "epoch": 5.239530988274707,
+      "grad_norm": 1.05986487865448,
+      "learning_rate": 0.0002,
+      "loss": 1.1502,
+      "step": 7820
+    },
+    {
+      "epoch": 5.2462311557788945,
+      "grad_norm": 0.8956538438796997,
+      "learning_rate": 0.0002,
+      "loss": 1.2517,
+      "step": 7830
+    },
+    {
+      "epoch": 5.252931323283082,
+      "grad_norm": 0.9568153619766235,
+      "learning_rate": 0.0002,
+      "loss": 1.2556,
+      "step": 7840
+    },
+    {
+      "epoch": 5.259631490787269,
+      "grad_norm": 1.0035018920898438,
+      "learning_rate": 0.0002,
+      "loss": 1.2442,
+      "step": 7850
+    },
+    {
+      "epoch": 5.266331658291457,
+      "grad_norm": 0.8554368615150452,
+      "learning_rate": 0.0002,
+      "loss": 1.2605,
+      "step": 7860
+    },
+    {
+      "epoch": 5.273031825795645,
+      "grad_norm": 0.9677708148956299,
+      "learning_rate": 0.0002,
+      "loss": 1.2799,
+      "step": 7870
+    },
+    {
+      "epoch": 5.279731993299833,
+      "grad_norm": 0.943606436252594,
+      "learning_rate": 0.0002,
+      "loss": 1.275,
+      "step": 7880
+    },
+    {
+      "epoch": 5.28643216080402,
+      "grad_norm": 1.0029335021972656,
+      "learning_rate": 0.0002,
+      "loss": 1.2335,
+      "step": 7890
+    },
+    {
+      "epoch": 5.293132328308207,
+      "grad_norm": 1.0164015293121338,
+      "learning_rate": 0.0002,
+      "loss": 1.2494,
+      "step": 7900
+    },
+    {
+      "epoch": 5.299832495812395,
+      "grad_norm": 0.8908365368843079,
+      "learning_rate": 0.0002,
+      "loss": 1.3117,
+      "step": 7910
+    },
+    {
+      "epoch": 5.306532663316583,
+      "grad_norm": 0.9307826161384583,
+      "learning_rate": 0.0002,
+      "loss": 1.2832,
+      "step": 7920
+    },
+    {
+      "epoch": 5.313232830820771,
+      "grad_norm": 1.0730371475219727,
+      "learning_rate": 0.0002,
+      "loss": 1.242,
+      "step": 7930
+    },
+    {
+      "epoch": 5.319932998324958,
+      "grad_norm": 0.844739556312561,
+      "learning_rate": 0.0002,
+      "loss": 1.2003,
+      "step": 7940
+    },
+    {
+      "epoch": 5.326633165829146,
+      "grad_norm": 1.275833010673523,
+      "learning_rate": 0.0002,
+      "loss": 1.2688,
+      "step": 7950
+    },
+    {
+      "epoch": 5.333333333333333,
+      "grad_norm": 0.9042661190032959,
+      "learning_rate": 0.0002,
+      "loss": 1.2957,
+      "step": 7960
+    },
+    {
+      "epoch": 5.340033500837521,
+      "grad_norm": 0.9374269247055054,
+      "learning_rate": 0.0002,
+      "loss": 1.2912,
+      "step": 7970
+    },
+    {
+      "epoch": 5.346733668341709,
+      "grad_norm": 1.033098578453064,
+      "learning_rate": 0.0002,
+      "loss": 1.2721,
+      "step": 7980
+    },
+    {
+      "epoch": 5.353433835845896,
+      "grad_norm": 1.062775731086731,
+      "learning_rate": 0.0002,
+      "loss": 1.3208,
+      "step": 7990
+    },
+    {
+      "epoch": 5.360134003350084,
+      "grad_norm": 1.1064317226409912,
+      "learning_rate": 0.0002,
+      "loss": 1.3065,
+      "step": 8000
+    },
+    {
+      "epoch": 5.366834170854271,
+      "grad_norm": 1.1114039421081543,
+      "learning_rate": 0.0002,
+      "loss": 1.2341,
+      "step": 8010
+    },
+    {
+      "epoch": 5.373534338358459,
+      "grad_norm": 1.0198014974594116,
+      "learning_rate": 0.0002,
+      "loss": 1.2255,
+      "step": 8020
+    },
+    {
+      "epoch": 5.380234505862647,
+      "grad_norm": 0.8443173170089722,
+      "learning_rate": 0.0002,
+      "loss": 1.2433,
+      "step": 8030
+    },
+    {
+      "epoch": 5.386934673366834,
+      "grad_norm": 1.000881314277649,
+      "learning_rate": 0.0002,
+      "loss": 1.206,
+      "step": 8040
+    },
+    {
+      "epoch": 5.393634840871022,
+      "grad_norm": 0.9874443411827087,
+      "learning_rate": 0.0002,
+      "loss": 1.2982,
+      "step": 8050
+    },
+    {
+      "epoch": 5.400335008375209,
+      "grad_norm": 0.9895344972610474,
+      "learning_rate": 0.0002,
+      "loss": 1.2289,
+      "step": 8060
+    },
+    {
+      "epoch": 5.407035175879397,
+      "grad_norm": 0.8595236539840698,
+      "learning_rate": 0.0002,
+      "loss": 1.249,
+      "step": 8070
+    },
+    {
+      "epoch": 5.413735343383585,
+      "grad_norm": 0.9523849487304688,
+      "learning_rate": 0.0002,
+      "loss": 1.2308,
+      "step": 8080
+    },
+    {
+      "epoch": 5.420435510887772,
+      "grad_norm": 1.0560476779937744,
+      "learning_rate": 0.0002,
+      "loss": 1.2343,
+      "step": 8090
+    },
+    {
+      "epoch": 5.42713567839196,
+      "grad_norm": 1.0893689393997192,
+      "learning_rate": 0.0002,
+      "loss": 1.2956,
+      "step": 8100
+    },
+    {
+      "epoch": 5.433835845896147,
+      "grad_norm": 0.9395513534545898,
+      "learning_rate": 0.0002,
+      "loss": 1.2846,
+      "step": 8110
+    },
+    {
+      "epoch": 5.440536013400335,
+      "grad_norm": 0.9364215135574341,
+      "learning_rate": 0.0002,
+      "loss": 1.3444,
+      "step": 8120
+    },
+    {
+      "epoch": 5.447236180904523,
+      "grad_norm": 0.9502208232879639,
+      "learning_rate": 0.0002,
+      "loss": 1.2944,
+      "step": 8130
+    },
+    {
+      "epoch": 5.45393634840871,
+      "grad_norm": 0.9559208154678345,
+      "learning_rate": 0.0002,
+      "loss": 1.2971,
+      "step": 8140
+    },
+    {
+      "epoch": 5.460636515912898,
+      "grad_norm": 0.9261730313301086,
+      "learning_rate": 0.0002,
+      "loss": 1.2495,
+      "step": 8150
+    },
+    {
+      "epoch": 5.467336683417085,
+      "grad_norm": 0.9832326173782349,
+      "learning_rate": 0.0002,
+      "loss": 1.2599,
+      "step": 8160
+    },
+    {
+      "epoch": 5.474036850921273,
+      "grad_norm": 1.065953016281128,
+      "learning_rate": 0.0002,
+      "loss": 1.2771,
+      "step": 8170
+    },
+    {
+      "epoch": 5.480737018425461,
+      "grad_norm": 0.9139469861984253,
+      "learning_rate": 0.0002,
+      "loss": 1.3617,
+      "step": 8180
+    },
+    {
+      "epoch": 5.4874371859296485,
+      "grad_norm": 1.2322484254837036,
+      "learning_rate": 0.0002,
+      "loss": 1.2197,
+      "step": 8190
+    },
+    {
+      "epoch": 5.494137353433836,
+      "grad_norm": 0.9722974896430969,
+      "learning_rate": 0.0002,
+      "loss": 1.2879,
+      "step": 8200
+    },
+    {
+      "epoch": 5.500837520938023,
+      "grad_norm": 0.9338926076889038,
+      "learning_rate": 0.0002,
+      "loss": 1.2664,
+      "step": 8210
+    },
+    {
+      "epoch": 5.507537688442211,
+      "grad_norm": 0.9283728003501892,
+      "learning_rate": 0.0002,
+      "loss": 1.2128,
+      "step": 8220
+    },
+    {
+      "epoch": 5.514237855946399,
+      "grad_norm": 1.0489585399627686,
+      "learning_rate": 0.0002,
+      "loss": 1.2141,
+      "step": 8230
+    },
+    {
+      "epoch": 5.5209380234505865,
+      "grad_norm": 0.9881814122200012,
+      "learning_rate": 0.0002,
+      "loss": 1.3257,
+      "step": 8240
+    },
+    {
+      "epoch": 5.527638190954773,
+      "grad_norm": 0.9274460077285767,
+      "learning_rate": 0.0002,
+      "loss": 1.2221,
+      "step": 8250
+    },
+    {
+      "epoch": 5.534338358458961,
+      "grad_norm": 0.8650718331336975,
+      "learning_rate": 0.0002,
+      "loss": 1.2241,
+      "step": 8260
+    },
+    {
+      "epoch": 5.541038525963149,
+      "grad_norm": 1.014069676399231,
+      "learning_rate": 0.0002,
+      "loss": 1.2462,
+      "step": 8270
+    },
+    {
+      "epoch": 5.547738693467337,
+      "grad_norm": 0.9212974905967712,
+      "learning_rate": 0.0002,
+      "loss": 1.3502,
+      "step": 8280
+    },
+    {
+      "epoch": 5.5544388609715245,
+      "grad_norm": 1.1235398054122925,
+      "learning_rate": 0.0002,
+      "loss": 1.2779,
+      "step": 8290
+    },
+    {
+      "epoch": 5.561139028475711,
+      "grad_norm": 0.961954653263092,
+      "learning_rate": 0.0002,
+      "loss": 1.306,
+      "step": 8300
+    },
+    {
+      "epoch": 5.567839195979899,
+      "grad_norm": 0.9386700391769409,
+      "learning_rate": 0.0002,
+      "loss": 1.2946,
+      "step": 8310
+    },
+    {
+      "epoch": 5.574539363484087,
+      "grad_norm": 1.01912522315979,
+      "learning_rate": 0.0002,
+      "loss": 1.313,
+      "step": 8320
+    },
+    {
+      "epoch": 5.581239530988275,
+      "grad_norm": 0.9851216077804565,
+      "learning_rate": 0.0002,
+      "loss": 1.3121,
+      "step": 8330
+    },
+    {
+      "epoch": 5.5879396984924625,
+      "grad_norm": 1.0138001441955566,
+      "learning_rate": 0.0002,
+      "loss": 1.3071,
+      "step": 8340
+    },
+    {
+      "epoch": 5.594639865996649,
+      "grad_norm": 0.9262447357177734,
+      "learning_rate": 0.0002,
+      "loss": 1.2257,
+      "step": 8350
+    },
+    {
+      "epoch": 5.601340033500837,
+      "grad_norm": 1.1322970390319824,
+      "learning_rate": 0.0002,
+      "loss": 1.2473,
+      "step": 8360
+    },
+    {
+      "epoch": 5.608040201005025,
+      "grad_norm": 1.1429349184036255,
+      "learning_rate": 0.0002,
+      "loss": 1.3098,
+      "step": 8370
+    },
+    {
+      "epoch": 5.614740368509213,
+      "grad_norm": 0.9130118489265442,
+      "learning_rate": 0.0002,
+      "loss": 1.2686,
+      "step": 8380
+    },
+    {
+      "epoch": 5.6214405360134005,
+      "grad_norm": 0.9651545882225037,
+      "learning_rate": 0.0002,
+      "loss": 1.2541,
+      "step": 8390
+    },
+    {
+      "epoch": 5.628140703517588,
+      "grad_norm": 0.9595398306846619,
+      "learning_rate": 0.0002,
+      "loss": 1.2799,
+      "step": 8400
+    },
+    {
+      "epoch": 5.634840871021775,
+      "grad_norm": 1.0049372911453247,
+      "learning_rate": 0.0002,
+      "loss": 1.3429,
+      "step": 8410
+    },
+    {
+      "epoch": 5.641541038525963,
+      "grad_norm": 1.082804560661316,
+      "learning_rate": 0.0002,
+      "loss": 1.3224,
+      "step": 8420
+    },
+    {
+      "epoch": 5.648241206030151,
+      "grad_norm": 0.9489204287528992,
+      "learning_rate": 0.0002,
+      "loss": 1.297,
+      "step": 8430
+    },
+    {
+      "epoch": 5.654941373534339,
+      "grad_norm": 0.9470235109329224,
+      "learning_rate": 0.0002,
+      "loss": 1.3424,
+      "step": 8440
+    },
+    {
+      "epoch": 5.661641541038526,
+      "grad_norm": 1.0662927627563477,
+      "learning_rate": 0.0002,
+      "loss": 1.3358,
+      "step": 8450
+    },
+    {
+      "epoch": 5.668341708542713,
+      "grad_norm": 0.9097877740859985,
+      "learning_rate": 0.0002,
+      "loss": 1.2973,
+      "step": 8460
+    },
+    {
+      "epoch": 5.675041876046901,
+      "grad_norm": 0.9740368127822876,
+      "learning_rate": 0.0002,
+      "loss": 1.3072,
+      "step": 8470
+    },
+    {
+      "epoch": 5.681742043551089,
+      "grad_norm": 0.9878810048103333,
+      "learning_rate": 0.0002,
+      "loss": 1.286,
+      "step": 8480
+    },
+    {
+      "epoch": 5.688442211055277,
+      "grad_norm": 1.148260474205017,
+      "learning_rate": 0.0002,
+      "loss": 1.208,
+      "step": 8490
+    },
+    {
+      "epoch": 5.695142378559464,
+      "grad_norm": 0.9632558822631836,
+      "learning_rate": 0.0002,
+      "loss": 1.2842,
+      "step": 8500
+    },
+    {
+      "epoch": 5.701842546063651,
+      "grad_norm": 0.876812756061554,
+      "learning_rate": 0.0002,
+      "loss": 1.2787,
+      "step": 8510
+    },
+    {
+      "epoch": 5.708542713567839,
+      "grad_norm": 1.0730829238891602,
+      "learning_rate": 0.0002,
+      "loss": 1.3186,
+      "step": 8520
+    },
+    {
+      "epoch": 5.715242881072027,
+      "grad_norm": 1.2239218950271606,
+      "learning_rate": 0.0002,
+      "loss": 1.2856,
+      "step": 8530
+    },
+    {
+      "epoch": 5.721943048576215,
+      "grad_norm": 0.9460835456848145,
+      "learning_rate": 0.0002,
+      "loss": 1.2717,
+      "step": 8540
+    },
+    {
+      "epoch": 5.728643216080402,
+      "grad_norm": 0.9086270928382874,
+      "learning_rate": 0.0002,
+      "loss": 1.3509,
+      "step": 8550
+    },
+    {
+      "epoch": 5.735343383584589,
+      "grad_norm": 1.0258867740631104,
+      "learning_rate": 0.0002,
+      "loss": 1.2971,
+      "step": 8560
+    },
+    {
+      "epoch": 5.742043551088777,
+      "grad_norm": 1.0543923377990723,
+      "learning_rate": 0.0002,
+      "loss": 1.3581,
+      "step": 8570
+    },
+    {
+      "epoch": 5.748743718592965,
+      "grad_norm": 0.9063900113105774,
+      "learning_rate": 0.0002,
+      "loss": 1.2988,
+      "step": 8580
+    },
+    {
+      "epoch": 5.755443886097153,
+      "grad_norm": 1.1838830709457397,
+      "learning_rate": 0.0002,
+      "loss": 1.3535,
+      "step": 8590
+    },
+    {
+      "epoch": 5.76214405360134,
+      "grad_norm": 0.9631859064102173,
+      "learning_rate": 0.0002,
+      "loss": 1.2655,
+      "step": 8600
+    },
+    {
+      "epoch": 5.768844221105527,
+      "grad_norm": 0.9702655673027039,
+      "learning_rate": 0.0002,
+      "loss": 1.276,
+      "step": 8610
+    },
+    {
+      "epoch": 5.775544388609715,
+      "grad_norm": 1.0591435432434082,
+      "learning_rate": 0.0002,
+      "loss": 1.3196,
+      "step": 8620
+    },
+    {
+      "epoch": 5.782244556113903,
+      "grad_norm": 0.9989570379257202,
+      "learning_rate": 0.0002,
+      "loss": 1.267,
+      "step": 8630
+    },
+    {
+      "epoch": 5.788944723618091,
+      "grad_norm": 1.0836435556411743,
+      "learning_rate": 0.0002,
+      "loss": 1.3227,
+      "step": 8640
+    },
+    {
+      "epoch": 5.795644891122278,
+      "grad_norm": 0.8832896947860718,
+      "learning_rate": 0.0002,
+      "loss": 1.3334,
+      "step": 8650
+    },
+    {
+      "epoch": 5.802345058626465,
+      "grad_norm": 1.0104607343673706,
+      "learning_rate": 0.0002,
+      "loss": 1.3214,
+      "step": 8660
+    },
+    {
+      "epoch": 5.809045226130653,
+      "grad_norm": 0.8375084400177002,
+      "learning_rate": 0.0002,
+      "loss": 1.2703,
+      "step": 8670
+    },
+    {
+      "epoch": 5.815745393634841,
+      "grad_norm": 1.1300716400146484,
+      "learning_rate": 0.0002,
+      "loss": 1.3554,
+      "step": 8680
+    },
+    {
+      "epoch": 5.822445561139029,
+      "grad_norm": 0.9311910271644592,
+      "learning_rate": 0.0002,
+      "loss": 1.3468,
+      "step": 8690
+    },
+    {
+      "epoch": 5.8291457286432165,
+      "grad_norm": 0.9488391876220703,
+      "learning_rate": 0.0002,
+      "loss": 1.2749,
+      "step": 8700
+    },
+    {
+      "epoch": 5.835845896147403,
+      "grad_norm": 0.9747629761695862,
+      "learning_rate": 0.0002,
+      "loss": 1.2281,
+      "step": 8710
+    },
+    {
+      "epoch": 5.842546063651591,
+      "grad_norm": 1.1029598712921143,
+      "learning_rate": 0.0002,
+      "loss": 1.2923,
+      "step": 8720
+    },
+    {
+      "epoch": 5.849246231155779,
+      "grad_norm": 1.0396875143051147,
+      "learning_rate": 0.0002,
+      "loss": 1.3613,
+      "step": 8730
+    },
+    {
+      "epoch": 5.855946398659967,
+      "grad_norm": 0.9259780645370483,
+      "learning_rate": 0.0002,
+      "loss": 1.3272,
+      "step": 8740
+    },
+    {
+      "epoch": 5.8626465661641545,
+      "grad_norm": 1.020033597946167,
+      "learning_rate": 0.0002,
+      "loss": 1.3236,
+      "step": 8750
+    },
+    {
+      "epoch": 5.869346733668341,
+      "grad_norm": 0.9191218614578247,
+      "learning_rate": 0.0002,
+      "loss": 1.3453,
+      "step": 8760
+    },
+    {
+      "epoch": 5.876046901172529,
+      "grad_norm": 1.1093107461929321,
+      "learning_rate": 0.0002,
+      "loss": 1.3012,
+      "step": 8770
+    },
+    {
+      "epoch": 5.882747068676717,
+      "grad_norm": 1.1626793146133423,
+      "learning_rate": 0.0002,
+      "loss": 1.2718,
+      "step": 8780
+    },
+    {
+      "epoch": 5.889447236180905,
+      "grad_norm": 0.9542945027351379,
+      "learning_rate": 0.0002,
+      "loss": 1.2969,
+      "step": 8790
+    },
+    {
+      "epoch": 5.8961474036850925,
+      "grad_norm": 0.9086058139801025,
+      "learning_rate": 0.0002,
+      "loss": 1.3134,
+      "step": 8800
+    },
+    {
+      "epoch": 5.902847571189279,
+      "grad_norm": 0.9249639511108398,
+      "learning_rate": 0.0002,
+      "loss": 1.2731,
+      "step": 8810
+    },
+    {
+      "epoch": 5.909547738693467,
+      "grad_norm": 0.9414396286010742,
+      "learning_rate": 0.0002,
+      "loss": 1.337,
+      "step": 8820
+    },
+    {
+      "epoch": 5.916247906197655,
+      "grad_norm": 0.9086037874221802,
+      "learning_rate": 0.0002,
+      "loss": 1.2865,
+      "step": 8830
+    },
+    {
+      "epoch": 5.922948073701843,
+      "grad_norm": 0.8685907125473022,
+      "learning_rate": 0.0002,
+      "loss": 1.2756,
+      "step": 8840
+    },
+    {
+      "epoch": 5.9296482412060305,
+      "grad_norm": 1.036419153213501,
+      "learning_rate": 0.0002,
+      "loss": 1.297,
+      "step": 8850
+    },
+    {
+      "epoch": 5.936348408710217,
+      "grad_norm": 1.0183674097061157,
+      "learning_rate": 0.0002,
+      "loss": 1.3207,
+      "step": 8860
+    },
+    {
+      "epoch": 5.943048576214405,
+      "grad_norm": 0.966444194316864,
+      "learning_rate": 0.0002,
+      "loss": 1.3922,
+      "step": 8870
+    },
+    {
+      "epoch": 5.949748743718593,
+      "grad_norm": 1.125693917274475,
+      "learning_rate": 0.0002,
+      "loss": 1.333,
+      "step": 8880
+    },
+    {
+      "epoch": 5.956448911222781,
+      "grad_norm": 0.9857436418533325,
+      "learning_rate": 0.0002,
+      "loss": 1.3116,
+      "step": 8890
+    },
+    {
+      "epoch": 5.9631490787269685,
+      "grad_norm": 0.9377069473266602,
+      "learning_rate": 0.0002,
+      "loss": 1.2526,
+      "step": 8900
+    },
+    {
+      "epoch": 5.969849246231155,
+      "grad_norm": 0.9493814706802368,
+      "learning_rate": 0.0002,
+      "loss": 1.3221,
+      "step": 8910
+    },
+    {
+      "epoch": 5.976549413735343,
+      "grad_norm": 0.8806208372116089,
+      "learning_rate": 0.0002,
+      "loss": 1.2516,
+      "step": 8920
+    },
+    {
+      "epoch": 5.983249581239531,
+      "grad_norm": 0.8727600574493408,
+      "learning_rate": 0.0002,
+      "loss": 1.2558,
+      "step": 8930
+    },
+    {
+      "epoch": 5.989949748743719,
+      "grad_norm": 0.9799810647964478,
+      "learning_rate": 0.0002,
+      "loss": 1.3538,
+      "step": 8940
+    },
+    {
+      "epoch": 5.9966499162479066,
+      "grad_norm": 0.9866513609886169,
+      "learning_rate": 0.0002,
+      "loss": 1.3323,
+      "step": 8950
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 2.0282373428344727,
+      "eval_runtime": 38.0375,
+      "eval_samples_per_second": 13.539,
+      "eval_steps_per_second": 1.709,
+      "step": 8955
+    },
+    {
+      "epoch": 6.0033500837520934,
+      "grad_norm": 0.8747885227203369,
+      "learning_rate": 0.0002,
+      "loss": 1.1768,
+      "step": 8960
+    },
+    {
+      "epoch": 6.010050251256281,
+      "grad_norm": 1.2512741088867188,
+      "learning_rate": 0.0002,
+      "loss": 1.0677,
+      "step": 8970
+    },
+    {
+      "epoch": 6.016750418760469,
+      "grad_norm": 1.06855309009552,
+      "learning_rate": 0.0002,
+      "loss": 1.1128,
+      "step": 8980
+    },
+    {
+      "epoch": 6.023450586264657,
+      "grad_norm": 1.1868711709976196,
+      "learning_rate": 0.0002,
+      "loss": 1.1382,
+      "step": 8990
+    },
+    {
+      "epoch": 6.030150753768845,
+      "grad_norm": 1.2984495162963867,
+      "learning_rate": 0.0002,
+      "loss": 1.1377,
+      "step": 9000
+    },
+    {
+      "epoch": 6.0368509212730315,
+      "grad_norm": 1.1147589683532715,
+      "learning_rate": 0.0002,
+      "loss": 1.0803,
+      "step": 9010
+    },
+    {
+      "epoch": 6.043551088777219,
+      "grad_norm": 1.3128414154052734,
+      "learning_rate": 0.0002,
+      "loss": 1.1244,
+      "step": 9020
+    },
+    {
+      "epoch": 6.050251256281407,
+      "grad_norm": 1.068290114402771,
+      "learning_rate": 0.0002,
+      "loss": 1.097,
+      "step": 9030
+    },
+    {
+      "epoch": 6.056951423785595,
+      "grad_norm": 1.1890562772750854,
+      "learning_rate": 0.0002,
+      "loss": 1.1764,
+      "step": 9040
+    },
+    {
+      "epoch": 6.063651591289783,
+      "grad_norm": 1.2121573686599731,
+      "learning_rate": 0.0002,
+      "loss": 1.1239,
+      "step": 9050
+    },
+    {
+      "epoch": 6.0703517587939695,
+      "grad_norm": 1.0860483646392822,
+      "learning_rate": 0.0002,
+      "loss": 1.0683,
+      "step": 9060
+    },
+    {
+      "epoch": 6.077051926298157,
+      "grad_norm": 1.1214599609375,
+      "learning_rate": 0.0002,
+      "loss": 1.1613,
+      "step": 9070
+    },
+    {
+      "epoch": 6.083752093802345,
+      "grad_norm": 1.147580862045288,
+      "learning_rate": 0.0002,
+      "loss": 1.1001,
+      "step": 9080
+    },
+    {
+      "epoch": 6.090452261306533,
+      "grad_norm": 1.3233155012130737,
+      "learning_rate": 0.0002,
+      "loss": 1.1154,
+      "step": 9090
+    },
+    {
+      "epoch": 6.097152428810721,
+      "grad_norm": 1.1869080066680908,
+      "learning_rate": 0.0002,
+      "loss": 1.1017,
+      "step": 9100
+    },
+    {
+      "epoch": 6.1038525963149075,
+      "grad_norm": 1.1695014238357544,
+      "learning_rate": 0.0002,
+      "loss": 1.1532,
+      "step": 9110
+    },
+    {
+      "epoch": 6.110552763819095,
+      "grad_norm": 1.1982251405715942,
+      "learning_rate": 0.0002,
+      "loss": 1.1366,
+      "step": 9120
+    },
+    {
+      "epoch": 6.117252931323283,
+      "grad_norm": 1.1426950693130493,
+      "learning_rate": 0.0002,
+      "loss": 1.0805,
+      "step": 9130
+    },
+    {
+      "epoch": 6.123953098827471,
+      "grad_norm": 1.2257394790649414,
+      "learning_rate": 0.0002,
+      "loss": 1.0801,
+      "step": 9140
+    },
+    {
+      "epoch": 6.130653266331659,
+      "grad_norm": 1.2932263612747192,
+      "learning_rate": 0.0002,
+      "loss": 1.1209,
+      "step": 9150
+    },
+    {
+      "epoch": 6.1373534338358455,
+      "grad_norm": 1.2617030143737793,
+      "learning_rate": 0.0002,
+      "loss": 1.0934,
+      "step": 9160
+    },
+    {
+      "epoch": 6.144053601340033,
+      "grad_norm": 1.1201422214508057,
+      "learning_rate": 0.0002,
+      "loss": 1.0551,
+      "step": 9170
+    },
+    {
+      "epoch": 6.150753768844221,
+      "grad_norm": 0.9625319838523865,
+      "learning_rate": 0.0002,
+      "loss": 1.1059,
+      "step": 9180
+    },
+    {
+      "epoch": 6.157453936348409,
+      "grad_norm": 1.0290048122406006,
+      "learning_rate": 0.0002,
+      "loss": 1.1397,
+      "step": 9190
+    },
+    {
+      "epoch": 6.164154103852597,
+      "grad_norm": 1.1137803792953491,
+      "learning_rate": 0.0002,
+      "loss": 1.1257,
+      "step": 9200
+    },
+    {
+      "epoch": 6.1708542713567835,
+      "grad_norm": 1.3674522638320923,
+      "learning_rate": 0.0002,
+      "loss": 1.1211,
+      "step": 9210
+    },
+    {
+      "epoch": 6.177554438860971,
+      "grad_norm": 1.182207703590393,
+      "learning_rate": 0.0002,
+      "loss": 1.0947,
+      "step": 9220
+    },
+    {
+      "epoch": 6.184254606365159,
+      "grad_norm": 1.0496711730957031,
+      "learning_rate": 0.0002,
+      "loss": 1.0838,
+      "step": 9230
+    },
+    {
+      "epoch": 6.190954773869347,
+      "grad_norm": 1.1899489164352417,
+      "learning_rate": 0.0002,
+      "loss": 1.0666,
+      "step": 9240
+    },
+    {
+      "epoch": 6.197654941373535,
+      "grad_norm": 1.2666147947311401,
+      "learning_rate": 0.0002,
+      "loss": 1.1633,
+      "step": 9250
+    },
+    {
+      "epoch": 6.204355108877722,
+      "grad_norm": 1.2013030052185059,
+      "learning_rate": 0.0002,
+      "loss": 1.1532,
+      "step": 9260
+    },
+    {
+      "epoch": 6.211055276381909,
+      "grad_norm": 1.3049768209457397,
+      "learning_rate": 0.0002,
+      "loss": 1.151,
+      "step": 9270
+    },
+    {
+      "epoch": 6.217755443886097,
+      "grad_norm": 1.1733006238937378,
+      "learning_rate": 0.0002,
+      "loss": 1.0741,
+      "step": 9280
+    },
+    {
+      "epoch": 6.224455611390285,
+      "grad_norm": 1.2742516994476318,
+      "learning_rate": 0.0002,
+      "loss": 1.0933,
+      "step": 9290
+    },
+    {
+      "epoch": 6.231155778894473,
+      "grad_norm": 1.110198974609375,
+      "learning_rate": 0.0002,
+      "loss": 1.1028,
+      "step": 9300
+    },
+    {
+      "epoch": 6.23785594639866,
+      "grad_norm": 1.159963607788086,
+      "learning_rate": 0.0002,
+      "loss": 1.1619,
+      "step": 9310
+    },
+    {
+      "epoch": 6.244556113902847,
+      "grad_norm": 1.302216649055481,
+      "learning_rate": 0.0002,
+      "loss": 1.0716,
+      "step": 9320
+    },
+    {
+      "epoch": 6.251256281407035,
+      "grad_norm": 1.2134063243865967,
+      "learning_rate": 0.0002,
+      "loss": 1.0694,
+      "step": 9330
+    },
+    {
+      "epoch": 6.257956448911223,
+      "grad_norm": 1.062682867050171,
+      "learning_rate": 0.0002,
+      "loss": 1.2151,
+      "step": 9340
+    },
+    {
+      "epoch": 6.264656616415411,
+      "grad_norm": 1.1568971872329712,
+      "learning_rate": 0.0002,
+      "loss": 1.148,
+      "step": 9350
+    },
+    {
+      "epoch": 6.271356783919598,
+      "grad_norm": 0.9914957880973816,
+      "learning_rate": 0.0002,
+      "loss": 1.094,
+      "step": 9360
+    },
+    {
+      "epoch": 6.278056951423785,
+      "grad_norm": 1.017250895500183,
+      "learning_rate": 0.0002,
+      "loss": 1.125,
+      "step": 9370
+    },
+    {
+      "epoch": 6.284757118927973,
+      "grad_norm": 1.1862834692001343,
+      "learning_rate": 0.0002,
+      "loss": 1.2177,
+      "step": 9380
+    },
+    {
+      "epoch": 6.291457286432161,
+      "grad_norm": 1.2834911346435547,
+      "learning_rate": 0.0002,
+      "loss": 0.9994,
+      "step": 9390
+    },
+    {
+      "epoch": 6.298157453936349,
+      "grad_norm": 1.3306856155395508,
+      "learning_rate": 0.0002,
+      "loss": 1.0922,
+      "step": 9400
+    },
+    {
+      "epoch": 6.304857621440536,
+      "grad_norm": 1.12908136844635,
+      "learning_rate": 0.0002,
+      "loss": 1.1136,
+      "step": 9410
+    },
+    {
+      "epoch": 6.311557788944723,
+      "grad_norm": 1.2157351970672607,
+      "learning_rate": 0.0002,
+      "loss": 1.1406,
+      "step": 9420
+    },
+    {
+      "epoch": 6.318257956448911,
+      "grad_norm": 1.121882677078247,
+      "learning_rate": 0.0002,
+      "loss": 1.1388,
+      "step": 9430
+    },
+    {
+      "epoch": 6.324958123953099,
+      "grad_norm": 1.3144481182098389,
+      "learning_rate": 0.0002,
+      "loss": 1.1648,
+      "step": 9440
+    },
+    {
+      "epoch": 6.331658291457287,
+      "grad_norm": 1.1946896314620972,
+      "learning_rate": 0.0002,
+      "loss": 1.1228,
+      "step": 9450
+    },
+    {
+      "epoch": 6.338358458961474,
+      "grad_norm": 1.1289668083190918,
+      "learning_rate": 0.0002,
+      "loss": 1.1613,
+      "step": 9460
+    },
+    {
+      "epoch": 6.345058626465661,
+      "grad_norm": 1.1065658330917358,
+      "learning_rate": 0.0002,
+      "loss": 1.1059,
+      "step": 9470
+    },
+    {
+      "epoch": 6.351758793969849,
+      "grad_norm": 1.0881422758102417,
+      "learning_rate": 0.0002,
+      "loss": 1.1431,
+      "step": 9480
+    },
+    {
+      "epoch": 6.358458961474037,
+      "grad_norm": 1.242676854133606,
+      "learning_rate": 0.0002,
+      "loss": 1.223,
+      "step": 9490
+    },
+    {
+      "epoch": 6.365159128978225,
+      "grad_norm": 0.9650855660438538,
+      "learning_rate": 0.0002,
+      "loss": 1.1379,
+      "step": 9500
+    },
+    {
+      "epoch": 6.371859296482412,
+      "grad_norm": 1.2845722436904907,
+      "learning_rate": 0.0002,
+      "loss": 1.0763,
+      "step": 9510
+    },
+    {
+      "epoch": 6.3785594639865995,
+      "grad_norm": 1.0327043533325195,
+      "learning_rate": 0.0002,
+      "loss": 1.1351,
+      "step": 9520
+    },
+    {
+      "epoch": 6.385259631490787,
+      "grad_norm": 1.0780898332595825,
+      "learning_rate": 0.0002,
+      "loss": 1.114,
+      "step": 9530
+    },
+    {
+      "epoch": 6.391959798994975,
+      "grad_norm": 1.4934027194976807,
+      "learning_rate": 0.0002,
+      "loss": 1.1579,
+      "step": 9540
+    },
+    {
+      "epoch": 6.398659966499163,
+      "grad_norm": 0.9882908463478088,
+      "learning_rate": 0.0002,
+      "loss": 1.1546,
+      "step": 9550
+    },
+    {
+      "epoch": 6.40536013400335,
+      "grad_norm": 1.3250664472579956,
+      "learning_rate": 0.0002,
+      "loss": 1.1145,
+      "step": 9560
+    },
+    {
+      "epoch": 6.4120603015075375,
+      "grad_norm": 1.1888482570648193,
+      "learning_rate": 0.0002,
+      "loss": 1.2333,
+      "step": 9570
+    },
+    {
+      "epoch": 6.418760469011725,
+      "grad_norm": 1.136496901512146,
+      "learning_rate": 0.0002,
+      "loss": 1.0892,
+      "step": 9580
+    },
+    {
+      "epoch": 6.425460636515913,
+      "grad_norm": 1.161360502243042,
+      "learning_rate": 0.0002,
+      "loss": 1.1674,
+      "step": 9590
+    },
+    {
+      "epoch": 6.432160804020101,
+      "grad_norm": 1.2034236192703247,
+      "learning_rate": 0.0002,
+      "loss": 1.1293,
+      "step": 9600
+    },
+    {
+      "epoch": 6.438860971524288,
+      "grad_norm": 1.0268361568450928,
+      "learning_rate": 0.0002,
+      "loss": 1.1059,
+      "step": 9610
+    },
+    {
+      "epoch": 6.4455611390284755,
+      "grad_norm": 1.2132930755615234,
+      "learning_rate": 0.0002,
+      "loss": 1.1732,
+      "step": 9620
+    },
+    {
+      "epoch": 6.452261306532663,
+      "grad_norm": 1.0773013830184937,
+      "learning_rate": 0.0002,
+      "loss": 1.1329,
+      "step": 9630
+    },
+    {
+      "epoch": 6.458961474036851,
+      "grad_norm": 1.3848375082015991,
+      "learning_rate": 0.0002,
+      "loss": 1.0822,
+      "step": 9640
+    },
+    {
+      "epoch": 6.465661641541039,
+      "grad_norm": 1.110495924949646,
+      "learning_rate": 0.0002,
+      "loss": 1.1778,
+      "step": 9650
+    },
+    {
+      "epoch": 6.472361809045226,
+      "grad_norm": 1.118093729019165,
+      "learning_rate": 0.0002,
+      "loss": 1.2022,
+      "step": 9660
+    },
+    {
+      "epoch": 6.4790619765494135,
+      "grad_norm": 1.2611900568008423,
+      "learning_rate": 0.0002,
+      "loss": 1.1222,
+      "step": 9670
+    },
+    {
+      "epoch": 6.485762144053601,
+      "grad_norm": 0.971754252910614,
+      "learning_rate": 0.0002,
+      "loss": 1.2138,
+      "step": 9680
+    },
+    {
+      "epoch": 6.492462311557789,
+      "grad_norm": 1.2615419626235962,
+      "learning_rate": 0.0002,
+      "loss": 1.1641,
+      "step": 9690
+    },
+    {
+      "epoch": 6.499162479061977,
+      "grad_norm": 1.1370900869369507,
+      "learning_rate": 0.0002,
+      "loss": 1.1412,
+      "step": 9700
+    },
+    {
+      "epoch": 6.505862646566165,
+      "grad_norm": 1.1815906763076782,
+      "learning_rate": 0.0002,
+      "loss": 1.186,
+      "step": 9710
+    },
+    {
+      "epoch": 6.5125628140703515,
+      "grad_norm": 1.3424339294433594,
+      "learning_rate": 0.0002,
+      "loss": 1.167,
+      "step": 9720
+    },
+    {
+      "epoch": 6.519262981574539,
+      "grad_norm": 1.2858397960662842,
+      "learning_rate": 0.0002,
+      "loss": 1.1602,
+      "step": 9730
+    },
+    {
+      "epoch": 6.525963149078727,
+      "grad_norm": 0.9578179121017456,
+      "learning_rate": 0.0002,
+      "loss": 1.178,
+      "step": 9740
+    },
+    {
+      "epoch": 6.532663316582915,
+      "grad_norm": 1.3105167150497437,
+      "learning_rate": 0.0002,
+      "loss": 1.1805,
+      "step": 9750
+    },
+    {
+      "epoch": 6.539363484087103,
+      "grad_norm": 1.0586575269699097,
+      "learning_rate": 0.0002,
+      "loss": 1.1899,
+      "step": 9760
+    },
+    {
+      "epoch": 6.54606365159129,
+      "grad_norm": 1.2122068405151367,
+      "learning_rate": 0.0002,
+      "loss": 1.095,
+      "step": 9770
+    },
+    {
+      "epoch": 6.552763819095477,
+      "grad_norm": 1.3088626861572266,
+      "learning_rate": 0.0002,
+      "loss": 1.1471,
+      "step": 9780
+    },
+    {
+      "epoch": 6.559463986599665,
+      "grad_norm": 1.194122076034546,
+      "learning_rate": 0.0002,
+      "loss": 1.1067,
+      "step": 9790
+    },
+    {
+      "epoch": 6.566164154103853,
+      "grad_norm": 1.1508387327194214,
+      "learning_rate": 0.0002,
+      "loss": 1.0967,
+      "step": 9800
+    },
+    {
+      "epoch": 6.572864321608041,
+      "grad_norm": 1.109228253364563,
+      "learning_rate": 0.0002,
+      "loss": 1.1694,
+      "step": 9810
+    },
+    {
+      "epoch": 6.579564489112228,
+      "grad_norm": 1.1607427597045898,
+      "learning_rate": 0.0002,
+      "loss": 1.1378,
+      "step": 9820
+    },
+    {
+      "epoch": 6.586264656616415,
+      "grad_norm": 1.174089789390564,
+      "learning_rate": 0.0002,
+      "loss": 1.1585,
+      "step": 9830
+    },
+    {
+      "epoch": 6.592964824120603,
+      "grad_norm": 1.1739521026611328,
+      "learning_rate": 0.0002,
+      "loss": 1.1385,
+      "step": 9840
+    },
+    {
+      "epoch": 6.599664991624791,
+      "grad_norm": 1.098528504371643,
+      "learning_rate": 0.0002,
+      "loss": 1.155,
+      "step": 9850
+    },
+    {
+      "epoch": 6.606365159128979,
+      "grad_norm": 1.0397740602493286,
+      "learning_rate": 0.0002,
+      "loss": 1.1359,
+      "step": 9860
+    },
+    {
+      "epoch": 6.613065326633166,
+      "grad_norm": 1.1087969541549683,
+      "learning_rate": 0.0002,
+      "loss": 1.1433,
+      "step": 9870
+    },
+    {
+      "epoch": 6.619765494137353,
+      "grad_norm": 1.2070481777191162,
+      "learning_rate": 0.0002,
+      "loss": 1.2356,
+      "step": 9880
+    },
+    {
+      "epoch": 6.626465661641541,
+      "grad_norm": 1.1115655899047852,
+      "learning_rate": 0.0002,
+      "loss": 1.1161,
+      "step": 9890
+    },
+    {
+      "epoch": 6.633165829145729,
+      "grad_norm": 1.2486097812652588,
+      "learning_rate": 0.0002,
+      "loss": 1.2163,
+      "step": 9900
+    },
+    {
+      "epoch": 6.639865996649917,
+      "grad_norm": 1.230380654335022,
+      "learning_rate": 0.0002,
+      "loss": 1.0984,
+      "step": 9910
+    },
+    {
+      "epoch": 6.646566164154104,
+      "grad_norm": 1.1479365825653076,
+      "learning_rate": 0.0002,
+      "loss": 1.1862,
+      "step": 9920
+    },
+    {
+      "epoch": 6.653266331658291,
+      "grad_norm": 1.0790960788726807,
+      "learning_rate": 0.0002,
+      "loss": 1.1139,
+      "step": 9930
+    },
+    {
+      "epoch": 6.659966499162479,
+      "grad_norm": 1.1157397031784058,
+      "learning_rate": 0.0002,
+      "loss": 1.2001,
+      "step": 9940
+    },
+    {
+      "epoch": 6.666666666666667,
+      "grad_norm": 1.3104028701782227,
+      "learning_rate": 0.0002,
+      "loss": 1.1085,
+      "step": 9950
+    },
+    {
+      "epoch": 6.673366834170855,
+      "grad_norm": 1.1727646589279175,
+      "learning_rate": 0.0002,
+      "loss": 1.2209,
+      "step": 9960
+    },
+    {
+      "epoch": 6.680067001675042,
+      "grad_norm": 1.2104284763336182,
+      "learning_rate": 0.0002,
+      "loss": 1.1671,
+      "step": 9970
+    },
+    {
+      "epoch": 6.686767169179229,
+      "grad_norm": 1.2023727893829346,
+      "learning_rate": 0.0002,
+      "loss": 1.1952,
+      "step": 9980
+    },
+    {
+      "epoch": 6.693467336683417,
+      "grad_norm": 1.0088225603103638,
+      "learning_rate": 0.0002,
+      "loss": 1.1385,
+      "step": 9990
+    },
+    {
+      "epoch": 6.700167504187605,
+      "grad_norm": 1.298015832901001,
+      "learning_rate": 0.0002,
+      "loss": 1.1314,
+      "step": 10000
+    },
+    {
+      "epoch": 6.706867671691793,
+      "grad_norm": 1.1315910816192627,
+      "learning_rate": 0.0002,
+      "loss": 1.1983,
+      "step": 10010
+    },
+    {
+      "epoch": 6.71356783919598,
+      "grad_norm": 1.1283273696899414,
+      "learning_rate": 0.0002,
+      "loss": 1.1679,
+      "step": 10020
+    },
+    {
+      "epoch": 6.720268006700167,
+      "grad_norm": 1.2564418315887451,
+      "learning_rate": 0.0002,
+      "loss": 1.1917,
+      "step": 10030
+    },
+    {
+      "epoch": 6.726968174204355,
+      "grad_norm": 1.0451353788375854,
+      "learning_rate": 0.0002,
+      "loss": 1.1788,
+      "step": 10040
+    },
+    {
+      "epoch": 6.733668341708543,
+      "grad_norm": 1.054793357849121,
+      "learning_rate": 0.0002,
+      "loss": 1.1905,
+      "step": 10050
+    },
+    {
+      "epoch": 6.740368509212731,
+      "grad_norm": 1.2741243839263916,
+      "learning_rate": 0.0002,
+      "loss": 1.1814,
+      "step": 10060
+    },
+    {
+      "epoch": 6.747068676716918,
+      "grad_norm": 1.1342514753341675,
+      "learning_rate": 0.0002,
+      "loss": 1.2015,
+      "step": 10070
+    },
+    {
+      "epoch": 6.7537688442211055,
+      "grad_norm": 1.0081498622894287,
+      "learning_rate": 0.0002,
+      "loss": 1.2587,
+      "step": 10080
+    },
+    {
+      "epoch": 6.760469011725293,
+      "grad_norm": 1.2164603471755981,
+      "learning_rate": 0.0002,
+      "loss": 1.1226,
+      "step": 10090
+    },
+    {
+      "epoch": 6.767169179229481,
+      "grad_norm": 1.2062463760375977,
+      "learning_rate": 0.0002,
+      "loss": 1.1353,
+      "step": 10100
+    },
+    {
+      "epoch": 6.773869346733669,
+      "grad_norm": 1.2255526781082153,
+      "learning_rate": 0.0002,
+      "loss": 1.2143,
+      "step": 10110
+    },
+    {
+      "epoch": 6.780569514237856,
+      "grad_norm": 1.08175790309906,
+      "learning_rate": 0.0002,
+      "loss": 1.1544,
+      "step": 10120
+    },
+    {
+      "epoch": 6.7872696817420435,
+      "grad_norm": 1.5781128406524658,
+      "learning_rate": 0.0002,
+      "loss": 1.1983,
+      "step": 10130
+    },
+    {
+      "epoch": 6.793969849246231,
+      "grad_norm": 1.0622451305389404,
+      "learning_rate": 0.0002,
+      "loss": 1.0994,
+      "step": 10140
+    },
+    {
+      "epoch": 6.800670016750419,
+      "grad_norm": 1.1591497659683228,
+      "learning_rate": 0.0002,
+      "loss": 1.2084,
+      "step": 10150
+    },
+    {
+      "epoch": 6.807370184254607,
+      "grad_norm": 1.0398483276367188,
+      "learning_rate": 0.0002,
+      "loss": 1.2203,
+      "step": 10160
+    },
+    {
+      "epoch": 6.814070351758794,
+      "grad_norm": 1.229132056236267,
+      "learning_rate": 0.0002,
+      "loss": 1.2249,
+      "step": 10170
+    },
+    {
+      "epoch": 6.8207705192629815,
+      "grad_norm": 1.0918090343475342,
+      "learning_rate": 0.0002,
+      "loss": 1.1789,
+      "step": 10180
+    },
+    {
+      "epoch": 6.827470686767169,
+      "grad_norm": 1.1543749570846558,
+      "learning_rate": 0.0002,
+      "loss": 1.1639,
+      "step": 10190
+    },
+    {
+      "epoch": 6.834170854271357,
+      "grad_norm": 1.1831817626953125,
+      "learning_rate": 0.0002,
+      "loss": 1.1353,
+      "step": 10200
+    },
+    {
+      "epoch": 6.840871021775545,
+      "grad_norm": 1.305327296257019,
+      "learning_rate": 0.0002,
+      "loss": 1.2565,
+      "step": 10210
+    },
+    {
+      "epoch": 6.847571189279732,
+      "grad_norm": 1.136720061302185,
+      "learning_rate": 0.0002,
+      "loss": 1.2037,
+      "step": 10220
+    },
+    {
+      "epoch": 6.8542713567839195,
+      "grad_norm": 1.2282346487045288,
+      "learning_rate": 0.0002,
+      "loss": 1.2256,
+      "step": 10230
+    },
+    {
+      "epoch": 6.860971524288107,
+      "grad_norm": 1.2457010746002197,
+      "learning_rate": 0.0002,
+      "loss": 1.1281,
+      "step": 10240
+    },
+    {
+      "epoch": 6.867671691792295,
+      "grad_norm": 1.2808631658554077,
+      "learning_rate": 0.0002,
+      "loss": 1.2762,
+      "step": 10250
+    },
+    {
+      "epoch": 6.874371859296483,
+      "grad_norm": 1.089066743850708,
+      "learning_rate": 0.0002,
+      "loss": 1.2213,
+      "step": 10260
+    },
+    {
+      "epoch": 6.88107202680067,
+      "grad_norm": 0.9543178081512451,
+      "learning_rate": 0.0002,
+      "loss": 1.2627,
+      "step": 10270
+    },
+    {
+      "epoch": 6.8877721943048575,
+      "grad_norm": 1.1149744987487793,
+      "learning_rate": 0.0002,
+      "loss": 1.1617,
+      "step": 10280
+    },
+    {
+      "epoch": 6.894472361809045,
+      "grad_norm": 1.0185538530349731,
+      "learning_rate": 0.0002,
+      "loss": 1.1134,
+      "step": 10290
+    },
+    {
+      "epoch": 6.901172529313233,
+      "grad_norm": 0.9954617619514465,
+      "learning_rate": 0.0002,
+      "loss": 1.217,
+      "step": 10300
+    },
+    {
+      "epoch": 6.907872696817421,
+      "grad_norm": 1.2581418752670288,
+      "learning_rate": 0.0002,
+      "loss": 1.1524,
+      "step": 10310
+    },
+    {
+      "epoch": 6.914572864321608,
+      "grad_norm": 1.2430983781814575,
+      "learning_rate": 0.0002,
+      "loss": 1.1942,
+      "step": 10320
+    },
+    {
+      "epoch": 6.921273031825796,
+      "grad_norm": 1.4937270879745483,
+      "learning_rate": 0.0002,
+      "loss": 1.1254,
+      "step": 10330
+    },
+    {
+      "epoch": 6.927973199329983,
+      "grad_norm": 1.1257144212722778,
+      "learning_rate": 0.0002,
+      "loss": 1.1519,
+      "step": 10340
+    },
+    {
+      "epoch": 6.934673366834171,
+      "grad_norm": 1.2068904638290405,
+      "learning_rate": 0.0002,
+      "loss": 1.2622,
+      "step": 10350
+    },
+    {
+      "epoch": 6.941373534338359,
+      "grad_norm": 1.0290757417678833,
+      "learning_rate": 0.0002,
+      "loss": 1.1349,
+      "step": 10360
+    },
+    {
+      "epoch": 6.948073701842546,
+      "grad_norm": 1.0070724487304688,
+      "learning_rate": 0.0002,
+      "loss": 1.1752,
+      "step": 10370
+    },
+    {
+      "epoch": 6.954773869346734,
+      "grad_norm": 0.9936357140541077,
+      "learning_rate": 0.0002,
+      "loss": 1.1838,
+      "step": 10380
+    },
+    {
+      "epoch": 6.961474036850921,
+      "grad_norm": 1.1063416004180908,
+      "learning_rate": 0.0002,
+      "loss": 1.2305,
+      "step": 10390
+    },
+    {
+      "epoch": 6.968174204355109,
+      "grad_norm": 1.5199986696243286,
+      "learning_rate": 0.0002,
+      "loss": 1.154,
+      "step": 10400
+    },
+    {
+      "epoch": 6.974874371859297,
+      "grad_norm": 1.160731554031372,
+      "learning_rate": 0.0002,
+      "loss": 1.1143,
+      "step": 10410
+    },
+    {
+      "epoch": 6.981574539363484,
+      "grad_norm": 1.084697961807251,
+      "learning_rate": 0.0002,
+      "loss": 1.2132,
+      "step": 10420
+    },
+    {
+      "epoch": 6.988274706867672,
+      "grad_norm": 1.1257576942443848,
+      "learning_rate": 0.0002,
+      "loss": 1.1995,
+      "step": 10430
+    },
+    {
+      "epoch": 6.994974874371859,
+      "grad_norm": 1.310616135597229,
+      "learning_rate": 0.0002,
+      "loss": 1.1197,
+      "step": 10440
+    },
+    {
+      "epoch": 6.99966499162479,
+      "eval_loss": 2.1203012466430664,
+      "eval_runtime": 37.936,
+      "eval_samples_per_second": 13.576,
+      "eval_steps_per_second": 1.713,
+      "step": 10447
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 11936,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 8,
+  "save_steps": 200,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 4.834869071000371e+17,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-10447/training_args.bin b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-10447/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..064f299b0f3d2a28f5b1f5c68ef32caab3e2dd49
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-10447/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aa7b586fc440d1e22156593e2dd4267d2bdcb8920a02fdf352ea29a9bec3dd94
+size 5560
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-11936/README.md b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-11936/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..835e31ab8469ee39ddc8b2b6b2143a8c66dad510
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-11936/README.md
@@ -0,0 +1,202 @@
+---
+base_model: meta-llama/Meta-Llama-3-8B-Instruct
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.13.1
\ No newline at end of file
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-11936/adapter_config.json b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-11936/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..406c5a08dc4a2a33b52c62a482f98c217c417215
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-11936/adapter_config.json
@@ -0,0 +1,29 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "meta-llama/Meta-Llama-3-8B-Instruct",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-11936/adapter_model.safetensors b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-11936/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..ab67859fd10c7865ae28b3816b5b9066ddb6cba0
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-11936/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9eef98d6245d2695d63ded5e3f460682daac0c7695520431d04d3d01b5fc9f15
+size 109069176
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-11936/optimizer.pt b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-11936/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..245cffb7b85a0008e22b6b24b80a62813244d76e
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-11936/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:849cf840d239789f3f568974fb379b6ddc17b287e99687426e6c464e93ef2bfc
+size 55532666
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-11936/rng_state.pth b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-11936/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..e92612298d3997192dc6db60274a0a76167c2922
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-11936/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:40b830f407e0023c6a767e982fcc0b9c16621f5f1933411e9c416dd8b1e8cd59
+size 14244
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-11936/scheduler.pt b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-11936/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..dec1ae9003899f214c493782eb2f9090beb6b746
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-11936/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a5845c9f2080be96d77ec57c3dc3a5e4a524b0af908d8d720010419b1753faf3
+size 1064
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-11936/special_tokens_map.json b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-11936/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..b43be96621d147110fb8a18b5776ec6e38516127
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-11936/special_tokens_map.json
@@ -0,0 +1,17 @@
+{
+  "bos_token": {
+    "content": "<|begin_of_text|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<|eot_id|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<|eot_id|>"
+}
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-11936/tokenizer.json b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-11936/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..86a33946b0c77216d2cce91bb28c8fada4a5e80b
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-11936/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3c5cf44023714fb39b05e71e425f8d7b92805ff73f7988b083b8c87f0bf87393
+size 17209961
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-11936/tokenizer_config.json b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-11936/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..061e40d9db3253624f86e8e364c15ef546527c9d
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-11936/tokenizer_config.json
@@ -0,0 +1,2063 @@
+{
+  "added_tokens_decoder": {
+    "128000": {
+      "content": "<|begin_of_text|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128001": {
+      "content": "<|end_of_text|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128002": {
+      "content": "<|reserved_special_token_0|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128003": {
+      "content": "<|reserved_special_token_1|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128004": {
+      "content": "<|reserved_special_token_2|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128005": {
+      "content": "<|reserved_special_token_3|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128006": {
+      "content": "<|start_header_id|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128007": {
+      "content": "<|end_header_id|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128008": {
+      "content": "<|reserved_special_token_4|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128009": {
+      "content": "<|eot_id|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128010": {
+      "content": "<|reserved_special_token_5|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128011": {
+      "content": "<|reserved_special_token_6|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128012": {
+      "content": "<|reserved_special_token_7|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128013": {
+      "content": "<|reserved_special_token_8|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128014": {
+      "content": "<|reserved_special_token_9|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128015": {
+      "content": "<|reserved_special_token_10|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128016": {
+      "content": "<|reserved_special_token_11|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128017": {
+      "content": "<|reserved_special_token_12|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128018": {
+      "content": "<|reserved_special_token_13|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128019": {
+      "content": "<|reserved_special_token_14|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128020": {
+      "content": "<|reserved_special_token_15|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128021": {
+      "content": "<|reserved_special_token_16|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128022": {
+      "content": "<|reserved_special_token_17|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128023": {
+      "content": "<|reserved_special_token_18|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128024": {
+      "content": "<|reserved_special_token_19|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128025": {
+      "content": "<|reserved_special_token_20|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128026": {
+      "content": "<|reserved_special_token_21|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128027": {
+      "content": "<|reserved_special_token_22|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128028": {
+      "content": "<|reserved_special_token_23|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128029": {
+      "content": "<|reserved_special_token_24|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128030": {
+      "content": "<|reserved_special_token_25|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128031": {
+      "content": "<|reserved_special_token_26|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128032": {
+      "content": "<|reserved_special_token_27|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128033": {
+      "content": "<|reserved_special_token_28|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128034": {
+      "content": "<|reserved_special_token_29|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128035": {
+      "content": "<|reserved_special_token_30|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128036": {
+      "content": "<|reserved_special_token_31|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128037": {
+      "content": "<|reserved_special_token_32|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128038": {
+      "content": "<|reserved_special_token_33|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128039": {
+      "content": "<|reserved_special_token_34|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128040": {
+      "content": "<|reserved_special_token_35|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128041": {
+      "content": "<|reserved_special_token_36|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128042": {
+      "content": "<|reserved_special_token_37|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128043": {
+      "content": "<|reserved_special_token_38|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128044": {
+      "content": "<|reserved_special_token_39|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128045": {
+      "content": "<|reserved_special_token_40|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128046": {
+      "content": "<|reserved_special_token_41|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128047": {
+      "content": "<|reserved_special_token_42|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128048": {
+      "content": "<|reserved_special_token_43|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128049": {
+      "content": "<|reserved_special_token_44|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128050": {
+      "content": "<|reserved_special_token_45|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128051": {
+      "content": "<|reserved_special_token_46|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128052": {
+      "content": "<|reserved_special_token_47|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128053": {
+      "content": "<|reserved_special_token_48|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128054": {
+      "content": "<|reserved_special_token_49|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128055": {
+      "content": "<|reserved_special_token_50|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128056": {
+      "content": "<|reserved_special_token_51|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128057": {
+      "content": "<|reserved_special_token_52|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128058": {
+      "content": "<|reserved_special_token_53|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128059": {
+      "content": "<|reserved_special_token_54|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128060": {
+      "content": "<|reserved_special_token_55|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128061": {
+      "content": "<|reserved_special_token_56|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128062": {
+      "content": "<|reserved_special_token_57|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128063": {
+      "content": "<|reserved_special_token_58|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128064": {
+      "content": "<|reserved_special_token_59|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128065": {
+      "content": "<|reserved_special_token_60|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128066": {
+      "content": "<|reserved_special_token_61|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128067": {
+      "content": "<|reserved_special_token_62|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128068": {
+      "content": "<|reserved_special_token_63|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128069": {
+      "content": "<|reserved_special_token_64|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128070": {
+      "content": "<|reserved_special_token_65|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128071": {
+      "content": "<|reserved_special_token_66|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128072": {
+      "content": "<|reserved_special_token_67|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128073": {
+      "content": "<|reserved_special_token_68|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128074": {
+      "content": "<|reserved_special_token_69|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128075": {
+      "content": "<|reserved_special_token_70|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128076": {
+      "content": "<|reserved_special_token_71|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128077": {
+      "content": "<|reserved_special_token_72|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128078": {
+      "content": "<|reserved_special_token_73|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128079": {
+      "content": "<|reserved_special_token_74|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128080": {
+      "content": "<|reserved_special_token_75|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128081": {
+      "content": "<|reserved_special_token_76|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128082": {
+      "content": "<|reserved_special_token_77|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128083": {
+      "content": "<|reserved_special_token_78|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128084": {
+      "content": "<|reserved_special_token_79|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128085": {
+      "content": "<|reserved_special_token_80|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128086": {
+      "content": "<|reserved_special_token_81|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128087": {
+      "content": "<|reserved_special_token_82|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128088": {
+      "content": "<|reserved_special_token_83|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128089": {
+      "content": "<|reserved_special_token_84|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128090": {
+      "content": "<|reserved_special_token_85|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128091": {
+      "content": "<|reserved_special_token_86|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128092": {
+      "content": "<|reserved_special_token_87|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128093": {
+      "content": "<|reserved_special_token_88|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128094": {
+      "content": "<|reserved_special_token_89|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128095": {
+      "content": "<|reserved_special_token_90|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128096": {
+      "content": "<|reserved_special_token_91|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128097": {
+      "content": "<|reserved_special_token_92|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128098": {
+      "content": "<|reserved_special_token_93|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128099": {
+      "content": "<|reserved_special_token_94|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128100": {
+      "content": "<|reserved_special_token_95|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128101": {
+      "content": "<|reserved_special_token_96|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128102": {
+      "content": "<|reserved_special_token_97|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128103": {
+      "content": "<|reserved_special_token_98|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128104": {
+      "content": "<|reserved_special_token_99|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128105": {
+      "content": "<|reserved_special_token_100|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128106": {
+      "content": "<|reserved_special_token_101|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128107": {
+      "content": "<|reserved_special_token_102|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128108": {
+      "content": "<|reserved_special_token_103|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128109": {
+      "content": "<|reserved_special_token_104|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128110": {
+      "content": "<|reserved_special_token_105|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128111": {
+      "content": "<|reserved_special_token_106|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128112": {
+      "content": "<|reserved_special_token_107|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128113": {
+      "content": "<|reserved_special_token_108|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128114": {
+      "content": "<|reserved_special_token_109|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128115": {
+      "content": "<|reserved_special_token_110|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128116": {
+      "content": "<|reserved_special_token_111|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128117": {
+      "content": "<|reserved_special_token_112|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128118": {
+      "content": "<|reserved_special_token_113|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128119": {
+      "content": "<|reserved_special_token_114|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128120": {
+      "content": "<|reserved_special_token_115|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128121": {
+      "content": "<|reserved_special_token_116|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128122": {
+      "content": "<|reserved_special_token_117|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128123": {
+      "content": "<|reserved_special_token_118|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128124": {
+      "content": "<|reserved_special_token_119|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128125": {
+      "content": "<|reserved_special_token_120|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128126": {
+      "content": "<|reserved_special_token_121|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128127": {
+      "content": "<|reserved_special_token_122|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128128": {
+      "content": "<|reserved_special_token_123|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128129": {
+      "content": "<|reserved_special_token_124|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128130": {
+      "content": "<|reserved_special_token_125|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128131": {
+      "content": "<|reserved_special_token_126|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128132": {
+      "content": "<|reserved_special_token_127|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128133": {
+      "content": "<|reserved_special_token_128|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128134": {
+      "content": "<|reserved_special_token_129|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128135": {
+      "content": "<|reserved_special_token_130|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128136": {
+      "content": "<|reserved_special_token_131|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128137": {
+      "content": "<|reserved_special_token_132|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128138": {
+      "content": "<|reserved_special_token_133|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128139": {
+      "content": "<|reserved_special_token_134|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128140": {
+      "content": "<|reserved_special_token_135|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128141": {
+      "content": "<|reserved_special_token_136|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128142": {
+      "content": "<|reserved_special_token_137|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128143": {
+      "content": "<|reserved_special_token_138|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128144": {
+      "content": "<|reserved_special_token_139|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128145": {
+      "content": "<|reserved_special_token_140|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128146": {
+      "content": "<|reserved_special_token_141|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128147": {
+      "content": "<|reserved_special_token_142|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128148": {
+      "content": "<|reserved_special_token_143|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128149": {
+      "content": "<|reserved_special_token_144|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128150": {
+      "content": "<|reserved_special_token_145|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128151": {
+      "content": "<|reserved_special_token_146|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128152": {
+      "content": "<|reserved_special_token_147|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128153": {
+      "content": "<|reserved_special_token_148|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128154": {
+      "content": "<|reserved_special_token_149|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128155": {
+      "content": "<|reserved_special_token_150|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128156": {
+      "content": "<|reserved_special_token_151|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128157": {
+      "content": "<|reserved_special_token_152|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128158": {
+      "content": "<|reserved_special_token_153|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128159": {
+      "content": "<|reserved_special_token_154|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128160": {
+      "content": "<|reserved_special_token_155|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128161": {
+      "content": "<|reserved_special_token_156|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128162": {
+      "content": "<|reserved_special_token_157|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128163": {
+      "content": "<|reserved_special_token_158|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128164": {
+      "content": "<|reserved_special_token_159|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128165": {
+      "content": "<|reserved_special_token_160|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128166": {
+      "content": "<|reserved_special_token_161|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128167": {
+      "content": "<|reserved_special_token_162|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128168": {
+      "content": "<|reserved_special_token_163|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128169": {
+      "content": "<|reserved_special_token_164|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128170": {
+      "content": "<|reserved_special_token_165|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128171": {
+      "content": "<|reserved_special_token_166|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128172": {
+      "content": "<|reserved_special_token_167|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128173": {
+      "content": "<|reserved_special_token_168|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128174": {
+      "content": "<|reserved_special_token_169|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128175": {
+      "content": "<|reserved_special_token_170|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128176": {
+      "content": "<|reserved_special_token_171|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128177": {
+      "content": "<|reserved_special_token_172|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128178": {
+      "content": "<|reserved_special_token_173|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128179": {
+      "content": "<|reserved_special_token_174|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128180": {
+      "content": "<|reserved_special_token_175|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128181": {
+      "content": "<|reserved_special_token_176|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128182": {
+      "content": "<|reserved_special_token_177|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128183": {
+      "content": "<|reserved_special_token_178|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128184": {
+      "content": "<|reserved_special_token_179|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128185": {
+      "content": "<|reserved_special_token_180|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128186": {
+      "content": "<|reserved_special_token_181|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128187": {
+      "content": "<|reserved_special_token_182|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128188": {
+      "content": "<|reserved_special_token_183|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128189": {
+      "content": "<|reserved_special_token_184|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128190": {
+      "content": "<|reserved_special_token_185|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128191": {
+      "content": "<|reserved_special_token_186|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128192": {
+      "content": "<|reserved_special_token_187|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128193": {
+      "content": "<|reserved_special_token_188|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128194": {
+      "content": "<|reserved_special_token_189|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128195": {
+      "content": "<|reserved_special_token_190|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128196": {
+      "content": "<|reserved_special_token_191|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128197": {
+      "content": "<|reserved_special_token_192|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128198": {
+      "content": "<|reserved_special_token_193|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128199": {
+      "content": "<|reserved_special_token_194|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128200": {
+      "content": "<|reserved_special_token_195|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128201": {
+      "content": "<|reserved_special_token_196|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128202": {
+      "content": "<|reserved_special_token_197|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128203": {
+      "content": "<|reserved_special_token_198|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128204": {
+      "content": "<|reserved_special_token_199|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128205": {
+      "content": "<|reserved_special_token_200|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128206": {
+      "content": "<|reserved_special_token_201|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128207": {
+      "content": "<|reserved_special_token_202|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128208": {
+      "content": "<|reserved_special_token_203|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128209": {
+      "content": "<|reserved_special_token_204|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128210": {
+      "content": "<|reserved_special_token_205|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128211": {
+      "content": "<|reserved_special_token_206|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128212": {
+      "content": "<|reserved_special_token_207|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128213": {
+      "content": "<|reserved_special_token_208|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128214": {
+      "content": "<|reserved_special_token_209|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128215": {
+      "content": "<|reserved_special_token_210|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128216": {
+      "content": "<|reserved_special_token_211|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128217": {
+      "content": "<|reserved_special_token_212|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128218": {
+      "content": "<|reserved_special_token_213|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128219": {
+      "content": "<|reserved_special_token_214|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128220": {
+      "content": "<|reserved_special_token_215|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128221": {
+      "content": "<|reserved_special_token_216|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128222": {
+      "content": "<|reserved_special_token_217|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128223": {
+      "content": "<|reserved_special_token_218|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128224": {
+      "content": "<|reserved_special_token_219|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128225": {
+      "content": "<|reserved_special_token_220|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128226": {
+      "content": "<|reserved_special_token_221|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128227": {
+      "content": "<|reserved_special_token_222|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128228": {
+      "content": "<|reserved_special_token_223|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128229": {
+      "content": "<|reserved_special_token_224|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128230": {
+      "content": "<|reserved_special_token_225|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128231": {
+      "content": "<|reserved_special_token_226|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128232": {
+      "content": "<|reserved_special_token_227|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128233": {
+      "content": "<|reserved_special_token_228|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128234": {
+      "content": "<|reserved_special_token_229|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128235": {
+      "content": "<|reserved_special_token_230|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128236": {
+      "content": "<|reserved_special_token_231|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128237": {
+      "content": "<|reserved_special_token_232|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128238": {
+      "content": "<|reserved_special_token_233|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128239": {
+      "content": "<|reserved_special_token_234|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128240": {
+      "content": "<|reserved_special_token_235|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128241": {
+      "content": "<|reserved_special_token_236|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128242": {
+      "content": "<|reserved_special_token_237|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128243": {
+      "content": "<|reserved_special_token_238|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128244": {
+      "content": "<|reserved_special_token_239|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128245": {
+      "content": "<|reserved_special_token_240|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128246": {
+      "content": "<|reserved_special_token_241|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128247": {
+      "content": "<|reserved_special_token_242|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128248": {
+      "content": "<|reserved_special_token_243|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128249": {
+      "content": "<|reserved_special_token_244|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128250": {
+      "content": "<|reserved_special_token_245|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128251": {
+      "content": "<|reserved_special_token_246|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128252": {
+      "content": "<|reserved_special_token_247|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128253": {
+      "content": "<|reserved_special_token_248|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128254": {
+      "content": "<|reserved_special_token_249|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128255": {
+      "content": "<|reserved_special_token_250|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<|begin_of_text|>",
+  "chat_template": "{% set loop_messages = messages %}{% for message in loop_messages %}{% set content = '<|start_header_id|>' + message['role'] + '<|end_header_id|>\n\n'+ message['content'] | trim + '<|eot_id|>' %}{% if loop.index0 == 0 %}{% set content = bos_token + content %}{% endif %}{{ content }}{% endfor %}{% if add_generation_prompt %}{{ '<|start_header_id|>assistant<|end_header_id|>\n\n' }}{% endif %}",
+  "clean_up_tokenization_spaces": true,
+  "eos_token": "<|eot_id|>",
+  "model_input_names": [
+    "input_ids",
+    "attention_mask"
+  ],
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<|eot_id|>",
+  "tokenizer_class": "PreTrainedTokenizerFast"
+}
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-11936/trainer_state.json b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-11936/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..5f39e7e95996f652f99a41c8de3df54d67a93298
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-11936/trainer_state.json
@@ -0,0 +1,8448 @@
+{
+  "best_metric": 1.8028968572616577,
+  "best_model_checkpoint": "outputs-001/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-2985",
+  "epoch": 7.997319932998325,
+  "eval_steps": 10,
+  "global_step": 11936,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.006700167504187605,
+      "grad_norm": 0.565915048122406,
+      "learning_rate": 0.0002,
+      "loss": 2.6189,
+      "step": 10
+    },
+    {
+      "epoch": 0.01340033500837521,
+      "grad_norm": 0.5004463791847229,
+      "learning_rate": 0.0002,
+      "loss": 2.3162,
+      "step": 20
+    },
+    {
+      "epoch": 0.020100502512562814,
+      "grad_norm": 0.511043906211853,
+      "learning_rate": 0.0002,
+      "loss": 2.0576,
+      "step": 30
+    },
+    {
+      "epoch": 0.02680067001675042,
+      "grad_norm": 0.47327178716659546,
+      "learning_rate": 0.0002,
+      "loss": 2.0085,
+      "step": 40
+    },
+    {
+      "epoch": 0.03350083752093802,
+      "grad_norm": 0.5511676669120789,
+      "learning_rate": 0.0002,
+      "loss": 2.0276,
+      "step": 50
+    },
+    {
+      "epoch": 0.04020100502512563,
+      "grad_norm": 0.4666278064250946,
+      "learning_rate": 0.0002,
+      "loss": 1.9075,
+      "step": 60
+    },
+    {
+      "epoch": 0.04690117252931323,
+      "grad_norm": 0.5310961008071899,
+      "learning_rate": 0.0002,
+      "loss": 1.8413,
+      "step": 70
+    },
+    {
+      "epoch": 0.05360134003350084,
+      "grad_norm": 0.5606027245521545,
+      "learning_rate": 0.0002,
+      "loss": 1.8711,
+      "step": 80
+    },
+    {
+      "epoch": 0.06030150753768844,
+      "grad_norm": 0.4934779703617096,
+      "learning_rate": 0.0002,
+      "loss": 1.9282,
+      "step": 90
+    },
+    {
+      "epoch": 0.06700167504187604,
+      "grad_norm": 0.4821869730949402,
+      "learning_rate": 0.0002,
+      "loss": 1.8925,
+      "step": 100
+    },
+    {
+      "epoch": 0.07370184254606366,
+      "grad_norm": 0.5262084603309631,
+      "learning_rate": 0.0002,
+      "loss": 1.8628,
+      "step": 110
+    },
+    {
+      "epoch": 0.08040201005025126,
+      "grad_norm": 0.3774230182170868,
+      "learning_rate": 0.0002,
+      "loss": 1.8347,
+      "step": 120
+    },
+    {
+      "epoch": 0.08710217755443886,
+      "grad_norm": 0.34137430787086487,
+      "learning_rate": 0.0002,
+      "loss": 1.8386,
+      "step": 130
+    },
+    {
+      "epoch": 0.09380234505862646,
+      "grad_norm": 0.407272070646286,
+      "learning_rate": 0.0002,
+      "loss": 1.861,
+      "step": 140
+    },
+    {
+      "epoch": 0.10050251256281408,
+      "grad_norm": 0.4011937975883484,
+      "learning_rate": 0.0002,
+      "loss": 1.8279,
+      "step": 150
+    },
+    {
+      "epoch": 0.10720268006700168,
+      "grad_norm": 0.4432467222213745,
+      "learning_rate": 0.0002,
+      "loss": 1.9317,
+      "step": 160
+    },
+    {
+      "epoch": 0.11390284757118928,
+      "grad_norm": 0.44030463695526123,
+      "learning_rate": 0.0002,
+      "loss": 1.8157,
+      "step": 170
+    },
+    {
+      "epoch": 0.12060301507537688,
+      "grad_norm": 0.3799569308757782,
+      "learning_rate": 0.0002,
+      "loss": 1.8534,
+      "step": 180
+    },
+    {
+      "epoch": 0.1273031825795645,
+      "grad_norm": 0.33721521496772766,
+      "learning_rate": 0.0002,
+      "loss": 1.7658,
+      "step": 190
+    },
+    {
+      "epoch": 0.13400335008375208,
+      "grad_norm": 0.4096226692199707,
+      "learning_rate": 0.0002,
+      "loss": 1.8269,
+      "step": 200
+    },
+    {
+      "epoch": 0.1407035175879397,
+      "grad_norm": 0.37374693155288696,
+      "learning_rate": 0.0002,
+      "loss": 1.802,
+      "step": 210
+    },
+    {
+      "epoch": 0.1474036850921273,
+      "grad_norm": 0.3249480128288269,
+      "learning_rate": 0.0002,
+      "loss": 1.8901,
+      "step": 220
+    },
+    {
+      "epoch": 0.1541038525963149,
+      "grad_norm": 0.3612042963504791,
+      "learning_rate": 0.0002,
+      "loss": 1.8163,
+      "step": 230
+    },
+    {
+      "epoch": 0.16080402010050251,
+      "grad_norm": 0.3686671257019043,
+      "learning_rate": 0.0002,
+      "loss": 1.7585,
+      "step": 240
+    },
+    {
+      "epoch": 0.16750418760469013,
+      "grad_norm": 0.3521044850349426,
+      "learning_rate": 0.0002,
+      "loss": 1.8365,
+      "step": 250
+    },
+    {
+      "epoch": 0.17420435510887772,
+      "grad_norm": 0.4073677361011505,
+      "learning_rate": 0.0002,
+      "loss": 1.8623,
+      "step": 260
+    },
+    {
+      "epoch": 0.18090452261306533,
+      "grad_norm": 0.34522193670272827,
+      "learning_rate": 0.0002,
+      "loss": 1.8026,
+      "step": 270
+    },
+    {
+      "epoch": 0.18760469011725292,
+      "grad_norm": 0.4121900498867035,
+      "learning_rate": 0.0002,
+      "loss": 1.8162,
+      "step": 280
+    },
+    {
+      "epoch": 0.19430485762144054,
+      "grad_norm": 0.3544778525829315,
+      "learning_rate": 0.0002,
+      "loss": 1.7976,
+      "step": 290
+    },
+    {
+      "epoch": 0.20100502512562815,
+      "grad_norm": 0.3482133448123932,
+      "learning_rate": 0.0002,
+      "loss": 1.8787,
+      "step": 300
+    },
+    {
+      "epoch": 0.20770519262981574,
+      "grad_norm": 0.3421826660633087,
+      "learning_rate": 0.0002,
+      "loss": 1.8578,
+      "step": 310
+    },
+    {
+      "epoch": 0.21440536013400335,
+      "grad_norm": 0.5024696588516235,
+      "learning_rate": 0.0002,
+      "loss": 1.8013,
+      "step": 320
+    },
+    {
+      "epoch": 0.22110552763819097,
+      "grad_norm": 0.36013063788414,
+      "learning_rate": 0.0002,
+      "loss": 1.8607,
+      "step": 330
+    },
+    {
+      "epoch": 0.22780569514237856,
+      "grad_norm": 0.3611244857311249,
+      "learning_rate": 0.0002,
+      "loss": 1.9075,
+      "step": 340
+    },
+    {
+      "epoch": 0.23450586264656617,
+      "grad_norm": 0.39244529604911804,
+      "learning_rate": 0.0002,
+      "loss": 1.8128,
+      "step": 350
+    },
+    {
+      "epoch": 0.24120603015075376,
+      "grad_norm": 0.3299325704574585,
+      "learning_rate": 0.0002,
+      "loss": 1.7885,
+      "step": 360
+    },
+    {
+      "epoch": 0.24790619765494137,
+      "grad_norm": 0.3994322419166565,
+      "learning_rate": 0.0002,
+      "loss": 1.8028,
+      "step": 370
+    },
+    {
+      "epoch": 0.254606365159129,
+      "grad_norm": 0.3559151887893677,
+      "learning_rate": 0.0002,
+      "loss": 1.8321,
+      "step": 380
+    },
+    {
+      "epoch": 0.2613065326633166,
+      "grad_norm": 0.3873756229877472,
+      "learning_rate": 0.0002,
+      "loss": 1.7802,
+      "step": 390
+    },
+    {
+      "epoch": 0.26800670016750416,
+      "grad_norm": 0.3710744082927704,
+      "learning_rate": 0.0002,
+      "loss": 1.7844,
+      "step": 400
+    },
+    {
+      "epoch": 0.2747068676716918,
+      "grad_norm": 0.3618465065956116,
+      "learning_rate": 0.0002,
+      "loss": 1.7789,
+      "step": 410
+    },
+    {
+      "epoch": 0.2814070351758794,
+      "grad_norm": 0.30063769221305847,
+      "learning_rate": 0.0002,
+      "loss": 1.8529,
+      "step": 420
+    },
+    {
+      "epoch": 0.288107202680067,
+      "grad_norm": 0.3695628345012665,
+      "learning_rate": 0.0002,
+      "loss": 1.7765,
+      "step": 430
+    },
+    {
+      "epoch": 0.2948073701842546,
+      "grad_norm": 0.31451135873794556,
+      "learning_rate": 0.0002,
+      "loss": 1.7982,
+      "step": 440
+    },
+    {
+      "epoch": 0.3015075376884422,
+      "grad_norm": 0.3959707021713257,
+      "learning_rate": 0.0002,
+      "loss": 1.7517,
+      "step": 450
+    },
+    {
+      "epoch": 0.3082077051926298,
+      "grad_norm": 0.33475354313850403,
+      "learning_rate": 0.0002,
+      "loss": 1.8142,
+      "step": 460
+    },
+    {
+      "epoch": 0.3149078726968174,
+      "grad_norm": 0.33933115005493164,
+      "learning_rate": 0.0002,
+      "loss": 1.8805,
+      "step": 470
+    },
+    {
+      "epoch": 0.32160804020100503,
+      "grad_norm": 0.3264943063259125,
+      "learning_rate": 0.0002,
+      "loss": 1.7564,
+      "step": 480
+    },
+    {
+      "epoch": 0.32830820770519265,
+      "grad_norm": 0.40188100934028625,
+      "learning_rate": 0.0002,
+      "loss": 1.8428,
+      "step": 490
+    },
+    {
+      "epoch": 0.33500837520938026,
+      "grad_norm": 0.37408649921417236,
+      "learning_rate": 0.0002,
+      "loss": 1.7624,
+      "step": 500
+    },
+    {
+      "epoch": 0.3417085427135678,
+      "grad_norm": 0.33925938606262207,
+      "learning_rate": 0.0002,
+      "loss": 1.7745,
+      "step": 510
+    },
+    {
+      "epoch": 0.34840871021775544,
+      "grad_norm": 0.36836713552474976,
+      "learning_rate": 0.0002,
+      "loss": 1.814,
+      "step": 520
+    },
+    {
+      "epoch": 0.35510887772194305,
+      "grad_norm": 0.37284499406814575,
+      "learning_rate": 0.0002,
+      "loss": 1.8037,
+      "step": 530
+    },
+    {
+      "epoch": 0.36180904522613067,
+      "grad_norm": 0.3192278742790222,
+      "learning_rate": 0.0002,
+      "loss": 1.8379,
+      "step": 540
+    },
+    {
+      "epoch": 0.3685092127303183,
+      "grad_norm": 0.30233290791511536,
+      "learning_rate": 0.0002,
+      "loss": 1.8702,
+      "step": 550
+    },
+    {
+      "epoch": 0.37520938023450584,
+      "grad_norm": 0.3340817391872406,
+      "learning_rate": 0.0002,
+      "loss": 1.8799,
+      "step": 560
+    },
+    {
+      "epoch": 0.38190954773869346,
+      "grad_norm": 0.32600095868110657,
+      "learning_rate": 0.0002,
+      "loss": 1.8404,
+      "step": 570
+    },
+    {
+      "epoch": 0.38860971524288107,
+      "grad_norm": 0.33711278438568115,
+      "learning_rate": 0.0002,
+      "loss": 1.7804,
+      "step": 580
+    },
+    {
+      "epoch": 0.3953098827470687,
+      "grad_norm": 0.34890690445899963,
+      "learning_rate": 0.0002,
+      "loss": 1.8445,
+      "step": 590
+    },
+    {
+      "epoch": 0.4020100502512563,
+      "grad_norm": 0.38238924741744995,
+      "learning_rate": 0.0002,
+      "loss": 1.8187,
+      "step": 600
+    },
+    {
+      "epoch": 0.40871021775544386,
+      "grad_norm": 0.34399354457855225,
+      "learning_rate": 0.0002,
+      "loss": 1.8111,
+      "step": 610
+    },
+    {
+      "epoch": 0.4154103852596315,
+      "grad_norm": 0.3346073627471924,
+      "learning_rate": 0.0002,
+      "loss": 1.8006,
+      "step": 620
+    },
+    {
+      "epoch": 0.4221105527638191,
+      "grad_norm": 0.3545648157596588,
+      "learning_rate": 0.0002,
+      "loss": 1.7705,
+      "step": 630
+    },
+    {
+      "epoch": 0.4288107202680067,
+      "grad_norm": 0.3378899097442627,
+      "learning_rate": 0.0002,
+      "loss": 1.8445,
+      "step": 640
+    },
+    {
+      "epoch": 0.4355108877721943,
+      "grad_norm": 0.3255569040775299,
+      "learning_rate": 0.0002,
+      "loss": 1.804,
+      "step": 650
+    },
+    {
+      "epoch": 0.44221105527638194,
+      "grad_norm": 0.34880587458610535,
+      "learning_rate": 0.0002,
+      "loss": 1.7679,
+      "step": 660
+    },
+    {
+      "epoch": 0.4489112227805695,
+      "grad_norm": 0.3402383625507355,
+      "learning_rate": 0.0002,
+      "loss": 1.7861,
+      "step": 670
+    },
+    {
+      "epoch": 0.4556113902847571,
+      "grad_norm": 0.3594033718109131,
+      "learning_rate": 0.0002,
+      "loss": 1.8131,
+      "step": 680
+    },
+    {
+      "epoch": 0.4623115577889447,
+      "grad_norm": 0.31000566482543945,
+      "learning_rate": 0.0002,
+      "loss": 1.8399,
+      "step": 690
+    },
+    {
+      "epoch": 0.46901172529313234,
+      "grad_norm": 0.37229061126708984,
+      "learning_rate": 0.0002,
+      "loss": 1.7521,
+      "step": 700
+    },
+    {
+      "epoch": 0.47571189279731996,
+      "grad_norm": 0.315801739692688,
+      "learning_rate": 0.0002,
+      "loss": 1.7779,
+      "step": 710
+    },
+    {
+      "epoch": 0.4824120603015075,
+      "grad_norm": 0.3220832645893097,
+      "learning_rate": 0.0002,
+      "loss": 1.7515,
+      "step": 720
+    },
+    {
+      "epoch": 0.48911222780569513,
+      "grad_norm": 0.3435456156730652,
+      "learning_rate": 0.0002,
+      "loss": 1.7181,
+      "step": 730
+    },
+    {
+      "epoch": 0.49581239530988275,
+      "grad_norm": 0.30380892753601074,
+      "learning_rate": 0.0002,
+      "loss": 1.8844,
+      "step": 740
+    },
+    {
+      "epoch": 0.5025125628140703,
+      "grad_norm": 0.3555026054382324,
+      "learning_rate": 0.0002,
+      "loss": 1.7792,
+      "step": 750
+    },
+    {
+      "epoch": 0.509212730318258,
+      "grad_norm": 0.3019855320453644,
+      "learning_rate": 0.0002,
+      "loss": 1.7714,
+      "step": 760
+    },
+    {
+      "epoch": 0.5159128978224455,
+      "grad_norm": 0.309111088514328,
+      "learning_rate": 0.0002,
+      "loss": 1.7962,
+      "step": 770
+    },
+    {
+      "epoch": 0.5226130653266332,
+      "grad_norm": 0.366020530462265,
+      "learning_rate": 0.0002,
+      "loss": 1.7913,
+      "step": 780
+    },
+    {
+      "epoch": 0.5293132328308208,
+      "grad_norm": 0.3267050087451935,
+      "learning_rate": 0.0002,
+      "loss": 1.8008,
+      "step": 790
+    },
+    {
+      "epoch": 0.5360134003350083,
+      "grad_norm": 0.34265750646591187,
+      "learning_rate": 0.0002,
+      "loss": 1.7397,
+      "step": 800
+    },
+    {
+      "epoch": 0.542713567839196,
+      "grad_norm": 0.313669890165329,
+      "learning_rate": 0.0002,
+      "loss": 1.8251,
+      "step": 810
+    },
+    {
+      "epoch": 0.5494137353433836,
+      "grad_norm": 0.3355236053466797,
+      "learning_rate": 0.0002,
+      "loss": 1.8369,
+      "step": 820
+    },
+    {
+      "epoch": 0.5561139028475712,
+      "grad_norm": 0.3186608552932739,
+      "learning_rate": 0.0002,
+      "loss": 1.7381,
+      "step": 830
+    },
+    {
+      "epoch": 0.5628140703517588,
+      "grad_norm": 0.30357518792152405,
+      "learning_rate": 0.0002,
+      "loss": 1.8034,
+      "step": 840
+    },
+    {
+      "epoch": 0.5695142378559463,
+      "grad_norm": 0.3990040123462677,
+      "learning_rate": 0.0002,
+      "loss": 1.769,
+      "step": 850
+    },
+    {
+      "epoch": 0.576214405360134,
+      "grad_norm": 0.34363803267478943,
+      "learning_rate": 0.0002,
+      "loss": 1.7482,
+      "step": 860
+    },
+    {
+      "epoch": 0.5829145728643216,
+      "grad_norm": 0.3757908046245575,
+      "learning_rate": 0.0002,
+      "loss": 1.8106,
+      "step": 870
+    },
+    {
+      "epoch": 0.5896147403685092,
+      "grad_norm": 0.3359757661819458,
+      "learning_rate": 0.0002,
+      "loss": 1.8104,
+      "step": 880
+    },
+    {
+      "epoch": 0.5963149078726968,
+      "grad_norm": 0.5555329918861389,
+      "learning_rate": 0.0002,
+      "loss": 1.7591,
+      "step": 890
+    },
+    {
+      "epoch": 0.6030150753768844,
+      "grad_norm": 0.4046323895454407,
+      "learning_rate": 0.0002,
+      "loss": 1.7715,
+      "step": 900
+    },
+    {
+      "epoch": 0.609715242881072,
+      "grad_norm": 0.29834219813346863,
+      "learning_rate": 0.0002,
+      "loss": 1.7998,
+      "step": 910
+    },
+    {
+      "epoch": 0.6164154103852596,
+      "grad_norm": 0.3241238594055176,
+      "learning_rate": 0.0002,
+      "loss": 1.7826,
+      "step": 920
+    },
+    {
+      "epoch": 0.6231155778894473,
+      "grad_norm": 0.35154739022254944,
+      "learning_rate": 0.0002,
+      "loss": 1.8342,
+      "step": 930
+    },
+    {
+      "epoch": 0.6298157453936348,
+      "grad_norm": 0.3287706673145294,
+      "learning_rate": 0.0002,
+      "loss": 1.8076,
+      "step": 940
+    },
+    {
+      "epoch": 0.6365159128978225,
+      "grad_norm": 0.35670626163482666,
+      "learning_rate": 0.0002,
+      "loss": 1.8038,
+      "step": 950
+    },
+    {
+      "epoch": 0.6432160804020101,
+      "grad_norm": 0.6114104986190796,
+      "learning_rate": 0.0002,
+      "loss": 1.869,
+      "step": 960
+    },
+    {
+      "epoch": 0.6499162479061976,
+      "grad_norm": 0.3186565041542053,
+      "learning_rate": 0.0002,
+      "loss": 1.8297,
+      "step": 970
+    },
+    {
+      "epoch": 0.6566164154103853,
+      "grad_norm": 0.27164125442504883,
+      "learning_rate": 0.0002,
+      "loss": 1.7539,
+      "step": 980
+    },
+    {
+      "epoch": 0.6633165829145728,
+      "grad_norm": 0.34407344460487366,
+      "learning_rate": 0.0002,
+      "loss": 1.8339,
+      "step": 990
+    },
+    {
+      "epoch": 0.6700167504187605,
+      "grad_norm": 0.368415892124176,
+      "learning_rate": 0.0002,
+      "loss": 1.855,
+      "step": 1000
+    },
+    {
+      "epoch": 0.6767169179229481,
+      "grad_norm": 0.3306390643119812,
+      "learning_rate": 0.0002,
+      "loss": 1.7821,
+      "step": 1010
+    },
+    {
+      "epoch": 0.6834170854271356,
+      "grad_norm": 0.3198648989200592,
+      "learning_rate": 0.0002,
+      "loss": 1.786,
+      "step": 1020
+    },
+    {
+      "epoch": 0.6901172529313233,
+      "grad_norm": 0.3092987537384033,
+      "learning_rate": 0.0002,
+      "loss": 1.816,
+      "step": 1030
+    },
+    {
+      "epoch": 0.6968174204355109,
+      "grad_norm": 0.3090653419494629,
+      "learning_rate": 0.0002,
+      "loss": 1.7689,
+      "step": 1040
+    },
+    {
+      "epoch": 0.7035175879396985,
+      "grad_norm": 0.3485880196094513,
+      "learning_rate": 0.0002,
+      "loss": 1.7544,
+      "step": 1050
+    },
+    {
+      "epoch": 0.7102177554438861,
+      "grad_norm": 0.35782721638679504,
+      "learning_rate": 0.0002,
+      "loss": 1.8187,
+      "step": 1060
+    },
+    {
+      "epoch": 0.7169179229480737,
+      "grad_norm": 0.34256869554519653,
+      "learning_rate": 0.0002,
+      "loss": 1.806,
+      "step": 1070
+    },
+    {
+      "epoch": 0.7236180904522613,
+      "grad_norm": 0.30461037158966064,
+      "learning_rate": 0.0002,
+      "loss": 1.7873,
+      "step": 1080
+    },
+    {
+      "epoch": 0.7303182579564489,
+      "grad_norm": 0.3398691713809967,
+      "learning_rate": 0.0002,
+      "loss": 1.7367,
+      "step": 1090
+    },
+    {
+      "epoch": 0.7370184254606366,
+      "grad_norm": 0.3180808126926422,
+      "learning_rate": 0.0002,
+      "loss": 1.8756,
+      "step": 1100
+    },
+    {
+      "epoch": 0.7437185929648241,
+      "grad_norm": 0.34400665760040283,
+      "learning_rate": 0.0002,
+      "loss": 1.6988,
+      "step": 1110
+    },
+    {
+      "epoch": 0.7504187604690117,
+      "grad_norm": 0.34244877099990845,
+      "learning_rate": 0.0002,
+      "loss": 1.7851,
+      "step": 1120
+    },
+    {
+      "epoch": 0.7571189279731994,
+      "grad_norm": 0.29946693778038025,
+      "learning_rate": 0.0002,
+      "loss": 1.7841,
+      "step": 1130
+    },
+    {
+      "epoch": 0.7638190954773869,
+      "grad_norm": 0.37547236680984497,
+      "learning_rate": 0.0002,
+      "loss": 1.7456,
+      "step": 1140
+    },
+    {
+      "epoch": 0.7705192629815746,
+      "grad_norm": 0.3263005018234253,
+      "learning_rate": 0.0002,
+      "loss": 1.8425,
+      "step": 1150
+    },
+    {
+      "epoch": 0.7772194304857621,
+      "grad_norm": 0.41363608837127686,
+      "learning_rate": 0.0002,
+      "loss": 1.7222,
+      "step": 1160
+    },
+    {
+      "epoch": 0.7839195979899497,
+      "grad_norm": 0.36267954111099243,
+      "learning_rate": 0.0002,
+      "loss": 1.7836,
+      "step": 1170
+    },
+    {
+      "epoch": 0.7906197654941374,
+      "grad_norm": 0.31789499521255493,
+      "learning_rate": 0.0002,
+      "loss": 1.9183,
+      "step": 1180
+    },
+    {
+      "epoch": 0.7973199329983249,
+      "grad_norm": 0.5708149075508118,
+      "learning_rate": 0.0002,
+      "loss": 1.78,
+      "step": 1190
+    },
+    {
+      "epoch": 0.8040201005025126,
+      "grad_norm": 0.322099506855011,
+      "learning_rate": 0.0002,
+      "loss": 1.6908,
+      "step": 1200
+    },
+    {
+      "epoch": 0.8107202680067002,
+      "grad_norm": 0.3419909179210663,
+      "learning_rate": 0.0002,
+      "loss": 1.7639,
+      "step": 1210
+    },
+    {
+      "epoch": 0.8174204355108877,
+      "grad_norm": 0.36286255717277527,
+      "learning_rate": 0.0002,
+      "loss": 1.7428,
+      "step": 1220
+    },
+    {
+      "epoch": 0.8241206030150754,
+      "grad_norm": 0.33992862701416016,
+      "learning_rate": 0.0002,
+      "loss": 1.8409,
+      "step": 1230
+    },
+    {
+      "epoch": 0.830820770519263,
+      "grad_norm": 0.32622793316841125,
+      "learning_rate": 0.0002,
+      "loss": 1.7507,
+      "step": 1240
+    },
+    {
+      "epoch": 0.8375209380234506,
+      "grad_norm": 0.3036167621612549,
+      "learning_rate": 0.0002,
+      "loss": 1.8098,
+      "step": 1250
+    },
+    {
+      "epoch": 0.8442211055276382,
+      "grad_norm": 0.3182215392589569,
+      "learning_rate": 0.0002,
+      "loss": 1.8094,
+      "step": 1260
+    },
+    {
+      "epoch": 0.8509212730318257,
+      "grad_norm": 0.3270018696784973,
+      "learning_rate": 0.0002,
+      "loss": 1.8017,
+      "step": 1270
+    },
+    {
+      "epoch": 0.8576214405360134,
+      "grad_norm": 0.32652342319488525,
+      "learning_rate": 0.0002,
+      "loss": 1.798,
+      "step": 1280
+    },
+    {
+      "epoch": 0.864321608040201,
+      "grad_norm": 0.3631329834461212,
+      "learning_rate": 0.0002,
+      "loss": 1.7448,
+      "step": 1290
+    },
+    {
+      "epoch": 0.8710217755443886,
+      "grad_norm": 0.36706018447875977,
+      "learning_rate": 0.0002,
+      "loss": 1.7,
+      "step": 1300
+    },
+    {
+      "epoch": 0.8777219430485762,
+      "grad_norm": 0.3347418010234833,
+      "learning_rate": 0.0002,
+      "loss": 1.8178,
+      "step": 1310
+    },
+    {
+      "epoch": 0.8844221105527639,
+      "grad_norm": 0.34371060132980347,
+      "learning_rate": 0.0002,
+      "loss": 1.7824,
+      "step": 1320
+    },
+    {
+      "epoch": 0.8911222780569514,
+      "grad_norm": 0.3029090166091919,
+      "learning_rate": 0.0002,
+      "loss": 1.783,
+      "step": 1330
+    },
+    {
+      "epoch": 0.897822445561139,
+      "grad_norm": 0.34700682759284973,
+      "learning_rate": 0.0002,
+      "loss": 1.8017,
+      "step": 1340
+    },
+    {
+      "epoch": 0.9045226130653267,
+      "grad_norm": 0.35574328899383545,
+      "learning_rate": 0.0002,
+      "loss": 1.7998,
+      "step": 1350
+    },
+    {
+      "epoch": 0.9112227805695142,
+      "grad_norm": 0.30928221344947815,
+      "learning_rate": 0.0002,
+      "loss": 1.7339,
+      "step": 1360
+    },
+    {
+      "epoch": 0.9179229480737019,
+      "grad_norm": 0.30652928352355957,
+      "learning_rate": 0.0002,
+      "loss": 1.7479,
+      "step": 1370
+    },
+    {
+      "epoch": 0.9246231155778895,
+      "grad_norm": 0.3838157653808594,
+      "learning_rate": 0.0002,
+      "loss": 1.7491,
+      "step": 1380
+    },
+    {
+      "epoch": 0.931323283082077,
+      "grad_norm": 0.31655240058898926,
+      "learning_rate": 0.0002,
+      "loss": 1.7977,
+      "step": 1390
+    },
+    {
+      "epoch": 0.9380234505862647,
+      "grad_norm": 0.41737303137779236,
+      "learning_rate": 0.0002,
+      "loss": 1.8175,
+      "step": 1400
+    },
+    {
+      "epoch": 0.9447236180904522,
+      "grad_norm": 0.3227267861366272,
+      "learning_rate": 0.0002,
+      "loss": 1.6811,
+      "step": 1410
+    },
+    {
+      "epoch": 0.9514237855946399,
+      "grad_norm": 0.3729925751686096,
+      "learning_rate": 0.0002,
+      "loss": 1.7343,
+      "step": 1420
+    },
+    {
+      "epoch": 0.9581239530988275,
+      "grad_norm": 0.30779409408569336,
+      "learning_rate": 0.0002,
+      "loss": 1.8221,
+      "step": 1430
+    },
+    {
+      "epoch": 0.964824120603015,
+      "grad_norm": 0.334379643201828,
+      "learning_rate": 0.0002,
+      "loss": 1.7972,
+      "step": 1440
+    },
+    {
+      "epoch": 0.9715242881072027,
+      "grad_norm": 0.3568236231803894,
+      "learning_rate": 0.0002,
+      "loss": 1.7141,
+      "step": 1450
+    },
+    {
+      "epoch": 0.9782244556113903,
+      "grad_norm": 0.33310577273368835,
+      "learning_rate": 0.0002,
+      "loss": 1.7541,
+      "step": 1460
+    },
+    {
+      "epoch": 0.9849246231155779,
+      "grad_norm": 0.2972261905670166,
+      "learning_rate": 0.0002,
+      "loss": 1.8511,
+      "step": 1470
+    },
+    {
+      "epoch": 0.9916247906197655,
+      "grad_norm": 0.3322717845439911,
+      "learning_rate": 0.0002,
+      "loss": 1.7654,
+      "step": 1480
+    },
+    {
+      "epoch": 0.998324958123953,
+      "grad_norm": 0.3276330828666687,
+      "learning_rate": 0.0002,
+      "loss": 1.8033,
+      "step": 1490
+    },
+    {
+      "epoch": 0.9996649916247906,
+      "eval_loss": 1.8036354780197144,
+      "eval_runtime": 37.8949,
+      "eval_samples_per_second": 13.59,
+      "eval_steps_per_second": 1.715,
+      "step": 1492
+    },
+    {
+      "epoch": 1.0050251256281406,
+      "grad_norm": 0.29252371191978455,
+      "learning_rate": 0.0002,
+      "loss": 1.7138,
+      "step": 1500
+    },
+    {
+      "epoch": 1.0117252931323284,
+      "grad_norm": 0.31607162952423096,
+      "learning_rate": 0.0002,
+      "loss": 1.8198,
+      "step": 1510
+    },
+    {
+      "epoch": 1.018425460636516,
+      "grad_norm": 0.32294467091560364,
+      "learning_rate": 0.0002,
+      "loss": 1.6779,
+      "step": 1520
+    },
+    {
+      "epoch": 1.0251256281407035,
+      "grad_norm": 0.3868017792701721,
+      "learning_rate": 0.0002,
+      "loss": 1.7919,
+      "step": 1530
+    },
+    {
+      "epoch": 1.031825795644891,
+      "grad_norm": 0.3178282082080841,
+      "learning_rate": 0.0002,
+      "loss": 1.7954,
+      "step": 1540
+    },
+    {
+      "epoch": 1.0385259631490786,
+      "grad_norm": 0.3706750273704529,
+      "learning_rate": 0.0002,
+      "loss": 1.7136,
+      "step": 1550
+    },
+    {
+      "epoch": 1.0452261306532664,
+      "grad_norm": 0.33930912613868713,
+      "learning_rate": 0.0002,
+      "loss": 1.7382,
+      "step": 1560
+    },
+    {
+      "epoch": 1.051926298157454,
+      "grad_norm": 0.33970504999160767,
+      "learning_rate": 0.0002,
+      "loss": 1.7602,
+      "step": 1570
+    },
+    {
+      "epoch": 1.0586264656616415,
+      "grad_norm": 0.42553383111953735,
+      "learning_rate": 0.0002,
+      "loss": 1.6573,
+      "step": 1580
+    },
+    {
+      "epoch": 1.065326633165829,
+      "grad_norm": 0.3772421181201935,
+      "learning_rate": 0.0002,
+      "loss": 1.645,
+      "step": 1590
+    },
+    {
+      "epoch": 1.0720268006700167,
+      "grad_norm": 0.34212902188301086,
+      "learning_rate": 0.0002,
+      "loss": 1.7362,
+      "step": 1600
+    },
+    {
+      "epoch": 1.0787269681742044,
+      "grad_norm": 0.3798283338546753,
+      "learning_rate": 0.0002,
+      "loss": 1.7057,
+      "step": 1610
+    },
+    {
+      "epoch": 1.085427135678392,
+      "grad_norm": 0.36909598112106323,
+      "learning_rate": 0.0002,
+      "loss": 1.7468,
+      "step": 1620
+    },
+    {
+      "epoch": 1.0921273031825796,
+      "grad_norm": 0.3344230651855469,
+      "learning_rate": 0.0002,
+      "loss": 1.7807,
+      "step": 1630
+    },
+    {
+      "epoch": 1.0988274706867671,
+      "grad_norm": 0.3862569332122803,
+      "learning_rate": 0.0002,
+      "loss": 1.7111,
+      "step": 1640
+    },
+    {
+      "epoch": 1.1055276381909547,
+      "grad_norm": 0.31188511848449707,
+      "learning_rate": 0.0002,
+      "loss": 1.7163,
+      "step": 1650
+    },
+    {
+      "epoch": 1.1122278056951425,
+      "grad_norm": 0.3563670814037323,
+      "learning_rate": 0.0002,
+      "loss": 1.7263,
+      "step": 1660
+    },
+    {
+      "epoch": 1.11892797319933,
+      "grad_norm": 0.35052165389060974,
+      "learning_rate": 0.0002,
+      "loss": 1.7718,
+      "step": 1670
+    },
+    {
+      "epoch": 1.1256281407035176,
+      "grad_norm": 0.3285699188709259,
+      "learning_rate": 0.0002,
+      "loss": 1.7601,
+      "step": 1680
+    },
+    {
+      "epoch": 1.1323283082077051,
+      "grad_norm": 0.3639393746852875,
+      "learning_rate": 0.0002,
+      "loss": 1.6877,
+      "step": 1690
+    },
+    {
+      "epoch": 1.1390284757118927,
+      "grad_norm": 0.3842753767967224,
+      "learning_rate": 0.0002,
+      "loss": 1.7719,
+      "step": 1700
+    },
+    {
+      "epoch": 1.1457286432160805,
+      "grad_norm": 0.3624933063983917,
+      "learning_rate": 0.0002,
+      "loss": 1.7002,
+      "step": 1710
+    },
+    {
+      "epoch": 1.152428810720268,
+      "grad_norm": 0.3641220033168793,
+      "learning_rate": 0.0002,
+      "loss": 1.7243,
+      "step": 1720
+    },
+    {
+      "epoch": 1.1591289782244556,
+      "grad_norm": 0.32765355706214905,
+      "learning_rate": 0.0002,
+      "loss": 1.752,
+      "step": 1730
+    },
+    {
+      "epoch": 1.1658291457286432,
+      "grad_norm": 0.34974896907806396,
+      "learning_rate": 0.0002,
+      "loss": 1.6556,
+      "step": 1740
+    },
+    {
+      "epoch": 1.1725293132328307,
+      "grad_norm": 0.3910926580429077,
+      "learning_rate": 0.0002,
+      "loss": 1.7273,
+      "step": 1750
+    },
+    {
+      "epoch": 1.1792294807370185,
+      "grad_norm": 0.3564300537109375,
+      "learning_rate": 0.0002,
+      "loss": 1.7204,
+      "step": 1760
+    },
+    {
+      "epoch": 1.185929648241206,
+      "grad_norm": 0.34822574257850647,
+      "learning_rate": 0.0002,
+      "loss": 1.746,
+      "step": 1770
+    },
+    {
+      "epoch": 1.1926298157453936,
+      "grad_norm": 0.36185044050216675,
+      "learning_rate": 0.0002,
+      "loss": 1.7256,
+      "step": 1780
+    },
+    {
+      "epoch": 1.1993299832495812,
+      "grad_norm": 0.34866711497306824,
+      "learning_rate": 0.0002,
+      "loss": 1.6431,
+      "step": 1790
+    },
+    {
+      "epoch": 1.2060301507537687,
+      "grad_norm": 0.4017769992351532,
+      "learning_rate": 0.0002,
+      "loss": 1.8084,
+      "step": 1800
+    },
+    {
+      "epoch": 1.2127303182579565,
+      "grad_norm": 0.32930681109428406,
+      "learning_rate": 0.0002,
+      "loss": 1.6985,
+      "step": 1810
+    },
+    {
+      "epoch": 1.219430485762144,
+      "grad_norm": 0.35951921343803406,
+      "learning_rate": 0.0002,
+      "loss": 1.7606,
+      "step": 1820
+    },
+    {
+      "epoch": 1.2261306532663316,
+      "grad_norm": 0.37366992235183716,
+      "learning_rate": 0.0002,
+      "loss": 1.6933,
+      "step": 1830
+    },
+    {
+      "epoch": 1.2328308207705192,
+      "grad_norm": 0.3565689027309418,
+      "learning_rate": 0.0002,
+      "loss": 1.6737,
+      "step": 1840
+    },
+    {
+      "epoch": 1.2395309882747068,
+      "grad_norm": 0.3692343533039093,
+      "learning_rate": 0.0002,
+      "loss": 1.8013,
+      "step": 1850
+    },
+    {
+      "epoch": 1.2462311557788945,
+      "grad_norm": 0.38426971435546875,
+      "learning_rate": 0.0002,
+      "loss": 1.736,
+      "step": 1860
+    },
+    {
+      "epoch": 1.252931323283082,
+      "grad_norm": 0.33559855818748474,
+      "learning_rate": 0.0002,
+      "loss": 1.7031,
+      "step": 1870
+    },
+    {
+      "epoch": 1.2596314907872697,
+      "grad_norm": 0.34181106090545654,
+      "learning_rate": 0.0002,
+      "loss": 1.7033,
+      "step": 1880
+    },
+    {
+      "epoch": 1.2663316582914572,
+      "grad_norm": 0.3916318416595459,
+      "learning_rate": 0.0002,
+      "loss": 1.7707,
+      "step": 1890
+    },
+    {
+      "epoch": 1.2730318257956448,
+      "grad_norm": 0.3887825012207031,
+      "learning_rate": 0.0002,
+      "loss": 1.6686,
+      "step": 1900
+    },
+    {
+      "epoch": 1.2797319932998326,
+      "grad_norm": 0.33583927154541016,
+      "learning_rate": 0.0002,
+      "loss": 1.7062,
+      "step": 1910
+    },
+    {
+      "epoch": 1.2864321608040201,
+      "grad_norm": 0.37639349699020386,
+      "learning_rate": 0.0002,
+      "loss": 1.717,
+      "step": 1920
+    },
+    {
+      "epoch": 1.2931323283082077,
+      "grad_norm": 0.38059428334236145,
+      "learning_rate": 0.0002,
+      "loss": 1.777,
+      "step": 1930
+    },
+    {
+      "epoch": 1.2998324958123952,
+      "grad_norm": 0.37253183126449585,
+      "learning_rate": 0.0002,
+      "loss": 1.6126,
+      "step": 1940
+    },
+    {
+      "epoch": 1.3065326633165828,
+      "grad_norm": 0.37371566891670227,
+      "learning_rate": 0.0002,
+      "loss": 1.6758,
+      "step": 1950
+    },
+    {
+      "epoch": 1.3132328308207706,
+      "grad_norm": 0.4080910086631775,
+      "learning_rate": 0.0002,
+      "loss": 1.6788,
+      "step": 1960
+    },
+    {
+      "epoch": 1.3199329983249581,
+      "grad_norm": 0.3174354135990143,
+      "learning_rate": 0.0002,
+      "loss": 1.6518,
+      "step": 1970
+    },
+    {
+      "epoch": 1.3266331658291457,
+      "grad_norm": 0.4518888294696808,
+      "learning_rate": 0.0002,
+      "loss": 1.7925,
+      "step": 1980
+    },
+    {
+      "epoch": 1.3333333333333333,
+      "grad_norm": 0.3627921938896179,
+      "learning_rate": 0.0002,
+      "loss": 1.7085,
+      "step": 1990
+    },
+    {
+      "epoch": 1.3400335008375208,
+      "grad_norm": 0.3655930161476135,
+      "learning_rate": 0.0002,
+      "loss": 1.7676,
+      "step": 2000
+    },
+    {
+      "epoch": 1.3467336683417086,
+      "grad_norm": 0.3509993255138397,
+      "learning_rate": 0.0002,
+      "loss": 1.7016,
+      "step": 2010
+    },
+    {
+      "epoch": 1.3534338358458962,
+      "grad_norm": 0.4281129240989685,
+      "learning_rate": 0.0002,
+      "loss": 1.7359,
+      "step": 2020
+    },
+    {
+      "epoch": 1.3601340033500837,
+      "grad_norm": 0.3821414113044739,
+      "learning_rate": 0.0002,
+      "loss": 1.6884,
+      "step": 2030
+    },
+    {
+      "epoch": 1.3668341708542713,
+      "grad_norm": 0.3907586336135864,
+      "learning_rate": 0.0002,
+      "loss": 1.7075,
+      "step": 2040
+    },
+    {
+      "epoch": 1.3735343383584588,
+      "grad_norm": 0.37792932987213135,
+      "learning_rate": 0.0002,
+      "loss": 1.7424,
+      "step": 2050
+    },
+    {
+      "epoch": 1.3802345058626466,
+      "grad_norm": 0.3693985641002655,
+      "learning_rate": 0.0002,
+      "loss": 1.7305,
+      "step": 2060
+    },
+    {
+      "epoch": 1.3869346733668342,
+      "grad_norm": 0.32275936007499695,
+      "learning_rate": 0.0002,
+      "loss": 1.7434,
+      "step": 2070
+    },
+    {
+      "epoch": 1.3936348408710217,
+      "grad_norm": 0.3789440095424652,
+      "learning_rate": 0.0002,
+      "loss": 1.6677,
+      "step": 2080
+    },
+    {
+      "epoch": 1.4003350083752093,
+      "grad_norm": 0.3638380467891693,
+      "learning_rate": 0.0002,
+      "loss": 1.6825,
+      "step": 2090
+    },
+    {
+      "epoch": 1.4070351758793969,
+      "grad_norm": 0.3495481610298157,
+      "learning_rate": 0.0002,
+      "loss": 1.6542,
+      "step": 2100
+    },
+    {
+      "epoch": 1.4137353433835846,
+      "grad_norm": 0.37920597195625305,
+      "learning_rate": 0.0002,
+      "loss": 1.7225,
+      "step": 2110
+    },
+    {
+      "epoch": 1.4204355108877722,
+      "grad_norm": 0.37218064069747925,
+      "learning_rate": 0.0002,
+      "loss": 1.7329,
+      "step": 2120
+    },
+    {
+      "epoch": 1.4271356783919598,
+      "grad_norm": 0.38074082136154175,
+      "learning_rate": 0.0002,
+      "loss": 1.799,
+      "step": 2130
+    },
+    {
+      "epoch": 1.4338358458961473,
+      "grad_norm": 0.3455527126789093,
+      "learning_rate": 0.0002,
+      "loss": 1.7403,
+      "step": 2140
+    },
+    {
+      "epoch": 1.4405360134003349,
+      "grad_norm": 0.3712003529071808,
+      "learning_rate": 0.0002,
+      "loss": 1.776,
+      "step": 2150
+    },
+    {
+      "epoch": 1.4472361809045227,
+      "grad_norm": 0.3786754906177521,
+      "learning_rate": 0.0002,
+      "loss": 1.7619,
+      "step": 2160
+    },
+    {
+      "epoch": 1.4539363484087102,
+      "grad_norm": 0.3879223167896271,
+      "learning_rate": 0.0002,
+      "loss": 1.68,
+      "step": 2170
+    },
+    {
+      "epoch": 1.4606365159128978,
+      "grad_norm": 0.38738805055618286,
+      "learning_rate": 0.0002,
+      "loss": 1.7,
+      "step": 2180
+    },
+    {
+      "epoch": 1.4673366834170856,
+      "grad_norm": 0.39768800139427185,
+      "learning_rate": 0.0002,
+      "loss": 1.7581,
+      "step": 2190
+    },
+    {
+      "epoch": 1.474036850921273,
+      "grad_norm": 0.4172441065311432,
+      "learning_rate": 0.0002,
+      "loss": 1.7671,
+      "step": 2200
+    },
+    {
+      "epoch": 1.4807370184254607,
+      "grad_norm": 0.4043174982070923,
+      "learning_rate": 0.0002,
+      "loss": 1.6736,
+      "step": 2210
+    },
+    {
+      "epoch": 1.4874371859296482,
+      "grad_norm": 0.3750883936882019,
+      "learning_rate": 0.0002,
+      "loss": 1.7444,
+      "step": 2220
+    },
+    {
+      "epoch": 1.4941373534338358,
+      "grad_norm": 0.3552253246307373,
+      "learning_rate": 0.0002,
+      "loss": 1.6861,
+      "step": 2230
+    },
+    {
+      "epoch": 1.5008375209380236,
+      "grad_norm": 0.34607139229774475,
+      "learning_rate": 0.0002,
+      "loss": 1.6471,
+      "step": 2240
+    },
+    {
+      "epoch": 1.507537688442211,
+      "grad_norm": 0.3406706750392914,
+      "learning_rate": 0.0002,
+      "loss": 1.6962,
+      "step": 2250
+    },
+    {
+      "epoch": 1.5142378559463987,
+      "grad_norm": 0.36654895544052124,
+      "learning_rate": 0.0002,
+      "loss": 1.7694,
+      "step": 2260
+    },
+    {
+      "epoch": 1.5209380234505863,
+      "grad_norm": 0.3914054334163666,
+      "learning_rate": 0.0002,
+      "loss": 1.6812,
+      "step": 2270
+    },
+    {
+      "epoch": 1.5276381909547738,
+      "grad_norm": 0.42012137174606323,
+      "learning_rate": 0.0002,
+      "loss": 1.6822,
+      "step": 2280
+    },
+    {
+      "epoch": 1.5343383584589616,
+      "grad_norm": 0.39563435316085815,
+      "learning_rate": 0.0002,
+      "loss": 1.697,
+      "step": 2290
+    },
+    {
+      "epoch": 1.541038525963149,
+      "grad_norm": 0.3508438766002655,
+      "learning_rate": 0.0002,
+      "loss": 1.7491,
+      "step": 2300
+    },
+    {
+      "epoch": 1.5477386934673367,
+      "grad_norm": 0.3785218596458435,
+      "learning_rate": 0.0002,
+      "loss": 1.7727,
+      "step": 2310
+    },
+    {
+      "epoch": 1.5544388609715243,
+      "grad_norm": 0.39377647638320923,
+      "learning_rate": 0.0002,
+      "loss": 1.6963,
+      "step": 2320
+    },
+    {
+      "epoch": 1.5611390284757118,
+      "grad_norm": 0.3391438126564026,
+      "learning_rate": 0.0002,
+      "loss": 1.7263,
+      "step": 2330
+    },
+    {
+      "epoch": 1.5678391959798996,
+      "grad_norm": 0.37944263219833374,
+      "learning_rate": 0.0002,
+      "loss": 1.7722,
+      "step": 2340
+    },
+    {
+      "epoch": 1.574539363484087,
+      "grad_norm": 0.3523491322994232,
+      "learning_rate": 0.0002,
+      "loss": 1.6371,
+      "step": 2350
+    },
+    {
+      "epoch": 1.5812395309882747,
+      "grad_norm": 0.3911575973033905,
+      "learning_rate": 0.0002,
+      "loss": 1.7583,
+      "step": 2360
+    },
+    {
+      "epoch": 1.5879396984924623,
+      "grad_norm": 0.33832186460494995,
+      "learning_rate": 0.0002,
+      "loss": 1.7117,
+      "step": 2370
+    },
+    {
+      "epoch": 1.5946398659966499,
+      "grad_norm": 0.3665979206562042,
+      "learning_rate": 0.0002,
+      "loss": 1.7701,
+      "step": 2380
+    },
+    {
+      "epoch": 1.6013400335008376,
+      "grad_norm": 0.3871748149394989,
+      "learning_rate": 0.0002,
+      "loss": 1.779,
+      "step": 2390
+    },
+    {
+      "epoch": 1.608040201005025,
+      "grad_norm": 0.3586967885494232,
+      "learning_rate": 0.0002,
+      "loss": 1.7109,
+      "step": 2400
+    },
+    {
+      "epoch": 1.6147403685092128,
+      "grad_norm": 0.3563673198223114,
+      "learning_rate": 0.0002,
+      "loss": 1.7096,
+      "step": 2410
+    },
+    {
+      "epoch": 1.6214405360134003,
+      "grad_norm": 0.37588971853256226,
+      "learning_rate": 0.0002,
+      "loss": 1.745,
+      "step": 2420
+    },
+    {
+      "epoch": 1.6281407035175879,
+      "grad_norm": 0.352556437253952,
+      "learning_rate": 0.0002,
+      "loss": 1.7086,
+      "step": 2430
+    },
+    {
+      "epoch": 1.6348408710217757,
+      "grad_norm": 0.3716259300708771,
+      "learning_rate": 0.0002,
+      "loss": 1.6547,
+      "step": 2440
+    },
+    {
+      "epoch": 1.641541038525963,
+      "grad_norm": 0.372001975774765,
+      "learning_rate": 0.0002,
+      "loss": 1.7033,
+      "step": 2450
+    },
+    {
+      "epoch": 1.6482412060301508,
+      "grad_norm": 0.3430042862892151,
+      "learning_rate": 0.0002,
+      "loss": 1.6584,
+      "step": 2460
+    },
+    {
+      "epoch": 1.6549413735343383,
+      "grad_norm": 0.3741483688354492,
+      "learning_rate": 0.0002,
+      "loss": 1.7217,
+      "step": 2470
+    },
+    {
+      "epoch": 1.661641541038526,
+      "grad_norm": 0.3610571324825287,
+      "learning_rate": 0.0002,
+      "loss": 1.7701,
+      "step": 2480
+    },
+    {
+      "epoch": 1.6683417085427137,
+      "grad_norm": 0.4204719066619873,
+      "learning_rate": 0.0002,
+      "loss": 1.7057,
+      "step": 2490
+    },
+    {
+      "epoch": 1.675041876046901,
+      "grad_norm": 0.3938186466693878,
+      "learning_rate": 0.0002,
+      "loss": 1.7954,
+      "step": 2500
+    },
+    {
+      "epoch": 1.6817420435510888,
+      "grad_norm": 0.3421435058116913,
+      "learning_rate": 0.0002,
+      "loss": 1.6633,
+      "step": 2510
+    },
+    {
+      "epoch": 1.6884422110552764,
+      "grad_norm": 0.42441412806510925,
+      "learning_rate": 0.0002,
+      "loss": 1.7996,
+      "step": 2520
+    },
+    {
+      "epoch": 1.695142378559464,
+      "grad_norm": 0.38071519136428833,
+      "learning_rate": 0.0002,
+      "loss": 1.7142,
+      "step": 2530
+    },
+    {
+      "epoch": 1.7018425460636517,
+      "grad_norm": 0.34078919887542725,
+      "learning_rate": 0.0002,
+      "loss": 1.7232,
+      "step": 2540
+    },
+    {
+      "epoch": 1.708542713567839,
+      "grad_norm": 0.412844181060791,
+      "learning_rate": 0.0002,
+      "loss": 1.7126,
+      "step": 2550
+    },
+    {
+      "epoch": 1.7152428810720268,
+      "grad_norm": 0.3753604292869568,
+      "learning_rate": 0.0002,
+      "loss": 1.7149,
+      "step": 2560
+    },
+    {
+      "epoch": 1.7219430485762144,
+      "grad_norm": 0.41588476300239563,
+      "learning_rate": 0.0002,
+      "loss": 1.7011,
+      "step": 2570
+    },
+    {
+      "epoch": 1.728643216080402,
+      "grad_norm": 0.35504111647605896,
+      "learning_rate": 0.0002,
+      "loss": 1.6427,
+      "step": 2580
+    },
+    {
+      "epoch": 1.7353433835845897,
+      "grad_norm": 0.36909720301628113,
+      "learning_rate": 0.0002,
+      "loss": 1.7296,
+      "step": 2590
+    },
+    {
+      "epoch": 1.742043551088777,
+      "grad_norm": 0.4149979054927826,
+      "learning_rate": 0.0002,
+      "loss": 1.7022,
+      "step": 2600
+    },
+    {
+      "epoch": 1.7487437185929648,
+      "grad_norm": 0.38859328627586365,
+      "learning_rate": 0.0002,
+      "loss": 1.77,
+      "step": 2610
+    },
+    {
+      "epoch": 1.7554438860971524,
+      "grad_norm": 0.36738792061805725,
+      "learning_rate": 0.0002,
+      "loss": 1.7036,
+      "step": 2620
+    },
+    {
+      "epoch": 1.76214405360134,
+      "grad_norm": 0.3968178927898407,
+      "learning_rate": 0.0002,
+      "loss": 1.764,
+      "step": 2630
+    },
+    {
+      "epoch": 1.7688442211055277,
+      "grad_norm": 0.3972901999950409,
+      "learning_rate": 0.0002,
+      "loss": 1.7687,
+      "step": 2640
+    },
+    {
+      "epoch": 1.775544388609715,
+      "grad_norm": 0.3949959874153137,
+      "learning_rate": 0.0002,
+      "loss": 1.6774,
+      "step": 2650
+    },
+    {
+      "epoch": 1.7822445561139029,
+      "grad_norm": 0.44074657559394836,
+      "learning_rate": 0.0002,
+      "loss": 1.7247,
+      "step": 2660
+    },
+    {
+      "epoch": 1.7889447236180904,
+      "grad_norm": 0.39743664860725403,
+      "learning_rate": 0.0002,
+      "loss": 1.7188,
+      "step": 2670
+    },
+    {
+      "epoch": 1.795644891122278,
+      "grad_norm": 0.3950406610965729,
+      "learning_rate": 0.0002,
+      "loss": 1.7258,
+      "step": 2680
+    },
+    {
+      "epoch": 1.8023450586264658,
+      "grad_norm": 0.3568263649940491,
+      "learning_rate": 0.0002,
+      "loss": 1.6906,
+      "step": 2690
+    },
+    {
+      "epoch": 1.809045226130653,
+      "grad_norm": 0.3819476366043091,
+      "learning_rate": 0.0002,
+      "loss": 1.6735,
+      "step": 2700
+    },
+    {
+      "epoch": 1.8157453936348409,
+      "grad_norm": 0.3480634391307831,
+      "learning_rate": 0.0002,
+      "loss": 1.7198,
+      "step": 2710
+    },
+    {
+      "epoch": 1.8224455611390284,
+      "grad_norm": 0.3875853419303894,
+      "learning_rate": 0.0002,
+      "loss": 1.7042,
+      "step": 2720
+    },
+    {
+      "epoch": 1.829145728643216,
+      "grad_norm": 0.3441337049007416,
+      "learning_rate": 0.0002,
+      "loss": 1.6988,
+      "step": 2730
+    },
+    {
+      "epoch": 1.8358458961474038,
+      "grad_norm": 0.35692882537841797,
+      "learning_rate": 0.0002,
+      "loss": 1.7647,
+      "step": 2740
+    },
+    {
+      "epoch": 1.8425460636515911,
+      "grad_norm": 0.36959215998649597,
+      "learning_rate": 0.0002,
+      "loss": 1.7033,
+      "step": 2750
+    },
+    {
+      "epoch": 1.849246231155779,
+      "grad_norm": 0.3893393278121948,
+      "learning_rate": 0.0002,
+      "loss": 1.7657,
+      "step": 2760
+    },
+    {
+      "epoch": 1.8559463986599665,
+      "grad_norm": 0.37817293405532837,
+      "learning_rate": 0.0002,
+      "loss": 1.7068,
+      "step": 2770
+    },
+    {
+      "epoch": 1.862646566164154,
+      "grad_norm": 0.36071285605430603,
+      "learning_rate": 0.0002,
+      "loss": 1.761,
+      "step": 2780
+    },
+    {
+      "epoch": 1.8693467336683418,
+      "grad_norm": 0.3758420944213867,
+      "learning_rate": 0.0002,
+      "loss": 1.7623,
+      "step": 2790
+    },
+    {
+      "epoch": 1.8760469011725294,
+      "grad_norm": 0.3889938294887543,
+      "learning_rate": 0.0002,
+      "loss": 1.6743,
+      "step": 2800
+    },
+    {
+      "epoch": 1.882747068676717,
+      "grad_norm": 0.34361857175827026,
+      "learning_rate": 0.0002,
+      "loss": 1.6151,
+      "step": 2810
+    },
+    {
+      "epoch": 1.8894472361809045,
+      "grad_norm": 0.39283323287963867,
+      "learning_rate": 0.0002,
+      "loss": 1.6038,
+      "step": 2820
+    },
+    {
+      "epoch": 1.896147403685092,
+      "grad_norm": 0.3919452726840973,
+      "learning_rate": 0.0002,
+      "loss": 1.7555,
+      "step": 2830
+    },
+    {
+      "epoch": 1.9028475711892798,
+      "grad_norm": 0.38215070962905884,
+      "learning_rate": 0.0002,
+      "loss": 1.673,
+      "step": 2840
+    },
+    {
+      "epoch": 1.9095477386934674,
+      "grad_norm": 0.4235064387321472,
+      "learning_rate": 0.0002,
+      "loss": 1.7044,
+      "step": 2850
+    },
+    {
+      "epoch": 1.916247906197655,
+      "grad_norm": 0.35694634914398193,
+      "learning_rate": 0.0002,
+      "loss": 1.7123,
+      "step": 2860
+    },
+    {
+      "epoch": 1.9229480737018425,
+      "grad_norm": 0.383492112159729,
+      "learning_rate": 0.0002,
+      "loss": 1.8128,
+      "step": 2870
+    },
+    {
+      "epoch": 1.92964824120603,
+      "grad_norm": 0.5945147275924683,
+      "learning_rate": 0.0002,
+      "loss": 1.7581,
+      "step": 2880
+    },
+    {
+      "epoch": 1.9363484087102178,
+      "grad_norm": 0.3367522358894348,
+      "learning_rate": 0.0002,
+      "loss": 1.7421,
+      "step": 2890
+    },
+    {
+      "epoch": 1.9430485762144054,
+      "grad_norm": 0.35300394892692566,
+      "learning_rate": 0.0002,
+      "loss": 1.6561,
+      "step": 2900
+    },
+    {
+      "epoch": 1.949748743718593,
+      "grad_norm": 0.38084495067596436,
+      "learning_rate": 0.0002,
+      "loss": 1.7033,
+      "step": 2910
+    },
+    {
+      "epoch": 1.9564489112227805,
+      "grad_norm": 0.37559160590171814,
+      "learning_rate": 0.0002,
+      "loss": 1.7132,
+      "step": 2920
+    },
+    {
+      "epoch": 1.963149078726968,
+      "grad_norm": 0.3661738336086273,
+      "learning_rate": 0.0002,
+      "loss": 1.6759,
+      "step": 2930
+    },
+    {
+      "epoch": 1.9698492462311559,
+      "grad_norm": 0.4073849320411682,
+      "learning_rate": 0.0002,
+      "loss": 1.7643,
+      "step": 2940
+    },
+    {
+      "epoch": 1.9765494137353434,
+      "grad_norm": 0.3723304271697998,
+      "learning_rate": 0.0002,
+      "loss": 1.6806,
+      "step": 2950
+    },
+    {
+      "epoch": 1.983249581239531,
+      "grad_norm": 0.3991098999977112,
+      "learning_rate": 0.0002,
+      "loss": 1.7611,
+      "step": 2960
+    },
+    {
+      "epoch": 1.9899497487437185,
+      "grad_norm": 0.3947085440158844,
+      "learning_rate": 0.0002,
+      "loss": 1.7263,
+      "step": 2970
+    },
+    {
+      "epoch": 1.996649916247906,
+      "grad_norm": 0.3786258399486542,
+      "learning_rate": 0.0002,
+      "loss": 1.7217,
+      "step": 2980
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 1.8028968572616577,
+      "eval_runtime": 37.8985,
+      "eval_samples_per_second": 13.589,
+      "eval_steps_per_second": 1.715,
+      "step": 2985
+    },
+    {
+      "epoch": 2.003350083752094,
+      "grad_norm": 0.34824079275131226,
+      "learning_rate": 0.0002,
+      "loss": 1.695,
+      "step": 2990
+    },
+    {
+      "epoch": 2.0100502512562812,
+      "grad_norm": 0.3394894003868103,
+      "learning_rate": 0.0002,
+      "loss": 1.5853,
+      "step": 3000
+    },
+    {
+      "epoch": 2.016750418760469,
+      "grad_norm": 0.36910977959632874,
+      "learning_rate": 0.0002,
+      "loss": 1.5783,
+      "step": 3010
+    },
+    {
+      "epoch": 2.023450586264657,
+      "grad_norm": 0.45000967383384705,
+      "learning_rate": 0.0002,
+      "loss": 1.6105,
+      "step": 3020
+    },
+    {
+      "epoch": 2.030150753768844,
+      "grad_norm": 0.3791407346725464,
+      "learning_rate": 0.0002,
+      "loss": 1.6019,
+      "step": 3030
+    },
+    {
+      "epoch": 2.036850921273032,
+      "grad_norm": 0.387321799993515,
+      "learning_rate": 0.0002,
+      "loss": 1.5832,
+      "step": 3040
+    },
+    {
+      "epoch": 2.0435510887772192,
+      "grad_norm": 0.4185757040977478,
+      "learning_rate": 0.0002,
+      "loss": 1.6834,
+      "step": 3050
+    },
+    {
+      "epoch": 2.050251256281407,
+      "grad_norm": 0.45110777020454407,
+      "learning_rate": 0.0002,
+      "loss": 1.5696,
+      "step": 3060
+    },
+    {
+      "epoch": 2.056951423785595,
+      "grad_norm": 0.42663660645484924,
+      "learning_rate": 0.0002,
+      "loss": 1.6231,
+      "step": 3070
+    },
+    {
+      "epoch": 2.063651591289782,
+      "grad_norm": 0.4546292722225189,
+      "learning_rate": 0.0002,
+      "loss": 1.6279,
+      "step": 3080
+    },
+    {
+      "epoch": 2.07035175879397,
+      "grad_norm": 0.3979759216308594,
+      "learning_rate": 0.0002,
+      "loss": 1.6141,
+      "step": 3090
+    },
+    {
+      "epoch": 2.0770519262981573,
+      "grad_norm": 0.43596673011779785,
+      "learning_rate": 0.0002,
+      "loss": 1.6343,
+      "step": 3100
+    },
+    {
+      "epoch": 2.083752093802345,
+      "grad_norm": 0.40120232105255127,
+      "learning_rate": 0.0002,
+      "loss": 1.5441,
+      "step": 3110
+    },
+    {
+      "epoch": 2.090452261306533,
+      "grad_norm": 0.44449281692504883,
+      "learning_rate": 0.0002,
+      "loss": 1.6309,
+      "step": 3120
+    },
+    {
+      "epoch": 2.09715242881072,
+      "grad_norm": 0.42672568559646606,
+      "learning_rate": 0.0002,
+      "loss": 1.5652,
+      "step": 3130
+    },
+    {
+      "epoch": 2.103852596314908,
+      "grad_norm": 0.4232690930366516,
+      "learning_rate": 0.0002,
+      "loss": 1.682,
+      "step": 3140
+    },
+    {
+      "epoch": 2.1105527638190953,
+      "grad_norm": 0.4299317002296448,
+      "learning_rate": 0.0002,
+      "loss": 1.624,
+      "step": 3150
+    },
+    {
+      "epoch": 2.117252931323283,
+      "grad_norm": 0.4067758023738861,
+      "learning_rate": 0.0002,
+      "loss": 1.6766,
+      "step": 3160
+    },
+    {
+      "epoch": 2.123953098827471,
+      "grad_norm": 0.4918815791606903,
+      "learning_rate": 0.0002,
+      "loss": 1.6759,
+      "step": 3170
+    },
+    {
+      "epoch": 2.130653266331658,
+      "grad_norm": 0.4140559732913971,
+      "learning_rate": 0.0002,
+      "loss": 1.6478,
+      "step": 3180
+    },
+    {
+      "epoch": 2.137353433835846,
+      "grad_norm": 0.4555995464324951,
+      "learning_rate": 0.0002,
+      "loss": 1.6641,
+      "step": 3190
+    },
+    {
+      "epoch": 2.1440536013400333,
+      "grad_norm": 0.42943915724754333,
+      "learning_rate": 0.0002,
+      "loss": 1.5888,
+      "step": 3200
+    },
+    {
+      "epoch": 2.150753768844221,
+      "grad_norm": 0.4730435013771057,
+      "learning_rate": 0.0002,
+      "loss": 1.5886,
+      "step": 3210
+    },
+    {
+      "epoch": 2.157453936348409,
+      "grad_norm": 0.43310216069221497,
+      "learning_rate": 0.0002,
+      "loss": 1.6022,
+      "step": 3220
+    },
+    {
+      "epoch": 2.164154103852596,
+      "grad_norm": 0.42054110765457153,
+      "learning_rate": 0.0002,
+      "loss": 1.6058,
+      "step": 3230
+    },
+    {
+      "epoch": 2.170854271356784,
+      "grad_norm": 0.4897233247756958,
+      "learning_rate": 0.0002,
+      "loss": 1.6749,
+      "step": 3240
+    },
+    {
+      "epoch": 2.1775544388609713,
+      "grad_norm": 0.42194533348083496,
+      "learning_rate": 0.0002,
+      "loss": 1.6983,
+      "step": 3250
+    },
+    {
+      "epoch": 2.184254606365159,
+      "grad_norm": 0.44494450092315674,
+      "learning_rate": 0.0002,
+      "loss": 1.6159,
+      "step": 3260
+    },
+    {
+      "epoch": 2.190954773869347,
+      "grad_norm": 0.43524879217147827,
+      "learning_rate": 0.0002,
+      "loss": 1.6977,
+      "step": 3270
+    },
+    {
+      "epoch": 2.1976549413735342,
+      "grad_norm": 0.4621117413043976,
+      "learning_rate": 0.0002,
+      "loss": 1.528,
+      "step": 3280
+    },
+    {
+      "epoch": 2.204355108877722,
+      "grad_norm": 0.4073285460472107,
+      "learning_rate": 0.0002,
+      "loss": 1.632,
+      "step": 3290
+    },
+    {
+      "epoch": 2.2110552763819094,
+      "grad_norm": 0.47868335247039795,
+      "learning_rate": 0.0002,
+      "loss": 1.6141,
+      "step": 3300
+    },
+    {
+      "epoch": 2.217755443886097,
+      "grad_norm": 0.4264970123767853,
+      "learning_rate": 0.0002,
+      "loss": 1.6857,
+      "step": 3310
+    },
+    {
+      "epoch": 2.224455611390285,
+      "grad_norm": 0.4491245150566101,
+      "learning_rate": 0.0002,
+      "loss": 1.5653,
+      "step": 3320
+    },
+    {
+      "epoch": 2.2311557788944723,
+      "grad_norm": 0.4010344445705414,
+      "learning_rate": 0.0002,
+      "loss": 1.5881,
+      "step": 3330
+    },
+    {
+      "epoch": 2.23785594639866,
+      "grad_norm": 0.4232759177684784,
+      "learning_rate": 0.0002,
+      "loss": 1.6684,
+      "step": 3340
+    },
+    {
+      "epoch": 2.2445561139028474,
+      "grad_norm": 0.5099776983261108,
+      "learning_rate": 0.0002,
+      "loss": 1.6336,
+      "step": 3350
+    },
+    {
+      "epoch": 2.251256281407035,
+      "grad_norm": 0.5223407745361328,
+      "learning_rate": 0.0002,
+      "loss": 1.6764,
+      "step": 3360
+    },
+    {
+      "epoch": 2.257956448911223,
+      "grad_norm": 0.47818470001220703,
+      "learning_rate": 0.0002,
+      "loss": 1.6625,
+      "step": 3370
+    },
+    {
+      "epoch": 2.2646566164154103,
+      "grad_norm": 0.4721255898475647,
+      "learning_rate": 0.0002,
+      "loss": 1.5946,
+      "step": 3380
+    },
+    {
+      "epoch": 2.271356783919598,
+      "grad_norm": 0.4113229513168335,
+      "learning_rate": 0.0002,
+      "loss": 1.5568,
+      "step": 3390
+    },
+    {
+      "epoch": 2.2780569514237854,
+      "grad_norm": 0.507080078125,
+      "learning_rate": 0.0002,
+      "loss": 1.6494,
+      "step": 3400
+    },
+    {
+      "epoch": 2.284757118927973,
+      "grad_norm": 0.4852292239665985,
+      "learning_rate": 0.0002,
+      "loss": 1.6183,
+      "step": 3410
+    },
+    {
+      "epoch": 2.291457286432161,
+      "grad_norm": 0.4503684341907501,
+      "learning_rate": 0.0002,
+      "loss": 1.6132,
+      "step": 3420
+    },
+    {
+      "epoch": 2.2981574539363483,
+      "grad_norm": 0.8359600305557251,
+      "learning_rate": 0.0002,
+      "loss": 1.6649,
+      "step": 3430
+    },
+    {
+      "epoch": 2.304857621440536,
+      "grad_norm": 0.44604045152664185,
+      "learning_rate": 0.0002,
+      "loss": 1.6644,
+      "step": 3440
+    },
+    {
+      "epoch": 2.3115577889447234,
+      "grad_norm": 0.45667049288749695,
+      "learning_rate": 0.0002,
+      "loss": 1.5972,
+      "step": 3450
+    },
+    {
+      "epoch": 2.318257956448911,
+      "grad_norm": 0.4879349172115326,
+      "learning_rate": 0.0002,
+      "loss": 1.6667,
+      "step": 3460
+    },
+    {
+      "epoch": 2.324958123953099,
+      "grad_norm": 0.4033963084220886,
+      "learning_rate": 0.0002,
+      "loss": 1.5804,
+      "step": 3470
+    },
+    {
+      "epoch": 2.3316582914572863,
+      "grad_norm": 0.44494301080703735,
+      "learning_rate": 0.0002,
+      "loss": 1.5838,
+      "step": 3480
+    },
+    {
+      "epoch": 2.338358458961474,
+      "grad_norm": 0.4794621765613556,
+      "learning_rate": 0.0002,
+      "loss": 1.5951,
+      "step": 3490
+    },
+    {
+      "epoch": 2.3450586264656614,
+      "grad_norm": 0.41404327750205994,
+      "learning_rate": 0.0002,
+      "loss": 1.6807,
+      "step": 3500
+    },
+    {
+      "epoch": 2.351758793969849,
+      "grad_norm": 0.4664851725101471,
+      "learning_rate": 0.0002,
+      "loss": 1.714,
+      "step": 3510
+    },
+    {
+      "epoch": 2.358458961474037,
+      "grad_norm": 0.4263697564601898,
+      "learning_rate": 0.0002,
+      "loss": 1.6537,
+      "step": 3520
+    },
+    {
+      "epoch": 2.3651591289782243,
+      "grad_norm": 0.5035167336463928,
+      "learning_rate": 0.0002,
+      "loss": 1.6551,
+      "step": 3530
+    },
+    {
+      "epoch": 2.371859296482412,
+      "grad_norm": 0.4380664527416229,
+      "learning_rate": 0.0002,
+      "loss": 1.6208,
+      "step": 3540
+    },
+    {
+      "epoch": 2.3785594639865995,
+      "grad_norm": 0.5227681994438171,
+      "learning_rate": 0.0002,
+      "loss": 1.634,
+      "step": 3550
+    },
+    {
+      "epoch": 2.3852596314907872,
+      "grad_norm": 0.4382302761077881,
+      "learning_rate": 0.0002,
+      "loss": 1.6146,
+      "step": 3560
+    },
+    {
+      "epoch": 2.391959798994975,
+      "grad_norm": 0.4392451047897339,
+      "learning_rate": 0.0002,
+      "loss": 1.5653,
+      "step": 3570
+    },
+    {
+      "epoch": 2.3986599664991624,
+      "grad_norm": 0.4372786581516266,
+      "learning_rate": 0.0002,
+      "loss": 1.6626,
+      "step": 3580
+    },
+    {
+      "epoch": 2.40536013400335,
+      "grad_norm": 0.5015502572059631,
+      "learning_rate": 0.0002,
+      "loss": 1.519,
+      "step": 3590
+    },
+    {
+      "epoch": 2.4120603015075375,
+      "grad_norm": 0.5653210878372192,
+      "learning_rate": 0.0002,
+      "loss": 1.588,
+      "step": 3600
+    },
+    {
+      "epoch": 2.4187604690117253,
+      "grad_norm": 0.53007972240448,
+      "learning_rate": 0.0002,
+      "loss": 1.6075,
+      "step": 3610
+    },
+    {
+      "epoch": 2.425460636515913,
+      "grad_norm": 0.4659176766872406,
+      "learning_rate": 0.0002,
+      "loss": 1.6421,
+      "step": 3620
+    },
+    {
+      "epoch": 2.4321608040201004,
+      "grad_norm": 0.5637837052345276,
+      "learning_rate": 0.0002,
+      "loss": 1.625,
+      "step": 3630
+    },
+    {
+      "epoch": 2.438860971524288,
+      "grad_norm": 0.4248391389846802,
+      "learning_rate": 0.0002,
+      "loss": 1.6168,
+      "step": 3640
+    },
+    {
+      "epoch": 2.4455611390284755,
+      "grad_norm": 0.44668248295783997,
+      "learning_rate": 0.0002,
+      "loss": 1.6822,
+      "step": 3650
+    },
+    {
+      "epoch": 2.4522613065326633,
+      "grad_norm": 0.43990179896354675,
+      "learning_rate": 0.0002,
+      "loss": 1.6417,
+      "step": 3660
+    },
+    {
+      "epoch": 2.458961474036851,
+      "grad_norm": 0.4532523453235626,
+      "learning_rate": 0.0002,
+      "loss": 1.6723,
+      "step": 3670
+    },
+    {
+      "epoch": 2.4656616415410384,
+      "grad_norm": 0.6605591773986816,
+      "learning_rate": 0.0002,
+      "loss": 1.6957,
+      "step": 3680
+    },
+    {
+      "epoch": 2.472361809045226,
+      "grad_norm": 0.4694533348083496,
+      "learning_rate": 0.0002,
+      "loss": 1.6159,
+      "step": 3690
+    },
+    {
+      "epoch": 2.4790619765494135,
+      "grad_norm": 0.4485011100769043,
+      "learning_rate": 0.0002,
+      "loss": 1.6239,
+      "step": 3700
+    },
+    {
+      "epoch": 2.4857621440536013,
+      "grad_norm": 0.4761785864830017,
+      "learning_rate": 0.0002,
+      "loss": 1.6834,
+      "step": 3710
+    },
+    {
+      "epoch": 2.492462311557789,
+      "grad_norm": 0.5116432309150696,
+      "learning_rate": 0.0002,
+      "loss": 1.6313,
+      "step": 3720
+    },
+    {
+      "epoch": 2.4991624790619764,
+      "grad_norm": 0.49523618817329407,
+      "learning_rate": 0.0002,
+      "loss": 1.5054,
+      "step": 3730
+    },
+    {
+      "epoch": 2.505862646566164,
+      "grad_norm": 0.43826380372047424,
+      "learning_rate": 0.0002,
+      "loss": 1.6249,
+      "step": 3740
+    },
+    {
+      "epoch": 2.5125628140703515,
+      "grad_norm": 0.4916154146194458,
+      "learning_rate": 0.0002,
+      "loss": 1.5762,
+      "step": 3750
+    },
+    {
+      "epoch": 2.5192629815745393,
+      "grad_norm": 0.5381299257278442,
+      "learning_rate": 0.0002,
+      "loss": 1.5157,
+      "step": 3760
+    },
+    {
+      "epoch": 2.525963149078727,
+      "grad_norm": 0.44947415590286255,
+      "learning_rate": 0.0002,
+      "loss": 1.6467,
+      "step": 3770
+    },
+    {
+      "epoch": 2.5326633165829144,
+      "grad_norm": 0.49979084730148315,
+      "learning_rate": 0.0002,
+      "loss": 1.67,
+      "step": 3780
+    },
+    {
+      "epoch": 2.539363484087102,
+      "grad_norm": 0.43046900629997253,
+      "learning_rate": 0.0002,
+      "loss": 1.622,
+      "step": 3790
+    },
+    {
+      "epoch": 2.5460636515912896,
+      "grad_norm": 0.4513470530509949,
+      "learning_rate": 0.0002,
+      "loss": 1.6789,
+      "step": 3800
+    },
+    {
+      "epoch": 2.5527638190954773,
+      "grad_norm": 0.49900051951408386,
+      "learning_rate": 0.0002,
+      "loss": 1.6335,
+      "step": 3810
+    },
+    {
+      "epoch": 2.559463986599665,
+      "grad_norm": 0.4348420202732086,
+      "learning_rate": 0.0002,
+      "loss": 1.6322,
+      "step": 3820
+    },
+    {
+      "epoch": 2.5661641541038525,
+      "grad_norm": 0.4684867560863495,
+      "learning_rate": 0.0002,
+      "loss": 1.6218,
+      "step": 3830
+    },
+    {
+      "epoch": 2.5728643216080402,
+      "grad_norm": 0.44430989027023315,
+      "learning_rate": 0.0002,
+      "loss": 1.6535,
+      "step": 3840
+    },
+    {
+      "epoch": 2.5795644891122276,
+      "grad_norm": 0.47375255823135376,
+      "learning_rate": 0.0002,
+      "loss": 1.5909,
+      "step": 3850
+    },
+    {
+      "epoch": 2.5862646566164154,
+      "grad_norm": 0.45493075251579285,
+      "learning_rate": 0.0002,
+      "loss": 1.6269,
+      "step": 3860
+    },
+    {
+      "epoch": 2.592964824120603,
+      "grad_norm": 0.4563275873661041,
+      "learning_rate": 0.0002,
+      "loss": 1.604,
+      "step": 3870
+    },
+    {
+      "epoch": 2.5996649916247905,
+      "grad_norm": 0.46060335636138916,
+      "learning_rate": 0.0002,
+      "loss": 1.642,
+      "step": 3880
+    },
+    {
+      "epoch": 2.6063651591289783,
+      "grad_norm": 0.4718867540359497,
+      "learning_rate": 0.0002,
+      "loss": 1.6302,
+      "step": 3890
+    },
+    {
+      "epoch": 2.6130653266331656,
+      "grad_norm": 0.41570305824279785,
+      "learning_rate": 0.0002,
+      "loss": 1.6242,
+      "step": 3900
+    },
+    {
+      "epoch": 2.6197654941373534,
+      "grad_norm": 0.4603121876716614,
+      "learning_rate": 0.0002,
+      "loss": 1.6401,
+      "step": 3910
+    },
+    {
+      "epoch": 2.626465661641541,
+      "grad_norm": 0.4734652638435364,
+      "learning_rate": 0.0002,
+      "loss": 1.6839,
+      "step": 3920
+    },
+    {
+      "epoch": 2.6331658291457285,
+      "grad_norm": 0.45348483324050903,
+      "learning_rate": 0.0002,
+      "loss": 1.5448,
+      "step": 3930
+    },
+    {
+      "epoch": 2.6398659966499163,
+      "grad_norm": 0.46559447050094604,
+      "learning_rate": 0.0002,
+      "loss": 1.6157,
+      "step": 3940
+    },
+    {
+      "epoch": 2.6465661641541036,
+      "grad_norm": 0.44113144278526306,
+      "learning_rate": 0.0002,
+      "loss": 1.7052,
+      "step": 3950
+    },
+    {
+      "epoch": 2.6532663316582914,
+      "grad_norm": 0.41415104269981384,
+      "learning_rate": 0.0002,
+      "loss": 1.6315,
+      "step": 3960
+    },
+    {
+      "epoch": 2.659966499162479,
+      "grad_norm": 0.48868080973625183,
+      "learning_rate": 0.0002,
+      "loss": 1.6589,
+      "step": 3970
+    },
+    {
+      "epoch": 2.6666666666666665,
+      "grad_norm": 0.49610549211502075,
+      "learning_rate": 0.0002,
+      "loss": 1.6211,
+      "step": 3980
+    },
+    {
+      "epoch": 2.6733668341708543,
+      "grad_norm": 0.4309130907058716,
+      "learning_rate": 0.0002,
+      "loss": 1.6235,
+      "step": 3990
+    },
+    {
+      "epoch": 2.6800670016750416,
+      "grad_norm": 0.4489327669143677,
+      "learning_rate": 0.0002,
+      "loss": 1.6452,
+      "step": 4000
+    },
+    {
+      "epoch": 2.6867671691792294,
+      "grad_norm": 0.5380139946937561,
+      "learning_rate": 0.0002,
+      "loss": 1.5744,
+      "step": 4010
+    },
+    {
+      "epoch": 2.693467336683417,
+      "grad_norm": 0.5076672434806824,
+      "learning_rate": 0.0002,
+      "loss": 1.6524,
+      "step": 4020
+    },
+    {
+      "epoch": 2.7001675041876045,
+      "grad_norm": 0.47620031237602234,
+      "learning_rate": 0.0002,
+      "loss": 1.636,
+      "step": 4030
+    },
+    {
+      "epoch": 2.7068676716917923,
+      "grad_norm": 0.48089155554771423,
+      "learning_rate": 0.0002,
+      "loss": 1.5543,
+      "step": 4040
+    },
+    {
+      "epoch": 2.7135678391959797,
+      "grad_norm": 0.5108814239501953,
+      "learning_rate": 0.0002,
+      "loss": 1.6396,
+      "step": 4050
+    },
+    {
+      "epoch": 2.7202680067001674,
+      "grad_norm": 0.4196513295173645,
+      "learning_rate": 0.0002,
+      "loss": 1.5905,
+      "step": 4060
+    },
+    {
+      "epoch": 2.726968174204355,
+      "grad_norm": 0.4574664831161499,
+      "learning_rate": 0.0002,
+      "loss": 1.686,
+      "step": 4070
+    },
+    {
+      "epoch": 2.7336683417085426,
+      "grad_norm": 0.4671640992164612,
+      "learning_rate": 0.0002,
+      "loss": 1.6234,
+      "step": 4080
+    },
+    {
+      "epoch": 2.7403685092127303,
+      "grad_norm": 0.49355530738830566,
+      "learning_rate": 0.0002,
+      "loss": 1.6827,
+      "step": 4090
+    },
+    {
+      "epoch": 2.7470686767169177,
+      "grad_norm": 0.46716663241386414,
+      "learning_rate": 0.0002,
+      "loss": 1.6999,
+      "step": 4100
+    },
+    {
+      "epoch": 2.7537688442211055,
+      "grad_norm": 0.45420581102371216,
+      "learning_rate": 0.0002,
+      "loss": 1.6463,
+      "step": 4110
+    },
+    {
+      "epoch": 2.7604690117252932,
+      "grad_norm": 0.4680487811565399,
+      "learning_rate": 0.0002,
+      "loss": 1.5718,
+      "step": 4120
+    },
+    {
+      "epoch": 2.7671691792294806,
+      "grad_norm": 0.5375032424926758,
+      "learning_rate": 0.0002,
+      "loss": 1.5968,
+      "step": 4130
+    },
+    {
+      "epoch": 2.7738693467336684,
+      "grad_norm": 0.46026280522346497,
+      "learning_rate": 0.0002,
+      "loss": 1.5254,
+      "step": 4140
+    },
+    {
+      "epoch": 2.7805695142378557,
+      "grad_norm": 0.43658447265625,
+      "learning_rate": 0.0002,
+      "loss": 1.6613,
+      "step": 4150
+    },
+    {
+      "epoch": 2.7872696817420435,
+      "grad_norm": 0.4935547113418579,
+      "learning_rate": 0.0002,
+      "loss": 1.6546,
+      "step": 4160
+    },
+    {
+      "epoch": 2.7939698492462313,
+      "grad_norm": 0.8167962431907654,
+      "learning_rate": 0.0002,
+      "loss": 1.5961,
+      "step": 4170
+    },
+    {
+      "epoch": 2.8006700167504186,
+      "grad_norm": 0.4289683997631073,
+      "learning_rate": 0.0002,
+      "loss": 1.6907,
+      "step": 4180
+    },
+    {
+      "epoch": 2.8073701842546064,
+      "grad_norm": 0.4569324254989624,
+      "learning_rate": 0.0002,
+      "loss": 1.6385,
+      "step": 4190
+    },
+    {
+      "epoch": 2.8140703517587937,
+      "grad_norm": 0.474795937538147,
+      "learning_rate": 0.0002,
+      "loss": 1.6077,
+      "step": 4200
+    },
+    {
+      "epoch": 2.8207705192629815,
+      "grad_norm": 0.44272229075431824,
+      "learning_rate": 0.0002,
+      "loss": 1.6223,
+      "step": 4210
+    },
+    {
+      "epoch": 2.8274706867671693,
+      "grad_norm": 0.525240957736969,
+      "learning_rate": 0.0002,
+      "loss": 1.6706,
+      "step": 4220
+    },
+    {
+      "epoch": 2.8341708542713566,
+      "grad_norm": 0.4802303910255432,
+      "learning_rate": 0.0002,
+      "loss": 1.7196,
+      "step": 4230
+    },
+    {
+      "epoch": 2.8408710217755444,
+      "grad_norm": 0.46400442719459534,
+      "learning_rate": 0.0002,
+      "loss": 1.6002,
+      "step": 4240
+    },
+    {
+      "epoch": 2.8475711892797317,
+      "grad_norm": 0.49884888529777527,
+      "learning_rate": 0.0002,
+      "loss": 1.6052,
+      "step": 4250
+    },
+    {
+      "epoch": 2.8542713567839195,
+      "grad_norm": 0.5015072226524353,
+      "learning_rate": 0.0002,
+      "loss": 1.6919,
+      "step": 4260
+    },
+    {
+      "epoch": 2.8609715242881073,
+      "grad_norm": 0.4335440695285797,
+      "learning_rate": 0.0002,
+      "loss": 1.6335,
+      "step": 4270
+    },
+    {
+      "epoch": 2.8676716917922946,
+      "grad_norm": 0.5131644606590271,
+      "learning_rate": 0.0002,
+      "loss": 1.5664,
+      "step": 4280
+    },
+    {
+      "epoch": 2.8743718592964824,
+      "grad_norm": 0.6977195739746094,
+      "learning_rate": 0.0002,
+      "loss": 1.6409,
+      "step": 4290
+    },
+    {
+      "epoch": 2.8810720268006698,
+      "grad_norm": 0.5133762955665588,
+      "learning_rate": 0.0002,
+      "loss": 1.7192,
+      "step": 4300
+    },
+    {
+      "epoch": 2.8877721943048575,
+      "grad_norm": 0.4737614393234253,
+      "learning_rate": 0.0002,
+      "loss": 1.6257,
+      "step": 4310
+    },
+    {
+      "epoch": 2.8944723618090453,
+      "grad_norm": 0.4580535590648651,
+      "learning_rate": 0.0002,
+      "loss": 1.6076,
+      "step": 4320
+    },
+    {
+      "epoch": 2.901172529313233,
+      "grad_norm": 0.43863341212272644,
+      "learning_rate": 0.0002,
+      "loss": 1.6538,
+      "step": 4330
+    },
+    {
+      "epoch": 2.9078726968174204,
+      "grad_norm": 0.4103737473487854,
+      "learning_rate": 0.0002,
+      "loss": 1.6091,
+      "step": 4340
+    },
+    {
+      "epoch": 2.914572864321608,
+      "grad_norm": 0.438014417886734,
+      "learning_rate": 0.0002,
+      "loss": 1.7106,
+      "step": 4350
+    },
+    {
+      "epoch": 2.9212730318257956,
+      "grad_norm": 0.5068213939666748,
+      "learning_rate": 0.0002,
+      "loss": 1.6025,
+      "step": 4360
+    },
+    {
+      "epoch": 2.9279731993299833,
+      "grad_norm": 0.45305484533309937,
+      "learning_rate": 0.0002,
+      "loss": 1.6426,
+      "step": 4370
+    },
+    {
+      "epoch": 2.934673366834171,
+      "grad_norm": 0.4612090289592743,
+      "learning_rate": 0.0002,
+      "loss": 1.5726,
+      "step": 4380
+    },
+    {
+      "epoch": 2.9413735343383585,
+      "grad_norm": 0.508736789226532,
+      "learning_rate": 0.0002,
+      "loss": 1.6536,
+      "step": 4390
+    },
+    {
+      "epoch": 2.948073701842546,
+      "grad_norm": 0.4924427270889282,
+      "learning_rate": 0.0002,
+      "loss": 1.6132,
+      "step": 4400
+    },
+    {
+      "epoch": 2.9547738693467336,
+      "grad_norm": 0.5707460641860962,
+      "learning_rate": 0.0002,
+      "loss": 1.7007,
+      "step": 4410
+    },
+    {
+      "epoch": 2.9614740368509214,
+      "grad_norm": 0.42270299792289734,
+      "learning_rate": 0.0002,
+      "loss": 1.6814,
+      "step": 4420
+    },
+    {
+      "epoch": 2.968174204355109,
+      "grad_norm": 0.4429931044578552,
+      "learning_rate": 0.0002,
+      "loss": 1.6644,
+      "step": 4430
+    },
+    {
+      "epoch": 2.9748743718592965,
+      "grad_norm": 0.49760574102401733,
+      "learning_rate": 0.0002,
+      "loss": 1.6251,
+      "step": 4440
+    },
+    {
+      "epoch": 2.981574539363484,
+      "grad_norm": 0.4558229148387909,
+      "learning_rate": 0.0002,
+      "loss": 1.6169,
+      "step": 4450
+    },
+    {
+      "epoch": 2.9882747068676716,
+      "grad_norm": 0.39848530292510986,
+      "learning_rate": 0.0002,
+      "loss": 1.6055,
+      "step": 4460
+    },
+    {
+      "epoch": 2.9949748743718594,
+      "grad_norm": 0.5224862098693848,
+      "learning_rate": 0.0002,
+      "loss": 1.6705,
+      "step": 4470
+    },
+    {
+      "epoch": 2.9996649916247904,
+      "eval_loss": 1.8228833675384521,
+      "eval_runtime": 37.9049,
+      "eval_samples_per_second": 13.587,
+      "eval_steps_per_second": 1.715,
+      "step": 4477
+    },
+    {
+      "epoch": 3.0016750418760467,
+      "grad_norm": 0.41169142723083496,
+      "learning_rate": 0.0002,
+      "loss": 1.6637,
+      "step": 4480
+    },
+    {
+      "epoch": 3.0083752093802345,
+      "grad_norm": 0.4865207374095917,
+      "learning_rate": 0.0002,
+      "loss": 1.5974,
+      "step": 4490
+    },
+    {
+      "epoch": 3.0150753768844223,
+      "grad_norm": 0.5462028384208679,
+      "learning_rate": 0.0002,
+      "loss": 1.5297,
+      "step": 4500
+    },
+    {
+      "epoch": 3.0217755443886096,
+      "grad_norm": 0.6169732809066772,
+      "learning_rate": 0.0002,
+      "loss": 1.5251,
+      "step": 4510
+    },
+    {
+      "epoch": 3.0284757118927974,
+      "grad_norm": 0.5667954087257385,
+      "learning_rate": 0.0002,
+      "loss": 1.5559,
+      "step": 4520
+    },
+    {
+      "epoch": 3.0351758793969847,
+      "grad_norm": 0.5758325457572937,
+      "learning_rate": 0.0002,
+      "loss": 1.5037,
+      "step": 4530
+    },
+    {
+      "epoch": 3.0418760469011725,
+      "grad_norm": 0.5220064520835876,
+      "learning_rate": 0.0002,
+      "loss": 1.4873,
+      "step": 4540
+    },
+    {
+      "epoch": 3.0485762144053603,
+      "grad_norm": 0.5469558835029602,
+      "learning_rate": 0.0002,
+      "loss": 1.5126,
+      "step": 4550
+    },
+    {
+      "epoch": 3.0552763819095476,
+      "grad_norm": 0.5680848956108093,
+      "learning_rate": 0.0002,
+      "loss": 1.4275,
+      "step": 4560
+    },
+    {
+      "epoch": 3.0619765494137354,
+      "grad_norm": 0.5906574726104736,
+      "learning_rate": 0.0002,
+      "loss": 1.5187,
+      "step": 4570
+    },
+    {
+      "epoch": 3.0686767169179228,
+      "grad_norm": 0.4725631773471832,
+      "learning_rate": 0.0002,
+      "loss": 1.4551,
+      "step": 4580
+    },
+    {
+      "epoch": 3.0753768844221105,
+      "grad_norm": 0.5273477435112,
+      "learning_rate": 0.0002,
+      "loss": 1.5083,
+      "step": 4590
+    },
+    {
+      "epoch": 3.0820770519262983,
+      "grad_norm": 0.5861203074455261,
+      "learning_rate": 0.0002,
+      "loss": 1.5154,
+      "step": 4600
+    },
+    {
+      "epoch": 3.0887772194304857,
+      "grad_norm": 0.5343965291976929,
+      "learning_rate": 0.0002,
+      "loss": 1.4924,
+      "step": 4610
+    },
+    {
+      "epoch": 3.0954773869346734,
+      "grad_norm": 0.5348150730133057,
+      "learning_rate": 0.0002,
+      "loss": 1.5608,
+      "step": 4620
+    },
+    {
+      "epoch": 3.102177554438861,
+      "grad_norm": 0.5971846580505371,
+      "learning_rate": 0.0002,
+      "loss": 1.5399,
+      "step": 4630
+    },
+    {
+      "epoch": 3.1088777219430486,
+      "grad_norm": 0.5203177332878113,
+      "learning_rate": 0.0002,
+      "loss": 1.4662,
+      "step": 4640
+    },
+    {
+      "epoch": 3.1155778894472363,
+      "grad_norm": 0.55289226770401,
+      "learning_rate": 0.0002,
+      "loss": 1.5805,
+      "step": 4650
+    },
+    {
+      "epoch": 3.1222780569514237,
+      "grad_norm": 0.6878530979156494,
+      "learning_rate": 0.0002,
+      "loss": 1.4745,
+      "step": 4660
+    },
+    {
+      "epoch": 3.1289782244556115,
+      "grad_norm": 0.6173256635665894,
+      "learning_rate": 0.0002,
+      "loss": 1.5335,
+      "step": 4670
+    },
+    {
+      "epoch": 3.135678391959799,
+      "grad_norm": 0.536796510219574,
+      "learning_rate": 0.0002,
+      "loss": 1.51,
+      "step": 4680
+    },
+    {
+      "epoch": 3.1423785594639866,
+      "grad_norm": 0.58846116065979,
+      "learning_rate": 0.0002,
+      "loss": 1.4713,
+      "step": 4690
+    },
+    {
+      "epoch": 3.1490787269681744,
+      "grad_norm": 0.645889401435852,
+      "learning_rate": 0.0002,
+      "loss": 1.5114,
+      "step": 4700
+    },
+    {
+      "epoch": 3.1557788944723617,
+      "grad_norm": 0.6118691563606262,
+      "learning_rate": 0.0002,
+      "loss": 1.4705,
+      "step": 4710
+    },
+    {
+      "epoch": 3.1624790619765495,
+      "grad_norm": 0.5189669132232666,
+      "learning_rate": 0.0002,
+      "loss": 1.5533,
+      "step": 4720
+    },
+    {
+      "epoch": 3.169179229480737,
+      "grad_norm": 0.5794713497161865,
+      "learning_rate": 0.0002,
+      "loss": 1.4769,
+      "step": 4730
+    },
+    {
+      "epoch": 3.1758793969849246,
+      "grad_norm": 0.6579326391220093,
+      "learning_rate": 0.0002,
+      "loss": 1.4849,
+      "step": 4740
+    },
+    {
+      "epoch": 3.1825795644891124,
+      "grad_norm": 0.5822742581367493,
+      "learning_rate": 0.0002,
+      "loss": 1.545,
+      "step": 4750
+    },
+    {
+      "epoch": 3.1892797319932997,
+      "grad_norm": 0.5475956201553345,
+      "learning_rate": 0.0002,
+      "loss": 1.4358,
+      "step": 4760
+    },
+    {
+      "epoch": 3.1959798994974875,
+      "grad_norm": 0.6743834018707275,
+      "learning_rate": 0.0002,
+      "loss": 1.4723,
+      "step": 4770
+    },
+    {
+      "epoch": 3.202680067001675,
+      "grad_norm": 0.6110585927963257,
+      "learning_rate": 0.0002,
+      "loss": 1.5161,
+      "step": 4780
+    },
+    {
+      "epoch": 3.2093802345058626,
+      "grad_norm": 0.5426181554794312,
+      "learning_rate": 0.0002,
+      "loss": 1.5455,
+      "step": 4790
+    },
+    {
+      "epoch": 3.2160804020100504,
+      "grad_norm": 0.6077824234962463,
+      "learning_rate": 0.0002,
+      "loss": 1.5315,
+      "step": 4800
+    },
+    {
+      "epoch": 3.2227805695142377,
+      "grad_norm": 0.5785858631134033,
+      "learning_rate": 0.0002,
+      "loss": 1.5314,
+      "step": 4810
+    },
+    {
+      "epoch": 3.2294807370184255,
+      "grad_norm": 0.6425958275794983,
+      "learning_rate": 0.0002,
+      "loss": 1.4041,
+      "step": 4820
+    },
+    {
+      "epoch": 3.236180904522613,
+      "grad_norm": 0.6607080698013306,
+      "learning_rate": 0.0002,
+      "loss": 1.4751,
+      "step": 4830
+    },
+    {
+      "epoch": 3.2428810720268006,
+      "grad_norm": 0.5385788679122925,
+      "learning_rate": 0.0002,
+      "loss": 1.5267,
+      "step": 4840
+    },
+    {
+      "epoch": 3.2495812395309884,
+      "grad_norm": 0.5630403757095337,
+      "learning_rate": 0.0002,
+      "loss": 1.4673,
+      "step": 4850
+    },
+    {
+      "epoch": 3.2562814070351758,
+      "grad_norm": 0.6340779662132263,
+      "learning_rate": 0.0002,
+      "loss": 1.5257,
+      "step": 4860
+    },
+    {
+      "epoch": 3.2629815745393635,
+      "grad_norm": 0.5305342674255371,
+      "learning_rate": 0.0002,
+      "loss": 1.5148,
+      "step": 4870
+    },
+    {
+      "epoch": 3.2696817420435513,
+      "grad_norm": 0.597670316696167,
+      "learning_rate": 0.0002,
+      "loss": 1.5162,
+      "step": 4880
+    },
+    {
+      "epoch": 3.2763819095477387,
+      "grad_norm": 0.665553867816925,
+      "learning_rate": 0.0002,
+      "loss": 1.5429,
+      "step": 4890
+    },
+    {
+      "epoch": 3.2830820770519265,
+      "grad_norm": 0.579767644405365,
+      "learning_rate": 0.0002,
+      "loss": 1.4607,
+      "step": 4900
+    },
+    {
+      "epoch": 3.289782244556114,
+      "grad_norm": 0.5512481331825256,
+      "learning_rate": 0.0002,
+      "loss": 1.4999,
+      "step": 4910
+    },
+    {
+      "epoch": 3.2964824120603016,
+      "grad_norm": 0.5916532278060913,
+      "learning_rate": 0.0002,
+      "loss": 1.5022,
+      "step": 4920
+    },
+    {
+      "epoch": 3.3031825795644894,
+      "grad_norm": 0.7521726489067078,
+      "learning_rate": 0.0002,
+      "loss": 1.4889,
+      "step": 4930
+    },
+    {
+      "epoch": 3.3098827470686767,
+      "grad_norm": 0.5352797508239746,
+      "learning_rate": 0.0002,
+      "loss": 1.4223,
+      "step": 4940
+    },
+    {
+      "epoch": 3.3165829145728645,
+      "grad_norm": 0.5950371623039246,
+      "learning_rate": 0.0002,
+      "loss": 1.5122,
+      "step": 4950
+    },
+    {
+      "epoch": 3.323283082077052,
+      "grad_norm": 0.8020477890968323,
+      "learning_rate": 0.0002,
+      "loss": 1.5072,
+      "step": 4960
+    },
+    {
+      "epoch": 3.3299832495812396,
+      "grad_norm": 0.6790024638175964,
+      "learning_rate": 0.0002,
+      "loss": 1.5422,
+      "step": 4970
+    },
+    {
+      "epoch": 3.3366834170854274,
+      "grad_norm": 0.687627375125885,
+      "learning_rate": 0.0002,
+      "loss": 1.5363,
+      "step": 4980
+    },
+    {
+      "epoch": 3.3433835845896147,
+      "grad_norm": 0.6094385385513306,
+      "learning_rate": 0.0002,
+      "loss": 1.5276,
+      "step": 4990
+    },
+    {
+      "epoch": 3.3500837520938025,
+      "grad_norm": 0.6541242003440857,
+      "learning_rate": 0.0002,
+      "loss": 1.549,
+      "step": 5000
+    },
+    {
+      "epoch": 3.35678391959799,
+      "grad_norm": 0.5560880303382874,
+      "learning_rate": 0.0002,
+      "loss": 1.6067,
+      "step": 5010
+    },
+    {
+      "epoch": 3.3634840871021776,
+      "grad_norm": 0.5440094470977783,
+      "learning_rate": 0.0002,
+      "loss": 1.5769,
+      "step": 5020
+    },
+    {
+      "epoch": 3.3701842546063654,
+      "grad_norm": 0.5749301314353943,
+      "learning_rate": 0.0002,
+      "loss": 1.6183,
+      "step": 5030
+    },
+    {
+      "epoch": 3.3768844221105527,
+      "grad_norm": 0.5919716954231262,
+      "learning_rate": 0.0002,
+      "loss": 1.4801,
+      "step": 5040
+    },
+    {
+      "epoch": 3.3835845896147405,
+      "grad_norm": 0.6331481337547302,
+      "learning_rate": 0.0002,
+      "loss": 1.5957,
+      "step": 5050
+    },
+    {
+      "epoch": 3.390284757118928,
+      "grad_norm": 0.5687161684036255,
+      "learning_rate": 0.0002,
+      "loss": 1.5188,
+      "step": 5060
+    },
+    {
+      "epoch": 3.3969849246231156,
+      "grad_norm": 0.6718577742576599,
+      "learning_rate": 0.0002,
+      "loss": 1.5702,
+      "step": 5070
+    },
+    {
+      "epoch": 3.4036850921273034,
+      "grad_norm": 0.5089324116706848,
+      "learning_rate": 0.0002,
+      "loss": 1.5577,
+      "step": 5080
+    },
+    {
+      "epoch": 3.4103852596314908,
+      "grad_norm": 0.5710174441337585,
+      "learning_rate": 0.0002,
+      "loss": 1.512,
+      "step": 5090
+    },
+    {
+      "epoch": 3.4170854271356785,
+      "grad_norm": 0.6670721173286438,
+      "learning_rate": 0.0002,
+      "loss": 1.5492,
+      "step": 5100
+    },
+    {
+      "epoch": 3.423785594639866,
+      "grad_norm": 0.6875665187835693,
+      "learning_rate": 0.0002,
+      "loss": 1.5227,
+      "step": 5110
+    },
+    {
+      "epoch": 3.4304857621440537,
+      "grad_norm": 0.5375880599021912,
+      "learning_rate": 0.0002,
+      "loss": 1.4496,
+      "step": 5120
+    },
+    {
+      "epoch": 3.4371859296482414,
+      "grad_norm": 0.6550399661064148,
+      "learning_rate": 0.0002,
+      "loss": 1.5527,
+      "step": 5130
+    },
+    {
+      "epoch": 3.4438860971524288,
+      "grad_norm": 0.5948067903518677,
+      "learning_rate": 0.0002,
+      "loss": 1.5687,
+      "step": 5140
+    },
+    {
+      "epoch": 3.4505862646566166,
+      "grad_norm": 0.6134477257728577,
+      "learning_rate": 0.0002,
+      "loss": 1.4813,
+      "step": 5150
+    },
+    {
+      "epoch": 3.457286432160804,
+      "grad_norm": 0.6506398320198059,
+      "learning_rate": 0.0002,
+      "loss": 1.5069,
+      "step": 5160
+    },
+    {
+      "epoch": 3.4639865996649917,
+      "grad_norm": 0.6060147881507874,
+      "learning_rate": 0.0002,
+      "loss": 1.4422,
+      "step": 5170
+    },
+    {
+      "epoch": 3.4706867671691795,
+      "grad_norm": 0.6173806190490723,
+      "learning_rate": 0.0002,
+      "loss": 1.5093,
+      "step": 5180
+    },
+    {
+      "epoch": 3.477386934673367,
+      "grad_norm": 0.6032607555389404,
+      "learning_rate": 0.0002,
+      "loss": 1.4975,
+      "step": 5190
+    },
+    {
+      "epoch": 3.4840871021775546,
+      "grad_norm": 0.5652492046356201,
+      "learning_rate": 0.0002,
+      "loss": 1.4979,
+      "step": 5200
+    },
+    {
+      "epoch": 3.490787269681742,
+      "grad_norm": 0.6168607473373413,
+      "learning_rate": 0.0002,
+      "loss": 1.4883,
+      "step": 5210
+    },
+    {
+      "epoch": 3.4974874371859297,
+      "grad_norm": 0.6170629262924194,
+      "learning_rate": 0.0002,
+      "loss": 1.5164,
+      "step": 5220
+    },
+    {
+      "epoch": 3.5041876046901175,
+      "grad_norm": 0.6926297545433044,
+      "learning_rate": 0.0002,
+      "loss": 1.4879,
+      "step": 5230
+    },
+    {
+      "epoch": 3.510887772194305,
+      "grad_norm": 0.6702437996864319,
+      "learning_rate": 0.0002,
+      "loss": 1.4982,
+      "step": 5240
+    },
+    {
+      "epoch": 3.5175879396984926,
+      "grad_norm": 0.5421436429023743,
+      "learning_rate": 0.0002,
+      "loss": 1.4986,
+      "step": 5250
+    },
+    {
+      "epoch": 3.52428810720268,
+      "grad_norm": 0.5726765990257263,
+      "learning_rate": 0.0002,
+      "loss": 1.4673,
+      "step": 5260
+    },
+    {
+      "epoch": 3.5309882747068677,
+      "grad_norm": 0.5685455203056335,
+      "learning_rate": 0.0002,
+      "loss": 1.5423,
+      "step": 5270
+    },
+    {
+      "epoch": 3.5376884422110555,
+      "grad_norm": 0.6018396019935608,
+      "learning_rate": 0.0002,
+      "loss": 1.4715,
+      "step": 5280
+    },
+    {
+      "epoch": 3.544388609715243,
+      "grad_norm": 0.5731932520866394,
+      "learning_rate": 0.0002,
+      "loss": 1.5451,
+      "step": 5290
+    },
+    {
+      "epoch": 3.5510887772194306,
+      "grad_norm": 0.6601519584655762,
+      "learning_rate": 0.0002,
+      "loss": 1.4752,
+      "step": 5300
+    },
+    {
+      "epoch": 3.557788944723618,
+      "grad_norm": 0.5545530319213867,
+      "learning_rate": 0.0002,
+      "loss": 1.5434,
+      "step": 5310
+    },
+    {
+      "epoch": 3.5644891122278057,
+      "grad_norm": 0.5998541116714478,
+      "learning_rate": 0.0002,
+      "loss": 1.5438,
+      "step": 5320
+    },
+    {
+      "epoch": 3.5711892797319935,
+      "grad_norm": 0.5651767253875732,
+      "learning_rate": 0.0002,
+      "loss": 1.56,
+      "step": 5330
+    },
+    {
+      "epoch": 3.577889447236181,
+      "grad_norm": 0.7425084114074707,
+      "learning_rate": 0.0002,
+      "loss": 1.4829,
+      "step": 5340
+    },
+    {
+      "epoch": 3.5845896147403686,
+      "grad_norm": 0.5770602226257324,
+      "learning_rate": 0.0002,
+      "loss": 1.5571,
+      "step": 5350
+    },
+    {
+      "epoch": 3.591289782244556,
+      "grad_norm": 0.54723060131073,
+      "learning_rate": 0.0002,
+      "loss": 1.458,
+      "step": 5360
+    },
+    {
+      "epoch": 3.5979899497487438,
+      "grad_norm": 0.6658238172531128,
+      "learning_rate": 0.0002,
+      "loss": 1.497,
+      "step": 5370
+    },
+    {
+      "epoch": 3.6046901172529315,
+      "grad_norm": 0.5787645578384399,
+      "learning_rate": 0.0002,
+      "loss": 1.5456,
+      "step": 5380
+    },
+    {
+      "epoch": 3.611390284757119,
+      "grad_norm": 0.594913125038147,
+      "learning_rate": 0.0002,
+      "loss": 1.5343,
+      "step": 5390
+    },
+    {
+      "epoch": 3.6180904522613067,
+      "grad_norm": 0.4964977502822876,
+      "learning_rate": 0.0002,
+      "loss": 1.4727,
+      "step": 5400
+    },
+    {
+      "epoch": 3.624790619765494,
+      "grad_norm": 0.6087527275085449,
+      "learning_rate": 0.0002,
+      "loss": 1.5062,
+      "step": 5410
+    },
+    {
+      "epoch": 3.6314907872696818,
+      "grad_norm": 0.6315323710441589,
+      "learning_rate": 0.0002,
+      "loss": 1.5098,
+      "step": 5420
+    },
+    {
+      "epoch": 3.6381909547738696,
+      "grad_norm": 0.574799120426178,
+      "learning_rate": 0.0002,
+      "loss": 1.4855,
+      "step": 5430
+    },
+    {
+      "epoch": 3.644891122278057,
+      "grad_norm": 0.5949277877807617,
+      "learning_rate": 0.0002,
+      "loss": 1.4595,
+      "step": 5440
+    },
+    {
+      "epoch": 3.6515912897822447,
+      "grad_norm": 0.5640677213668823,
+      "learning_rate": 0.0002,
+      "loss": 1.4816,
+      "step": 5450
+    },
+    {
+      "epoch": 3.658291457286432,
+      "grad_norm": 0.6198237538337708,
+      "learning_rate": 0.0002,
+      "loss": 1.525,
+      "step": 5460
+    },
+    {
+      "epoch": 3.66499162479062,
+      "grad_norm": 0.6902034878730774,
+      "learning_rate": 0.0002,
+      "loss": 1.5676,
+      "step": 5470
+    },
+    {
+      "epoch": 3.6716917922948076,
+      "grad_norm": 0.5686674118041992,
+      "learning_rate": 0.0002,
+      "loss": 1.544,
+      "step": 5480
+    },
+    {
+      "epoch": 3.678391959798995,
+      "grad_norm": 0.6532107591629028,
+      "learning_rate": 0.0002,
+      "loss": 1.5255,
+      "step": 5490
+    },
+    {
+      "epoch": 3.6850921273031827,
+      "grad_norm": 0.5790849924087524,
+      "learning_rate": 0.0002,
+      "loss": 1.5767,
+      "step": 5500
+    },
+    {
+      "epoch": 3.69179229480737,
+      "grad_norm": 0.6055065393447876,
+      "learning_rate": 0.0002,
+      "loss": 1.5507,
+      "step": 5510
+    },
+    {
+      "epoch": 3.698492462311558,
+      "grad_norm": 0.5630605816841125,
+      "learning_rate": 0.0002,
+      "loss": 1.4656,
+      "step": 5520
+    },
+    {
+      "epoch": 3.7051926298157456,
+      "grad_norm": 0.6005825996398926,
+      "learning_rate": 0.0002,
+      "loss": 1.537,
+      "step": 5530
+    },
+    {
+      "epoch": 3.711892797319933,
+      "grad_norm": 0.6553038954734802,
+      "learning_rate": 0.0002,
+      "loss": 1.5313,
+      "step": 5540
+    },
+    {
+      "epoch": 3.7185929648241207,
+      "grad_norm": 0.5601094961166382,
+      "learning_rate": 0.0002,
+      "loss": 1.4943,
+      "step": 5550
+    },
+    {
+      "epoch": 3.725293132328308,
+      "grad_norm": 0.6598808169364929,
+      "learning_rate": 0.0002,
+      "loss": 1.5161,
+      "step": 5560
+    },
+    {
+      "epoch": 3.731993299832496,
+      "grad_norm": 0.5506255626678467,
+      "learning_rate": 0.0002,
+      "loss": 1.5345,
+      "step": 5570
+    },
+    {
+      "epoch": 3.7386934673366836,
+      "grad_norm": 0.6001223921775818,
+      "learning_rate": 0.0002,
+      "loss": 1.4805,
+      "step": 5580
+    },
+    {
+      "epoch": 3.745393634840871,
+      "grad_norm": 0.6287297606468201,
+      "learning_rate": 0.0002,
+      "loss": 1.4652,
+      "step": 5590
+    },
+    {
+      "epoch": 3.7520938023450587,
+      "grad_norm": 0.6253238916397095,
+      "learning_rate": 0.0002,
+      "loss": 1.5246,
+      "step": 5600
+    },
+    {
+      "epoch": 3.758793969849246,
+      "grad_norm": 0.5713174939155579,
+      "learning_rate": 0.0002,
+      "loss": 1.5691,
+      "step": 5610
+    },
+    {
+      "epoch": 3.765494137353434,
+      "grad_norm": 0.6198310852050781,
+      "learning_rate": 0.0002,
+      "loss": 1.5661,
+      "step": 5620
+    },
+    {
+      "epoch": 3.7721943048576216,
+      "grad_norm": 0.5941224098205566,
+      "learning_rate": 0.0002,
+      "loss": 1.5448,
+      "step": 5630
+    },
+    {
+      "epoch": 3.778894472361809,
+      "grad_norm": 0.606002151966095,
+      "learning_rate": 0.0002,
+      "loss": 1.4925,
+      "step": 5640
+    },
+    {
+      "epoch": 3.7855946398659968,
+      "grad_norm": 0.6540704965591431,
+      "learning_rate": 0.0002,
+      "loss": 1.5182,
+      "step": 5650
+    },
+    {
+      "epoch": 3.792294807370184,
+      "grad_norm": 0.6147415041923523,
+      "learning_rate": 0.0002,
+      "loss": 1.5903,
+      "step": 5660
+    },
+    {
+      "epoch": 3.798994974874372,
+      "grad_norm": 0.5649605393409729,
+      "learning_rate": 0.0002,
+      "loss": 1.5329,
+      "step": 5670
+    },
+    {
+      "epoch": 3.8056951423785597,
+      "grad_norm": 0.6788773536682129,
+      "learning_rate": 0.0002,
+      "loss": 1.5747,
+      "step": 5680
+    },
+    {
+      "epoch": 3.812395309882747,
+      "grad_norm": 0.6581860780715942,
+      "learning_rate": 0.0002,
+      "loss": 1.535,
+      "step": 5690
+    },
+    {
+      "epoch": 3.819095477386935,
+      "grad_norm": 0.5529348850250244,
+      "learning_rate": 0.0002,
+      "loss": 1.4587,
+      "step": 5700
+    },
+    {
+      "epoch": 3.825795644891122,
+      "grad_norm": 0.6320232152938843,
+      "learning_rate": 0.0002,
+      "loss": 1.5594,
+      "step": 5710
+    },
+    {
+      "epoch": 3.83249581239531,
+      "grad_norm": 0.6529698371887207,
+      "learning_rate": 0.0002,
+      "loss": 1.4696,
+      "step": 5720
+    },
+    {
+      "epoch": 3.8391959798994977,
+      "grad_norm": 0.5983362793922424,
+      "learning_rate": 0.0002,
+      "loss": 1.5854,
+      "step": 5730
+    },
+    {
+      "epoch": 3.845896147403685,
+      "grad_norm": 0.6335684061050415,
+      "learning_rate": 0.0002,
+      "loss": 1.465,
+      "step": 5740
+    },
+    {
+      "epoch": 3.852596314907873,
+      "grad_norm": 0.700446605682373,
+      "learning_rate": 0.0002,
+      "loss": 1.5545,
+      "step": 5750
+    },
+    {
+      "epoch": 3.85929648241206,
+      "grad_norm": 0.6092597842216492,
+      "learning_rate": 0.0002,
+      "loss": 1.5707,
+      "step": 5760
+    },
+    {
+      "epoch": 3.865996649916248,
+      "grad_norm": 0.564146101474762,
+      "learning_rate": 0.0002,
+      "loss": 1.5729,
+      "step": 5770
+    },
+    {
+      "epoch": 3.8726968174204357,
+      "grad_norm": 0.615275502204895,
+      "learning_rate": 0.0002,
+      "loss": 1.5872,
+      "step": 5780
+    },
+    {
+      "epoch": 3.879396984924623,
+      "grad_norm": 0.6685376763343811,
+      "learning_rate": 0.0002,
+      "loss": 1.5142,
+      "step": 5790
+    },
+    {
+      "epoch": 3.886097152428811,
+      "grad_norm": 0.6116922497749329,
+      "learning_rate": 0.0002,
+      "loss": 1.4752,
+      "step": 5800
+    },
+    {
+      "epoch": 3.892797319932998,
+      "grad_norm": 0.5486813187599182,
+      "learning_rate": 0.0002,
+      "loss": 1.5179,
+      "step": 5810
+    },
+    {
+      "epoch": 3.899497487437186,
+      "grad_norm": 0.6208204030990601,
+      "learning_rate": 0.0002,
+      "loss": 1.5167,
+      "step": 5820
+    },
+    {
+      "epoch": 3.9061976549413737,
+      "grad_norm": 0.6500625014305115,
+      "learning_rate": 0.0002,
+      "loss": 1.5334,
+      "step": 5830
+    },
+    {
+      "epoch": 3.912897822445561,
+      "grad_norm": 0.5948089361190796,
+      "learning_rate": 0.0002,
+      "loss": 1.4716,
+      "step": 5840
+    },
+    {
+      "epoch": 3.919597989949749,
+      "grad_norm": 0.7210732698440552,
+      "learning_rate": 0.0002,
+      "loss": 1.6011,
+      "step": 5850
+    },
+    {
+      "epoch": 3.926298157453936,
+      "grad_norm": 0.6662322878837585,
+      "learning_rate": 0.0002,
+      "loss": 1.5519,
+      "step": 5860
+    },
+    {
+      "epoch": 3.932998324958124,
+      "grad_norm": 0.5613839626312256,
+      "learning_rate": 0.0002,
+      "loss": 1.5656,
+      "step": 5870
+    },
+    {
+      "epoch": 3.9396984924623117,
+      "grad_norm": 0.6069002151489258,
+      "learning_rate": 0.0002,
+      "loss": 1.544,
+      "step": 5880
+    },
+    {
+      "epoch": 3.946398659966499,
+      "grad_norm": 0.7075562477111816,
+      "learning_rate": 0.0002,
+      "loss": 1.6745,
+      "step": 5890
+    },
+    {
+      "epoch": 3.953098827470687,
+      "grad_norm": 0.6316173076629639,
+      "learning_rate": 0.0002,
+      "loss": 1.5391,
+      "step": 5900
+    },
+    {
+      "epoch": 3.959798994974874,
+      "grad_norm": 0.5716308355331421,
+      "learning_rate": 0.0002,
+      "loss": 1.6314,
+      "step": 5910
+    },
+    {
+      "epoch": 3.966499162479062,
+      "grad_norm": 0.6800096035003662,
+      "learning_rate": 0.0002,
+      "loss": 1.5947,
+      "step": 5920
+    },
+    {
+      "epoch": 3.9731993299832498,
+      "grad_norm": 0.6057983040809631,
+      "learning_rate": 0.0002,
+      "loss": 1.5189,
+      "step": 5930
+    },
+    {
+      "epoch": 3.979899497487437,
+      "grad_norm": 0.5938987731933594,
+      "learning_rate": 0.0002,
+      "loss": 1.5431,
+      "step": 5940
+    },
+    {
+      "epoch": 3.986599664991625,
+      "grad_norm": 0.6963576674461365,
+      "learning_rate": 0.0002,
+      "loss": 1.5111,
+      "step": 5950
+    },
+    {
+      "epoch": 3.993299832495812,
+      "grad_norm": 0.6279940009117126,
+      "learning_rate": 0.0002,
+      "loss": 1.5521,
+      "step": 5960
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 0.7161159515380859,
+      "learning_rate": 0.0002,
+      "loss": 1.5974,
+      "step": 5970
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 1.8655421733856201,
+      "eval_runtime": 37.9276,
+      "eval_samples_per_second": 13.579,
+      "eval_steps_per_second": 1.714,
+      "step": 5970
+    },
+    {
+      "epoch": 4.006700167504188,
+      "grad_norm": 0.7380476593971252,
+      "learning_rate": 0.0002,
+      "loss": 1.3666,
+      "step": 5980
+    },
+    {
+      "epoch": 4.013400335008376,
+      "grad_norm": 0.7148947715759277,
+      "learning_rate": 0.0002,
+      "loss": 1.3913,
+      "step": 5990
+    },
+    {
+      "epoch": 4.0201005025125625,
+      "grad_norm": 0.6177082657814026,
+      "learning_rate": 0.0002,
+      "loss": 1.4204,
+      "step": 6000
+    },
+    {
+      "epoch": 4.02680067001675,
+      "grad_norm": 0.8552946448326111,
+      "learning_rate": 0.0002,
+      "loss": 1.4421,
+      "step": 6010
+    },
+    {
+      "epoch": 4.033500837520938,
+      "grad_norm": 0.8033416271209717,
+      "learning_rate": 0.0002,
+      "loss": 1.4342,
+      "step": 6020
+    },
+    {
+      "epoch": 4.040201005025126,
+      "grad_norm": 0.8501318097114563,
+      "learning_rate": 0.0002,
+      "loss": 1.4092,
+      "step": 6030
+    },
+    {
+      "epoch": 4.046901172529314,
+      "grad_norm": 0.6981393098831177,
+      "learning_rate": 0.0002,
+      "loss": 1.3367,
+      "step": 6040
+    },
+    {
+      "epoch": 4.0536013400335005,
+      "grad_norm": 0.7227180600166321,
+      "learning_rate": 0.0002,
+      "loss": 1.3925,
+      "step": 6050
+    },
+    {
+      "epoch": 4.060301507537688,
+      "grad_norm": 0.6923989653587341,
+      "learning_rate": 0.0002,
+      "loss": 1.4007,
+      "step": 6060
+    },
+    {
+      "epoch": 4.067001675041876,
+      "grad_norm": 0.879779040813446,
+      "learning_rate": 0.0002,
+      "loss": 1.3837,
+      "step": 6070
+    },
+    {
+      "epoch": 4.073701842546064,
+      "grad_norm": 0.8184754848480225,
+      "learning_rate": 0.0002,
+      "loss": 1.4383,
+      "step": 6080
+    },
+    {
+      "epoch": 4.080402010050252,
+      "grad_norm": 0.8211342692375183,
+      "learning_rate": 0.0002,
+      "loss": 1.3128,
+      "step": 6090
+    },
+    {
+      "epoch": 4.0871021775544385,
+      "grad_norm": 0.7542396783828735,
+      "learning_rate": 0.0002,
+      "loss": 1.3892,
+      "step": 6100
+    },
+    {
+      "epoch": 4.093802345058626,
+      "grad_norm": 0.6631066799163818,
+      "learning_rate": 0.0002,
+      "loss": 1.3607,
+      "step": 6110
+    },
+    {
+      "epoch": 4.100502512562814,
+      "grad_norm": 0.6728386282920837,
+      "learning_rate": 0.0002,
+      "loss": 1.3275,
+      "step": 6120
+    },
+    {
+      "epoch": 4.107202680067002,
+      "grad_norm": 0.681851863861084,
+      "learning_rate": 0.0002,
+      "loss": 1.3443,
+      "step": 6130
+    },
+    {
+      "epoch": 4.11390284757119,
+      "grad_norm": 0.8757794499397278,
+      "learning_rate": 0.0002,
+      "loss": 1.3486,
+      "step": 6140
+    },
+    {
+      "epoch": 4.1206030150753765,
+      "grad_norm": 0.6567301750183105,
+      "learning_rate": 0.0002,
+      "loss": 1.351,
+      "step": 6150
+    },
+    {
+      "epoch": 4.127303182579564,
+      "grad_norm": 0.7950329184532166,
+      "learning_rate": 0.0002,
+      "loss": 1.3824,
+      "step": 6160
+    },
+    {
+      "epoch": 4.134003350083752,
+      "grad_norm": 0.7545644044876099,
+      "learning_rate": 0.0002,
+      "loss": 1.3738,
+      "step": 6170
+    },
+    {
+      "epoch": 4.14070351758794,
+      "grad_norm": 0.7172710299491882,
+      "learning_rate": 0.0002,
+      "loss": 1.4214,
+      "step": 6180
+    },
+    {
+      "epoch": 4.147403685092128,
+      "grad_norm": 0.7040584087371826,
+      "learning_rate": 0.0002,
+      "loss": 1.4091,
+      "step": 6190
+    },
+    {
+      "epoch": 4.1541038525963145,
+      "grad_norm": 0.7482913732528687,
+      "learning_rate": 0.0002,
+      "loss": 1.4149,
+      "step": 6200
+    },
+    {
+      "epoch": 4.160804020100502,
+      "grad_norm": 0.8523276448249817,
+      "learning_rate": 0.0002,
+      "loss": 1.3227,
+      "step": 6210
+    },
+    {
+      "epoch": 4.16750418760469,
+      "grad_norm": 0.6672041416168213,
+      "learning_rate": 0.0002,
+      "loss": 1.4194,
+      "step": 6220
+    },
+    {
+      "epoch": 4.174204355108878,
+      "grad_norm": 0.7523500919342041,
+      "learning_rate": 0.0002,
+      "loss": 1.3953,
+      "step": 6230
+    },
+    {
+      "epoch": 4.180904522613066,
+      "grad_norm": 0.8085253834724426,
+      "learning_rate": 0.0002,
+      "loss": 1.371,
+      "step": 6240
+    },
+    {
+      "epoch": 4.187604690117253,
+      "grad_norm": 0.789450466632843,
+      "learning_rate": 0.0002,
+      "loss": 1.3293,
+      "step": 6250
+    },
+    {
+      "epoch": 4.19430485762144,
+      "grad_norm": 0.7502310872077942,
+      "learning_rate": 0.0002,
+      "loss": 1.3539,
+      "step": 6260
+    },
+    {
+      "epoch": 4.201005025125628,
+      "grad_norm": 0.7397456765174866,
+      "learning_rate": 0.0002,
+      "loss": 1.3415,
+      "step": 6270
+    },
+    {
+      "epoch": 4.207705192629816,
+      "grad_norm": 0.6921947002410889,
+      "learning_rate": 0.0002,
+      "loss": 1.3963,
+      "step": 6280
+    },
+    {
+      "epoch": 4.214405360134004,
+      "grad_norm": 0.9334571957588196,
+      "learning_rate": 0.0002,
+      "loss": 1.3125,
+      "step": 6290
+    },
+    {
+      "epoch": 4.221105527638191,
+      "grad_norm": 0.725799024105072,
+      "learning_rate": 0.0002,
+      "loss": 1.3612,
+      "step": 6300
+    },
+    {
+      "epoch": 4.227805695142378,
+      "grad_norm": 0.8290495872497559,
+      "learning_rate": 0.0002,
+      "loss": 1.4217,
+      "step": 6310
+    },
+    {
+      "epoch": 4.234505862646566,
+      "grad_norm": 0.688983678817749,
+      "learning_rate": 0.0002,
+      "loss": 1.4135,
+      "step": 6320
+    },
+    {
+      "epoch": 4.241206030150754,
+      "grad_norm": 0.8620913028717041,
+      "learning_rate": 0.0002,
+      "loss": 1.3807,
+      "step": 6330
+    },
+    {
+      "epoch": 4.247906197654942,
+      "grad_norm": 0.8008657693862915,
+      "learning_rate": 0.0002,
+      "loss": 1.3738,
+      "step": 6340
+    },
+    {
+      "epoch": 4.254606365159129,
+      "grad_norm": 0.7379199266433716,
+      "learning_rate": 0.0002,
+      "loss": 1.4005,
+      "step": 6350
+    },
+    {
+      "epoch": 4.261306532663316,
+      "grad_norm": 0.7842815518379211,
+      "learning_rate": 0.0002,
+      "loss": 1.426,
+      "step": 6360
+    },
+    {
+      "epoch": 4.268006700167504,
+      "grad_norm": 0.812600314617157,
+      "learning_rate": 0.0002,
+      "loss": 1.4262,
+      "step": 6370
+    },
+    {
+      "epoch": 4.274706867671692,
+      "grad_norm": 0.7852841019630432,
+      "learning_rate": 0.0002,
+      "loss": 1.4028,
+      "step": 6380
+    },
+    {
+      "epoch": 4.28140703517588,
+      "grad_norm": 1.0377534627914429,
+      "learning_rate": 0.0002,
+      "loss": 1.3722,
+      "step": 6390
+    },
+    {
+      "epoch": 4.288107202680067,
+      "grad_norm": 1.03935706615448,
+      "learning_rate": 0.0002,
+      "loss": 1.3755,
+      "step": 6400
+    },
+    {
+      "epoch": 4.294807370184254,
+      "grad_norm": 0.7244732975959778,
+      "learning_rate": 0.0002,
+      "loss": 1.3961,
+      "step": 6410
+    },
+    {
+      "epoch": 4.301507537688442,
+      "grad_norm": 0.7137406468391418,
+      "learning_rate": 0.0002,
+      "loss": 1.4608,
+      "step": 6420
+    },
+    {
+      "epoch": 4.30820770519263,
+      "grad_norm": 0.7492543458938599,
+      "learning_rate": 0.0002,
+      "loss": 1.4461,
+      "step": 6430
+    },
+    {
+      "epoch": 4.314907872696818,
+      "grad_norm": 0.7065439224243164,
+      "learning_rate": 0.0002,
+      "loss": 1.4562,
+      "step": 6440
+    },
+    {
+      "epoch": 4.321608040201005,
+      "grad_norm": 0.7786989808082581,
+      "learning_rate": 0.0002,
+      "loss": 1.4246,
+      "step": 6450
+    },
+    {
+      "epoch": 4.328308207705192,
+      "grad_norm": 0.7369208335876465,
+      "learning_rate": 0.0002,
+      "loss": 1.3098,
+      "step": 6460
+    },
+    {
+      "epoch": 4.33500837520938,
+      "grad_norm": 0.7412346005439758,
+      "learning_rate": 0.0002,
+      "loss": 1.3686,
+      "step": 6470
+    },
+    {
+      "epoch": 4.341708542713568,
+      "grad_norm": 0.780927300453186,
+      "learning_rate": 0.0002,
+      "loss": 1.4087,
+      "step": 6480
+    },
+    {
+      "epoch": 4.348408710217756,
+      "grad_norm": 0.8320930600166321,
+      "learning_rate": 0.0002,
+      "loss": 1.3628,
+      "step": 6490
+    },
+    {
+      "epoch": 4.355108877721943,
+      "grad_norm": 0.6871094703674316,
+      "learning_rate": 0.0002,
+      "loss": 1.3715,
+      "step": 6500
+    },
+    {
+      "epoch": 4.36180904522613,
+      "grad_norm": 0.6751559972763062,
+      "learning_rate": 0.0002,
+      "loss": 1.3257,
+      "step": 6510
+    },
+    {
+      "epoch": 4.368509212730318,
+      "grad_norm": 0.7723976969718933,
+      "learning_rate": 0.0002,
+      "loss": 1.4311,
+      "step": 6520
+    },
+    {
+      "epoch": 4.375209380234506,
+      "grad_norm": 0.7915401458740234,
+      "learning_rate": 0.0002,
+      "loss": 1.4086,
+      "step": 6530
+    },
+    {
+      "epoch": 4.381909547738694,
+      "grad_norm": 0.7329102754592896,
+      "learning_rate": 0.0002,
+      "loss": 1.3973,
+      "step": 6540
+    },
+    {
+      "epoch": 4.388609715242881,
+      "grad_norm": 0.7388760447502136,
+      "learning_rate": 0.0002,
+      "loss": 1.447,
+      "step": 6550
+    },
+    {
+      "epoch": 4.3953098827470685,
+      "grad_norm": 0.8282579183578491,
+      "learning_rate": 0.0002,
+      "loss": 1.4378,
+      "step": 6560
+    },
+    {
+      "epoch": 4.402010050251256,
+      "grad_norm": 0.7192724347114563,
+      "learning_rate": 0.0002,
+      "loss": 1.3923,
+      "step": 6570
+    },
+    {
+      "epoch": 4.408710217755444,
+      "grad_norm": 0.746526837348938,
+      "learning_rate": 0.0002,
+      "loss": 1.4141,
+      "step": 6580
+    },
+    {
+      "epoch": 4.415410385259632,
+      "grad_norm": 0.8738046288490295,
+      "learning_rate": 0.0002,
+      "loss": 1.33,
+      "step": 6590
+    },
+    {
+      "epoch": 4.422110552763819,
+      "grad_norm": 0.8408458828926086,
+      "learning_rate": 0.0002,
+      "loss": 1.3995,
+      "step": 6600
+    },
+    {
+      "epoch": 4.4288107202680065,
+      "grad_norm": 0.8110666275024414,
+      "learning_rate": 0.0002,
+      "loss": 1.4148,
+      "step": 6610
+    },
+    {
+      "epoch": 4.435510887772194,
+      "grad_norm": 0.8602406978607178,
+      "learning_rate": 0.0002,
+      "loss": 1.441,
+      "step": 6620
+    },
+    {
+      "epoch": 4.442211055276382,
+      "grad_norm": 0.7549102902412415,
+      "learning_rate": 0.0002,
+      "loss": 1.4319,
+      "step": 6630
+    },
+    {
+      "epoch": 4.44891122278057,
+      "grad_norm": 0.7831804156303406,
+      "learning_rate": 0.0002,
+      "loss": 1.388,
+      "step": 6640
+    },
+    {
+      "epoch": 4.455611390284757,
+      "grad_norm": 0.7269673943519592,
+      "learning_rate": 0.0002,
+      "loss": 1.4283,
+      "step": 6650
+    },
+    {
+      "epoch": 4.4623115577889445,
+      "grad_norm": 0.7397838830947876,
+      "learning_rate": 0.0002,
+      "loss": 1.4132,
+      "step": 6660
+    },
+    {
+      "epoch": 4.469011725293132,
+      "grad_norm": 0.713707447052002,
+      "learning_rate": 0.0002,
+      "loss": 1.3174,
+      "step": 6670
+    },
+    {
+      "epoch": 4.47571189279732,
+      "grad_norm": 0.7525581121444702,
+      "learning_rate": 0.0002,
+      "loss": 1.3406,
+      "step": 6680
+    },
+    {
+      "epoch": 4.482412060301508,
+      "grad_norm": 0.8030191659927368,
+      "learning_rate": 0.0002,
+      "loss": 1.4283,
+      "step": 6690
+    },
+    {
+      "epoch": 4.489112227805695,
+      "grad_norm": 0.7469439506530762,
+      "learning_rate": 0.0002,
+      "loss": 1.4586,
+      "step": 6700
+    },
+    {
+      "epoch": 4.4958123953098825,
+      "grad_norm": 0.7743868231773376,
+      "learning_rate": 0.0002,
+      "loss": 1.367,
+      "step": 6710
+    },
+    {
+      "epoch": 4.50251256281407,
+      "grad_norm": 0.6539737582206726,
+      "learning_rate": 0.0002,
+      "loss": 1.3439,
+      "step": 6720
+    },
+    {
+      "epoch": 4.509212730318258,
+      "grad_norm": 0.825818657875061,
+      "learning_rate": 0.0002,
+      "loss": 1.4513,
+      "step": 6730
+    },
+    {
+      "epoch": 4.515912897822446,
+      "grad_norm": 0.8048575520515442,
+      "learning_rate": 0.0002,
+      "loss": 1.3984,
+      "step": 6740
+    },
+    {
+      "epoch": 4.522613065326633,
+      "grad_norm": 0.7828766107559204,
+      "learning_rate": 0.0002,
+      "loss": 1.3923,
+      "step": 6750
+    },
+    {
+      "epoch": 4.5293132328308205,
+      "grad_norm": 0.7406010031700134,
+      "learning_rate": 0.0002,
+      "loss": 1.3886,
+      "step": 6760
+    },
+    {
+      "epoch": 4.536013400335008,
+      "grad_norm": 0.840345561504364,
+      "learning_rate": 0.0002,
+      "loss": 1.3109,
+      "step": 6770
+    },
+    {
+      "epoch": 4.542713567839196,
+      "grad_norm": 0.8492622971534729,
+      "learning_rate": 0.0002,
+      "loss": 1.4808,
+      "step": 6780
+    },
+    {
+      "epoch": 4.549413735343384,
+      "grad_norm": 0.7130163908004761,
+      "learning_rate": 0.0002,
+      "loss": 1.4384,
+      "step": 6790
+    },
+    {
+      "epoch": 4.556113902847571,
+      "grad_norm": 0.8454728126525879,
+      "learning_rate": 0.0002,
+      "loss": 1.4531,
+      "step": 6800
+    },
+    {
+      "epoch": 4.562814070351759,
+      "grad_norm": 0.7847645282745361,
+      "learning_rate": 0.0002,
+      "loss": 1.3239,
+      "step": 6810
+    },
+    {
+      "epoch": 4.569514237855946,
+      "grad_norm": 0.7245864272117615,
+      "learning_rate": 0.0002,
+      "loss": 1.4181,
+      "step": 6820
+    },
+    {
+      "epoch": 4.576214405360134,
+      "grad_norm": 0.768893301486969,
+      "learning_rate": 0.0002,
+      "loss": 1.3233,
+      "step": 6830
+    },
+    {
+      "epoch": 4.582914572864322,
+      "grad_norm": 0.8028400540351868,
+      "learning_rate": 0.0002,
+      "loss": 1.3932,
+      "step": 6840
+    },
+    {
+      "epoch": 4.589614740368509,
+      "grad_norm": 0.763945460319519,
+      "learning_rate": 0.0002,
+      "loss": 1.3745,
+      "step": 6850
+    },
+    {
+      "epoch": 4.596314907872697,
+      "grad_norm": 0.7417685389518738,
+      "learning_rate": 0.0002,
+      "loss": 1.4797,
+      "step": 6860
+    },
+    {
+      "epoch": 4.603015075376884,
+      "grad_norm": 0.7603038549423218,
+      "learning_rate": 0.0002,
+      "loss": 1.4468,
+      "step": 6870
+    },
+    {
+      "epoch": 4.609715242881072,
+      "grad_norm": 0.7981528043746948,
+      "learning_rate": 0.0002,
+      "loss": 1.4095,
+      "step": 6880
+    },
+    {
+      "epoch": 4.61641541038526,
+      "grad_norm": 0.8077111840248108,
+      "learning_rate": 0.0002,
+      "loss": 1.3963,
+      "step": 6890
+    },
+    {
+      "epoch": 4.623115577889447,
+      "grad_norm": 0.8778454065322876,
+      "learning_rate": 0.0002,
+      "loss": 1.4721,
+      "step": 6900
+    },
+    {
+      "epoch": 4.629815745393635,
+      "grad_norm": 0.8620710372924805,
+      "learning_rate": 0.0002,
+      "loss": 1.3758,
+      "step": 6910
+    },
+    {
+      "epoch": 4.636515912897822,
+      "grad_norm": 0.7486072778701782,
+      "learning_rate": 0.0002,
+      "loss": 1.344,
+      "step": 6920
+    },
+    {
+      "epoch": 4.64321608040201,
+      "grad_norm": 0.7493042945861816,
+      "learning_rate": 0.0002,
+      "loss": 1.3913,
+      "step": 6930
+    },
+    {
+      "epoch": 4.649916247906198,
+      "grad_norm": 0.7388978600502014,
+      "learning_rate": 0.0002,
+      "loss": 1.397,
+      "step": 6940
+    },
+    {
+      "epoch": 4.656616415410385,
+      "grad_norm": 0.798530638217926,
+      "learning_rate": 0.0002,
+      "loss": 1.3593,
+      "step": 6950
+    },
+    {
+      "epoch": 4.663316582914573,
+      "grad_norm": 0.7929500937461853,
+      "learning_rate": 0.0002,
+      "loss": 1.3982,
+      "step": 6960
+    },
+    {
+      "epoch": 4.67001675041876,
+      "grad_norm": 0.9186785221099854,
+      "learning_rate": 0.0002,
+      "loss": 1.4183,
+      "step": 6970
+    },
+    {
+      "epoch": 4.676716917922948,
+      "grad_norm": 1.1103485822677612,
+      "learning_rate": 0.0002,
+      "loss": 1.3955,
+      "step": 6980
+    },
+    {
+      "epoch": 4.683417085427136,
+      "grad_norm": 0.8000466823577881,
+      "learning_rate": 0.0002,
+      "loss": 1.3941,
+      "step": 6990
+    },
+    {
+      "epoch": 4.690117252931323,
+      "grad_norm": 0.7520599961280823,
+      "learning_rate": 0.0002,
+      "loss": 1.371,
+      "step": 7000
+    },
+    {
+      "epoch": 4.696817420435511,
+      "grad_norm": 0.7971973419189453,
+      "learning_rate": 0.0002,
+      "loss": 1.4582,
+      "step": 7010
+    },
+    {
+      "epoch": 4.703517587939698,
+      "grad_norm": 0.7363343834877014,
+      "learning_rate": 0.0002,
+      "loss": 1.3682,
+      "step": 7020
+    },
+    {
+      "epoch": 4.710217755443886,
+      "grad_norm": 0.8268865942955017,
+      "learning_rate": 0.0002,
+      "loss": 1.3889,
+      "step": 7030
+    },
+    {
+      "epoch": 4.716917922948074,
+      "grad_norm": 0.7054963111877441,
+      "learning_rate": 0.0002,
+      "loss": 1.4382,
+      "step": 7040
+    },
+    {
+      "epoch": 4.723618090452261,
+      "grad_norm": 0.8196262121200562,
+      "learning_rate": 0.0002,
+      "loss": 1.4578,
+      "step": 7050
+    },
+    {
+      "epoch": 4.730318257956449,
+      "grad_norm": 0.8276031017303467,
+      "learning_rate": 0.0002,
+      "loss": 1.365,
+      "step": 7060
+    },
+    {
+      "epoch": 4.7370184254606365,
+      "grad_norm": 0.8248157501220703,
+      "learning_rate": 0.0002,
+      "loss": 1.3887,
+      "step": 7070
+    },
+    {
+      "epoch": 4.743718592964824,
+      "grad_norm": 0.8937979936599731,
+      "learning_rate": 0.0002,
+      "loss": 1.4193,
+      "step": 7080
+    },
+    {
+      "epoch": 4.750418760469012,
+      "grad_norm": 0.7984238862991333,
+      "learning_rate": 0.0002,
+      "loss": 1.4334,
+      "step": 7090
+    },
+    {
+      "epoch": 4.757118927973199,
+      "grad_norm": 0.9495313763618469,
+      "learning_rate": 0.0002,
+      "loss": 1.4385,
+      "step": 7100
+    },
+    {
+      "epoch": 4.763819095477387,
+      "grad_norm": 0.8598204255104065,
+      "learning_rate": 0.0002,
+      "loss": 1.4504,
+      "step": 7110
+    },
+    {
+      "epoch": 4.7705192629815745,
+      "grad_norm": 0.8951472640037537,
+      "learning_rate": 0.0002,
+      "loss": 1.3969,
+      "step": 7120
+    },
+    {
+      "epoch": 4.777219430485762,
+      "grad_norm": 0.9110309481620789,
+      "learning_rate": 0.0002,
+      "loss": 1.4339,
+      "step": 7130
+    },
+    {
+      "epoch": 4.78391959798995,
+      "grad_norm": 0.7929584980010986,
+      "learning_rate": 0.0002,
+      "loss": 1.4001,
+      "step": 7140
+    },
+    {
+      "epoch": 4.790619765494137,
+      "grad_norm": 0.7415322661399841,
+      "learning_rate": 0.0002,
+      "loss": 1.467,
+      "step": 7150
+    },
+    {
+      "epoch": 4.797319932998325,
+      "grad_norm": 0.7504757046699524,
+      "learning_rate": 0.0002,
+      "loss": 1.5107,
+      "step": 7160
+    },
+    {
+      "epoch": 4.8040201005025125,
+      "grad_norm": 0.7166924476623535,
+      "learning_rate": 0.0002,
+      "loss": 1.3736,
+      "step": 7170
+    },
+    {
+      "epoch": 4.8107202680067,
+      "grad_norm": 0.7728400826454163,
+      "learning_rate": 0.0002,
+      "loss": 1.4088,
+      "step": 7180
+    },
+    {
+      "epoch": 4.817420435510888,
+      "grad_norm": 0.7992154955863953,
+      "learning_rate": 0.0002,
+      "loss": 1.3814,
+      "step": 7190
+    },
+    {
+      "epoch": 4.824120603015075,
+      "grad_norm": 0.8655321002006531,
+      "learning_rate": 0.0002,
+      "loss": 1.3958,
+      "step": 7200
+    },
+    {
+      "epoch": 4.830820770519263,
+      "grad_norm": 0.7672632336616516,
+      "learning_rate": 0.0002,
+      "loss": 1.3837,
+      "step": 7210
+    },
+    {
+      "epoch": 4.8375209380234505,
+      "grad_norm": 0.708416223526001,
+      "learning_rate": 0.0002,
+      "loss": 1.4578,
+      "step": 7220
+    },
+    {
+      "epoch": 4.844221105527638,
+      "grad_norm": 0.8914081454277039,
+      "learning_rate": 0.0002,
+      "loss": 1.5413,
+      "step": 7230
+    },
+    {
+      "epoch": 4.850921273031826,
+      "grad_norm": 0.7141931653022766,
+      "learning_rate": 0.0002,
+      "loss": 1.3569,
+      "step": 7240
+    },
+    {
+      "epoch": 4.857621440536013,
+      "grad_norm": 0.6913040280342102,
+      "learning_rate": 0.0002,
+      "loss": 1.4532,
+      "step": 7250
+    },
+    {
+      "epoch": 4.864321608040201,
+      "grad_norm": 0.7871233820915222,
+      "learning_rate": 0.0002,
+      "loss": 1.3912,
+      "step": 7260
+    },
+    {
+      "epoch": 4.8710217755443885,
+      "grad_norm": 0.8466277122497559,
+      "learning_rate": 0.0002,
+      "loss": 1.3688,
+      "step": 7270
+    },
+    {
+      "epoch": 4.877721943048576,
+      "grad_norm": 0.8492183685302734,
+      "learning_rate": 0.0002,
+      "loss": 1.33,
+      "step": 7280
+    },
+    {
+      "epoch": 4.884422110552764,
+      "grad_norm": 0.8339574933052063,
+      "learning_rate": 0.0002,
+      "loss": 1.3744,
+      "step": 7290
+    },
+    {
+      "epoch": 4.891122278056951,
+      "grad_norm": 0.787022590637207,
+      "learning_rate": 0.0002,
+      "loss": 1.4157,
+      "step": 7300
+    },
+    {
+      "epoch": 4.897822445561139,
+      "grad_norm": 0.8877332806587219,
+      "learning_rate": 0.0002,
+      "loss": 1.3725,
+      "step": 7310
+    },
+    {
+      "epoch": 4.9045226130653266,
+      "grad_norm": 0.744989812374115,
+      "learning_rate": 0.0002,
+      "loss": 1.3968,
+      "step": 7320
+    },
+    {
+      "epoch": 4.911222780569514,
+      "grad_norm": 0.8027268648147583,
+      "learning_rate": 0.0002,
+      "loss": 1.4421,
+      "step": 7330
+    },
+    {
+      "epoch": 4.917922948073702,
+      "grad_norm": 0.6437455415725708,
+      "learning_rate": 0.0002,
+      "loss": 1.425,
+      "step": 7340
+    },
+    {
+      "epoch": 4.924623115577889,
+      "grad_norm": 0.685999870300293,
+      "learning_rate": 0.0002,
+      "loss": 1.4829,
+      "step": 7350
+    },
+    {
+      "epoch": 4.931323283082077,
+      "grad_norm": 0.9086187481880188,
+      "learning_rate": 0.0002,
+      "loss": 1.4352,
+      "step": 7360
+    },
+    {
+      "epoch": 4.938023450586265,
+      "grad_norm": 0.8272411227226257,
+      "learning_rate": 0.0002,
+      "loss": 1.4245,
+      "step": 7370
+    },
+    {
+      "epoch": 4.944723618090452,
+      "grad_norm": 0.9227852821350098,
+      "learning_rate": 0.0002,
+      "loss": 1.4226,
+      "step": 7380
+    },
+    {
+      "epoch": 4.95142378559464,
+      "grad_norm": 0.7688441276550293,
+      "learning_rate": 0.0002,
+      "loss": 1.3643,
+      "step": 7390
+    },
+    {
+      "epoch": 4.958123953098827,
+      "grad_norm": 0.8662643432617188,
+      "learning_rate": 0.0002,
+      "loss": 1.4491,
+      "step": 7400
+    },
+    {
+      "epoch": 4.964824120603015,
+      "grad_norm": 0.9234127998352051,
+      "learning_rate": 0.0002,
+      "loss": 1.4194,
+      "step": 7410
+    },
+    {
+      "epoch": 4.971524288107203,
+      "grad_norm": 0.9131470918655396,
+      "learning_rate": 0.0002,
+      "loss": 1.4009,
+      "step": 7420
+    },
+    {
+      "epoch": 4.97822445561139,
+      "grad_norm": 0.7377504110336304,
+      "learning_rate": 0.0002,
+      "loss": 1.4544,
+      "step": 7430
+    },
+    {
+      "epoch": 4.984924623115578,
+      "grad_norm": 0.8762801289558411,
+      "learning_rate": 0.0002,
+      "loss": 1.4008,
+      "step": 7440
+    },
+    {
+      "epoch": 4.991624790619765,
+      "grad_norm": 0.7919872999191284,
+      "learning_rate": 0.0002,
+      "loss": 1.4304,
+      "step": 7450
+    },
+    {
+      "epoch": 4.998324958123953,
+      "grad_norm": 0.7144299149513245,
+      "learning_rate": 0.0002,
+      "loss": 1.3817,
+      "step": 7460
+    },
+    {
+      "epoch": 4.99966499162479,
+      "eval_loss": 1.9291157722473145,
+      "eval_runtime": 37.9831,
+      "eval_samples_per_second": 13.559,
+      "eval_steps_per_second": 1.711,
+      "step": 7462
+    },
+    {
+      "epoch": 5.005025125628141,
+      "grad_norm": 0.7860151529312134,
+      "learning_rate": 0.0002,
+      "loss": 1.2753,
+      "step": 7470
+    },
+    {
+      "epoch": 5.011725293132328,
+      "grad_norm": 0.9418314695358276,
+      "learning_rate": 0.0002,
+      "loss": 1.2149,
+      "step": 7480
+    },
+    {
+      "epoch": 5.018425460636516,
+      "grad_norm": 0.8474572896957397,
+      "learning_rate": 0.0002,
+      "loss": 1.1966,
+      "step": 7490
+    },
+    {
+      "epoch": 5.025125628140704,
+      "grad_norm": 1.0724040269851685,
+      "learning_rate": 0.0002,
+      "loss": 1.2111,
+      "step": 7500
+    },
+    {
+      "epoch": 5.031825795644891,
+      "grad_norm": 0.9109148979187012,
+      "learning_rate": 0.0002,
+      "loss": 1.2228,
+      "step": 7510
+    },
+    {
+      "epoch": 5.038525963149079,
+      "grad_norm": 1.0088659524917603,
+      "learning_rate": 0.0002,
+      "loss": 1.2239,
+      "step": 7520
+    },
+    {
+      "epoch": 5.045226130653266,
+      "grad_norm": 1.1421623229980469,
+      "learning_rate": 0.0002,
+      "loss": 1.2156,
+      "step": 7530
+    },
+    {
+      "epoch": 5.051926298157454,
+      "grad_norm": 0.9219902157783508,
+      "learning_rate": 0.0002,
+      "loss": 1.1739,
+      "step": 7540
+    },
+    {
+      "epoch": 5.058626465661642,
+      "grad_norm": 0.9150987863540649,
+      "learning_rate": 0.0002,
+      "loss": 1.2686,
+      "step": 7550
+    },
+    {
+      "epoch": 5.065326633165829,
+      "grad_norm": 0.8889328241348267,
+      "learning_rate": 0.0002,
+      "loss": 1.2068,
+      "step": 7560
+    },
+    {
+      "epoch": 5.072026800670017,
+      "grad_norm": 0.9751363396644592,
+      "learning_rate": 0.0002,
+      "loss": 1.276,
+      "step": 7570
+    },
+    {
+      "epoch": 5.078726968174204,
+      "grad_norm": 0.8603123426437378,
+      "learning_rate": 0.0002,
+      "loss": 1.2078,
+      "step": 7580
+    },
+    {
+      "epoch": 5.085427135678392,
+      "grad_norm": 0.8910616636276245,
+      "learning_rate": 0.0002,
+      "loss": 1.2175,
+      "step": 7590
+    },
+    {
+      "epoch": 5.09212730318258,
+      "grad_norm": 1.1128392219543457,
+      "learning_rate": 0.0002,
+      "loss": 1.2475,
+      "step": 7600
+    },
+    {
+      "epoch": 5.098827470686767,
+      "grad_norm": 0.9480258822441101,
+      "learning_rate": 0.0002,
+      "loss": 1.3065,
+      "step": 7610
+    },
+    {
+      "epoch": 5.105527638190955,
+      "grad_norm": 0.906958818435669,
+      "learning_rate": 0.0002,
+      "loss": 1.193,
+      "step": 7620
+    },
+    {
+      "epoch": 5.1122278056951425,
+      "grad_norm": 0.8741167187690735,
+      "learning_rate": 0.0002,
+      "loss": 1.2223,
+      "step": 7630
+    },
+    {
+      "epoch": 5.11892797319933,
+      "grad_norm": 0.966268002986908,
+      "learning_rate": 0.0002,
+      "loss": 1.2126,
+      "step": 7640
+    },
+    {
+      "epoch": 5.125628140703517,
+      "grad_norm": 0.9124358892440796,
+      "learning_rate": 0.0002,
+      "loss": 1.2782,
+      "step": 7650
+    },
+    {
+      "epoch": 5.132328308207705,
+      "grad_norm": 1.0436606407165527,
+      "learning_rate": 0.0002,
+      "loss": 1.3004,
+      "step": 7660
+    },
+    {
+      "epoch": 5.139028475711893,
+      "grad_norm": 0.9217309355735779,
+      "learning_rate": 0.0002,
+      "loss": 1.2675,
+      "step": 7670
+    },
+    {
+      "epoch": 5.1457286432160805,
+      "grad_norm": 1.344765543937683,
+      "learning_rate": 0.0002,
+      "loss": 1.2502,
+      "step": 7680
+    },
+    {
+      "epoch": 5.152428810720268,
+      "grad_norm": 1.0730723142623901,
+      "learning_rate": 0.0002,
+      "loss": 1.2416,
+      "step": 7690
+    },
+    {
+      "epoch": 5.159128978224456,
+      "grad_norm": 0.9321247339248657,
+      "learning_rate": 0.0002,
+      "loss": 1.1888,
+      "step": 7700
+    },
+    {
+      "epoch": 5.165829145728643,
+      "grad_norm": 0.8482614755630493,
+      "learning_rate": 0.0002,
+      "loss": 1.1941,
+      "step": 7710
+    },
+    {
+      "epoch": 5.172529313232831,
+      "grad_norm": 0.8274452686309814,
+      "learning_rate": 0.0002,
+      "loss": 1.2668,
+      "step": 7720
+    },
+    {
+      "epoch": 5.1792294807370185,
+      "grad_norm": 0.9120376706123352,
+      "learning_rate": 0.0002,
+      "loss": 1.1972,
+      "step": 7730
+    },
+    {
+      "epoch": 5.185929648241206,
+      "grad_norm": 1.0062892436981201,
+      "learning_rate": 0.0002,
+      "loss": 1.1648,
+      "step": 7740
+    },
+    {
+      "epoch": 5.192629815745394,
+      "grad_norm": 0.9521504640579224,
+      "learning_rate": 0.0002,
+      "loss": 1.2199,
+      "step": 7750
+    },
+    {
+      "epoch": 5.199329983249581,
+      "grad_norm": 0.8800198435783386,
+      "learning_rate": 0.0002,
+      "loss": 1.2855,
+      "step": 7760
+    },
+    {
+      "epoch": 5.206030150753769,
+      "grad_norm": 0.9749179482460022,
+      "learning_rate": 0.0002,
+      "loss": 1.2535,
+      "step": 7770
+    },
+    {
+      "epoch": 5.2127303182579565,
+      "grad_norm": 0.9441686868667603,
+      "learning_rate": 0.0002,
+      "loss": 1.2975,
+      "step": 7780
+    },
+    {
+      "epoch": 5.219430485762144,
+      "grad_norm": 0.9114066362380981,
+      "learning_rate": 0.0002,
+      "loss": 1.256,
+      "step": 7790
+    },
+    {
+      "epoch": 5.226130653266332,
+      "grad_norm": 0.9851446151733398,
+      "learning_rate": 0.0002,
+      "loss": 1.2621,
+      "step": 7800
+    },
+    {
+      "epoch": 5.232830820770519,
+      "grad_norm": 0.9526297450065613,
+      "learning_rate": 0.0002,
+      "loss": 1.2502,
+      "step": 7810
+    },
+    {
+      "epoch": 5.239530988274707,
+      "grad_norm": 1.05986487865448,
+      "learning_rate": 0.0002,
+      "loss": 1.1502,
+      "step": 7820
+    },
+    {
+      "epoch": 5.2462311557788945,
+      "grad_norm": 0.8956538438796997,
+      "learning_rate": 0.0002,
+      "loss": 1.2517,
+      "step": 7830
+    },
+    {
+      "epoch": 5.252931323283082,
+      "grad_norm": 0.9568153619766235,
+      "learning_rate": 0.0002,
+      "loss": 1.2556,
+      "step": 7840
+    },
+    {
+      "epoch": 5.259631490787269,
+      "grad_norm": 1.0035018920898438,
+      "learning_rate": 0.0002,
+      "loss": 1.2442,
+      "step": 7850
+    },
+    {
+      "epoch": 5.266331658291457,
+      "grad_norm": 0.8554368615150452,
+      "learning_rate": 0.0002,
+      "loss": 1.2605,
+      "step": 7860
+    },
+    {
+      "epoch": 5.273031825795645,
+      "grad_norm": 0.9677708148956299,
+      "learning_rate": 0.0002,
+      "loss": 1.2799,
+      "step": 7870
+    },
+    {
+      "epoch": 5.279731993299833,
+      "grad_norm": 0.943606436252594,
+      "learning_rate": 0.0002,
+      "loss": 1.275,
+      "step": 7880
+    },
+    {
+      "epoch": 5.28643216080402,
+      "grad_norm": 1.0029335021972656,
+      "learning_rate": 0.0002,
+      "loss": 1.2335,
+      "step": 7890
+    },
+    {
+      "epoch": 5.293132328308207,
+      "grad_norm": 1.0164015293121338,
+      "learning_rate": 0.0002,
+      "loss": 1.2494,
+      "step": 7900
+    },
+    {
+      "epoch": 5.299832495812395,
+      "grad_norm": 0.8908365368843079,
+      "learning_rate": 0.0002,
+      "loss": 1.3117,
+      "step": 7910
+    },
+    {
+      "epoch": 5.306532663316583,
+      "grad_norm": 0.9307826161384583,
+      "learning_rate": 0.0002,
+      "loss": 1.2832,
+      "step": 7920
+    },
+    {
+      "epoch": 5.313232830820771,
+      "grad_norm": 1.0730371475219727,
+      "learning_rate": 0.0002,
+      "loss": 1.242,
+      "step": 7930
+    },
+    {
+      "epoch": 5.319932998324958,
+      "grad_norm": 0.844739556312561,
+      "learning_rate": 0.0002,
+      "loss": 1.2003,
+      "step": 7940
+    },
+    {
+      "epoch": 5.326633165829146,
+      "grad_norm": 1.275833010673523,
+      "learning_rate": 0.0002,
+      "loss": 1.2688,
+      "step": 7950
+    },
+    {
+      "epoch": 5.333333333333333,
+      "grad_norm": 0.9042661190032959,
+      "learning_rate": 0.0002,
+      "loss": 1.2957,
+      "step": 7960
+    },
+    {
+      "epoch": 5.340033500837521,
+      "grad_norm": 0.9374269247055054,
+      "learning_rate": 0.0002,
+      "loss": 1.2912,
+      "step": 7970
+    },
+    {
+      "epoch": 5.346733668341709,
+      "grad_norm": 1.033098578453064,
+      "learning_rate": 0.0002,
+      "loss": 1.2721,
+      "step": 7980
+    },
+    {
+      "epoch": 5.353433835845896,
+      "grad_norm": 1.062775731086731,
+      "learning_rate": 0.0002,
+      "loss": 1.3208,
+      "step": 7990
+    },
+    {
+      "epoch": 5.360134003350084,
+      "grad_norm": 1.1064317226409912,
+      "learning_rate": 0.0002,
+      "loss": 1.3065,
+      "step": 8000
+    },
+    {
+      "epoch": 5.366834170854271,
+      "grad_norm": 1.1114039421081543,
+      "learning_rate": 0.0002,
+      "loss": 1.2341,
+      "step": 8010
+    },
+    {
+      "epoch": 5.373534338358459,
+      "grad_norm": 1.0198014974594116,
+      "learning_rate": 0.0002,
+      "loss": 1.2255,
+      "step": 8020
+    },
+    {
+      "epoch": 5.380234505862647,
+      "grad_norm": 0.8443173170089722,
+      "learning_rate": 0.0002,
+      "loss": 1.2433,
+      "step": 8030
+    },
+    {
+      "epoch": 5.386934673366834,
+      "grad_norm": 1.000881314277649,
+      "learning_rate": 0.0002,
+      "loss": 1.206,
+      "step": 8040
+    },
+    {
+      "epoch": 5.393634840871022,
+      "grad_norm": 0.9874443411827087,
+      "learning_rate": 0.0002,
+      "loss": 1.2982,
+      "step": 8050
+    },
+    {
+      "epoch": 5.400335008375209,
+      "grad_norm": 0.9895344972610474,
+      "learning_rate": 0.0002,
+      "loss": 1.2289,
+      "step": 8060
+    },
+    {
+      "epoch": 5.407035175879397,
+      "grad_norm": 0.8595236539840698,
+      "learning_rate": 0.0002,
+      "loss": 1.249,
+      "step": 8070
+    },
+    {
+      "epoch": 5.413735343383585,
+      "grad_norm": 0.9523849487304688,
+      "learning_rate": 0.0002,
+      "loss": 1.2308,
+      "step": 8080
+    },
+    {
+      "epoch": 5.420435510887772,
+      "grad_norm": 1.0560476779937744,
+      "learning_rate": 0.0002,
+      "loss": 1.2343,
+      "step": 8090
+    },
+    {
+      "epoch": 5.42713567839196,
+      "grad_norm": 1.0893689393997192,
+      "learning_rate": 0.0002,
+      "loss": 1.2956,
+      "step": 8100
+    },
+    {
+      "epoch": 5.433835845896147,
+      "grad_norm": 0.9395513534545898,
+      "learning_rate": 0.0002,
+      "loss": 1.2846,
+      "step": 8110
+    },
+    {
+      "epoch": 5.440536013400335,
+      "grad_norm": 0.9364215135574341,
+      "learning_rate": 0.0002,
+      "loss": 1.3444,
+      "step": 8120
+    },
+    {
+      "epoch": 5.447236180904523,
+      "grad_norm": 0.9502208232879639,
+      "learning_rate": 0.0002,
+      "loss": 1.2944,
+      "step": 8130
+    },
+    {
+      "epoch": 5.45393634840871,
+      "grad_norm": 0.9559208154678345,
+      "learning_rate": 0.0002,
+      "loss": 1.2971,
+      "step": 8140
+    },
+    {
+      "epoch": 5.460636515912898,
+      "grad_norm": 0.9261730313301086,
+      "learning_rate": 0.0002,
+      "loss": 1.2495,
+      "step": 8150
+    },
+    {
+      "epoch": 5.467336683417085,
+      "grad_norm": 0.9832326173782349,
+      "learning_rate": 0.0002,
+      "loss": 1.2599,
+      "step": 8160
+    },
+    {
+      "epoch": 5.474036850921273,
+      "grad_norm": 1.065953016281128,
+      "learning_rate": 0.0002,
+      "loss": 1.2771,
+      "step": 8170
+    },
+    {
+      "epoch": 5.480737018425461,
+      "grad_norm": 0.9139469861984253,
+      "learning_rate": 0.0002,
+      "loss": 1.3617,
+      "step": 8180
+    },
+    {
+      "epoch": 5.4874371859296485,
+      "grad_norm": 1.2322484254837036,
+      "learning_rate": 0.0002,
+      "loss": 1.2197,
+      "step": 8190
+    },
+    {
+      "epoch": 5.494137353433836,
+      "grad_norm": 0.9722974896430969,
+      "learning_rate": 0.0002,
+      "loss": 1.2879,
+      "step": 8200
+    },
+    {
+      "epoch": 5.500837520938023,
+      "grad_norm": 0.9338926076889038,
+      "learning_rate": 0.0002,
+      "loss": 1.2664,
+      "step": 8210
+    },
+    {
+      "epoch": 5.507537688442211,
+      "grad_norm": 0.9283728003501892,
+      "learning_rate": 0.0002,
+      "loss": 1.2128,
+      "step": 8220
+    },
+    {
+      "epoch": 5.514237855946399,
+      "grad_norm": 1.0489585399627686,
+      "learning_rate": 0.0002,
+      "loss": 1.2141,
+      "step": 8230
+    },
+    {
+      "epoch": 5.5209380234505865,
+      "grad_norm": 0.9881814122200012,
+      "learning_rate": 0.0002,
+      "loss": 1.3257,
+      "step": 8240
+    },
+    {
+      "epoch": 5.527638190954773,
+      "grad_norm": 0.9274460077285767,
+      "learning_rate": 0.0002,
+      "loss": 1.2221,
+      "step": 8250
+    },
+    {
+      "epoch": 5.534338358458961,
+      "grad_norm": 0.8650718331336975,
+      "learning_rate": 0.0002,
+      "loss": 1.2241,
+      "step": 8260
+    },
+    {
+      "epoch": 5.541038525963149,
+      "grad_norm": 1.014069676399231,
+      "learning_rate": 0.0002,
+      "loss": 1.2462,
+      "step": 8270
+    },
+    {
+      "epoch": 5.547738693467337,
+      "grad_norm": 0.9212974905967712,
+      "learning_rate": 0.0002,
+      "loss": 1.3502,
+      "step": 8280
+    },
+    {
+      "epoch": 5.5544388609715245,
+      "grad_norm": 1.1235398054122925,
+      "learning_rate": 0.0002,
+      "loss": 1.2779,
+      "step": 8290
+    },
+    {
+      "epoch": 5.561139028475711,
+      "grad_norm": 0.961954653263092,
+      "learning_rate": 0.0002,
+      "loss": 1.306,
+      "step": 8300
+    },
+    {
+      "epoch": 5.567839195979899,
+      "grad_norm": 0.9386700391769409,
+      "learning_rate": 0.0002,
+      "loss": 1.2946,
+      "step": 8310
+    },
+    {
+      "epoch": 5.574539363484087,
+      "grad_norm": 1.01912522315979,
+      "learning_rate": 0.0002,
+      "loss": 1.313,
+      "step": 8320
+    },
+    {
+      "epoch": 5.581239530988275,
+      "grad_norm": 0.9851216077804565,
+      "learning_rate": 0.0002,
+      "loss": 1.3121,
+      "step": 8330
+    },
+    {
+      "epoch": 5.5879396984924625,
+      "grad_norm": 1.0138001441955566,
+      "learning_rate": 0.0002,
+      "loss": 1.3071,
+      "step": 8340
+    },
+    {
+      "epoch": 5.594639865996649,
+      "grad_norm": 0.9262447357177734,
+      "learning_rate": 0.0002,
+      "loss": 1.2257,
+      "step": 8350
+    },
+    {
+      "epoch": 5.601340033500837,
+      "grad_norm": 1.1322970390319824,
+      "learning_rate": 0.0002,
+      "loss": 1.2473,
+      "step": 8360
+    },
+    {
+      "epoch": 5.608040201005025,
+      "grad_norm": 1.1429349184036255,
+      "learning_rate": 0.0002,
+      "loss": 1.3098,
+      "step": 8370
+    },
+    {
+      "epoch": 5.614740368509213,
+      "grad_norm": 0.9130118489265442,
+      "learning_rate": 0.0002,
+      "loss": 1.2686,
+      "step": 8380
+    },
+    {
+      "epoch": 5.6214405360134005,
+      "grad_norm": 0.9651545882225037,
+      "learning_rate": 0.0002,
+      "loss": 1.2541,
+      "step": 8390
+    },
+    {
+      "epoch": 5.628140703517588,
+      "grad_norm": 0.9595398306846619,
+      "learning_rate": 0.0002,
+      "loss": 1.2799,
+      "step": 8400
+    },
+    {
+      "epoch": 5.634840871021775,
+      "grad_norm": 1.0049372911453247,
+      "learning_rate": 0.0002,
+      "loss": 1.3429,
+      "step": 8410
+    },
+    {
+      "epoch": 5.641541038525963,
+      "grad_norm": 1.082804560661316,
+      "learning_rate": 0.0002,
+      "loss": 1.3224,
+      "step": 8420
+    },
+    {
+      "epoch": 5.648241206030151,
+      "grad_norm": 0.9489204287528992,
+      "learning_rate": 0.0002,
+      "loss": 1.297,
+      "step": 8430
+    },
+    {
+      "epoch": 5.654941373534339,
+      "grad_norm": 0.9470235109329224,
+      "learning_rate": 0.0002,
+      "loss": 1.3424,
+      "step": 8440
+    },
+    {
+      "epoch": 5.661641541038526,
+      "grad_norm": 1.0662927627563477,
+      "learning_rate": 0.0002,
+      "loss": 1.3358,
+      "step": 8450
+    },
+    {
+      "epoch": 5.668341708542713,
+      "grad_norm": 0.9097877740859985,
+      "learning_rate": 0.0002,
+      "loss": 1.2973,
+      "step": 8460
+    },
+    {
+      "epoch": 5.675041876046901,
+      "grad_norm": 0.9740368127822876,
+      "learning_rate": 0.0002,
+      "loss": 1.3072,
+      "step": 8470
+    },
+    {
+      "epoch": 5.681742043551089,
+      "grad_norm": 0.9878810048103333,
+      "learning_rate": 0.0002,
+      "loss": 1.286,
+      "step": 8480
+    },
+    {
+      "epoch": 5.688442211055277,
+      "grad_norm": 1.148260474205017,
+      "learning_rate": 0.0002,
+      "loss": 1.208,
+      "step": 8490
+    },
+    {
+      "epoch": 5.695142378559464,
+      "grad_norm": 0.9632558822631836,
+      "learning_rate": 0.0002,
+      "loss": 1.2842,
+      "step": 8500
+    },
+    {
+      "epoch": 5.701842546063651,
+      "grad_norm": 0.876812756061554,
+      "learning_rate": 0.0002,
+      "loss": 1.2787,
+      "step": 8510
+    },
+    {
+      "epoch": 5.708542713567839,
+      "grad_norm": 1.0730829238891602,
+      "learning_rate": 0.0002,
+      "loss": 1.3186,
+      "step": 8520
+    },
+    {
+      "epoch": 5.715242881072027,
+      "grad_norm": 1.2239218950271606,
+      "learning_rate": 0.0002,
+      "loss": 1.2856,
+      "step": 8530
+    },
+    {
+      "epoch": 5.721943048576215,
+      "grad_norm": 0.9460835456848145,
+      "learning_rate": 0.0002,
+      "loss": 1.2717,
+      "step": 8540
+    },
+    {
+      "epoch": 5.728643216080402,
+      "grad_norm": 0.9086270928382874,
+      "learning_rate": 0.0002,
+      "loss": 1.3509,
+      "step": 8550
+    },
+    {
+      "epoch": 5.735343383584589,
+      "grad_norm": 1.0258867740631104,
+      "learning_rate": 0.0002,
+      "loss": 1.2971,
+      "step": 8560
+    },
+    {
+      "epoch": 5.742043551088777,
+      "grad_norm": 1.0543923377990723,
+      "learning_rate": 0.0002,
+      "loss": 1.3581,
+      "step": 8570
+    },
+    {
+      "epoch": 5.748743718592965,
+      "grad_norm": 0.9063900113105774,
+      "learning_rate": 0.0002,
+      "loss": 1.2988,
+      "step": 8580
+    },
+    {
+      "epoch": 5.755443886097153,
+      "grad_norm": 1.1838830709457397,
+      "learning_rate": 0.0002,
+      "loss": 1.3535,
+      "step": 8590
+    },
+    {
+      "epoch": 5.76214405360134,
+      "grad_norm": 0.9631859064102173,
+      "learning_rate": 0.0002,
+      "loss": 1.2655,
+      "step": 8600
+    },
+    {
+      "epoch": 5.768844221105527,
+      "grad_norm": 0.9702655673027039,
+      "learning_rate": 0.0002,
+      "loss": 1.276,
+      "step": 8610
+    },
+    {
+      "epoch": 5.775544388609715,
+      "grad_norm": 1.0591435432434082,
+      "learning_rate": 0.0002,
+      "loss": 1.3196,
+      "step": 8620
+    },
+    {
+      "epoch": 5.782244556113903,
+      "grad_norm": 0.9989570379257202,
+      "learning_rate": 0.0002,
+      "loss": 1.267,
+      "step": 8630
+    },
+    {
+      "epoch": 5.788944723618091,
+      "grad_norm": 1.0836435556411743,
+      "learning_rate": 0.0002,
+      "loss": 1.3227,
+      "step": 8640
+    },
+    {
+      "epoch": 5.795644891122278,
+      "grad_norm": 0.8832896947860718,
+      "learning_rate": 0.0002,
+      "loss": 1.3334,
+      "step": 8650
+    },
+    {
+      "epoch": 5.802345058626465,
+      "grad_norm": 1.0104607343673706,
+      "learning_rate": 0.0002,
+      "loss": 1.3214,
+      "step": 8660
+    },
+    {
+      "epoch": 5.809045226130653,
+      "grad_norm": 0.8375084400177002,
+      "learning_rate": 0.0002,
+      "loss": 1.2703,
+      "step": 8670
+    },
+    {
+      "epoch": 5.815745393634841,
+      "grad_norm": 1.1300716400146484,
+      "learning_rate": 0.0002,
+      "loss": 1.3554,
+      "step": 8680
+    },
+    {
+      "epoch": 5.822445561139029,
+      "grad_norm": 0.9311910271644592,
+      "learning_rate": 0.0002,
+      "loss": 1.3468,
+      "step": 8690
+    },
+    {
+      "epoch": 5.8291457286432165,
+      "grad_norm": 0.9488391876220703,
+      "learning_rate": 0.0002,
+      "loss": 1.2749,
+      "step": 8700
+    },
+    {
+      "epoch": 5.835845896147403,
+      "grad_norm": 0.9747629761695862,
+      "learning_rate": 0.0002,
+      "loss": 1.2281,
+      "step": 8710
+    },
+    {
+      "epoch": 5.842546063651591,
+      "grad_norm": 1.1029598712921143,
+      "learning_rate": 0.0002,
+      "loss": 1.2923,
+      "step": 8720
+    },
+    {
+      "epoch": 5.849246231155779,
+      "grad_norm": 1.0396875143051147,
+      "learning_rate": 0.0002,
+      "loss": 1.3613,
+      "step": 8730
+    },
+    {
+      "epoch": 5.855946398659967,
+      "grad_norm": 0.9259780645370483,
+      "learning_rate": 0.0002,
+      "loss": 1.3272,
+      "step": 8740
+    },
+    {
+      "epoch": 5.8626465661641545,
+      "grad_norm": 1.020033597946167,
+      "learning_rate": 0.0002,
+      "loss": 1.3236,
+      "step": 8750
+    },
+    {
+      "epoch": 5.869346733668341,
+      "grad_norm": 0.9191218614578247,
+      "learning_rate": 0.0002,
+      "loss": 1.3453,
+      "step": 8760
+    },
+    {
+      "epoch": 5.876046901172529,
+      "grad_norm": 1.1093107461929321,
+      "learning_rate": 0.0002,
+      "loss": 1.3012,
+      "step": 8770
+    },
+    {
+      "epoch": 5.882747068676717,
+      "grad_norm": 1.1626793146133423,
+      "learning_rate": 0.0002,
+      "loss": 1.2718,
+      "step": 8780
+    },
+    {
+      "epoch": 5.889447236180905,
+      "grad_norm": 0.9542945027351379,
+      "learning_rate": 0.0002,
+      "loss": 1.2969,
+      "step": 8790
+    },
+    {
+      "epoch": 5.8961474036850925,
+      "grad_norm": 0.9086058139801025,
+      "learning_rate": 0.0002,
+      "loss": 1.3134,
+      "step": 8800
+    },
+    {
+      "epoch": 5.902847571189279,
+      "grad_norm": 0.9249639511108398,
+      "learning_rate": 0.0002,
+      "loss": 1.2731,
+      "step": 8810
+    },
+    {
+      "epoch": 5.909547738693467,
+      "grad_norm": 0.9414396286010742,
+      "learning_rate": 0.0002,
+      "loss": 1.337,
+      "step": 8820
+    },
+    {
+      "epoch": 5.916247906197655,
+      "grad_norm": 0.9086037874221802,
+      "learning_rate": 0.0002,
+      "loss": 1.2865,
+      "step": 8830
+    },
+    {
+      "epoch": 5.922948073701843,
+      "grad_norm": 0.8685907125473022,
+      "learning_rate": 0.0002,
+      "loss": 1.2756,
+      "step": 8840
+    },
+    {
+      "epoch": 5.9296482412060305,
+      "grad_norm": 1.036419153213501,
+      "learning_rate": 0.0002,
+      "loss": 1.297,
+      "step": 8850
+    },
+    {
+      "epoch": 5.936348408710217,
+      "grad_norm": 1.0183674097061157,
+      "learning_rate": 0.0002,
+      "loss": 1.3207,
+      "step": 8860
+    },
+    {
+      "epoch": 5.943048576214405,
+      "grad_norm": 0.966444194316864,
+      "learning_rate": 0.0002,
+      "loss": 1.3922,
+      "step": 8870
+    },
+    {
+      "epoch": 5.949748743718593,
+      "grad_norm": 1.125693917274475,
+      "learning_rate": 0.0002,
+      "loss": 1.333,
+      "step": 8880
+    },
+    {
+      "epoch": 5.956448911222781,
+      "grad_norm": 0.9857436418533325,
+      "learning_rate": 0.0002,
+      "loss": 1.3116,
+      "step": 8890
+    },
+    {
+      "epoch": 5.9631490787269685,
+      "grad_norm": 0.9377069473266602,
+      "learning_rate": 0.0002,
+      "loss": 1.2526,
+      "step": 8900
+    },
+    {
+      "epoch": 5.969849246231155,
+      "grad_norm": 0.9493814706802368,
+      "learning_rate": 0.0002,
+      "loss": 1.3221,
+      "step": 8910
+    },
+    {
+      "epoch": 5.976549413735343,
+      "grad_norm": 0.8806208372116089,
+      "learning_rate": 0.0002,
+      "loss": 1.2516,
+      "step": 8920
+    },
+    {
+      "epoch": 5.983249581239531,
+      "grad_norm": 0.8727600574493408,
+      "learning_rate": 0.0002,
+      "loss": 1.2558,
+      "step": 8930
+    },
+    {
+      "epoch": 5.989949748743719,
+      "grad_norm": 0.9799810647964478,
+      "learning_rate": 0.0002,
+      "loss": 1.3538,
+      "step": 8940
+    },
+    {
+      "epoch": 5.9966499162479066,
+      "grad_norm": 0.9866513609886169,
+      "learning_rate": 0.0002,
+      "loss": 1.3323,
+      "step": 8950
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 2.0282373428344727,
+      "eval_runtime": 38.0375,
+      "eval_samples_per_second": 13.539,
+      "eval_steps_per_second": 1.709,
+      "step": 8955
+    },
+    {
+      "epoch": 6.0033500837520934,
+      "grad_norm": 0.8747885227203369,
+      "learning_rate": 0.0002,
+      "loss": 1.1768,
+      "step": 8960
+    },
+    {
+      "epoch": 6.010050251256281,
+      "grad_norm": 1.2512741088867188,
+      "learning_rate": 0.0002,
+      "loss": 1.0677,
+      "step": 8970
+    },
+    {
+      "epoch": 6.016750418760469,
+      "grad_norm": 1.06855309009552,
+      "learning_rate": 0.0002,
+      "loss": 1.1128,
+      "step": 8980
+    },
+    {
+      "epoch": 6.023450586264657,
+      "grad_norm": 1.1868711709976196,
+      "learning_rate": 0.0002,
+      "loss": 1.1382,
+      "step": 8990
+    },
+    {
+      "epoch": 6.030150753768845,
+      "grad_norm": 1.2984495162963867,
+      "learning_rate": 0.0002,
+      "loss": 1.1377,
+      "step": 9000
+    },
+    {
+      "epoch": 6.0368509212730315,
+      "grad_norm": 1.1147589683532715,
+      "learning_rate": 0.0002,
+      "loss": 1.0803,
+      "step": 9010
+    },
+    {
+      "epoch": 6.043551088777219,
+      "grad_norm": 1.3128414154052734,
+      "learning_rate": 0.0002,
+      "loss": 1.1244,
+      "step": 9020
+    },
+    {
+      "epoch": 6.050251256281407,
+      "grad_norm": 1.068290114402771,
+      "learning_rate": 0.0002,
+      "loss": 1.097,
+      "step": 9030
+    },
+    {
+      "epoch": 6.056951423785595,
+      "grad_norm": 1.1890562772750854,
+      "learning_rate": 0.0002,
+      "loss": 1.1764,
+      "step": 9040
+    },
+    {
+      "epoch": 6.063651591289783,
+      "grad_norm": 1.2121573686599731,
+      "learning_rate": 0.0002,
+      "loss": 1.1239,
+      "step": 9050
+    },
+    {
+      "epoch": 6.0703517587939695,
+      "grad_norm": 1.0860483646392822,
+      "learning_rate": 0.0002,
+      "loss": 1.0683,
+      "step": 9060
+    },
+    {
+      "epoch": 6.077051926298157,
+      "grad_norm": 1.1214599609375,
+      "learning_rate": 0.0002,
+      "loss": 1.1613,
+      "step": 9070
+    },
+    {
+      "epoch": 6.083752093802345,
+      "grad_norm": 1.147580862045288,
+      "learning_rate": 0.0002,
+      "loss": 1.1001,
+      "step": 9080
+    },
+    {
+      "epoch": 6.090452261306533,
+      "grad_norm": 1.3233155012130737,
+      "learning_rate": 0.0002,
+      "loss": 1.1154,
+      "step": 9090
+    },
+    {
+      "epoch": 6.097152428810721,
+      "grad_norm": 1.1869080066680908,
+      "learning_rate": 0.0002,
+      "loss": 1.1017,
+      "step": 9100
+    },
+    {
+      "epoch": 6.1038525963149075,
+      "grad_norm": 1.1695014238357544,
+      "learning_rate": 0.0002,
+      "loss": 1.1532,
+      "step": 9110
+    },
+    {
+      "epoch": 6.110552763819095,
+      "grad_norm": 1.1982251405715942,
+      "learning_rate": 0.0002,
+      "loss": 1.1366,
+      "step": 9120
+    },
+    {
+      "epoch": 6.117252931323283,
+      "grad_norm": 1.1426950693130493,
+      "learning_rate": 0.0002,
+      "loss": 1.0805,
+      "step": 9130
+    },
+    {
+      "epoch": 6.123953098827471,
+      "grad_norm": 1.2257394790649414,
+      "learning_rate": 0.0002,
+      "loss": 1.0801,
+      "step": 9140
+    },
+    {
+      "epoch": 6.130653266331659,
+      "grad_norm": 1.2932263612747192,
+      "learning_rate": 0.0002,
+      "loss": 1.1209,
+      "step": 9150
+    },
+    {
+      "epoch": 6.1373534338358455,
+      "grad_norm": 1.2617030143737793,
+      "learning_rate": 0.0002,
+      "loss": 1.0934,
+      "step": 9160
+    },
+    {
+      "epoch": 6.144053601340033,
+      "grad_norm": 1.1201422214508057,
+      "learning_rate": 0.0002,
+      "loss": 1.0551,
+      "step": 9170
+    },
+    {
+      "epoch": 6.150753768844221,
+      "grad_norm": 0.9625319838523865,
+      "learning_rate": 0.0002,
+      "loss": 1.1059,
+      "step": 9180
+    },
+    {
+      "epoch": 6.157453936348409,
+      "grad_norm": 1.0290048122406006,
+      "learning_rate": 0.0002,
+      "loss": 1.1397,
+      "step": 9190
+    },
+    {
+      "epoch": 6.164154103852597,
+      "grad_norm": 1.1137803792953491,
+      "learning_rate": 0.0002,
+      "loss": 1.1257,
+      "step": 9200
+    },
+    {
+      "epoch": 6.1708542713567835,
+      "grad_norm": 1.3674522638320923,
+      "learning_rate": 0.0002,
+      "loss": 1.1211,
+      "step": 9210
+    },
+    {
+      "epoch": 6.177554438860971,
+      "grad_norm": 1.182207703590393,
+      "learning_rate": 0.0002,
+      "loss": 1.0947,
+      "step": 9220
+    },
+    {
+      "epoch": 6.184254606365159,
+      "grad_norm": 1.0496711730957031,
+      "learning_rate": 0.0002,
+      "loss": 1.0838,
+      "step": 9230
+    },
+    {
+      "epoch": 6.190954773869347,
+      "grad_norm": 1.1899489164352417,
+      "learning_rate": 0.0002,
+      "loss": 1.0666,
+      "step": 9240
+    },
+    {
+      "epoch": 6.197654941373535,
+      "grad_norm": 1.2666147947311401,
+      "learning_rate": 0.0002,
+      "loss": 1.1633,
+      "step": 9250
+    },
+    {
+      "epoch": 6.204355108877722,
+      "grad_norm": 1.2013030052185059,
+      "learning_rate": 0.0002,
+      "loss": 1.1532,
+      "step": 9260
+    },
+    {
+      "epoch": 6.211055276381909,
+      "grad_norm": 1.3049768209457397,
+      "learning_rate": 0.0002,
+      "loss": 1.151,
+      "step": 9270
+    },
+    {
+      "epoch": 6.217755443886097,
+      "grad_norm": 1.1733006238937378,
+      "learning_rate": 0.0002,
+      "loss": 1.0741,
+      "step": 9280
+    },
+    {
+      "epoch": 6.224455611390285,
+      "grad_norm": 1.2742516994476318,
+      "learning_rate": 0.0002,
+      "loss": 1.0933,
+      "step": 9290
+    },
+    {
+      "epoch": 6.231155778894473,
+      "grad_norm": 1.110198974609375,
+      "learning_rate": 0.0002,
+      "loss": 1.1028,
+      "step": 9300
+    },
+    {
+      "epoch": 6.23785594639866,
+      "grad_norm": 1.159963607788086,
+      "learning_rate": 0.0002,
+      "loss": 1.1619,
+      "step": 9310
+    },
+    {
+      "epoch": 6.244556113902847,
+      "grad_norm": 1.302216649055481,
+      "learning_rate": 0.0002,
+      "loss": 1.0716,
+      "step": 9320
+    },
+    {
+      "epoch": 6.251256281407035,
+      "grad_norm": 1.2134063243865967,
+      "learning_rate": 0.0002,
+      "loss": 1.0694,
+      "step": 9330
+    },
+    {
+      "epoch": 6.257956448911223,
+      "grad_norm": 1.062682867050171,
+      "learning_rate": 0.0002,
+      "loss": 1.2151,
+      "step": 9340
+    },
+    {
+      "epoch": 6.264656616415411,
+      "grad_norm": 1.1568971872329712,
+      "learning_rate": 0.0002,
+      "loss": 1.148,
+      "step": 9350
+    },
+    {
+      "epoch": 6.271356783919598,
+      "grad_norm": 0.9914957880973816,
+      "learning_rate": 0.0002,
+      "loss": 1.094,
+      "step": 9360
+    },
+    {
+      "epoch": 6.278056951423785,
+      "grad_norm": 1.017250895500183,
+      "learning_rate": 0.0002,
+      "loss": 1.125,
+      "step": 9370
+    },
+    {
+      "epoch": 6.284757118927973,
+      "grad_norm": 1.1862834692001343,
+      "learning_rate": 0.0002,
+      "loss": 1.2177,
+      "step": 9380
+    },
+    {
+      "epoch": 6.291457286432161,
+      "grad_norm": 1.2834911346435547,
+      "learning_rate": 0.0002,
+      "loss": 0.9994,
+      "step": 9390
+    },
+    {
+      "epoch": 6.298157453936349,
+      "grad_norm": 1.3306856155395508,
+      "learning_rate": 0.0002,
+      "loss": 1.0922,
+      "step": 9400
+    },
+    {
+      "epoch": 6.304857621440536,
+      "grad_norm": 1.12908136844635,
+      "learning_rate": 0.0002,
+      "loss": 1.1136,
+      "step": 9410
+    },
+    {
+      "epoch": 6.311557788944723,
+      "grad_norm": 1.2157351970672607,
+      "learning_rate": 0.0002,
+      "loss": 1.1406,
+      "step": 9420
+    },
+    {
+      "epoch": 6.318257956448911,
+      "grad_norm": 1.121882677078247,
+      "learning_rate": 0.0002,
+      "loss": 1.1388,
+      "step": 9430
+    },
+    {
+      "epoch": 6.324958123953099,
+      "grad_norm": 1.3144481182098389,
+      "learning_rate": 0.0002,
+      "loss": 1.1648,
+      "step": 9440
+    },
+    {
+      "epoch": 6.331658291457287,
+      "grad_norm": 1.1946896314620972,
+      "learning_rate": 0.0002,
+      "loss": 1.1228,
+      "step": 9450
+    },
+    {
+      "epoch": 6.338358458961474,
+      "grad_norm": 1.1289668083190918,
+      "learning_rate": 0.0002,
+      "loss": 1.1613,
+      "step": 9460
+    },
+    {
+      "epoch": 6.345058626465661,
+      "grad_norm": 1.1065658330917358,
+      "learning_rate": 0.0002,
+      "loss": 1.1059,
+      "step": 9470
+    },
+    {
+      "epoch": 6.351758793969849,
+      "grad_norm": 1.0881422758102417,
+      "learning_rate": 0.0002,
+      "loss": 1.1431,
+      "step": 9480
+    },
+    {
+      "epoch": 6.358458961474037,
+      "grad_norm": 1.242676854133606,
+      "learning_rate": 0.0002,
+      "loss": 1.223,
+      "step": 9490
+    },
+    {
+      "epoch": 6.365159128978225,
+      "grad_norm": 0.9650855660438538,
+      "learning_rate": 0.0002,
+      "loss": 1.1379,
+      "step": 9500
+    },
+    {
+      "epoch": 6.371859296482412,
+      "grad_norm": 1.2845722436904907,
+      "learning_rate": 0.0002,
+      "loss": 1.0763,
+      "step": 9510
+    },
+    {
+      "epoch": 6.3785594639865995,
+      "grad_norm": 1.0327043533325195,
+      "learning_rate": 0.0002,
+      "loss": 1.1351,
+      "step": 9520
+    },
+    {
+      "epoch": 6.385259631490787,
+      "grad_norm": 1.0780898332595825,
+      "learning_rate": 0.0002,
+      "loss": 1.114,
+      "step": 9530
+    },
+    {
+      "epoch": 6.391959798994975,
+      "grad_norm": 1.4934027194976807,
+      "learning_rate": 0.0002,
+      "loss": 1.1579,
+      "step": 9540
+    },
+    {
+      "epoch": 6.398659966499163,
+      "grad_norm": 0.9882908463478088,
+      "learning_rate": 0.0002,
+      "loss": 1.1546,
+      "step": 9550
+    },
+    {
+      "epoch": 6.40536013400335,
+      "grad_norm": 1.3250664472579956,
+      "learning_rate": 0.0002,
+      "loss": 1.1145,
+      "step": 9560
+    },
+    {
+      "epoch": 6.4120603015075375,
+      "grad_norm": 1.1888482570648193,
+      "learning_rate": 0.0002,
+      "loss": 1.2333,
+      "step": 9570
+    },
+    {
+      "epoch": 6.418760469011725,
+      "grad_norm": 1.136496901512146,
+      "learning_rate": 0.0002,
+      "loss": 1.0892,
+      "step": 9580
+    },
+    {
+      "epoch": 6.425460636515913,
+      "grad_norm": 1.161360502243042,
+      "learning_rate": 0.0002,
+      "loss": 1.1674,
+      "step": 9590
+    },
+    {
+      "epoch": 6.432160804020101,
+      "grad_norm": 1.2034236192703247,
+      "learning_rate": 0.0002,
+      "loss": 1.1293,
+      "step": 9600
+    },
+    {
+      "epoch": 6.438860971524288,
+      "grad_norm": 1.0268361568450928,
+      "learning_rate": 0.0002,
+      "loss": 1.1059,
+      "step": 9610
+    },
+    {
+      "epoch": 6.4455611390284755,
+      "grad_norm": 1.2132930755615234,
+      "learning_rate": 0.0002,
+      "loss": 1.1732,
+      "step": 9620
+    },
+    {
+      "epoch": 6.452261306532663,
+      "grad_norm": 1.0773013830184937,
+      "learning_rate": 0.0002,
+      "loss": 1.1329,
+      "step": 9630
+    },
+    {
+      "epoch": 6.458961474036851,
+      "grad_norm": 1.3848375082015991,
+      "learning_rate": 0.0002,
+      "loss": 1.0822,
+      "step": 9640
+    },
+    {
+      "epoch": 6.465661641541039,
+      "grad_norm": 1.110495924949646,
+      "learning_rate": 0.0002,
+      "loss": 1.1778,
+      "step": 9650
+    },
+    {
+      "epoch": 6.472361809045226,
+      "grad_norm": 1.118093729019165,
+      "learning_rate": 0.0002,
+      "loss": 1.2022,
+      "step": 9660
+    },
+    {
+      "epoch": 6.4790619765494135,
+      "grad_norm": 1.2611900568008423,
+      "learning_rate": 0.0002,
+      "loss": 1.1222,
+      "step": 9670
+    },
+    {
+      "epoch": 6.485762144053601,
+      "grad_norm": 0.971754252910614,
+      "learning_rate": 0.0002,
+      "loss": 1.2138,
+      "step": 9680
+    },
+    {
+      "epoch": 6.492462311557789,
+      "grad_norm": 1.2615419626235962,
+      "learning_rate": 0.0002,
+      "loss": 1.1641,
+      "step": 9690
+    },
+    {
+      "epoch": 6.499162479061977,
+      "grad_norm": 1.1370900869369507,
+      "learning_rate": 0.0002,
+      "loss": 1.1412,
+      "step": 9700
+    },
+    {
+      "epoch": 6.505862646566165,
+      "grad_norm": 1.1815906763076782,
+      "learning_rate": 0.0002,
+      "loss": 1.186,
+      "step": 9710
+    },
+    {
+      "epoch": 6.5125628140703515,
+      "grad_norm": 1.3424339294433594,
+      "learning_rate": 0.0002,
+      "loss": 1.167,
+      "step": 9720
+    },
+    {
+      "epoch": 6.519262981574539,
+      "grad_norm": 1.2858397960662842,
+      "learning_rate": 0.0002,
+      "loss": 1.1602,
+      "step": 9730
+    },
+    {
+      "epoch": 6.525963149078727,
+      "grad_norm": 0.9578179121017456,
+      "learning_rate": 0.0002,
+      "loss": 1.178,
+      "step": 9740
+    },
+    {
+      "epoch": 6.532663316582915,
+      "grad_norm": 1.3105167150497437,
+      "learning_rate": 0.0002,
+      "loss": 1.1805,
+      "step": 9750
+    },
+    {
+      "epoch": 6.539363484087103,
+      "grad_norm": 1.0586575269699097,
+      "learning_rate": 0.0002,
+      "loss": 1.1899,
+      "step": 9760
+    },
+    {
+      "epoch": 6.54606365159129,
+      "grad_norm": 1.2122068405151367,
+      "learning_rate": 0.0002,
+      "loss": 1.095,
+      "step": 9770
+    },
+    {
+      "epoch": 6.552763819095477,
+      "grad_norm": 1.3088626861572266,
+      "learning_rate": 0.0002,
+      "loss": 1.1471,
+      "step": 9780
+    },
+    {
+      "epoch": 6.559463986599665,
+      "grad_norm": 1.194122076034546,
+      "learning_rate": 0.0002,
+      "loss": 1.1067,
+      "step": 9790
+    },
+    {
+      "epoch": 6.566164154103853,
+      "grad_norm": 1.1508387327194214,
+      "learning_rate": 0.0002,
+      "loss": 1.0967,
+      "step": 9800
+    },
+    {
+      "epoch": 6.572864321608041,
+      "grad_norm": 1.109228253364563,
+      "learning_rate": 0.0002,
+      "loss": 1.1694,
+      "step": 9810
+    },
+    {
+      "epoch": 6.579564489112228,
+      "grad_norm": 1.1607427597045898,
+      "learning_rate": 0.0002,
+      "loss": 1.1378,
+      "step": 9820
+    },
+    {
+      "epoch": 6.586264656616415,
+      "grad_norm": 1.174089789390564,
+      "learning_rate": 0.0002,
+      "loss": 1.1585,
+      "step": 9830
+    },
+    {
+      "epoch": 6.592964824120603,
+      "grad_norm": 1.1739521026611328,
+      "learning_rate": 0.0002,
+      "loss": 1.1385,
+      "step": 9840
+    },
+    {
+      "epoch": 6.599664991624791,
+      "grad_norm": 1.098528504371643,
+      "learning_rate": 0.0002,
+      "loss": 1.155,
+      "step": 9850
+    },
+    {
+      "epoch": 6.606365159128979,
+      "grad_norm": 1.0397740602493286,
+      "learning_rate": 0.0002,
+      "loss": 1.1359,
+      "step": 9860
+    },
+    {
+      "epoch": 6.613065326633166,
+      "grad_norm": 1.1087969541549683,
+      "learning_rate": 0.0002,
+      "loss": 1.1433,
+      "step": 9870
+    },
+    {
+      "epoch": 6.619765494137353,
+      "grad_norm": 1.2070481777191162,
+      "learning_rate": 0.0002,
+      "loss": 1.2356,
+      "step": 9880
+    },
+    {
+      "epoch": 6.626465661641541,
+      "grad_norm": 1.1115655899047852,
+      "learning_rate": 0.0002,
+      "loss": 1.1161,
+      "step": 9890
+    },
+    {
+      "epoch": 6.633165829145729,
+      "grad_norm": 1.2486097812652588,
+      "learning_rate": 0.0002,
+      "loss": 1.2163,
+      "step": 9900
+    },
+    {
+      "epoch": 6.639865996649917,
+      "grad_norm": 1.230380654335022,
+      "learning_rate": 0.0002,
+      "loss": 1.0984,
+      "step": 9910
+    },
+    {
+      "epoch": 6.646566164154104,
+      "grad_norm": 1.1479365825653076,
+      "learning_rate": 0.0002,
+      "loss": 1.1862,
+      "step": 9920
+    },
+    {
+      "epoch": 6.653266331658291,
+      "grad_norm": 1.0790960788726807,
+      "learning_rate": 0.0002,
+      "loss": 1.1139,
+      "step": 9930
+    },
+    {
+      "epoch": 6.659966499162479,
+      "grad_norm": 1.1157397031784058,
+      "learning_rate": 0.0002,
+      "loss": 1.2001,
+      "step": 9940
+    },
+    {
+      "epoch": 6.666666666666667,
+      "grad_norm": 1.3104028701782227,
+      "learning_rate": 0.0002,
+      "loss": 1.1085,
+      "step": 9950
+    },
+    {
+      "epoch": 6.673366834170855,
+      "grad_norm": 1.1727646589279175,
+      "learning_rate": 0.0002,
+      "loss": 1.2209,
+      "step": 9960
+    },
+    {
+      "epoch": 6.680067001675042,
+      "grad_norm": 1.2104284763336182,
+      "learning_rate": 0.0002,
+      "loss": 1.1671,
+      "step": 9970
+    },
+    {
+      "epoch": 6.686767169179229,
+      "grad_norm": 1.2023727893829346,
+      "learning_rate": 0.0002,
+      "loss": 1.1952,
+      "step": 9980
+    },
+    {
+      "epoch": 6.693467336683417,
+      "grad_norm": 1.0088225603103638,
+      "learning_rate": 0.0002,
+      "loss": 1.1385,
+      "step": 9990
+    },
+    {
+      "epoch": 6.700167504187605,
+      "grad_norm": 1.298015832901001,
+      "learning_rate": 0.0002,
+      "loss": 1.1314,
+      "step": 10000
+    },
+    {
+      "epoch": 6.706867671691793,
+      "grad_norm": 1.1315910816192627,
+      "learning_rate": 0.0002,
+      "loss": 1.1983,
+      "step": 10010
+    },
+    {
+      "epoch": 6.71356783919598,
+      "grad_norm": 1.1283273696899414,
+      "learning_rate": 0.0002,
+      "loss": 1.1679,
+      "step": 10020
+    },
+    {
+      "epoch": 6.720268006700167,
+      "grad_norm": 1.2564418315887451,
+      "learning_rate": 0.0002,
+      "loss": 1.1917,
+      "step": 10030
+    },
+    {
+      "epoch": 6.726968174204355,
+      "grad_norm": 1.0451353788375854,
+      "learning_rate": 0.0002,
+      "loss": 1.1788,
+      "step": 10040
+    },
+    {
+      "epoch": 6.733668341708543,
+      "grad_norm": 1.054793357849121,
+      "learning_rate": 0.0002,
+      "loss": 1.1905,
+      "step": 10050
+    },
+    {
+      "epoch": 6.740368509212731,
+      "grad_norm": 1.2741243839263916,
+      "learning_rate": 0.0002,
+      "loss": 1.1814,
+      "step": 10060
+    },
+    {
+      "epoch": 6.747068676716918,
+      "grad_norm": 1.1342514753341675,
+      "learning_rate": 0.0002,
+      "loss": 1.2015,
+      "step": 10070
+    },
+    {
+      "epoch": 6.7537688442211055,
+      "grad_norm": 1.0081498622894287,
+      "learning_rate": 0.0002,
+      "loss": 1.2587,
+      "step": 10080
+    },
+    {
+      "epoch": 6.760469011725293,
+      "grad_norm": 1.2164603471755981,
+      "learning_rate": 0.0002,
+      "loss": 1.1226,
+      "step": 10090
+    },
+    {
+      "epoch": 6.767169179229481,
+      "grad_norm": 1.2062463760375977,
+      "learning_rate": 0.0002,
+      "loss": 1.1353,
+      "step": 10100
+    },
+    {
+      "epoch": 6.773869346733669,
+      "grad_norm": 1.2255526781082153,
+      "learning_rate": 0.0002,
+      "loss": 1.2143,
+      "step": 10110
+    },
+    {
+      "epoch": 6.780569514237856,
+      "grad_norm": 1.08175790309906,
+      "learning_rate": 0.0002,
+      "loss": 1.1544,
+      "step": 10120
+    },
+    {
+      "epoch": 6.7872696817420435,
+      "grad_norm": 1.5781128406524658,
+      "learning_rate": 0.0002,
+      "loss": 1.1983,
+      "step": 10130
+    },
+    {
+      "epoch": 6.793969849246231,
+      "grad_norm": 1.0622451305389404,
+      "learning_rate": 0.0002,
+      "loss": 1.0994,
+      "step": 10140
+    },
+    {
+      "epoch": 6.800670016750419,
+      "grad_norm": 1.1591497659683228,
+      "learning_rate": 0.0002,
+      "loss": 1.2084,
+      "step": 10150
+    },
+    {
+      "epoch": 6.807370184254607,
+      "grad_norm": 1.0398483276367188,
+      "learning_rate": 0.0002,
+      "loss": 1.2203,
+      "step": 10160
+    },
+    {
+      "epoch": 6.814070351758794,
+      "grad_norm": 1.229132056236267,
+      "learning_rate": 0.0002,
+      "loss": 1.2249,
+      "step": 10170
+    },
+    {
+      "epoch": 6.8207705192629815,
+      "grad_norm": 1.0918090343475342,
+      "learning_rate": 0.0002,
+      "loss": 1.1789,
+      "step": 10180
+    },
+    {
+      "epoch": 6.827470686767169,
+      "grad_norm": 1.1543749570846558,
+      "learning_rate": 0.0002,
+      "loss": 1.1639,
+      "step": 10190
+    },
+    {
+      "epoch": 6.834170854271357,
+      "grad_norm": 1.1831817626953125,
+      "learning_rate": 0.0002,
+      "loss": 1.1353,
+      "step": 10200
+    },
+    {
+      "epoch": 6.840871021775545,
+      "grad_norm": 1.305327296257019,
+      "learning_rate": 0.0002,
+      "loss": 1.2565,
+      "step": 10210
+    },
+    {
+      "epoch": 6.847571189279732,
+      "grad_norm": 1.136720061302185,
+      "learning_rate": 0.0002,
+      "loss": 1.2037,
+      "step": 10220
+    },
+    {
+      "epoch": 6.8542713567839195,
+      "grad_norm": 1.2282346487045288,
+      "learning_rate": 0.0002,
+      "loss": 1.2256,
+      "step": 10230
+    },
+    {
+      "epoch": 6.860971524288107,
+      "grad_norm": 1.2457010746002197,
+      "learning_rate": 0.0002,
+      "loss": 1.1281,
+      "step": 10240
+    },
+    {
+      "epoch": 6.867671691792295,
+      "grad_norm": 1.2808631658554077,
+      "learning_rate": 0.0002,
+      "loss": 1.2762,
+      "step": 10250
+    },
+    {
+      "epoch": 6.874371859296483,
+      "grad_norm": 1.089066743850708,
+      "learning_rate": 0.0002,
+      "loss": 1.2213,
+      "step": 10260
+    },
+    {
+      "epoch": 6.88107202680067,
+      "grad_norm": 0.9543178081512451,
+      "learning_rate": 0.0002,
+      "loss": 1.2627,
+      "step": 10270
+    },
+    {
+      "epoch": 6.8877721943048575,
+      "grad_norm": 1.1149744987487793,
+      "learning_rate": 0.0002,
+      "loss": 1.1617,
+      "step": 10280
+    },
+    {
+      "epoch": 6.894472361809045,
+      "grad_norm": 1.0185538530349731,
+      "learning_rate": 0.0002,
+      "loss": 1.1134,
+      "step": 10290
+    },
+    {
+      "epoch": 6.901172529313233,
+      "grad_norm": 0.9954617619514465,
+      "learning_rate": 0.0002,
+      "loss": 1.217,
+      "step": 10300
+    },
+    {
+      "epoch": 6.907872696817421,
+      "grad_norm": 1.2581418752670288,
+      "learning_rate": 0.0002,
+      "loss": 1.1524,
+      "step": 10310
+    },
+    {
+      "epoch": 6.914572864321608,
+      "grad_norm": 1.2430983781814575,
+      "learning_rate": 0.0002,
+      "loss": 1.1942,
+      "step": 10320
+    },
+    {
+      "epoch": 6.921273031825796,
+      "grad_norm": 1.4937270879745483,
+      "learning_rate": 0.0002,
+      "loss": 1.1254,
+      "step": 10330
+    },
+    {
+      "epoch": 6.927973199329983,
+      "grad_norm": 1.1257144212722778,
+      "learning_rate": 0.0002,
+      "loss": 1.1519,
+      "step": 10340
+    },
+    {
+      "epoch": 6.934673366834171,
+      "grad_norm": 1.2068904638290405,
+      "learning_rate": 0.0002,
+      "loss": 1.2622,
+      "step": 10350
+    },
+    {
+      "epoch": 6.941373534338359,
+      "grad_norm": 1.0290757417678833,
+      "learning_rate": 0.0002,
+      "loss": 1.1349,
+      "step": 10360
+    },
+    {
+      "epoch": 6.948073701842546,
+      "grad_norm": 1.0070724487304688,
+      "learning_rate": 0.0002,
+      "loss": 1.1752,
+      "step": 10370
+    },
+    {
+      "epoch": 6.954773869346734,
+      "grad_norm": 0.9936357140541077,
+      "learning_rate": 0.0002,
+      "loss": 1.1838,
+      "step": 10380
+    },
+    {
+      "epoch": 6.961474036850921,
+      "grad_norm": 1.1063416004180908,
+      "learning_rate": 0.0002,
+      "loss": 1.2305,
+      "step": 10390
+    },
+    {
+      "epoch": 6.968174204355109,
+      "grad_norm": 1.5199986696243286,
+      "learning_rate": 0.0002,
+      "loss": 1.154,
+      "step": 10400
+    },
+    {
+      "epoch": 6.974874371859297,
+      "grad_norm": 1.160731554031372,
+      "learning_rate": 0.0002,
+      "loss": 1.1143,
+      "step": 10410
+    },
+    {
+      "epoch": 6.981574539363484,
+      "grad_norm": 1.084697961807251,
+      "learning_rate": 0.0002,
+      "loss": 1.2132,
+      "step": 10420
+    },
+    {
+      "epoch": 6.988274706867672,
+      "grad_norm": 1.1257576942443848,
+      "learning_rate": 0.0002,
+      "loss": 1.1995,
+      "step": 10430
+    },
+    {
+      "epoch": 6.994974874371859,
+      "grad_norm": 1.310616135597229,
+      "learning_rate": 0.0002,
+      "loss": 1.1197,
+      "step": 10440
+    },
+    {
+      "epoch": 6.99966499162479,
+      "eval_loss": 2.1203012466430664,
+      "eval_runtime": 37.936,
+      "eval_samples_per_second": 13.576,
+      "eval_steps_per_second": 1.713,
+      "step": 10447
+    },
+    {
+      "epoch": 7.001675041876047,
+      "grad_norm": 1.0176491737365723,
+      "learning_rate": 0.0002,
+      "loss": 1.1118,
+      "step": 10450
+    },
+    {
+      "epoch": 7.008375209380235,
+      "grad_norm": 1.602665662765503,
+      "learning_rate": 0.0002,
+      "loss": 0.9932,
+      "step": 10460
+    },
+    {
+      "epoch": 7.015075376884422,
+      "grad_norm": 1.2909572124481201,
+      "learning_rate": 0.0002,
+      "loss": 0.9818,
+      "step": 10470
+    },
+    {
+      "epoch": 7.02177554438861,
+      "grad_norm": 1.2601855993270874,
+      "learning_rate": 0.0002,
+      "loss": 0.9398,
+      "step": 10480
+    },
+    {
+      "epoch": 7.028475711892797,
+      "grad_norm": 1.3437587022781372,
+      "learning_rate": 0.0002,
+      "loss": 0.9241,
+      "step": 10490
+    },
+    {
+      "epoch": 7.035175879396985,
+      "grad_norm": 1.2220089435577393,
+      "learning_rate": 0.0002,
+      "loss": 1.0031,
+      "step": 10500
+    },
+    {
+      "epoch": 7.041876046901173,
+      "grad_norm": 1.3392685651779175,
+      "learning_rate": 0.0002,
+      "loss": 0.9428,
+      "step": 10510
+    },
+    {
+      "epoch": 7.04857621440536,
+      "grad_norm": 1.3902767896652222,
+      "learning_rate": 0.0002,
+      "loss": 0.9566,
+      "step": 10520
+    },
+    {
+      "epoch": 7.055276381909548,
+      "grad_norm": 1.4098035097122192,
+      "learning_rate": 0.0002,
+      "loss": 0.9993,
+      "step": 10530
+    },
+    {
+      "epoch": 7.061976549413735,
+      "grad_norm": 1.38866126537323,
+      "learning_rate": 0.0002,
+      "loss": 0.9683,
+      "step": 10540
+    },
+    {
+      "epoch": 7.068676716917923,
+      "grad_norm": 1.3638999462127686,
+      "learning_rate": 0.0002,
+      "loss": 0.9961,
+      "step": 10550
+    },
+    {
+      "epoch": 7.075376884422111,
+      "grad_norm": 1.3181349039077759,
+      "learning_rate": 0.0002,
+      "loss": 0.9698,
+      "step": 10560
+    },
+    {
+      "epoch": 7.082077051926298,
+      "grad_norm": 1.2528657913208008,
+      "learning_rate": 0.0002,
+      "loss": 0.9963,
+      "step": 10570
+    },
+    {
+      "epoch": 7.088777219430486,
+      "grad_norm": 1.4326589107513428,
+      "learning_rate": 0.0002,
+      "loss": 0.9624,
+      "step": 10580
+    },
+    {
+      "epoch": 7.0954773869346734,
+      "grad_norm": 1.4312337636947632,
+      "learning_rate": 0.0002,
+      "loss": 1.0432,
+      "step": 10590
+    },
+    {
+      "epoch": 7.102177554438861,
+      "grad_norm": 1.7286990880966187,
+      "learning_rate": 0.0002,
+      "loss": 0.9704,
+      "step": 10600
+    },
+    {
+      "epoch": 7.108877721943049,
+      "grad_norm": 1.1248762607574463,
+      "learning_rate": 0.0002,
+      "loss": 0.9954,
+      "step": 10610
+    },
+    {
+      "epoch": 7.115577889447236,
+      "grad_norm": 1.278731346130371,
+      "learning_rate": 0.0002,
+      "loss": 0.9736,
+      "step": 10620
+    },
+    {
+      "epoch": 7.122278056951424,
+      "grad_norm": 1.53670072555542,
+      "learning_rate": 0.0002,
+      "loss": 0.9885,
+      "step": 10630
+    },
+    {
+      "epoch": 7.1289782244556115,
+      "grad_norm": 1.268069863319397,
+      "learning_rate": 0.0002,
+      "loss": 0.9573,
+      "step": 10640
+    },
+    {
+      "epoch": 7.135678391959799,
+      "grad_norm": 1.5072290897369385,
+      "learning_rate": 0.0002,
+      "loss": 0.9716,
+      "step": 10650
+    },
+    {
+      "epoch": 7.142378559463987,
+      "grad_norm": 1.5552845001220703,
+      "learning_rate": 0.0002,
+      "loss": 0.9871,
+      "step": 10660
+    },
+    {
+      "epoch": 7.149078726968174,
+      "grad_norm": 1.2643769979476929,
+      "learning_rate": 0.0002,
+      "loss": 1.0029,
+      "step": 10670
+    },
+    {
+      "epoch": 7.155778894472362,
+      "grad_norm": 1.347589373588562,
+      "learning_rate": 0.0002,
+      "loss": 0.9848,
+      "step": 10680
+    },
+    {
+      "epoch": 7.1624790619765495,
+      "grad_norm": 1.364700436592102,
+      "learning_rate": 0.0002,
+      "loss": 1.0193,
+      "step": 10690
+    },
+    {
+      "epoch": 7.169179229480737,
+      "grad_norm": 1.4375768899917603,
+      "learning_rate": 0.0002,
+      "loss": 1.0057,
+      "step": 10700
+    },
+    {
+      "epoch": 7.175879396984925,
+      "grad_norm": 1.426047444343567,
+      "learning_rate": 0.0002,
+      "loss": 1.0108,
+      "step": 10710
+    },
+    {
+      "epoch": 7.182579564489112,
+      "grad_norm": 1.5208740234375,
+      "learning_rate": 0.0002,
+      "loss": 0.9985,
+      "step": 10720
+    },
+    {
+      "epoch": 7.1892797319933,
+      "grad_norm": 1.4713337421417236,
+      "learning_rate": 0.0002,
+      "loss": 1.0535,
+      "step": 10730
+    },
+    {
+      "epoch": 7.1959798994974875,
+      "grad_norm": 1.3042446374893188,
+      "learning_rate": 0.0002,
+      "loss": 0.9481,
+      "step": 10740
+    },
+    {
+      "epoch": 7.202680067001675,
+      "grad_norm": 1.2290682792663574,
+      "learning_rate": 0.0002,
+      "loss": 0.9813,
+      "step": 10750
+    },
+    {
+      "epoch": 7.209380234505863,
+      "grad_norm": 1.1152390241622925,
+      "learning_rate": 0.0002,
+      "loss": 1.0437,
+      "step": 10760
+    },
+    {
+      "epoch": 7.21608040201005,
+      "grad_norm": 1.3632242679595947,
+      "learning_rate": 0.0002,
+      "loss": 0.9557,
+      "step": 10770
+    },
+    {
+      "epoch": 7.222780569514238,
+      "grad_norm": 1.2406541109085083,
+      "learning_rate": 0.0002,
+      "loss": 0.9915,
+      "step": 10780
+    },
+    {
+      "epoch": 7.2294807370184255,
+      "grad_norm": 1.1965205669403076,
+      "learning_rate": 0.0002,
+      "loss": 1.0138,
+      "step": 10790
+    },
+    {
+      "epoch": 7.236180904522613,
+      "grad_norm": 1.2895352840423584,
+      "learning_rate": 0.0002,
+      "loss": 1.0186,
+      "step": 10800
+    },
+    {
+      "epoch": 7.242881072026801,
+      "grad_norm": 1.388366937637329,
+      "learning_rate": 0.0002,
+      "loss": 0.9616,
+      "step": 10810
+    },
+    {
+      "epoch": 7.249581239530988,
+      "grad_norm": 1.1411796808242798,
+      "learning_rate": 0.0002,
+      "loss": 0.992,
+      "step": 10820
+    },
+    {
+      "epoch": 7.256281407035176,
+      "grad_norm": 1.2220646142959595,
+      "learning_rate": 0.0002,
+      "loss": 0.942,
+      "step": 10830
+    },
+    {
+      "epoch": 7.2629815745393635,
+      "grad_norm": 1.495492696762085,
+      "learning_rate": 0.0002,
+      "loss": 1.0136,
+      "step": 10840
+    },
+    {
+      "epoch": 7.269681742043551,
+      "grad_norm": 1.1395213603973389,
+      "learning_rate": 0.0002,
+      "loss": 1.0475,
+      "step": 10850
+    },
+    {
+      "epoch": 7.276381909547739,
+      "grad_norm": 1.3826487064361572,
+      "learning_rate": 0.0002,
+      "loss": 1.0134,
+      "step": 10860
+    },
+    {
+      "epoch": 7.283082077051926,
+      "grad_norm": 1.4356474876403809,
+      "learning_rate": 0.0002,
+      "loss": 0.944,
+      "step": 10870
+    },
+    {
+      "epoch": 7.289782244556114,
+      "grad_norm": 1.3617557287216187,
+      "learning_rate": 0.0002,
+      "loss": 0.971,
+      "step": 10880
+    },
+    {
+      "epoch": 7.296482412060302,
+      "grad_norm": 1.585394024848938,
+      "learning_rate": 0.0002,
+      "loss": 0.9772,
+      "step": 10890
+    },
+    {
+      "epoch": 7.303182579564489,
+      "grad_norm": 1.1442821025848389,
+      "learning_rate": 0.0002,
+      "loss": 1.0204,
+      "step": 10900
+    },
+    {
+      "epoch": 7.309882747068677,
+      "grad_norm": 1.4712985754013062,
+      "learning_rate": 0.0002,
+      "loss": 0.9608,
+      "step": 10910
+    },
+    {
+      "epoch": 7.316582914572864,
+      "grad_norm": 1.2864325046539307,
+      "learning_rate": 0.0002,
+      "loss": 1.0027,
+      "step": 10920
+    },
+    {
+      "epoch": 7.323283082077052,
+      "grad_norm": 1.308010458946228,
+      "learning_rate": 0.0002,
+      "loss": 1.0341,
+      "step": 10930
+    },
+    {
+      "epoch": 7.32998324958124,
+      "grad_norm": 1.4371414184570312,
+      "learning_rate": 0.0002,
+      "loss": 1.0096,
+      "step": 10940
+    },
+    {
+      "epoch": 7.336683417085427,
+      "grad_norm": 1.5968806743621826,
+      "learning_rate": 0.0002,
+      "loss": 0.9999,
+      "step": 10950
+    },
+    {
+      "epoch": 7.343383584589615,
+      "grad_norm": 1.3592339754104614,
+      "learning_rate": 0.0002,
+      "loss": 0.9611,
+      "step": 10960
+    },
+    {
+      "epoch": 7.350083752093802,
+      "grad_norm": 1.225520133972168,
+      "learning_rate": 0.0002,
+      "loss": 1.0505,
+      "step": 10970
+    },
+    {
+      "epoch": 7.35678391959799,
+      "grad_norm": 1.3138031959533691,
+      "learning_rate": 0.0002,
+      "loss": 0.9702,
+      "step": 10980
+    },
+    {
+      "epoch": 7.363484087102178,
+      "grad_norm": 1.2601540088653564,
+      "learning_rate": 0.0002,
+      "loss": 1.0062,
+      "step": 10990
+    },
+    {
+      "epoch": 7.370184254606365,
+      "grad_norm": 1.3437113761901855,
+      "learning_rate": 0.0002,
+      "loss": 1.0508,
+      "step": 11000
+    },
+    {
+      "epoch": 7.376884422110553,
+      "grad_norm": 1.0681836605072021,
+      "learning_rate": 0.0002,
+      "loss": 0.9649,
+      "step": 11010
+    },
+    {
+      "epoch": 7.38358458961474,
+      "grad_norm": 1.415852665901184,
+      "learning_rate": 0.0002,
+      "loss": 1.0025,
+      "step": 11020
+    },
+    {
+      "epoch": 7.390284757118928,
+      "grad_norm": 1.3147039413452148,
+      "learning_rate": 0.0002,
+      "loss": 1.0532,
+      "step": 11030
+    },
+    {
+      "epoch": 7.396984924623116,
+      "grad_norm": 1.4778614044189453,
+      "learning_rate": 0.0002,
+      "loss": 1.1188,
+      "step": 11040
+    },
+    {
+      "epoch": 7.403685092127303,
+      "grad_norm": 1.3137797117233276,
+      "learning_rate": 0.0002,
+      "loss": 0.9917,
+      "step": 11050
+    },
+    {
+      "epoch": 7.410385259631491,
+      "grad_norm": 1.1917701959609985,
+      "learning_rate": 0.0002,
+      "loss": 1.0115,
+      "step": 11060
+    },
+    {
+      "epoch": 7.417085427135678,
+      "grad_norm": 1.3999699354171753,
+      "learning_rate": 0.0002,
+      "loss": 1.0668,
+      "step": 11070
+    },
+    {
+      "epoch": 7.423785594639866,
+      "grad_norm": 1.3817322254180908,
+      "learning_rate": 0.0002,
+      "loss": 1.0311,
+      "step": 11080
+    },
+    {
+      "epoch": 7.430485762144054,
+      "grad_norm": 1.2086812257766724,
+      "learning_rate": 0.0002,
+      "loss": 1.0086,
+      "step": 11090
+    },
+    {
+      "epoch": 7.437185929648241,
+      "grad_norm": 1.1938024759292603,
+      "learning_rate": 0.0002,
+      "loss": 1.0002,
+      "step": 11100
+    },
+    {
+      "epoch": 7.443886097152429,
+      "grad_norm": 1.1543669700622559,
+      "learning_rate": 0.0002,
+      "loss": 1.0463,
+      "step": 11110
+    },
+    {
+      "epoch": 7.450586264656616,
+      "grad_norm": 1.158841848373413,
+      "learning_rate": 0.0002,
+      "loss": 1.0814,
+      "step": 11120
+    },
+    {
+      "epoch": 7.457286432160804,
+      "grad_norm": 1.2777763605117798,
+      "learning_rate": 0.0002,
+      "loss": 1.0022,
+      "step": 11130
+    },
+    {
+      "epoch": 7.463986599664992,
+      "grad_norm": 1.3375903367996216,
+      "learning_rate": 0.0002,
+      "loss": 1.0162,
+      "step": 11140
+    },
+    {
+      "epoch": 7.4706867671691795,
+      "grad_norm": 1.5573794841766357,
+      "learning_rate": 0.0002,
+      "loss": 0.9995,
+      "step": 11150
+    },
+    {
+      "epoch": 7.477386934673367,
+      "grad_norm": 1.3869640827178955,
+      "learning_rate": 0.0002,
+      "loss": 1.0698,
+      "step": 11160
+    },
+    {
+      "epoch": 7.484087102177554,
+      "grad_norm": 1.310341238975525,
+      "learning_rate": 0.0002,
+      "loss": 1.0217,
+      "step": 11170
+    },
+    {
+      "epoch": 7.490787269681742,
+      "grad_norm": 1.4249892234802246,
+      "learning_rate": 0.0002,
+      "loss": 1.0365,
+      "step": 11180
+    },
+    {
+      "epoch": 7.49748743718593,
+      "grad_norm": 1.4025905132293701,
+      "learning_rate": 0.0002,
+      "loss": 1.0346,
+      "step": 11190
+    },
+    {
+      "epoch": 7.5041876046901175,
+      "grad_norm": 1.4361881017684937,
+      "learning_rate": 0.0002,
+      "loss": 1.0328,
+      "step": 11200
+    },
+    {
+      "epoch": 7.510887772194305,
+      "grad_norm": 1.1791380643844604,
+      "learning_rate": 0.0002,
+      "loss": 1.0316,
+      "step": 11210
+    },
+    {
+      "epoch": 7.517587939698492,
+      "grad_norm": 1.415075421333313,
+      "learning_rate": 0.0002,
+      "loss": 1.0621,
+      "step": 11220
+    },
+    {
+      "epoch": 7.52428810720268,
+      "grad_norm": 1.3633697032928467,
+      "learning_rate": 0.0002,
+      "loss": 1.0265,
+      "step": 11230
+    },
+    {
+      "epoch": 7.530988274706868,
+      "grad_norm": 1.2803648710250854,
+      "learning_rate": 0.0002,
+      "loss": 0.983,
+      "step": 11240
+    },
+    {
+      "epoch": 7.5376884422110555,
+      "grad_norm": 1.4032878875732422,
+      "learning_rate": 0.0002,
+      "loss": 1.0121,
+      "step": 11250
+    },
+    {
+      "epoch": 7.544388609715243,
+      "grad_norm": 1.4507378339767456,
+      "learning_rate": 0.0002,
+      "loss": 1.032,
+      "step": 11260
+    },
+    {
+      "epoch": 7.55108877721943,
+      "grad_norm": 1.227613925933838,
+      "learning_rate": 0.0002,
+      "loss": 1.0221,
+      "step": 11270
+    },
+    {
+      "epoch": 7.557788944723618,
+      "grad_norm": 1.2620965242385864,
+      "learning_rate": 0.0002,
+      "loss": 1.0288,
+      "step": 11280
+    },
+    {
+      "epoch": 7.564489112227806,
+      "grad_norm": 1.3917821645736694,
+      "learning_rate": 0.0002,
+      "loss": 1.1263,
+      "step": 11290
+    },
+    {
+      "epoch": 7.5711892797319935,
+      "grad_norm": 1.5720019340515137,
+      "learning_rate": 0.0002,
+      "loss": 1.0628,
+      "step": 11300
+    },
+    {
+      "epoch": 7.577889447236181,
+      "grad_norm": 1.376694917678833,
+      "learning_rate": 0.0002,
+      "loss": 1.0537,
+      "step": 11310
+    },
+    {
+      "epoch": 7.584589614740368,
+      "grad_norm": 1.4403680562973022,
+      "learning_rate": 0.0002,
+      "loss": 1.012,
+      "step": 11320
+    },
+    {
+      "epoch": 7.591289782244556,
+      "grad_norm": 1.6306934356689453,
+      "learning_rate": 0.0002,
+      "loss": 1.0318,
+      "step": 11330
+    },
+    {
+      "epoch": 7.597989949748744,
+      "grad_norm": 1.2361583709716797,
+      "learning_rate": 0.0002,
+      "loss": 1.0297,
+      "step": 11340
+    },
+    {
+      "epoch": 7.6046901172529315,
+      "grad_norm": 1.2658058404922485,
+      "learning_rate": 0.0002,
+      "loss": 1.022,
+      "step": 11350
+    },
+    {
+      "epoch": 7.611390284757119,
+      "grad_norm": 1.5335173606872559,
+      "learning_rate": 0.0002,
+      "loss": 1.0357,
+      "step": 11360
+    },
+    {
+      "epoch": 7.618090452261306,
+      "grad_norm": 1.3432948589324951,
+      "learning_rate": 0.0002,
+      "loss": 1.0689,
+      "step": 11370
+    },
+    {
+      "epoch": 7.624790619765494,
+      "grad_norm": 1.374617338180542,
+      "learning_rate": 0.0002,
+      "loss": 1.0613,
+      "step": 11380
+    },
+    {
+      "epoch": 7.631490787269682,
+      "grad_norm": 1.3790302276611328,
+      "learning_rate": 0.0002,
+      "loss": 1.0722,
+      "step": 11390
+    },
+    {
+      "epoch": 7.63819095477387,
+      "grad_norm": 1.4256713390350342,
+      "learning_rate": 0.0002,
+      "loss": 1.0078,
+      "step": 11400
+    },
+    {
+      "epoch": 7.644891122278057,
+      "grad_norm": 1.271228313446045,
+      "learning_rate": 0.0002,
+      "loss": 1.0314,
+      "step": 11410
+    },
+    {
+      "epoch": 7.651591289782244,
+      "grad_norm": 1.432146668434143,
+      "learning_rate": 0.0002,
+      "loss": 1.0678,
+      "step": 11420
+    },
+    {
+      "epoch": 7.658291457286432,
+      "grad_norm": 1.2698006629943848,
+      "learning_rate": 0.0002,
+      "loss": 1.0496,
+      "step": 11430
+    },
+    {
+      "epoch": 7.66499162479062,
+      "grad_norm": 1.439496636390686,
+      "learning_rate": 0.0002,
+      "loss": 1.0678,
+      "step": 11440
+    },
+    {
+      "epoch": 7.671691792294808,
+      "grad_norm": 1.2079370021820068,
+      "learning_rate": 0.0002,
+      "loss": 1.0344,
+      "step": 11450
+    },
+    {
+      "epoch": 7.678391959798995,
+      "grad_norm": 1.310412049293518,
+      "learning_rate": 0.0002,
+      "loss": 1.047,
+      "step": 11460
+    },
+    {
+      "epoch": 7.685092127303182,
+      "grad_norm": 1.413438320159912,
+      "learning_rate": 0.0002,
+      "loss": 1.0524,
+      "step": 11470
+    },
+    {
+      "epoch": 7.69179229480737,
+      "grad_norm": 1.2390344142913818,
+      "learning_rate": 0.0002,
+      "loss": 0.9965,
+      "step": 11480
+    },
+    {
+      "epoch": 7.698492462311558,
+      "grad_norm": 1.3902971744537354,
+      "learning_rate": 0.0002,
+      "loss": 1.0477,
+      "step": 11490
+    },
+    {
+      "epoch": 7.705192629815746,
+      "grad_norm": 1.1194908618927002,
+      "learning_rate": 0.0002,
+      "loss": 0.9811,
+      "step": 11500
+    },
+    {
+      "epoch": 7.711892797319933,
+      "grad_norm": 1.625697374343872,
+      "learning_rate": 0.0002,
+      "loss": 1.0959,
+      "step": 11510
+    },
+    {
+      "epoch": 7.71859296482412,
+      "grad_norm": 1.339687466621399,
+      "learning_rate": 0.0002,
+      "loss": 1.0285,
+      "step": 11520
+    },
+    {
+      "epoch": 7.725293132328308,
+      "grad_norm": 1.336680293083191,
+      "learning_rate": 0.0002,
+      "loss": 1.032,
+      "step": 11530
+    },
+    {
+      "epoch": 7.731993299832496,
+      "grad_norm": 1.3316529989242554,
+      "learning_rate": 0.0002,
+      "loss": 1.0915,
+      "step": 11540
+    },
+    {
+      "epoch": 7.738693467336684,
+      "grad_norm": 1.2593837976455688,
+      "learning_rate": 0.0002,
+      "loss": 1.0031,
+      "step": 11550
+    },
+    {
+      "epoch": 7.745393634840871,
+      "grad_norm": 1.2159652709960938,
+      "learning_rate": 0.0002,
+      "loss": 1.019,
+      "step": 11560
+    },
+    {
+      "epoch": 7.752093802345058,
+      "grad_norm": 1.2919223308563232,
+      "learning_rate": 0.0002,
+      "loss": 1.0554,
+      "step": 11570
+    },
+    {
+      "epoch": 7.758793969849246,
+      "grad_norm": 1.2574092149734497,
+      "learning_rate": 0.0002,
+      "loss": 1.0072,
+      "step": 11580
+    },
+    {
+      "epoch": 7.765494137353434,
+      "grad_norm": 1.228236436843872,
+      "learning_rate": 0.0002,
+      "loss": 1.0379,
+      "step": 11590
+    },
+    {
+      "epoch": 7.772194304857622,
+      "grad_norm": 1.3790936470031738,
+      "learning_rate": 0.0002,
+      "loss": 1.0494,
+      "step": 11600
+    },
+    {
+      "epoch": 7.778894472361809,
+      "grad_norm": 1.419376015663147,
+      "learning_rate": 0.0002,
+      "loss": 1.0256,
+      "step": 11610
+    },
+    {
+      "epoch": 7.785594639865996,
+      "grad_norm": 1.3336344957351685,
+      "learning_rate": 0.0002,
+      "loss": 1.0985,
+      "step": 11620
+    },
+    {
+      "epoch": 7.792294807370184,
+      "grad_norm": 1.5279520750045776,
+      "learning_rate": 0.0002,
+      "loss": 1.107,
+      "step": 11630
+    },
+    {
+      "epoch": 7.798994974874372,
+      "grad_norm": 1.4296824932098389,
+      "learning_rate": 0.0002,
+      "loss": 1.1197,
+      "step": 11640
+    },
+    {
+      "epoch": 7.80569514237856,
+      "grad_norm": 1.279316782951355,
+      "learning_rate": 0.0002,
+      "loss": 1.128,
+      "step": 11650
+    },
+    {
+      "epoch": 7.812395309882747,
+      "grad_norm": 1.2764557600021362,
+      "learning_rate": 0.0002,
+      "loss": 1.0738,
+      "step": 11660
+    },
+    {
+      "epoch": 7.819095477386934,
+      "grad_norm": 1.1745330095291138,
+      "learning_rate": 0.0002,
+      "loss": 1.1157,
+      "step": 11670
+    },
+    {
+      "epoch": 7.825795644891122,
+      "grad_norm": 1.3440991640090942,
+      "learning_rate": 0.0002,
+      "loss": 1.1071,
+      "step": 11680
+    },
+    {
+      "epoch": 7.83249581239531,
+      "grad_norm": 1.1923167705535889,
+      "learning_rate": 0.0002,
+      "loss": 1.0751,
+      "step": 11690
+    },
+    {
+      "epoch": 7.839195979899498,
+      "grad_norm": 1.2679530382156372,
+      "learning_rate": 0.0002,
+      "loss": 1.0656,
+      "step": 11700
+    },
+    {
+      "epoch": 7.8458961474036855,
+      "grad_norm": 1.3027020692825317,
+      "learning_rate": 0.0002,
+      "loss": 1.0504,
+      "step": 11710
+    },
+    {
+      "epoch": 7.852596314907872,
+      "grad_norm": 1.4565616846084595,
+      "learning_rate": 0.0002,
+      "loss": 1.0496,
+      "step": 11720
+    },
+    {
+      "epoch": 7.85929648241206,
+      "grad_norm": 1.3157920837402344,
+      "learning_rate": 0.0002,
+      "loss": 1.0487,
+      "step": 11730
+    },
+    {
+      "epoch": 7.865996649916248,
+      "grad_norm": 1.3120285272598267,
+      "learning_rate": 0.0002,
+      "loss": 1.0876,
+      "step": 11740
+    },
+    {
+      "epoch": 7.872696817420436,
+      "grad_norm": 1.2625858783721924,
+      "learning_rate": 0.0002,
+      "loss": 1.0248,
+      "step": 11750
+    },
+    {
+      "epoch": 7.8793969849246235,
+      "grad_norm": 1.3911317586898804,
+      "learning_rate": 0.0002,
+      "loss": 1.0047,
+      "step": 11760
+    },
+    {
+      "epoch": 7.88609715242881,
+      "grad_norm": 1.2151232957839966,
+      "learning_rate": 0.0002,
+      "loss": 1.0436,
+      "step": 11770
+    },
+    {
+      "epoch": 7.892797319932998,
+      "grad_norm": 1.5171650648117065,
+      "learning_rate": 0.0002,
+      "loss": 1.0526,
+      "step": 11780
+    },
+    {
+      "epoch": 7.899497487437186,
+      "grad_norm": 1.4308419227600098,
+      "learning_rate": 0.0002,
+      "loss": 1.0957,
+      "step": 11790
+    },
+    {
+      "epoch": 7.906197654941374,
+      "grad_norm": 1.2683558464050293,
+      "learning_rate": 0.0002,
+      "loss": 1.0346,
+      "step": 11800
+    },
+    {
+      "epoch": 7.9128978224455615,
+      "grad_norm": 1.6872674226760864,
+      "learning_rate": 0.0002,
+      "loss": 1.1134,
+      "step": 11810
+    },
+    {
+      "epoch": 7.919597989949748,
+      "grad_norm": 1.3805692195892334,
+      "learning_rate": 0.0002,
+      "loss": 1.0077,
+      "step": 11820
+    },
+    {
+      "epoch": 7.926298157453936,
+      "grad_norm": 1.3342738151550293,
+      "learning_rate": 0.0002,
+      "loss": 1.1098,
+      "step": 11830
+    },
+    {
+      "epoch": 7.932998324958124,
+      "grad_norm": 1.308137059211731,
+      "learning_rate": 0.0002,
+      "loss": 1.0446,
+      "step": 11840
+    },
+    {
+      "epoch": 7.939698492462312,
+      "grad_norm": 1.3725523948669434,
+      "learning_rate": 0.0002,
+      "loss": 1.1032,
+      "step": 11850
+    },
+    {
+      "epoch": 7.9463986599664995,
+      "grad_norm": 1.1458995342254639,
+      "learning_rate": 0.0002,
+      "loss": 1.0492,
+      "step": 11860
+    },
+    {
+      "epoch": 7.953098827470686,
+      "grad_norm": 1.4715759754180908,
+      "learning_rate": 0.0002,
+      "loss": 1.0968,
+      "step": 11870
+    },
+    {
+      "epoch": 7.959798994974874,
+      "grad_norm": 1.1897743940353394,
+      "learning_rate": 0.0002,
+      "loss": 1.1408,
+      "step": 11880
+    },
+    {
+      "epoch": 7.966499162479062,
+      "grad_norm": 1.3386842012405396,
+      "learning_rate": 0.0002,
+      "loss": 1.0628,
+      "step": 11890
+    },
+    {
+      "epoch": 7.97319932998325,
+      "grad_norm": 1.3611114025115967,
+      "learning_rate": 0.0002,
+      "loss": 1.0463,
+      "step": 11900
+    },
+    {
+      "epoch": 7.9798994974874375,
+      "grad_norm": 1.1429232358932495,
+      "learning_rate": 0.0002,
+      "loss": 1.1283,
+      "step": 11910
+    },
+    {
+      "epoch": 7.986599664991624,
+      "grad_norm": 1.4848406314849854,
+      "learning_rate": 0.0002,
+      "loss": 1.1153,
+      "step": 11920
+    },
+    {
+      "epoch": 7.993299832495812,
+      "grad_norm": 1.3205432891845703,
+      "learning_rate": 0.0002,
+      "loss": 1.1159,
+      "step": 11930
+    },
+    {
+      "epoch": 7.997319932998325,
+      "eval_loss": 2.273319959640503,
+      "eval_runtime": 37.9333,
+      "eval_samples_per_second": 13.576,
+      "eval_steps_per_second": 1.714,
+      "step": 11936
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 11936,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 8,
+  "save_steps": 200,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 5.52371354213548e+17,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-11936/training_args.bin b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-11936/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..064f299b0f3d2a28f5b1f5c68ef32caab3e2dd49
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-11936/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aa7b586fc440d1e22156593e2dd4267d2bdcb8920a02fdf352ea29a9bec3dd94
+size 5560
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-1492/README.md b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-1492/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..835e31ab8469ee39ddc8b2b6b2143a8c66dad510
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-1492/README.md
@@ -0,0 +1,202 @@
+---
+base_model: meta-llama/Meta-Llama-3-8B-Instruct
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.13.1
\ No newline at end of file
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-1492/adapter_config.json b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-1492/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..406c5a08dc4a2a33b52c62a482f98c217c417215
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-1492/adapter_config.json
@@ -0,0 +1,29 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "meta-llama/Meta-Llama-3-8B-Instruct",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-1492/adapter_model.safetensors b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-1492/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..dd239e42afc491dece8497032500d4eb9b33a024
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-1492/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:73e5e994f25eeb647e9eaf23119c1032c2dfc7e32d980c8862eff32d0a60cd25
+size 109069176
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-1492/optimizer.pt b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-1492/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..620fa8c94b3f0e09c590632f8933d9f7426387b1
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-1492/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2f6239078bcb5ed53dc80356c24ddbd2bac0fcb98c4f6db364b6bf14f0e94f37
+size 55532666
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-1492/rng_state.pth b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-1492/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..0ff627ce7a7574c57726bcd7f8dd33ef923e735b
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-1492/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d416b20e85177f56c04997a90b643a2900ee505cb41f99c2d2b2e23d88b8dd8f
+size 14244
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-1492/scheduler.pt b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-1492/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..574851d73f1f423bc7fd3b74ef9bc88c905f3754
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-1492/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3c77dbd02d51b31cf00f6638f73d175ce25655ac869b78c1454b4a1c11a6fdae
+size 1064
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-1492/special_tokens_map.json b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-1492/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..b43be96621d147110fb8a18b5776ec6e38516127
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-1492/special_tokens_map.json
@@ -0,0 +1,17 @@
+{
+  "bos_token": {
+    "content": "<|begin_of_text|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<|eot_id|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<|eot_id|>"
+}
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-1492/tokenizer.json b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-1492/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..86a33946b0c77216d2cce91bb28c8fada4a5e80b
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-1492/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3c5cf44023714fb39b05e71e425f8d7b92805ff73f7988b083b8c87f0bf87393
+size 17209961
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-1492/tokenizer_config.json b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-1492/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..061e40d9db3253624f86e8e364c15ef546527c9d
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-1492/tokenizer_config.json
@@ -0,0 +1,2063 @@
+{
+  "added_tokens_decoder": {
+    "128000": {
+      "content": "<|begin_of_text|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128001": {
+      "content": "<|end_of_text|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128002": {
+      "content": "<|reserved_special_token_0|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128003": {
+      "content": "<|reserved_special_token_1|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128004": {
+      "content": "<|reserved_special_token_2|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128005": {
+      "content": "<|reserved_special_token_3|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128006": {
+      "content": "<|start_header_id|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128007": {
+      "content": "<|end_header_id|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128008": {
+      "content": "<|reserved_special_token_4|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128009": {
+      "content": "<|eot_id|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128010": {
+      "content": "<|reserved_special_token_5|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128011": {
+      "content": "<|reserved_special_token_6|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128012": {
+      "content": "<|reserved_special_token_7|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128013": {
+      "content": "<|reserved_special_token_8|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128014": {
+      "content": "<|reserved_special_token_9|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128015": {
+      "content": "<|reserved_special_token_10|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128016": {
+      "content": "<|reserved_special_token_11|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128017": {
+      "content": "<|reserved_special_token_12|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128018": {
+      "content": "<|reserved_special_token_13|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128019": {
+      "content": "<|reserved_special_token_14|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128020": {
+      "content": "<|reserved_special_token_15|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128021": {
+      "content": "<|reserved_special_token_16|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128022": {
+      "content": "<|reserved_special_token_17|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128023": {
+      "content": "<|reserved_special_token_18|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128024": {
+      "content": "<|reserved_special_token_19|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128025": {
+      "content": "<|reserved_special_token_20|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128026": {
+      "content": "<|reserved_special_token_21|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128027": {
+      "content": "<|reserved_special_token_22|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128028": {
+      "content": "<|reserved_special_token_23|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128029": {
+      "content": "<|reserved_special_token_24|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128030": {
+      "content": "<|reserved_special_token_25|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128031": {
+      "content": "<|reserved_special_token_26|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128032": {
+      "content": "<|reserved_special_token_27|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128033": {
+      "content": "<|reserved_special_token_28|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128034": {
+      "content": "<|reserved_special_token_29|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128035": {
+      "content": "<|reserved_special_token_30|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128036": {
+      "content": "<|reserved_special_token_31|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128037": {
+      "content": "<|reserved_special_token_32|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128038": {
+      "content": "<|reserved_special_token_33|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128039": {
+      "content": "<|reserved_special_token_34|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128040": {
+      "content": "<|reserved_special_token_35|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128041": {
+      "content": "<|reserved_special_token_36|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128042": {
+      "content": "<|reserved_special_token_37|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128043": {
+      "content": "<|reserved_special_token_38|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128044": {
+      "content": "<|reserved_special_token_39|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128045": {
+      "content": "<|reserved_special_token_40|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128046": {
+      "content": "<|reserved_special_token_41|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128047": {
+      "content": "<|reserved_special_token_42|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128048": {
+      "content": "<|reserved_special_token_43|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128049": {
+      "content": "<|reserved_special_token_44|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128050": {
+      "content": "<|reserved_special_token_45|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128051": {
+      "content": "<|reserved_special_token_46|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128052": {
+      "content": "<|reserved_special_token_47|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128053": {
+      "content": "<|reserved_special_token_48|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128054": {
+      "content": "<|reserved_special_token_49|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128055": {
+      "content": "<|reserved_special_token_50|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128056": {
+      "content": "<|reserved_special_token_51|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128057": {
+      "content": "<|reserved_special_token_52|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128058": {
+      "content": "<|reserved_special_token_53|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128059": {
+      "content": "<|reserved_special_token_54|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128060": {
+      "content": "<|reserved_special_token_55|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128061": {
+      "content": "<|reserved_special_token_56|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128062": {
+      "content": "<|reserved_special_token_57|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128063": {
+      "content": "<|reserved_special_token_58|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128064": {
+      "content": "<|reserved_special_token_59|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128065": {
+      "content": "<|reserved_special_token_60|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128066": {
+      "content": "<|reserved_special_token_61|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128067": {
+      "content": "<|reserved_special_token_62|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128068": {
+      "content": "<|reserved_special_token_63|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128069": {
+      "content": "<|reserved_special_token_64|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128070": {
+      "content": "<|reserved_special_token_65|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128071": {
+      "content": "<|reserved_special_token_66|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128072": {
+      "content": "<|reserved_special_token_67|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128073": {
+      "content": "<|reserved_special_token_68|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128074": {
+      "content": "<|reserved_special_token_69|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128075": {
+      "content": "<|reserved_special_token_70|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128076": {
+      "content": "<|reserved_special_token_71|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128077": {
+      "content": "<|reserved_special_token_72|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128078": {
+      "content": "<|reserved_special_token_73|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128079": {
+      "content": "<|reserved_special_token_74|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128080": {
+      "content": "<|reserved_special_token_75|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128081": {
+      "content": "<|reserved_special_token_76|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128082": {
+      "content": "<|reserved_special_token_77|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128083": {
+      "content": "<|reserved_special_token_78|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128084": {
+      "content": "<|reserved_special_token_79|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128085": {
+      "content": "<|reserved_special_token_80|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128086": {
+      "content": "<|reserved_special_token_81|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128087": {
+      "content": "<|reserved_special_token_82|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128088": {
+      "content": "<|reserved_special_token_83|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128089": {
+      "content": "<|reserved_special_token_84|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128090": {
+      "content": "<|reserved_special_token_85|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128091": {
+      "content": "<|reserved_special_token_86|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128092": {
+      "content": "<|reserved_special_token_87|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128093": {
+      "content": "<|reserved_special_token_88|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128094": {
+      "content": "<|reserved_special_token_89|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128095": {
+      "content": "<|reserved_special_token_90|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128096": {
+      "content": "<|reserved_special_token_91|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128097": {
+      "content": "<|reserved_special_token_92|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128098": {
+      "content": "<|reserved_special_token_93|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128099": {
+      "content": "<|reserved_special_token_94|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128100": {
+      "content": "<|reserved_special_token_95|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128101": {
+      "content": "<|reserved_special_token_96|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128102": {
+      "content": "<|reserved_special_token_97|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128103": {
+      "content": "<|reserved_special_token_98|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128104": {
+      "content": "<|reserved_special_token_99|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128105": {
+      "content": "<|reserved_special_token_100|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128106": {
+      "content": "<|reserved_special_token_101|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128107": {
+      "content": "<|reserved_special_token_102|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128108": {
+      "content": "<|reserved_special_token_103|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128109": {
+      "content": "<|reserved_special_token_104|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128110": {
+      "content": "<|reserved_special_token_105|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128111": {
+      "content": "<|reserved_special_token_106|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128112": {
+      "content": "<|reserved_special_token_107|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128113": {
+      "content": "<|reserved_special_token_108|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128114": {
+      "content": "<|reserved_special_token_109|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128115": {
+      "content": "<|reserved_special_token_110|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128116": {
+      "content": "<|reserved_special_token_111|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128117": {
+      "content": "<|reserved_special_token_112|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128118": {
+      "content": "<|reserved_special_token_113|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128119": {
+      "content": "<|reserved_special_token_114|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128120": {
+      "content": "<|reserved_special_token_115|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128121": {
+      "content": "<|reserved_special_token_116|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128122": {
+      "content": "<|reserved_special_token_117|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128123": {
+      "content": "<|reserved_special_token_118|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128124": {
+      "content": "<|reserved_special_token_119|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128125": {
+      "content": "<|reserved_special_token_120|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128126": {
+      "content": "<|reserved_special_token_121|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128127": {
+      "content": "<|reserved_special_token_122|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128128": {
+      "content": "<|reserved_special_token_123|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128129": {
+      "content": "<|reserved_special_token_124|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128130": {
+      "content": "<|reserved_special_token_125|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128131": {
+      "content": "<|reserved_special_token_126|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128132": {
+      "content": "<|reserved_special_token_127|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128133": {
+      "content": "<|reserved_special_token_128|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128134": {
+      "content": "<|reserved_special_token_129|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128135": {
+      "content": "<|reserved_special_token_130|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128136": {
+      "content": "<|reserved_special_token_131|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128137": {
+      "content": "<|reserved_special_token_132|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128138": {
+      "content": "<|reserved_special_token_133|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128139": {
+      "content": "<|reserved_special_token_134|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128140": {
+      "content": "<|reserved_special_token_135|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128141": {
+      "content": "<|reserved_special_token_136|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128142": {
+      "content": "<|reserved_special_token_137|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128143": {
+      "content": "<|reserved_special_token_138|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128144": {
+      "content": "<|reserved_special_token_139|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128145": {
+      "content": "<|reserved_special_token_140|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128146": {
+      "content": "<|reserved_special_token_141|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128147": {
+      "content": "<|reserved_special_token_142|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128148": {
+      "content": "<|reserved_special_token_143|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128149": {
+      "content": "<|reserved_special_token_144|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128150": {
+      "content": "<|reserved_special_token_145|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128151": {
+      "content": "<|reserved_special_token_146|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128152": {
+      "content": "<|reserved_special_token_147|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128153": {
+      "content": "<|reserved_special_token_148|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128154": {
+      "content": "<|reserved_special_token_149|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128155": {
+      "content": "<|reserved_special_token_150|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128156": {
+      "content": "<|reserved_special_token_151|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128157": {
+      "content": "<|reserved_special_token_152|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128158": {
+      "content": "<|reserved_special_token_153|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128159": {
+      "content": "<|reserved_special_token_154|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128160": {
+      "content": "<|reserved_special_token_155|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128161": {
+      "content": "<|reserved_special_token_156|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128162": {
+      "content": "<|reserved_special_token_157|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128163": {
+      "content": "<|reserved_special_token_158|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128164": {
+      "content": "<|reserved_special_token_159|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128165": {
+      "content": "<|reserved_special_token_160|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128166": {
+      "content": "<|reserved_special_token_161|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128167": {
+      "content": "<|reserved_special_token_162|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128168": {
+      "content": "<|reserved_special_token_163|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128169": {
+      "content": "<|reserved_special_token_164|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128170": {
+      "content": "<|reserved_special_token_165|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128171": {
+      "content": "<|reserved_special_token_166|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128172": {
+      "content": "<|reserved_special_token_167|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128173": {
+      "content": "<|reserved_special_token_168|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128174": {
+      "content": "<|reserved_special_token_169|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128175": {
+      "content": "<|reserved_special_token_170|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128176": {
+      "content": "<|reserved_special_token_171|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128177": {
+      "content": "<|reserved_special_token_172|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128178": {
+      "content": "<|reserved_special_token_173|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128179": {
+      "content": "<|reserved_special_token_174|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128180": {
+      "content": "<|reserved_special_token_175|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128181": {
+      "content": "<|reserved_special_token_176|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128182": {
+      "content": "<|reserved_special_token_177|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128183": {
+      "content": "<|reserved_special_token_178|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128184": {
+      "content": "<|reserved_special_token_179|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128185": {
+      "content": "<|reserved_special_token_180|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128186": {
+      "content": "<|reserved_special_token_181|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128187": {
+      "content": "<|reserved_special_token_182|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128188": {
+      "content": "<|reserved_special_token_183|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128189": {
+      "content": "<|reserved_special_token_184|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128190": {
+      "content": "<|reserved_special_token_185|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128191": {
+      "content": "<|reserved_special_token_186|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128192": {
+      "content": "<|reserved_special_token_187|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128193": {
+      "content": "<|reserved_special_token_188|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128194": {
+      "content": "<|reserved_special_token_189|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128195": {
+      "content": "<|reserved_special_token_190|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128196": {
+      "content": "<|reserved_special_token_191|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128197": {
+      "content": "<|reserved_special_token_192|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128198": {
+      "content": "<|reserved_special_token_193|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128199": {
+      "content": "<|reserved_special_token_194|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128200": {
+      "content": "<|reserved_special_token_195|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128201": {
+      "content": "<|reserved_special_token_196|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128202": {
+      "content": "<|reserved_special_token_197|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128203": {
+      "content": "<|reserved_special_token_198|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128204": {
+      "content": "<|reserved_special_token_199|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128205": {
+      "content": "<|reserved_special_token_200|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128206": {
+      "content": "<|reserved_special_token_201|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128207": {
+      "content": "<|reserved_special_token_202|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128208": {
+      "content": "<|reserved_special_token_203|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128209": {
+      "content": "<|reserved_special_token_204|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128210": {
+      "content": "<|reserved_special_token_205|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128211": {
+      "content": "<|reserved_special_token_206|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128212": {
+      "content": "<|reserved_special_token_207|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128213": {
+      "content": "<|reserved_special_token_208|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128214": {
+      "content": "<|reserved_special_token_209|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128215": {
+      "content": "<|reserved_special_token_210|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128216": {
+      "content": "<|reserved_special_token_211|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128217": {
+      "content": "<|reserved_special_token_212|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128218": {
+      "content": "<|reserved_special_token_213|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128219": {
+      "content": "<|reserved_special_token_214|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128220": {
+      "content": "<|reserved_special_token_215|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128221": {
+      "content": "<|reserved_special_token_216|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128222": {
+      "content": "<|reserved_special_token_217|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128223": {
+      "content": "<|reserved_special_token_218|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128224": {
+      "content": "<|reserved_special_token_219|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128225": {
+      "content": "<|reserved_special_token_220|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128226": {
+      "content": "<|reserved_special_token_221|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128227": {
+      "content": "<|reserved_special_token_222|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128228": {
+      "content": "<|reserved_special_token_223|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128229": {
+      "content": "<|reserved_special_token_224|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128230": {
+      "content": "<|reserved_special_token_225|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128231": {
+      "content": "<|reserved_special_token_226|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128232": {
+      "content": "<|reserved_special_token_227|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128233": {
+      "content": "<|reserved_special_token_228|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128234": {
+      "content": "<|reserved_special_token_229|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128235": {
+      "content": "<|reserved_special_token_230|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128236": {
+      "content": "<|reserved_special_token_231|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128237": {
+      "content": "<|reserved_special_token_232|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128238": {
+      "content": "<|reserved_special_token_233|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128239": {
+      "content": "<|reserved_special_token_234|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128240": {
+      "content": "<|reserved_special_token_235|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128241": {
+      "content": "<|reserved_special_token_236|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128242": {
+      "content": "<|reserved_special_token_237|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128243": {
+      "content": "<|reserved_special_token_238|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128244": {
+      "content": "<|reserved_special_token_239|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128245": {
+      "content": "<|reserved_special_token_240|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128246": {
+      "content": "<|reserved_special_token_241|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128247": {
+      "content": "<|reserved_special_token_242|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128248": {
+      "content": "<|reserved_special_token_243|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128249": {
+      "content": "<|reserved_special_token_244|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128250": {
+      "content": "<|reserved_special_token_245|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128251": {
+      "content": "<|reserved_special_token_246|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128252": {
+      "content": "<|reserved_special_token_247|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128253": {
+      "content": "<|reserved_special_token_248|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128254": {
+      "content": "<|reserved_special_token_249|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128255": {
+      "content": "<|reserved_special_token_250|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<|begin_of_text|>",
+  "chat_template": "{% set loop_messages = messages %}{% for message in loop_messages %}{% set content = '<|start_header_id|>' + message['role'] + '<|end_header_id|>\n\n'+ message['content'] | trim + '<|eot_id|>' %}{% if loop.index0 == 0 %}{% set content = bos_token + content %}{% endif %}{{ content }}{% endfor %}{% if add_generation_prompt %}{{ '<|start_header_id|>assistant<|end_header_id|>\n\n' }}{% endif %}",
+  "clean_up_tokenization_spaces": true,
+  "eos_token": "<|eot_id|>",
+  "model_input_names": [
+    "input_ids",
+    "attention_mask"
+  ],
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<|eot_id|>",
+  "tokenizer_class": "PreTrainedTokenizerFast"
+}
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-1492/trainer_state.json b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-1492/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..b48ac6f9b0866be7a6a8882bab01a12c6bafe113
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-1492/trainer_state.json
@@ -0,0 +1,1084 @@
+{
+  "best_metric": 1.8036354780197144,
+  "best_model_checkpoint": "outputs-001/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-1492",
+  "epoch": 0.9996649916247906,
+  "eval_steps": 10,
+  "global_step": 1492,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.006700167504187605,
+      "grad_norm": 0.565915048122406,
+      "learning_rate": 0.0002,
+      "loss": 2.6189,
+      "step": 10
+    },
+    {
+      "epoch": 0.01340033500837521,
+      "grad_norm": 0.5004463791847229,
+      "learning_rate": 0.0002,
+      "loss": 2.3162,
+      "step": 20
+    },
+    {
+      "epoch": 0.020100502512562814,
+      "grad_norm": 0.511043906211853,
+      "learning_rate": 0.0002,
+      "loss": 2.0576,
+      "step": 30
+    },
+    {
+      "epoch": 0.02680067001675042,
+      "grad_norm": 0.47327178716659546,
+      "learning_rate": 0.0002,
+      "loss": 2.0085,
+      "step": 40
+    },
+    {
+      "epoch": 0.03350083752093802,
+      "grad_norm": 0.5511676669120789,
+      "learning_rate": 0.0002,
+      "loss": 2.0276,
+      "step": 50
+    },
+    {
+      "epoch": 0.04020100502512563,
+      "grad_norm": 0.4666278064250946,
+      "learning_rate": 0.0002,
+      "loss": 1.9075,
+      "step": 60
+    },
+    {
+      "epoch": 0.04690117252931323,
+      "grad_norm": 0.5310961008071899,
+      "learning_rate": 0.0002,
+      "loss": 1.8413,
+      "step": 70
+    },
+    {
+      "epoch": 0.05360134003350084,
+      "grad_norm": 0.5606027245521545,
+      "learning_rate": 0.0002,
+      "loss": 1.8711,
+      "step": 80
+    },
+    {
+      "epoch": 0.06030150753768844,
+      "grad_norm": 0.4934779703617096,
+      "learning_rate": 0.0002,
+      "loss": 1.9282,
+      "step": 90
+    },
+    {
+      "epoch": 0.06700167504187604,
+      "grad_norm": 0.4821869730949402,
+      "learning_rate": 0.0002,
+      "loss": 1.8925,
+      "step": 100
+    },
+    {
+      "epoch": 0.07370184254606366,
+      "grad_norm": 0.5262084603309631,
+      "learning_rate": 0.0002,
+      "loss": 1.8628,
+      "step": 110
+    },
+    {
+      "epoch": 0.08040201005025126,
+      "grad_norm": 0.3774230182170868,
+      "learning_rate": 0.0002,
+      "loss": 1.8347,
+      "step": 120
+    },
+    {
+      "epoch": 0.08710217755443886,
+      "grad_norm": 0.34137430787086487,
+      "learning_rate": 0.0002,
+      "loss": 1.8386,
+      "step": 130
+    },
+    {
+      "epoch": 0.09380234505862646,
+      "grad_norm": 0.407272070646286,
+      "learning_rate": 0.0002,
+      "loss": 1.861,
+      "step": 140
+    },
+    {
+      "epoch": 0.10050251256281408,
+      "grad_norm": 0.4011937975883484,
+      "learning_rate": 0.0002,
+      "loss": 1.8279,
+      "step": 150
+    },
+    {
+      "epoch": 0.10720268006700168,
+      "grad_norm": 0.4432467222213745,
+      "learning_rate": 0.0002,
+      "loss": 1.9317,
+      "step": 160
+    },
+    {
+      "epoch": 0.11390284757118928,
+      "grad_norm": 0.44030463695526123,
+      "learning_rate": 0.0002,
+      "loss": 1.8157,
+      "step": 170
+    },
+    {
+      "epoch": 0.12060301507537688,
+      "grad_norm": 0.3799569308757782,
+      "learning_rate": 0.0002,
+      "loss": 1.8534,
+      "step": 180
+    },
+    {
+      "epoch": 0.1273031825795645,
+      "grad_norm": 0.33721521496772766,
+      "learning_rate": 0.0002,
+      "loss": 1.7658,
+      "step": 190
+    },
+    {
+      "epoch": 0.13400335008375208,
+      "grad_norm": 0.4096226692199707,
+      "learning_rate": 0.0002,
+      "loss": 1.8269,
+      "step": 200
+    },
+    {
+      "epoch": 0.1407035175879397,
+      "grad_norm": 0.37374693155288696,
+      "learning_rate": 0.0002,
+      "loss": 1.802,
+      "step": 210
+    },
+    {
+      "epoch": 0.1474036850921273,
+      "grad_norm": 0.3249480128288269,
+      "learning_rate": 0.0002,
+      "loss": 1.8901,
+      "step": 220
+    },
+    {
+      "epoch": 0.1541038525963149,
+      "grad_norm": 0.3612042963504791,
+      "learning_rate": 0.0002,
+      "loss": 1.8163,
+      "step": 230
+    },
+    {
+      "epoch": 0.16080402010050251,
+      "grad_norm": 0.3686671257019043,
+      "learning_rate": 0.0002,
+      "loss": 1.7585,
+      "step": 240
+    },
+    {
+      "epoch": 0.16750418760469013,
+      "grad_norm": 0.3521044850349426,
+      "learning_rate": 0.0002,
+      "loss": 1.8365,
+      "step": 250
+    },
+    {
+      "epoch": 0.17420435510887772,
+      "grad_norm": 0.4073677361011505,
+      "learning_rate": 0.0002,
+      "loss": 1.8623,
+      "step": 260
+    },
+    {
+      "epoch": 0.18090452261306533,
+      "grad_norm": 0.34522193670272827,
+      "learning_rate": 0.0002,
+      "loss": 1.8026,
+      "step": 270
+    },
+    {
+      "epoch": 0.18760469011725292,
+      "grad_norm": 0.4121900498867035,
+      "learning_rate": 0.0002,
+      "loss": 1.8162,
+      "step": 280
+    },
+    {
+      "epoch": 0.19430485762144054,
+      "grad_norm": 0.3544778525829315,
+      "learning_rate": 0.0002,
+      "loss": 1.7976,
+      "step": 290
+    },
+    {
+      "epoch": 0.20100502512562815,
+      "grad_norm": 0.3482133448123932,
+      "learning_rate": 0.0002,
+      "loss": 1.8787,
+      "step": 300
+    },
+    {
+      "epoch": 0.20770519262981574,
+      "grad_norm": 0.3421826660633087,
+      "learning_rate": 0.0002,
+      "loss": 1.8578,
+      "step": 310
+    },
+    {
+      "epoch": 0.21440536013400335,
+      "grad_norm": 0.5024696588516235,
+      "learning_rate": 0.0002,
+      "loss": 1.8013,
+      "step": 320
+    },
+    {
+      "epoch": 0.22110552763819097,
+      "grad_norm": 0.36013063788414,
+      "learning_rate": 0.0002,
+      "loss": 1.8607,
+      "step": 330
+    },
+    {
+      "epoch": 0.22780569514237856,
+      "grad_norm": 0.3611244857311249,
+      "learning_rate": 0.0002,
+      "loss": 1.9075,
+      "step": 340
+    },
+    {
+      "epoch": 0.23450586264656617,
+      "grad_norm": 0.39244529604911804,
+      "learning_rate": 0.0002,
+      "loss": 1.8128,
+      "step": 350
+    },
+    {
+      "epoch": 0.24120603015075376,
+      "grad_norm": 0.3299325704574585,
+      "learning_rate": 0.0002,
+      "loss": 1.7885,
+      "step": 360
+    },
+    {
+      "epoch": 0.24790619765494137,
+      "grad_norm": 0.3994322419166565,
+      "learning_rate": 0.0002,
+      "loss": 1.8028,
+      "step": 370
+    },
+    {
+      "epoch": 0.254606365159129,
+      "grad_norm": 0.3559151887893677,
+      "learning_rate": 0.0002,
+      "loss": 1.8321,
+      "step": 380
+    },
+    {
+      "epoch": 0.2613065326633166,
+      "grad_norm": 0.3873756229877472,
+      "learning_rate": 0.0002,
+      "loss": 1.7802,
+      "step": 390
+    },
+    {
+      "epoch": 0.26800670016750416,
+      "grad_norm": 0.3710744082927704,
+      "learning_rate": 0.0002,
+      "loss": 1.7844,
+      "step": 400
+    },
+    {
+      "epoch": 0.2747068676716918,
+      "grad_norm": 0.3618465065956116,
+      "learning_rate": 0.0002,
+      "loss": 1.7789,
+      "step": 410
+    },
+    {
+      "epoch": 0.2814070351758794,
+      "grad_norm": 0.30063769221305847,
+      "learning_rate": 0.0002,
+      "loss": 1.8529,
+      "step": 420
+    },
+    {
+      "epoch": 0.288107202680067,
+      "grad_norm": 0.3695628345012665,
+      "learning_rate": 0.0002,
+      "loss": 1.7765,
+      "step": 430
+    },
+    {
+      "epoch": 0.2948073701842546,
+      "grad_norm": 0.31451135873794556,
+      "learning_rate": 0.0002,
+      "loss": 1.7982,
+      "step": 440
+    },
+    {
+      "epoch": 0.3015075376884422,
+      "grad_norm": 0.3959707021713257,
+      "learning_rate": 0.0002,
+      "loss": 1.7517,
+      "step": 450
+    },
+    {
+      "epoch": 0.3082077051926298,
+      "grad_norm": 0.33475354313850403,
+      "learning_rate": 0.0002,
+      "loss": 1.8142,
+      "step": 460
+    },
+    {
+      "epoch": 0.3149078726968174,
+      "grad_norm": 0.33933115005493164,
+      "learning_rate": 0.0002,
+      "loss": 1.8805,
+      "step": 470
+    },
+    {
+      "epoch": 0.32160804020100503,
+      "grad_norm": 0.3264943063259125,
+      "learning_rate": 0.0002,
+      "loss": 1.7564,
+      "step": 480
+    },
+    {
+      "epoch": 0.32830820770519265,
+      "grad_norm": 0.40188100934028625,
+      "learning_rate": 0.0002,
+      "loss": 1.8428,
+      "step": 490
+    },
+    {
+      "epoch": 0.33500837520938026,
+      "grad_norm": 0.37408649921417236,
+      "learning_rate": 0.0002,
+      "loss": 1.7624,
+      "step": 500
+    },
+    {
+      "epoch": 0.3417085427135678,
+      "grad_norm": 0.33925938606262207,
+      "learning_rate": 0.0002,
+      "loss": 1.7745,
+      "step": 510
+    },
+    {
+      "epoch": 0.34840871021775544,
+      "grad_norm": 0.36836713552474976,
+      "learning_rate": 0.0002,
+      "loss": 1.814,
+      "step": 520
+    },
+    {
+      "epoch": 0.35510887772194305,
+      "grad_norm": 0.37284499406814575,
+      "learning_rate": 0.0002,
+      "loss": 1.8037,
+      "step": 530
+    },
+    {
+      "epoch": 0.36180904522613067,
+      "grad_norm": 0.3192278742790222,
+      "learning_rate": 0.0002,
+      "loss": 1.8379,
+      "step": 540
+    },
+    {
+      "epoch": 0.3685092127303183,
+      "grad_norm": 0.30233290791511536,
+      "learning_rate": 0.0002,
+      "loss": 1.8702,
+      "step": 550
+    },
+    {
+      "epoch": 0.37520938023450584,
+      "grad_norm": 0.3340817391872406,
+      "learning_rate": 0.0002,
+      "loss": 1.8799,
+      "step": 560
+    },
+    {
+      "epoch": 0.38190954773869346,
+      "grad_norm": 0.32600095868110657,
+      "learning_rate": 0.0002,
+      "loss": 1.8404,
+      "step": 570
+    },
+    {
+      "epoch": 0.38860971524288107,
+      "grad_norm": 0.33711278438568115,
+      "learning_rate": 0.0002,
+      "loss": 1.7804,
+      "step": 580
+    },
+    {
+      "epoch": 0.3953098827470687,
+      "grad_norm": 0.34890690445899963,
+      "learning_rate": 0.0002,
+      "loss": 1.8445,
+      "step": 590
+    },
+    {
+      "epoch": 0.4020100502512563,
+      "grad_norm": 0.38238924741744995,
+      "learning_rate": 0.0002,
+      "loss": 1.8187,
+      "step": 600
+    },
+    {
+      "epoch": 0.40871021775544386,
+      "grad_norm": 0.34399354457855225,
+      "learning_rate": 0.0002,
+      "loss": 1.8111,
+      "step": 610
+    },
+    {
+      "epoch": 0.4154103852596315,
+      "grad_norm": 0.3346073627471924,
+      "learning_rate": 0.0002,
+      "loss": 1.8006,
+      "step": 620
+    },
+    {
+      "epoch": 0.4221105527638191,
+      "grad_norm": 0.3545648157596588,
+      "learning_rate": 0.0002,
+      "loss": 1.7705,
+      "step": 630
+    },
+    {
+      "epoch": 0.4288107202680067,
+      "grad_norm": 0.3378899097442627,
+      "learning_rate": 0.0002,
+      "loss": 1.8445,
+      "step": 640
+    },
+    {
+      "epoch": 0.4355108877721943,
+      "grad_norm": 0.3255569040775299,
+      "learning_rate": 0.0002,
+      "loss": 1.804,
+      "step": 650
+    },
+    {
+      "epoch": 0.44221105527638194,
+      "grad_norm": 0.34880587458610535,
+      "learning_rate": 0.0002,
+      "loss": 1.7679,
+      "step": 660
+    },
+    {
+      "epoch": 0.4489112227805695,
+      "grad_norm": 0.3402383625507355,
+      "learning_rate": 0.0002,
+      "loss": 1.7861,
+      "step": 670
+    },
+    {
+      "epoch": 0.4556113902847571,
+      "grad_norm": 0.3594033718109131,
+      "learning_rate": 0.0002,
+      "loss": 1.8131,
+      "step": 680
+    },
+    {
+      "epoch": 0.4623115577889447,
+      "grad_norm": 0.31000566482543945,
+      "learning_rate": 0.0002,
+      "loss": 1.8399,
+      "step": 690
+    },
+    {
+      "epoch": 0.46901172529313234,
+      "grad_norm": 0.37229061126708984,
+      "learning_rate": 0.0002,
+      "loss": 1.7521,
+      "step": 700
+    },
+    {
+      "epoch": 0.47571189279731996,
+      "grad_norm": 0.315801739692688,
+      "learning_rate": 0.0002,
+      "loss": 1.7779,
+      "step": 710
+    },
+    {
+      "epoch": 0.4824120603015075,
+      "grad_norm": 0.3220832645893097,
+      "learning_rate": 0.0002,
+      "loss": 1.7515,
+      "step": 720
+    },
+    {
+      "epoch": 0.48911222780569513,
+      "grad_norm": 0.3435456156730652,
+      "learning_rate": 0.0002,
+      "loss": 1.7181,
+      "step": 730
+    },
+    {
+      "epoch": 0.49581239530988275,
+      "grad_norm": 0.30380892753601074,
+      "learning_rate": 0.0002,
+      "loss": 1.8844,
+      "step": 740
+    },
+    {
+      "epoch": 0.5025125628140703,
+      "grad_norm": 0.3555026054382324,
+      "learning_rate": 0.0002,
+      "loss": 1.7792,
+      "step": 750
+    },
+    {
+      "epoch": 0.509212730318258,
+      "grad_norm": 0.3019855320453644,
+      "learning_rate": 0.0002,
+      "loss": 1.7714,
+      "step": 760
+    },
+    {
+      "epoch": 0.5159128978224455,
+      "grad_norm": 0.309111088514328,
+      "learning_rate": 0.0002,
+      "loss": 1.7962,
+      "step": 770
+    },
+    {
+      "epoch": 0.5226130653266332,
+      "grad_norm": 0.366020530462265,
+      "learning_rate": 0.0002,
+      "loss": 1.7913,
+      "step": 780
+    },
+    {
+      "epoch": 0.5293132328308208,
+      "grad_norm": 0.3267050087451935,
+      "learning_rate": 0.0002,
+      "loss": 1.8008,
+      "step": 790
+    },
+    {
+      "epoch": 0.5360134003350083,
+      "grad_norm": 0.34265750646591187,
+      "learning_rate": 0.0002,
+      "loss": 1.7397,
+      "step": 800
+    },
+    {
+      "epoch": 0.542713567839196,
+      "grad_norm": 0.313669890165329,
+      "learning_rate": 0.0002,
+      "loss": 1.8251,
+      "step": 810
+    },
+    {
+      "epoch": 0.5494137353433836,
+      "grad_norm": 0.3355236053466797,
+      "learning_rate": 0.0002,
+      "loss": 1.8369,
+      "step": 820
+    },
+    {
+      "epoch": 0.5561139028475712,
+      "grad_norm": 0.3186608552932739,
+      "learning_rate": 0.0002,
+      "loss": 1.7381,
+      "step": 830
+    },
+    {
+      "epoch": 0.5628140703517588,
+      "grad_norm": 0.30357518792152405,
+      "learning_rate": 0.0002,
+      "loss": 1.8034,
+      "step": 840
+    },
+    {
+      "epoch": 0.5695142378559463,
+      "grad_norm": 0.3990040123462677,
+      "learning_rate": 0.0002,
+      "loss": 1.769,
+      "step": 850
+    },
+    {
+      "epoch": 0.576214405360134,
+      "grad_norm": 0.34363803267478943,
+      "learning_rate": 0.0002,
+      "loss": 1.7482,
+      "step": 860
+    },
+    {
+      "epoch": 0.5829145728643216,
+      "grad_norm": 0.3757908046245575,
+      "learning_rate": 0.0002,
+      "loss": 1.8106,
+      "step": 870
+    },
+    {
+      "epoch": 0.5896147403685092,
+      "grad_norm": 0.3359757661819458,
+      "learning_rate": 0.0002,
+      "loss": 1.8104,
+      "step": 880
+    },
+    {
+      "epoch": 0.5963149078726968,
+      "grad_norm": 0.5555329918861389,
+      "learning_rate": 0.0002,
+      "loss": 1.7591,
+      "step": 890
+    },
+    {
+      "epoch": 0.6030150753768844,
+      "grad_norm": 0.4046323895454407,
+      "learning_rate": 0.0002,
+      "loss": 1.7715,
+      "step": 900
+    },
+    {
+      "epoch": 0.609715242881072,
+      "grad_norm": 0.29834219813346863,
+      "learning_rate": 0.0002,
+      "loss": 1.7998,
+      "step": 910
+    },
+    {
+      "epoch": 0.6164154103852596,
+      "grad_norm": 0.3241238594055176,
+      "learning_rate": 0.0002,
+      "loss": 1.7826,
+      "step": 920
+    },
+    {
+      "epoch": 0.6231155778894473,
+      "grad_norm": 0.35154739022254944,
+      "learning_rate": 0.0002,
+      "loss": 1.8342,
+      "step": 930
+    },
+    {
+      "epoch": 0.6298157453936348,
+      "grad_norm": 0.3287706673145294,
+      "learning_rate": 0.0002,
+      "loss": 1.8076,
+      "step": 940
+    },
+    {
+      "epoch": 0.6365159128978225,
+      "grad_norm": 0.35670626163482666,
+      "learning_rate": 0.0002,
+      "loss": 1.8038,
+      "step": 950
+    },
+    {
+      "epoch": 0.6432160804020101,
+      "grad_norm": 0.6114104986190796,
+      "learning_rate": 0.0002,
+      "loss": 1.869,
+      "step": 960
+    },
+    {
+      "epoch": 0.6499162479061976,
+      "grad_norm": 0.3186565041542053,
+      "learning_rate": 0.0002,
+      "loss": 1.8297,
+      "step": 970
+    },
+    {
+      "epoch": 0.6566164154103853,
+      "grad_norm": 0.27164125442504883,
+      "learning_rate": 0.0002,
+      "loss": 1.7539,
+      "step": 980
+    },
+    {
+      "epoch": 0.6633165829145728,
+      "grad_norm": 0.34407344460487366,
+      "learning_rate": 0.0002,
+      "loss": 1.8339,
+      "step": 990
+    },
+    {
+      "epoch": 0.6700167504187605,
+      "grad_norm": 0.368415892124176,
+      "learning_rate": 0.0002,
+      "loss": 1.855,
+      "step": 1000
+    },
+    {
+      "epoch": 0.6767169179229481,
+      "grad_norm": 0.3306390643119812,
+      "learning_rate": 0.0002,
+      "loss": 1.7821,
+      "step": 1010
+    },
+    {
+      "epoch": 0.6834170854271356,
+      "grad_norm": 0.3198648989200592,
+      "learning_rate": 0.0002,
+      "loss": 1.786,
+      "step": 1020
+    },
+    {
+      "epoch": 0.6901172529313233,
+      "grad_norm": 0.3092987537384033,
+      "learning_rate": 0.0002,
+      "loss": 1.816,
+      "step": 1030
+    },
+    {
+      "epoch": 0.6968174204355109,
+      "grad_norm": 0.3090653419494629,
+      "learning_rate": 0.0002,
+      "loss": 1.7689,
+      "step": 1040
+    },
+    {
+      "epoch": 0.7035175879396985,
+      "grad_norm": 0.3485880196094513,
+      "learning_rate": 0.0002,
+      "loss": 1.7544,
+      "step": 1050
+    },
+    {
+      "epoch": 0.7102177554438861,
+      "grad_norm": 0.35782721638679504,
+      "learning_rate": 0.0002,
+      "loss": 1.8187,
+      "step": 1060
+    },
+    {
+      "epoch": 0.7169179229480737,
+      "grad_norm": 0.34256869554519653,
+      "learning_rate": 0.0002,
+      "loss": 1.806,
+      "step": 1070
+    },
+    {
+      "epoch": 0.7236180904522613,
+      "grad_norm": 0.30461037158966064,
+      "learning_rate": 0.0002,
+      "loss": 1.7873,
+      "step": 1080
+    },
+    {
+      "epoch": 0.7303182579564489,
+      "grad_norm": 0.3398691713809967,
+      "learning_rate": 0.0002,
+      "loss": 1.7367,
+      "step": 1090
+    },
+    {
+      "epoch": 0.7370184254606366,
+      "grad_norm": 0.3180808126926422,
+      "learning_rate": 0.0002,
+      "loss": 1.8756,
+      "step": 1100
+    },
+    {
+      "epoch": 0.7437185929648241,
+      "grad_norm": 0.34400665760040283,
+      "learning_rate": 0.0002,
+      "loss": 1.6988,
+      "step": 1110
+    },
+    {
+      "epoch": 0.7504187604690117,
+      "grad_norm": 0.34244877099990845,
+      "learning_rate": 0.0002,
+      "loss": 1.7851,
+      "step": 1120
+    },
+    {
+      "epoch": 0.7571189279731994,
+      "grad_norm": 0.29946693778038025,
+      "learning_rate": 0.0002,
+      "loss": 1.7841,
+      "step": 1130
+    },
+    {
+      "epoch": 0.7638190954773869,
+      "grad_norm": 0.37547236680984497,
+      "learning_rate": 0.0002,
+      "loss": 1.7456,
+      "step": 1140
+    },
+    {
+      "epoch": 0.7705192629815746,
+      "grad_norm": 0.3263005018234253,
+      "learning_rate": 0.0002,
+      "loss": 1.8425,
+      "step": 1150
+    },
+    {
+      "epoch": 0.7772194304857621,
+      "grad_norm": 0.41363608837127686,
+      "learning_rate": 0.0002,
+      "loss": 1.7222,
+      "step": 1160
+    },
+    {
+      "epoch": 0.7839195979899497,
+      "grad_norm": 0.36267954111099243,
+      "learning_rate": 0.0002,
+      "loss": 1.7836,
+      "step": 1170
+    },
+    {
+      "epoch": 0.7906197654941374,
+      "grad_norm": 0.31789499521255493,
+      "learning_rate": 0.0002,
+      "loss": 1.9183,
+      "step": 1180
+    },
+    {
+      "epoch": 0.7973199329983249,
+      "grad_norm": 0.5708149075508118,
+      "learning_rate": 0.0002,
+      "loss": 1.78,
+      "step": 1190
+    },
+    {
+      "epoch": 0.8040201005025126,
+      "grad_norm": 0.322099506855011,
+      "learning_rate": 0.0002,
+      "loss": 1.6908,
+      "step": 1200
+    },
+    {
+      "epoch": 0.8107202680067002,
+      "grad_norm": 0.3419909179210663,
+      "learning_rate": 0.0002,
+      "loss": 1.7639,
+      "step": 1210
+    },
+    {
+      "epoch": 0.8174204355108877,
+      "grad_norm": 0.36286255717277527,
+      "learning_rate": 0.0002,
+      "loss": 1.7428,
+      "step": 1220
+    },
+    {
+      "epoch": 0.8241206030150754,
+      "grad_norm": 0.33992862701416016,
+      "learning_rate": 0.0002,
+      "loss": 1.8409,
+      "step": 1230
+    },
+    {
+      "epoch": 0.830820770519263,
+      "grad_norm": 0.32622793316841125,
+      "learning_rate": 0.0002,
+      "loss": 1.7507,
+      "step": 1240
+    },
+    {
+      "epoch": 0.8375209380234506,
+      "grad_norm": 0.3036167621612549,
+      "learning_rate": 0.0002,
+      "loss": 1.8098,
+      "step": 1250
+    },
+    {
+      "epoch": 0.8442211055276382,
+      "grad_norm": 0.3182215392589569,
+      "learning_rate": 0.0002,
+      "loss": 1.8094,
+      "step": 1260
+    },
+    {
+      "epoch": 0.8509212730318257,
+      "grad_norm": 0.3270018696784973,
+      "learning_rate": 0.0002,
+      "loss": 1.8017,
+      "step": 1270
+    },
+    {
+      "epoch": 0.8576214405360134,
+      "grad_norm": 0.32652342319488525,
+      "learning_rate": 0.0002,
+      "loss": 1.798,
+      "step": 1280
+    },
+    {
+      "epoch": 0.864321608040201,
+      "grad_norm": 0.3631329834461212,
+      "learning_rate": 0.0002,
+      "loss": 1.7448,
+      "step": 1290
+    },
+    {
+      "epoch": 0.8710217755443886,
+      "grad_norm": 0.36706018447875977,
+      "learning_rate": 0.0002,
+      "loss": 1.7,
+      "step": 1300
+    },
+    {
+      "epoch": 0.8777219430485762,
+      "grad_norm": 0.3347418010234833,
+      "learning_rate": 0.0002,
+      "loss": 1.8178,
+      "step": 1310
+    },
+    {
+      "epoch": 0.8844221105527639,
+      "grad_norm": 0.34371060132980347,
+      "learning_rate": 0.0002,
+      "loss": 1.7824,
+      "step": 1320
+    },
+    {
+      "epoch": 0.8911222780569514,
+      "grad_norm": 0.3029090166091919,
+      "learning_rate": 0.0002,
+      "loss": 1.783,
+      "step": 1330
+    },
+    {
+      "epoch": 0.897822445561139,
+      "grad_norm": 0.34700682759284973,
+      "learning_rate": 0.0002,
+      "loss": 1.8017,
+      "step": 1340
+    },
+    {
+      "epoch": 0.9045226130653267,
+      "grad_norm": 0.35574328899383545,
+      "learning_rate": 0.0002,
+      "loss": 1.7998,
+      "step": 1350
+    },
+    {
+      "epoch": 0.9112227805695142,
+      "grad_norm": 0.30928221344947815,
+      "learning_rate": 0.0002,
+      "loss": 1.7339,
+      "step": 1360
+    },
+    {
+      "epoch": 0.9179229480737019,
+      "grad_norm": 0.30652928352355957,
+      "learning_rate": 0.0002,
+      "loss": 1.7479,
+      "step": 1370
+    },
+    {
+      "epoch": 0.9246231155778895,
+      "grad_norm": 0.3838157653808594,
+      "learning_rate": 0.0002,
+      "loss": 1.7491,
+      "step": 1380
+    },
+    {
+      "epoch": 0.931323283082077,
+      "grad_norm": 0.31655240058898926,
+      "learning_rate": 0.0002,
+      "loss": 1.7977,
+      "step": 1390
+    },
+    {
+      "epoch": 0.9380234505862647,
+      "grad_norm": 0.41737303137779236,
+      "learning_rate": 0.0002,
+      "loss": 1.8175,
+      "step": 1400
+    },
+    {
+      "epoch": 0.9447236180904522,
+      "grad_norm": 0.3227267861366272,
+      "learning_rate": 0.0002,
+      "loss": 1.6811,
+      "step": 1410
+    },
+    {
+      "epoch": 0.9514237855946399,
+      "grad_norm": 0.3729925751686096,
+      "learning_rate": 0.0002,
+      "loss": 1.7343,
+      "step": 1420
+    },
+    {
+      "epoch": 0.9581239530988275,
+      "grad_norm": 0.30779409408569336,
+      "learning_rate": 0.0002,
+      "loss": 1.8221,
+      "step": 1430
+    },
+    {
+      "epoch": 0.964824120603015,
+      "grad_norm": 0.334379643201828,
+      "learning_rate": 0.0002,
+      "loss": 1.7972,
+      "step": 1440
+    },
+    {
+      "epoch": 0.9715242881072027,
+      "grad_norm": 0.3568236231803894,
+      "learning_rate": 0.0002,
+      "loss": 1.7141,
+      "step": 1450
+    },
+    {
+      "epoch": 0.9782244556113903,
+      "grad_norm": 0.33310577273368835,
+      "learning_rate": 0.0002,
+      "loss": 1.7541,
+      "step": 1460
+    },
+    {
+      "epoch": 0.9849246231155779,
+      "grad_norm": 0.2972261905670166,
+      "learning_rate": 0.0002,
+      "loss": 1.8511,
+      "step": 1470
+    },
+    {
+      "epoch": 0.9916247906197655,
+      "grad_norm": 0.3322717845439911,
+      "learning_rate": 0.0002,
+      "loss": 1.7654,
+      "step": 1480
+    },
+    {
+      "epoch": 0.998324958123953,
+      "grad_norm": 0.3276330828666687,
+      "learning_rate": 0.0002,
+      "loss": 1.8033,
+      "step": 1490
+    },
+    {
+      "epoch": 0.9996649916247906,
+      "eval_loss": 1.8036354780197144,
+      "eval_runtime": 37.8949,
+      "eval_samples_per_second": 13.59,
+      "eval_steps_per_second": 1.715,
+      "step": 1492
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 11936,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 8,
+  "save_steps": 200,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 6.906955815714816e+16,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-1492/training_args.bin b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-1492/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..064f299b0f3d2a28f5b1f5c68ef32caab3e2dd49
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-1492/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aa7b586fc440d1e22156593e2dd4267d2bdcb8920a02fdf352ea29a9bec3dd94
+size 5560
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-2985/README.md b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-2985/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..835e31ab8469ee39ddc8b2b6b2143a8c66dad510
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-2985/README.md
@@ -0,0 +1,202 @@
+---
+base_model: meta-llama/Meta-Llama-3-8B-Instruct
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.13.1
\ No newline at end of file
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-2985/adapter_config.json b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-2985/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..406c5a08dc4a2a33b52c62a482f98c217c417215
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-2985/adapter_config.json
@@ -0,0 +1,29 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "meta-llama/Meta-Llama-3-8B-Instruct",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-2985/adapter_model.safetensors b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-2985/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..eb6e7189f2f644eaba0f037e36f9798ed96f6276
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-2985/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bfd96df8ee104e7b3825aa8521a0d4d68984970c2c0b69c6892e57ee4766a75d
+size 109069176
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-2985/optimizer.pt b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-2985/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..5ed0048a717946e1b10e202a18778ce3546f396f
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-2985/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:636991cdadb24d5433244c04a05ff58a08f7790337aee34bee6465e0857fcf64
+size 55532666
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-2985/rng_state.pth b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-2985/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..feb1aca0d4082b105b3b7b388da02b234494d28a
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-2985/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:334295595baafc3a59c6c9ae498247797c3e2460976c08b7917b610520d97d56
+size 14244
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-2985/scheduler.pt b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-2985/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..79e491e499bd9b2ad182b3c4b2638bd93a77c563
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-2985/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dd6e6b7e1db1e8263475596c6777128aabdf2ae74656407a474b8e1ecceb9c9d
+size 1064
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-2985/special_tokens_map.json b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-2985/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..b43be96621d147110fb8a18b5776ec6e38516127
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-2985/special_tokens_map.json
@@ -0,0 +1,17 @@
+{
+  "bos_token": {
+    "content": "<|begin_of_text|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<|eot_id|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<|eot_id|>"
+}
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-2985/tokenizer.json b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-2985/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..86a33946b0c77216d2cce91bb28c8fada4a5e80b
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-2985/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3c5cf44023714fb39b05e71e425f8d7b92805ff73f7988b083b8c87f0bf87393
+size 17209961
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-2985/tokenizer_config.json b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-2985/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..061e40d9db3253624f86e8e364c15ef546527c9d
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-2985/tokenizer_config.json
@@ -0,0 +1,2063 @@
+{
+  "added_tokens_decoder": {
+    "128000": {
+      "content": "<|begin_of_text|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128001": {
+      "content": "<|end_of_text|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128002": {
+      "content": "<|reserved_special_token_0|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128003": {
+      "content": "<|reserved_special_token_1|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128004": {
+      "content": "<|reserved_special_token_2|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128005": {
+      "content": "<|reserved_special_token_3|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128006": {
+      "content": "<|start_header_id|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128007": {
+      "content": "<|end_header_id|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128008": {
+      "content": "<|reserved_special_token_4|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128009": {
+      "content": "<|eot_id|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128010": {
+      "content": "<|reserved_special_token_5|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128011": {
+      "content": "<|reserved_special_token_6|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128012": {
+      "content": "<|reserved_special_token_7|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128013": {
+      "content": "<|reserved_special_token_8|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128014": {
+      "content": "<|reserved_special_token_9|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128015": {
+      "content": "<|reserved_special_token_10|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128016": {
+      "content": "<|reserved_special_token_11|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128017": {
+      "content": "<|reserved_special_token_12|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128018": {
+      "content": "<|reserved_special_token_13|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128019": {
+      "content": "<|reserved_special_token_14|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128020": {
+      "content": "<|reserved_special_token_15|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128021": {
+      "content": "<|reserved_special_token_16|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128022": {
+      "content": "<|reserved_special_token_17|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128023": {
+      "content": "<|reserved_special_token_18|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128024": {
+      "content": "<|reserved_special_token_19|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128025": {
+      "content": "<|reserved_special_token_20|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128026": {
+      "content": "<|reserved_special_token_21|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128027": {
+      "content": "<|reserved_special_token_22|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128028": {
+      "content": "<|reserved_special_token_23|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128029": {
+      "content": "<|reserved_special_token_24|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128030": {
+      "content": "<|reserved_special_token_25|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128031": {
+      "content": "<|reserved_special_token_26|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128032": {
+      "content": "<|reserved_special_token_27|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128033": {
+      "content": "<|reserved_special_token_28|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128034": {
+      "content": "<|reserved_special_token_29|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128035": {
+      "content": "<|reserved_special_token_30|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128036": {
+      "content": "<|reserved_special_token_31|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128037": {
+      "content": "<|reserved_special_token_32|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128038": {
+      "content": "<|reserved_special_token_33|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128039": {
+      "content": "<|reserved_special_token_34|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128040": {
+      "content": "<|reserved_special_token_35|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128041": {
+      "content": "<|reserved_special_token_36|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128042": {
+      "content": "<|reserved_special_token_37|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128043": {
+      "content": "<|reserved_special_token_38|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128044": {
+      "content": "<|reserved_special_token_39|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128045": {
+      "content": "<|reserved_special_token_40|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128046": {
+      "content": "<|reserved_special_token_41|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128047": {
+      "content": "<|reserved_special_token_42|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128048": {
+      "content": "<|reserved_special_token_43|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128049": {
+      "content": "<|reserved_special_token_44|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128050": {
+      "content": "<|reserved_special_token_45|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128051": {
+      "content": "<|reserved_special_token_46|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128052": {
+      "content": "<|reserved_special_token_47|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128053": {
+      "content": "<|reserved_special_token_48|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128054": {
+      "content": "<|reserved_special_token_49|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128055": {
+      "content": "<|reserved_special_token_50|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128056": {
+      "content": "<|reserved_special_token_51|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128057": {
+      "content": "<|reserved_special_token_52|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128058": {
+      "content": "<|reserved_special_token_53|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128059": {
+      "content": "<|reserved_special_token_54|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128060": {
+      "content": "<|reserved_special_token_55|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128061": {
+      "content": "<|reserved_special_token_56|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128062": {
+      "content": "<|reserved_special_token_57|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128063": {
+      "content": "<|reserved_special_token_58|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128064": {
+      "content": "<|reserved_special_token_59|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128065": {
+      "content": "<|reserved_special_token_60|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128066": {
+      "content": "<|reserved_special_token_61|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128067": {
+      "content": "<|reserved_special_token_62|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128068": {
+      "content": "<|reserved_special_token_63|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128069": {
+      "content": "<|reserved_special_token_64|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128070": {
+      "content": "<|reserved_special_token_65|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128071": {
+      "content": "<|reserved_special_token_66|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128072": {
+      "content": "<|reserved_special_token_67|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128073": {
+      "content": "<|reserved_special_token_68|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128074": {
+      "content": "<|reserved_special_token_69|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128075": {
+      "content": "<|reserved_special_token_70|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128076": {
+      "content": "<|reserved_special_token_71|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128077": {
+      "content": "<|reserved_special_token_72|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128078": {
+      "content": "<|reserved_special_token_73|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128079": {
+      "content": "<|reserved_special_token_74|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128080": {
+      "content": "<|reserved_special_token_75|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128081": {
+      "content": "<|reserved_special_token_76|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128082": {
+      "content": "<|reserved_special_token_77|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128083": {
+      "content": "<|reserved_special_token_78|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128084": {
+      "content": "<|reserved_special_token_79|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128085": {
+      "content": "<|reserved_special_token_80|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128086": {
+      "content": "<|reserved_special_token_81|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128087": {
+      "content": "<|reserved_special_token_82|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128088": {
+      "content": "<|reserved_special_token_83|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128089": {
+      "content": "<|reserved_special_token_84|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128090": {
+      "content": "<|reserved_special_token_85|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128091": {
+      "content": "<|reserved_special_token_86|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128092": {
+      "content": "<|reserved_special_token_87|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128093": {
+      "content": "<|reserved_special_token_88|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128094": {
+      "content": "<|reserved_special_token_89|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128095": {
+      "content": "<|reserved_special_token_90|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128096": {
+      "content": "<|reserved_special_token_91|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128097": {
+      "content": "<|reserved_special_token_92|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128098": {
+      "content": "<|reserved_special_token_93|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128099": {
+      "content": "<|reserved_special_token_94|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128100": {
+      "content": "<|reserved_special_token_95|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128101": {
+      "content": "<|reserved_special_token_96|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128102": {
+      "content": "<|reserved_special_token_97|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128103": {
+      "content": "<|reserved_special_token_98|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128104": {
+      "content": "<|reserved_special_token_99|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128105": {
+      "content": "<|reserved_special_token_100|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128106": {
+      "content": "<|reserved_special_token_101|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128107": {
+      "content": "<|reserved_special_token_102|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128108": {
+      "content": "<|reserved_special_token_103|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128109": {
+      "content": "<|reserved_special_token_104|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128110": {
+      "content": "<|reserved_special_token_105|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128111": {
+      "content": "<|reserved_special_token_106|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128112": {
+      "content": "<|reserved_special_token_107|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128113": {
+      "content": "<|reserved_special_token_108|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128114": {
+      "content": "<|reserved_special_token_109|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128115": {
+      "content": "<|reserved_special_token_110|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128116": {
+      "content": "<|reserved_special_token_111|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128117": {
+      "content": "<|reserved_special_token_112|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128118": {
+      "content": "<|reserved_special_token_113|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128119": {
+      "content": "<|reserved_special_token_114|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128120": {
+      "content": "<|reserved_special_token_115|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128121": {
+      "content": "<|reserved_special_token_116|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128122": {
+      "content": "<|reserved_special_token_117|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128123": {
+      "content": "<|reserved_special_token_118|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128124": {
+      "content": "<|reserved_special_token_119|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128125": {
+      "content": "<|reserved_special_token_120|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128126": {
+      "content": "<|reserved_special_token_121|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128127": {
+      "content": "<|reserved_special_token_122|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128128": {
+      "content": "<|reserved_special_token_123|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128129": {
+      "content": "<|reserved_special_token_124|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128130": {
+      "content": "<|reserved_special_token_125|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128131": {
+      "content": "<|reserved_special_token_126|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128132": {
+      "content": "<|reserved_special_token_127|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128133": {
+      "content": "<|reserved_special_token_128|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128134": {
+      "content": "<|reserved_special_token_129|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128135": {
+      "content": "<|reserved_special_token_130|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128136": {
+      "content": "<|reserved_special_token_131|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128137": {
+      "content": "<|reserved_special_token_132|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128138": {
+      "content": "<|reserved_special_token_133|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128139": {
+      "content": "<|reserved_special_token_134|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128140": {
+      "content": "<|reserved_special_token_135|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128141": {
+      "content": "<|reserved_special_token_136|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128142": {
+      "content": "<|reserved_special_token_137|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128143": {
+      "content": "<|reserved_special_token_138|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128144": {
+      "content": "<|reserved_special_token_139|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128145": {
+      "content": "<|reserved_special_token_140|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128146": {
+      "content": "<|reserved_special_token_141|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128147": {
+      "content": "<|reserved_special_token_142|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128148": {
+      "content": "<|reserved_special_token_143|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128149": {
+      "content": "<|reserved_special_token_144|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128150": {
+      "content": "<|reserved_special_token_145|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128151": {
+      "content": "<|reserved_special_token_146|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128152": {
+      "content": "<|reserved_special_token_147|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128153": {
+      "content": "<|reserved_special_token_148|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128154": {
+      "content": "<|reserved_special_token_149|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128155": {
+      "content": "<|reserved_special_token_150|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128156": {
+      "content": "<|reserved_special_token_151|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128157": {
+      "content": "<|reserved_special_token_152|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128158": {
+      "content": "<|reserved_special_token_153|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128159": {
+      "content": "<|reserved_special_token_154|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128160": {
+      "content": "<|reserved_special_token_155|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128161": {
+      "content": "<|reserved_special_token_156|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128162": {
+      "content": "<|reserved_special_token_157|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128163": {
+      "content": "<|reserved_special_token_158|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128164": {
+      "content": "<|reserved_special_token_159|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128165": {
+      "content": "<|reserved_special_token_160|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128166": {
+      "content": "<|reserved_special_token_161|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128167": {
+      "content": "<|reserved_special_token_162|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128168": {
+      "content": "<|reserved_special_token_163|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128169": {
+      "content": "<|reserved_special_token_164|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128170": {
+      "content": "<|reserved_special_token_165|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128171": {
+      "content": "<|reserved_special_token_166|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128172": {
+      "content": "<|reserved_special_token_167|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128173": {
+      "content": "<|reserved_special_token_168|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128174": {
+      "content": "<|reserved_special_token_169|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128175": {
+      "content": "<|reserved_special_token_170|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128176": {
+      "content": "<|reserved_special_token_171|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128177": {
+      "content": "<|reserved_special_token_172|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128178": {
+      "content": "<|reserved_special_token_173|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128179": {
+      "content": "<|reserved_special_token_174|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128180": {
+      "content": "<|reserved_special_token_175|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128181": {
+      "content": "<|reserved_special_token_176|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128182": {
+      "content": "<|reserved_special_token_177|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128183": {
+      "content": "<|reserved_special_token_178|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128184": {
+      "content": "<|reserved_special_token_179|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128185": {
+      "content": "<|reserved_special_token_180|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128186": {
+      "content": "<|reserved_special_token_181|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128187": {
+      "content": "<|reserved_special_token_182|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128188": {
+      "content": "<|reserved_special_token_183|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128189": {
+      "content": "<|reserved_special_token_184|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128190": {
+      "content": "<|reserved_special_token_185|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128191": {
+      "content": "<|reserved_special_token_186|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128192": {
+      "content": "<|reserved_special_token_187|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128193": {
+      "content": "<|reserved_special_token_188|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128194": {
+      "content": "<|reserved_special_token_189|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128195": {
+      "content": "<|reserved_special_token_190|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128196": {
+      "content": "<|reserved_special_token_191|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128197": {
+      "content": "<|reserved_special_token_192|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128198": {
+      "content": "<|reserved_special_token_193|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128199": {
+      "content": "<|reserved_special_token_194|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128200": {
+      "content": "<|reserved_special_token_195|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128201": {
+      "content": "<|reserved_special_token_196|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128202": {
+      "content": "<|reserved_special_token_197|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128203": {
+      "content": "<|reserved_special_token_198|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128204": {
+      "content": "<|reserved_special_token_199|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128205": {
+      "content": "<|reserved_special_token_200|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128206": {
+      "content": "<|reserved_special_token_201|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128207": {
+      "content": "<|reserved_special_token_202|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128208": {
+      "content": "<|reserved_special_token_203|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128209": {
+      "content": "<|reserved_special_token_204|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128210": {
+      "content": "<|reserved_special_token_205|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128211": {
+      "content": "<|reserved_special_token_206|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128212": {
+      "content": "<|reserved_special_token_207|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128213": {
+      "content": "<|reserved_special_token_208|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128214": {
+      "content": "<|reserved_special_token_209|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128215": {
+      "content": "<|reserved_special_token_210|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128216": {
+      "content": "<|reserved_special_token_211|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128217": {
+      "content": "<|reserved_special_token_212|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128218": {
+      "content": "<|reserved_special_token_213|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128219": {
+      "content": "<|reserved_special_token_214|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128220": {
+      "content": "<|reserved_special_token_215|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128221": {
+      "content": "<|reserved_special_token_216|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128222": {
+      "content": "<|reserved_special_token_217|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128223": {
+      "content": "<|reserved_special_token_218|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128224": {
+      "content": "<|reserved_special_token_219|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128225": {
+      "content": "<|reserved_special_token_220|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128226": {
+      "content": "<|reserved_special_token_221|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128227": {
+      "content": "<|reserved_special_token_222|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128228": {
+      "content": "<|reserved_special_token_223|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128229": {
+      "content": "<|reserved_special_token_224|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128230": {
+      "content": "<|reserved_special_token_225|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128231": {
+      "content": "<|reserved_special_token_226|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128232": {
+      "content": "<|reserved_special_token_227|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128233": {
+      "content": "<|reserved_special_token_228|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128234": {
+      "content": "<|reserved_special_token_229|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128235": {
+      "content": "<|reserved_special_token_230|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128236": {
+      "content": "<|reserved_special_token_231|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128237": {
+      "content": "<|reserved_special_token_232|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128238": {
+      "content": "<|reserved_special_token_233|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128239": {
+      "content": "<|reserved_special_token_234|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128240": {
+      "content": "<|reserved_special_token_235|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128241": {
+      "content": "<|reserved_special_token_236|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128242": {
+      "content": "<|reserved_special_token_237|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128243": {
+      "content": "<|reserved_special_token_238|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128244": {
+      "content": "<|reserved_special_token_239|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128245": {
+      "content": "<|reserved_special_token_240|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128246": {
+      "content": "<|reserved_special_token_241|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128247": {
+      "content": "<|reserved_special_token_242|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128248": {
+      "content": "<|reserved_special_token_243|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128249": {
+      "content": "<|reserved_special_token_244|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128250": {
+      "content": "<|reserved_special_token_245|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128251": {
+      "content": "<|reserved_special_token_246|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128252": {
+      "content": "<|reserved_special_token_247|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128253": {
+      "content": "<|reserved_special_token_248|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128254": {
+      "content": "<|reserved_special_token_249|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128255": {
+      "content": "<|reserved_special_token_250|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<|begin_of_text|>",
+  "chat_template": "{% set loop_messages = messages %}{% for message in loop_messages %}{% set content = '<|start_header_id|>' + message['role'] + '<|end_header_id|>\n\n'+ message['content'] | trim + '<|eot_id|>' %}{% if loop.index0 == 0 %}{% set content = bos_token + content %}{% endif %}{{ content }}{% endfor %}{% if add_generation_prompt %}{{ '<|start_header_id|>assistant<|end_header_id|>\n\n' }}{% endif %}",
+  "clean_up_tokenization_spaces": true,
+  "eos_token": "<|eot_id|>",
+  "model_input_names": [
+    "input_ids",
+    "attention_mask"
+  ],
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<|eot_id|>",
+  "tokenizer_class": "PreTrainedTokenizerFast"
+}
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-2985/trainer_state.json b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-2985/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..dbb477671bbe8bee88068baadcbb083f6c7bc743
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-2985/trainer_state.json
@@ -0,0 +1,2135 @@
+{
+  "best_metric": 1.8028968572616577,
+  "best_model_checkpoint": "outputs-001/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-2985",
+  "epoch": 2.0,
+  "eval_steps": 10,
+  "global_step": 2985,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.006700167504187605,
+      "grad_norm": 0.565915048122406,
+      "learning_rate": 0.0002,
+      "loss": 2.6189,
+      "step": 10
+    },
+    {
+      "epoch": 0.01340033500837521,
+      "grad_norm": 0.5004463791847229,
+      "learning_rate": 0.0002,
+      "loss": 2.3162,
+      "step": 20
+    },
+    {
+      "epoch": 0.020100502512562814,
+      "grad_norm": 0.511043906211853,
+      "learning_rate": 0.0002,
+      "loss": 2.0576,
+      "step": 30
+    },
+    {
+      "epoch": 0.02680067001675042,
+      "grad_norm": 0.47327178716659546,
+      "learning_rate": 0.0002,
+      "loss": 2.0085,
+      "step": 40
+    },
+    {
+      "epoch": 0.03350083752093802,
+      "grad_norm": 0.5511676669120789,
+      "learning_rate": 0.0002,
+      "loss": 2.0276,
+      "step": 50
+    },
+    {
+      "epoch": 0.04020100502512563,
+      "grad_norm": 0.4666278064250946,
+      "learning_rate": 0.0002,
+      "loss": 1.9075,
+      "step": 60
+    },
+    {
+      "epoch": 0.04690117252931323,
+      "grad_norm": 0.5310961008071899,
+      "learning_rate": 0.0002,
+      "loss": 1.8413,
+      "step": 70
+    },
+    {
+      "epoch": 0.05360134003350084,
+      "grad_norm": 0.5606027245521545,
+      "learning_rate": 0.0002,
+      "loss": 1.8711,
+      "step": 80
+    },
+    {
+      "epoch": 0.06030150753768844,
+      "grad_norm": 0.4934779703617096,
+      "learning_rate": 0.0002,
+      "loss": 1.9282,
+      "step": 90
+    },
+    {
+      "epoch": 0.06700167504187604,
+      "grad_norm": 0.4821869730949402,
+      "learning_rate": 0.0002,
+      "loss": 1.8925,
+      "step": 100
+    },
+    {
+      "epoch": 0.07370184254606366,
+      "grad_norm": 0.5262084603309631,
+      "learning_rate": 0.0002,
+      "loss": 1.8628,
+      "step": 110
+    },
+    {
+      "epoch": 0.08040201005025126,
+      "grad_norm": 0.3774230182170868,
+      "learning_rate": 0.0002,
+      "loss": 1.8347,
+      "step": 120
+    },
+    {
+      "epoch": 0.08710217755443886,
+      "grad_norm": 0.34137430787086487,
+      "learning_rate": 0.0002,
+      "loss": 1.8386,
+      "step": 130
+    },
+    {
+      "epoch": 0.09380234505862646,
+      "grad_norm": 0.407272070646286,
+      "learning_rate": 0.0002,
+      "loss": 1.861,
+      "step": 140
+    },
+    {
+      "epoch": 0.10050251256281408,
+      "grad_norm": 0.4011937975883484,
+      "learning_rate": 0.0002,
+      "loss": 1.8279,
+      "step": 150
+    },
+    {
+      "epoch": 0.10720268006700168,
+      "grad_norm": 0.4432467222213745,
+      "learning_rate": 0.0002,
+      "loss": 1.9317,
+      "step": 160
+    },
+    {
+      "epoch": 0.11390284757118928,
+      "grad_norm": 0.44030463695526123,
+      "learning_rate": 0.0002,
+      "loss": 1.8157,
+      "step": 170
+    },
+    {
+      "epoch": 0.12060301507537688,
+      "grad_norm": 0.3799569308757782,
+      "learning_rate": 0.0002,
+      "loss": 1.8534,
+      "step": 180
+    },
+    {
+      "epoch": 0.1273031825795645,
+      "grad_norm": 0.33721521496772766,
+      "learning_rate": 0.0002,
+      "loss": 1.7658,
+      "step": 190
+    },
+    {
+      "epoch": 0.13400335008375208,
+      "grad_norm": 0.4096226692199707,
+      "learning_rate": 0.0002,
+      "loss": 1.8269,
+      "step": 200
+    },
+    {
+      "epoch": 0.1407035175879397,
+      "grad_norm": 0.37374693155288696,
+      "learning_rate": 0.0002,
+      "loss": 1.802,
+      "step": 210
+    },
+    {
+      "epoch": 0.1474036850921273,
+      "grad_norm": 0.3249480128288269,
+      "learning_rate": 0.0002,
+      "loss": 1.8901,
+      "step": 220
+    },
+    {
+      "epoch": 0.1541038525963149,
+      "grad_norm": 0.3612042963504791,
+      "learning_rate": 0.0002,
+      "loss": 1.8163,
+      "step": 230
+    },
+    {
+      "epoch": 0.16080402010050251,
+      "grad_norm": 0.3686671257019043,
+      "learning_rate": 0.0002,
+      "loss": 1.7585,
+      "step": 240
+    },
+    {
+      "epoch": 0.16750418760469013,
+      "grad_norm": 0.3521044850349426,
+      "learning_rate": 0.0002,
+      "loss": 1.8365,
+      "step": 250
+    },
+    {
+      "epoch": 0.17420435510887772,
+      "grad_norm": 0.4073677361011505,
+      "learning_rate": 0.0002,
+      "loss": 1.8623,
+      "step": 260
+    },
+    {
+      "epoch": 0.18090452261306533,
+      "grad_norm": 0.34522193670272827,
+      "learning_rate": 0.0002,
+      "loss": 1.8026,
+      "step": 270
+    },
+    {
+      "epoch": 0.18760469011725292,
+      "grad_norm": 0.4121900498867035,
+      "learning_rate": 0.0002,
+      "loss": 1.8162,
+      "step": 280
+    },
+    {
+      "epoch": 0.19430485762144054,
+      "grad_norm": 0.3544778525829315,
+      "learning_rate": 0.0002,
+      "loss": 1.7976,
+      "step": 290
+    },
+    {
+      "epoch": 0.20100502512562815,
+      "grad_norm": 0.3482133448123932,
+      "learning_rate": 0.0002,
+      "loss": 1.8787,
+      "step": 300
+    },
+    {
+      "epoch": 0.20770519262981574,
+      "grad_norm": 0.3421826660633087,
+      "learning_rate": 0.0002,
+      "loss": 1.8578,
+      "step": 310
+    },
+    {
+      "epoch": 0.21440536013400335,
+      "grad_norm": 0.5024696588516235,
+      "learning_rate": 0.0002,
+      "loss": 1.8013,
+      "step": 320
+    },
+    {
+      "epoch": 0.22110552763819097,
+      "grad_norm": 0.36013063788414,
+      "learning_rate": 0.0002,
+      "loss": 1.8607,
+      "step": 330
+    },
+    {
+      "epoch": 0.22780569514237856,
+      "grad_norm": 0.3611244857311249,
+      "learning_rate": 0.0002,
+      "loss": 1.9075,
+      "step": 340
+    },
+    {
+      "epoch": 0.23450586264656617,
+      "grad_norm": 0.39244529604911804,
+      "learning_rate": 0.0002,
+      "loss": 1.8128,
+      "step": 350
+    },
+    {
+      "epoch": 0.24120603015075376,
+      "grad_norm": 0.3299325704574585,
+      "learning_rate": 0.0002,
+      "loss": 1.7885,
+      "step": 360
+    },
+    {
+      "epoch": 0.24790619765494137,
+      "grad_norm": 0.3994322419166565,
+      "learning_rate": 0.0002,
+      "loss": 1.8028,
+      "step": 370
+    },
+    {
+      "epoch": 0.254606365159129,
+      "grad_norm": 0.3559151887893677,
+      "learning_rate": 0.0002,
+      "loss": 1.8321,
+      "step": 380
+    },
+    {
+      "epoch": 0.2613065326633166,
+      "grad_norm": 0.3873756229877472,
+      "learning_rate": 0.0002,
+      "loss": 1.7802,
+      "step": 390
+    },
+    {
+      "epoch": 0.26800670016750416,
+      "grad_norm": 0.3710744082927704,
+      "learning_rate": 0.0002,
+      "loss": 1.7844,
+      "step": 400
+    },
+    {
+      "epoch": 0.2747068676716918,
+      "grad_norm": 0.3618465065956116,
+      "learning_rate": 0.0002,
+      "loss": 1.7789,
+      "step": 410
+    },
+    {
+      "epoch": 0.2814070351758794,
+      "grad_norm": 0.30063769221305847,
+      "learning_rate": 0.0002,
+      "loss": 1.8529,
+      "step": 420
+    },
+    {
+      "epoch": 0.288107202680067,
+      "grad_norm": 0.3695628345012665,
+      "learning_rate": 0.0002,
+      "loss": 1.7765,
+      "step": 430
+    },
+    {
+      "epoch": 0.2948073701842546,
+      "grad_norm": 0.31451135873794556,
+      "learning_rate": 0.0002,
+      "loss": 1.7982,
+      "step": 440
+    },
+    {
+      "epoch": 0.3015075376884422,
+      "grad_norm": 0.3959707021713257,
+      "learning_rate": 0.0002,
+      "loss": 1.7517,
+      "step": 450
+    },
+    {
+      "epoch": 0.3082077051926298,
+      "grad_norm": 0.33475354313850403,
+      "learning_rate": 0.0002,
+      "loss": 1.8142,
+      "step": 460
+    },
+    {
+      "epoch": 0.3149078726968174,
+      "grad_norm": 0.33933115005493164,
+      "learning_rate": 0.0002,
+      "loss": 1.8805,
+      "step": 470
+    },
+    {
+      "epoch": 0.32160804020100503,
+      "grad_norm": 0.3264943063259125,
+      "learning_rate": 0.0002,
+      "loss": 1.7564,
+      "step": 480
+    },
+    {
+      "epoch": 0.32830820770519265,
+      "grad_norm": 0.40188100934028625,
+      "learning_rate": 0.0002,
+      "loss": 1.8428,
+      "step": 490
+    },
+    {
+      "epoch": 0.33500837520938026,
+      "grad_norm": 0.37408649921417236,
+      "learning_rate": 0.0002,
+      "loss": 1.7624,
+      "step": 500
+    },
+    {
+      "epoch": 0.3417085427135678,
+      "grad_norm": 0.33925938606262207,
+      "learning_rate": 0.0002,
+      "loss": 1.7745,
+      "step": 510
+    },
+    {
+      "epoch": 0.34840871021775544,
+      "grad_norm": 0.36836713552474976,
+      "learning_rate": 0.0002,
+      "loss": 1.814,
+      "step": 520
+    },
+    {
+      "epoch": 0.35510887772194305,
+      "grad_norm": 0.37284499406814575,
+      "learning_rate": 0.0002,
+      "loss": 1.8037,
+      "step": 530
+    },
+    {
+      "epoch": 0.36180904522613067,
+      "grad_norm": 0.3192278742790222,
+      "learning_rate": 0.0002,
+      "loss": 1.8379,
+      "step": 540
+    },
+    {
+      "epoch": 0.3685092127303183,
+      "grad_norm": 0.30233290791511536,
+      "learning_rate": 0.0002,
+      "loss": 1.8702,
+      "step": 550
+    },
+    {
+      "epoch": 0.37520938023450584,
+      "grad_norm": 0.3340817391872406,
+      "learning_rate": 0.0002,
+      "loss": 1.8799,
+      "step": 560
+    },
+    {
+      "epoch": 0.38190954773869346,
+      "grad_norm": 0.32600095868110657,
+      "learning_rate": 0.0002,
+      "loss": 1.8404,
+      "step": 570
+    },
+    {
+      "epoch": 0.38860971524288107,
+      "grad_norm": 0.33711278438568115,
+      "learning_rate": 0.0002,
+      "loss": 1.7804,
+      "step": 580
+    },
+    {
+      "epoch": 0.3953098827470687,
+      "grad_norm": 0.34890690445899963,
+      "learning_rate": 0.0002,
+      "loss": 1.8445,
+      "step": 590
+    },
+    {
+      "epoch": 0.4020100502512563,
+      "grad_norm": 0.38238924741744995,
+      "learning_rate": 0.0002,
+      "loss": 1.8187,
+      "step": 600
+    },
+    {
+      "epoch": 0.40871021775544386,
+      "grad_norm": 0.34399354457855225,
+      "learning_rate": 0.0002,
+      "loss": 1.8111,
+      "step": 610
+    },
+    {
+      "epoch": 0.4154103852596315,
+      "grad_norm": 0.3346073627471924,
+      "learning_rate": 0.0002,
+      "loss": 1.8006,
+      "step": 620
+    },
+    {
+      "epoch": 0.4221105527638191,
+      "grad_norm": 0.3545648157596588,
+      "learning_rate": 0.0002,
+      "loss": 1.7705,
+      "step": 630
+    },
+    {
+      "epoch": 0.4288107202680067,
+      "grad_norm": 0.3378899097442627,
+      "learning_rate": 0.0002,
+      "loss": 1.8445,
+      "step": 640
+    },
+    {
+      "epoch": 0.4355108877721943,
+      "grad_norm": 0.3255569040775299,
+      "learning_rate": 0.0002,
+      "loss": 1.804,
+      "step": 650
+    },
+    {
+      "epoch": 0.44221105527638194,
+      "grad_norm": 0.34880587458610535,
+      "learning_rate": 0.0002,
+      "loss": 1.7679,
+      "step": 660
+    },
+    {
+      "epoch": 0.4489112227805695,
+      "grad_norm": 0.3402383625507355,
+      "learning_rate": 0.0002,
+      "loss": 1.7861,
+      "step": 670
+    },
+    {
+      "epoch": 0.4556113902847571,
+      "grad_norm": 0.3594033718109131,
+      "learning_rate": 0.0002,
+      "loss": 1.8131,
+      "step": 680
+    },
+    {
+      "epoch": 0.4623115577889447,
+      "grad_norm": 0.31000566482543945,
+      "learning_rate": 0.0002,
+      "loss": 1.8399,
+      "step": 690
+    },
+    {
+      "epoch": 0.46901172529313234,
+      "grad_norm": 0.37229061126708984,
+      "learning_rate": 0.0002,
+      "loss": 1.7521,
+      "step": 700
+    },
+    {
+      "epoch": 0.47571189279731996,
+      "grad_norm": 0.315801739692688,
+      "learning_rate": 0.0002,
+      "loss": 1.7779,
+      "step": 710
+    },
+    {
+      "epoch": 0.4824120603015075,
+      "grad_norm": 0.3220832645893097,
+      "learning_rate": 0.0002,
+      "loss": 1.7515,
+      "step": 720
+    },
+    {
+      "epoch": 0.48911222780569513,
+      "grad_norm": 0.3435456156730652,
+      "learning_rate": 0.0002,
+      "loss": 1.7181,
+      "step": 730
+    },
+    {
+      "epoch": 0.49581239530988275,
+      "grad_norm": 0.30380892753601074,
+      "learning_rate": 0.0002,
+      "loss": 1.8844,
+      "step": 740
+    },
+    {
+      "epoch": 0.5025125628140703,
+      "grad_norm": 0.3555026054382324,
+      "learning_rate": 0.0002,
+      "loss": 1.7792,
+      "step": 750
+    },
+    {
+      "epoch": 0.509212730318258,
+      "grad_norm": 0.3019855320453644,
+      "learning_rate": 0.0002,
+      "loss": 1.7714,
+      "step": 760
+    },
+    {
+      "epoch": 0.5159128978224455,
+      "grad_norm": 0.309111088514328,
+      "learning_rate": 0.0002,
+      "loss": 1.7962,
+      "step": 770
+    },
+    {
+      "epoch": 0.5226130653266332,
+      "grad_norm": 0.366020530462265,
+      "learning_rate": 0.0002,
+      "loss": 1.7913,
+      "step": 780
+    },
+    {
+      "epoch": 0.5293132328308208,
+      "grad_norm": 0.3267050087451935,
+      "learning_rate": 0.0002,
+      "loss": 1.8008,
+      "step": 790
+    },
+    {
+      "epoch": 0.5360134003350083,
+      "grad_norm": 0.34265750646591187,
+      "learning_rate": 0.0002,
+      "loss": 1.7397,
+      "step": 800
+    },
+    {
+      "epoch": 0.542713567839196,
+      "grad_norm": 0.313669890165329,
+      "learning_rate": 0.0002,
+      "loss": 1.8251,
+      "step": 810
+    },
+    {
+      "epoch": 0.5494137353433836,
+      "grad_norm": 0.3355236053466797,
+      "learning_rate": 0.0002,
+      "loss": 1.8369,
+      "step": 820
+    },
+    {
+      "epoch": 0.5561139028475712,
+      "grad_norm": 0.3186608552932739,
+      "learning_rate": 0.0002,
+      "loss": 1.7381,
+      "step": 830
+    },
+    {
+      "epoch": 0.5628140703517588,
+      "grad_norm": 0.30357518792152405,
+      "learning_rate": 0.0002,
+      "loss": 1.8034,
+      "step": 840
+    },
+    {
+      "epoch": 0.5695142378559463,
+      "grad_norm": 0.3990040123462677,
+      "learning_rate": 0.0002,
+      "loss": 1.769,
+      "step": 850
+    },
+    {
+      "epoch": 0.576214405360134,
+      "grad_norm": 0.34363803267478943,
+      "learning_rate": 0.0002,
+      "loss": 1.7482,
+      "step": 860
+    },
+    {
+      "epoch": 0.5829145728643216,
+      "grad_norm": 0.3757908046245575,
+      "learning_rate": 0.0002,
+      "loss": 1.8106,
+      "step": 870
+    },
+    {
+      "epoch": 0.5896147403685092,
+      "grad_norm": 0.3359757661819458,
+      "learning_rate": 0.0002,
+      "loss": 1.8104,
+      "step": 880
+    },
+    {
+      "epoch": 0.5963149078726968,
+      "grad_norm": 0.5555329918861389,
+      "learning_rate": 0.0002,
+      "loss": 1.7591,
+      "step": 890
+    },
+    {
+      "epoch": 0.6030150753768844,
+      "grad_norm": 0.4046323895454407,
+      "learning_rate": 0.0002,
+      "loss": 1.7715,
+      "step": 900
+    },
+    {
+      "epoch": 0.609715242881072,
+      "grad_norm": 0.29834219813346863,
+      "learning_rate": 0.0002,
+      "loss": 1.7998,
+      "step": 910
+    },
+    {
+      "epoch": 0.6164154103852596,
+      "grad_norm": 0.3241238594055176,
+      "learning_rate": 0.0002,
+      "loss": 1.7826,
+      "step": 920
+    },
+    {
+      "epoch": 0.6231155778894473,
+      "grad_norm": 0.35154739022254944,
+      "learning_rate": 0.0002,
+      "loss": 1.8342,
+      "step": 930
+    },
+    {
+      "epoch": 0.6298157453936348,
+      "grad_norm": 0.3287706673145294,
+      "learning_rate": 0.0002,
+      "loss": 1.8076,
+      "step": 940
+    },
+    {
+      "epoch": 0.6365159128978225,
+      "grad_norm": 0.35670626163482666,
+      "learning_rate": 0.0002,
+      "loss": 1.8038,
+      "step": 950
+    },
+    {
+      "epoch": 0.6432160804020101,
+      "grad_norm": 0.6114104986190796,
+      "learning_rate": 0.0002,
+      "loss": 1.869,
+      "step": 960
+    },
+    {
+      "epoch": 0.6499162479061976,
+      "grad_norm": 0.3186565041542053,
+      "learning_rate": 0.0002,
+      "loss": 1.8297,
+      "step": 970
+    },
+    {
+      "epoch": 0.6566164154103853,
+      "grad_norm": 0.27164125442504883,
+      "learning_rate": 0.0002,
+      "loss": 1.7539,
+      "step": 980
+    },
+    {
+      "epoch": 0.6633165829145728,
+      "grad_norm": 0.34407344460487366,
+      "learning_rate": 0.0002,
+      "loss": 1.8339,
+      "step": 990
+    },
+    {
+      "epoch": 0.6700167504187605,
+      "grad_norm": 0.368415892124176,
+      "learning_rate": 0.0002,
+      "loss": 1.855,
+      "step": 1000
+    },
+    {
+      "epoch": 0.6767169179229481,
+      "grad_norm": 0.3306390643119812,
+      "learning_rate": 0.0002,
+      "loss": 1.7821,
+      "step": 1010
+    },
+    {
+      "epoch": 0.6834170854271356,
+      "grad_norm": 0.3198648989200592,
+      "learning_rate": 0.0002,
+      "loss": 1.786,
+      "step": 1020
+    },
+    {
+      "epoch": 0.6901172529313233,
+      "grad_norm": 0.3092987537384033,
+      "learning_rate": 0.0002,
+      "loss": 1.816,
+      "step": 1030
+    },
+    {
+      "epoch": 0.6968174204355109,
+      "grad_norm": 0.3090653419494629,
+      "learning_rate": 0.0002,
+      "loss": 1.7689,
+      "step": 1040
+    },
+    {
+      "epoch": 0.7035175879396985,
+      "grad_norm": 0.3485880196094513,
+      "learning_rate": 0.0002,
+      "loss": 1.7544,
+      "step": 1050
+    },
+    {
+      "epoch": 0.7102177554438861,
+      "grad_norm": 0.35782721638679504,
+      "learning_rate": 0.0002,
+      "loss": 1.8187,
+      "step": 1060
+    },
+    {
+      "epoch": 0.7169179229480737,
+      "grad_norm": 0.34256869554519653,
+      "learning_rate": 0.0002,
+      "loss": 1.806,
+      "step": 1070
+    },
+    {
+      "epoch": 0.7236180904522613,
+      "grad_norm": 0.30461037158966064,
+      "learning_rate": 0.0002,
+      "loss": 1.7873,
+      "step": 1080
+    },
+    {
+      "epoch": 0.7303182579564489,
+      "grad_norm": 0.3398691713809967,
+      "learning_rate": 0.0002,
+      "loss": 1.7367,
+      "step": 1090
+    },
+    {
+      "epoch": 0.7370184254606366,
+      "grad_norm": 0.3180808126926422,
+      "learning_rate": 0.0002,
+      "loss": 1.8756,
+      "step": 1100
+    },
+    {
+      "epoch": 0.7437185929648241,
+      "grad_norm": 0.34400665760040283,
+      "learning_rate": 0.0002,
+      "loss": 1.6988,
+      "step": 1110
+    },
+    {
+      "epoch": 0.7504187604690117,
+      "grad_norm": 0.34244877099990845,
+      "learning_rate": 0.0002,
+      "loss": 1.7851,
+      "step": 1120
+    },
+    {
+      "epoch": 0.7571189279731994,
+      "grad_norm": 0.29946693778038025,
+      "learning_rate": 0.0002,
+      "loss": 1.7841,
+      "step": 1130
+    },
+    {
+      "epoch": 0.7638190954773869,
+      "grad_norm": 0.37547236680984497,
+      "learning_rate": 0.0002,
+      "loss": 1.7456,
+      "step": 1140
+    },
+    {
+      "epoch": 0.7705192629815746,
+      "grad_norm": 0.3263005018234253,
+      "learning_rate": 0.0002,
+      "loss": 1.8425,
+      "step": 1150
+    },
+    {
+      "epoch": 0.7772194304857621,
+      "grad_norm": 0.41363608837127686,
+      "learning_rate": 0.0002,
+      "loss": 1.7222,
+      "step": 1160
+    },
+    {
+      "epoch": 0.7839195979899497,
+      "grad_norm": 0.36267954111099243,
+      "learning_rate": 0.0002,
+      "loss": 1.7836,
+      "step": 1170
+    },
+    {
+      "epoch": 0.7906197654941374,
+      "grad_norm": 0.31789499521255493,
+      "learning_rate": 0.0002,
+      "loss": 1.9183,
+      "step": 1180
+    },
+    {
+      "epoch": 0.7973199329983249,
+      "grad_norm": 0.5708149075508118,
+      "learning_rate": 0.0002,
+      "loss": 1.78,
+      "step": 1190
+    },
+    {
+      "epoch": 0.8040201005025126,
+      "grad_norm": 0.322099506855011,
+      "learning_rate": 0.0002,
+      "loss": 1.6908,
+      "step": 1200
+    },
+    {
+      "epoch": 0.8107202680067002,
+      "grad_norm": 0.3419909179210663,
+      "learning_rate": 0.0002,
+      "loss": 1.7639,
+      "step": 1210
+    },
+    {
+      "epoch": 0.8174204355108877,
+      "grad_norm": 0.36286255717277527,
+      "learning_rate": 0.0002,
+      "loss": 1.7428,
+      "step": 1220
+    },
+    {
+      "epoch": 0.8241206030150754,
+      "grad_norm": 0.33992862701416016,
+      "learning_rate": 0.0002,
+      "loss": 1.8409,
+      "step": 1230
+    },
+    {
+      "epoch": 0.830820770519263,
+      "grad_norm": 0.32622793316841125,
+      "learning_rate": 0.0002,
+      "loss": 1.7507,
+      "step": 1240
+    },
+    {
+      "epoch": 0.8375209380234506,
+      "grad_norm": 0.3036167621612549,
+      "learning_rate": 0.0002,
+      "loss": 1.8098,
+      "step": 1250
+    },
+    {
+      "epoch": 0.8442211055276382,
+      "grad_norm": 0.3182215392589569,
+      "learning_rate": 0.0002,
+      "loss": 1.8094,
+      "step": 1260
+    },
+    {
+      "epoch": 0.8509212730318257,
+      "grad_norm": 0.3270018696784973,
+      "learning_rate": 0.0002,
+      "loss": 1.8017,
+      "step": 1270
+    },
+    {
+      "epoch": 0.8576214405360134,
+      "grad_norm": 0.32652342319488525,
+      "learning_rate": 0.0002,
+      "loss": 1.798,
+      "step": 1280
+    },
+    {
+      "epoch": 0.864321608040201,
+      "grad_norm": 0.3631329834461212,
+      "learning_rate": 0.0002,
+      "loss": 1.7448,
+      "step": 1290
+    },
+    {
+      "epoch": 0.8710217755443886,
+      "grad_norm": 0.36706018447875977,
+      "learning_rate": 0.0002,
+      "loss": 1.7,
+      "step": 1300
+    },
+    {
+      "epoch": 0.8777219430485762,
+      "grad_norm": 0.3347418010234833,
+      "learning_rate": 0.0002,
+      "loss": 1.8178,
+      "step": 1310
+    },
+    {
+      "epoch": 0.8844221105527639,
+      "grad_norm": 0.34371060132980347,
+      "learning_rate": 0.0002,
+      "loss": 1.7824,
+      "step": 1320
+    },
+    {
+      "epoch": 0.8911222780569514,
+      "grad_norm": 0.3029090166091919,
+      "learning_rate": 0.0002,
+      "loss": 1.783,
+      "step": 1330
+    },
+    {
+      "epoch": 0.897822445561139,
+      "grad_norm": 0.34700682759284973,
+      "learning_rate": 0.0002,
+      "loss": 1.8017,
+      "step": 1340
+    },
+    {
+      "epoch": 0.9045226130653267,
+      "grad_norm": 0.35574328899383545,
+      "learning_rate": 0.0002,
+      "loss": 1.7998,
+      "step": 1350
+    },
+    {
+      "epoch": 0.9112227805695142,
+      "grad_norm": 0.30928221344947815,
+      "learning_rate": 0.0002,
+      "loss": 1.7339,
+      "step": 1360
+    },
+    {
+      "epoch": 0.9179229480737019,
+      "grad_norm": 0.30652928352355957,
+      "learning_rate": 0.0002,
+      "loss": 1.7479,
+      "step": 1370
+    },
+    {
+      "epoch": 0.9246231155778895,
+      "grad_norm": 0.3838157653808594,
+      "learning_rate": 0.0002,
+      "loss": 1.7491,
+      "step": 1380
+    },
+    {
+      "epoch": 0.931323283082077,
+      "grad_norm": 0.31655240058898926,
+      "learning_rate": 0.0002,
+      "loss": 1.7977,
+      "step": 1390
+    },
+    {
+      "epoch": 0.9380234505862647,
+      "grad_norm": 0.41737303137779236,
+      "learning_rate": 0.0002,
+      "loss": 1.8175,
+      "step": 1400
+    },
+    {
+      "epoch": 0.9447236180904522,
+      "grad_norm": 0.3227267861366272,
+      "learning_rate": 0.0002,
+      "loss": 1.6811,
+      "step": 1410
+    },
+    {
+      "epoch": 0.9514237855946399,
+      "grad_norm": 0.3729925751686096,
+      "learning_rate": 0.0002,
+      "loss": 1.7343,
+      "step": 1420
+    },
+    {
+      "epoch": 0.9581239530988275,
+      "grad_norm": 0.30779409408569336,
+      "learning_rate": 0.0002,
+      "loss": 1.8221,
+      "step": 1430
+    },
+    {
+      "epoch": 0.964824120603015,
+      "grad_norm": 0.334379643201828,
+      "learning_rate": 0.0002,
+      "loss": 1.7972,
+      "step": 1440
+    },
+    {
+      "epoch": 0.9715242881072027,
+      "grad_norm": 0.3568236231803894,
+      "learning_rate": 0.0002,
+      "loss": 1.7141,
+      "step": 1450
+    },
+    {
+      "epoch": 0.9782244556113903,
+      "grad_norm": 0.33310577273368835,
+      "learning_rate": 0.0002,
+      "loss": 1.7541,
+      "step": 1460
+    },
+    {
+      "epoch": 0.9849246231155779,
+      "grad_norm": 0.2972261905670166,
+      "learning_rate": 0.0002,
+      "loss": 1.8511,
+      "step": 1470
+    },
+    {
+      "epoch": 0.9916247906197655,
+      "grad_norm": 0.3322717845439911,
+      "learning_rate": 0.0002,
+      "loss": 1.7654,
+      "step": 1480
+    },
+    {
+      "epoch": 0.998324958123953,
+      "grad_norm": 0.3276330828666687,
+      "learning_rate": 0.0002,
+      "loss": 1.8033,
+      "step": 1490
+    },
+    {
+      "epoch": 0.9996649916247906,
+      "eval_loss": 1.8036354780197144,
+      "eval_runtime": 37.8949,
+      "eval_samples_per_second": 13.59,
+      "eval_steps_per_second": 1.715,
+      "step": 1492
+    },
+    {
+      "epoch": 1.0050251256281406,
+      "grad_norm": 0.29252371191978455,
+      "learning_rate": 0.0002,
+      "loss": 1.7138,
+      "step": 1500
+    },
+    {
+      "epoch": 1.0117252931323284,
+      "grad_norm": 0.31607162952423096,
+      "learning_rate": 0.0002,
+      "loss": 1.8198,
+      "step": 1510
+    },
+    {
+      "epoch": 1.018425460636516,
+      "grad_norm": 0.32294467091560364,
+      "learning_rate": 0.0002,
+      "loss": 1.6779,
+      "step": 1520
+    },
+    {
+      "epoch": 1.0251256281407035,
+      "grad_norm": 0.3868017792701721,
+      "learning_rate": 0.0002,
+      "loss": 1.7919,
+      "step": 1530
+    },
+    {
+      "epoch": 1.031825795644891,
+      "grad_norm": 0.3178282082080841,
+      "learning_rate": 0.0002,
+      "loss": 1.7954,
+      "step": 1540
+    },
+    {
+      "epoch": 1.0385259631490786,
+      "grad_norm": 0.3706750273704529,
+      "learning_rate": 0.0002,
+      "loss": 1.7136,
+      "step": 1550
+    },
+    {
+      "epoch": 1.0452261306532664,
+      "grad_norm": 0.33930912613868713,
+      "learning_rate": 0.0002,
+      "loss": 1.7382,
+      "step": 1560
+    },
+    {
+      "epoch": 1.051926298157454,
+      "grad_norm": 0.33970504999160767,
+      "learning_rate": 0.0002,
+      "loss": 1.7602,
+      "step": 1570
+    },
+    {
+      "epoch": 1.0586264656616415,
+      "grad_norm": 0.42553383111953735,
+      "learning_rate": 0.0002,
+      "loss": 1.6573,
+      "step": 1580
+    },
+    {
+      "epoch": 1.065326633165829,
+      "grad_norm": 0.3772421181201935,
+      "learning_rate": 0.0002,
+      "loss": 1.645,
+      "step": 1590
+    },
+    {
+      "epoch": 1.0720268006700167,
+      "grad_norm": 0.34212902188301086,
+      "learning_rate": 0.0002,
+      "loss": 1.7362,
+      "step": 1600
+    },
+    {
+      "epoch": 1.0787269681742044,
+      "grad_norm": 0.3798283338546753,
+      "learning_rate": 0.0002,
+      "loss": 1.7057,
+      "step": 1610
+    },
+    {
+      "epoch": 1.085427135678392,
+      "grad_norm": 0.36909598112106323,
+      "learning_rate": 0.0002,
+      "loss": 1.7468,
+      "step": 1620
+    },
+    {
+      "epoch": 1.0921273031825796,
+      "grad_norm": 0.3344230651855469,
+      "learning_rate": 0.0002,
+      "loss": 1.7807,
+      "step": 1630
+    },
+    {
+      "epoch": 1.0988274706867671,
+      "grad_norm": 0.3862569332122803,
+      "learning_rate": 0.0002,
+      "loss": 1.7111,
+      "step": 1640
+    },
+    {
+      "epoch": 1.1055276381909547,
+      "grad_norm": 0.31188511848449707,
+      "learning_rate": 0.0002,
+      "loss": 1.7163,
+      "step": 1650
+    },
+    {
+      "epoch": 1.1122278056951425,
+      "grad_norm": 0.3563670814037323,
+      "learning_rate": 0.0002,
+      "loss": 1.7263,
+      "step": 1660
+    },
+    {
+      "epoch": 1.11892797319933,
+      "grad_norm": 0.35052165389060974,
+      "learning_rate": 0.0002,
+      "loss": 1.7718,
+      "step": 1670
+    },
+    {
+      "epoch": 1.1256281407035176,
+      "grad_norm": 0.3285699188709259,
+      "learning_rate": 0.0002,
+      "loss": 1.7601,
+      "step": 1680
+    },
+    {
+      "epoch": 1.1323283082077051,
+      "grad_norm": 0.3639393746852875,
+      "learning_rate": 0.0002,
+      "loss": 1.6877,
+      "step": 1690
+    },
+    {
+      "epoch": 1.1390284757118927,
+      "grad_norm": 0.3842753767967224,
+      "learning_rate": 0.0002,
+      "loss": 1.7719,
+      "step": 1700
+    },
+    {
+      "epoch": 1.1457286432160805,
+      "grad_norm": 0.3624933063983917,
+      "learning_rate": 0.0002,
+      "loss": 1.7002,
+      "step": 1710
+    },
+    {
+      "epoch": 1.152428810720268,
+      "grad_norm": 0.3641220033168793,
+      "learning_rate": 0.0002,
+      "loss": 1.7243,
+      "step": 1720
+    },
+    {
+      "epoch": 1.1591289782244556,
+      "grad_norm": 0.32765355706214905,
+      "learning_rate": 0.0002,
+      "loss": 1.752,
+      "step": 1730
+    },
+    {
+      "epoch": 1.1658291457286432,
+      "grad_norm": 0.34974896907806396,
+      "learning_rate": 0.0002,
+      "loss": 1.6556,
+      "step": 1740
+    },
+    {
+      "epoch": 1.1725293132328307,
+      "grad_norm": 0.3910926580429077,
+      "learning_rate": 0.0002,
+      "loss": 1.7273,
+      "step": 1750
+    },
+    {
+      "epoch": 1.1792294807370185,
+      "grad_norm": 0.3564300537109375,
+      "learning_rate": 0.0002,
+      "loss": 1.7204,
+      "step": 1760
+    },
+    {
+      "epoch": 1.185929648241206,
+      "grad_norm": 0.34822574257850647,
+      "learning_rate": 0.0002,
+      "loss": 1.746,
+      "step": 1770
+    },
+    {
+      "epoch": 1.1926298157453936,
+      "grad_norm": 0.36185044050216675,
+      "learning_rate": 0.0002,
+      "loss": 1.7256,
+      "step": 1780
+    },
+    {
+      "epoch": 1.1993299832495812,
+      "grad_norm": 0.34866711497306824,
+      "learning_rate": 0.0002,
+      "loss": 1.6431,
+      "step": 1790
+    },
+    {
+      "epoch": 1.2060301507537687,
+      "grad_norm": 0.4017769992351532,
+      "learning_rate": 0.0002,
+      "loss": 1.8084,
+      "step": 1800
+    },
+    {
+      "epoch": 1.2127303182579565,
+      "grad_norm": 0.32930681109428406,
+      "learning_rate": 0.0002,
+      "loss": 1.6985,
+      "step": 1810
+    },
+    {
+      "epoch": 1.219430485762144,
+      "grad_norm": 0.35951921343803406,
+      "learning_rate": 0.0002,
+      "loss": 1.7606,
+      "step": 1820
+    },
+    {
+      "epoch": 1.2261306532663316,
+      "grad_norm": 0.37366992235183716,
+      "learning_rate": 0.0002,
+      "loss": 1.6933,
+      "step": 1830
+    },
+    {
+      "epoch": 1.2328308207705192,
+      "grad_norm": 0.3565689027309418,
+      "learning_rate": 0.0002,
+      "loss": 1.6737,
+      "step": 1840
+    },
+    {
+      "epoch": 1.2395309882747068,
+      "grad_norm": 0.3692343533039093,
+      "learning_rate": 0.0002,
+      "loss": 1.8013,
+      "step": 1850
+    },
+    {
+      "epoch": 1.2462311557788945,
+      "grad_norm": 0.38426971435546875,
+      "learning_rate": 0.0002,
+      "loss": 1.736,
+      "step": 1860
+    },
+    {
+      "epoch": 1.252931323283082,
+      "grad_norm": 0.33559855818748474,
+      "learning_rate": 0.0002,
+      "loss": 1.7031,
+      "step": 1870
+    },
+    {
+      "epoch": 1.2596314907872697,
+      "grad_norm": 0.34181106090545654,
+      "learning_rate": 0.0002,
+      "loss": 1.7033,
+      "step": 1880
+    },
+    {
+      "epoch": 1.2663316582914572,
+      "grad_norm": 0.3916318416595459,
+      "learning_rate": 0.0002,
+      "loss": 1.7707,
+      "step": 1890
+    },
+    {
+      "epoch": 1.2730318257956448,
+      "grad_norm": 0.3887825012207031,
+      "learning_rate": 0.0002,
+      "loss": 1.6686,
+      "step": 1900
+    },
+    {
+      "epoch": 1.2797319932998326,
+      "grad_norm": 0.33583927154541016,
+      "learning_rate": 0.0002,
+      "loss": 1.7062,
+      "step": 1910
+    },
+    {
+      "epoch": 1.2864321608040201,
+      "grad_norm": 0.37639349699020386,
+      "learning_rate": 0.0002,
+      "loss": 1.717,
+      "step": 1920
+    },
+    {
+      "epoch": 1.2931323283082077,
+      "grad_norm": 0.38059428334236145,
+      "learning_rate": 0.0002,
+      "loss": 1.777,
+      "step": 1930
+    },
+    {
+      "epoch": 1.2998324958123952,
+      "grad_norm": 0.37253183126449585,
+      "learning_rate": 0.0002,
+      "loss": 1.6126,
+      "step": 1940
+    },
+    {
+      "epoch": 1.3065326633165828,
+      "grad_norm": 0.37371566891670227,
+      "learning_rate": 0.0002,
+      "loss": 1.6758,
+      "step": 1950
+    },
+    {
+      "epoch": 1.3132328308207706,
+      "grad_norm": 0.4080910086631775,
+      "learning_rate": 0.0002,
+      "loss": 1.6788,
+      "step": 1960
+    },
+    {
+      "epoch": 1.3199329983249581,
+      "grad_norm": 0.3174354135990143,
+      "learning_rate": 0.0002,
+      "loss": 1.6518,
+      "step": 1970
+    },
+    {
+      "epoch": 1.3266331658291457,
+      "grad_norm": 0.4518888294696808,
+      "learning_rate": 0.0002,
+      "loss": 1.7925,
+      "step": 1980
+    },
+    {
+      "epoch": 1.3333333333333333,
+      "grad_norm": 0.3627921938896179,
+      "learning_rate": 0.0002,
+      "loss": 1.7085,
+      "step": 1990
+    },
+    {
+      "epoch": 1.3400335008375208,
+      "grad_norm": 0.3655930161476135,
+      "learning_rate": 0.0002,
+      "loss": 1.7676,
+      "step": 2000
+    },
+    {
+      "epoch": 1.3467336683417086,
+      "grad_norm": 0.3509993255138397,
+      "learning_rate": 0.0002,
+      "loss": 1.7016,
+      "step": 2010
+    },
+    {
+      "epoch": 1.3534338358458962,
+      "grad_norm": 0.4281129240989685,
+      "learning_rate": 0.0002,
+      "loss": 1.7359,
+      "step": 2020
+    },
+    {
+      "epoch": 1.3601340033500837,
+      "grad_norm": 0.3821414113044739,
+      "learning_rate": 0.0002,
+      "loss": 1.6884,
+      "step": 2030
+    },
+    {
+      "epoch": 1.3668341708542713,
+      "grad_norm": 0.3907586336135864,
+      "learning_rate": 0.0002,
+      "loss": 1.7075,
+      "step": 2040
+    },
+    {
+      "epoch": 1.3735343383584588,
+      "grad_norm": 0.37792932987213135,
+      "learning_rate": 0.0002,
+      "loss": 1.7424,
+      "step": 2050
+    },
+    {
+      "epoch": 1.3802345058626466,
+      "grad_norm": 0.3693985641002655,
+      "learning_rate": 0.0002,
+      "loss": 1.7305,
+      "step": 2060
+    },
+    {
+      "epoch": 1.3869346733668342,
+      "grad_norm": 0.32275936007499695,
+      "learning_rate": 0.0002,
+      "loss": 1.7434,
+      "step": 2070
+    },
+    {
+      "epoch": 1.3936348408710217,
+      "grad_norm": 0.3789440095424652,
+      "learning_rate": 0.0002,
+      "loss": 1.6677,
+      "step": 2080
+    },
+    {
+      "epoch": 1.4003350083752093,
+      "grad_norm": 0.3638380467891693,
+      "learning_rate": 0.0002,
+      "loss": 1.6825,
+      "step": 2090
+    },
+    {
+      "epoch": 1.4070351758793969,
+      "grad_norm": 0.3495481610298157,
+      "learning_rate": 0.0002,
+      "loss": 1.6542,
+      "step": 2100
+    },
+    {
+      "epoch": 1.4137353433835846,
+      "grad_norm": 0.37920597195625305,
+      "learning_rate": 0.0002,
+      "loss": 1.7225,
+      "step": 2110
+    },
+    {
+      "epoch": 1.4204355108877722,
+      "grad_norm": 0.37218064069747925,
+      "learning_rate": 0.0002,
+      "loss": 1.7329,
+      "step": 2120
+    },
+    {
+      "epoch": 1.4271356783919598,
+      "grad_norm": 0.38074082136154175,
+      "learning_rate": 0.0002,
+      "loss": 1.799,
+      "step": 2130
+    },
+    {
+      "epoch": 1.4338358458961473,
+      "grad_norm": 0.3455527126789093,
+      "learning_rate": 0.0002,
+      "loss": 1.7403,
+      "step": 2140
+    },
+    {
+      "epoch": 1.4405360134003349,
+      "grad_norm": 0.3712003529071808,
+      "learning_rate": 0.0002,
+      "loss": 1.776,
+      "step": 2150
+    },
+    {
+      "epoch": 1.4472361809045227,
+      "grad_norm": 0.3786754906177521,
+      "learning_rate": 0.0002,
+      "loss": 1.7619,
+      "step": 2160
+    },
+    {
+      "epoch": 1.4539363484087102,
+      "grad_norm": 0.3879223167896271,
+      "learning_rate": 0.0002,
+      "loss": 1.68,
+      "step": 2170
+    },
+    {
+      "epoch": 1.4606365159128978,
+      "grad_norm": 0.38738805055618286,
+      "learning_rate": 0.0002,
+      "loss": 1.7,
+      "step": 2180
+    },
+    {
+      "epoch": 1.4673366834170856,
+      "grad_norm": 0.39768800139427185,
+      "learning_rate": 0.0002,
+      "loss": 1.7581,
+      "step": 2190
+    },
+    {
+      "epoch": 1.474036850921273,
+      "grad_norm": 0.4172441065311432,
+      "learning_rate": 0.0002,
+      "loss": 1.7671,
+      "step": 2200
+    },
+    {
+      "epoch": 1.4807370184254607,
+      "grad_norm": 0.4043174982070923,
+      "learning_rate": 0.0002,
+      "loss": 1.6736,
+      "step": 2210
+    },
+    {
+      "epoch": 1.4874371859296482,
+      "grad_norm": 0.3750883936882019,
+      "learning_rate": 0.0002,
+      "loss": 1.7444,
+      "step": 2220
+    },
+    {
+      "epoch": 1.4941373534338358,
+      "grad_norm": 0.3552253246307373,
+      "learning_rate": 0.0002,
+      "loss": 1.6861,
+      "step": 2230
+    },
+    {
+      "epoch": 1.5008375209380236,
+      "grad_norm": 0.34607139229774475,
+      "learning_rate": 0.0002,
+      "loss": 1.6471,
+      "step": 2240
+    },
+    {
+      "epoch": 1.507537688442211,
+      "grad_norm": 0.3406706750392914,
+      "learning_rate": 0.0002,
+      "loss": 1.6962,
+      "step": 2250
+    },
+    {
+      "epoch": 1.5142378559463987,
+      "grad_norm": 0.36654895544052124,
+      "learning_rate": 0.0002,
+      "loss": 1.7694,
+      "step": 2260
+    },
+    {
+      "epoch": 1.5209380234505863,
+      "grad_norm": 0.3914054334163666,
+      "learning_rate": 0.0002,
+      "loss": 1.6812,
+      "step": 2270
+    },
+    {
+      "epoch": 1.5276381909547738,
+      "grad_norm": 0.42012137174606323,
+      "learning_rate": 0.0002,
+      "loss": 1.6822,
+      "step": 2280
+    },
+    {
+      "epoch": 1.5343383584589616,
+      "grad_norm": 0.39563435316085815,
+      "learning_rate": 0.0002,
+      "loss": 1.697,
+      "step": 2290
+    },
+    {
+      "epoch": 1.541038525963149,
+      "grad_norm": 0.3508438766002655,
+      "learning_rate": 0.0002,
+      "loss": 1.7491,
+      "step": 2300
+    },
+    {
+      "epoch": 1.5477386934673367,
+      "grad_norm": 0.3785218596458435,
+      "learning_rate": 0.0002,
+      "loss": 1.7727,
+      "step": 2310
+    },
+    {
+      "epoch": 1.5544388609715243,
+      "grad_norm": 0.39377647638320923,
+      "learning_rate": 0.0002,
+      "loss": 1.6963,
+      "step": 2320
+    },
+    {
+      "epoch": 1.5611390284757118,
+      "grad_norm": 0.3391438126564026,
+      "learning_rate": 0.0002,
+      "loss": 1.7263,
+      "step": 2330
+    },
+    {
+      "epoch": 1.5678391959798996,
+      "grad_norm": 0.37944263219833374,
+      "learning_rate": 0.0002,
+      "loss": 1.7722,
+      "step": 2340
+    },
+    {
+      "epoch": 1.574539363484087,
+      "grad_norm": 0.3523491322994232,
+      "learning_rate": 0.0002,
+      "loss": 1.6371,
+      "step": 2350
+    },
+    {
+      "epoch": 1.5812395309882747,
+      "grad_norm": 0.3911575973033905,
+      "learning_rate": 0.0002,
+      "loss": 1.7583,
+      "step": 2360
+    },
+    {
+      "epoch": 1.5879396984924623,
+      "grad_norm": 0.33832186460494995,
+      "learning_rate": 0.0002,
+      "loss": 1.7117,
+      "step": 2370
+    },
+    {
+      "epoch": 1.5946398659966499,
+      "grad_norm": 0.3665979206562042,
+      "learning_rate": 0.0002,
+      "loss": 1.7701,
+      "step": 2380
+    },
+    {
+      "epoch": 1.6013400335008376,
+      "grad_norm": 0.3871748149394989,
+      "learning_rate": 0.0002,
+      "loss": 1.779,
+      "step": 2390
+    },
+    {
+      "epoch": 1.608040201005025,
+      "grad_norm": 0.3586967885494232,
+      "learning_rate": 0.0002,
+      "loss": 1.7109,
+      "step": 2400
+    },
+    {
+      "epoch": 1.6147403685092128,
+      "grad_norm": 0.3563673198223114,
+      "learning_rate": 0.0002,
+      "loss": 1.7096,
+      "step": 2410
+    },
+    {
+      "epoch": 1.6214405360134003,
+      "grad_norm": 0.37588971853256226,
+      "learning_rate": 0.0002,
+      "loss": 1.745,
+      "step": 2420
+    },
+    {
+      "epoch": 1.6281407035175879,
+      "grad_norm": 0.352556437253952,
+      "learning_rate": 0.0002,
+      "loss": 1.7086,
+      "step": 2430
+    },
+    {
+      "epoch": 1.6348408710217757,
+      "grad_norm": 0.3716259300708771,
+      "learning_rate": 0.0002,
+      "loss": 1.6547,
+      "step": 2440
+    },
+    {
+      "epoch": 1.641541038525963,
+      "grad_norm": 0.372001975774765,
+      "learning_rate": 0.0002,
+      "loss": 1.7033,
+      "step": 2450
+    },
+    {
+      "epoch": 1.6482412060301508,
+      "grad_norm": 0.3430042862892151,
+      "learning_rate": 0.0002,
+      "loss": 1.6584,
+      "step": 2460
+    },
+    {
+      "epoch": 1.6549413735343383,
+      "grad_norm": 0.3741483688354492,
+      "learning_rate": 0.0002,
+      "loss": 1.7217,
+      "step": 2470
+    },
+    {
+      "epoch": 1.661641541038526,
+      "grad_norm": 0.3610571324825287,
+      "learning_rate": 0.0002,
+      "loss": 1.7701,
+      "step": 2480
+    },
+    {
+      "epoch": 1.6683417085427137,
+      "grad_norm": 0.4204719066619873,
+      "learning_rate": 0.0002,
+      "loss": 1.7057,
+      "step": 2490
+    },
+    {
+      "epoch": 1.675041876046901,
+      "grad_norm": 0.3938186466693878,
+      "learning_rate": 0.0002,
+      "loss": 1.7954,
+      "step": 2500
+    },
+    {
+      "epoch": 1.6817420435510888,
+      "grad_norm": 0.3421435058116913,
+      "learning_rate": 0.0002,
+      "loss": 1.6633,
+      "step": 2510
+    },
+    {
+      "epoch": 1.6884422110552764,
+      "grad_norm": 0.42441412806510925,
+      "learning_rate": 0.0002,
+      "loss": 1.7996,
+      "step": 2520
+    },
+    {
+      "epoch": 1.695142378559464,
+      "grad_norm": 0.38071519136428833,
+      "learning_rate": 0.0002,
+      "loss": 1.7142,
+      "step": 2530
+    },
+    {
+      "epoch": 1.7018425460636517,
+      "grad_norm": 0.34078919887542725,
+      "learning_rate": 0.0002,
+      "loss": 1.7232,
+      "step": 2540
+    },
+    {
+      "epoch": 1.708542713567839,
+      "grad_norm": 0.412844181060791,
+      "learning_rate": 0.0002,
+      "loss": 1.7126,
+      "step": 2550
+    },
+    {
+      "epoch": 1.7152428810720268,
+      "grad_norm": 0.3753604292869568,
+      "learning_rate": 0.0002,
+      "loss": 1.7149,
+      "step": 2560
+    },
+    {
+      "epoch": 1.7219430485762144,
+      "grad_norm": 0.41588476300239563,
+      "learning_rate": 0.0002,
+      "loss": 1.7011,
+      "step": 2570
+    },
+    {
+      "epoch": 1.728643216080402,
+      "grad_norm": 0.35504111647605896,
+      "learning_rate": 0.0002,
+      "loss": 1.6427,
+      "step": 2580
+    },
+    {
+      "epoch": 1.7353433835845897,
+      "grad_norm": 0.36909720301628113,
+      "learning_rate": 0.0002,
+      "loss": 1.7296,
+      "step": 2590
+    },
+    {
+      "epoch": 1.742043551088777,
+      "grad_norm": 0.4149979054927826,
+      "learning_rate": 0.0002,
+      "loss": 1.7022,
+      "step": 2600
+    },
+    {
+      "epoch": 1.7487437185929648,
+      "grad_norm": 0.38859328627586365,
+      "learning_rate": 0.0002,
+      "loss": 1.77,
+      "step": 2610
+    },
+    {
+      "epoch": 1.7554438860971524,
+      "grad_norm": 0.36738792061805725,
+      "learning_rate": 0.0002,
+      "loss": 1.7036,
+      "step": 2620
+    },
+    {
+      "epoch": 1.76214405360134,
+      "grad_norm": 0.3968178927898407,
+      "learning_rate": 0.0002,
+      "loss": 1.764,
+      "step": 2630
+    },
+    {
+      "epoch": 1.7688442211055277,
+      "grad_norm": 0.3972901999950409,
+      "learning_rate": 0.0002,
+      "loss": 1.7687,
+      "step": 2640
+    },
+    {
+      "epoch": 1.775544388609715,
+      "grad_norm": 0.3949959874153137,
+      "learning_rate": 0.0002,
+      "loss": 1.6774,
+      "step": 2650
+    },
+    {
+      "epoch": 1.7822445561139029,
+      "grad_norm": 0.44074657559394836,
+      "learning_rate": 0.0002,
+      "loss": 1.7247,
+      "step": 2660
+    },
+    {
+      "epoch": 1.7889447236180904,
+      "grad_norm": 0.39743664860725403,
+      "learning_rate": 0.0002,
+      "loss": 1.7188,
+      "step": 2670
+    },
+    {
+      "epoch": 1.795644891122278,
+      "grad_norm": 0.3950406610965729,
+      "learning_rate": 0.0002,
+      "loss": 1.7258,
+      "step": 2680
+    },
+    {
+      "epoch": 1.8023450586264658,
+      "grad_norm": 0.3568263649940491,
+      "learning_rate": 0.0002,
+      "loss": 1.6906,
+      "step": 2690
+    },
+    {
+      "epoch": 1.809045226130653,
+      "grad_norm": 0.3819476366043091,
+      "learning_rate": 0.0002,
+      "loss": 1.6735,
+      "step": 2700
+    },
+    {
+      "epoch": 1.8157453936348409,
+      "grad_norm": 0.3480634391307831,
+      "learning_rate": 0.0002,
+      "loss": 1.7198,
+      "step": 2710
+    },
+    {
+      "epoch": 1.8224455611390284,
+      "grad_norm": 0.3875853419303894,
+      "learning_rate": 0.0002,
+      "loss": 1.7042,
+      "step": 2720
+    },
+    {
+      "epoch": 1.829145728643216,
+      "grad_norm": 0.3441337049007416,
+      "learning_rate": 0.0002,
+      "loss": 1.6988,
+      "step": 2730
+    },
+    {
+      "epoch": 1.8358458961474038,
+      "grad_norm": 0.35692882537841797,
+      "learning_rate": 0.0002,
+      "loss": 1.7647,
+      "step": 2740
+    },
+    {
+      "epoch": 1.8425460636515911,
+      "grad_norm": 0.36959215998649597,
+      "learning_rate": 0.0002,
+      "loss": 1.7033,
+      "step": 2750
+    },
+    {
+      "epoch": 1.849246231155779,
+      "grad_norm": 0.3893393278121948,
+      "learning_rate": 0.0002,
+      "loss": 1.7657,
+      "step": 2760
+    },
+    {
+      "epoch": 1.8559463986599665,
+      "grad_norm": 0.37817293405532837,
+      "learning_rate": 0.0002,
+      "loss": 1.7068,
+      "step": 2770
+    },
+    {
+      "epoch": 1.862646566164154,
+      "grad_norm": 0.36071285605430603,
+      "learning_rate": 0.0002,
+      "loss": 1.761,
+      "step": 2780
+    },
+    {
+      "epoch": 1.8693467336683418,
+      "grad_norm": 0.3758420944213867,
+      "learning_rate": 0.0002,
+      "loss": 1.7623,
+      "step": 2790
+    },
+    {
+      "epoch": 1.8760469011725294,
+      "grad_norm": 0.3889938294887543,
+      "learning_rate": 0.0002,
+      "loss": 1.6743,
+      "step": 2800
+    },
+    {
+      "epoch": 1.882747068676717,
+      "grad_norm": 0.34361857175827026,
+      "learning_rate": 0.0002,
+      "loss": 1.6151,
+      "step": 2810
+    },
+    {
+      "epoch": 1.8894472361809045,
+      "grad_norm": 0.39283323287963867,
+      "learning_rate": 0.0002,
+      "loss": 1.6038,
+      "step": 2820
+    },
+    {
+      "epoch": 1.896147403685092,
+      "grad_norm": 0.3919452726840973,
+      "learning_rate": 0.0002,
+      "loss": 1.7555,
+      "step": 2830
+    },
+    {
+      "epoch": 1.9028475711892798,
+      "grad_norm": 0.38215070962905884,
+      "learning_rate": 0.0002,
+      "loss": 1.673,
+      "step": 2840
+    },
+    {
+      "epoch": 1.9095477386934674,
+      "grad_norm": 0.4235064387321472,
+      "learning_rate": 0.0002,
+      "loss": 1.7044,
+      "step": 2850
+    },
+    {
+      "epoch": 1.916247906197655,
+      "grad_norm": 0.35694634914398193,
+      "learning_rate": 0.0002,
+      "loss": 1.7123,
+      "step": 2860
+    },
+    {
+      "epoch": 1.9229480737018425,
+      "grad_norm": 0.383492112159729,
+      "learning_rate": 0.0002,
+      "loss": 1.8128,
+      "step": 2870
+    },
+    {
+      "epoch": 1.92964824120603,
+      "grad_norm": 0.5945147275924683,
+      "learning_rate": 0.0002,
+      "loss": 1.7581,
+      "step": 2880
+    },
+    {
+      "epoch": 1.9363484087102178,
+      "grad_norm": 0.3367522358894348,
+      "learning_rate": 0.0002,
+      "loss": 1.7421,
+      "step": 2890
+    },
+    {
+      "epoch": 1.9430485762144054,
+      "grad_norm": 0.35300394892692566,
+      "learning_rate": 0.0002,
+      "loss": 1.6561,
+      "step": 2900
+    },
+    {
+      "epoch": 1.949748743718593,
+      "grad_norm": 0.38084495067596436,
+      "learning_rate": 0.0002,
+      "loss": 1.7033,
+      "step": 2910
+    },
+    {
+      "epoch": 1.9564489112227805,
+      "grad_norm": 0.37559160590171814,
+      "learning_rate": 0.0002,
+      "loss": 1.7132,
+      "step": 2920
+    },
+    {
+      "epoch": 1.963149078726968,
+      "grad_norm": 0.3661738336086273,
+      "learning_rate": 0.0002,
+      "loss": 1.6759,
+      "step": 2930
+    },
+    {
+      "epoch": 1.9698492462311559,
+      "grad_norm": 0.4073849320411682,
+      "learning_rate": 0.0002,
+      "loss": 1.7643,
+      "step": 2940
+    },
+    {
+      "epoch": 1.9765494137353434,
+      "grad_norm": 0.3723304271697998,
+      "learning_rate": 0.0002,
+      "loss": 1.6806,
+      "step": 2950
+    },
+    {
+      "epoch": 1.983249581239531,
+      "grad_norm": 0.3991098999977112,
+      "learning_rate": 0.0002,
+      "loss": 1.7611,
+      "step": 2960
+    },
+    {
+      "epoch": 1.9899497487437185,
+      "grad_norm": 0.3947085440158844,
+      "learning_rate": 0.0002,
+      "loss": 1.7263,
+      "step": 2970
+    },
+    {
+      "epoch": 1.996649916247906,
+      "grad_norm": 0.3786258399486542,
+      "learning_rate": 0.0002,
+      "loss": 1.7217,
+      "step": 2980
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 1.8028968572616577,
+      "eval_runtime": 37.8985,
+      "eval_samples_per_second": 13.589,
+      "eval_steps_per_second": 1.715,
+      "step": 2985
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 11936,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 8,
+  "save_steps": 200,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 1.3813911631429632e+17,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-2985/training_args.bin b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-2985/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..064f299b0f3d2a28f5b1f5c68ef32caab3e2dd49
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-2985/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aa7b586fc440d1e22156593e2dd4267d2bdcb8920a02fdf352ea29a9bec3dd94
+size 5560
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-4477/README.md b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-4477/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..835e31ab8469ee39ddc8b2b6b2143a8c66dad510
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-4477/README.md
@@ -0,0 +1,202 @@
+---
+base_model: meta-llama/Meta-Llama-3-8B-Instruct
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.13.1
\ No newline at end of file
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-4477/adapter_config.json b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-4477/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..406c5a08dc4a2a33b52c62a482f98c217c417215
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-4477/adapter_config.json
@@ -0,0 +1,29 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "meta-llama/Meta-Llama-3-8B-Instruct",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-4477/adapter_model.safetensors b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-4477/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..9db6a6e39dad5a1aeda5f62bb72072dfee810f36
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-4477/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1a6250b488c4c4dbef4033e080f9cdecb7ac533c6ad1de44a10fbfdcb20da6bf
+size 109069176
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-4477/optimizer.pt b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-4477/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..e45c82df3bf4ccf8bcd0253221f3b6ecc030d86f
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-4477/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c993acdc25f216c826a96c6519846f644d89b0060cca1151dbb3e0f3226eaea5
+size 55532666
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-4477/rng_state.pth b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-4477/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..0fefd235b9fd5d25f75ddf110535ddec5c539fe4
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-4477/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8fbc45d51dbc1471231cadd0032584621fb86eecfcde0d03ab47e4dcd3c68af5
+size 14244
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-4477/scheduler.pt b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-4477/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..feb46514db6c2bb40a9629c6d795c96be175d35b
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-4477/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d2c1d004afcbbbebcfb68a9727530fff405d175a6ca8b3f7efcd34b9885fe831
+size 1064
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-4477/special_tokens_map.json b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-4477/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..b43be96621d147110fb8a18b5776ec6e38516127
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-4477/special_tokens_map.json
@@ -0,0 +1,17 @@
+{
+  "bos_token": {
+    "content": "<|begin_of_text|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<|eot_id|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<|eot_id|>"
+}
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-4477/tokenizer.json b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-4477/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..86a33946b0c77216d2cce91bb28c8fada4a5e80b
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-4477/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3c5cf44023714fb39b05e71e425f8d7b92805ff73f7988b083b8c87f0bf87393
+size 17209961
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-4477/tokenizer_config.json b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-4477/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..061e40d9db3253624f86e8e364c15ef546527c9d
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-4477/tokenizer_config.json
@@ -0,0 +1,2063 @@
+{
+  "added_tokens_decoder": {
+    "128000": {
+      "content": "<|begin_of_text|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128001": {
+      "content": "<|end_of_text|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128002": {
+      "content": "<|reserved_special_token_0|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128003": {
+      "content": "<|reserved_special_token_1|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128004": {
+      "content": "<|reserved_special_token_2|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128005": {
+      "content": "<|reserved_special_token_3|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128006": {
+      "content": "<|start_header_id|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128007": {
+      "content": "<|end_header_id|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128008": {
+      "content": "<|reserved_special_token_4|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128009": {
+      "content": "<|eot_id|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128010": {
+      "content": "<|reserved_special_token_5|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128011": {
+      "content": "<|reserved_special_token_6|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128012": {
+      "content": "<|reserved_special_token_7|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128013": {
+      "content": "<|reserved_special_token_8|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128014": {
+      "content": "<|reserved_special_token_9|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128015": {
+      "content": "<|reserved_special_token_10|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128016": {
+      "content": "<|reserved_special_token_11|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128017": {
+      "content": "<|reserved_special_token_12|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128018": {
+      "content": "<|reserved_special_token_13|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128019": {
+      "content": "<|reserved_special_token_14|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128020": {
+      "content": "<|reserved_special_token_15|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128021": {
+      "content": "<|reserved_special_token_16|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128022": {
+      "content": "<|reserved_special_token_17|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128023": {
+      "content": "<|reserved_special_token_18|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128024": {
+      "content": "<|reserved_special_token_19|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128025": {
+      "content": "<|reserved_special_token_20|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128026": {
+      "content": "<|reserved_special_token_21|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128027": {
+      "content": "<|reserved_special_token_22|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128028": {
+      "content": "<|reserved_special_token_23|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128029": {
+      "content": "<|reserved_special_token_24|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128030": {
+      "content": "<|reserved_special_token_25|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128031": {
+      "content": "<|reserved_special_token_26|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128032": {
+      "content": "<|reserved_special_token_27|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128033": {
+      "content": "<|reserved_special_token_28|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128034": {
+      "content": "<|reserved_special_token_29|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128035": {
+      "content": "<|reserved_special_token_30|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128036": {
+      "content": "<|reserved_special_token_31|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128037": {
+      "content": "<|reserved_special_token_32|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128038": {
+      "content": "<|reserved_special_token_33|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128039": {
+      "content": "<|reserved_special_token_34|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128040": {
+      "content": "<|reserved_special_token_35|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128041": {
+      "content": "<|reserved_special_token_36|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128042": {
+      "content": "<|reserved_special_token_37|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128043": {
+      "content": "<|reserved_special_token_38|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128044": {
+      "content": "<|reserved_special_token_39|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128045": {
+      "content": "<|reserved_special_token_40|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128046": {
+      "content": "<|reserved_special_token_41|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128047": {
+      "content": "<|reserved_special_token_42|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128048": {
+      "content": "<|reserved_special_token_43|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128049": {
+      "content": "<|reserved_special_token_44|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128050": {
+      "content": "<|reserved_special_token_45|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128051": {
+      "content": "<|reserved_special_token_46|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128052": {
+      "content": "<|reserved_special_token_47|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128053": {
+      "content": "<|reserved_special_token_48|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128054": {
+      "content": "<|reserved_special_token_49|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128055": {
+      "content": "<|reserved_special_token_50|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128056": {
+      "content": "<|reserved_special_token_51|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128057": {
+      "content": "<|reserved_special_token_52|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128058": {
+      "content": "<|reserved_special_token_53|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128059": {
+      "content": "<|reserved_special_token_54|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128060": {
+      "content": "<|reserved_special_token_55|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128061": {
+      "content": "<|reserved_special_token_56|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128062": {
+      "content": "<|reserved_special_token_57|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128063": {
+      "content": "<|reserved_special_token_58|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128064": {
+      "content": "<|reserved_special_token_59|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128065": {
+      "content": "<|reserved_special_token_60|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128066": {
+      "content": "<|reserved_special_token_61|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128067": {
+      "content": "<|reserved_special_token_62|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128068": {
+      "content": "<|reserved_special_token_63|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128069": {
+      "content": "<|reserved_special_token_64|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128070": {
+      "content": "<|reserved_special_token_65|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128071": {
+      "content": "<|reserved_special_token_66|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128072": {
+      "content": "<|reserved_special_token_67|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128073": {
+      "content": "<|reserved_special_token_68|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128074": {
+      "content": "<|reserved_special_token_69|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128075": {
+      "content": "<|reserved_special_token_70|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128076": {
+      "content": "<|reserved_special_token_71|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128077": {
+      "content": "<|reserved_special_token_72|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128078": {
+      "content": "<|reserved_special_token_73|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128079": {
+      "content": "<|reserved_special_token_74|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128080": {
+      "content": "<|reserved_special_token_75|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128081": {
+      "content": "<|reserved_special_token_76|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128082": {
+      "content": "<|reserved_special_token_77|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128083": {
+      "content": "<|reserved_special_token_78|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128084": {
+      "content": "<|reserved_special_token_79|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128085": {
+      "content": "<|reserved_special_token_80|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128086": {
+      "content": "<|reserved_special_token_81|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128087": {
+      "content": "<|reserved_special_token_82|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128088": {
+      "content": "<|reserved_special_token_83|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128089": {
+      "content": "<|reserved_special_token_84|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128090": {
+      "content": "<|reserved_special_token_85|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128091": {
+      "content": "<|reserved_special_token_86|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128092": {
+      "content": "<|reserved_special_token_87|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128093": {
+      "content": "<|reserved_special_token_88|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128094": {
+      "content": "<|reserved_special_token_89|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128095": {
+      "content": "<|reserved_special_token_90|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128096": {
+      "content": "<|reserved_special_token_91|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128097": {
+      "content": "<|reserved_special_token_92|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128098": {
+      "content": "<|reserved_special_token_93|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128099": {
+      "content": "<|reserved_special_token_94|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128100": {
+      "content": "<|reserved_special_token_95|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128101": {
+      "content": "<|reserved_special_token_96|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128102": {
+      "content": "<|reserved_special_token_97|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128103": {
+      "content": "<|reserved_special_token_98|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128104": {
+      "content": "<|reserved_special_token_99|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128105": {
+      "content": "<|reserved_special_token_100|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128106": {
+      "content": "<|reserved_special_token_101|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128107": {
+      "content": "<|reserved_special_token_102|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128108": {
+      "content": "<|reserved_special_token_103|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128109": {
+      "content": "<|reserved_special_token_104|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128110": {
+      "content": "<|reserved_special_token_105|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128111": {
+      "content": "<|reserved_special_token_106|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128112": {
+      "content": "<|reserved_special_token_107|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128113": {
+      "content": "<|reserved_special_token_108|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128114": {
+      "content": "<|reserved_special_token_109|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128115": {
+      "content": "<|reserved_special_token_110|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128116": {
+      "content": "<|reserved_special_token_111|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128117": {
+      "content": "<|reserved_special_token_112|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128118": {
+      "content": "<|reserved_special_token_113|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128119": {
+      "content": "<|reserved_special_token_114|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128120": {
+      "content": "<|reserved_special_token_115|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128121": {
+      "content": "<|reserved_special_token_116|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128122": {
+      "content": "<|reserved_special_token_117|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128123": {
+      "content": "<|reserved_special_token_118|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128124": {
+      "content": "<|reserved_special_token_119|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128125": {
+      "content": "<|reserved_special_token_120|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128126": {
+      "content": "<|reserved_special_token_121|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128127": {
+      "content": "<|reserved_special_token_122|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128128": {
+      "content": "<|reserved_special_token_123|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128129": {
+      "content": "<|reserved_special_token_124|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128130": {
+      "content": "<|reserved_special_token_125|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128131": {
+      "content": "<|reserved_special_token_126|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128132": {
+      "content": "<|reserved_special_token_127|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128133": {
+      "content": "<|reserved_special_token_128|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128134": {
+      "content": "<|reserved_special_token_129|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128135": {
+      "content": "<|reserved_special_token_130|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128136": {
+      "content": "<|reserved_special_token_131|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128137": {
+      "content": "<|reserved_special_token_132|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128138": {
+      "content": "<|reserved_special_token_133|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128139": {
+      "content": "<|reserved_special_token_134|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128140": {
+      "content": "<|reserved_special_token_135|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128141": {
+      "content": "<|reserved_special_token_136|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128142": {
+      "content": "<|reserved_special_token_137|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128143": {
+      "content": "<|reserved_special_token_138|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128144": {
+      "content": "<|reserved_special_token_139|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128145": {
+      "content": "<|reserved_special_token_140|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128146": {
+      "content": "<|reserved_special_token_141|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128147": {
+      "content": "<|reserved_special_token_142|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128148": {
+      "content": "<|reserved_special_token_143|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128149": {
+      "content": "<|reserved_special_token_144|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128150": {
+      "content": "<|reserved_special_token_145|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128151": {
+      "content": "<|reserved_special_token_146|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128152": {
+      "content": "<|reserved_special_token_147|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128153": {
+      "content": "<|reserved_special_token_148|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128154": {
+      "content": "<|reserved_special_token_149|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128155": {
+      "content": "<|reserved_special_token_150|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128156": {
+      "content": "<|reserved_special_token_151|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128157": {
+      "content": "<|reserved_special_token_152|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128158": {
+      "content": "<|reserved_special_token_153|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128159": {
+      "content": "<|reserved_special_token_154|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128160": {
+      "content": "<|reserved_special_token_155|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128161": {
+      "content": "<|reserved_special_token_156|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128162": {
+      "content": "<|reserved_special_token_157|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128163": {
+      "content": "<|reserved_special_token_158|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128164": {
+      "content": "<|reserved_special_token_159|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128165": {
+      "content": "<|reserved_special_token_160|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128166": {
+      "content": "<|reserved_special_token_161|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128167": {
+      "content": "<|reserved_special_token_162|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128168": {
+      "content": "<|reserved_special_token_163|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128169": {
+      "content": "<|reserved_special_token_164|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128170": {
+      "content": "<|reserved_special_token_165|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128171": {
+      "content": "<|reserved_special_token_166|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128172": {
+      "content": "<|reserved_special_token_167|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128173": {
+      "content": "<|reserved_special_token_168|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128174": {
+      "content": "<|reserved_special_token_169|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128175": {
+      "content": "<|reserved_special_token_170|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128176": {
+      "content": "<|reserved_special_token_171|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128177": {
+      "content": "<|reserved_special_token_172|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128178": {
+      "content": "<|reserved_special_token_173|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128179": {
+      "content": "<|reserved_special_token_174|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128180": {
+      "content": "<|reserved_special_token_175|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128181": {
+      "content": "<|reserved_special_token_176|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128182": {
+      "content": "<|reserved_special_token_177|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128183": {
+      "content": "<|reserved_special_token_178|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128184": {
+      "content": "<|reserved_special_token_179|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128185": {
+      "content": "<|reserved_special_token_180|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128186": {
+      "content": "<|reserved_special_token_181|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128187": {
+      "content": "<|reserved_special_token_182|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128188": {
+      "content": "<|reserved_special_token_183|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128189": {
+      "content": "<|reserved_special_token_184|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128190": {
+      "content": "<|reserved_special_token_185|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128191": {
+      "content": "<|reserved_special_token_186|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128192": {
+      "content": "<|reserved_special_token_187|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128193": {
+      "content": "<|reserved_special_token_188|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128194": {
+      "content": "<|reserved_special_token_189|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128195": {
+      "content": "<|reserved_special_token_190|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128196": {
+      "content": "<|reserved_special_token_191|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128197": {
+      "content": "<|reserved_special_token_192|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128198": {
+      "content": "<|reserved_special_token_193|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128199": {
+      "content": "<|reserved_special_token_194|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128200": {
+      "content": "<|reserved_special_token_195|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128201": {
+      "content": "<|reserved_special_token_196|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128202": {
+      "content": "<|reserved_special_token_197|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128203": {
+      "content": "<|reserved_special_token_198|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128204": {
+      "content": "<|reserved_special_token_199|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128205": {
+      "content": "<|reserved_special_token_200|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128206": {
+      "content": "<|reserved_special_token_201|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128207": {
+      "content": "<|reserved_special_token_202|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128208": {
+      "content": "<|reserved_special_token_203|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128209": {
+      "content": "<|reserved_special_token_204|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128210": {
+      "content": "<|reserved_special_token_205|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128211": {
+      "content": "<|reserved_special_token_206|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128212": {
+      "content": "<|reserved_special_token_207|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128213": {
+      "content": "<|reserved_special_token_208|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128214": {
+      "content": "<|reserved_special_token_209|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128215": {
+      "content": "<|reserved_special_token_210|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128216": {
+      "content": "<|reserved_special_token_211|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128217": {
+      "content": "<|reserved_special_token_212|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128218": {
+      "content": "<|reserved_special_token_213|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128219": {
+      "content": "<|reserved_special_token_214|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128220": {
+      "content": "<|reserved_special_token_215|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128221": {
+      "content": "<|reserved_special_token_216|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128222": {
+      "content": "<|reserved_special_token_217|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128223": {
+      "content": "<|reserved_special_token_218|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128224": {
+      "content": "<|reserved_special_token_219|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128225": {
+      "content": "<|reserved_special_token_220|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128226": {
+      "content": "<|reserved_special_token_221|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128227": {
+      "content": "<|reserved_special_token_222|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128228": {
+      "content": "<|reserved_special_token_223|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128229": {
+      "content": "<|reserved_special_token_224|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128230": {
+      "content": "<|reserved_special_token_225|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128231": {
+      "content": "<|reserved_special_token_226|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128232": {
+      "content": "<|reserved_special_token_227|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128233": {
+      "content": "<|reserved_special_token_228|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128234": {
+      "content": "<|reserved_special_token_229|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128235": {
+      "content": "<|reserved_special_token_230|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128236": {
+      "content": "<|reserved_special_token_231|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128237": {
+      "content": "<|reserved_special_token_232|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128238": {
+      "content": "<|reserved_special_token_233|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128239": {
+      "content": "<|reserved_special_token_234|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128240": {
+      "content": "<|reserved_special_token_235|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128241": {
+      "content": "<|reserved_special_token_236|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128242": {
+      "content": "<|reserved_special_token_237|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128243": {
+      "content": "<|reserved_special_token_238|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128244": {
+      "content": "<|reserved_special_token_239|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128245": {
+      "content": "<|reserved_special_token_240|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128246": {
+      "content": "<|reserved_special_token_241|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128247": {
+      "content": "<|reserved_special_token_242|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128248": {
+      "content": "<|reserved_special_token_243|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128249": {
+      "content": "<|reserved_special_token_244|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128250": {
+      "content": "<|reserved_special_token_245|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128251": {
+      "content": "<|reserved_special_token_246|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128252": {
+      "content": "<|reserved_special_token_247|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128253": {
+      "content": "<|reserved_special_token_248|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128254": {
+      "content": "<|reserved_special_token_249|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128255": {
+      "content": "<|reserved_special_token_250|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<|begin_of_text|>",
+  "chat_template": "{% set loop_messages = messages %}{% for message in loop_messages %}{% set content = '<|start_header_id|>' + message['role'] + '<|end_header_id|>\n\n'+ message['content'] | trim + '<|eot_id|>' %}{% if loop.index0 == 0 %}{% set content = bos_token + content %}{% endif %}{{ content }}{% endfor %}{% if add_generation_prompt %}{{ '<|start_header_id|>assistant<|end_header_id|>\n\n' }}{% endif %}",
+  "clean_up_tokenization_spaces": true,
+  "eos_token": "<|eot_id|>",
+  "model_input_names": [
+    "input_ids",
+    "attention_mask"
+  ],
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<|eot_id|>",
+  "tokenizer_class": "PreTrainedTokenizerFast"
+}
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-4477/trainer_state.json b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-4477/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..591961ec116b7fa7cf876189377ccdbd3b9f8172
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-4477/trainer_state.json
@@ -0,0 +1,3186 @@
+{
+  "best_metric": 1.8028968572616577,
+  "best_model_checkpoint": "outputs-001/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-2985",
+  "epoch": 2.9996649916247904,
+  "eval_steps": 10,
+  "global_step": 4477,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.006700167504187605,
+      "grad_norm": 0.565915048122406,
+      "learning_rate": 0.0002,
+      "loss": 2.6189,
+      "step": 10
+    },
+    {
+      "epoch": 0.01340033500837521,
+      "grad_norm": 0.5004463791847229,
+      "learning_rate": 0.0002,
+      "loss": 2.3162,
+      "step": 20
+    },
+    {
+      "epoch": 0.020100502512562814,
+      "grad_norm": 0.511043906211853,
+      "learning_rate": 0.0002,
+      "loss": 2.0576,
+      "step": 30
+    },
+    {
+      "epoch": 0.02680067001675042,
+      "grad_norm": 0.47327178716659546,
+      "learning_rate": 0.0002,
+      "loss": 2.0085,
+      "step": 40
+    },
+    {
+      "epoch": 0.03350083752093802,
+      "grad_norm": 0.5511676669120789,
+      "learning_rate": 0.0002,
+      "loss": 2.0276,
+      "step": 50
+    },
+    {
+      "epoch": 0.04020100502512563,
+      "grad_norm": 0.4666278064250946,
+      "learning_rate": 0.0002,
+      "loss": 1.9075,
+      "step": 60
+    },
+    {
+      "epoch": 0.04690117252931323,
+      "grad_norm": 0.5310961008071899,
+      "learning_rate": 0.0002,
+      "loss": 1.8413,
+      "step": 70
+    },
+    {
+      "epoch": 0.05360134003350084,
+      "grad_norm": 0.5606027245521545,
+      "learning_rate": 0.0002,
+      "loss": 1.8711,
+      "step": 80
+    },
+    {
+      "epoch": 0.06030150753768844,
+      "grad_norm": 0.4934779703617096,
+      "learning_rate": 0.0002,
+      "loss": 1.9282,
+      "step": 90
+    },
+    {
+      "epoch": 0.06700167504187604,
+      "grad_norm": 0.4821869730949402,
+      "learning_rate": 0.0002,
+      "loss": 1.8925,
+      "step": 100
+    },
+    {
+      "epoch": 0.07370184254606366,
+      "grad_norm": 0.5262084603309631,
+      "learning_rate": 0.0002,
+      "loss": 1.8628,
+      "step": 110
+    },
+    {
+      "epoch": 0.08040201005025126,
+      "grad_norm": 0.3774230182170868,
+      "learning_rate": 0.0002,
+      "loss": 1.8347,
+      "step": 120
+    },
+    {
+      "epoch": 0.08710217755443886,
+      "grad_norm": 0.34137430787086487,
+      "learning_rate": 0.0002,
+      "loss": 1.8386,
+      "step": 130
+    },
+    {
+      "epoch": 0.09380234505862646,
+      "grad_norm": 0.407272070646286,
+      "learning_rate": 0.0002,
+      "loss": 1.861,
+      "step": 140
+    },
+    {
+      "epoch": 0.10050251256281408,
+      "grad_norm": 0.4011937975883484,
+      "learning_rate": 0.0002,
+      "loss": 1.8279,
+      "step": 150
+    },
+    {
+      "epoch": 0.10720268006700168,
+      "grad_norm": 0.4432467222213745,
+      "learning_rate": 0.0002,
+      "loss": 1.9317,
+      "step": 160
+    },
+    {
+      "epoch": 0.11390284757118928,
+      "grad_norm": 0.44030463695526123,
+      "learning_rate": 0.0002,
+      "loss": 1.8157,
+      "step": 170
+    },
+    {
+      "epoch": 0.12060301507537688,
+      "grad_norm": 0.3799569308757782,
+      "learning_rate": 0.0002,
+      "loss": 1.8534,
+      "step": 180
+    },
+    {
+      "epoch": 0.1273031825795645,
+      "grad_norm": 0.33721521496772766,
+      "learning_rate": 0.0002,
+      "loss": 1.7658,
+      "step": 190
+    },
+    {
+      "epoch": 0.13400335008375208,
+      "grad_norm": 0.4096226692199707,
+      "learning_rate": 0.0002,
+      "loss": 1.8269,
+      "step": 200
+    },
+    {
+      "epoch": 0.1407035175879397,
+      "grad_norm": 0.37374693155288696,
+      "learning_rate": 0.0002,
+      "loss": 1.802,
+      "step": 210
+    },
+    {
+      "epoch": 0.1474036850921273,
+      "grad_norm": 0.3249480128288269,
+      "learning_rate": 0.0002,
+      "loss": 1.8901,
+      "step": 220
+    },
+    {
+      "epoch": 0.1541038525963149,
+      "grad_norm": 0.3612042963504791,
+      "learning_rate": 0.0002,
+      "loss": 1.8163,
+      "step": 230
+    },
+    {
+      "epoch": 0.16080402010050251,
+      "grad_norm": 0.3686671257019043,
+      "learning_rate": 0.0002,
+      "loss": 1.7585,
+      "step": 240
+    },
+    {
+      "epoch": 0.16750418760469013,
+      "grad_norm": 0.3521044850349426,
+      "learning_rate": 0.0002,
+      "loss": 1.8365,
+      "step": 250
+    },
+    {
+      "epoch": 0.17420435510887772,
+      "grad_norm": 0.4073677361011505,
+      "learning_rate": 0.0002,
+      "loss": 1.8623,
+      "step": 260
+    },
+    {
+      "epoch": 0.18090452261306533,
+      "grad_norm": 0.34522193670272827,
+      "learning_rate": 0.0002,
+      "loss": 1.8026,
+      "step": 270
+    },
+    {
+      "epoch": 0.18760469011725292,
+      "grad_norm": 0.4121900498867035,
+      "learning_rate": 0.0002,
+      "loss": 1.8162,
+      "step": 280
+    },
+    {
+      "epoch": 0.19430485762144054,
+      "grad_norm": 0.3544778525829315,
+      "learning_rate": 0.0002,
+      "loss": 1.7976,
+      "step": 290
+    },
+    {
+      "epoch": 0.20100502512562815,
+      "grad_norm": 0.3482133448123932,
+      "learning_rate": 0.0002,
+      "loss": 1.8787,
+      "step": 300
+    },
+    {
+      "epoch": 0.20770519262981574,
+      "grad_norm": 0.3421826660633087,
+      "learning_rate": 0.0002,
+      "loss": 1.8578,
+      "step": 310
+    },
+    {
+      "epoch": 0.21440536013400335,
+      "grad_norm": 0.5024696588516235,
+      "learning_rate": 0.0002,
+      "loss": 1.8013,
+      "step": 320
+    },
+    {
+      "epoch": 0.22110552763819097,
+      "grad_norm": 0.36013063788414,
+      "learning_rate": 0.0002,
+      "loss": 1.8607,
+      "step": 330
+    },
+    {
+      "epoch": 0.22780569514237856,
+      "grad_norm": 0.3611244857311249,
+      "learning_rate": 0.0002,
+      "loss": 1.9075,
+      "step": 340
+    },
+    {
+      "epoch": 0.23450586264656617,
+      "grad_norm": 0.39244529604911804,
+      "learning_rate": 0.0002,
+      "loss": 1.8128,
+      "step": 350
+    },
+    {
+      "epoch": 0.24120603015075376,
+      "grad_norm": 0.3299325704574585,
+      "learning_rate": 0.0002,
+      "loss": 1.7885,
+      "step": 360
+    },
+    {
+      "epoch": 0.24790619765494137,
+      "grad_norm": 0.3994322419166565,
+      "learning_rate": 0.0002,
+      "loss": 1.8028,
+      "step": 370
+    },
+    {
+      "epoch": 0.254606365159129,
+      "grad_norm": 0.3559151887893677,
+      "learning_rate": 0.0002,
+      "loss": 1.8321,
+      "step": 380
+    },
+    {
+      "epoch": 0.2613065326633166,
+      "grad_norm": 0.3873756229877472,
+      "learning_rate": 0.0002,
+      "loss": 1.7802,
+      "step": 390
+    },
+    {
+      "epoch": 0.26800670016750416,
+      "grad_norm": 0.3710744082927704,
+      "learning_rate": 0.0002,
+      "loss": 1.7844,
+      "step": 400
+    },
+    {
+      "epoch": 0.2747068676716918,
+      "grad_norm": 0.3618465065956116,
+      "learning_rate": 0.0002,
+      "loss": 1.7789,
+      "step": 410
+    },
+    {
+      "epoch": 0.2814070351758794,
+      "grad_norm": 0.30063769221305847,
+      "learning_rate": 0.0002,
+      "loss": 1.8529,
+      "step": 420
+    },
+    {
+      "epoch": 0.288107202680067,
+      "grad_norm": 0.3695628345012665,
+      "learning_rate": 0.0002,
+      "loss": 1.7765,
+      "step": 430
+    },
+    {
+      "epoch": 0.2948073701842546,
+      "grad_norm": 0.31451135873794556,
+      "learning_rate": 0.0002,
+      "loss": 1.7982,
+      "step": 440
+    },
+    {
+      "epoch": 0.3015075376884422,
+      "grad_norm": 0.3959707021713257,
+      "learning_rate": 0.0002,
+      "loss": 1.7517,
+      "step": 450
+    },
+    {
+      "epoch": 0.3082077051926298,
+      "grad_norm": 0.33475354313850403,
+      "learning_rate": 0.0002,
+      "loss": 1.8142,
+      "step": 460
+    },
+    {
+      "epoch": 0.3149078726968174,
+      "grad_norm": 0.33933115005493164,
+      "learning_rate": 0.0002,
+      "loss": 1.8805,
+      "step": 470
+    },
+    {
+      "epoch": 0.32160804020100503,
+      "grad_norm": 0.3264943063259125,
+      "learning_rate": 0.0002,
+      "loss": 1.7564,
+      "step": 480
+    },
+    {
+      "epoch": 0.32830820770519265,
+      "grad_norm": 0.40188100934028625,
+      "learning_rate": 0.0002,
+      "loss": 1.8428,
+      "step": 490
+    },
+    {
+      "epoch": 0.33500837520938026,
+      "grad_norm": 0.37408649921417236,
+      "learning_rate": 0.0002,
+      "loss": 1.7624,
+      "step": 500
+    },
+    {
+      "epoch": 0.3417085427135678,
+      "grad_norm": 0.33925938606262207,
+      "learning_rate": 0.0002,
+      "loss": 1.7745,
+      "step": 510
+    },
+    {
+      "epoch": 0.34840871021775544,
+      "grad_norm": 0.36836713552474976,
+      "learning_rate": 0.0002,
+      "loss": 1.814,
+      "step": 520
+    },
+    {
+      "epoch": 0.35510887772194305,
+      "grad_norm": 0.37284499406814575,
+      "learning_rate": 0.0002,
+      "loss": 1.8037,
+      "step": 530
+    },
+    {
+      "epoch": 0.36180904522613067,
+      "grad_norm": 0.3192278742790222,
+      "learning_rate": 0.0002,
+      "loss": 1.8379,
+      "step": 540
+    },
+    {
+      "epoch": 0.3685092127303183,
+      "grad_norm": 0.30233290791511536,
+      "learning_rate": 0.0002,
+      "loss": 1.8702,
+      "step": 550
+    },
+    {
+      "epoch": 0.37520938023450584,
+      "grad_norm": 0.3340817391872406,
+      "learning_rate": 0.0002,
+      "loss": 1.8799,
+      "step": 560
+    },
+    {
+      "epoch": 0.38190954773869346,
+      "grad_norm": 0.32600095868110657,
+      "learning_rate": 0.0002,
+      "loss": 1.8404,
+      "step": 570
+    },
+    {
+      "epoch": 0.38860971524288107,
+      "grad_norm": 0.33711278438568115,
+      "learning_rate": 0.0002,
+      "loss": 1.7804,
+      "step": 580
+    },
+    {
+      "epoch": 0.3953098827470687,
+      "grad_norm": 0.34890690445899963,
+      "learning_rate": 0.0002,
+      "loss": 1.8445,
+      "step": 590
+    },
+    {
+      "epoch": 0.4020100502512563,
+      "grad_norm": 0.38238924741744995,
+      "learning_rate": 0.0002,
+      "loss": 1.8187,
+      "step": 600
+    },
+    {
+      "epoch": 0.40871021775544386,
+      "grad_norm": 0.34399354457855225,
+      "learning_rate": 0.0002,
+      "loss": 1.8111,
+      "step": 610
+    },
+    {
+      "epoch": 0.4154103852596315,
+      "grad_norm": 0.3346073627471924,
+      "learning_rate": 0.0002,
+      "loss": 1.8006,
+      "step": 620
+    },
+    {
+      "epoch": 0.4221105527638191,
+      "grad_norm": 0.3545648157596588,
+      "learning_rate": 0.0002,
+      "loss": 1.7705,
+      "step": 630
+    },
+    {
+      "epoch": 0.4288107202680067,
+      "grad_norm": 0.3378899097442627,
+      "learning_rate": 0.0002,
+      "loss": 1.8445,
+      "step": 640
+    },
+    {
+      "epoch": 0.4355108877721943,
+      "grad_norm": 0.3255569040775299,
+      "learning_rate": 0.0002,
+      "loss": 1.804,
+      "step": 650
+    },
+    {
+      "epoch": 0.44221105527638194,
+      "grad_norm": 0.34880587458610535,
+      "learning_rate": 0.0002,
+      "loss": 1.7679,
+      "step": 660
+    },
+    {
+      "epoch": 0.4489112227805695,
+      "grad_norm": 0.3402383625507355,
+      "learning_rate": 0.0002,
+      "loss": 1.7861,
+      "step": 670
+    },
+    {
+      "epoch": 0.4556113902847571,
+      "grad_norm": 0.3594033718109131,
+      "learning_rate": 0.0002,
+      "loss": 1.8131,
+      "step": 680
+    },
+    {
+      "epoch": 0.4623115577889447,
+      "grad_norm": 0.31000566482543945,
+      "learning_rate": 0.0002,
+      "loss": 1.8399,
+      "step": 690
+    },
+    {
+      "epoch": 0.46901172529313234,
+      "grad_norm": 0.37229061126708984,
+      "learning_rate": 0.0002,
+      "loss": 1.7521,
+      "step": 700
+    },
+    {
+      "epoch": 0.47571189279731996,
+      "grad_norm": 0.315801739692688,
+      "learning_rate": 0.0002,
+      "loss": 1.7779,
+      "step": 710
+    },
+    {
+      "epoch": 0.4824120603015075,
+      "grad_norm": 0.3220832645893097,
+      "learning_rate": 0.0002,
+      "loss": 1.7515,
+      "step": 720
+    },
+    {
+      "epoch": 0.48911222780569513,
+      "grad_norm": 0.3435456156730652,
+      "learning_rate": 0.0002,
+      "loss": 1.7181,
+      "step": 730
+    },
+    {
+      "epoch": 0.49581239530988275,
+      "grad_norm": 0.30380892753601074,
+      "learning_rate": 0.0002,
+      "loss": 1.8844,
+      "step": 740
+    },
+    {
+      "epoch": 0.5025125628140703,
+      "grad_norm": 0.3555026054382324,
+      "learning_rate": 0.0002,
+      "loss": 1.7792,
+      "step": 750
+    },
+    {
+      "epoch": 0.509212730318258,
+      "grad_norm": 0.3019855320453644,
+      "learning_rate": 0.0002,
+      "loss": 1.7714,
+      "step": 760
+    },
+    {
+      "epoch": 0.5159128978224455,
+      "grad_norm": 0.309111088514328,
+      "learning_rate": 0.0002,
+      "loss": 1.7962,
+      "step": 770
+    },
+    {
+      "epoch": 0.5226130653266332,
+      "grad_norm": 0.366020530462265,
+      "learning_rate": 0.0002,
+      "loss": 1.7913,
+      "step": 780
+    },
+    {
+      "epoch": 0.5293132328308208,
+      "grad_norm": 0.3267050087451935,
+      "learning_rate": 0.0002,
+      "loss": 1.8008,
+      "step": 790
+    },
+    {
+      "epoch": 0.5360134003350083,
+      "grad_norm": 0.34265750646591187,
+      "learning_rate": 0.0002,
+      "loss": 1.7397,
+      "step": 800
+    },
+    {
+      "epoch": 0.542713567839196,
+      "grad_norm": 0.313669890165329,
+      "learning_rate": 0.0002,
+      "loss": 1.8251,
+      "step": 810
+    },
+    {
+      "epoch": 0.5494137353433836,
+      "grad_norm": 0.3355236053466797,
+      "learning_rate": 0.0002,
+      "loss": 1.8369,
+      "step": 820
+    },
+    {
+      "epoch": 0.5561139028475712,
+      "grad_norm": 0.3186608552932739,
+      "learning_rate": 0.0002,
+      "loss": 1.7381,
+      "step": 830
+    },
+    {
+      "epoch": 0.5628140703517588,
+      "grad_norm": 0.30357518792152405,
+      "learning_rate": 0.0002,
+      "loss": 1.8034,
+      "step": 840
+    },
+    {
+      "epoch": 0.5695142378559463,
+      "grad_norm": 0.3990040123462677,
+      "learning_rate": 0.0002,
+      "loss": 1.769,
+      "step": 850
+    },
+    {
+      "epoch": 0.576214405360134,
+      "grad_norm": 0.34363803267478943,
+      "learning_rate": 0.0002,
+      "loss": 1.7482,
+      "step": 860
+    },
+    {
+      "epoch": 0.5829145728643216,
+      "grad_norm": 0.3757908046245575,
+      "learning_rate": 0.0002,
+      "loss": 1.8106,
+      "step": 870
+    },
+    {
+      "epoch": 0.5896147403685092,
+      "grad_norm": 0.3359757661819458,
+      "learning_rate": 0.0002,
+      "loss": 1.8104,
+      "step": 880
+    },
+    {
+      "epoch": 0.5963149078726968,
+      "grad_norm": 0.5555329918861389,
+      "learning_rate": 0.0002,
+      "loss": 1.7591,
+      "step": 890
+    },
+    {
+      "epoch": 0.6030150753768844,
+      "grad_norm": 0.4046323895454407,
+      "learning_rate": 0.0002,
+      "loss": 1.7715,
+      "step": 900
+    },
+    {
+      "epoch": 0.609715242881072,
+      "grad_norm": 0.29834219813346863,
+      "learning_rate": 0.0002,
+      "loss": 1.7998,
+      "step": 910
+    },
+    {
+      "epoch": 0.6164154103852596,
+      "grad_norm": 0.3241238594055176,
+      "learning_rate": 0.0002,
+      "loss": 1.7826,
+      "step": 920
+    },
+    {
+      "epoch": 0.6231155778894473,
+      "grad_norm": 0.35154739022254944,
+      "learning_rate": 0.0002,
+      "loss": 1.8342,
+      "step": 930
+    },
+    {
+      "epoch": 0.6298157453936348,
+      "grad_norm": 0.3287706673145294,
+      "learning_rate": 0.0002,
+      "loss": 1.8076,
+      "step": 940
+    },
+    {
+      "epoch": 0.6365159128978225,
+      "grad_norm": 0.35670626163482666,
+      "learning_rate": 0.0002,
+      "loss": 1.8038,
+      "step": 950
+    },
+    {
+      "epoch": 0.6432160804020101,
+      "grad_norm": 0.6114104986190796,
+      "learning_rate": 0.0002,
+      "loss": 1.869,
+      "step": 960
+    },
+    {
+      "epoch": 0.6499162479061976,
+      "grad_norm": 0.3186565041542053,
+      "learning_rate": 0.0002,
+      "loss": 1.8297,
+      "step": 970
+    },
+    {
+      "epoch": 0.6566164154103853,
+      "grad_norm": 0.27164125442504883,
+      "learning_rate": 0.0002,
+      "loss": 1.7539,
+      "step": 980
+    },
+    {
+      "epoch": 0.6633165829145728,
+      "grad_norm": 0.34407344460487366,
+      "learning_rate": 0.0002,
+      "loss": 1.8339,
+      "step": 990
+    },
+    {
+      "epoch": 0.6700167504187605,
+      "grad_norm": 0.368415892124176,
+      "learning_rate": 0.0002,
+      "loss": 1.855,
+      "step": 1000
+    },
+    {
+      "epoch": 0.6767169179229481,
+      "grad_norm": 0.3306390643119812,
+      "learning_rate": 0.0002,
+      "loss": 1.7821,
+      "step": 1010
+    },
+    {
+      "epoch": 0.6834170854271356,
+      "grad_norm": 0.3198648989200592,
+      "learning_rate": 0.0002,
+      "loss": 1.786,
+      "step": 1020
+    },
+    {
+      "epoch": 0.6901172529313233,
+      "grad_norm": 0.3092987537384033,
+      "learning_rate": 0.0002,
+      "loss": 1.816,
+      "step": 1030
+    },
+    {
+      "epoch": 0.6968174204355109,
+      "grad_norm": 0.3090653419494629,
+      "learning_rate": 0.0002,
+      "loss": 1.7689,
+      "step": 1040
+    },
+    {
+      "epoch": 0.7035175879396985,
+      "grad_norm": 0.3485880196094513,
+      "learning_rate": 0.0002,
+      "loss": 1.7544,
+      "step": 1050
+    },
+    {
+      "epoch": 0.7102177554438861,
+      "grad_norm": 0.35782721638679504,
+      "learning_rate": 0.0002,
+      "loss": 1.8187,
+      "step": 1060
+    },
+    {
+      "epoch": 0.7169179229480737,
+      "grad_norm": 0.34256869554519653,
+      "learning_rate": 0.0002,
+      "loss": 1.806,
+      "step": 1070
+    },
+    {
+      "epoch": 0.7236180904522613,
+      "grad_norm": 0.30461037158966064,
+      "learning_rate": 0.0002,
+      "loss": 1.7873,
+      "step": 1080
+    },
+    {
+      "epoch": 0.7303182579564489,
+      "grad_norm": 0.3398691713809967,
+      "learning_rate": 0.0002,
+      "loss": 1.7367,
+      "step": 1090
+    },
+    {
+      "epoch": 0.7370184254606366,
+      "grad_norm": 0.3180808126926422,
+      "learning_rate": 0.0002,
+      "loss": 1.8756,
+      "step": 1100
+    },
+    {
+      "epoch": 0.7437185929648241,
+      "grad_norm": 0.34400665760040283,
+      "learning_rate": 0.0002,
+      "loss": 1.6988,
+      "step": 1110
+    },
+    {
+      "epoch": 0.7504187604690117,
+      "grad_norm": 0.34244877099990845,
+      "learning_rate": 0.0002,
+      "loss": 1.7851,
+      "step": 1120
+    },
+    {
+      "epoch": 0.7571189279731994,
+      "grad_norm": 0.29946693778038025,
+      "learning_rate": 0.0002,
+      "loss": 1.7841,
+      "step": 1130
+    },
+    {
+      "epoch": 0.7638190954773869,
+      "grad_norm": 0.37547236680984497,
+      "learning_rate": 0.0002,
+      "loss": 1.7456,
+      "step": 1140
+    },
+    {
+      "epoch": 0.7705192629815746,
+      "grad_norm": 0.3263005018234253,
+      "learning_rate": 0.0002,
+      "loss": 1.8425,
+      "step": 1150
+    },
+    {
+      "epoch": 0.7772194304857621,
+      "grad_norm": 0.41363608837127686,
+      "learning_rate": 0.0002,
+      "loss": 1.7222,
+      "step": 1160
+    },
+    {
+      "epoch": 0.7839195979899497,
+      "grad_norm": 0.36267954111099243,
+      "learning_rate": 0.0002,
+      "loss": 1.7836,
+      "step": 1170
+    },
+    {
+      "epoch": 0.7906197654941374,
+      "grad_norm": 0.31789499521255493,
+      "learning_rate": 0.0002,
+      "loss": 1.9183,
+      "step": 1180
+    },
+    {
+      "epoch": 0.7973199329983249,
+      "grad_norm": 0.5708149075508118,
+      "learning_rate": 0.0002,
+      "loss": 1.78,
+      "step": 1190
+    },
+    {
+      "epoch": 0.8040201005025126,
+      "grad_norm": 0.322099506855011,
+      "learning_rate": 0.0002,
+      "loss": 1.6908,
+      "step": 1200
+    },
+    {
+      "epoch": 0.8107202680067002,
+      "grad_norm": 0.3419909179210663,
+      "learning_rate": 0.0002,
+      "loss": 1.7639,
+      "step": 1210
+    },
+    {
+      "epoch": 0.8174204355108877,
+      "grad_norm": 0.36286255717277527,
+      "learning_rate": 0.0002,
+      "loss": 1.7428,
+      "step": 1220
+    },
+    {
+      "epoch": 0.8241206030150754,
+      "grad_norm": 0.33992862701416016,
+      "learning_rate": 0.0002,
+      "loss": 1.8409,
+      "step": 1230
+    },
+    {
+      "epoch": 0.830820770519263,
+      "grad_norm": 0.32622793316841125,
+      "learning_rate": 0.0002,
+      "loss": 1.7507,
+      "step": 1240
+    },
+    {
+      "epoch": 0.8375209380234506,
+      "grad_norm": 0.3036167621612549,
+      "learning_rate": 0.0002,
+      "loss": 1.8098,
+      "step": 1250
+    },
+    {
+      "epoch": 0.8442211055276382,
+      "grad_norm": 0.3182215392589569,
+      "learning_rate": 0.0002,
+      "loss": 1.8094,
+      "step": 1260
+    },
+    {
+      "epoch": 0.8509212730318257,
+      "grad_norm": 0.3270018696784973,
+      "learning_rate": 0.0002,
+      "loss": 1.8017,
+      "step": 1270
+    },
+    {
+      "epoch": 0.8576214405360134,
+      "grad_norm": 0.32652342319488525,
+      "learning_rate": 0.0002,
+      "loss": 1.798,
+      "step": 1280
+    },
+    {
+      "epoch": 0.864321608040201,
+      "grad_norm": 0.3631329834461212,
+      "learning_rate": 0.0002,
+      "loss": 1.7448,
+      "step": 1290
+    },
+    {
+      "epoch": 0.8710217755443886,
+      "grad_norm": 0.36706018447875977,
+      "learning_rate": 0.0002,
+      "loss": 1.7,
+      "step": 1300
+    },
+    {
+      "epoch": 0.8777219430485762,
+      "grad_norm": 0.3347418010234833,
+      "learning_rate": 0.0002,
+      "loss": 1.8178,
+      "step": 1310
+    },
+    {
+      "epoch": 0.8844221105527639,
+      "grad_norm": 0.34371060132980347,
+      "learning_rate": 0.0002,
+      "loss": 1.7824,
+      "step": 1320
+    },
+    {
+      "epoch": 0.8911222780569514,
+      "grad_norm": 0.3029090166091919,
+      "learning_rate": 0.0002,
+      "loss": 1.783,
+      "step": 1330
+    },
+    {
+      "epoch": 0.897822445561139,
+      "grad_norm": 0.34700682759284973,
+      "learning_rate": 0.0002,
+      "loss": 1.8017,
+      "step": 1340
+    },
+    {
+      "epoch": 0.9045226130653267,
+      "grad_norm": 0.35574328899383545,
+      "learning_rate": 0.0002,
+      "loss": 1.7998,
+      "step": 1350
+    },
+    {
+      "epoch": 0.9112227805695142,
+      "grad_norm": 0.30928221344947815,
+      "learning_rate": 0.0002,
+      "loss": 1.7339,
+      "step": 1360
+    },
+    {
+      "epoch": 0.9179229480737019,
+      "grad_norm": 0.30652928352355957,
+      "learning_rate": 0.0002,
+      "loss": 1.7479,
+      "step": 1370
+    },
+    {
+      "epoch": 0.9246231155778895,
+      "grad_norm": 0.3838157653808594,
+      "learning_rate": 0.0002,
+      "loss": 1.7491,
+      "step": 1380
+    },
+    {
+      "epoch": 0.931323283082077,
+      "grad_norm": 0.31655240058898926,
+      "learning_rate": 0.0002,
+      "loss": 1.7977,
+      "step": 1390
+    },
+    {
+      "epoch": 0.9380234505862647,
+      "grad_norm": 0.41737303137779236,
+      "learning_rate": 0.0002,
+      "loss": 1.8175,
+      "step": 1400
+    },
+    {
+      "epoch": 0.9447236180904522,
+      "grad_norm": 0.3227267861366272,
+      "learning_rate": 0.0002,
+      "loss": 1.6811,
+      "step": 1410
+    },
+    {
+      "epoch": 0.9514237855946399,
+      "grad_norm": 0.3729925751686096,
+      "learning_rate": 0.0002,
+      "loss": 1.7343,
+      "step": 1420
+    },
+    {
+      "epoch": 0.9581239530988275,
+      "grad_norm": 0.30779409408569336,
+      "learning_rate": 0.0002,
+      "loss": 1.8221,
+      "step": 1430
+    },
+    {
+      "epoch": 0.964824120603015,
+      "grad_norm": 0.334379643201828,
+      "learning_rate": 0.0002,
+      "loss": 1.7972,
+      "step": 1440
+    },
+    {
+      "epoch": 0.9715242881072027,
+      "grad_norm": 0.3568236231803894,
+      "learning_rate": 0.0002,
+      "loss": 1.7141,
+      "step": 1450
+    },
+    {
+      "epoch": 0.9782244556113903,
+      "grad_norm": 0.33310577273368835,
+      "learning_rate": 0.0002,
+      "loss": 1.7541,
+      "step": 1460
+    },
+    {
+      "epoch": 0.9849246231155779,
+      "grad_norm": 0.2972261905670166,
+      "learning_rate": 0.0002,
+      "loss": 1.8511,
+      "step": 1470
+    },
+    {
+      "epoch": 0.9916247906197655,
+      "grad_norm": 0.3322717845439911,
+      "learning_rate": 0.0002,
+      "loss": 1.7654,
+      "step": 1480
+    },
+    {
+      "epoch": 0.998324958123953,
+      "grad_norm": 0.3276330828666687,
+      "learning_rate": 0.0002,
+      "loss": 1.8033,
+      "step": 1490
+    },
+    {
+      "epoch": 0.9996649916247906,
+      "eval_loss": 1.8036354780197144,
+      "eval_runtime": 37.8949,
+      "eval_samples_per_second": 13.59,
+      "eval_steps_per_second": 1.715,
+      "step": 1492
+    },
+    {
+      "epoch": 1.0050251256281406,
+      "grad_norm": 0.29252371191978455,
+      "learning_rate": 0.0002,
+      "loss": 1.7138,
+      "step": 1500
+    },
+    {
+      "epoch": 1.0117252931323284,
+      "grad_norm": 0.31607162952423096,
+      "learning_rate": 0.0002,
+      "loss": 1.8198,
+      "step": 1510
+    },
+    {
+      "epoch": 1.018425460636516,
+      "grad_norm": 0.32294467091560364,
+      "learning_rate": 0.0002,
+      "loss": 1.6779,
+      "step": 1520
+    },
+    {
+      "epoch": 1.0251256281407035,
+      "grad_norm": 0.3868017792701721,
+      "learning_rate": 0.0002,
+      "loss": 1.7919,
+      "step": 1530
+    },
+    {
+      "epoch": 1.031825795644891,
+      "grad_norm": 0.3178282082080841,
+      "learning_rate": 0.0002,
+      "loss": 1.7954,
+      "step": 1540
+    },
+    {
+      "epoch": 1.0385259631490786,
+      "grad_norm": 0.3706750273704529,
+      "learning_rate": 0.0002,
+      "loss": 1.7136,
+      "step": 1550
+    },
+    {
+      "epoch": 1.0452261306532664,
+      "grad_norm": 0.33930912613868713,
+      "learning_rate": 0.0002,
+      "loss": 1.7382,
+      "step": 1560
+    },
+    {
+      "epoch": 1.051926298157454,
+      "grad_norm": 0.33970504999160767,
+      "learning_rate": 0.0002,
+      "loss": 1.7602,
+      "step": 1570
+    },
+    {
+      "epoch": 1.0586264656616415,
+      "grad_norm": 0.42553383111953735,
+      "learning_rate": 0.0002,
+      "loss": 1.6573,
+      "step": 1580
+    },
+    {
+      "epoch": 1.065326633165829,
+      "grad_norm": 0.3772421181201935,
+      "learning_rate": 0.0002,
+      "loss": 1.645,
+      "step": 1590
+    },
+    {
+      "epoch": 1.0720268006700167,
+      "grad_norm": 0.34212902188301086,
+      "learning_rate": 0.0002,
+      "loss": 1.7362,
+      "step": 1600
+    },
+    {
+      "epoch": 1.0787269681742044,
+      "grad_norm": 0.3798283338546753,
+      "learning_rate": 0.0002,
+      "loss": 1.7057,
+      "step": 1610
+    },
+    {
+      "epoch": 1.085427135678392,
+      "grad_norm": 0.36909598112106323,
+      "learning_rate": 0.0002,
+      "loss": 1.7468,
+      "step": 1620
+    },
+    {
+      "epoch": 1.0921273031825796,
+      "grad_norm": 0.3344230651855469,
+      "learning_rate": 0.0002,
+      "loss": 1.7807,
+      "step": 1630
+    },
+    {
+      "epoch": 1.0988274706867671,
+      "grad_norm": 0.3862569332122803,
+      "learning_rate": 0.0002,
+      "loss": 1.7111,
+      "step": 1640
+    },
+    {
+      "epoch": 1.1055276381909547,
+      "grad_norm": 0.31188511848449707,
+      "learning_rate": 0.0002,
+      "loss": 1.7163,
+      "step": 1650
+    },
+    {
+      "epoch": 1.1122278056951425,
+      "grad_norm": 0.3563670814037323,
+      "learning_rate": 0.0002,
+      "loss": 1.7263,
+      "step": 1660
+    },
+    {
+      "epoch": 1.11892797319933,
+      "grad_norm": 0.35052165389060974,
+      "learning_rate": 0.0002,
+      "loss": 1.7718,
+      "step": 1670
+    },
+    {
+      "epoch": 1.1256281407035176,
+      "grad_norm": 0.3285699188709259,
+      "learning_rate": 0.0002,
+      "loss": 1.7601,
+      "step": 1680
+    },
+    {
+      "epoch": 1.1323283082077051,
+      "grad_norm": 0.3639393746852875,
+      "learning_rate": 0.0002,
+      "loss": 1.6877,
+      "step": 1690
+    },
+    {
+      "epoch": 1.1390284757118927,
+      "grad_norm": 0.3842753767967224,
+      "learning_rate": 0.0002,
+      "loss": 1.7719,
+      "step": 1700
+    },
+    {
+      "epoch": 1.1457286432160805,
+      "grad_norm": 0.3624933063983917,
+      "learning_rate": 0.0002,
+      "loss": 1.7002,
+      "step": 1710
+    },
+    {
+      "epoch": 1.152428810720268,
+      "grad_norm": 0.3641220033168793,
+      "learning_rate": 0.0002,
+      "loss": 1.7243,
+      "step": 1720
+    },
+    {
+      "epoch": 1.1591289782244556,
+      "grad_norm": 0.32765355706214905,
+      "learning_rate": 0.0002,
+      "loss": 1.752,
+      "step": 1730
+    },
+    {
+      "epoch": 1.1658291457286432,
+      "grad_norm": 0.34974896907806396,
+      "learning_rate": 0.0002,
+      "loss": 1.6556,
+      "step": 1740
+    },
+    {
+      "epoch": 1.1725293132328307,
+      "grad_norm": 0.3910926580429077,
+      "learning_rate": 0.0002,
+      "loss": 1.7273,
+      "step": 1750
+    },
+    {
+      "epoch": 1.1792294807370185,
+      "grad_norm": 0.3564300537109375,
+      "learning_rate": 0.0002,
+      "loss": 1.7204,
+      "step": 1760
+    },
+    {
+      "epoch": 1.185929648241206,
+      "grad_norm": 0.34822574257850647,
+      "learning_rate": 0.0002,
+      "loss": 1.746,
+      "step": 1770
+    },
+    {
+      "epoch": 1.1926298157453936,
+      "grad_norm": 0.36185044050216675,
+      "learning_rate": 0.0002,
+      "loss": 1.7256,
+      "step": 1780
+    },
+    {
+      "epoch": 1.1993299832495812,
+      "grad_norm": 0.34866711497306824,
+      "learning_rate": 0.0002,
+      "loss": 1.6431,
+      "step": 1790
+    },
+    {
+      "epoch": 1.2060301507537687,
+      "grad_norm": 0.4017769992351532,
+      "learning_rate": 0.0002,
+      "loss": 1.8084,
+      "step": 1800
+    },
+    {
+      "epoch": 1.2127303182579565,
+      "grad_norm": 0.32930681109428406,
+      "learning_rate": 0.0002,
+      "loss": 1.6985,
+      "step": 1810
+    },
+    {
+      "epoch": 1.219430485762144,
+      "grad_norm": 0.35951921343803406,
+      "learning_rate": 0.0002,
+      "loss": 1.7606,
+      "step": 1820
+    },
+    {
+      "epoch": 1.2261306532663316,
+      "grad_norm": 0.37366992235183716,
+      "learning_rate": 0.0002,
+      "loss": 1.6933,
+      "step": 1830
+    },
+    {
+      "epoch": 1.2328308207705192,
+      "grad_norm": 0.3565689027309418,
+      "learning_rate": 0.0002,
+      "loss": 1.6737,
+      "step": 1840
+    },
+    {
+      "epoch": 1.2395309882747068,
+      "grad_norm": 0.3692343533039093,
+      "learning_rate": 0.0002,
+      "loss": 1.8013,
+      "step": 1850
+    },
+    {
+      "epoch": 1.2462311557788945,
+      "grad_norm": 0.38426971435546875,
+      "learning_rate": 0.0002,
+      "loss": 1.736,
+      "step": 1860
+    },
+    {
+      "epoch": 1.252931323283082,
+      "grad_norm": 0.33559855818748474,
+      "learning_rate": 0.0002,
+      "loss": 1.7031,
+      "step": 1870
+    },
+    {
+      "epoch": 1.2596314907872697,
+      "grad_norm": 0.34181106090545654,
+      "learning_rate": 0.0002,
+      "loss": 1.7033,
+      "step": 1880
+    },
+    {
+      "epoch": 1.2663316582914572,
+      "grad_norm": 0.3916318416595459,
+      "learning_rate": 0.0002,
+      "loss": 1.7707,
+      "step": 1890
+    },
+    {
+      "epoch": 1.2730318257956448,
+      "grad_norm": 0.3887825012207031,
+      "learning_rate": 0.0002,
+      "loss": 1.6686,
+      "step": 1900
+    },
+    {
+      "epoch": 1.2797319932998326,
+      "grad_norm": 0.33583927154541016,
+      "learning_rate": 0.0002,
+      "loss": 1.7062,
+      "step": 1910
+    },
+    {
+      "epoch": 1.2864321608040201,
+      "grad_norm": 0.37639349699020386,
+      "learning_rate": 0.0002,
+      "loss": 1.717,
+      "step": 1920
+    },
+    {
+      "epoch": 1.2931323283082077,
+      "grad_norm": 0.38059428334236145,
+      "learning_rate": 0.0002,
+      "loss": 1.777,
+      "step": 1930
+    },
+    {
+      "epoch": 1.2998324958123952,
+      "grad_norm": 0.37253183126449585,
+      "learning_rate": 0.0002,
+      "loss": 1.6126,
+      "step": 1940
+    },
+    {
+      "epoch": 1.3065326633165828,
+      "grad_norm": 0.37371566891670227,
+      "learning_rate": 0.0002,
+      "loss": 1.6758,
+      "step": 1950
+    },
+    {
+      "epoch": 1.3132328308207706,
+      "grad_norm": 0.4080910086631775,
+      "learning_rate": 0.0002,
+      "loss": 1.6788,
+      "step": 1960
+    },
+    {
+      "epoch": 1.3199329983249581,
+      "grad_norm": 0.3174354135990143,
+      "learning_rate": 0.0002,
+      "loss": 1.6518,
+      "step": 1970
+    },
+    {
+      "epoch": 1.3266331658291457,
+      "grad_norm": 0.4518888294696808,
+      "learning_rate": 0.0002,
+      "loss": 1.7925,
+      "step": 1980
+    },
+    {
+      "epoch": 1.3333333333333333,
+      "grad_norm": 0.3627921938896179,
+      "learning_rate": 0.0002,
+      "loss": 1.7085,
+      "step": 1990
+    },
+    {
+      "epoch": 1.3400335008375208,
+      "grad_norm": 0.3655930161476135,
+      "learning_rate": 0.0002,
+      "loss": 1.7676,
+      "step": 2000
+    },
+    {
+      "epoch": 1.3467336683417086,
+      "grad_norm": 0.3509993255138397,
+      "learning_rate": 0.0002,
+      "loss": 1.7016,
+      "step": 2010
+    },
+    {
+      "epoch": 1.3534338358458962,
+      "grad_norm": 0.4281129240989685,
+      "learning_rate": 0.0002,
+      "loss": 1.7359,
+      "step": 2020
+    },
+    {
+      "epoch": 1.3601340033500837,
+      "grad_norm": 0.3821414113044739,
+      "learning_rate": 0.0002,
+      "loss": 1.6884,
+      "step": 2030
+    },
+    {
+      "epoch": 1.3668341708542713,
+      "grad_norm": 0.3907586336135864,
+      "learning_rate": 0.0002,
+      "loss": 1.7075,
+      "step": 2040
+    },
+    {
+      "epoch": 1.3735343383584588,
+      "grad_norm": 0.37792932987213135,
+      "learning_rate": 0.0002,
+      "loss": 1.7424,
+      "step": 2050
+    },
+    {
+      "epoch": 1.3802345058626466,
+      "grad_norm": 0.3693985641002655,
+      "learning_rate": 0.0002,
+      "loss": 1.7305,
+      "step": 2060
+    },
+    {
+      "epoch": 1.3869346733668342,
+      "grad_norm": 0.32275936007499695,
+      "learning_rate": 0.0002,
+      "loss": 1.7434,
+      "step": 2070
+    },
+    {
+      "epoch": 1.3936348408710217,
+      "grad_norm": 0.3789440095424652,
+      "learning_rate": 0.0002,
+      "loss": 1.6677,
+      "step": 2080
+    },
+    {
+      "epoch": 1.4003350083752093,
+      "grad_norm": 0.3638380467891693,
+      "learning_rate": 0.0002,
+      "loss": 1.6825,
+      "step": 2090
+    },
+    {
+      "epoch": 1.4070351758793969,
+      "grad_norm": 0.3495481610298157,
+      "learning_rate": 0.0002,
+      "loss": 1.6542,
+      "step": 2100
+    },
+    {
+      "epoch": 1.4137353433835846,
+      "grad_norm": 0.37920597195625305,
+      "learning_rate": 0.0002,
+      "loss": 1.7225,
+      "step": 2110
+    },
+    {
+      "epoch": 1.4204355108877722,
+      "grad_norm": 0.37218064069747925,
+      "learning_rate": 0.0002,
+      "loss": 1.7329,
+      "step": 2120
+    },
+    {
+      "epoch": 1.4271356783919598,
+      "grad_norm": 0.38074082136154175,
+      "learning_rate": 0.0002,
+      "loss": 1.799,
+      "step": 2130
+    },
+    {
+      "epoch": 1.4338358458961473,
+      "grad_norm": 0.3455527126789093,
+      "learning_rate": 0.0002,
+      "loss": 1.7403,
+      "step": 2140
+    },
+    {
+      "epoch": 1.4405360134003349,
+      "grad_norm": 0.3712003529071808,
+      "learning_rate": 0.0002,
+      "loss": 1.776,
+      "step": 2150
+    },
+    {
+      "epoch": 1.4472361809045227,
+      "grad_norm": 0.3786754906177521,
+      "learning_rate": 0.0002,
+      "loss": 1.7619,
+      "step": 2160
+    },
+    {
+      "epoch": 1.4539363484087102,
+      "grad_norm": 0.3879223167896271,
+      "learning_rate": 0.0002,
+      "loss": 1.68,
+      "step": 2170
+    },
+    {
+      "epoch": 1.4606365159128978,
+      "grad_norm": 0.38738805055618286,
+      "learning_rate": 0.0002,
+      "loss": 1.7,
+      "step": 2180
+    },
+    {
+      "epoch": 1.4673366834170856,
+      "grad_norm": 0.39768800139427185,
+      "learning_rate": 0.0002,
+      "loss": 1.7581,
+      "step": 2190
+    },
+    {
+      "epoch": 1.474036850921273,
+      "grad_norm": 0.4172441065311432,
+      "learning_rate": 0.0002,
+      "loss": 1.7671,
+      "step": 2200
+    },
+    {
+      "epoch": 1.4807370184254607,
+      "grad_norm": 0.4043174982070923,
+      "learning_rate": 0.0002,
+      "loss": 1.6736,
+      "step": 2210
+    },
+    {
+      "epoch": 1.4874371859296482,
+      "grad_norm": 0.3750883936882019,
+      "learning_rate": 0.0002,
+      "loss": 1.7444,
+      "step": 2220
+    },
+    {
+      "epoch": 1.4941373534338358,
+      "grad_norm": 0.3552253246307373,
+      "learning_rate": 0.0002,
+      "loss": 1.6861,
+      "step": 2230
+    },
+    {
+      "epoch": 1.5008375209380236,
+      "grad_norm": 0.34607139229774475,
+      "learning_rate": 0.0002,
+      "loss": 1.6471,
+      "step": 2240
+    },
+    {
+      "epoch": 1.507537688442211,
+      "grad_norm": 0.3406706750392914,
+      "learning_rate": 0.0002,
+      "loss": 1.6962,
+      "step": 2250
+    },
+    {
+      "epoch": 1.5142378559463987,
+      "grad_norm": 0.36654895544052124,
+      "learning_rate": 0.0002,
+      "loss": 1.7694,
+      "step": 2260
+    },
+    {
+      "epoch": 1.5209380234505863,
+      "grad_norm": 0.3914054334163666,
+      "learning_rate": 0.0002,
+      "loss": 1.6812,
+      "step": 2270
+    },
+    {
+      "epoch": 1.5276381909547738,
+      "grad_norm": 0.42012137174606323,
+      "learning_rate": 0.0002,
+      "loss": 1.6822,
+      "step": 2280
+    },
+    {
+      "epoch": 1.5343383584589616,
+      "grad_norm": 0.39563435316085815,
+      "learning_rate": 0.0002,
+      "loss": 1.697,
+      "step": 2290
+    },
+    {
+      "epoch": 1.541038525963149,
+      "grad_norm": 0.3508438766002655,
+      "learning_rate": 0.0002,
+      "loss": 1.7491,
+      "step": 2300
+    },
+    {
+      "epoch": 1.5477386934673367,
+      "grad_norm": 0.3785218596458435,
+      "learning_rate": 0.0002,
+      "loss": 1.7727,
+      "step": 2310
+    },
+    {
+      "epoch": 1.5544388609715243,
+      "grad_norm": 0.39377647638320923,
+      "learning_rate": 0.0002,
+      "loss": 1.6963,
+      "step": 2320
+    },
+    {
+      "epoch": 1.5611390284757118,
+      "grad_norm": 0.3391438126564026,
+      "learning_rate": 0.0002,
+      "loss": 1.7263,
+      "step": 2330
+    },
+    {
+      "epoch": 1.5678391959798996,
+      "grad_norm": 0.37944263219833374,
+      "learning_rate": 0.0002,
+      "loss": 1.7722,
+      "step": 2340
+    },
+    {
+      "epoch": 1.574539363484087,
+      "grad_norm": 0.3523491322994232,
+      "learning_rate": 0.0002,
+      "loss": 1.6371,
+      "step": 2350
+    },
+    {
+      "epoch": 1.5812395309882747,
+      "grad_norm": 0.3911575973033905,
+      "learning_rate": 0.0002,
+      "loss": 1.7583,
+      "step": 2360
+    },
+    {
+      "epoch": 1.5879396984924623,
+      "grad_norm": 0.33832186460494995,
+      "learning_rate": 0.0002,
+      "loss": 1.7117,
+      "step": 2370
+    },
+    {
+      "epoch": 1.5946398659966499,
+      "grad_norm": 0.3665979206562042,
+      "learning_rate": 0.0002,
+      "loss": 1.7701,
+      "step": 2380
+    },
+    {
+      "epoch": 1.6013400335008376,
+      "grad_norm": 0.3871748149394989,
+      "learning_rate": 0.0002,
+      "loss": 1.779,
+      "step": 2390
+    },
+    {
+      "epoch": 1.608040201005025,
+      "grad_norm": 0.3586967885494232,
+      "learning_rate": 0.0002,
+      "loss": 1.7109,
+      "step": 2400
+    },
+    {
+      "epoch": 1.6147403685092128,
+      "grad_norm": 0.3563673198223114,
+      "learning_rate": 0.0002,
+      "loss": 1.7096,
+      "step": 2410
+    },
+    {
+      "epoch": 1.6214405360134003,
+      "grad_norm": 0.37588971853256226,
+      "learning_rate": 0.0002,
+      "loss": 1.745,
+      "step": 2420
+    },
+    {
+      "epoch": 1.6281407035175879,
+      "grad_norm": 0.352556437253952,
+      "learning_rate": 0.0002,
+      "loss": 1.7086,
+      "step": 2430
+    },
+    {
+      "epoch": 1.6348408710217757,
+      "grad_norm": 0.3716259300708771,
+      "learning_rate": 0.0002,
+      "loss": 1.6547,
+      "step": 2440
+    },
+    {
+      "epoch": 1.641541038525963,
+      "grad_norm": 0.372001975774765,
+      "learning_rate": 0.0002,
+      "loss": 1.7033,
+      "step": 2450
+    },
+    {
+      "epoch": 1.6482412060301508,
+      "grad_norm": 0.3430042862892151,
+      "learning_rate": 0.0002,
+      "loss": 1.6584,
+      "step": 2460
+    },
+    {
+      "epoch": 1.6549413735343383,
+      "grad_norm": 0.3741483688354492,
+      "learning_rate": 0.0002,
+      "loss": 1.7217,
+      "step": 2470
+    },
+    {
+      "epoch": 1.661641541038526,
+      "grad_norm": 0.3610571324825287,
+      "learning_rate": 0.0002,
+      "loss": 1.7701,
+      "step": 2480
+    },
+    {
+      "epoch": 1.6683417085427137,
+      "grad_norm": 0.4204719066619873,
+      "learning_rate": 0.0002,
+      "loss": 1.7057,
+      "step": 2490
+    },
+    {
+      "epoch": 1.675041876046901,
+      "grad_norm": 0.3938186466693878,
+      "learning_rate": 0.0002,
+      "loss": 1.7954,
+      "step": 2500
+    },
+    {
+      "epoch": 1.6817420435510888,
+      "grad_norm": 0.3421435058116913,
+      "learning_rate": 0.0002,
+      "loss": 1.6633,
+      "step": 2510
+    },
+    {
+      "epoch": 1.6884422110552764,
+      "grad_norm": 0.42441412806510925,
+      "learning_rate": 0.0002,
+      "loss": 1.7996,
+      "step": 2520
+    },
+    {
+      "epoch": 1.695142378559464,
+      "grad_norm": 0.38071519136428833,
+      "learning_rate": 0.0002,
+      "loss": 1.7142,
+      "step": 2530
+    },
+    {
+      "epoch": 1.7018425460636517,
+      "grad_norm": 0.34078919887542725,
+      "learning_rate": 0.0002,
+      "loss": 1.7232,
+      "step": 2540
+    },
+    {
+      "epoch": 1.708542713567839,
+      "grad_norm": 0.412844181060791,
+      "learning_rate": 0.0002,
+      "loss": 1.7126,
+      "step": 2550
+    },
+    {
+      "epoch": 1.7152428810720268,
+      "grad_norm": 0.3753604292869568,
+      "learning_rate": 0.0002,
+      "loss": 1.7149,
+      "step": 2560
+    },
+    {
+      "epoch": 1.7219430485762144,
+      "grad_norm": 0.41588476300239563,
+      "learning_rate": 0.0002,
+      "loss": 1.7011,
+      "step": 2570
+    },
+    {
+      "epoch": 1.728643216080402,
+      "grad_norm": 0.35504111647605896,
+      "learning_rate": 0.0002,
+      "loss": 1.6427,
+      "step": 2580
+    },
+    {
+      "epoch": 1.7353433835845897,
+      "grad_norm": 0.36909720301628113,
+      "learning_rate": 0.0002,
+      "loss": 1.7296,
+      "step": 2590
+    },
+    {
+      "epoch": 1.742043551088777,
+      "grad_norm": 0.4149979054927826,
+      "learning_rate": 0.0002,
+      "loss": 1.7022,
+      "step": 2600
+    },
+    {
+      "epoch": 1.7487437185929648,
+      "grad_norm": 0.38859328627586365,
+      "learning_rate": 0.0002,
+      "loss": 1.77,
+      "step": 2610
+    },
+    {
+      "epoch": 1.7554438860971524,
+      "grad_norm": 0.36738792061805725,
+      "learning_rate": 0.0002,
+      "loss": 1.7036,
+      "step": 2620
+    },
+    {
+      "epoch": 1.76214405360134,
+      "grad_norm": 0.3968178927898407,
+      "learning_rate": 0.0002,
+      "loss": 1.764,
+      "step": 2630
+    },
+    {
+      "epoch": 1.7688442211055277,
+      "grad_norm": 0.3972901999950409,
+      "learning_rate": 0.0002,
+      "loss": 1.7687,
+      "step": 2640
+    },
+    {
+      "epoch": 1.775544388609715,
+      "grad_norm": 0.3949959874153137,
+      "learning_rate": 0.0002,
+      "loss": 1.6774,
+      "step": 2650
+    },
+    {
+      "epoch": 1.7822445561139029,
+      "grad_norm": 0.44074657559394836,
+      "learning_rate": 0.0002,
+      "loss": 1.7247,
+      "step": 2660
+    },
+    {
+      "epoch": 1.7889447236180904,
+      "grad_norm": 0.39743664860725403,
+      "learning_rate": 0.0002,
+      "loss": 1.7188,
+      "step": 2670
+    },
+    {
+      "epoch": 1.795644891122278,
+      "grad_norm": 0.3950406610965729,
+      "learning_rate": 0.0002,
+      "loss": 1.7258,
+      "step": 2680
+    },
+    {
+      "epoch": 1.8023450586264658,
+      "grad_norm": 0.3568263649940491,
+      "learning_rate": 0.0002,
+      "loss": 1.6906,
+      "step": 2690
+    },
+    {
+      "epoch": 1.809045226130653,
+      "grad_norm": 0.3819476366043091,
+      "learning_rate": 0.0002,
+      "loss": 1.6735,
+      "step": 2700
+    },
+    {
+      "epoch": 1.8157453936348409,
+      "grad_norm": 0.3480634391307831,
+      "learning_rate": 0.0002,
+      "loss": 1.7198,
+      "step": 2710
+    },
+    {
+      "epoch": 1.8224455611390284,
+      "grad_norm": 0.3875853419303894,
+      "learning_rate": 0.0002,
+      "loss": 1.7042,
+      "step": 2720
+    },
+    {
+      "epoch": 1.829145728643216,
+      "grad_norm": 0.3441337049007416,
+      "learning_rate": 0.0002,
+      "loss": 1.6988,
+      "step": 2730
+    },
+    {
+      "epoch": 1.8358458961474038,
+      "grad_norm": 0.35692882537841797,
+      "learning_rate": 0.0002,
+      "loss": 1.7647,
+      "step": 2740
+    },
+    {
+      "epoch": 1.8425460636515911,
+      "grad_norm": 0.36959215998649597,
+      "learning_rate": 0.0002,
+      "loss": 1.7033,
+      "step": 2750
+    },
+    {
+      "epoch": 1.849246231155779,
+      "grad_norm": 0.3893393278121948,
+      "learning_rate": 0.0002,
+      "loss": 1.7657,
+      "step": 2760
+    },
+    {
+      "epoch": 1.8559463986599665,
+      "grad_norm": 0.37817293405532837,
+      "learning_rate": 0.0002,
+      "loss": 1.7068,
+      "step": 2770
+    },
+    {
+      "epoch": 1.862646566164154,
+      "grad_norm": 0.36071285605430603,
+      "learning_rate": 0.0002,
+      "loss": 1.761,
+      "step": 2780
+    },
+    {
+      "epoch": 1.8693467336683418,
+      "grad_norm": 0.3758420944213867,
+      "learning_rate": 0.0002,
+      "loss": 1.7623,
+      "step": 2790
+    },
+    {
+      "epoch": 1.8760469011725294,
+      "grad_norm": 0.3889938294887543,
+      "learning_rate": 0.0002,
+      "loss": 1.6743,
+      "step": 2800
+    },
+    {
+      "epoch": 1.882747068676717,
+      "grad_norm": 0.34361857175827026,
+      "learning_rate": 0.0002,
+      "loss": 1.6151,
+      "step": 2810
+    },
+    {
+      "epoch": 1.8894472361809045,
+      "grad_norm": 0.39283323287963867,
+      "learning_rate": 0.0002,
+      "loss": 1.6038,
+      "step": 2820
+    },
+    {
+      "epoch": 1.896147403685092,
+      "grad_norm": 0.3919452726840973,
+      "learning_rate": 0.0002,
+      "loss": 1.7555,
+      "step": 2830
+    },
+    {
+      "epoch": 1.9028475711892798,
+      "grad_norm": 0.38215070962905884,
+      "learning_rate": 0.0002,
+      "loss": 1.673,
+      "step": 2840
+    },
+    {
+      "epoch": 1.9095477386934674,
+      "grad_norm": 0.4235064387321472,
+      "learning_rate": 0.0002,
+      "loss": 1.7044,
+      "step": 2850
+    },
+    {
+      "epoch": 1.916247906197655,
+      "grad_norm": 0.35694634914398193,
+      "learning_rate": 0.0002,
+      "loss": 1.7123,
+      "step": 2860
+    },
+    {
+      "epoch": 1.9229480737018425,
+      "grad_norm": 0.383492112159729,
+      "learning_rate": 0.0002,
+      "loss": 1.8128,
+      "step": 2870
+    },
+    {
+      "epoch": 1.92964824120603,
+      "grad_norm": 0.5945147275924683,
+      "learning_rate": 0.0002,
+      "loss": 1.7581,
+      "step": 2880
+    },
+    {
+      "epoch": 1.9363484087102178,
+      "grad_norm": 0.3367522358894348,
+      "learning_rate": 0.0002,
+      "loss": 1.7421,
+      "step": 2890
+    },
+    {
+      "epoch": 1.9430485762144054,
+      "grad_norm": 0.35300394892692566,
+      "learning_rate": 0.0002,
+      "loss": 1.6561,
+      "step": 2900
+    },
+    {
+      "epoch": 1.949748743718593,
+      "grad_norm": 0.38084495067596436,
+      "learning_rate": 0.0002,
+      "loss": 1.7033,
+      "step": 2910
+    },
+    {
+      "epoch": 1.9564489112227805,
+      "grad_norm": 0.37559160590171814,
+      "learning_rate": 0.0002,
+      "loss": 1.7132,
+      "step": 2920
+    },
+    {
+      "epoch": 1.963149078726968,
+      "grad_norm": 0.3661738336086273,
+      "learning_rate": 0.0002,
+      "loss": 1.6759,
+      "step": 2930
+    },
+    {
+      "epoch": 1.9698492462311559,
+      "grad_norm": 0.4073849320411682,
+      "learning_rate": 0.0002,
+      "loss": 1.7643,
+      "step": 2940
+    },
+    {
+      "epoch": 1.9765494137353434,
+      "grad_norm": 0.3723304271697998,
+      "learning_rate": 0.0002,
+      "loss": 1.6806,
+      "step": 2950
+    },
+    {
+      "epoch": 1.983249581239531,
+      "grad_norm": 0.3991098999977112,
+      "learning_rate": 0.0002,
+      "loss": 1.7611,
+      "step": 2960
+    },
+    {
+      "epoch": 1.9899497487437185,
+      "grad_norm": 0.3947085440158844,
+      "learning_rate": 0.0002,
+      "loss": 1.7263,
+      "step": 2970
+    },
+    {
+      "epoch": 1.996649916247906,
+      "grad_norm": 0.3786258399486542,
+      "learning_rate": 0.0002,
+      "loss": 1.7217,
+      "step": 2980
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 1.8028968572616577,
+      "eval_runtime": 37.8985,
+      "eval_samples_per_second": 13.589,
+      "eval_steps_per_second": 1.715,
+      "step": 2985
+    },
+    {
+      "epoch": 2.003350083752094,
+      "grad_norm": 0.34824079275131226,
+      "learning_rate": 0.0002,
+      "loss": 1.695,
+      "step": 2990
+    },
+    {
+      "epoch": 2.0100502512562812,
+      "grad_norm": 0.3394894003868103,
+      "learning_rate": 0.0002,
+      "loss": 1.5853,
+      "step": 3000
+    },
+    {
+      "epoch": 2.016750418760469,
+      "grad_norm": 0.36910977959632874,
+      "learning_rate": 0.0002,
+      "loss": 1.5783,
+      "step": 3010
+    },
+    {
+      "epoch": 2.023450586264657,
+      "grad_norm": 0.45000967383384705,
+      "learning_rate": 0.0002,
+      "loss": 1.6105,
+      "step": 3020
+    },
+    {
+      "epoch": 2.030150753768844,
+      "grad_norm": 0.3791407346725464,
+      "learning_rate": 0.0002,
+      "loss": 1.6019,
+      "step": 3030
+    },
+    {
+      "epoch": 2.036850921273032,
+      "grad_norm": 0.387321799993515,
+      "learning_rate": 0.0002,
+      "loss": 1.5832,
+      "step": 3040
+    },
+    {
+      "epoch": 2.0435510887772192,
+      "grad_norm": 0.4185757040977478,
+      "learning_rate": 0.0002,
+      "loss": 1.6834,
+      "step": 3050
+    },
+    {
+      "epoch": 2.050251256281407,
+      "grad_norm": 0.45110777020454407,
+      "learning_rate": 0.0002,
+      "loss": 1.5696,
+      "step": 3060
+    },
+    {
+      "epoch": 2.056951423785595,
+      "grad_norm": 0.42663660645484924,
+      "learning_rate": 0.0002,
+      "loss": 1.6231,
+      "step": 3070
+    },
+    {
+      "epoch": 2.063651591289782,
+      "grad_norm": 0.4546292722225189,
+      "learning_rate": 0.0002,
+      "loss": 1.6279,
+      "step": 3080
+    },
+    {
+      "epoch": 2.07035175879397,
+      "grad_norm": 0.3979759216308594,
+      "learning_rate": 0.0002,
+      "loss": 1.6141,
+      "step": 3090
+    },
+    {
+      "epoch": 2.0770519262981573,
+      "grad_norm": 0.43596673011779785,
+      "learning_rate": 0.0002,
+      "loss": 1.6343,
+      "step": 3100
+    },
+    {
+      "epoch": 2.083752093802345,
+      "grad_norm": 0.40120232105255127,
+      "learning_rate": 0.0002,
+      "loss": 1.5441,
+      "step": 3110
+    },
+    {
+      "epoch": 2.090452261306533,
+      "grad_norm": 0.44449281692504883,
+      "learning_rate": 0.0002,
+      "loss": 1.6309,
+      "step": 3120
+    },
+    {
+      "epoch": 2.09715242881072,
+      "grad_norm": 0.42672568559646606,
+      "learning_rate": 0.0002,
+      "loss": 1.5652,
+      "step": 3130
+    },
+    {
+      "epoch": 2.103852596314908,
+      "grad_norm": 0.4232690930366516,
+      "learning_rate": 0.0002,
+      "loss": 1.682,
+      "step": 3140
+    },
+    {
+      "epoch": 2.1105527638190953,
+      "grad_norm": 0.4299317002296448,
+      "learning_rate": 0.0002,
+      "loss": 1.624,
+      "step": 3150
+    },
+    {
+      "epoch": 2.117252931323283,
+      "grad_norm": 0.4067758023738861,
+      "learning_rate": 0.0002,
+      "loss": 1.6766,
+      "step": 3160
+    },
+    {
+      "epoch": 2.123953098827471,
+      "grad_norm": 0.4918815791606903,
+      "learning_rate": 0.0002,
+      "loss": 1.6759,
+      "step": 3170
+    },
+    {
+      "epoch": 2.130653266331658,
+      "grad_norm": 0.4140559732913971,
+      "learning_rate": 0.0002,
+      "loss": 1.6478,
+      "step": 3180
+    },
+    {
+      "epoch": 2.137353433835846,
+      "grad_norm": 0.4555995464324951,
+      "learning_rate": 0.0002,
+      "loss": 1.6641,
+      "step": 3190
+    },
+    {
+      "epoch": 2.1440536013400333,
+      "grad_norm": 0.42943915724754333,
+      "learning_rate": 0.0002,
+      "loss": 1.5888,
+      "step": 3200
+    },
+    {
+      "epoch": 2.150753768844221,
+      "grad_norm": 0.4730435013771057,
+      "learning_rate": 0.0002,
+      "loss": 1.5886,
+      "step": 3210
+    },
+    {
+      "epoch": 2.157453936348409,
+      "grad_norm": 0.43310216069221497,
+      "learning_rate": 0.0002,
+      "loss": 1.6022,
+      "step": 3220
+    },
+    {
+      "epoch": 2.164154103852596,
+      "grad_norm": 0.42054110765457153,
+      "learning_rate": 0.0002,
+      "loss": 1.6058,
+      "step": 3230
+    },
+    {
+      "epoch": 2.170854271356784,
+      "grad_norm": 0.4897233247756958,
+      "learning_rate": 0.0002,
+      "loss": 1.6749,
+      "step": 3240
+    },
+    {
+      "epoch": 2.1775544388609713,
+      "grad_norm": 0.42194533348083496,
+      "learning_rate": 0.0002,
+      "loss": 1.6983,
+      "step": 3250
+    },
+    {
+      "epoch": 2.184254606365159,
+      "grad_norm": 0.44494450092315674,
+      "learning_rate": 0.0002,
+      "loss": 1.6159,
+      "step": 3260
+    },
+    {
+      "epoch": 2.190954773869347,
+      "grad_norm": 0.43524879217147827,
+      "learning_rate": 0.0002,
+      "loss": 1.6977,
+      "step": 3270
+    },
+    {
+      "epoch": 2.1976549413735342,
+      "grad_norm": 0.4621117413043976,
+      "learning_rate": 0.0002,
+      "loss": 1.528,
+      "step": 3280
+    },
+    {
+      "epoch": 2.204355108877722,
+      "grad_norm": 0.4073285460472107,
+      "learning_rate": 0.0002,
+      "loss": 1.632,
+      "step": 3290
+    },
+    {
+      "epoch": 2.2110552763819094,
+      "grad_norm": 0.47868335247039795,
+      "learning_rate": 0.0002,
+      "loss": 1.6141,
+      "step": 3300
+    },
+    {
+      "epoch": 2.217755443886097,
+      "grad_norm": 0.4264970123767853,
+      "learning_rate": 0.0002,
+      "loss": 1.6857,
+      "step": 3310
+    },
+    {
+      "epoch": 2.224455611390285,
+      "grad_norm": 0.4491245150566101,
+      "learning_rate": 0.0002,
+      "loss": 1.5653,
+      "step": 3320
+    },
+    {
+      "epoch": 2.2311557788944723,
+      "grad_norm": 0.4010344445705414,
+      "learning_rate": 0.0002,
+      "loss": 1.5881,
+      "step": 3330
+    },
+    {
+      "epoch": 2.23785594639866,
+      "grad_norm": 0.4232759177684784,
+      "learning_rate": 0.0002,
+      "loss": 1.6684,
+      "step": 3340
+    },
+    {
+      "epoch": 2.2445561139028474,
+      "grad_norm": 0.5099776983261108,
+      "learning_rate": 0.0002,
+      "loss": 1.6336,
+      "step": 3350
+    },
+    {
+      "epoch": 2.251256281407035,
+      "grad_norm": 0.5223407745361328,
+      "learning_rate": 0.0002,
+      "loss": 1.6764,
+      "step": 3360
+    },
+    {
+      "epoch": 2.257956448911223,
+      "grad_norm": 0.47818470001220703,
+      "learning_rate": 0.0002,
+      "loss": 1.6625,
+      "step": 3370
+    },
+    {
+      "epoch": 2.2646566164154103,
+      "grad_norm": 0.4721255898475647,
+      "learning_rate": 0.0002,
+      "loss": 1.5946,
+      "step": 3380
+    },
+    {
+      "epoch": 2.271356783919598,
+      "grad_norm": 0.4113229513168335,
+      "learning_rate": 0.0002,
+      "loss": 1.5568,
+      "step": 3390
+    },
+    {
+      "epoch": 2.2780569514237854,
+      "grad_norm": 0.507080078125,
+      "learning_rate": 0.0002,
+      "loss": 1.6494,
+      "step": 3400
+    },
+    {
+      "epoch": 2.284757118927973,
+      "grad_norm": 0.4852292239665985,
+      "learning_rate": 0.0002,
+      "loss": 1.6183,
+      "step": 3410
+    },
+    {
+      "epoch": 2.291457286432161,
+      "grad_norm": 0.4503684341907501,
+      "learning_rate": 0.0002,
+      "loss": 1.6132,
+      "step": 3420
+    },
+    {
+      "epoch": 2.2981574539363483,
+      "grad_norm": 0.8359600305557251,
+      "learning_rate": 0.0002,
+      "loss": 1.6649,
+      "step": 3430
+    },
+    {
+      "epoch": 2.304857621440536,
+      "grad_norm": 0.44604045152664185,
+      "learning_rate": 0.0002,
+      "loss": 1.6644,
+      "step": 3440
+    },
+    {
+      "epoch": 2.3115577889447234,
+      "grad_norm": 0.45667049288749695,
+      "learning_rate": 0.0002,
+      "loss": 1.5972,
+      "step": 3450
+    },
+    {
+      "epoch": 2.318257956448911,
+      "grad_norm": 0.4879349172115326,
+      "learning_rate": 0.0002,
+      "loss": 1.6667,
+      "step": 3460
+    },
+    {
+      "epoch": 2.324958123953099,
+      "grad_norm": 0.4033963084220886,
+      "learning_rate": 0.0002,
+      "loss": 1.5804,
+      "step": 3470
+    },
+    {
+      "epoch": 2.3316582914572863,
+      "grad_norm": 0.44494301080703735,
+      "learning_rate": 0.0002,
+      "loss": 1.5838,
+      "step": 3480
+    },
+    {
+      "epoch": 2.338358458961474,
+      "grad_norm": 0.4794621765613556,
+      "learning_rate": 0.0002,
+      "loss": 1.5951,
+      "step": 3490
+    },
+    {
+      "epoch": 2.3450586264656614,
+      "grad_norm": 0.41404327750205994,
+      "learning_rate": 0.0002,
+      "loss": 1.6807,
+      "step": 3500
+    },
+    {
+      "epoch": 2.351758793969849,
+      "grad_norm": 0.4664851725101471,
+      "learning_rate": 0.0002,
+      "loss": 1.714,
+      "step": 3510
+    },
+    {
+      "epoch": 2.358458961474037,
+      "grad_norm": 0.4263697564601898,
+      "learning_rate": 0.0002,
+      "loss": 1.6537,
+      "step": 3520
+    },
+    {
+      "epoch": 2.3651591289782243,
+      "grad_norm": 0.5035167336463928,
+      "learning_rate": 0.0002,
+      "loss": 1.6551,
+      "step": 3530
+    },
+    {
+      "epoch": 2.371859296482412,
+      "grad_norm": 0.4380664527416229,
+      "learning_rate": 0.0002,
+      "loss": 1.6208,
+      "step": 3540
+    },
+    {
+      "epoch": 2.3785594639865995,
+      "grad_norm": 0.5227681994438171,
+      "learning_rate": 0.0002,
+      "loss": 1.634,
+      "step": 3550
+    },
+    {
+      "epoch": 2.3852596314907872,
+      "grad_norm": 0.4382302761077881,
+      "learning_rate": 0.0002,
+      "loss": 1.6146,
+      "step": 3560
+    },
+    {
+      "epoch": 2.391959798994975,
+      "grad_norm": 0.4392451047897339,
+      "learning_rate": 0.0002,
+      "loss": 1.5653,
+      "step": 3570
+    },
+    {
+      "epoch": 2.3986599664991624,
+      "grad_norm": 0.4372786581516266,
+      "learning_rate": 0.0002,
+      "loss": 1.6626,
+      "step": 3580
+    },
+    {
+      "epoch": 2.40536013400335,
+      "grad_norm": 0.5015502572059631,
+      "learning_rate": 0.0002,
+      "loss": 1.519,
+      "step": 3590
+    },
+    {
+      "epoch": 2.4120603015075375,
+      "grad_norm": 0.5653210878372192,
+      "learning_rate": 0.0002,
+      "loss": 1.588,
+      "step": 3600
+    },
+    {
+      "epoch": 2.4187604690117253,
+      "grad_norm": 0.53007972240448,
+      "learning_rate": 0.0002,
+      "loss": 1.6075,
+      "step": 3610
+    },
+    {
+      "epoch": 2.425460636515913,
+      "grad_norm": 0.4659176766872406,
+      "learning_rate": 0.0002,
+      "loss": 1.6421,
+      "step": 3620
+    },
+    {
+      "epoch": 2.4321608040201004,
+      "grad_norm": 0.5637837052345276,
+      "learning_rate": 0.0002,
+      "loss": 1.625,
+      "step": 3630
+    },
+    {
+      "epoch": 2.438860971524288,
+      "grad_norm": 0.4248391389846802,
+      "learning_rate": 0.0002,
+      "loss": 1.6168,
+      "step": 3640
+    },
+    {
+      "epoch": 2.4455611390284755,
+      "grad_norm": 0.44668248295783997,
+      "learning_rate": 0.0002,
+      "loss": 1.6822,
+      "step": 3650
+    },
+    {
+      "epoch": 2.4522613065326633,
+      "grad_norm": 0.43990179896354675,
+      "learning_rate": 0.0002,
+      "loss": 1.6417,
+      "step": 3660
+    },
+    {
+      "epoch": 2.458961474036851,
+      "grad_norm": 0.4532523453235626,
+      "learning_rate": 0.0002,
+      "loss": 1.6723,
+      "step": 3670
+    },
+    {
+      "epoch": 2.4656616415410384,
+      "grad_norm": 0.6605591773986816,
+      "learning_rate": 0.0002,
+      "loss": 1.6957,
+      "step": 3680
+    },
+    {
+      "epoch": 2.472361809045226,
+      "grad_norm": 0.4694533348083496,
+      "learning_rate": 0.0002,
+      "loss": 1.6159,
+      "step": 3690
+    },
+    {
+      "epoch": 2.4790619765494135,
+      "grad_norm": 0.4485011100769043,
+      "learning_rate": 0.0002,
+      "loss": 1.6239,
+      "step": 3700
+    },
+    {
+      "epoch": 2.4857621440536013,
+      "grad_norm": 0.4761785864830017,
+      "learning_rate": 0.0002,
+      "loss": 1.6834,
+      "step": 3710
+    },
+    {
+      "epoch": 2.492462311557789,
+      "grad_norm": 0.5116432309150696,
+      "learning_rate": 0.0002,
+      "loss": 1.6313,
+      "step": 3720
+    },
+    {
+      "epoch": 2.4991624790619764,
+      "grad_norm": 0.49523618817329407,
+      "learning_rate": 0.0002,
+      "loss": 1.5054,
+      "step": 3730
+    },
+    {
+      "epoch": 2.505862646566164,
+      "grad_norm": 0.43826380372047424,
+      "learning_rate": 0.0002,
+      "loss": 1.6249,
+      "step": 3740
+    },
+    {
+      "epoch": 2.5125628140703515,
+      "grad_norm": 0.4916154146194458,
+      "learning_rate": 0.0002,
+      "loss": 1.5762,
+      "step": 3750
+    },
+    {
+      "epoch": 2.5192629815745393,
+      "grad_norm": 0.5381299257278442,
+      "learning_rate": 0.0002,
+      "loss": 1.5157,
+      "step": 3760
+    },
+    {
+      "epoch": 2.525963149078727,
+      "grad_norm": 0.44947415590286255,
+      "learning_rate": 0.0002,
+      "loss": 1.6467,
+      "step": 3770
+    },
+    {
+      "epoch": 2.5326633165829144,
+      "grad_norm": 0.49979084730148315,
+      "learning_rate": 0.0002,
+      "loss": 1.67,
+      "step": 3780
+    },
+    {
+      "epoch": 2.539363484087102,
+      "grad_norm": 0.43046900629997253,
+      "learning_rate": 0.0002,
+      "loss": 1.622,
+      "step": 3790
+    },
+    {
+      "epoch": 2.5460636515912896,
+      "grad_norm": 0.4513470530509949,
+      "learning_rate": 0.0002,
+      "loss": 1.6789,
+      "step": 3800
+    },
+    {
+      "epoch": 2.5527638190954773,
+      "grad_norm": 0.49900051951408386,
+      "learning_rate": 0.0002,
+      "loss": 1.6335,
+      "step": 3810
+    },
+    {
+      "epoch": 2.559463986599665,
+      "grad_norm": 0.4348420202732086,
+      "learning_rate": 0.0002,
+      "loss": 1.6322,
+      "step": 3820
+    },
+    {
+      "epoch": 2.5661641541038525,
+      "grad_norm": 0.4684867560863495,
+      "learning_rate": 0.0002,
+      "loss": 1.6218,
+      "step": 3830
+    },
+    {
+      "epoch": 2.5728643216080402,
+      "grad_norm": 0.44430989027023315,
+      "learning_rate": 0.0002,
+      "loss": 1.6535,
+      "step": 3840
+    },
+    {
+      "epoch": 2.5795644891122276,
+      "grad_norm": 0.47375255823135376,
+      "learning_rate": 0.0002,
+      "loss": 1.5909,
+      "step": 3850
+    },
+    {
+      "epoch": 2.5862646566164154,
+      "grad_norm": 0.45493075251579285,
+      "learning_rate": 0.0002,
+      "loss": 1.6269,
+      "step": 3860
+    },
+    {
+      "epoch": 2.592964824120603,
+      "grad_norm": 0.4563275873661041,
+      "learning_rate": 0.0002,
+      "loss": 1.604,
+      "step": 3870
+    },
+    {
+      "epoch": 2.5996649916247905,
+      "grad_norm": 0.46060335636138916,
+      "learning_rate": 0.0002,
+      "loss": 1.642,
+      "step": 3880
+    },
+    {
+      "epoch": 2.6063651591289783,
+      "grad_norm": 0.4718867540359497,
+      "learning_rate": 0.0002,
+      "loss": 1.6302,
+      "step": 3890
+    },
+    {
+      "epoch": 2.6130653266331656,
+      "grad_norm": 0.41570305824279785,
+      "learning_rate": 0.0002,
+      "loss": 1.6242,
+      "step": 3900
+    },
+    {
+      "epoch": 2.6197654941373534,
+      "grad_norm": 0.4603121876716614,
+      "learning_rate": 0.0002,
+      "loss": 1.6401,
+      "step": 3910
+    },
+    {
+      "epoch": 2.626465661641541,
+      "grad_norm": 0.4734652638435364,
+      "learning_rate": 0.0002,
+      "loss": 1.6839,
+      "step": 3920
+    },
+    {
+      "epoch": 2.6331658291457285,
+      "grad_norm": 0.45348483324050903,
+      "learning_rate": 0.0002,
+      "loss": 1.5448,
+      "step": 3930
+    },
+    {
+      "epoch": 2.6398659966499163,
+      "grad_norm": 0.46559447050094604,
+      "learning_rate": 0.0002,
+      "loss": 1.6157,
+      "step": 3940
+    },
+    {
+      "epoch": 2.6465661641541036,
+      "grad_norm": 0.44113144278526306,
+      "learning_rate": 0.0002,
+      "loss": 1.7052,
+      "step": 3950
+    },
+    {
+      "epoch": 2.6532663316582914,
+      "grad_norm": 0.41415104269981384,
+      "learning_rate": 0.0002,
+      "loss": 1.6315,
+      "step": 3960
+    },
+    {
+      "epoch": 2.659966499162479,
+      "grad_norm": 0.48868080973625183,
+      "learning_rate": 0.0002,
+      "loss": 1.6589,
+      "step": 3970
+    },
+    {
+      "epoch": 2.6666666666666665,
+      "grad_norm": 0.49610549211502075,
+      "learning_rate": 0.0002,
+      "loss": 1.6211,
+      "step": 3980
+    },
+    {
+      "epoch": 2.6733668341708543,
+      "grad_norm": 0.4309130907058716,
+      "learning_rate": 0.0002,
+      "loss": 1.6235,
+      "step": 3990
+    },
+    {
+      "epoch": 2.6800670016750416,
+      "grad_norm": 0.4489327669143677,
+      "learning_rate": 0.0002,
+      "loss": 1.6452,
+      "step": 4000
+    },
+    {
+      "epoch": 2.6867671691792294,
+      "grad_norm": 0.5380139946937561,
+      "learning_rate": 0.0002,
+      "loss": 1.5744,
+      "step": 4010
+    },
+    {
+      "epoch": 2.693467336683417,
+      "grad_norm": 0.5076672434806824,
+      "learning_rate": 0.0002,
+      "loss": 1.6524,
+      "step": 4020
+    },
+    {
+      "epoch": 2.7001675041876045,
+      "grad_norm": 0.47620031237602234,
+      "learning_rate": 0.0002,
+      "loss": 1.636,
+      "step": 4030
+    },
+    {
+      "epoch": 2.7068676716917923,
+      "grad_norm": 0.48089155554771423,
+      "learning_rate": 0.0002,
+      "loss": 1.5543,
+      "step": 4040
+    },
+    {
+      "epoch": 2.7135678391959797,
+      "grad_norm": 0.5108814239501953,
+      "learning_rate": 0.0002,
+      "loss": 1.6396,
+      "step": 4050
+    },
+    {
+      "epoch": 2.7202680067001674,
+      "grad_norm": 0.4196513295173645,
+      "learning_rate": 0.0002,
+      "loss": 1.5905,
+      "step": 4060
+    },
+    {
+      "epoch": 2.726968174204355,
+      "grad_norm": 0.4574664831161499,
+      "learning_rate": 0.0002,
+      "loss": 1.686,
+      "step": 4070
+    },
+    {
+      "epoch": 2.7336683417085426,
+      "grad_norm": 0.4671640992164612,
+      "learning_rate": 0.0002,
+      "loss": 1.6234,
+      "step": 4080
+    },
+    {
+      "epoch": 2.7403685092127303,
+      "grad_norm": 0.49355530738830566,
+      "learning_rate": 0.0002,
+      "loss": 1.6827,
+      "step": 4090
+    },
+    {
+      "epoch": 2.7470686767169177,
+      "grad_norm": 0.46716663241386414,
+      "learning_rate": 0.0002,
+      "loss": 1.6999,
+      "step": 4100
+    },
+    {
+      "epoch": 2.7537688442211055,
+      "grad_norm": 0.45420581102371216,
+      "learning_rate": 0.0002,
+      "loss": 1.6463,
+      "step": 4110
+    },
+    {
+      "epoch": 2.7604690117252932,
+      "grad_norm": 0.4680487811565399,
+      "learning_rate": 0.0002,
+      "loss": 1.5718,
+      "step": 4120
+    },
+    {
+      "epoch": 2.7671691792294806,
+      "grad_norm": 0.5375032424926758,
+      "learning_rate": 0.0002,
+      "loss": 1.5968,
+      "step": 4130
+    },
+    {
+      "epoch": 2.7738693467336684,
+      "grad_norm": 0.46026280522346497,
+      "learning_rate": 0.0002,
+      "loss": 1.5254,
+      "step": 4140
+    },
+    {
+      "epoch": 2.7805695142378557,
+      "grad_norm": 0.43658447265625,
+      "learning_rate": 0.0002,
+      "loss": 1.6613,
+      "step": 4150
+    },
+    {
+      "epoch": 2.7872696817420435,
+      "grad_norm": 0.4935547113418579,
+      "learning_rate": 0.0002,
+      "loss": 1.6546,
+      "step": 4160
+    },
+    {
+      "epoch": 2.7939698492462313,
+      "grad_norm": 0.8167962431907654,
+      "learning_rate": 0.0002,
+      "loss": 1.5961,
+      "step": 4170
+    },
+    {
+      "epoch": 2.8006700167504186,
+      "grad_norm": 0.4289683997631073,
+      "learning_rate": 0.0002,
+      "loss": 1.6907,
+      "step": 4180
+    },
+    {
+      "epoch": 2.8073701842546064,
+      "grad_norm": 0.4569324254989624,
+      "learning_rate": 0.0002,
+      "loss": 1.6385,
+      "step": 4190
+    },
+    {
+      "epoch": 2.8140703517587937,
+      "grad_norm": 0.474795937538147,
+      "learning_rate": 0.0002,
+      "loss": 1.6077,
+      "step": 4200
+    },
+    {
+      "epoch": 2.8207705192629815,
+      "grad_norm": 0.44272229075431824,
+      "learning_rate": 0.0002,
+      "loss": 1.6223,
+      "step": 4210
+    },
+    {
+      "epoch": 2.8274706867671693,
+      "grad_norm": 0.525240957736969,
+      "learning_rate": 0.0002,
+      "loss": 1.6706,
+      "step": 4220
+    },
+    {
+      "epoch": 2.8341708542713566,
+      "grad_norm": 0.4802303910255432,
+      "learning_rate": 0.0002,
+      "loss": 1.7196,
+      "step": 4230
+    },
+    {
+      "epoch": 2.8408710217755444,
+      "grad_norm": 0.46400442719459534,
+      "learning_rate": 0.0002,
+      "loss": 1.6002,
+      "step": 4240
+    },
+    {
+      "epoch": 2.8475711892797317,
+      "grad_norm": 0.49884888529777527,
+      "learning_rate": 0.0002,
+      "loss": 1.6052,
+      "step": 4250
+    },
+    {
+      "epoch": 2.8542713567839195,
+      "grad_norm": 0.5015072226524353,
+      "learning_rate": 0.0002,
+      "loss": 1.6919,
+      "step": 4260
+    },
+    {
+      "epoch": 2.8609715242881073,
+      "grad_norm": 0.4335440695285797,
+      "learning_rate": 0.0002,
+      "loss": 1.6335,
+      "step": 4270
+    },
+    {
+      "epoch": 2.8676716917922946,
+      "grad_norm": 0.5131644606590271,
+      "learning_rate": 0.0002,
+      "loss": 1.5664,
+      "step": 4280
+    },
+    {
+      "epoch": 2.8743718592964824,
+      "grad_norm": 0.6977195739746094,
+      "learning_rate": 0.0002,
+      "loss": 1.6409,
+      "step": 4290
+    },
+    {
+      "epoch": 2.8810720268006698,
+      "grad_norm": 0.5133762955665588,
+      "learning_rate": 0.0002,
+      "loss": 1.7192,
+      "step": 4300
+    },
+    {
+      "epoch": 2.8877721943048575,
+      "grad_norm": 0.4737614393234253,
+      "learning_rate": 0.0002,
+      "loss": 1.6257,
+      "step": 4310
+    },
+    {
+      "epoch": 2.8944723618090453,
+      "grad_norm": 0.4580535590648651,
+      "learning_rate": 0.0002,
+      "loss": 1.6076,
+      "step": 4320
+    },
+    {
+      "epoch": 2.901172529313233,
+      "grad_norm": 0.43863341212272644,
+      "learning_rate": 0.0002,
+      "loss": 1.6538,
+      "step": 4330
+    },
+    {
+      "epoch": 2.9078726968174204,
+      "grad_norm": 0.4103737473487854,
+      "learning_rate": 0.0002,
+      "loss": 1.6091,
+      "step": 4340
+    },
+    {
+      "epoch": 2.914572864321608,
+      "grad_norm": 0.438014417886734,
+      "learning_rate": 0.0002,
+      "loss": 1.7106,
+      "step": 4350
+    },
+    {
+      "epoch": 2.9212730318257956,
+      "grad_norm": 0.5068213939666748,
+      "learning_rate": 0.0002,
+      "loss": 1.6025,
+      "step": 4360
+    },
+    {
+      "epoch": 2.9279731993299833,
+      "grad_norm": 0.45305484533309937,
+      "learning_rate": 0.0002,
+      "loss": 1.6426,
+      "step": 4370
+    },
+    {
+      "epoch": 2.934673366834171,
+      "grad_norm": 0.4612090289592743,
+      "learning_rate": 0.0002,
+      "loss": 1.5726,
+      "step": 4380
+    },
+    {
+      "epoch": 2.9413735343383585,
+      "grad_norm": 0.508736789226532,
+      "learning_rate": 0.0002,
+      "loss": 1.6536,
+      "step": 4390
+    },
+    {
+      "epoch": 2.948073701842546,
+      "grad_norm": 0.4924427270889282,
+      "learning_rate": 0.0002,
+      "loss": 1.6132,
+      "step": 4400
+    },
+    {
+      "epoch": 2.9547738693467336,
+      "grad_norm": 0.5707460641860962,
+      "learning_rate": 0.0002,
+      "loss": 1.7007,
+      "step": 4410
+    },
+    {
+      "epoch": 2.9614740368509214,
+      "grad_norm": 0.42270299792289734,
+      "learning_rate": 0.0002,
+      "loss": 1.6814,
+      "step": 4420
+    },
+    {
+      "epoch": 2.968174204355109,
+      "grad_norm": 0.4429931044578552,
+      "learning_rate": 0.0002,
+      "loss": 1.6644,
+      "step": 4430
+    },
+    {
+      "epoch": 2.9748743718592965,
+      "grad_norm": 0.49760574102401733,
+      "learning_rate": 0.0002,
+      "loss": 1.6251,
+      "step": 4440
+    },
+    {
+      "epoch": 2.981574539363484,
+      "grad_norm": 0.4558229148387909,
+      "learning_rate": 0.0002,
+      "loss": 1.6169,
+      "step": 4450
+    },
+    {
+      "epoch": 2.9882747068676716,
+      "grad_norm": 0.39848530292510986,
+      "learning_rate": 0.0002,
+      "loss": 1.6055,
+      "step": 4460
+    },
+    {
+      "epoch": 2.9949748743718594,
+      "grad_norm": 0.5224862098693848,
+      "learning_rate": 0.0002,
+      "loss": 1.6705,
+      "step": 4470
+    },
+    {
+      "epoch": 2.9996649916247904,
+      "eval_loss": 1.8228833675384521,
+      "eval_runtime": 37.9049,
+      "eval_samples_per_second": 13.587,
+      "eval_steps_per_second": 1.715,
+      "step": 4477
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 11936,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 8,
+  "save_steps": 200,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 2.0720867447144448e+17,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-4477/training_args.bin b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-4477/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..064f299b0f3d2a28f5b1f5c68ef32caab3e2dd49
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-4477/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aa7b586fc440d1e22156593e2dd4267d2bdcb8920a02fdf352ea29a9bec3dd94
+size 5560
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-5970/README.md b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-5970/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..835e31ab8469ee39ddc8b2b6b2143a8c66dad510
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-5970/README.md
@@ -0,0 +1,202 @@
+---
+base_model: meta-llama/Meta-Llama-3-8B-Instruct
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.13.1
\ No newline at end of file
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-5970/adapter_config.json b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-5970/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..406c5a08dc4a2a33b52c62a482f98c217c417215
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-5970/adapter_config.json
@@ -0,0 +1,29 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "meta-llama/Meta-Llama-3-8B-Instruct",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-5970/adapter_model.safetensors b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-5970/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..187483575890af464d9fc34a5c6041a1d913c030
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-5970/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:09843f37ac51f23607d4fc013c537ae32d4a4610a3e68536ef0c6fd0de56067f
+size 109069176
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-5970/optimizer.pt b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-5970/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..bae881a0c8a101634bb2e8482068e7815fd9da88
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-5970/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bce883ace8bdd886f8b3a5dcf93aa4a8675f7bd47a446c777aeecb5881fad069
+size 55532666
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-5970/rng_state.pth b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-5970/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..943a6bd2ee8cbed6782eaa5a484edf6695650762
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-5970/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:48a4e09a09c5f85de9778f65259b862561985aa5a2edaf77c99c39901e891ff3
+size 14244
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-5970/scheduler.pt b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-5970/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..5dfc75f10e9f51117e5f561a5d0b15bda67213a3
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-5970/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:001d71f239fa6dc6a8238c31d4375718af287b278554a1c1582559d2cad2bb41
+size 1064
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-5970/special_tokens_map.json b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-5970/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..b43be96621d147110fb8a18b5776ec6e38516127
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-5970/special_tokens_map.json
@@ -0,0 +1,17 @@
+{
+  "bos_token": {
+    "content": "<|begin_of_text|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<|eot_id|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<|eot_id|>"
+}
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-5970/tokenizer.json b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-5970/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..86a33946b0c77216d2cce91bb28c8fada4a5e80b
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-5970/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3c5cf44023714fb39b05e71e425f8d7b92805ff73f7988b083b8c87f0bf87393
+size 17209961
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-5970/tokenizer_config.json b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-5970/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..061e40d9db3253624f86e8e364c15ef546527c9d
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-5970/tokenizer_config.json
@@ -0,0 +1,2063 @@
+{
+  "added_tokens_decoder": {
+    "128000": {
+      "content": "<|begin_of_text|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128001": {
+      "content": "<|end_of_text|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128002": {
+      "content": "<|reserved_special_token_0|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128003": {
+      "content": "<|reserved_special_token_1|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128004": {
+      "content": "<|reserved_special_token_2|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128005": {
+      "content": "<|reserved_special_token_3|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128006": {
+      "content": "<|start_header_id|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128007": {
+      "content": "<|end_header_id|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128008": {
+      "content": "<|reserved_special_token_4|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128009": {
+      "content": "<|eot_id|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128010": {
+      "content": "<|reserved_special_token_5|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128011": {
+      "content": "<|reserved_special_token_6|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128012": {
+      "content": "<|reserved_special_token_7|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128013": {
+      "content": "<|reserved_special_token_8|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128014": {
+      "content": "<|reserved_special_token_9|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128015": {
+      "content": "<|reserved_special_token_10|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128016": {
+      "content": "<|reserved_special_token_11|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128017": {
+      "content": "<|reserved_special_token_12|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128018": {
+      "content": "<|reserved_special_token_13|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128019": {
+      "content": "<|reserved_special_token_14|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128020": {
+      "content": "<|reserved_special_token_15|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128021": {
+      "content": "<|reserved_special_token_16|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128022": {
+      "content": "<|reserved_special_token_17|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128023": {
+      "content": "<|reserved_special_token_18|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128024": {
+      "content": "<|reserved_special_token_19|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128025": {
+      "content": "<|reserved_special_token_20|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128026": {
+      "content": "<|reserved_special_token_21|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128027": {
+      "content": "<|reserved_special_token_22|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128028": {
+      "content": "<|reserved_special_token_23|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128029": {
+      "content": "<|reserved_special_token_24|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128030": {
+      "content": "<|reserved_special_token_25|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128031": {
+      "content": "<|reserved_special_token_26|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128032": {
+      "content": "<|reserved_special_token_27|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128033": {
+      "content": "<|reserved_special_token_28|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128034": {
+      "content": "<|reserved_special_token_29|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128035": {
+      "content": "<|reserved_special_token_30|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128036": {
+      "content": "<|reserved_special_token_31|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128037": {
+      "content": "<|reserved_special_token_32|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128038": {
+      "content": "<|reserved_special_token_33|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128039": {
+      "content": "<|reserved_special_token_34|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128040": {
+      "content": "<|reserved_special_token_35|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128041": {
+      "content": "<|reserved_special_token_36|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128042": {
+      "content": "<|reserved_special_token_37|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128043": {
+      "content": "<|reserved_special_token_38|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128044": {
+      "content": "<|reserved_special_token_39|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128045": {
+      "content": "<|reserved_special_token_40|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128046": {
+      "content": "<|reserved_special_token_41|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128047": {
+      "content": "<|reserved_special_token_42|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128048": {
+      "content": "<|reserved_special_token_43|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128049": {
+      "content": "<|reserved_special_token_44|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128050": {
+      "content": "<|reserved_special_token_45|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128051": {
+      "content": "<|reserved_special_token_46|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128052": {
+      "content": "<|reserved_special_token_47|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128053": {
+      "content": "<|reserved_special_token_48|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128054": {
+      "content": "<|reserved_special_token_49|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128055": {
+      "content": "<|reserved_special_token_50|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128056": {
+      "content": "<|reserved_special_token_51|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128057": {
+      "content": "<|reserved_special_token_52|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128058": {
+      "content": "<|reserved_special_token_53|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128059": {
+      "content": "<|reserved_special_token_54|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128060": {
+      "content": "<|reserved_special_token_55|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128061": {
+      "content": "<|reserved_special_token_56|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128062": {
+      "content": "<|reserved_special_token_57|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128063": {
+      "content": "<|reserved_special_token_58|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128064": {
+      "content": "<|reserved_special_token_59|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128065": {
+      "content": "<|reserved_special_token_60|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128066": {
+      "content": "<|reserved_special_token_61|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128067": {
+      "content": "<|reserved_special_token_62|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128068": {
+      "content": "<|reserved_special_token_63|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128069": {
+      "content": "<|reserved_special_token_64|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128070": {
+      "content": "<|reserved_special_token_65|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128071": {
+      "content": "<|reserved_special_token_66|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128072": {
+      "content": "<|reserved_special_token_67|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128073": {
+      "content": "<|reserved_special_token_68|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128074": {
+      "content": "<|reserved_special_token_69|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128075": {
+      "content": "<|reserved_special_token_70|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128076": {
+      "content": "<|reserved_special_token_71|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128077": {
+      "content": "<|reserved_special_token_72|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128078": {
+      "content": "<|reserved_special_token_73|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128079": {
+      "content": "<|reserved_special_token_74|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128080": {
+      "content": "<|reserved_special_token_75|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128081": {
+      "content": "<|reserved_special_token_76|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128082": {
+      "content": "<|reserved_special_token_77|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128083": {
+      "content": "<|reserved_special_token_78|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128084": {
+      "content": "<|reserved_special_token_79|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128085": {
+      "content": "<|reserved_special_token_80|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128086": {
+      "content": "<|reserved_special_token_81|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128087": {
+      "content": "<|reserved_special_token_82|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128088": {
+      "content": "<|reserved_special_token_83|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128089": {
+      "content": "<|reserved_special_token_84|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128090": {
+      "content": "<|reserved_special_token_85|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128091": {
+      "content": "<|reserved_special_token_86|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128092": {
+      "content": "<|reserved_special_token_87|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128093": {
+      "content": "<|reserved_special_token_88|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128094": {
+      "content": "<|reserved_special_token_89|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128095": {
+      "content": "<|reserved_special_token_90|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128096": {
+      "content": "<|reserved_special_token_91|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128097": {
+      "content": "<|reserved_special_token_92|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128098": {
+      "content": "<|reserved_special_token_93|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128099": {
+      "content": "<|reserved_special_token_94|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128100": {
+      "content": "<|reserved_special_token_95|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128101": {
+      "content": "<|reserved_special_token_96|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128102": {
+      "content": "<|reserved_special_token_97|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128103": {
+      "content": "<|reserved_special_token_98|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128104": {
+      "content": "<|reserved_special_token_99|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128105": {
+      "content": "<|reserved_special_token_100|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128106": {
+      "content": "<|reserved_special_token_101|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128107": {
+      "content": "<|reserved_special_token_102|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128108": {
+      "content": "<|reserved_special_token_103|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128109": {
+      "content": "<|reserved_special_token_104|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128110": {
+      "content": "<|reserved_special_token_105|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128111": {
+      "content": "<|reserved_special_token_106|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128112": {
+      "content": "<|reserved_special_token_107|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128113": {
+      "content": "<|reserved_special_token_108|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128114": {
+      "content": "<|reserved_special_token_109|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128115": {
+      "content": "<|reserved_special_token_110|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128116": {
+      "content": "<|reserved_special_token_111|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128117": {
+      "content": "<|reserved_special_token_112|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128118": {
+      "content": "<|reserved_special_token_113|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128119": {
+      "content": "<|reserved_special_token_114|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128120": {
+      "content": "<|reserved_special_token_115|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128121": {
+      "content": "<|reserved_special_token_116|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128122": {
+      "content": "<|reserved_special_token_117|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128123": {
+      "content": "<|reserved_special_token_118|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128124": {
+      "content": "<|reserved_special_token_119|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128125": {
+      "content": "<|reserved_special_token_120|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128126": {
+      "content": "<|reserved_special_token_121|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128127": {
+      "content": "<|reserved_special_token_122|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128128": {
+      "content": "<|reserved_special_token_123|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128129": {
+      "content": "<|reserved_special_token_124|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128130": {
+      "content": "<|reserved_special_token_125|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128131": {
+      "content": "<|reserved_special_token_126|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128132": {
+      "content": "<|reserved_special_token_127|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128133": {
+      "content": "<|reserved_special_token_128|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128134": {
+      "content": "<|reserved_special_token_129|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128135": {
+      "content": "<|reserved_special_token_130|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128136": {
+      "content": "<|reserved_special_token_131|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128137": {
+      "content": "<|reserved_special_token_132|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128138": {
+      "content": "<|reserved_special_token_133|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128139": {
+      "content": "<|reserved_special_token_134|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128140": {
+      "content": "<|reserved_special_token_135|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128141": {
+      "content": "<|reserved_special_token_136|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128142": {
+      "content": "<|reserved_special_token_137|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128143": {
+      "content": "<|reserved_special_token_138|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128144": {
+      "content": "<|reserved_special_token_139|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128145": {
+      "content": "<|reserved_special_token_140|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128146": {
+      "content": "<|reserved_special_token_141|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128147": {
+      "content": "<|reserved_special_token_142|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128148": {
+      "content": "<|reserved_special_token_143|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128149": {
+      "content": "<|reserved_special_token_144|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128150": {
+      "content": "<|reserved_special_token_145|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128151": {
+      "content": "<|reserved_special_token_146|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128152": {
+      "content": "<|reserved_special_token_147|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128153": {
+      "content": "<|reserved_special_token_148|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128154": {
+      "content": "<|reserved_special_token_149|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128155": {
+      "content": "<|reserved_special_token_150|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128156": {
+      "content": "<|reserved_special_token_151|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128157": {
+      "content": "<|reserved_special_token_152|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128158": {
+      "content": "<|reserved_special_token_153|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128159": {
+      "content": "<|reserved_special_token_154|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128160": {
+      "content": "<|reserved_special_token_155|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128161": {
+      "content": "<|reserved_special_token_156|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128162": {
+      "content": "<|reserved_special_token_157|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128163": {
+      "content": "<|reserved_special_token_158|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128164": {
+      "content": "<|reserved_special_token_159|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128165": {
+      "content": "<|reserved_special_token_160|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128166": {
+      "content": "<|reserved_special_token_161|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128167": {
+      "content": "<|reserved_special_token_162|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128168": {
+      "content": "<|reserved_special_token_163|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128169": {
+      "content": "<|reserved_special_token_164|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128170": {
+      "content": "<|reserved_special_token_165|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128171": {
+      "content": "<|reserved_special_token_166|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128172": {
+      "content": "<|reserved_special_token_167|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128173": {
+      "content": "<|reserved_special_token_168|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128174": {
+      "content": "<|reserved_special_token_169|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128175": {
+      "content": "<|reserved_special_token_170|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128176": {
+      "content": "<|reserved_special_token_171|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128177": {
+      "content": "<|reserved_special_token_172|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128178": {
+      "content": "<|reserved_special_token_173|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128179": {
+      "content": "<|reserved_special_token_174|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128180": {
+      "content": "<|reserved_special_token_175|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128181": {
+      "content": "<|reserved_special_token_176|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128182": {
+      "content": "<|reserved_special_token_177|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128183": {
+      "content": "<|reserved_special_token_178|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128184": {
+      "content": "<|reserved_special_token_179|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128185": {
+      "content": "<|reserved_special_token_180|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128186": {
+      "content": "<|reserved_special_token_181|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128187": {
+      "content": "<|reserved_special_token_182|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128188": {
+      "content": "<|reserved_special_token_183|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128189": {
+      "content": "<|reserved_special_token_184|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128190": {
+      "content": "<|reserved_special_token_185|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128191": {
+      "content": "<|reserved_special_token_186|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128192": {
+      "content": "<|reserved_special_token_187|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128193": {
+      "content": "<|reserved_special_token_188|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128194": {
+      "content": "<|reserved_special_token_189|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128195": {
+      "content": "<|reserved_special_token_190|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128196": {
+      "content": "<|reserved_special_token_191|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128197": {
+      "content": "<|reserved_special_token_192|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128198": {
+      "content": "<|reserved_special_token_193|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128199": {
+      "content": "<|reserved_special_token_194|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128200": {
+      "content": "<|reserved_special_token_195|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128201": {
+      "content": "<|reserved_special_token_196|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128202": {
+      "content": "<|reserved_special_token_197|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128203": {
+      "content": "<|reserved_special_token_198|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128204": {
+      "content": "<|reserved_special_token_199|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128205": {
+      "content": "<|reserved_special_token_200|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128206": {
+      "content": "<|reserved_special_token_201|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128207": {
+      "content": "<|reserved_special_token_202|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128208": {
+      "content": "<|reserved_special_token_203|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128209": {
+      "content": "<|reserved_special_token_204|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128210": {
+      "content": "<|reserved_special_token_205|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128211": {
+      "content": "<|reserved_special_token_206|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128212": {
+      "content": "<|reserved_special_token_207|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128213": {
+      "content": "<|reserved_special_token_208|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128214": {
+      "content": "<|reserved_special_token_209|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128215": {
+      "content": "<|reserved_special_token_210|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128216": {
+      "content": "<|reserved_special_token_211|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128217": {
+      "content": "<|reserved_special_token_212|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128218": {
+      "content": "<|reserved_special_token_213|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128219": {
+      "content": "<|reserved_special_token_214|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128220": {
+      "content": "<|reserved_special_token_215|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128221": {
+      "content": "<|reserved_special_token_216|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128222": {
+      "content": "<|reserved_special_token_217|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128223": {
+      "content": "<|reserved_special_token_218|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128224": {
+      "content": "<|reserved_special_token_219|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128225": {
+      "content": "<|reserved_special_token_220|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128226": {
+      "content": "<|reserved_special_token_221|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128227": {
+      "content": "<|reserved_special_token_222|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128228": {
+      "content": "<|reserved_special_token_223|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128229": {
+      "content": "<|reserved_special_token_224|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128230": {
+      "content": "<|reserved_special_token_225|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128231": {
+      "content": "<|reserved_special_token_226|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128232": {
+      "content": "<|reserved_special_token_227|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128233": {
+      "content": "<|reserved_special_token_228|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128234": {
+      "content": "<|reserved_special_token_229|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128235": {
+      "content": "<|reserved_special_token_230|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128236": {
+      "content": "<|reserved_special_token_231|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128237": {
+      "content": "<|reserved_special_token_232|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128238": {
+      "content": "<|reserved_special_token_233|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128239": {
+      "content": "<|reserved_special_token_234|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128240": {
+      "content": "<|reserved_special_token_235|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128241": {
+      "content": "<|reserved_special_token_236|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128242": {
+      "content": "<|reserved_special_token_237|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128243": {
+      "content": "<|reserved_special_token_238|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128244": {
+      "content": "<|reserved_special_token_239|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128245": {
+      "content": "<|reserved_special_token_240|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128246": {
+      "content": "<|reserved_special_token_241|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128247": {
+      "content": "<|reserved_special_token_242|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128248": {
+      "content": "<|reserved_special_token_243|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128249": {
+      "content": "<|reserved_special_token_244|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128250": {
+      "content": "<|reserved_special_token_245|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128251": {
+      "content": "<|reserved_special_token_246|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128252": {
+      "content": "<|reserved_special_token_247|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128253": {
+      "content": "<|reserved_special_token_248|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128254": {
+      "content": "<|reserved_special_token_249|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128255": {
+      "content": "<|reserved_special_token_250|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<|begin_of_text|>",
+  "chat_template": "{% set loop_messages = messages %}{% for message in loop_messages %}{% set content = '<|start_header_id|>' + message['role'] + '<|end_header_id|>\n\n'+ message['content'] | trim + '<|eot_id|>' %}{% if loop.index0 == 0 %}{% set content = bos_token + content %}{% endif %}{{ content }}{% endfor %}{% if add_generation_prompt %}{{ '<|start_header_id|>assistant<|end_header_id|>\n\n' }}{% endif %}",
+  "clean_up_tokenization_spaces": true,
+  "eos_token": "<|eot_id|>",
+  "model_input_names": [
+    "input_ids",
+    "attention_mask"
+  ],
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<|eot_id|>",
+  "tokenizer_class": "PreTrainedTokenizerFast"
+}
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-5970/trainer_state.json b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-5970/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..9f9251c68727836b1d9a7287d6fd442eeaf6d65f
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-5970/trainer_state.json
@@ -0,0 +1,4244 @@
+{
+  "best_metric": 1.8028968572616577,
+  "best_model_checkpoint": "outputs-001/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-2985",
+  "epoch": 4.0,
+  "eval_steps": 10,
+  "global_step": 5970,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.006700167504187605,
+      "grad_norm": 0.565915048122406,
+      "learning_rate": 0.0002,
+      "loss": 2.6189,
+      "step": 10
+    },
+    {
+      "epoch": 0.01340033500837521,
+      "grad_norm": 0.5004463791847229,
+      "learning_rate": 0.0002,
+      "loss": 2.3162,
+      "step": 20
+    },
+    {
+      "epoch": 0.020100502512562814,
+      "grad_norm": 0.511043906211853,
+      "learning_rate": 0.0002,
+      "loss": 2.0576,
+      "step": 30
+    },
+    {
+      "epoch": 0.02680067001675042,
+      "grad_norm": 0.47327178716659546,
+      "learning_rate": 0.0002,
+      "loss": 2.0085,
+      "step": 40
+    },
+    {
+      "epoch": 0.03350083752093802,
+      "grad_norm": 0.5511676669120789,
+      "learning_rate": 0.0002,
+      "loss": 2.0276,
+      "step": 50
+    },
+    {
+      "epoch": 0.04020100502512563,
+      "grad_norm": 0.4666278064250946,
+      "learning_rate": 0.0002,
+      "loss": 1.9075,
+      "step": 60
+    },
+    {
+      "epoch": 0.04690117252931323,
+      "grad_norm": 0.5310961008071899,
+      "learning_rate": 0.0002,
+      "loss": 1.8413,
+      "step": 70
+    },
+    {
+      "epoch": 0.05360134003350084,
+      "grad_norm": 0.5606027245521545,
+      "learning_rate": 0.0002,
+      "loss": 1.8711,
+      "step": 80
+    },
+    {
+      "epoch": 0.06030150753768844,
+      "grad_norm": 0.4934779703617096,
+      "learning_rate": 0.0002,
+      "loss": 1.9282,
+      "step": 90
+    },
+    {
+      "epoch": 0.06700167504187604,
+      "grad_norm": 0.4821869730949402,
+      "learning_rate": 0.0002,
+      "loss": 1.8925,
+      "step": 100
+    },
+    {
+      "epoch": 0.07370184254606366,
+      "grad_norm": 0.5262084603309631,
+      "learning_rate": 0.0002,
+      "loss": 1.8628,
+      "step": 110
+    },
+    {
+      "epoch": 0.08040201005025126,
+      "grad_norm": 0.3774230182170868,
+      "learning_rate": 0.0002,
+      "loss": 1.8347,
+      "step": 120
+    },
+    {
+      "epoch": 0.08710217755443886,
+      "grad_norm": 0.34137430787086487,
+      "learning_rate": 0.0002,
+      "loss": 1.8386,
+      "step": 130
+    },
+    {
+      "epoch": 0.09380234505862646,
+      "grad_norm": 0.407272070646286,
+      "learning_rate": 0.0002,
+      "loss": 1.861,
+      "step": 140
+    },
+    {
+      "epoch": 0.10050251256281408,
+      "grad_norm": 0.4011937975883484,
+      "learning_rate": 0.0002,
+      "loss": 1.8279,
+      "step": 150
+    },
+    {
+      "epoch": 0.10720268006700168,
+      "grad_norm": 0.4432467222213745,
+      "learning_rate": 0.0002,
+      "loss": 1.9317,
+      "step": 160
+    },
+    {
+      "epoch": 0.11390284757118928,
+      "grad_norm": 0.44030463695526123,
+      "learning_rate": 0.0002,
+      "loss": 1.8157,
+      "step": 170
+    },
+    {
+      "epoch": 0.12060301507537688,
+      "grad_norm": 0.3799569308757782,
+      "learning_rate": 0.0002,
+      "loss": 1.8534,
+      "step": 180
+    },
+    {
+      "epoch": 0.1273031825795645,
+      "grad_norm": 0.33721521496772766,
+      "learning_rate": 0.0002,
+      "loss": 1.7658,
+      "step": 190
+    },
+    {
+      "epoch": 0.13400335008375208,
+      "grad_norm": 0.4096226692199707,
+      "learning_rate": 0.0002,
+      "loss": 1.8269,
+      "step": 200
+    },
+    {
+      "epoch": 0.1407035175879397,
+      "grad_norm": 0.37374693155288696,
+      "learning_rate": 0.0002,
+      "loss": 1.802,
+      "step": 210
+    },
+    {
+      "epoch": 0.1474036850921273,
+      "grad_norm": 0.3249480128288269,
+      "learning_rate": 0.0002,
+      "loss": 1.8901,
+      "step": 220
+    },
+    {
+      "epoch": 0.1541038525963149,
+      "grad_norm": 0.3612042963504791,
+      "learning_rate": 0.0002,
+      "loss": 1.8163,
+      "step": 230
+    },
+    {
+      "epoch": 0.16080402010050251,
+      "grad_norm": 0.3686671257019043,
+      "learning_rate": 0.0002,
+      "loss": 1.7585,
+      "step": 240
+    },
+    {
+      "epoch": 0.16750418760469013,
+      "grad_norm": 0.3521044850349426,
+      "learning_rate": 0.0002,
+      "loss": 1.8365,
+      "step": 250
+    },
+    {
+      "epoch": 0.17420435510887772,
+      "grad_norm": 0.4073677361011505,
+      "learning_rate": 0.0002,
+      "loss": 1.8623,
+      "step": 260
+    },
+    {
+      "epoch": 0.18090452261306533,
+      "grad_norm": 0.34522193670272827,
+      "learning_rate": 0.0002,
+      "loss": 1.8026,
+      "step": 270
+    },
+    {
+      "epoch": 0.18760469011725292,
+      "grad_norm": 0.4121900498867035,
+      "learning_rate": 0.0002,
+      "loss": 1.8162,
+      "step": 280
+    },
+    {
+      "epoch": 0.19430485762144054,
+      "grad_norm": 0.3544778525829315,
+      "learning_rate": 0.0002,
+      "loss": 1.7976,
+      "step": 290
+    },
+    {
+      "epoch": 0.20100502512562815,
+      "grad_norm": 0.3482133448123932,
+      "learning_rate": 0.0002,
+      "loss": 1.8787,
+      "step": 300
+    },
+    {
+      "epoch": 0.20770519262981574,
+      "grad_norm": 0.3421826660633087,
+      "learning_rate": 0.0002,
+      "loss": 1.8578,
+      "step": 310
+    },
+    {
+      "epoch": 0.21440536013400335,
+      "grad_norm": 0.5024696588516235,
+      "learning_rate": 0.0002,
+      "loss": 1.8013,
+      "step": 320
+    },
+    {
+      "epoch": 0.22110552763819097,
+      "grad_norm": 0.36013063788414,
+      "learning_rate": 0.0002,
+      "loss": 1.8607,
+      "step": 330
+    },
+    {
+      "epoch": 0.22780569514237856,
+      "grad_norm": 0.3611244857311249,
+      "learning_rate": 0.0002,
+      "loss": 1.9075,
+      "step": 340
+    },
+    {
+      "epoch": 0.23450586264656617,
+      "grad_norm": 0.39244529604911804,
+      "learning_rate": 0.0002,
+      "loss": 1.8128,
+      "step": 350
+    },
+    {
+      "epoch": 0.24120603015075376,
+      "grad_norm": 0.3299325704574585,
+      "learning_rate": 0.0002,
+      "loss": 1.7885,
+      "step": 360
+    },
+    {
+      "epoch": 0.24790619765494137,
+      "grad_norm": 0.3994322419166565,
+      "learning_rate": 0.0002,
+      "loss": 1.8028,
+      "step": 370
+    },
+    {
+      "epoch": 0.254606365159129,
+      "grad_norm": 0.3559151887893677,
+      "learning_rate": 0.0002,
+      "loss": 1.8321,
+      "step": 380
+    },
+    {
+      "epoch": 0.2613065326633166,
+      "grad_norm": 0.3873756229877472,
+      "learning_rate": 0.0002,
+      "loss": 1.7802,
+      "step": 390
+    },
+    {
+      "epoch": 0.26800670016750416,
+      "grad_norm": 0.3710744082927704,
+      "learning_rate": 0.0002,
+      "loss": 1.7844,
+      "step": 400
+    },
+    {
+      "epoch": 0.2747068676716918,
+      "grad_norm": 0.3618465065956116,
+      "learning_rate": 0.0002,
+      "loss": 1.7789,
+      "step": 410
+    },
+    {
+      "epoch": 0.2814070351758794,
+      "grad_norm": 0.30063769221305847,
+      "learning_rate": 0.0002,
+      "loss": 1.8529,
+      "step": 420
+    },
+    {
+      "epoch": 0.288107202680067,
+      "grad_norm": 0.3695628345012665,
+      "learning_rate": 0.0002,
+      "loss": 1.7765,
+      "step": 430
+    },
+    {
+      "epoch": 0.2948073701842546,
+      "grad_norm": 0.31451135873794556,
+      "learning_rate": 0.0002,
+      "loss": 1.7982,
+      "step": 440
+    },
+    {
+      "epoch": 0.3015075376884422,
+      "grad_norm": 0.3959707021713257,
+      "learning_rate": 0.0002,
+      "loss": 1.7517,
+      "step": 450
+    },
+    {
+      "epoch": 0.3082077051926298,
+      "grad_norm": 0.33475354313850403,
+      "learning_rate": 0.0002,
+      "loss": 1.8142,
+      "step": 460
+    },
+    {
+      "epoch": 0.3149078726968174,
+      "grad_norm": 0.33933115005493164,
+      "learning_rate": 0.0002,
+      "loss": 1.8805,
+      "step": 470
+    },
+    {
+      "epoch": 0.32160804020100503,
+      "grad_norm": 0.3264943063259125,
+      "learning_rate": 0.0002,
+      "loss": 1.7564,
+      "step": 480
+    },
+    {
+      "epoch": 0.32830820770519265,
+      "grad_norm": 0.40188100934028625,
+      "learning_rate": 0.0002,
+      "loss": 1.8428,
+      "step": 490
+    },
+    {
+      "epoch": 0.33500837520938026,
+      "grad_norm": 0.37408649921417236,
+      "learning_rate": 0.0002,
+      "loss": 1.7624,
+      "step": 500
+    },
+    {
+      "epoch": 0.3417085427135678,
+      "grad_norm": 0.33925938606262207,
+      "learning_rate": 0.0002,
+      "loss": 1.7745,
+      "step": 510
+    },
+    {
+      "epoch": 0.34840871021775544,
+      "grad_norm": 0.36836713552474976,
+      "learning_rate": 0.0002,
+      "loss": 1.814,
+      "step": 520
+    },
+    {
+      "epoch": 0.35510887772194305,
+      "grad_norm": 0.37284499406814575,
+      "learning_rate": 0.0002,
+      "loss": 1.8037,
+      "step": 530
+    },
+    {
+      "epoch": 0.36180904522613067,
+      "grad_norm": 0.3192278742790222,
+      "learning_rate": 0.0002,
+      "loss": 1.8379,
+      "step": 540
+    },
+    {
+      "epoch": 0.3685092127303183,
+      "grad_norm": 0.30233290791511536,
+      "learning_rate": 0.0002,
+      "loss": 1.8702,
+      "step": 550
+    },
+    {
+      "epoch": 0.37520938023450584,
+      "grad_norm": 0.3340817391872406,
+      "learning_rate": 0.0002,
+      "loss": 1.8799,
+      "step": 560
+    },
+    {
+      "epoch": 0.38190954773869346,
+      "grad_norm": 0.32600095868110657,
+      "learning_rate": 0.0002,
+      "loss": 1.8404,
+      "step": 570
+    },
+    {
+      "epoch": 0.38860971524288107,
+      "grad_norm": 0.33711278438568115,
+      "learning_rate": 0.0002,
+      "loss": 1.7804,
+      "step": 580
+    },
+    {
+      "epoch": 0.3953098827470687,
+      "grad_norm": 0.34890690445899963,
+      "learning_rate": 0.0002,
+      "loss": 1.8445,
+      "step": 590
+    },
+    {
+      "epoch": 0.4020100502512563,
+      "grad_norm": 0.38238924741744995,
+      "learning_rate": 0.0002,
+      "loss": 1.8187,
+      "step": 600
+    },
+    {
+      "epoch": 0.40871021775544386,
+      "grad_norm": 0.34399354457855225,
+      "learning_rate": 0.0002,
+      "loss": 1.8111,
+      "step": 610
+    },
+    {
+      "epoch": 0.4154103852596315,
+      "grad_norm": 0.3346073627471924,
+      "learning_rate": 0.0002,
+      "loss": 1.8006,
+      "step": 620
+    },
+    {
+      "epoch": 0.4221105527638191,
+      "grad_norm": 0.3545648157596588,
+      "learning_rate": 0.0002,
+      "loss": 1.7705,
+      "step": 630
+    },
+    {
+      "epoch": 0.4288107202680067,
+      "grad_norm": 0.3378899097442627,
+      "learning_rate": 0.0002,
+      "loss": 1.8445,
+      "step": 640
+    },
+    {
+      "epoch": 0.4355108877721943,
+      "grad_norm": 0.3255569040775299,
+      "learning_rate": 0.0002,
+      "loss": 1.804,
+      "step": 650
+    },
+    {
+      "epoch": 0.44221105527638194,
+      "grad_norm": 0.34880587458610535,
+      "learning_rate": 0.0002,
+      "loss": 1.7679,
+      "step": 660
+    },
+    {
+      "epoch": 0.4489112227805695,
+      "grad_norm": 0.3402383625507355,
+      "learning_rate": 0.0002,
+      "loss": 1.7861,
+      "step": 670
+    },
+    {
+      "epoch": 0.4556113902847571,
+      "grad_norm": 0.3594033718109131,
+      "learning_rate": 0.0002,
+      "loss": 1.8131,
+      "step": 680
+    },
+    {
+      "epoch": 0.4623115577889447,
+      "grad_norm": 0.31000566482543945,
+      "learning_rate": 0.0002,
+      "loss": 1.8399,
+      "step": 690
+    },
+    {
+      "epoch": 0.46901172529313234,
+      "grad_norm": 0.37229061126708984,
+      "learning_rate": 0.0002,
+      "loss": 1.7521,
+      "step": 700
+    },
+    {
+      "epoch": 0.47571189279731996,
+      "grad_norm": 0.315801739692688,
+      "learning_rate": 0.0002,
+      "loss": 1.7779,
+      "step": 710
+    },
+    {
+      "epoch": 0.4824120603015075,
+      "grad_norm": 0.3220832645893097,
+      "learning_rate": 0.0002,
+      "loss": 1.7515,
+      "step": 720
+    },
+    {
+      "epoch": 0.48911222780569513,
+      "grad_norm": 0.3435456156730652,
+      "learning_rate": 0.0002,
+      "loss": 1.7181,
+      "step": 730
+    },
+    {
+      "epoch": 0.49581239530988275,
+      "grad_norm": 0.30380892753601074,
+      "learning_rate": 0.0002,
+      "loss": 1.8844,
+      "step": 740
+    },
+    {
+      "epoch": 0.5025125628140703,
+      "grad_norm": 0.3555026054382324,
+      "learning_rate": 0.0002,
+      "loss": 1.7792,
+      "step": 750
+    },
+    {
+      "epoch": 0.509212730318258,
+      "grad_norm": 0.3019855320453644,
+      "learning_rate": 0.0002,
+      "loss": 1.7714,
+      "step": 760
+    },
+    {
+      "epoch": 0.5159128978224455,
+      "grad_norm": 0.309111088514328,
+      "learning_rate": 0.0002,
+      "loss": 1.7962,
+      "step": 770
+    },
+    {
+      "epoch": 0.5226130653266332,
+      "grad_norm": 0.366020530462265,
+      "learning_rate": 0.0002,
+      "loss": 1.7913,
+      "step": 780
+    },
+    {
+      "epoch": 0.5293132328308208,
+      "grad_norm": 0.3267050087451935,
+      "learning_rate": 0.0002,
+      "loss": 1.8008,
+      "step": 790
+    },
+    {
+      "epoch": 0.5360134003350083,
+      "grad_norm": 0.34265750646591187,
+      "learning_rate": 0.0002,
+      "loss": 1.7397,
+      "step": 800
+    },
+    {
+      "epoch": 0.542713567839196,
+      "grad_norm": 0.313669890165329,
+      "learning_rate": 0.0002,
+      "loss": 1.8251,
+      "step": 810
+    },
+    {
+      "epoch": 0.5494137353433836,
+      "grad_norm": 0.3355236053466797,
+      "learning_rate": 0.0002,
+      "loss": 1.8369,
+      "step": 820
+    },
+    {
+      "epoch": 0.5561139028475712,
+      "grad_norm": 0.3186608552932739,
+      "learning_rate": 0.0002,
+      "loss": 1.7381,
+      "step": 830
+    },
+    {
+      "epoch": 0.5628140703517588,
+      "grad_norm": 0.30357518792152405,
+      "learning_rate": 0.0002,
+      "loss": 1.8034,
+      "step": 840
+    },
+    {
+      "epoch": 0.5695142378559463,
+      "grad_norm": 0.3990040123462677,
+      "learning_rate": 0.0002,
+      "loss": 1.769,
+      "step": 850
+    },
+    {
+      "epoch": 0.576214405360134,
+      "grad_norm": 0.34363803267478943,
+      "learning_rate": 0.0002,
+      "loss": 1.7482,
+      "step": 860
+    },
+    {
+      "epoch": 0.5829145728643216,
+      "grad_norm": 0.3757908046245575,
+      "learning_rate": 0.0002,
+      "loss": 1.8106,
+      "step": 870
+    },
+    {
+      "epoch": 0.5896147403685092,
+      "grad_norm": 0.3359757661819458,
+      "learning_rate": 0.0002,
+      "loss": 1.8104,
+      "step": 880
+    },
+    {
+      "epoch": 0.5963149078726968,
+      "grad_norm": 0.5555329918861389,
+      "learning_rate": 0.0002,
+      "loss": 1.7591,
+      "step": 890
+    },
+    {
+      "epoch": 0.6030150753768844,
+      "grad_norm": 0.4046323895454407,
+      "learning_rate": 0.0002,
+      "loss": 1.7715,
+      "step": 900
+    },
+    {
+      "epoch": 0.609715242881072,
+      "grad_norm": 0.29834219813346863,
+      "learning_rate": 0.0002,
+      "loss": 1.7998,
+      "step": 910
+    },
+    {
+      "epoch": 0.6164154103852596,
+      "grad_norm": 0.3241238594055176,
+      "learning_rate": 0.0002,
+      "loss": 1.7826,
+      "step": 920
+    },
+    {
+      "epoch": 0.6231155778894473,
+      "grad_norm": 0.35154739022254944,
+      "learning_rate": 0.0002,
+      "loss": 1.8342,
+      "step": 930
+    },
+    {
+      "epoch": 0.6298157453936348,
+      "grad_norm": 0.3287706673145294,
+      "learning_rate": 0.0002,
+      "loss": 1.8076,
+      "step": 940
+    },
+    {
+      "epoch": 0.6365159128978225,
+      "grad_norm": 0.35670626163482666,
+      "learning_rate": 0.0002,
+      "loss": 1.8038,
+      "step": 950
+    },
+    {
+      "epoch": 0.6432160804020101,
+      "grad_norm": 0.6114104986190796,
+      "learning_rate": 0.0002,
+      "loss": 1.869,
+      "step": 960
+    },
+    {
+      "epoch": 0.6499162479061976,
+      "grad_norm": 0.3186565041542053,
+      "learning_rate": 0.0002,
+      "loss": 1.8297,
+      "step": 970
+    },
+    {
+      "epoch": 0.6566164154103853,
+      "grad_norm": 0.27164125442504883,
+      "learning_rate": 0.0002,
+      "loss": 1.7539,
+      "step": 980
+    },
+    {
+      "epoch": 0.6633165829145728,
+      "grad_norm": 0.34407344460487366,
+      "learning_rate": 0.0002,
+      "loss": 1.8339,
+      "step": 990
+    },
+    {
+      "epoch": 0.6700167504187605,
+      "grad_norm": 0.368415892124176,
+      "learning_rate": 0.0002,
+      "loss": 1.855,
+      "step": 1000
+    },
+    {
+      "epoch": 0.6767169179229481,
+      "grad_norm": 0.3306390643119812,
+      "learning_rate": 0.0002,
+      "loss": 1.7821,
+      "step": 1010
+    },
+    {
+      "epoch": 0.6834170854271356,
+      "grad_norm": 0.3198648989200592,
+      "learning_rate": 0.0002,
+      "loss": 1.786,
+      "step": 1020
+    },
+    {
+      "epoch": 0.6901172529313233,
+      "grad_norm": 0.3092987537384033,
+      "learning_rate": 0.0002,
+      "loss": 1.816,
+      "step": 1030
+    },
+    {
+      "epoch": 0.6968174204355109,
+      "grad_norm": 0.3090653419494629,
+      "learning_rate": 0.0002,
+      "loss": 1.7689,
+      "step": 1040
+    },
+    {
+      "epoch": 0.7035175879396985,
+      "grad_norm": 0.3485880196094513,
+      "learning_rate": 0.0002,
+      "loss": 1.7544,
+      "step": 1050
+    },
+    {
+      "epoch": 0.7102177554438861,
+      "grad_norm": 0.35782721638679504,
+      "learning_rate": 0.0002,
+      "loss": 1.8187,
+      "step": 1060
+    },
+    {
+      "epoch": 0.7169179229480737,
+      "grad_norm": 0.34256869554519653,
+      "learning_rate": 0.0002,
+      "loss": 1.806,
+      "step": 1070
+    },
+    {
+      "epoch": 0.7236180904522613,
+      "grad_norm": 0.30461037158966064,
+      "learning_rate": 0.0002,
+      "loss": 1.7873,
+      "step": 1080
+    },
+    {
+      "epoch": 0.7303182579564489,
+      "grad_norm": 0.3398691713809967,
+      "learning_rate": 0.0002,
+      "loss": 1.7367,
+      "step": 1090
+    },
+    {
+      "epoch": 0.7370184254606366,
+      "grad_norm": 0.3180808126926422,
+      "learning_rate": 0.0002,
+      "loss": 1.8756,
+      "step": 1100
+    },
+    {
+      "epoch": 0.7437185929648241,
+      "grad_norm": 0.34400665760040283,
+      "learning_rate": 0.0002,
+      "loss": 1.6988,
+      "step": 1110
+    },
+    {
+      "epoch": 0.7504187604690117,
+      "grad_norm": 0.34244877099990845,
+      "learning_rate": 0.0002,
+      "loss": 1.7851,
+      "step": 1120
+    },
+    {
+      "epoch": 0.7571189279731994,
+      "grad_norm": 0.29946693778038025,
+      "learning_rate": 0.0002,
+      "loss": 1.7841,
+      "step": 1130
+    },
+    {
+      "epoch": 0.7638190954773869,
+      "grad_norm": 0.37547236680984497,
+      "learning_rate": 0.0002,
+      "loss": 1.7456,
+      "step": 1140
+    },
+    {
+      "epoch": 0.7705192629815746,
+      "grad_norm": 0.3263005018234253,
+      "learning_rate": 0.0002,
+      "loss": 1.8425,
+      "step": 1150
+    },
+    {
+      "epoch": 0.7772194304857621,
+      "grad_norm": 0.41363608837127686,
+      "learning_rate": 0.0002,
+      "loss": 1.7222,
+      "step": 1160
+    },
+    {
+      "epoch": 0.7839195979899497,
+      "grad_norm": 0.36267954111099243,
+      "learning_rate": 0.0002,
+      "loss": 1.7836,
+      "step": 1170
+    },
+    {
+      "epoch": 0.7906197654941374,
+      "grad_norm": 0.31789499521255493,
+      "learning_rate": 0.0002,
+      "loss": 1.9183,
+      "step": 1180
+    },
+    {
+      "epoch": 0.7973199329983249,
+      "grad_norm": 0.5708149075508118,
+      "learning_rate": 0.0002,
+      "loss": 1.78,
+      "step": 1190
+    },
+    {
+      "epoch": 0.8040201005025126,
+      "grad_norm": 0.322099506855011,
+      "learning_rate": 0.0002,
+      "loss": 1.6908,
+      "step": 1200
+    },
+    {
+      "epoch": 0.8107202680067002,
+      "grad_norm": 0.3419909179210663,
+      "learning_rate": 0.0002,
+      "loss": 1.7639,
+      "step": 1210
+    },
+    {
+      "epoch": 0.8174204355108877,
+      "grad_norm": 0.36286255717277527,
+      "learning_rate": 0.0002,
+      "loss": 1.7428,
+      "step": 1220
+    },
+    {
+      "epoch": 0.8241206030150754,
+      "grad_norm": 0.33992862701416016,
+      "learning_rate": 0.0002,
+      "loss": 1.8409,
+      "step": 1230
+    },
+    {
+      "epoch": 0.830820770519263,
+      "grad_norm": 0.32622793316841125,
+      "learning_rate": 0.0002,
+      "loss": 1.7507,
+      "step": 1240
+    },
+    {
+      "epoch": 0.8375209380234506,
+      "grad_norm": 0.3036167621612549,
+      "learning_rate": 0.0002,
+      "loss": 1.8098,
+      "step": 1250
+    },
+    {
+      "epoch": 0.8442211055276382,
+      "grad_norm": 0.3182215392589569,
+      "learning_rate": 0.0002,
+      "loss": 1.8094,
+      "step": 1260
+    },
+    {
+      "epoch": 0.8509212730318257,
+      "grad_norm": 0.3270018696784973,
+      "learning_rate": 0.0002,
+      "loss": 1.8017,
+      "step": 1270
+    },
+    {
+      "epoch": 0.8576214405360134,
+      "grad_norm": 0.32652342319488525,
+      "learning_rate": 0.0002,
+      "loss": 1.798,
+      "step": 1280
+    },
+    {
+      "epoch": 0.864321608040201,
+      "grad_norm": 0.3631329834461212,
+      "learning_rate": 0.0002,
+      "loss": 1.7448,
+      "step": 1290
+    },
+    {
+      "epoch": 0.8710217755443886,
+      "grad_norm": 0.36706018447875977,
+      "learning_rate": 0.0002,
+      "loss": 1.7,
+      "step": 1300
+    },
+    {
+      "epoch": 0.8777219430485762,
+      "grad_norm": 0.3347418010234833,
+      "learning_rate": 0.0002,
+      "loss": 1.8178,
+      "step": 1310
+    },
+    {
+      "epoch": 0.8844221105527639,
+      "grad_norm": 0.34371060132980347,
+      "learning_rate": 0.0002,
+      "loss": 1.7824,
+      "step": 1320
+    },
+    {
+      "epoch": 0.8911222780569514,
+      "grad_norm": 0.3029090166091919,
+      "learning_rate": 0.0002,
+      "loss": 1.783,
+      "step": 1330
+    },
+    {
+      "epoch": 0.897822445561139,
+      "grad_norm": 0.34700682759284973,
+      "learning_rate": 0.0002,
+      "loss": 1.8017,
+      "step": 1340
+    },
+    {
+      "epoch": 0.9045226130653267,
+      "grad_norm": 0.35574328899383545,
+      "learning_rate": 0.0002,
+      "loss": 1.7998,
+      "step": 1350
+    },
+    {
+      "epoch": 0.9112227805695142,
+      "grad_norm": 0.30928221344947815,
+      "learning_rate": 0.0002,
+      "loss": 1.7339,
+      "step": 1360
+    },
+    {
+      "epoch": 0.9179229480737019,
+      "grad_norm": 0.30652928352355957,
+      "learning_rate": 0.0002,
+      "loss": 1.7479,
+      "step": 1370
+    },
+    {
+      "epoch": 0.9246231155778895,
+      "grad_norm": 0.3838157653808594,
+      "learning_rate": 0.0002,
+      "loss": 1.7491,
+      "step": 1380
+    },
+    {
+      "epoch": 0.931323283082077,
+      "grad_norm": 0.31655240058898926,
+      "learning_rate": 0.0002,
+      "loss": 1.7977,
+      "step": 1390
+    },
+    {
+      "epoch": 0.9380234505862647,
+      "grad_norm": 0.41737303137779236,
+      "learning_rate": 0.0002,
+      "loss": 1.8175,
+      "step": 1400
+    },
+    {
+      "epoch": 0.9447236180904522,
+      "grad_norm": 0.3227267861366272,
+      "learning_rate": 0.0002,
+      "loss": 1.6811,
+      "step": 1410
+    },
+    {
+      "epoch": 0.9514237855946399,
+      "grad_norm": 0.3729925751686096,
+      "learning_rate": 0.0002,
+      "loss": 1.7343,
+      "step": 1420
+    },
+    {
+      "epoch": 0.9581239530988275,
+      "grad_norm": 0.30779409408569336,
+      "learning_rate": 0.0002,
+      "loss": 1.8221,
+      "step": 1430
+    },
+    {
+      "epoch": 0.964824120603015,
+      "grad_norm": 0.334379643201828,
+      "learning_rate": 0.0002,
+      "loss": 1.7972,
+      "step": 1440
+    },
+    {
+      "epoch": 0.9715242881072027,
+      "grad_norm": 0.3568236231803894,
+      "learning_rate": 0.0002,
+      "loss": 1.7141,
+      "step": 1450
+    },
+    {
+      "epoch": 0.9782244556113903,
+      "grad_norm": 0.33310577273368835,
+      "learning_rate": 0.0002,
+      "loss": 1.7541,
+      "step": 1460
+    },
+    {
+      "epoch": 0.9849246231155779,
+      "grad_norm": 0.2972261905670166,
+      "learning_rate": 0.0002,
+      "loss": 1.8511,
+      "step": 1470
+    },
+    {
+      "epoch": 0.9916247906197655,
+      "grad_norm": 0.3322717845439911,
+      "learning_rate": 0.0002,
+      "loss": 1.7654,
+      "step": 1480
+    },
+    {
+      "epoch": 0.998324958123953,
+      "grad_norm": 0.3276330828666687,
+      "learning_rate": 0.0002,
+      "loss": 1.8033,
+      "step": 1490
+    },
+    {
+      "epoch": 0.9996649916247906,
+      "eval_loss": 1.8036354780197144,
+      "eval_runtime": 37.8949,
+      "eval_samples_per_second": 13.59,
+      "eval_steps_per_second": 1.715,
+      "step": 1492
+    },
+    {
+      "epoch": 1.0050251256281406,
+      "grad_norm": 0.29252371191978455,
+      "learning_rate": 0.0002,
+      "loss": 1.7138,
+      "step": 1500
+    },
+    {
+      "epoch": 1.0117252931323284,
+      "grad_norm": 0.31607162952423096,
+      "learning_rate": 0.0002,
+      "loss": 1.8198,
+      "step": 1510
+    },
+    {
+      "epoch": 1.018425460636516,
+      "grad_norm": 0.32294467091560364,
+      "learning_rate": 0.0002,
+      "loss": 1.6779,
+      "step": 1520
+    },
+    {
+      "epoch": 1.0251256281407035,
+      "grad_norm": 0.3868017792701721,
+      "learning_rate": 0.0002,
+      "loss": 1.7919,
+      "step": 1530
+    },
+    {
+      "epoch": 1.031825795644891,
+      "grad_norm": 0.3178282082080841,
+      "learning_rate": 0.0002,
+      "loss": 1.7954,
+      "step": 1540
+    },
+    {
+      "epoch": 1.0385259631490786,
+      "grad_norm": 0.3706750273704529,
+      "learning_rate": 0.0002,
+      "loss": 1.7136,
+      "step": 1550
+    },
+    {
+      "epoch": 1.0452261306532664,
+      "grad_norm": 0.33930912613868713,
+      "learning_rate": 0.0002,
+      "loss": 1.7382,
+      "step": 1560
+    },
+    {
+      "epoch": 1.051926298157454,
+      "grad_norm": 0.33970504999160767,
+      "learning_rate": 0.0002,
+      "loss": 1.7602,
+      "step": 1570
+    },
+    {
+      "epoch": 1.0586264656616415,
+      "grad_norm": 0.42553383111953735,
+      "learning_rate": 0.0002,
+      "loss": 1.6573,
+      "step": 1580
+    },
+    {
+      "epoch": 1.065326633165829,
+      "grad_norm": 0.3772421181201935,
+      "learning_rate": 0.0002,
+      "loss": 1.645,
+      "step": 1590
+    },
+    {
+      "epoch": 1.0720268006700167,
+      "grad_norm": 0.34212902188301086,
+      "learning_rate": 0.0002,
+      "loss": 1.7362,
+      "step": 1600
+    },
+    {
+      "epoch": 1.0787269681742044,
+      "grad_norm": 0.3798283338546753,
+      "learning_rate": 0.0002,
+      "loss": 1.7057,
+      "step": 1610
+    },
+    {
+      "epoch": 1.085427135678392,
+      "grad_norm": 0.36909598112106323,
+      "learning_rate": 0.0002,
+      "loss": 1.7468,
+      "step": 1620
+    },
+    {
+      "epoch": 1.0921273031825796,
+      "grad_norm": 0.3344230651855469,
+      "learning_rate": 0.0002,
+      "loss": 1.7807,
+      "step": 1630
+    },
+    {
+      "epoch": 1.0988274706867671,
+      "grad_norm": 0.3862569332122803,
+      "learning_rate": 0.0002,
+      "loss": 1.7111,
+      "step": 1640
+    },
+    {
+      "epoch": 1.1055276381909547,
+      "grad_norm": 0.31188511848449707,
+      "learning_rate": 0.0002,
+      "loss": 1.7163,
+      "step": 1650
+    },
+    {
+      "epoch": 1.1122278056951425,
+      "grad_norm": 0.3563670814037323,
+      "learning_rate": 0.0002,
+      "loss": 1.7263,
+      "step": 1660
+    },
+    {
+      "epoch": 1.11892797319933,
+      "grad_norm": 0.35052165389060974,
+      "learning_rate": 0.0002,
+      "loss": 1.7718,
+      "step": 1670
+    },
+    {
+      "epoch": 1.1256281407035176,
+      "grad_norm": 0.3285699188709259,
+      "learning_rate": 0.0002,
+      "loss": 1.7601,
+      "step": 1680
+    },
+    {
+      "epoch": 1.1323283082077051,
+      "grad_norm": 0.3639393746852875,
+      "learning_rate": 0.0002,
+      "loss": 1.6877,
+      "step": 1690
+    },
+    {
+      "epoch": 1.1390284757118927,
+      "grad_norm": 0.3842753767967224,
+      "learning_rate": 0.0002,
+      "loss": 1.7719,
+      "step": 1700
+    },
+    {
+      "epoch": 1.1457286432160805,
+      "grad_norm": 0.3624933063983917,
+      "learning_rate": 0.0002,
+      "loss": 1.7002,
+      "step": 1710
+    },
+    {
+      "epoch": 1.152428810720268,
+      "grad_norm": 0.3641220033168793,
+      "learning_rate": 0.0002,
+      "loss": 1.7243,
+      "step": 1720
+    },
+    {
+      "epoch": 1.1591289782244556,
+      "grad_norm": 0.32765355706214905,
+      "learning_rate": 0.0002,
+      "loss": 1.752,
+      "step": 1730
+    },
+    {
+      "epoch": 1.1658291457286432,
+      "grad_norm": 0.34974896907806396,
+      "learning_rate": 0.0002,
+      "loss": 1.6556,
+      "step": 1740
+    },
+    {
+      "epoch": 1.1725293132328307,
+      "grad_norm": 0.3910926580429077,
+      "learning_rate": 0.0002,
+      "loss": 1.7273,
+      "step": 1750
+    },
+    {
+      "epoch": 1.1792294807370185,
+      "grad_norm": 0.3564300537109375,
+      "learning_rate": 0.0002,
+      "loss": 1.7204,
+      "step": 1760
+    },
+    {
+      "epoch": 1.185929648241206,
+      "grad_norm": 0.34822574257850647,
+      "learning_rate": 0.0002,
+      "loss": 1.746,
+      "step": 1770
+    },
+    {
+      "epoch": 1.1926298157453936,
+      "grad_norm": 0.36185044050216675,
+      "learning_rate": 0.0002,
+      "loss": 1.7256,
+      "step": 1780
+    },
+    {
+      "epoch": 1.1993299832495812,
+      "grad_norm": 0.34866711497306824,
+      "learning_rate": 0.0002,
+      "loss": 1.6431,
+      "step": 1790
+    },
+    {
+      "epoch": 1.2060301507537687,
+      "grad_norm": 0.4017769992351532,
+      "learning_rate": 0.0002,
+      "loss": 1.8084,
+      "step": 1800
+    },
+    {
+      "epoch": 1.2127303182579565,
+      "grad_norm": 0.32930681109428406,
+      "learning_rate": 0.0002,
+      "loss": 1.6985,
+      "step": 1810
+    },
+    {
+      "epoch": 1.219430485762144,
+      "grad_norm": 0.35951921343803406,
+      "learning_rate": 0.0002,
+      "loss": 1.7606,
+      "step": 1820
+    },
+    {
+      "epoch": 1.2261306532663316,
+      "grad_norm": 0.37366992235183716,
+      "learning_rate": 0.0002,
+      "loss": 1.6933,
+      "step": 1830
+    },
+    {
+      "epoch": 1.2328308207705192,
+      "grad_norm": 0.3565689027309418,
+      "learning_rate": 0.0002,
+      "loss": 1.6737,
+      "step": 1840
+    },
+    {
+      "epoch": 1.2395309882747068,
+      "grad_norm": 0.3692343533039093,
+      "learning_rate": 0.0002,
+      "loss": 1.8013,
+      "step": 1850
+    },
+    {
+      "epoch": 1.2462311557788945,
+      "grad_norm": 0.38426971435546875,
+      "learning_rate": 0.0002,
+      "loss": 1.736,
+      "step": 1860
+    },
+    {
+      "epoch": 1.252931323283082,
+      "grad_norm": 0.33559855818748474,
+      "learning_rate": 0.0002,
+      "loss": 1.7031,
+      "step": 1870
+    },
+    {
+      "epoch": 1.2596314907872697,
+      "grad_norm": 0.34181106090545654,
+      "learning_rate": 0.0002,
+      "loss": 1.7033,
+      "step": 1880
+    },
+    {
+      "epoch": 1.2663316582914572,
+      "grad_norm": 0.3916318416595459,
+      "learning_rate": 0.0002,
+      "loss": 1.7707,
+      "step": 1890
+    },
+    {
+      "epoch": 1.2730318257956448,
+      "grad_norm": 0.3887825012207031,
+      "learning_rate": 0.0002,
+      "loss": 1.6686,
+      "step": 1900
+    },
+    {
+      "epoch": 1.2797319932998326,
+      "grad_norm": 0.33583927154541016,
+      "learning_rate": 0.0002,
+      "loss": 1.7062,
+      "step": 1910
+    },
+    {
+      "epoch": 1.2864321608040201,
+      "grad_norm": 0.37639349699020386,
+      "learning_rate": 0.0002,
+      "loss": 1.717,
+      "step": 1920
+    },
+    {
+      "epoch": 1.2931323283082077,
+      "grad_norm": 0.38059428334236145,
+      "learning_rate": 0.0002,
+      "loss": 1.777,
+      "step": 1930
+    },
+    {
+      "epoch": 1.2998324958123952,
+      "grad_norm": 0.37253183126449585,
+      "learning_rate": 0.0002,
+      "loss": 1.6126,
+      "step": 1940
+    },
+    {
+      "epoch": 1.3065326633165828,
+      "grad_norm": 0.37371566891670227,
+      "learning_rate": 0.0002,
+      "loss": 1.6758,
+      "step": 1950
+    },
+    {
+      "epoch": 1.3132328308207706,
+      "grad_norm": 0.4080910086631775,
+      "learning_rate": 0.0002,
+      "loss": 1.6788,
+      "step": 1960
+    },
+    {
+      "epoch": 1.3199329983249581,
+      "grad_norm": 0.3174354135990143,
+      "learning_rate": 0.0002,
+      "loss": 1.6518,
+      "step": 1970
+    },
+    {
+      "epoch": 1.3266331658291457,
+      "grad_norm": 0.4518888294696808,
+      "learning_rate": 0.0002,
+      "loss": 1.7925,
+      "step": 1980
+    },
+    {
+      "epoch": 1.3333333333333333,
+      "grad_norm": 0.3627921938896179,
+      "learning_rate": 0.0002,
+      "loss": 1.7085,
+      "step": 1990
+    },
+    {
+      "epoch": 1.3400335008375208,
+      "grad_norm": 0.3655930161476135,
+      "learning_rate": 0.0002,
+      "loss": 1.7676,
+      "step": 2000
+    },
+    {
+      "epoch": 1.3467336683417086,
+      "grad_norm": 0.3509993255138397,
+      "learning_rate": 0.0002,
+      "loss": 1.7016,
+      "step": 2010
+    },
+    {
+      "epoch": 1.3534338358458962,
+      "grad_norm": 0.4281129240989685,
+      "learning_rate": 0.0002,
+      "loss": 1.7359,
+      "step": 2020
+    },
+    {
+      "epoch": 1.3601340033500837,
+      "grad_norm": 0.3821414113044739,
+      "learning_rate": 0.0002,
+      "loss": 1.6884,
+      "step": 2030
+    },
+    {
+      "epoch": 1.3668341708542713,
+      "grad_norm": 0.3907586336135864,
+      "learning_rate": 0.0002,
+      "loss": 1.7075,
+      "step": 2040
+    },
+    {
+      "epoch": 1.3735343383584588,
+      "grad_norm": 0.37792932987213135,
+      "learning_rate": 0.0002,
+      "loss": 1.7424,
+      "step": 2050
+    },
+    {
+      "epoch": 1.3802345058626466,
+      "grad_norm": 0.3693985641002655,
+      "learning_rate": 0.0002,
+      "loss": 1.7305,
+      "step": 2060
+    },
+    {
+      "epoch": 1.3869346733668342,
+      "grad_norm": 0.32275936007499695,
+      "learning_rate": 0.0002,
+      "loss": 1.7434,
+      "step": 2070
+    },
+    {
+      "epoch": 1.3936348408710217,
+      "grad_norm": 0.3789440095424652,
+      "learning_rate": 0.0002,
+      "loss": 1.6677,
+      "step": 2080
+    },
+    {
+      "epoch": 1.4003350083752093,
+      "grad_norm": 0.3638380467891693,
+      "learning_rate": 0.0002,
+      "loss": 1.6825,
+      "step": 2090
+    },
+    {
+      "epoch": 1.4070351758793969,
+      "grad_norm": 0.3495481610298157,
+      "learning_rate": 0.0002,
+      "loss": 1.6542,
+      "step": 2100
+    },
+    {
+      "epoch": 1.4137353433835846,
+      "grad_norm": 0.37920597195625305,
+      "learning_rate": 0.0002,
+      "loss": 1.7225,
+      "step": 2110
+    },
+    {
+      "epoch": 1.4204355108877722,
+      "grad_norm": 0.37218064069747925,
+      "learning_rate": 0.0002,
+      "loss": 1.7329,
+      "step": 2120
+    },
+    {
+      "epoch": 1.4271356783919598,
+      "grad_norm": 0.38074082136154175,
+      "learning_rate": 0.0002,
+      "loss": 1.799,
+      "step": 2130
+    },
+    {
+      "epoch": 1.4338358458961473,
+      "grad_norm": 0.3455527126789093,
+      "learning_rate": 0.0002,
+      "loss": 1.7403,
+      "step": 2140
+    },
+    {
+      "epoch": 1.4405360134003349,
+      "grad_norm": 0.3712003529071808,
+      "learning_rate": 0.0002,
+      "loss": 1.776,
+      "step": 2150
+    },
+    {
+      "epoch": 1.4472361809045227,
+      "grad_norm": 0.3786754906177521,
+      "learning_rate": 0.0002,
+      "loss": 1.7619,
+      "step": 2160
+    },
+    {
+      "epoch": 1.4539363484087102,
+      "grad_norm": 0.3879223167896271,
+      "learning_rate": 0.0002,
+      "loss": 1.68,
+      "step": 2170
+    },
+    {
+      "epoch": 1.4606365159128978,
+      "grad_norm": 0.38738805055618286,
+      "learning_rate": 0.0002,
+      "loss": 1.7,
+      "step": 2180
+    },
+    {
+      "epoch": 1.4673366834170856,
+      "grad_norm": 0.39768800139427185,
+      "learning_rate": 0.0002,
+      "loss": 1.7581,
+      "step": 2190
+    },
+    {
+      "epoch": 1.474036850921273,
+      "grad_norm": 0.4172441065311432,
+      "learning_rate": 0.0002,
+      "loss": 1.7671,
+      "step": 2200
+    },
+    {
+      "epoch": 1.4807370184254607,
+      "grad_norm": 0.4043174982070923,
+      "learning_rate": 0.0002,
+      "loss": 1.6736,
+      "step": 2210
+    },
+    {
+      "epoch": 1.4874371859296482,
+      "grad_norm": 0.3750883936882019,
+      "learning_rate": 0.0002,
+      "loss": 1.7444,
+      "step": 2220
+    },
+    {
+      "epoch": 1.4941373534338358,
+      "grad_norm": 0.3552253246307373,
+      "learning_rate": 0.0002,
+      "loss": 1.6861,
+      "step": 2230
+    },
+    {
+      "epoch": 1.5008375209380236,
+      "grad_norm": 0.34607139229774475,
+      "learning_rate": 0.0002,
+      "loss": 1.6471,
+      "step": 2240
+    },
+    {
+      "epoch": 1.507537688442211,
+      "grad_norm": 0.3406706750392914,
+      "learning_rate": 0.0002,
+      "loss": 1.6962,
+      "step": 2250
+    },
+    {
+      "epoch": 1.5142378559463987,
+      "grad_norm": 0.36654895544052124,
+      "learning_rate": 0.0002,
+      "loss": 1.7694,
+      "step": 2260
+    },
+    {
+      "epoch": 1.5209380234505863,
+      "grad_norm": 0.3914054334163666,
+      "learning_rate": 0.0002,
+      "loss": 1.6812,
+      "step": 2270
+    },
+    {
+      "epoch": 1.5276381909547738,
+      "grad_norm": 0.42012137174606323,
+      "learning_rate": 0.0002,
+      "loss": 1.6822,
+      "step": 2280
+    },
+    {
+      "epoch": 1.5343383584589616,
+      "grad_norm": 0.39563435316085815,
+      "learning_rate": 0.0002,
+      "loss": 1.697,
+      "step": 2290
+    },
+    {
+      "epoch": 1.541038525963149,
+      "grad_norm": 0.3508438766002655,
+      "learning_rate": 0.0002,
+      "loss": 1.7491,
+      "step": 2300
+    },
+    {
+      "epoch": 1.5477386934673367,
+      "grad_norm": 0.3785218596458435,
+      "learning_rate": 0.0002,
+      "loss": 1.7727,
+      "step": 2310
+    },
+    {
+      "epoch": 1.5544388609715243,
+      "grad_norm": 0.39377647638320923,
+      "learning_rate": 0.0002,
+      "loss": 1.6963,
+      "step": 2320
+    },
+    {
+      "epoch": 1.5611390284757118,
+      "grad_norm": 0.3391438126564026,
+      "learning_rate": 0.0002,
+      "loss": 1.7263,
+      "step": 2330
+    },
+    {
+      "epoch": 1.5678391959798996,
+      "grad_norm": 0.37944263219833374,
+      "learning_rate": 0.0002,
+      "loss": 1.7722,
+      "step": 2340
+    },
+    {
+      "epoch": 1.574539363484087,
+      "grad_norm": 0.3523491322994232,
+      "learning_rate": 0.0002,
+      "loss": 1.6371,
+      "step": 2350
+    },
+    {
+      "epoch": 1.5812395309882747,
+      "grad_norm": 0.3911575973033905,
+      "learning_rate": 0.0002,
+      "loss": 1.7583,
+      "step": 2360
+    },
+    {
+      "epoch": 1.5879396984924623,
+      "grad_norm": 0.33832186460494995,
+      "learning_rate": 0.0002,
+      "loss": 1.7117,
+      "step": 2370
+    },
+    {
+      "epoch": 1.5946398659966499,
+      "grad_norm": 0.3665979206562042,
+      "learning_rate": 0.0002,
+      "loss": 1.7701,
+      "step": 2380
+    },
+    {
+      "epoch": 1.6013400335008376,
+      "grad_norm": 0.3871748149394989,
+      "learning_rate": 0.0002,
+      "loss": 1.779,
+      "step": 2390
+    },
+    {
+      "epoch": 1.608040201005025,
+      "grad_norm": 0.3586967885494232,
+      "learning_rate": 0.0002,
+      "loss": 1.7109,
+      "step": 2400
+    },
+    {
+      "epoch": 1.6147403685092128,
+      "grad_norm": 0.3563673198223114,
+      "learning_rate": 0.0002,
+      "loss": 1.7096,
+      "step": 2410
+    },
+    {
+      "epoch": 1.6214405360134003,
+      "grad_norm": 0.37588971853256226,
+      "learning_rate": 0.0002,
+      "loss": 1.745,
+      "step": 2420
+    },
+    {
+      "epoch": 1.6281407035175879,
+      "grad_norm": 0.352556437253952,
+      "learning_rate": 0.0002,
+      "loss": 1.7086,
+      "step": 2430
+    },
+    {
+      "epoch": 1.6348408710217757,
+      "grad_norm": 0.3716259300708771,
+      "learning_rate": 0.0002,
+      "loss": 1.6547,
+      "step": 2440
+    },
+    {
+      "epoch": 1.641541038525963,
+      "grad_norm": 0.372001975774765,
+      "learning_rate": 0.0002,
+      "loss": 1.7033,
+      "step": 2450
+    },
+    {
+      "epoch": 1.6482412060301508,
+      "grad_norm": 0.3430042862892151,
+      "learning_rate": 0.0002,
+      "loss": 1.6584,
+      "step": 2460
+    },
+    {
+      "epoch": 1.6549413735343383,
+      "grad_norm": 0.3741483688354492,
+      "learning_rate": 0.0002,
+      "loss": 1.7217,
+      "step": 2470
+    },
+    {
+      "epoch": 1.661641541038526,
+      "grad_norm": 0.3610571324825287,
+      "learning_rate": 0.0002,
+      "loss": 1.7701,
+      "step": 2480
+    },
+    {
+      "epoch": 1.6683417085427137,
+      "grad_norm": 0.4204719066619873,
+      "learning_rate": 0.0002,
+      "loss": 1.7057,
+      "step": 2490
+    },
+    {
+      "epoch": 1.675041876046901,
+      "grad_norm": 0.3938186466693878,
+      "learning_rate": 0.0002,
+      "loss": 1.7954,
+      "step": 2500
+    },
+    {
+      "epoch": 1.6817420435510888,
+      "grad_norm": 0.3421435058116913,
+      "learning_rate": 0.0002,
+      "loss": 1.6633,
+      "step": 2510
+    },
+    {
+      "epoch": 1.6884422110552764,
+      "grad_norm": 0.42441412806510925,
+      "learning_rate": 0.0002,
+      "loss": 1.7996,
+      "step": 2520
+    },
+    {
+      "epoch": 1.695142378559464,
+      "grad_norm": 0.38071519136428833,
+      "learning_rate": 0.0002,
+      "loss": 1.7142,
+      "step": 2530
+    },
+    {
+      "epoch": 1.7018425460636517,
+      "grad_norm": 0.34078919887542725,
+      "learning_rate": 0.0002,
+      "loss": 1.7232,
+      "step": 2540
+    },
+    {
+      "epoch": 1.708542713567839,
+      "grad_norm": 0.412844181060791,
+      "learning_rate": 0.0002,
+      "loss": 1.7126,
+      "step": 2550
+    },
+    {
+      "epoch": 1.7152428810720268,
+      "grad_norm": 0.3753604292869568,
+      "learning_rate": 0.0002,
+      "loss": 1.7149,
+      "step": 2560
+    },
+    {
+      "epoch": 1.7219430485762144,
+      "grad_norm": 0.41588476300239563,
+      "learning_rate": 0.0002,
+      "loss": 1.7011,
+      "step": 2570
+    },
+    {
+      "epoch": 1.728643216080402,
+      "grad_norm": 0.35504111647605896,
+      "learning_rate": 0.0002,
+      "loss": 1.6427,
+      "step": 2580
+    },
+    {
+      "epoch": 1.7353433835845897,
+      "grad_norm": 0.36909720301628113,
+      "learning_rate": 0.0002,
+      "loss": 1.7296,
+      "step": 2590
+    },
+    {
+      "epoch": 1.742043551088777,
+      "grad_norm": 0.4149979054927826,
+      "learning_rate": 0.0002,
+      "loss": 1.7022,
+      "step": 2600
+    },
+    {
+      "epoch": 1.7487437185929648,
+      "grad_norm": 0.38859328627586365,
+      "learning_rate": 0.0002,
+      "loss": 1.77,
+      "step": 2610
+    },
+    {
+      "epoch": 1.7554438860971524,
+      "grad_norm": 0.36738792061805725,
+      "learning_rate": 0.0002,
+      "loss": 1.7036,
+      "step": 2620
+    },
+    {
+      "epoch": 1.76214405360134,
+      "grad_norm": 0.3968178927898407,
+      "learning_rate": 0.0002,
+      "loss": 1.764,
+      "step": 2630
+    },
+    {
+      "epoch": 1.7688442211055277,
+      "grad_norm": 0.3972901999950409,
+      "learning_rate": 0.0002,
+      "loss": 1.7687,
+      "step": 2640
+    },
+    {
+      "epoch": 1.775544388609715,
+      "grad_norm": 0.3949959874153137,
+      "learning_rate": 0.0002,
+      "loss": 1.6774,
+      "step": 2650
+    },
+    {
+      "epoch": 1.7822445561139029,
+      "grad_norm": 0.44074657559394836,
+      "learning_rate": 0.0002,
+      "loss": 1.7247,
+      "step": 2660
+    },
+    {
+      "epoch": 1.7889447236180904,
+      "grad_norm": 0.39743664860725403,
+      "learning_rate": 0.0002,
+      "loss": 1.7188,
+      "step": 2670
+    },
+    {
+      "epoch": 1.795644891122278,
+      "grad_norm": 0.3950406610965729,
+      "learning_rate": 0.0002,
+      "loss": 1.7258,
+      "step": 2680
+    },
+    {
+      "epoch": 1.8023450586264658,
+      "grad_norm": 0.3568263649940491,
+      "learning_rate": 0.0002,
+      "loss": 1.6906,
+      "step": 2690
+    },
+    {
+      "epoch": 1.809045226130653,
+      "grad_norm": 0.3819476366043091,
+      "learning_rate": 0.0002,
+      "loss": 1.6735,
+      "step": 2700
+    },
+    {
+      "epoch": 1.8157453936348409,
+      "grad_norm": 0.3480634391307831,
+      "learning_rate": 0.0002,
+      "loss": 1.7198,
+      "step": 2710
+    },
+    {
+      "epoch": 1.8224455611390284,
+      "grad_norm": 0.3875853419303894,
+      "learning_rate": 0.0002,
+      "loss": 1.7042,
+      "step": 2720
+    },
+    {
+      "epoch": 1.829145728643216,
+      "grad_norm": 0.3441337049007416,
+      "learning_rate": 0.0002,
+      "loss": 1.6988,
+      "step": 2730
+    },
+    {
+      "epoch": 1.8358458961474038,
+      "grad_norm": 0.35692882537841797,
+      "learning_rate": 0.0002,
+      "loss": 1.7647,
+      "step": 2740
+    },
+    {
+      "epoch": 1.8425460636515911,
+      "grad_norm": 0.36959215998649597,
+      "learning_rate": 0.0002,
+      "loss": 1.7033,
+      "step": 2750
+    },
+    {
+      "epoch": 1.849246231155779,
+      "grad_norm": 0.3893393278121948,
+      "learning_rate": 0.0002,
+      "loss": 1.7657,
+      "step": 2760
+    },
+    {
+      "epoch": 1.8559463986599665,
+      "grad_norm": 0.37817293405532837,
+      "learning_rate": 0.0002,
+      "loss": 1.7068,
+      "step": 2770
+    },
+    {
+      "epoch": 1.862646566164154,
+      "grad_norm": 0.36071285605430603,
+      "learning_rate": 0.0002,
+      "loss": 1.761,
+      "step": 2780
+    },
+    {
+      "epoch": 1.8693467336683418,
+      "grad_norm": 0.3758420944213867,
+      "learning_rate": 0.0002,
+      "loss": 1.7623,
+      "step": 2790
+    },
+    {
+      "epoch": 1.8760469011725294,
+      "grad_norm": 0.3889938294887543,
+      "learning_rate": 0.0002,
+      "loss": 1.6743,
+      "step": 2800
+    },
+    {
+      "epoch": 1.882747068676717,
+      "grad_norm": 0.34361857175827026,
+      "learning_rate": 0.0002,
+      "loss": 1.6151,
+      "step": 2810
+    },
+    {
+      "epoch": 1.8894472361809045,
+      "grad_norm": 0.39283323287963867,
+      "learning_rate": 0.0002,
+      "loss": 1.6038,
+      "step": 2820
+    },
+    {
+      "epoch": 1.896147403685092,
+      "grad_norm": 0.3919452726840973,
+      "learning_rate": 0.0002,
+      "loss": 1.7555,
+      "step": 2830
+    },
+    {
+      "epoch": 1.9028475711892798,
+      "grad_norm": 0.38215070962905884,
+      "learning_rate": 0.0002,
+      "loss": 1.673,
+      "step": 2840
+    },
+    {
+      "epoch": 1.9095477386934674,
+      "grad_norm": 0.4235064387321472,
+      "learning_rate": 0.0002,
+      "loss": 1.7044,
+      "step": 2850
+    },
+    {
+      "epoch": 1.916247906197655,
+      "grad_norm": 0.35694634914398193,
+      "learning_rate": 0.0002,
+      "loss": 1.7123,
+      "step": 2860
+    },
+    {
+      "epoch": 1.9229480737018425,
+      "grad_norm": 0.383492112159729,
+      "learning_rate": 0.0002,
+      "loss": 1.8128,
+      "step": 2870
+    },
+    {
+      "epoch": 1.92964824120603,
+      "grad_norm": 0.5945147275924683,
+      "learning_rate": 0.0002,
+      "loss": 1.7581,
+      "step": 2880
+    },
+    {
+      "epoch": 1.9363484087102178,
+      "grad_norm": 0.3367522358894348,
+      "learning_rate": 0.0002,
+      "loss": 1.7421,
+      "step": 2890
+    },
+    {
+      "epoch": 1.9430485762144054,
+      "grad_norm": 0.35300394892692566,
+      "learning_rate": 0.0002,
+      "loss": 1.6561,
+      "step": 2900
+    },
+    {
+      "epoch": 1.949748743718593,
+      "grad_norm": 0.38084495067596436,
+      "learning_rate": 0.0002,
+      "loss": 1.7033,
+      "step": 2910
+    },
+    {
+      "epoch": 1.9564489112227805,
+      "grad_norm": 0.37559160590171814,
+      "learning_rate": 0.0002,
+      "loss": 1.7132,
+      "step": 2920
+    },
+    {
+      "epoch": 1.963149078726968,
+      "grad_norm": 0.3661738336086273,
+      "learning_rate": 0.0002,
+      "loss": 1.6759,
+      "step": 2930
+    },
+    {
+      "epoch": 1.9698492462311559,
+      "grad_norm": 0.4073849320411682,
+      "learning_rate": 0.0002,
+      "loss": 1.7643,
+      "step": 2940
+    },
+    {
+      "epoch": 1.9765494137353434,
+      "grad_norm": 0.3723304271697998,
+      "learning_rate": 0.0002,
+      "loss": 1.6806,
+      "step": 2950
+    },
+    {
+      "epoch": 1.983249581239531,
+      "grad_norm": 0.3991098999977112,
+      "learning_rate": 0.0002,
+      "loss": 1.7611,
+      "step": 2960
+    },
+    {
+      "epoch": 1.9899497487437185,
+      "grad_norm": 0.3947085440158844,
+      "learning_rate": 0.0002,
+      "loss": 1.7263,
+      "step": 2970
+    },
+    {
+      "epoch": 1.996649916247906,
+      "grad_norm": 0.3786258399486542,
+      "learning_rate": 0.0002,
+      "loss": 1.7217,
+      "step": 2980
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 1.8028968572616577,
+      "eval_runtime": 37.8985,
+      "eval_samples_per_second": 13.589,
+      "eval_steps_per_second": 1.715,
+      "step": 2985
+    },
+    {
+      "epoch": 2.003350083752094,
+      "grad_norm": 0.34824079275131226,
+      "learning_rate": 0.0002,
+      "loss": 1.695,
+      "step": 2990
+    },
+    {
+      "epoch": 2.0100502512562812,
+      "grad_norm": 0.3394894003868103,
+      "learning_rate": 0.0002,
+      "loss": 1.5853,
+      "step": 3000
+    },
+    {
+      "epoch": 2.016750418760469,
+      "grad_norm": 0.36910977959632874,
+      "learning_rate": 0.0002,
+      "loss": 1.5783,
+      "step": 3010
+    },
+    {
+      "epoch": 2.023450586264657,
+      "grad_norm": 0.45000967383384705,
+      "learning_rate": 0.0002,
+      "loss": 1.6105,
+      "step": 3020
+    },
+    {
+      "epoch": 2.030150753768844,
+      "grad_norm": 0.3791407346725464,
+      "learning_rate": 0.0002,
+      "loss": 1.6019,
+      "step": 3030
+    },
+    {
+      "epoch": 2.036850921273032,
+      "grad_norm": 0.387321799993515,
+      "learning_rate": 0.0002,
+      "loss": 1.5832,
+      "step": 3040
+    },
+    {
+      "epoch": 2.0435510887772192,
+      "grad_norm": 0.4185757040977478,
+      "learning_rate": 0.0002,
+      "loss": 1.6834,
+      "step": 3050
+    },
+    {
+      "epoch": 2.050251256281407,
+      "grad_norm": 0.45110777020454407,
+      "learning_rate": 0.0002,
+      "loss": 1.5696,
+      "step": 3060
+    },
+    {
+      "epoch": 2.056951423785595,
+      "grad_norm": 0.42663660645484924,
+      "learning_rate": 0.0002,
+      "loss": 1.6231,
+      "step": 3070
+    },
+    {
+      "epoch": 2.063651591289782,
+      "grad_norm": 0.4546292722225189,
+      "learning_rate": 0.0002,
+      "loss": 1.6279,
+      "step": 3080
+    },
+    {
+      "epoch": 2.07035175879397,
+      "grad_norm": 0.3979759216308594,
+      "learning_rate": 0.0002,
+      "loss": 1.6141,
+      "step": 3090
+    },
+    {
+      "epoch": 2.0770519262981573,
+      "grad_norm": 0.43596673011779785,
+      "learning_rate": 0.0002,
+      "loss": 1.6343,
+      "step": 3100
+    },
+    {
+      "epoch": 2.083752093802345,
+      "grad_norm": 0.40120232105255127,
+      "learning_rate": 0.0002,
+      "loss": 1.5441,
+      "step": 3110
+    },
+    {
+      "epoch": 2.090452261306533,
+      "grad_norm": 0.44449281692504883,
+      "learning_rate": 0.0002,
+      "loss": 1.6309,
+      "step": 3120
+    },
+    {
+      "epoch": 2.09715242881072,
+      "grad_norm": 0.42672568559646606,
+      "learning_rate": 0.0002,
+      "loss": 1.5652,
+      "step": 3130
+    },
+    {
+      "epoch": 2.103852596314908,
+      "grad_norm": 0.4232690930366516,
+      "learning_rate": 0.0002,
+      "loss": 1.682,
+      "step": 3140
+    },
+    {
+      "epoch": 2.1105527638190953,
+      "grad_norm": 0.4299317002296448,
+      "learning_rate": 0.0002,
+      "loss": 1.624,
+      "step": 3150
+    },
+    {
+      "epoch": 2.117252931323283,
+      "grad_norm": 0.4067758023738861,
+      "learning_rate": 0.0002,
+      "loss": 1.6766,
+      "step": 3160
+    },
+    {
+      "epoch": 2.123953098827471,
+      "grad_norm": 0.4918815791606903,
+      "learning_rate": 0.0002,
+      "loss": 1.6759,
+      "step": 3170
+    },
+    {
+      "epoch": 2.130653266331658,
+      "grad_norm": 0.4140559732913971,
+      "learning_rate": 0.0002,
+      "loss": 1.6478,
+      "step": 3180
+    },
+    {
+      "epoch": 2.137353433835846,
+      "grad_norm": 0.4555995464324951,
+      "learning_rate": 0.0002,
+      "loss": 1.6641,
+      "step": 3190
+    },
+    {
+      "epoch": 2.1440536013400333,
+      "grad_norm": 0.42943915724754333,
+      "learning_rate": 0.0002,
+      "loss": 1.5888,
+      "step": 3200
+    },
+    {
+      "epoch": 2.150753768844221,
+      "grad_norm": 0.4730435013771057,
+      "learning_rate": 0.0002,
+      "loss": 1.5886,
+      "step": 3210
+    },
+    {
+      "epoch": 2.157453936348409,
+      "grad_norm": 0.43310216069221497,
+      "learning_rate": 0.0002,
+      "loss": 1.6022,
+      "step": 3220
+    },
+    {
+      "epoch": 2.164154103852596,
+      "grad_norm": 0.42054110765457153,
+      "learning_rate": 0.0002,
+      "loss": 1.6058,
+      "step": 3230
+    },
+    {
+      "epoch": 2.170854271356784,
+      "grad_norm": 0.4897233247756958,
+      "learning_rate": 0.0002,
+      "loss": 1.6749,
+      "step": 3240
+    },
+    {
+      "epoch": 2.1775544388609713,
+      "grad_norm": 0.42194533348083496,
+      "learning_rate": 0.0002,
+      "loss": 1.6983,
+      "step": 3250
+    },
+    {
+      "epoch": 2.184254606365159,
+      "grad_norm": 0.44494450092315674,
+      "learning_rate": 0.0002,
+      "loss": 1.6159,
+      "step": 3260
+    },
+    {
+      "epoch": 2.190954773869347,
+      "grad_norm": 0.43524879217147827,
+      "learning_rate": 0.0002,
+      "loss": 1.6977,
+      "step": 3270
+    },
+    {
+      "epoch": 2.1976549413735342,
+      "grad_norm": 0.4621117413043976,
+      "learning_rate": 0.0002,
+      "loss": 1.528,
+      "step": 3280
+    },
+    {
+      "epoch": 2.204355108877722,
+      "grad_norm": 0.4073285460472107,
+      "learning_rate": 0.0002,
+      "loss": 1.632,
+      "step": 3290
+    },
+    {
+      "epoch": 2.2110552763819094,
+      "grad_norm": 0.47868335247039795,
+      "learning_rate": 0.0002,
+      "loss": 1.6141,
+      "step": 3300
+    },
+    {
+      "epoch": 2.217755443886097,
+      "grad_norm": 0.4264970123767853,
+      "learning_rate": 0.0002,
+      "loss": 1.6857,
+      "step": 3310
+    },
+    {
+      "epoch": 2.224455611390285,
+      "grad_norm": 0.4491245150566101,
+      "learning_rate": 0.0002,
+      "loss": 1.5653,
+      "step": 3320
+    },
+    {
+      "epoch": 2.2311557788944723,
+      "grad_norm": 0.4010344445705414,
+      "learning_rate": 0.0002,
+      "loss": 1.5881,
+      "step": 3330
+    },
+    {
+      "epoch": 2.23785594639866,
+      "grad_norm": 0.4232759177684784,
+      "learning_rate": 0.0002,
+      "loss": 1.6684,
+      "step": 3340
+    },
+    {
+      "epoch": 2.2445561139028474,
+      "grad_norm": 0.5099776983261108,
+      "learning_rate": 0.0002,
+      "loss": 1.6336,
+      "step": 3350
+    },
+    {
+      "epoch": 2.251256281407035,
+      "grad_norm": 0.5223407745361328,
+      "learning_rate": 0.0002,
+      "loss": 1.6764,
+      "step": 3360
+    },
+    {
+      "epoch": 2.257956448911223,
+      "grad_norm": 0.47818470001220703,
+      "learning_rate": 0.0002,
+      "loss": 1.6625,
+      "step": 3370
+    },
+    {
+      "epoch": 2.2646566164154103,
+      "grad_norm": 0.4721255898475647,
+      "learning_rate": 0.0002,
+      "loss": 1.5946,
+      "step": 3380
+    },
+    {
+      "epoch": 2.271356783919598,
+      "grad_norm": 0.4113229513168335,
+      "learning_rate": 0.0002,
+      "loss": 1.5568,
+      "step": 3390
+    },
+    {
+      "epoch": 2.2780569514237854,
+      "grad_norm": 0.507080078125,
+      "learning_rate": 0.0002,
+      "loss": 1.6494,
+      "step": 3400
+    },
+    {
+      "epoch": 2.284757118927973,
+      "grad_norm": 0.4852292239665985,
+      "learning_rate": 0.0002,
+      "loss": 1.6183,
+      "step": 3410
+    },
+    {
+      "epoch": 2.291457286432161,
+      "grad_norm": 0.4503684341907501,
+      "learning_rate": 0.0002,
+      "loss": 1.6132,
+      "step": 3420
+    },
+    {
+      "epoch": 2.2981574539363483,
+      "grad_norm": 0.8359600305557251,
+      "learning_rate": 0.0002,
+      "loss": 1.6649,
+      "step": 3430
+    },
+    {
+      "epoch": 2.304857621440536,
+      "grad_norm": 0.44604045152664185,
+      "learning_rate": 0.0002,
+      "loss": 1.6644,
+      "step": 3440
+    },
+    {
+      "epoch": 2.3115577889447234,
+      "grad_norm": 0.45667049288749695,
+      "learning_rate": 0.0002,
+      "loss": 1.5972,
+      "step": 3450
+    },
+    {
+      "epoch": 2.318257956448911,
+      "grad_norm": 0.4879349172115326,
+      "learning_rate": 0.0002,
+      "loss": 1.6667,
+      "step": 3460
+    },
+    {
+      "epoch": 2.324958123953099,
+      "grad_norm": 0.4033963084220886,
+      "learning_rate": 0.0002,
+      "loss": 1.5804,
+      "step": 3470
+    },
+    {
+      "epoch": 2.3316582914572863,
+      "grad_norm": 0.44494301080703735,
+      "learning_rate": 0.0002,
+      "loss": 1.5838,
+      "step": 3480
+    },
+    {
+      "epoch": 2.338358458961474,
+      "grad_norm": 0.4794621765613556,
+      "learning_rate": 0.0002,
+      "loss": 1.5951,
+      "step": 3490
+    },
+    {
+      "epoch": 2.3450586264656614,
+      "grad_norm": 0.41404327750205994,
+      "learning_rate": 0.0002,
+      "loss": 1.6807,
+      "step": 3500
+    },
+    {
+      "epoch": 2.351758793969849,
+      "grad_norm": 0.4664851725101471,
+      "learning_rate": 0.0002,
+      "loss": 1.714,
+      "step": 3510
+    },
+    {
+      "epoch": 2.358458961474037,
+      "grad_norm": 0.4263697564601898,
+      "learning_rate": 0.0002,
+      "loss": 1.6537,
+      "step": 3520
+    },
+    {
+      "epoch": 2.3651591289782243,
+      "grad_norm": 0.5035167336463928,
+      "learning_rate": 0.0002,
+      "loss": 1.6551,
+      "step": 3530
+    },
+    {
+      "epoch": 2.371859296482412,
+      "grad_norm": 0.4380664527416229,
+      "learning_rate": 0.0002,
+      "loss": 1.6208,
+      "step": 3540
+    },
+    {
+      "epoch": 2.3785594639865995,
+      "grad_norm": 0.5227681994438171,
+      "learning_rate": 0.0002,
+      "loss": 1.634,
+      "step": 3550
+    },
+    {
+      "epoch": 2.3852596314907872,
+      "grad_norm": 0.4382302761077881,
+      "learning_rate": 0.0002,
+      "loss": 1.6146,
+      "step": 3560
+    },
+    {
+      "epoch": 2.391959798994975,
+      "grad_norm": 0.4392451047897339,
+      "learning_rate": 0.0002,
+      "loss": 1.5653,
+      "step": 3570
+    },
+    {
+      "epoch": 2.3986599664991624,
+      "grad_norm": 0.4372786581516266,
+      "learning_rate": 0.0002,
+      "loss": 1.6626,
+      "step": 3580
+    },
+    {
+      "epoch": 2.40536013400335,
+      "grad_norm": 0.5015502572059631,
+      "learning_rate": 0.0002,
+      "loss": 1.519,
+      "step": 3590
+    },
+    {
+      "epoch": 2.4120603015075375,
+      "grad_norm": 0.5653210878372192,
+      "learning_rate": 0.0002,
+      "loss": 1.588,
+      "step": 3600
+    },
+    {
+      "epoch": 2.4187604690117253,
+      "grad_norm": 0.53007972240448,
+      "learning_rate": 0.0002,
+      "loss": 1.6075,
+      "step": 3610
+    },
+    {
+      "epoch": 2.425460636515913,
+      "grad_norm": 0.4659176766872406,
+      "learning_rate": 0.0002,
+      "loss": 1.6421,
+      "step": 3620
+    },
+    {
+      "epoch": 2.4321608040201004,
+      "grad_norm": 0.5637837052345276,
+      "learning_rate": 0.0002,
+      "loss": 1.625,
+      "step": 3630
+    },
+    {
+      "epoch": 2.438860971524288,
+      "grad_norm": 0.4248391389846802,
+      "learning_rate": 0.0002,
+      "loss": 1.6168,
+      "step": 3640
+    },
+    {
+      "epoch": 2.4455611390284755,
+      "grad_norm": 0.44668248295783997,
+      "learning_rate": 0.0002,
+      "loss": 1.6822,
+      "step": 3650
+    },
+    {
+      "epoch": 2.4522613065326633,
+      "grad_norm": 0.43990179896354675,
+      "learning_rate": 0.0002,
+      "loss": 1.6417,
+      "step": 3660
+    },
+    {
+      "epoch": 2.458961474036851,
+      "grad_norm": 0.4532523453235626,
+      "learning_rate": 0.0002,
+      "loss": 1.6723,
+      "step": 3670
+    },
+    {
+      "epoch": 2.4656616415410384,
+      "grad_norm": 0.6605591773986816,
+      "learning_rate": 0.0002,
+      "loss": 1.6957,
+      "step": 3680
+    },
+    {
+      "epoch": 2.472361809045226,
+      "grad_norm": 0.4694533348083496,
+      "learning_rate": 0.0002,
+      "loss": 1.6159,
+      "step": 3690
+    },
+    {
+      "epoch": 2.4790619765494135,
+      "grad_norm": 0.4485011100769043,
+      "learning_rate": 0.0002,
+      "loss": 1.6239,
+      "step": 3700
+    },
+    {
+      "epoch": 2.4857621440536013,
+      "grad_norm": 0.4761785864830017,
+      "learning_rate": 0.0002,
+      "loss": 1.6834,
+      "step": 3710
+    },
+    {
+      "epoch": 2.492462311557789,
+      "grad_norm": 0.5116432309150696,
+      "learning_rate": 0.0002,
+      "loss": 1.6313,
+      "step": 3720
+    },
+    {
+      "epoch": 2.4991624790619764,
+      "grad_norm": 0.49523618817329407,
+      "learning_rate": 0.0002,
+      "loss": 1.5054,
+      "step": 3730
+    },
+    {
+      "epoch": 2.505862646566164,
+      "grad_norm": 0.43826380372047424,
+      "learning_rate": 0.0002,
+      "loss": 1.6249,
+      "step": 3740
+    },
+    {
+      "epoch": 2.5125628140703515,
+      "grad_norm": 0.4916154146194458,
+      "learning_rate": 0.0002,
+      "loss": 1.5762,
+      "step": 3750
+    },
+    {
+      "epoch": 2.5192629815745393,
+      "grad_norm": 0.5381299257278442,
+      "learning_rate": 0.0002,
+      "loss": 1.5157,
+      "step": 3760
+    },
+    {
+      "epoch": 2.525963149078727,
+      "grad_norm": 0.44947415590286255,
+      "learning_rate": 0.0002,
+      "loss": 1.6467,
+      "step": 3770
+    },
+    {
+      "epoch": 2.5326633165829144,
+      "grad_norm": 0.49979084730148315,
+      "learning_rate": 0.0002,
+      "loss": 1.67,
+      "step": 3780
+    },
+    {
+      "epoch": 2.539363484087102,
+      "grad_norm": 0.43046900629997253,
+      "learning_rate": 0.0002,
+      "loss": 1.622,
+      "step": 3790
+    },
+    {
+      "epoch": 2.5460636515912896,
+      "grad_norm": 0.4513470530509949,
+      "learning_rate": 0.0002,
+      "loss": 1.6789,
+      "step": 3800
+    },
+    {
+      "epoch": 2.5527638190954773,
+      "grad_norm": 0.49900051951408386,
+      "learning_rate": 0.0002,
+      "loss": 1.6335,
+      "step": 3810
+    },
+    {
+      "epoch": 2.559463986599665,
+      "grad_norm": 0.4348420202732086,
+      "learning_rate": 0.0002,
+      "loss": 1.6322,
+      "step": 3820
+    },
+    {
+      "epoch": 2.5661641541038525,
+      "grad_norm": 0.4684867560863495,
+      "learning_rate": 0.0002,
+      "loss": 1.6218,
+      "step": 3830
+    },
+    {
+      "epoch": 2.5728643216080402,
+      "grad_norm": 0.44430989027023315,
+      "learning_rate": 0.0002,
+      "loss": 1.6535,
+      "step": 3840
+    },
+    {
+      "epoch": 2.5795644891122276,
+      "grad_norm": 0.47375255823135376,
+      "learning_rate": 0.0002,
+      "loss": 1.5909,
+      "step": 3850
+    },
+    {
+      "epoch": 2.5862646566164154,
+      "grad_norm": 0.45493075251579285,
+      "learning_rate": 0.0002,
+      "loss": 1.6269,
+      "step": 3860
+    },
+    {
+      "epoch": 2.592964824120603,
+      "grad_norm": 0.4563275873661041,
+      "learning_rate": 0.0002,
+      "loss": 1.604,
+      "step": 3870
+    },
+    {
+      "epoch": 2.5996649916247905,
+      "grad_norm": 0.46060335636138916,
+      "learning_rate": 0.0002,
+      "loss": 1.642,
+      "step": 3880
+    },
+    {
+      "epoch": 2.6063651591289783,
+      "grad_norm": 0.4718867540359497,
+      "learning_rate": 0.0002,
+      "loss": 1.6302,
+      "step": 3890
+    },
+    {
+      "epoch": 2.6130653266331656,
+      "grad_norm": 0.41570305824279785,
+      "learning_rate": 0.0002,
+      "loss": 1.6242,
+      "step": 3900
+    },
+    {
+      "epoch": 2.6197654941373534,
+      "grad_norm": 0.4603121876716614,
+      "learning_rate": 0.0002,
+      "loss": 1.6401,
+      "step": 3910
+    },
+    {
+      "epoch": 2.626465661641541,
+      "grad_norm": 0.4734652638435364,
+      "learning_rate": 0.0002,
+      "loss": 1.6839,
+      "step": 3920
+    },
+    {
+      "epoch": 2.6331658291457285,
+      "grad_norm": 0.45348483324050903,
+      "learning_rate": 0.0002,
+      "loss": 1.5448,
+      "step": 3930
+    },
+    {
+      "epoch": 2.6398659966499163,
+      "grad_norm": 0.46559447050094604,
+      "learning_rate": 0.0002,
+      "loss": 1.6157,
+      "step": 3940
+    },
+    {
+      "epoch": 2.6465661641541036,
+      "grad_norm": 0.44113144278526306,
+      "learning_rate": 0.0002,
+      "loss": 1.7052,
+      "step": 3950
+    },
+    {
+      "epoch": 2.6532663316582914,
+      "grad_norm": 0.41415104269981384,
+      "learning_rate": 0.0002,
+      "loss": 1.6315,
+      "step": 3960
+    },
+    {
+      "epoch": 2.659966499162479,
+      "grad_norm": 0.48868080973625183,
+      "learning_rate": 0.0002,
+      "loss": 1.6589,
+      "step": 3970
+    },
+    {
+      "epoch": 2.6666666666666665,
+      "grad_norm": 0.49610549211502075,
+      "learning_rate": 0.0002,
+      "loss": 1.6211,
+      "step": 3980
+    },
+    {
+      "epoch": 2.6733668341708543,
+      "grad_norm": 0.4309130907058716,
+      "learning_rate": 0.0002,
+      "loss": 1.6235,
+      "step": 3990
+    },
+    {
+      "epoch": 2.6800670016750416,
+      "grad_norm": 0.4489327669143677,
+      "learning_rate": 0.0002,
+      "loss": 1.6452,
+      "step": 4000
+    },
+    {
+      "epoch": 2.6867671691792294,
+      "grad_norm": 0.5380139946937561,
+      "learning_rate": 0.0002,
+      "loss": 1.5744,
+      "step": 4010
+    },
+    {
+      "epoch": 2.693467336683417,
+      "grad_norm": 0.5076672434806824,
+      "learning_rate": 0.0002,
+      "loss": 1.6524,
+      "step": 4020
+    },
+    {
+      "epoch": 2.7001675041876045,
+      "grad_norm": 0.47620031237602234,
+      "learning_rate": 0.0002,
+      "loss": 1.636,
+      "step": 4030
+    },
+    {
+      "epoch": 2.7068676716917923,
+      "grad_norm": 0.48089155554771423,
+      "learning_rate": 0.0002,
+      "loss": 1.5543,
+      "step": 4040
+    },
+    {
+      "epoch": 2.7135678391959797,
+      "grad_norm": 0.5108814239501953,
+      "learning_rate": 0.0002,
+      "loss": 1.6396,
+      "step": 4050
+    },
+    {
+      "epoch": 2.7202680067001674,
+      "grad_norm": 0.4196513295173645,
+      "learning_rate": 0.0002,
+      "loss": 1.5905,
+      "step": 4060
+    },
+    {
+      "epoch": 2.726968174204355,
+      "grad_norm": 0.4574664831161499,
+      "learning_rate": 0.0002,
+      "loss": 1.686,
+      "step": 4070
+    },
+    {
+      "epoch": 2.7336683417085426,
+      "grad_norm": 0.4671640992164612,
+      "learning_rate": 0.0002,
+      "loss": 1.6234,
+      "step": 4080
+    },
+    {
+      "epoch": 2.7403685092127303,
+      "grad_norm": 0.49355530738830566,
+      "learning_rate": 0.0002,
+      "loss": 1.6827,
+      "step": 4090
+    },
+    {
+      "epoch": 2.7470686767169177,
+      "grad_norm": 0.46716663241386414,
+      "learning_rate": 0.0002,
+      "loss": 1.6999,
+      "step": 4100
+    },
+    {
+      "epoch": 2.7537688442211055,
+      "grad_norm": 0.45420581102371216,
+      "learning_rate": 0.0002,
+      "loss": 1.6463,
+      "step": 4110
+    },
+    {
+      "epoch": 2.7604690117252932,
+      "grad_norm": 0.4680487811565399,
+      "learning_rate": 0.0002,
+      "loss": 1.5718,
+      "step": 4120
+    },
+    {
+      "epoch": 2.7671691792294806,
+      "grad_norm": 0.5375032424926758,
+      "learning_rate": 0.0002,
+      "loss": 1.5968,
+      "step": 4130
+    },
+    {
+      "epoch": 2.7738693467336684,
+      "grad_norm": 0.46026280522346497,
+      "learning_rate": 0.0002,
+      "loss": 1.5254,
+      "step": 4140
+    },
+    {
+      "epoch": 2.7805695142378557,
+      "grad_norm": 0.43658447265625,
+      "learning_rate": 0.0002,
+      "loss": 1.6613,
+      "step": 4150
+    },
+    {
+      "epoch": 2.7872696817420435,
+      "grad_norm": 0.4935547113418579,
+      "learning_rate": 0.0002,
+      "loss": 1.6546,
+      "step": 4160
+    },
+    {
+      "epoch": 2.7939698492462313,
+      "grad_norm": 0.8167962431907654,
+      "learning_rate": 0.0002,
+      "loss": 1.5961,
+      "step": 4170
+    },
+    {
+      "epoch": 2.8006700167504186,
+      "grad_norm": 0.4289683997631073,
+      "learning_rate": 0.0002,
+      "loss": 1.6907,
+      "step": 4180
+    },
+    {
+      "epoch": 2.8073701842546064,
+      "grad_norm": 0.4569324254989624,
+      "learning_rate": 0.0002,
+      "loss": 1.6385,
+      "step": 4190
+    },
+    {
+      "epoch": 2.8140703517587937,
+      "grad_norm": 0.474795937538147,
+      "learning_rate": 0.0002,
+      "loss": 1.6077,
+      "step": 4200
+    },
+    {
+      "epoch": 2.8207705192629815,
+      "grad_norm": 0.44272229075431824,
+      "learning_rate": 0.0002,
+      "loss": 1.6223,
+      "step": 4210
+    },
+    {
+      "epoch": 2.8274706867671693,
+      "grad_norm": 0.525240957736969,
+      "learning_rate": 0.0002,
+      "loss": 1.6706,
+      "step": 4220
+    },
+    {
+      "epoch": 2.8341708542713566,
+      "grad_norm": 0.4802303910255432,
+      "learning_rate": 0.0002,
+      "loss": 1.7196,
+      "step": 4230
+    },
+    {
+      "epoch": 2.8408710217755444,
+      "grad_norm": 0.46400442719459534,
+      "learning_rate": 0.0002,
+      "loss": 1.6002,
+      "step": 4240
+    },
+    {
+      "epoch": 2.8475711892797317,
+      "grad_norm": 0.49884888529777527,
+      "learning_rate": 0.0002,
+      "loss": 1.6052,
+      "step": 4250
+    },
+    {
+      "epoch": 2.8542713567839195,
+      "grad_norm": 0.5015072226524353,
+      "learning_rate": 0.0002,
+      "loss": 1.6919,
+      "step": 4260
+    },
+    {
+      "epoch": 2.8609715242881073,
+      "grad_norm": 0.4335440695285797,
+      "learning_rate": 0.0002,
+      "loss": 1.6335,
+      "step": 4270
+    },
+    {
+      "epoch": 2.8676716917922946,
+      "grad_norm": 0.5131644606590271,
+      "learning_rate": 0.0002,
+      "loss": 1.5664,
+      "step": 4280
+    },
+    {
+      "epoch": 2.8743718592964824,
+      "grad_norm": 0.6977195739746094,
+      "learning_rate": 0.0002,
+      "loss": 1.6409,
+      "step": 4290
+    },
+    {
+      "epoch": 2.8810720268006698,
+      "grad_norm": 0.5133762955665588,
+      "learning_rate": 0.0002,
+      "loss": 1.7192,
+      "step": 4300
+    },
+    {
+      "epoch": 2.8877721943048575,
+      "grad_norm": 0.4737614393234253,
+      "learning_rate": 0.0002,
+      "loss": 1.6257,
+      "step": 4310
+    },
+    {
+      "epoch": 2.8944723618090453,
+      "grad_norm": 0.4580535590648651,
+      "learning_rate": 0.0002,
+      "loss": 1.6076,
+      "step": 4320
+    },
+    {
+      "epoch": 2.901172529313233,
+      "grad_norm": 0.43863341212272644,
+      "learning_rate": 0.0002,
+      "loss": 1.6538,
+      "step": 4330
+    },
+    {
+      "epoch": 2.9078726968174204,
+      "grad_norm": 0.4103737473487854,
+      "learning_rate": 0.0002,
+      "loss": 1.6091,
+      "step": 4340
+    },
+    {
+      "epoch": 2.914572864321608,
+      "grad_norm": 0.438014417886734,
+      "learning_rate": 0.0002,
+      "loss": 1.7106,
+      "step": 4350
+    },
+    {
+      "epoch": 2.9212730318257956,
+      "grad_norm": 0.5068213939666748,
+      "learning_rate": 0.0002,
+      "loss": 1.6025,
+      "step": 4360
+    },
+    {
+      "epoch": 2.9279731993299833,
+      "grad_norm": 0.45305484533309937,
+      "learning_rate": 0.0002,
+      "loss": 1.6426,
+      "step": 4370
+    },
+    {
+      "epoch": 2.934673366834171,
+      "grad_norm": 0.4612090289592743,
+      "learning_rate": 0.0002,
+      "loss": 1.5726,
+      "step": 4380
+    },
+    {
+      "epoch": 2.9413735343383585,
+      "grad_norm": 0.508736789226532,
+      "learning_rate": 0.0002,
+      "loss": 1.6536,
+      "step": 4390
+    },
+    {
+      "epoch": 2.948073701842546,
+      "grad_norm": 0.4924427270889282,
+      "learning_rate": 0.0002,
+      "loss": 1.6132,
+      "step": 4400
+    },
+    {
+      "epoch": 2.9547738693467336,
+      "grad_norm": 0.5707460641860962,
+      "learning_rate": 0.0002,
+      "loss": 1.7007,
+      "step": 4410
+    },
+    {
+      "epoch": 2.9614740368509214,
+      "grad_norm": 0.42270299792289734,
+      "learning_rate": 0.0002,
+      "loss": 1.6814,
+      "step": 4420
+    },
+    {
+      "epoch": 2.968174204355109,
+      "grad_norm": 0.4429931044578552,
+      "learning_rate": 0.0002,
+      "loss": 1.6644,
+      "step": 4430
+    },
+    {
+      "epoch": 2.9748743718592965,
+      "grad_norm": 0.49760574102401733,
+      "learning_rate": 0.0002,
+      "loss": 1.6251,
+      "step": 4440
+    },
+    {
+      "epoch": 2.981574539363484,
+      "grad_norm": 0.4558229148387909,
+      "learning_rate": 0.0002,
+      "loss": 1.6169,
+      "step": 4450
+    },
+    {
+      "epoch": 2.9882747068676716,
+      "grad_norm": 0.39848530292510986,
+      "learning_rate": 0.0002,
+      "loss": 1.6055,
+      "step": 4460
+    },
+    {
+      "epoch": 2.9949748743718594,
+      "grad_norm": 0.5224862098693848,
+      "learning_rate": 0.0002,
+      "loss": 1.6705,
+      "step": 4470
+    },
+    {
+      "epoch": 2.9996649916247904,
+      "eval_loss": 1.8228833675384521,
+      "eval_runtime": 37.9049,
+      "eval_samples_per_second": 13.587,
+      "eval_steps_per_second": 1.715,
+      "step": 4477
+    },
+    {
+      "epoch": 3.0016750418760467,
+      "grad_norm": 0.41169142723083496,
+      "learning_rate": 0.0002,
+      "loss": 1.6637,
+      "step": 4480
+    },
+    {
+      "epoch": 3.0083752093802345,
+      "grad_norm": 0.4865207374095917,
+      "learning_rate": 0.0002,
+      "loss": 1.5974,
+      "step": 4490
+    },
+    {
+      "epoch": 3.0150753768844223,
+      "grad_norm": 0.5462028384208679,
+      "learning_rate": 0.0002,
+      "loss": 1.5297,
+      "step": 4500
+    },
+    {
+      "epoch": 3.0217755443886096,
+      "grad_norm": 0.6169732809066772,
+      "learning_rate": 0.0002,
+      "loss": 1.5251,
+      "step": 4510
+    },
+    {
+      "epoch": 3.0284757118927974,
+      "grad_norm": 0.5667954087257385,
+      "learning_rate": 0.0002,
+      "loss": 1.5559,
+      "step": 4520
+    },
+    {
+      "epoch": 3.0351758793969847,
+      "grad_norm": 0.5758325457572937,
+      "learning_rate": 0.0002,
+      "loss": 1.5037,
+      "step": 4530
+    },
+    {
+      "epoch": 3.0418760469011725,
+      "grad_norm": 0.5220064520835876,
+      "learning_rate": 0.0002,
+      "loss": 1.4873,
+      "step": 4540
+    },
+    {
+      "epoch": 3.0485762144053603,
+      "grad_norm": 0.5469558835029602,
+      "learning_rate": 0.0002,
+      "loss": 1.5126,
+      "step": 4550
+    },
+    {
+      "epoch": 3.0552763819095476,
+      "grad_norm": 0.5680848956108093,
+      "learning_rate": 0.0002,
+      "loss": 1.4275,
+      "step": 4560
+    },
+    {
+      "epoch": 3.0619765494137354,
+      "grad_norm": 0.5906574726104736,
+      "learning_rate": 0.0002,
+      "loss": 1.5187,
+      "step": 4570
+    },
+    {
+      "epoch": 3.0686767169179228,
+      "grad_norm": 0.4725631773471832,
+      "learning_rate": 0.0002,
+      "loss": 1.4551,
+      "step": 4580
+    },
+    {
+      "epoch": 3.0753768844221105,
+      "grad_norm": 0.5273477435112,
+      "learning_rate": 0.0002,
+      "loss": 1.5083,
+      "step": 4590
+    },
+    {
+      "epoch": 3.0820770519262983,
+      "grad_norm": 0.5861203074455261,
+      "learning_rate": 0.0002,
+      "loss": 1.5154,
+      "step": 4600
+    },
+    {
+      "epoch": 3.0887772194304857,
+      "grad_norm": 0.5343965291976929,
+      "learning_rate": 0.0002,
+      "loss": 1.4924,
+      "step": 4610
+    },
+    {
+      "epoch": 3.0954773869346734,
+      "grad_norm": 0.5348150730133057,
+      "learning_rate": 0.0002,
+      "loss": 1.5608,
+      "step": 4620
+    },
+    {
+      "epoch": 3.102177554438861,
+      "grad_norm": 0.5971846580505371,
+      "learning_rate": 0.0002,
+      "loss": 1.5399,
+      "step": 4630
+    },
+    {
+      "epoch": 3.1088777219430486,
+      "grad_norm": 0.5203177332878113,
+      "learning_rate": 0.0002,
+      "loss": 1.4662,
+      "step": 4640
+    },
+    {
+      "epoch": 3.1155778894472363,
+      "grad_norm": 0.55289226770401,
+      "learning_rate": 0.0002,
+      "loss": 1.5805,
+      "step": 4650
+    },
+    {
+      "epoch": 3.1222780569514237,
+      "grad_norm": 0.6878530979156494,
+      "learning_rate": 0.0002,
+      "loss": 1.4745,
+      "step": 4660
+    },
+    {
+      "epoch": 3.1289782244556115,
+      "grad_norm": 0.6173256635665894,
+      "learning_rate": 0.0002,
+      "loss": 1.5335,
+      "step": 4670
+    },
+    {
+      "epoch": 3.135678391959799,
+      "grad_norm": 0.536796510219574,
+      "learning_rate": 0.0002,
+      "loss": 1.51,
+      "step": 4680
+    },
+    {
+      "epoch": 3.1423785594639866,
+      "grad_norm": 0.58846116065979,
+      "learning_rate": 0.0002,
+      "loss": 1.4713,
+      "step": 4690
+    },
+    {
+      "epoch": 3.1490787269681744,
+      "grad_norm": 0.645889401435852,
+      "learning_rate": 0.0002,
+      "loss": 1.5114,
+      "step": 4700
+    },
+    {
+      "epoch": 3.1557788944723617,
+      "grad_norm": 0.6118691563606262,
+      "learning_rate": 0.0002,
+      "loss": 1.4705,
+      "step": 4710
+    },
+    {
+      "epoch": 3.1624790619765495,
+      "grad_norm": 0.5189669132232666,
+      "learning_rate": 0.0002,
+      "loss": 1.5533,
+      "step": 4720
+    },
+    {
+      "epoch": 3.169179229480737,
+      "grad_norm": 0.5794713497161865,
+      "learning_rate": 0.0002,
+      "loss": 1.4769,
+      "step": 4730
+    },
+    {
+      "epoch": 3.1758793969849246,
+      "grad_norm": 0.6579326391220093,
+      "learning_rate": 0.0002,
+      "loss": 1.4849,
+      "step": 4740
+    },
+    {
+      "epoch": 3.1825795644891124,
+      "grad_norm": 0.5822742581367493,
+      "learning_rate": 0.0002,
+      "loss": 1.545,
+      "step": 4750
+    },
+    {
+      "epoch": 3.1892797319932997,
+      "grad_norm": 0.5475956201553345,
+      "learning_rate": 0.0002,
+      "loss": 1.4358,
+      "step": 4760
+    },
+    {
+      "epoch": 3.1959798994974875,
+      "grad_norm": 0.6743834018707275,
+      "learning_rate": 0.0002,
+      "loss": 1.4723,
+      "step": 4770
+    },
+    {
+      "epoch": 3.202680067001675,
+      "grad_norm": 0.6110585927963257,
+      "learning_rate": 0.0002,
+      "loss": 1.5161,
+      "step": 4780
+    },
+    {
+      "epoch": 3.2093802345058626,
+      "grad_norm": 0.5426181554794312,
+      "learning_rate": 0.0002,
+      "loss": 1.5455,
+      "step": 4790
+    },
+    {
+      "epoch": 3.2160804020100504,
+      "grad_norm": 0.6077824234962463,
+      "learning_rate": 0.0002,
+      "loss": 1.5315,
+      "step": 4800
+    },
+    {
+      "epoch": 3.2227805695142377,
+      "grad_norm": 0.5785858631134033,
+      "learning_rate": 0.0002,
+      "loss": 1.5314,
+      "step": 4810
+    },
+    {
+      "epoch": 3.2294807370184255,
+      "grad_norm": 0.6425958275794983,
+      "learning_rate": 0.0002,
+      "loss": 1.4041,
+      "step": 4820
+    },
+    {
+      "epoch": 3.236180904522613,
+      "grad_norm": 0.6607080698013306,
+      "learning_rate": 0.0002,
+      "loss": 1.4751,
+      "step": 4830
+    },
+    {
+      "epoch": 3.2428810720268006,
+      "grad_norm": 0.5385788679122925,
+      "learning_rate": 0.0002,
+      "loss": 1.5267,
+      "step": 4840
+    },
+    {
+      "epoch": 3.2495812395309884,
+      "grad_norm": 0.5630403757095337,
+      "learning_rate": 0.0002,
+      "loss": 1.4673,
+      "step": 4850
+    },
+    {
+      "epoch": 3.2562814070351758,
+      "grad_norm": 0.6340779662132263,
+      "learning_rate": 0.0002,
+      "loss": 1.5257,
+      "step": 4860
+    },
+    {
+      "epoch": 3.2629815745393635,
+      "grad_norm": 0.5305342674255371,
+      "learning_rate": 0.0002,
+      "loss": 1.5148,
+      "step": 4870
+    },
+    {
+      "epoch": 3.2696817420435513,
+      "grad_norm": 0.597670316696167,
+      "learning_rate": 0.0002,
+      "loss": 1.5162,
+      "step": 4880
+    },
+    {
+      "epoch": 3.2763819095477387,
+      "grad_norm": 0.665553867816925,
+      "learning_rate": 0.0002,
+      "loss": 1.5429,
+      "step": 4890
+    },
+    {
+      "epoch": 3.2830820770519265,
+      "grad_norm": 0.579767644405365,
+      "learning_rate": 0.0002,
+      "loss": 1.4607,
+      "step": 4900
+    },
+    {
+      "epoch": 3.289782244556114,
+      "grad_norm": 0.5512481331825256,
+      "learning_rate": 0.0002,
+      "loss": 1.4999,
+      "step": 4910
+    },
+    {
+      "epoch": 3.2964824120603016,
+      "grad_norm": 0.5916532278060913,
+      "learning_rate": 0.0002,
+      "loss": 1.5022,
+      "step": 4920
+    },
+    {
+      "epoch": 3.3031825795644894,
+      "grad_norm": 0.7521726489067078,
+      "learning_rate": 0.0002,
+      "loss": 1.4889,
+      "step": 4930
+    },
+    {
+      "epoch": 3.3098827470686767,
+      "grad_norm": 0.5352797508239746,
+      "learning_rate": 0.0002,
+      "loss": 1.4223,
+      "step": 4940
+    },
+    {
+      "epoch": 3.3165829145728645,
+      "grad_norm": 0.5950371623039246,
+      "learning_rate": 0.0002,
+      "loss": 1.5122,
+      "step": 4950
+    },
+    {
+      "epoch": 3.323283082077052,
+      "grad_norm": 0.8020477890968323,
+      "learning_rate": 0.0002,
+      "loss": 1.5072,
+      "step": 4960
+    },
+    {
+      "epoch": 3.3299832495812396,
+      "grad_norm": 0.6790024638175964,
+      "learning_rate": 0.0002,
+      "loss": 1.5422,
+      "step": 4970
+    },
+    {
+      "epoch": 3.3366834170854274,
+      "grad_norm": 0.687627375125885,
+      "learning_rate": 0.0002,
+      "loss": 1.5363,
+      "step": 4980
+    },
+    {
+      "epoch": 3.3433835845896147,
+      "grad_norm": 0.6094385385513306,
+      "learning_rate": 0.0002,
+      "loss": 1.5276,
+      "step": 4990
+    },
+    {
+      "epoch": 3.3500837520938025,
+      "grad_norm": 0.6541242003440857,
+      "learning_rate": 0.0002,
+      "loss": 1.549,
+      "step": 5000
+    },
+    {
+      "epoch": 3.35678391959799,
+      "grad_norm": 0.5560880303382874,
+      "learning_rate": 0.0002,
+      "loss": 1.6067,
+      "step": 5010
+    },
+    {
+      "epoch": 3.3634840871021776,
+      "grad_norm": 0.5440094470977783,
+      "learning_rate": 0.0002,
+      "loss": 1.5769,
+      "step": 5020
+    },
+    {
+      "epoch": 3.3701842546063654,
+      "grad_norm": 0.5749301314353943,
+      "learning_rate": 0.0002,
+      "loss": 1.6183,
+      "step": 5030
+    },
+    {
+      "epoch": 3.3768844221105527,
+      "grad_norm": 0.5919716954231262,
+      "learning_rate": 0.0002,
+      "loss": 1.4801,
+      "step": 5040
+    },
+    {
+      "epoch": 3.3835845896147405,
+      "grad_norm": 0.6331481337547302,
+      "learning_rate": 0.0002,
+      "loss": 1.5957,
+      "step": 5050
+    },
+    {
+      "epoch": 3.390284757118928,
+      "grad_norm": 0.5687161684036255,
+      "learning_rate": 0.0002,
+      "loss": 1.5188,
+      "step": 5060
+    },
+    {
+      "epoch": 3.3969849246231156,
+      "grad_norm": 0.6718577742576599,
+      "learning_rate": 0.0002,
+      "loss": 1.5702,
+      "step": 5070
+    },
+    {
+      "epoch": 3.4036850921273034,
+      "grad_norm": 0.5089324116706848,
+      "learning_rate": 0.0002,
+      "loss": 1.5577,
+      "step": 5080
+    },
+    {
+      "epoch": 3.4103852596314908,
+      "grad_norm": 0.5710174441337585,
+      "learning_rate": 0.0002,
+      "loss": 1.512,
+      "step": 5090
+    },
+    {
+      "epoch": 3.4170854271356785,
+      "grad_norm": 0.6670721173286438,
+      "learning_rate": 0.0002,
+      "loss": 1.5492,
+      "step": 5100
+    },
+    {
+      "epoch": 3.423785594639866,
+      "grad_norm": 0.6875665187835693,
+      "learning_rate": 0.0002,
+      "loss": 1.5227,
+      "step": 5110
+    },
+    {
+      "epoch": 3.4304857621440537,
+      "grad_norm": 0.5375880599021912,
+      "learning_rate": 0.0002,
+      "loss": 1.4496,
+      "step": 5120
+    },
+    {
+      "epoch": 3.4371859296482414,
+      "grad_norm": 0.6550399661064148,
+      "learning_rate": 0.0002,
+      "loss": 1.5527,
+      "step": 5130
+    },
+    {
+      "epoch": 3.4438860971524288,
+      "grad_norm": 0.5948067903518677,
+      "learning_rate": 0.0002,
+      "loss": 1.5687,
+      "step": 5140
+    },
+    {
+      "epoch": 3.4505862646566166,
+      "grad_norm": 0.6134477257728577,
+      "learning_rate": 0.0002,
+      "loss": 1.4813,
+      "step": 5150
+    },
+    {
+      "epoch": 3.457286432160804,
+      "grad_norm": 0.6506398320198059,
+      "learning_rate": 0.0002,
+      "loss": 1.5069,
+      "step": 5160
+    },
+    {
+      "epoch": 3.4639865996649917,
+      "grad_norm": 0.6060147881507874,
+      "learning_rate": 0.0002,
+      "loss": 1.4422,
+      "step": 5170
+    },
+    {
+      "epoch": 3.4706867671691795,
+      "grad_norm": 0.6173806190490723,
+      "learning_rate": 0.0002,
+      "loss": 1.5093,
+      "step": 5180
+    },
+    {
+      "epoch": 3.477386934673367,
+      "grad_norm": 0.6032607555389404,
+      "learning_rate": 0.0002,
+      "loss": 1.4975,
+      "step": 5190
+    },
+    {
+      "epoch": 3.4840871021775546,
+      "grad_norm": 0.5652492046356201,
+      "learning_rate": 0.0002,
+      "loss": 1.4979,
+      "step": 5200
+    },
+    {
+      "epoch": 3.490787269681742,
+      "grad_norm": 0.6168607473373413,
+      "learning_rate": 0.0002,
+      "loss": 1.4883,
+      "step": 5210
+    },
+    {
+      "epoch": 3.4974874371859297,
+      "grad_norm": 0.6170629262924194,
+      "learning_rate": 0.0002,
+      "loss": 1.5164,
+      "step": 5220
+    },
+    {
+      "epoch": 3.5041876046901175,
+      "grad_norm": 0.6926297545433044,
+      "learning_rate": 0.0002,
+      "loss": 1.4879,
+      "step": 5230
+    },
+    {
+      "epoch": 3.510887772194305,
+      "grad_norm": 0.6702437996864319,
+      "learning_rate": 0.0002,
+      "loss": 1.4982,
+      "step": 5240
+    },
+    {
+      "epoch": 3.5175879396984926,
+      "grad_norm": 0.5421436429023743,
+      "learning_rate": 0.0002,
+      "loss": 1.4986,
+      "step": 5250
+    },
+    {
+      "epoch": 3.52428810720268,
+      "grad_norm": 0.5726765990257263,
+      "learning_rate": 0.0002,
+      "loss": 1.4673,
+      "step": 5260
+    },
+    {
+      "epoch": 3.5309882747068677,
+      "grad_norm": 0.5685455203056335,
+      "learning_rate": 0.0002,
+      "loss": 1.5423,
+      "step": 5270
+    },
+    {
+      "epoch": 3.5376884422110555,
+      "grad_norm": 0.6018396019935608,
+      "learning_rate": 0.0002,
+      "loss": 1.4715,
+      "step": 5280
+    },
+    {
+      "epoch": 3.544388609715243,
+      "grad_norm": 0.5731932520866394,
+      "learning_rate": 0.0002,
+      "loss": 1.5451,
+      "step": 5290
+    },
+    {
+      "epoch": 3.5510887772194306,
+      "grad_norm": 0.6601519584655762,
+      "learning_rate": 0.0002,
+      "loss": 1.4752,
+      "step": 5300
+    },
+    {
+      "epoch": 3.557788944723618,
+      "grad_norm": 0.5545530319213867,
+      "learning_rate": 0.0002,
+      "loss": 1.5434,
+      "step": 5310
+    },
+    {
+      "epoch": 3.5644891122278057,
+      "grad_norm": 0.5998541116714478,
+      "learning_rate": 0.0002,
+      "loss": 1.5438,
+      "step": 5320
+    },
+    {
+      "epoch": 3.5711892797319935,
+      "grad_norm": 0.5651767253875732,
+      "learning_rate": 0.0002,
+      "loss": 1.56,
+      "step": 5330
+    },
+    {
+      "epoch": 3.577889447236181,
+      "grad_norm": 0.7425084114074707,
+      "learning_rate": 0.0002,
+      "loss": 1.4829,
+      "step": 5340
+    },
+    {
+      "epoch": 3.5845896147403686,
+      "grad_norm": 0.5770602226257324,
+      "learning_rate": 0.0002,
+      "loss": 1.5571,
+      "step": 5350
+    },
+    {
+      "epoch": 3.591289782244556,
+      "grad_norm": 0.54723060131073,
+      "learning_rate": 0.0002,
+      "loss": 1.458,
+      "step": 5360
+    },
+    {
+      "epoch": 3.5979899497487438,
+      "grad_norm": 0.6658238172531128,
+      "learning_rate": 0.0002,
+      "loss": 1.497,
+      "step": 5370
+    },
+    {
+      "epoch": 3.6046901172529315,
+      "grad_norm": 0.5787645578384399,
+      "learning_rate": 0.0002,
+      "loss": 1.5456,
+      "step": 5380
+    },
+    {
+      "epoch": 3.611390284757119,
+      "grad_norm": 0.594913125038147,
+      "learning_rate": 0.0002,
+      "loss": 1.5343,
+      "step": 5390
+    },
+    {
+      "epoch": 3.6180904522613067,
+      "grad_norm": 0.4964977502822876,
+      "learning_rate": 0.0002,
+      "loss": 1.4727,
+      "step": 5400
+    },
+    {
+      "epoch": 3.624790619765494,
+      "grad_norm": 0.6087527275085449,
+      "learning_rate": 0.0002,
+      "loss": 1.5062,
+      "step": 5410
+    },
+    {
+      "epoch": 3.6314907872696818,
+      "grad_norm": 0.6315323710441589,
+      "learning_rate": 0.0002,
+      "loss": 1.5098,
+      "step": 5420
+    },
+    {
+      "epoch": 3.6381909547738696,
+      "grad_norm": 0.574799120426178,
+      "learning_rate": 0.0002,
+      "loss": 1.4855,
+      "step": 5430
+    },
+    {
+      "epoch": 3.644891122278057,
+      "grad_norm": 0.5949277877807617,
+      "learning_rate": 0.0002,
+      "loss": 1.4595,
+      "step": 5440
+    },
+    {
+      "epoch": 3.6515912897822447,
+      "grad_norm": 0.5640677213668823,
+      "learning_rate": 0.0002,
+      "loss": 1.4816,
+      "step": 5450
+    },
+    {
+      "epoch": 3.658291457286432,
+      "grad_norm": 0.6198237538337708,
+      "learning_rate": 0.0002,
+      "loss": 1.525,
+      "step": 5460
+    },
+    {
+      "epoch": 3.66499162479062,
+      "grad_norm": 0.6902034878730774,
+      "learning_rate": 0.0002,
+      "loss": 1.5676,
+      "step": 5470
+    },
+    {
+      "epoch": 3.6716917922948076,
+      "grad_norm": 0.5686674118041992,
+      "learning_rate": 0.0002,
+      "loss": 1.544,
+      "step": 5480
+    },
+    {
+      "epoch": 3.678391959798995,
+      "grad_norm": 0.6532107591629028,
+      "learning_rate": 0.0002,
+      "loss": 1.5255,
+      "step": 5490
+    },
+    {
+      "epoch": 3.6850921273031827,
+      "grad_norm": 0.5790849924087524,
+      "learning_rate": 0.0002,
+      "loss": 1.5767,
+      "step": 5500
+    },
+    {
+      "epoch": 3.69179229480737,
+      "grad_norm": 0.6055065393447876,
+      "learning_rate": 0.0002,
+      "loss": 1.5507,
+      "step": 5510
+    },
+    {
+      "epoch": 3.698492462311558,
+      "grad_norm": 0.5630605816841125,
+      "learning_rate": 0.0002,
+      "loss": 1.4656,
+      "step": 5520
+    },
+    {
+      "epoch": 3.7051926298157456,
+      "grad_norm": 0.6005825996398926,
+      "learning_rate": 0.0002,
+      "loss": 1.537,
+      "step": 5530
+    },
+    {
+      "epoch": 3.711892797319933,
+      "grad_norm": 0.6553038954734802,
+      "learning_rate": 0.0002,
+      "loss": 1.5313,
+      "step": 5540
+    },
+    {
+      "epoch": 3.7185929648241207,
+      "grad_norm": 0.5601094961166382,
+      "learning_rate": 0.0002,
+      "loss": 1.4943,
+      "step": 5550
+    },
+    {
+      "epoch": 3.725293132328308,
+      "grad_norm": 0.6598808169364929,
+      "learning_rate": 0.0002,
+      "loss": 1.5161,
+      "step": 5560
+    },
+    {
+      "epoch": 3.731993299832496,
+      "grad_norm": 0.5506255626678467,
+      "learning_rate": 0.0002,
+      "loss": 1.5345,
+      "step": 5570
+    },
+    {
+      "epoch": 3.7386934673366836,
+      "grad_norm": 0.6001223921775818,
+      "learning_rate": 0.0002,
+      "loss": 1.4805,
+      "step": 5580
+    },
+    {
+      "epoch": 3.745393634840871,
+      "grad_norm": 0.6287297606468201,
+      "learning_rate": 0.0002,
+      "loss": 1.4652,
+      "step": 5590
+    },
+    {
+      "epoch": 3.7520938023450587,
+      "grad_norm": 0.6253238916397095,
+      "learning_rate": 0.0002,
+      "loss": 1.5246,
+      "step": 5600
+    },
+    {
+      "epoch": 3.758793969849246,
+      "grad_norm": 0.5713174939155579,
+      "learning_rate": 0.0002,
+      "loss": 1.5691,
+      "step": 5610
+    },
+    {
+      "epoch": 3.765494137353434,
+      "grad_norm": 0.6198310852050781,
+      "learning_rate": 0.0002,
+      "loss": 1.5661,
+      "step": 5620
+    },
+    {
+      "epoch": 3.7721943048576216,
+      "grad_norm": 0.5941224098205566,
+      "learning_rate": 0.0002,
+      "loss": 1.5448,
+      "step": 5630
+    },
+    {
+      "epoch": 3.778894472361809,
+      "grad_norm": 0.606002151966095,
+      "learning_rate": 0.0002,
+      "loss": 1.4925,
+      "step": 5640
+    },
+    {
+      "epoch": 3.7855946398659968,
+      "grad_norm": 0.6540704965591431,
+      "learning_rate": 0.0002,
+      "loss": 1.5182,
+      "step": 5650
+    },
+    {
+      "epoch": 3.792294807370184,
+      "grad_norm": 0.6147415041923523,
+      "learning_rate": 0.0002,
+      "loss": 1.5903,
+      "step": 5660
+    },
+    {
+      "epoch": 3.798994974874372,
+      "grad_norm": 0.5649605393409729,
+      "learning_rate": 0.0002,
+      "loss": 1.5329,
+      "step": 5670
+    },
+    {
+      "epoch": 3.8056951423785597,
+      "grad_norm": 0.6788773536682129,
+      "learning_rate": 0.0002,
+      "loss": 1.5747,
+      "step": 5680
+    },
+    {
+      "epoch": 3.812395309882747,
+      "grad_norm": 0.6581860780715942,
+      "learning_rate": 0.0002,
+      "loss": 1.535,
+      "step": 5690
+    },
+    {
+      "epoch": 3.819095477386935,
+      "grad_norm": 0.5529348850250244,
+      "learning_rate": 0.0002,
+      "loss": 1.4587,
+      "step": 5700
+    },
+    {
+      "epoch": 3.825795644891122,
+      "grad_norm": 0.6320232152938843,
+      "learning_rate": 0.0002,
+      "loss": 1.5594,
+      "step": 5710
+    },
+    {
+      "epoch": 3.83249581239531,
+      "grad_norm": 0.6529698371887207,
+      "learning_rate": 0.0002,
+      "loss": 1.4696,
+      "step": 5720
+    },
+    {
+      "epoch": 3.8391959798994977,
+      "grad_norm": 0.5983362793922424,
+      "learning_rate": 0.0002,
+      "loss": 1.5854,
+      "step": 5730
+    },
+    {
+      "epoch": 3.845896147403685,
+      "grad_norm": 0.6335684061050415,
+      "learning_rate": 0.0002,
+      "loss": 1.465,
+      "step": 5740
+    },
+    {
+      "epoch": 3.852596314907873,
+      "grad_norm": 0.700446605682373,
+      "learning_rate": 0.0002,
+      "loss": 1.5545,
+      "step": 5750
+    },
+    {
+      "epoch": 3.85929648241206,
+      "grad_norm": 0.6092597842216492,
+      "learning_rate": 0.0002,
+      "loss": 1.5707,
+      "step": 5760
+    },
+    {
+      "epoch": 3.865996649916248,
+      "grad_norm": 0.564146101474762,
+      "learning_rate": 0.0002,
+      "loss": 1.5729,
+      "step": 5770
+    },
+    {
+      "epoch": 3.8726968174204357,
+      "grad_norm": 0.615275502204895,
+      "learning_rate": 0.0002,
+      "loss": 1.5872,
+      "step": 5780
+    },
+    {
+      "epoch": 3.879396984924623,
+      "grad_norm": 0.6685376763343811,
+      "learning_rate": 0.0002,
+      "loss": 1.5142,
+      "step": 5790
+    },
+    {
+      "epoch": 3.886097152428811,
+      "grad_norm": 0.6116922497749329,
+      "learning_rate": 0.0002,
+      "loss": 1.4752,
+      "step": 5800
+    },
+    {
+      "epoch": 3.892797319932998,
+      "grad_norm": 0.5486813187599182,
+      "learning_rate": 0.0002,
+      "loss": 1.5179,
+      "step": 5810
+    },
+    {
+      "epoch": 3.899497487437186,
+      "grad_norm": 0.6208204030990601,
+      "learning_rate": 0.0002,
+      "loss": 1.5167,
+      "step": 5820
+    },
+    {
+      "epoch": 3.9061976549413737,
+      "grad_norm": 0.6500625014305115,
+      "learning_rate": 0.0002,
+      "loss": 1.5334,
+      "step": 5830
+    },
+    {
+      "epoch": 3.912897822445561,
+      "grad_norm": 0.5948089361190796,
+      "learning_rate": 0.0002,
+      "loss": 1.4716,
+      "step": 5840
+    },
+    {
+      "epoch": 3.919597989949749,
+      "grad_norm": 0.7210732698440552,
+      "learning_rate": 0.0002,
+      "loss": 1.6011,
+      "step": 5850
+    },
+    {
+      "epoch": 3.926298157453936,
+      "grad_norm": 0.6662322878837585,
+      "learning_rate": 0.0002,
+      "loss": 1.5519,
+      "step": 5860
+    },
+    {
+      "epoch": 3.932998324958124,
+      "grad_norm": 0.5613839626312256,
+      "learning_rate": 0.0002,
+      "loss": 1.5656,
+      "step": 5870
+    },
+    {
+      "epoch": 3.9396984924623117,
+      "grad_norm": 0.6069002151489258,
+      "learning_rate": 0.0002,
+      "loss": 1.544,
+      "step": 5880
+    },
+    {
+      "epoch": 3.946398659966499,
+      "grad_norm": 0.7075562477111816,
+      "learning_rate": 0.0002,
+      "loss": 1.6745,
+      "step": 5890
+    },
+    {
+      "epoch": 3.953098827470687,
+      "grad_norm": 0.6316173076629639,
+      "learning_rate": 0.0002,
+      "loss": 1.5391,
+      "step": 5900
+    },
+    {
+      "epoch": 3.959798994974874,
+      "grad_norm": 0.5716308355331421,
+      "learning_rate": 0.0002,
+      "loss": 1.6314,
+      "step": 5910
+    },
+    {
+      "epoch": 3.966499162479062,
+      "grad_norm": 0.6800096035003662,
+      "learning_rate": 0.0002,
+      "loss": 1.5947,
+      "step": 5920
+    },
+    {
+      "epoch": 3.9731993299832498,
+      "grad_norm": 0.6057983040809631,
+      "learning_rate": 0.0002,
+      "loss": 1.5189,
+      "step": 5930
+    },
+    {
+      "epoch": 3.979899497487437,
+      "grad_norm": 0.5938987731933594,
+      "learning_rate": 0.0002,
+      "loss": 1.5431,
+      "step": 5940
+    },
+    {
+      "epoch": 3.986599664991625,
+      "grad_norm": 0.6963576674461365,
+      "learning_rate": 0.0002,
+      "loss": 1.5111,
+      "step": 5950
+    },
+    {
+      "epoch": 3.993299832495812,
+      "grad_norm": 0.6279940009117126,
+      "learning_rate": 0.0002,
+      "loss": 1.5521,
+      "step": 5960
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 0.7161159515380859,
+      "learning_rate": 0.0002,
+      "loss": 1.5974,
+      "step": 5970
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 1.8655421733856201,
+      "eval_runtime": 37.9276,
+      "eval_samples_per_second": 13.579,
+      "eval_steps_per_second": 1.714,
+      "step": 5970
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 11936,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 8,
+  "save_steps": 200,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 2.7627823262859264e+17,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-5970/training_args.bin b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-5970/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..064f299b0f3d2a28f5b1f5c68ef32caab3e2dd49
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-5970/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aa7b586fc440d1e22156593e2dd4267d2bdcb8920a02fdf352ea29a9bec3dd94
+size 5560
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-7462/README.md b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-7462/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..835e31ab8469ee39ddc8b2b6b2143a8c66dad510
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-7462/README.md
@@ -0,0 +1,202 @@
+---
+base_model: meta-llama/Meta-Llama-3-8B-Instruct
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.13.1
\ No newline at end of file
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-7462/adapter_config.json b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-7462/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..406c5a08dc4a2a33b52c62a482f98c217c417215
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-7462/adapter_config.json
@@ -0,0 +1,29 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "meta-llama/Meta-Llama-3-8B-Instruct",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-7462/adapter_model.safetensors b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-7462/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..2091ccdd675cfe239b2d7d6ac2a9c115915cf019
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-7462/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4a6dc8f1eb63b254d61cc48121629221832e83035d7cbbea32f7a1c7e5613918
+size 109069176
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-7462/optimizer.pt b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-7462/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..7970f235a5307df1b83c9903f9352cdde671a838
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-7462/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6af6af9b3e1d5fc139cd7f77d43795b275d0cffe9c1e94b132559552babf620b
+size 55532666
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-7462/rng_state.pth b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-7462/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..a68ba21ac26150c6f30a31fadacd57f2eb69cb77
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-7462/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:111d23dcfd260fbf001922a777fa691a7c513e3e31685b60c2713a289382a171
+size 14244
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-7462/scheduler.pt b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-7462/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..17f10eb44ce7821cb7f52d06750abb17355b54c3
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-7462/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:27652f182b842c7ed5e0b2abe3cf6cd30fac1baf0f5622b65fbc7d55d4a9a6c6
+size 1064
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-7462/special_tokens_map.json b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-7462/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..b43be96621d147110fb8a18b5776ec6e38516127
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-7462/special_tokens_map.json
@@ -0,0 +1,17 @@
+{
+  "bos_token": {
+    "content": "<|begin_of_text|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<|eot_id|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<|eot_id|>"
+}
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-7462/tokenizer.json b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-7462/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..86a33946b0c77216d2cce91bb28c8fada4a5e80b
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-7462/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3c5cf44023714fb39b05e71e425f8d7b92805ff73f7988b083b8c87f0bf87393
+size 17209961
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-7462/tokenizer_config.json b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-7462/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..061e40d9db3253624f86e8e364c15ef546527c9d
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-7462/tokenizer_config.json
@@ -0,0 +1,2063 @@
+{
+  "added_tokens_decoder": {
+    "128000": {
+      "content": "<|begin_of_text|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128001": {
+      "content": "<|end_of_text|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128002": {
+      "content": "<|reserved_special_token_0|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128003": {
+      "content": "<|reserved_special_token_1|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128004": {
+      "content": "<|reserved_special_token_2|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128005": {
+      "content": "<|reserved_special_token_3|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128006": {
+      "content": "<|start_header_id|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128007": {
+      "content": "<|end_header_id|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128008": {
+      "content": "<|reserved_special_token_4|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128009": {
+      "content": "<|eot_id|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128010": {
+      "content": "<|reserved_special_token_5|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128011": {
+      "content": "<|reserved_special_token_6|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128012": {
+      "content": "<|reserved_special_token_7|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128013": {
+      "content": "<|reserved_special_token_8|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128014": {
+      "content": "<|reserved_special_token_9|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128015": {
+      "content": "<|reserved_special_token_10|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128016": {
+      "content": "<|reserved_special_token_11|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128017": {
+      "content": "<|reserved_special_token_12|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128018": {
+      "content": "<|reserved_special_token_13|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128019": {
+      "content": "<|reserved_special_token_14|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128020": {
+      "content": "<|reserved_special_token_15|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128021": {
+      "content": "<|reserved_special_token_16|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128022": {
+      "content": "<|reserved_special_token_17|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128023": {
+      "content": "<|reserved_special_token_18|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128024": {
+      "content": "<|reserved_special_token_19|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128025": {
+      "content": "<|reserved_special_token_20|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128026": {
+      "content": "<|reserved_special_token_21|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128027": {
+      "content": "<|reserved_special_token_22|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128028": {
+      "content": "<|reserved_special_token_23|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128029": {
+      "content": "<|reserved_special_token_24|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128030": {
+      "content": "<|reserved_special_token_25|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128031": {
+      "content": "<|reserved_special_token_26|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128032": {
+      "content": "<|reserved_special_token_27|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128033": {
+      "content": "<|reserved_special_token_28|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128034": {
+      "content": "<|reserved_special_token_29|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128035": {
+      "content": "<|reserved_special_token_30|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128036": {
+      "content": "<|reserved_special_token_31|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128037": {
+      "content": "<|reserved_special_token_32|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128038": {
+      "content": "<|reserved_special_token_33|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128039": {
+      "content": "<|reserved_special_token_34|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128040": {
+      "content": "<|reserved_special_token_35|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128041": {
+      "content": "<|reserved_special_token_36|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128042": {
+      "content": "<|reserved_special_token_37|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128043": {
+      "content": "<|reserved_special_token_38|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128044": {
+      "content": "<|reserved_special_token_39|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128045": {
+      "content": "<|reserved_special_token_40|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128046": {
+      "content": "<|reserved_special_token_41|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128047": {
+      "content": "<|reserved_special_token_42|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128048": {
+      "content": "<|reserved_special_token_43|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128049": {
+      "content": "<|reserved_special_token_44|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128050": {
+      "content": "<|reserved_special_token_45|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128051": {
+      "content": "<|reserved_special_token_46|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128052": {
+      "content": "<|reserved_special_token_47|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128053": {
+      "content": "<|reserved_special_token_48|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128054": {
+      "content": "<|reserved_special_token_49|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128055": {
+      "content": "<|reserved_special_token_50|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128056": {
+      "content": "<|reserved_special_token_51|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128057": {
+      "content": "<|reserved_special_token_52|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128058": {
+      "content": "<|reserved_special_token_53|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128059": {
+      "content": "<|reserved_special_token_54|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128060": {
+      "content": "<|reserved_special_token_55|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128061": {
+      "content": "<|reserved_special_token_56|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128062": {
+      "content": "<|reserved_special_token_57|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128063": {
+      "content": "<|reserved_special_token_58|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128064": {
+      "content": "<|reserved_special_token_59|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128065": {
+      "content": "<|reserved_special_token_60|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128066": {
+      "content": "<|reserved_special_token_61|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128067": {
+      "content": "<|reserved_special_token_62|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128068": {
+      "content": "<|reserved_special_token_63|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128069": {
+      "content": "<|reserved_special_token_64|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128070": {
+      "content": "<|reserved_special_token_65|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128071": {
+      "content": "<|reserved_special_token_66|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128072": {
+      "content": "<|reserved_special_token_67|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128073": {
+      "content": "<|reserved_special_token_68|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128074": {
+      "content": "<|reserved_special_token_69|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128075": {
+      "content": "<|reserved_special_token_70|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128076": {
+      "content": "<|reserved_special_token_71|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128077": {
+      "content": "<|reserved_special_token_72|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128078": {
+      "content": "<|reserved_special_token_73|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128079": {
+      "content": "<|reserved_special_token_74|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128080": {
+      "content": "<|reserved_special_token_75|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128081": {
+      "content": "<|reserved_special_token_76|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128082": {
+      "content": "<|reserved_special_token_77|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128083": {
+      "content": "<|reserved_special_token_78|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128084": {
+      "content": "<|reserved_special_token_79|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128085": {
+      "content": "<|reserved_special_token_80|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128086": {
+      "content": "<|reserved_special_token_81|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128087": {
+      "content": "<|reserved_special_token_82|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128088": {
+      "content": "<|reserved_special_token_83|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128089": {
+      "content": "<|reserved_special_token_84|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128090": {
+      "content": "<|reserved_special_token_85|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128091": {
+      "content": "<|reserved_special_token_86|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128092": {
+      "content": "<|reserved_special_token_87|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128093": {
+      "content": "<|reserved_special_token_88|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128094": {
+      "content": "<|reserved_special_token_89|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128095": {
+      "content": "<|reserved_special_token_90|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128096": {
+      "content": "<|reserved_special_token_91|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128097": {
+      "content": "<|reserved_special_token_92|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128098": {
+      "content": "<|reserved_special_token_93|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128099": {
+      "content": "<|reserved_special_token_94|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128100": {
+      "content": "<|reserved_special_token_95|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128101": {
+      "content": "<|reserved_special_token_96|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128102": {
+      "content": "<|reserved_special_token_97|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128103": {
+      "content": "<|reserved_special_token_98|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128104": {
+      "content": "<|reserved_special_token_99|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128105": {
+      "content": "<|reserved_special_token_100|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128106": {
+      "content": "<|reserved_special_token_101|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128107": {
+      "content": "<|reserved_special_token_102|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128108": {
+      "content": "<|reserved_special_token_103|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128109": {
+      "content": "<|reserved_special_token_104|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128110": {
+      "content": "<|reserved_special_token_105|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128111": {
+      "content": "<|reserved_special_token_106|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128112": {
+      "content": "<|reserved_special_token_107|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128113": {
+      "content": "<|reserved_special_token_108|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128114": {
+      "content": "<|reserved_special_token_109|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128115": {
+      "content": "<|reserved_special_token_110|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128116": {
+      "content": "<|reserved_special_token_111|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128117": {
+      "content": "<|reserved_special_token_112|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128118": {
+      "content": "<|reserved_special_token_113|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128119": {
+      "content": "<|reserved_special_token_114|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128120": {
+      "content": "<|reserved_special_token_115|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128121": {
+      "content": "<|reserved_special_token_116|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128122": {
+      "content": "<|reserved_special_token_117|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128123": {
+      "content": "<|reserved_special_token_118|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128124": {
+      "content": "<|reserved_special_token_119|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128125": {
+      "content": "<|reserved_special_token_120|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128126": {
+      "content": "<|reserved_special_token_121|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128127": {
+      "content": "<|reserved_special_token_122|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128128": {
+      "content": "<|reserved_special_token_123|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128129": {
+      "content": "<|reserved_special_token_124|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128130": {
+      "content": "<|reserved_special_token_125|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128131": {
+      "content": "<|reserved_special_token_126|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128132": {
+      "content": "<|reserved_special_token_127|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128133": {
+      "content": "<|reserved_special_token_128|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128134": {
+      "content": "<|reserved_special_token_129|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128135": {
+      "content": "<|reserved_special_token_130|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128136": {
+      "content": "<|reserved_special_token_131|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128137": {
+      "content": "<|reserved_special_token_132|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128138": {
+      "content": "<|reserved_special_token_133|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128139": {
+      "content": "<|reserved_special_token_134|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128140": {
+      "content": "<|reserved_special_token_135|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128141": {
+      "content": "<|reserved_special_token_136|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128142": {
+      "content": "<|reserved_special_token_137|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128143": {
+      "content": "<|reserved_special_token_138|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128144": {
+      "content": "<|reserved_special_token_139|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128145": {
+      "content": "<|reserved_special_token_140|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128146": {
+      "content": "<|reserved_special_token_141|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128147": {
+      "content": "<|reserved_special_token_142|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128148": {
+      "content": "<|reserved_special_token_143|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128149": {
+      "content": "<|reserved_special_token_144|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128150": {
+      "content": "<|reserved_special_token_145|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128151": {
+      "content": "<|reserved_special_token_146|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128152": {
+      "content": "<|reserved_special_token_147|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128153": {
+      "content": "<|reserved_special_token_148|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128154": {
+      "content": "<|reserved_special_token_149|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128155": {
+      "content": "<|reserved_special_token_150|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128156": {
+      "content": "<|reserved_special_token_151|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128157": {
+      "content": "<|reserved_special_token_152|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128158": {
+      "content": "<|reserved_special_token_153|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128159": {
+      "content": "<|reserved_special_token_154|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128160": {
+      "content": "<|reserved_special_token_155|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128161": {
+      "content": "<|reserved_special_token_156|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128162": {
+      "content": "<|reserved_special_token_157|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128163": {
+      "content": "<|reserved_special_token_158|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128164": {
+      "content": "<|reserved_special_token_159|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128165": {
+      "content": "<|reserved_special_token_160|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128166": {
+      "content": "<|reserved_special_token_161|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128167": {
+      "content": "<|reserved_special_token_162|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128168": {
+      "content": "<|reserved_special_token_163|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128169": {
+      "content": "<|reserved_special_token_164|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128170": {
+      "content": "<|reserved_special_token_165|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128171": {
+      "content": "<|reserved_special_token_166|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128172": {
+      "content": "<|reserved_special_token_167|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128173": {
+      "content": "<|reserved_special_token_168|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128174": {
+      "content": "<|reserved_special_token_169|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128175": {
+      "content": "<|reserved_special_token_170|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128176": {
+      "content": "<|reserved_special_token_171|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128177": {
+      "content": "<|reserved_special_token_172|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128178": {
+      "content": "<|reserved_special_token_173|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128179": {
+      "content": "<|reserved_special_token_174|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128180": {
+      "content": "<|reserved_special_token_175|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128181": {
+      "content": "<|reserved_special_token_176|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128182": {
+      "content": "<|reserved_special_token_177|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128183": {
+      "content": "<|reserved_special_token_178|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128184": {
+      "content": "<|reserved_special_token_179|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128185": {
+      "content": "<|reserved_special_token_180|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128186": {
+      "content": "<|reserved_special_token_181|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128187": {
+      "content": "<|reserved_special_token_182|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128188": {
+      "content": "<|reserved_special_token_183|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128189": {
+      "content": "<|reserved_special_token_184|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128190": {
+      "content": "<|reserved_special_token_185|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128191": {
+      "content": "<|reserved_special_token_186|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128192": {
+      "content": "<|reserved_special_token_187|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128193": {
+      "content": "<|reserved_special_token_188|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128194": {
+      "content": "<|reserved_special_token_189|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128195": {
+      "content": "<|reserved_special_token_190|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128196": {
+      "content": "<|reserved_special_token_191|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128197": {
+      "content": "<|reserved_special_token_192|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128198": {
+      "content": "<|reserved_special_token_193|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128199": {
+      "content": "<|reserved_special_token_194|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128200": {
+      "content": "<|reserved_special_token_195|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128201": {
+      "content": "<|reserved_special_token_196|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128202": {
+      "content": "<|reserved_special_token_197|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128203": {
+      "content": "<|reserved_special_token_198|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128204": {
+      "content": "<|reserved_special_token_199|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128205": {
+      "content": "<|reserved_special_token_200|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128206": {
+      "content": "<|reserved_special_token_201|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128207": {
+      "content": "<|reserved_special_token_202|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128208": {
+      "content": "<|reserved_special_token_203|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128209": {
+      "content": "<|reserved_special_token_204|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128210": {
+      "content": "<|reserved_special_token_205|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128211": {
+      "content": "<|reserved_special_token_206|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128212": {
+      "content": "<|reserved_special_token_207|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128213": {
+      "content": "<|reserved_special_token_208|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128214": {
+      "content": "<|reserved_special_token_209|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128215": {
+      "content": "<|reserved_special_token_210|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128216": {
+      "content": "<|reserved_special_token_211|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128217": {
+      "content": "<|reserved_special_token_212|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128218": {
+      "content": "<|reserved_special_token_213|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128219": {
+      "content": "<|reserved_special_token_214|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128220": {
+      "content": "<|reserved_special_token_215|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128221": {
+      "content": "<|reserved_special_token_216|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128222": {
+      "content": "<|reserved_special_token_217|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128223": {
+      "content": "<|reserved_special_token_218|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128224": {
+      "content": "<|reserved_special_token_219|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128225": {
+      "content": "<|reserved_special_token_220|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128226": {
+      "content": "<|reserved_special_token_221|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128227": {
+      "content": "<|reserved_special_token_222|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128228": {
+      "content": "<|reserved_special_token_223|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128229": {
+      "content": "<|reserved_special_token_224|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128230": {
+      "content": "<|reserved_special_token_225|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128231": {
+      "content": "<|reserved_special_token_226|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128232": {
+      "content": "<|reserved_special_token_227|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128233": {
+      "content": "<|reserved_special_token_228|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128234": {
+      "content": "<|reserved_special_token_229|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128235": {
+      "content": "<|reserved_special_token_230|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128236": {
+      "content": "<|reserved_special_token_231|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128237": {
+      "content": "<|reserved_special_token_232|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128238": {
+      "content": "<|reserved_special_token_233|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128239": {
+      "content": "<|reserved_special_token_234|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128240": {
+      "content": "<|reserved_special_token_235|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128241": {
+      "content": "<|reserved_special_token_236|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128242": {
+      "content": "<|reserved_special_token_237|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128243": {
+      "content": "<|reserved_special_token_238|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128244": {
+      "content": "<|reserved_special_token_239|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128245": {
+      "content": "<|reserved_special_token_240|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128246": {
+      "content": "<|reserved_special_token_241|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128247": {
+      "content": "<|reserved_special_token_242|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128248": {
+      "content": "<|reserved_special_token_243|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128249": {
+      "content": "<|reserved_special_token_244|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128250": {
+      "content": "<|reserved_special_token_245|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128251": {
+      "content": "<|reserved_special_token_246|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128252": {
+      "content": "<|reserved_special_token_247|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128253": {
+      "content": "<|reserved_special_token_248|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128254": {
+      "content": "<|reserved_special_token_249|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128255": {
+      "content": "<|reserved_special_token_250|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<|begin_of_text|>",
+  "chat_template": "{% set loop_messages = messages %}{% for message in loop_messages %}{% set content = '<|start_header_id|>' + message['role'] + '<|end_header_id|>\n\n'+ message['content'] | trim + '<|eot_id|>' %}{% if loop.index0 == 0 %}{% set content = bos_token + content %}{% endif %}{{ content }}{% endfor %}{% if add_generation_prompt %}{{ '<|start_header_id|>assistant<|end_header_id|>\n\n' }}{% endif %}",
+  "clean_up_tokenization_spaces": true,
+  "eos_token": "<|eot_id|>",
+  "model_input_names": [
+    "input_ids",
+    "attention_mask"
+  ],
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<|eot_id|>",
+  "tokenizer_class": "PreTrainedTokenizerFast"
+}
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-7462/trainer_state.json b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-7462/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..27ab2ae4a415f5a320f4d7807d8619740f871418
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-7462/trainer_state.json
@@ -0,0 +1,5295 @@
+{
+  "best_metric": 1.8028968572616577,
+  "best_model_checkpoint": "outputs-001/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-2985",
+  "epoch": 4.99966499162479,
+  "eval_steps": 10,
+  "global_step": 7462,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.006700167504187605,
+      "grad_norm": 0.565915048122406,
+      "learning_rate": 0.0002,
+      "loss": 2.6189,
+      "step": 10
+    },
+    {
+      "epoch": 0.01340033500837521,
+      "grad_norm": 0.5004463791847229,
+      "learning_rate": 0.0002,
+      "loss": 2.3162,
+      "step": 20
+    },
+    {
+      "epoch": 0.020100502512562814,
+      "grad_norm": 0.511043906211853,
+      "learning_rate": 0.0002,
+      "loss": 2.0576,
+      "step": 30
+    },
+    {
+      "epoch": 0.02680067001675042,
+      "grad_norm": 0.47327178716659546,
+      "learning_rate": 0.0002,
+      "loss": 2.0085,
+      "step": 40
+    },
+    {
+      "epoch": 0.03350083752093802,
+      "grad_norm": 0.5511676669120789,
+      "learning_rate": 0.0002,
+      "loss": 2.0276,
+      "step": 50
+    },
+    {
+      "epoch": 0.04020100502512563,
+      "grad_norm": 0.4666278064250946,
+      "learning_rate": 0.0002,
+      "loss": 1.9075,
+      "step": 60
+    },
+    {
+      "epoch": 0.04690117252931323,
+      "grad_norm": 0.5310961008071899,
+      "learning_rate": 0.0002,
+      "loss": 1.8413,
+      "step": 70
+    },
+    {
+      "epoch": 0.05360134003350084,
+      "grad_norm": 0.5606027245521545,
+      "learning_rate": 0.0002,
+      "loss": 1.8711,
+      "step": 80
+    },
+    {
+      "epoch": 0.06030150753768844,
+      "grad_norm": 0.4934779703617096,
+      "learning_rate": 0.0002,
+      "loss": 1.9282,
+      "step": 90
+    },
+    {
+      "epoch": 0.06700167504187604,
+      "grad_norm": 0.4821869730949402,
+      "learning_rate": 0.0002,
+      "loss": 1.8925,
+      "step": 100
+    },
+    {
+      "epoch": 0.07370184254606366,
+      "grad_norm": 0.5262084603309631,
+      "learning_rate": 0.0002,
+      "loss": 1.8628,
+      "step": 110
+    },
+    {
+      "epoch": 0.08040201005025126,
+      "grad_norm": 0.3774230182170868,
+      "learning_rate": 0.0002,
+      "loss": 1.8347,
+      "step": 120
+    },
+    {
+      "epoch": 0.08710217755443886,
+      "grad_norm": 0.34137430787086487,
+      "learning_rate": 0.0002,
+      "loss": 1.8386,
+      "step": 130
+    },
+    {
+      "epoch": 0.09380234505862646,
+      "grad_norm": 0.407272070646286,
+      "learning_rate": 0.0002,
+      "loss": 1.861,
+      "step": 140
+    },
+    {
+      "epoch": 0.10050251256281408,
+      "grad_norm": 0.4011937975883484,
+      "learning_rate": 0.0002,
+      "loss": 1.8279,
+      "step": 150
+    },
+    {
+      "epoch": 0.10720268006700168,
+      "grad_norm": 0.4432467222213745,
+      "learning_rate": 0.0002,
+      "loss": 1.9317,
+      "step": 160
+    },
+    {
+      "epoch": 0.11390284757118928,
+      "grad_norm": 0.44030463695526123,
+      "learning_rate": 0.0002,
+      "loss": 1.8157,
+      "step": 170
+    },
+    {
+      "epoch": 0.12060301507537688,
+      "grad_norm": 0.3799569308757782,
+      "learning_rate": 0.0002,
+      "loss": 1.8534,
+      "step": 180
+    },
+    {
+      "epoch": 0.1273031825795645,
+      "grad_norm": 0.33721521496772766,
+      "learning_rate": 0.0002,
+      "loss": 1.7658,
+      "step": 190
+    },
+    {
+      "epoch": 0.13400335008375208,
+      "grad_norm": 0.4096226692199707,
+      "learning_rate": 0.0002,
+      "loss": 1.8269,
+      "step": 200
+    },
+    {
+      "epoch": 0.1407035175879397,
+      "grad_norm": 0.37374693155288696,
+      "learning_rate": 0.0002,
+      "loss": 1.802,
+      "step": 210
+    },
+    {
+      "epoch": 0.1474036850921273,
+      "grad_norm": 0.3249480128288269,
+      "learning_rate": 0.0002,
+      "loss": 1.8901,
+      "step": 220
+    },
+    {
+      "epoch": 0.1541038525963149,
+      "grad_norm": 0.3612042963504791,
+      "learning_rate": 0.0002,
+      "loss": 1.8163,
+      "step": 230
+    },
+    {
+      "epoch": 0.16080402010050251,
+      "grad_norm": 0.3686671257019043,
+      "learning_rate": 0.0002,
+      "loss": 1.7585,
+      "step": 240
+    },
+    {
+      "epoch": 0.16750418760469013,
+      "grad_norm": 0.3521044850349426,
+      "learning_rate": 0.0002,
+      "loss": 1.8365,
+      "step": 250
+    },
+    {
+      "epoch": 0.17420435510887772,
+      "grad_norm": 0.4073677361011505,
+      "learning_rate": 0.0002,
+      "loss": 1.8623,
+      "step": 260
+    },
+    {
+      "epoch": 0.18090452261306533,
+      "grad_norm": 0.34522193670272827,
+      "learning_rate": 0.0002,
+      "loss": 1.8026,
+      "step": 270
+    },
+    {
+      "epoch": 0.18760469011725292,
+      "grad_norm": 0.4121900498867035,
+      "learning_rate": 0.0002,
+      "loss": 1.8162,
+      "step": 280
+    },
+    {
+      "epoch": 0.19430485762144054,
+      "grad_norm": 0.3544778525829315,
+      "learning_rate": 0.0002,
+      "loss": 1.7976,
+      "step": 290
+    },
+    {
+      "epoch": 0.20100502512562815,
+      "grad_norm": 0.3482133448123932,
+      "learning_rate": 0.0002,
+      "loss": 1.8787,
+      "step": 300
+    },
+    {
+      "epoch": 0.20770519262981574,
+      "grad_norm": 0.3421826660633087,
+      "learning_rate": 0.0002,
+      "loss": 1.8578,
+      "step": 310
+    },
+    {
+      "epoch": 0.21440536013400335,
+      "grad_norm": 0.5024696588516235,
+      "learning_rate": 0.0002,
+      "loss": 1.8013,
+      "step": 320
+    },
+    {
+      "epoch": 0.22110552763819097,
+      "grad_norm": 0.36013063788414,
+      "learning_rate": 0.0002,
+      "loss": 1.8607,
+      "step": 330
+    },
+    {
+      "epoch": 0.22780569514237856,
+      "grad_norm": 0.3611244857311249,
+      "learning_rate": 0.0002,
+      "loss": 1.9075,
+      "step": 340
+    },
+    {
+      "epoch": 0.23450586264656617,
+      "grad_norm": 0.39244529604911804,
+      "learning_rate": 0.0002,
+      "loss": 1.8128,
+      "step": 350
+    },
+    {
+      "epoch": 0.24120603015075376,
+      "grad_norm": 0.3299325704574585,
+      "learning_rate": 0.0002,
+      "loss": 1.7885,
+      "step": 360
+    },
+    {
+      "epoch": 0.24790619765494137,
+      "grad_norm": 0.3994322419166565,
+      "learning_rate": 0.0002,
+      "loss": 1.8028,
+      "step": 370
+    },
+    {
+      "epoch": 0.254606365159129,
+      "grad_norm": 0.3559151887893677,
+      "learning_rate": 0.0002,
+      "loss": 1.8321,
+      "step": 380
+    },
+    {
+      "epoch": 0.2613065326633166,
+      "grad_norm": 0.3873756229877472,
+      "learning_rate": 0.0002,
+      "loss": 1.7802,
+      "step": 390
+    },
+    {
+      "epoch": 0.26800670016750416,
+      "grad_norm": 0.3710744082927704,
+      "learning_rate": 0.0002,
+      "loss": 1.7844,
+      "step": 400
+    },
+    {
+      "epoch": 0.2747068676716918,
+      "grad_norm": 0.3618465065956116,
+      "learning_rate": 0.0002,
+      "loss": 1.7789,
+      "step": 410
+    },
+    {
+      "epoch": 0.2814070351758794,
+      "grad_norm": 0.30063769221305847,
+      "learning_rate": 0.0002,
+      "loss": 1.8529,
+      "step": 420
+    },
+    {
+      "epoch": 0.288107202680067,
+      "grad_norm": 0.3695628345012665,
+      "learning_rate": 0.0002,
+      "loss": 1.7765,
+      "step": 430
+    },
+    {
+      "epoch": 0.2948073701842546,
+      "grad_norm": 0.31451135873794556,
+      "learning_rate": 0.0002,
+      "loss": 1.7982,
+      "step": 440
+    },
+    {
+      "epoch": 0.3015075376884422,
+      "grad_norm": 0.3959707021713257,
+      "learning_rate": 0.0002,
+      "loss": 1.7517,
+      "step": 450
+    },
+    {
+      "epoch": 0.3082077051926298,
+      "grad_norm": 0.33475354313850403,
+      "learning_rate": 0.0002,
+      "loss": 1.8142,
+      "step": 460
+    },
+    {
+      "epoch": 0.3149078726968174,
+      "grad_norm": 0.33933115005493164,
+      "learning_rate": 0.0002,
+      "loss": 1.8805,
+      "step": 470
+    },
+    {
+      "epoch": 0.32160804020100503,
+      "grad_norm": 0.3264943063259125,
+      "learning_rate": 0.0002,
+      "loss": 1.7564,
+      "step": 480
+    },
+    {
+      "epoch": 0.32830820770519265,
+      "grad_norm": 0.40188100934028625,
+      "learning_rate": 0.0002,
+      "loss": 1.8428,
+      "step": 490
+    },
+    {
+      "epoch": 0.33500837520938026,
+      "grad_norm": 0.37408649921417236,
+      "learning_rate": 0.0002,
+      "loss": 1.7624,
+      "step": 500
+    },
+    {
+      "epoch": 0.3417085427135678,
+      "grad_norm": 0.33925938606262207,
+      "learning_rate": 0.0002,
+      "loss": 1.7745,
+      "step": 510
+    },
+    {
+      "epoch": 0.34840871021775544,
+      "grad_norm": 0.36836713552474976,
+      "learning_rate": 0.0002,
+      "loss": 1.814,
+      "step": 520
+    },
+    {
+      "epoch": 0.35510887772194305,
+      "grad_norm": 0.37284499406814575,
+      "learning_rate": 0.0002,
+      "loss": 1.8037,
+      "step": 530
+    },
+    {
+      "epoch": 0.36180904522613067,
+      "grad_norm": 0.3192278742790222,
+      "learning_rate": 0.0002,
+      "loss": 1.8379,
+      "step": 540
+    },
+    {
+      "epoch": 0.3685092127303183,
+      "grad_norm": 0.30233290791511536,
+      "learning_rate": 0.0002,
+      "loss": 1.8702,
+      "step": 550
+    },
+    {
+      "epoch": 0.37520938023450584,
+      "grad_norm": 0.3340817391872406,
+      "learning_rate": 0.0002,
+      "loss": 1.8799,
+      "step": 560
+    },
+    {
+      "epoch": 0.38190954773869346,
+      "grad_norm": 0.32600095868110657,
+      "learning_rate": 0.0002,
+      "loss": 1.8404,
+      "step": 570
+    },
+    {
+      "epoch": 0.38860971524288107,
+      "grad_norm": 0.33711278438568115,
+      "learning_rate": 0.0002,
+      "loss": 1.7804,
+      "step": 580
+    },
+    {
+      "epoch": 0.3953098827470687,
+      "grad_norm": 0.34890690445899963,
+      "learning_rate": 0.0002,
+      "loss": 1.8445,
+      "step": 590
+    },
+    {
+      "epoch": 0.4020100502512563,
+      "grad_norm": 0.38238924741744995,
+      "learning_rate": 0.0002,
+      "loss": 1.8187,
+      "step": 600
+    },
+    {
+      "epoch": 0.40871021775544386,
+      "grad_norm": 0.34399354457855225,
+      "learning_rate": 0.0002,
+      "loss": 1.8111,
+      "step": 610
+    },
+    {
+      "epoch": 0.4154103852596315,
+      "grad_norm": 0.3346073627471924,
+      "learning_rate": 0.0002,
+      "loss": 1.8006,
+      "step": 620
+    },
+    {
+      "epoch": 0.4221105527638191,
+      "grad_norm": 0.3545648157596588,
+      "learning_rate": 0.0002,
+      "loss": 1.7705,
+      "step": 630
+    },
+    {
+      "epoch": 0.4288107202680067,
+      "grad_norm": 0.3378899097442627,
+      "learning_rate": 0.0002,
+      "loss": 1.8445,
+      "step": 640
+    },
+    {
+      "epoch": 0.4355108877721943,
+      "grad_norm": 0.3255569040775299,
+      "learning_rate": 0.0002,
+      "loss": 1.804,
+      "step": 650
+    },
+    {
+      "epoch": 0.44221105527638194,
+      "grad_norm": 0.34880587458610535,
+      "learning_rate": 0.0002,
+      "loss": 1.7679,
+      "step": 660
+    },
+    {
+      "epoch": 0.4489112227805695,
+      "grad_norm": 0.3402383625507355,
+      "learning_rate": 0.0002,
+      "loss": 1.7861,
+      "step": 670
+    },
+    {
+      "epoch": 0.4556113902847571,
+      "grad_norm": 0.3594033718109131,
+      "learning_rate": 0.0002,
+      "loss": 1.8131,
+      "step": 680
+    },
+    {
+      "epoch": 0.4623115577889447,
+      "grad_norm": 0.31000566482543945,
+      "learning_rate": 0.0002,
+      "loss": 1.8399,
+      "step": 690
+    },
+    {
+      "epoch": 0.46901172529313234,
+      "grad_norm": 0.37229061126708984,
+      "learning_rate": 0.0002,
+      "loss": 1.7521,
+      "step": 700
+    },
+    {
+      "epoch": 0.47571189279731996,
+      "grad_norm": 0.315801739692688,
+      "learning_rate": 0.0002,
+      "loss": 1.7779,
+      "step": 710
+    },
+    {
+      "epoch": 0.4824120603015075,
+      "grad_norm": 0.3220832645893097,
+      "learning_rate": 0.0002,
+      "loss": 1.7515,
+      "step": 720
+    },
+    {
+      "epoch": 0.48911222780569513,
+      "grad_norm": 0.3435456156730652,
+      "learning_rate": 0.0002,
+      "loss": 1.7181,
+      "step": 730
+    },
+    {
+      "epoch": 0.49581239530988275,
+      "grad_norm": 0.30380892753601074,
+      "learning_rate": 0.0002,
+      "loss": 1.8844,
+      "step": 740
+    },
+    {
+      "epoch": 0.5025125628140703,
+      "grad_norm": 0.3555026054382324,
+      "learning_rate": 0.0002,
+      "loss": 1.7792,
+      "step": 750
+    },
+    {
+      "epoch": 0.509212730318258,
+      "grad_norm": 0.3019855320453644,
+      "learning_rate": 0.0002,
+      "loss": 1.7714,
+      "step": 760
+    },
+    {
+      "epoch": 0.5159128978224455,
+      "grad_norm": 0.309111088514328,
+      "learning_rate": 0.0002,
+      "loss": 1.7962,
+      "step": 770
+    },
+    {
+      "epoch": 0.5226130653266332,
+      "grad_norm": 0.366020530462265,
+      "learning_rate": 0.0002,
+      "loss": 1.7913,
+      "step": 780
+    },
+    {
+      "epoch": 0.5293132328308208,
+      "grad_norm": 0.3267050087451935,
+      "learning_rate": 0.0002,
+      "loss": 1.8008,
+      "step": 790
+    },
+    {
+      "epoch": 0.5360134003350083,
+      "grad_norm": 0.34265750646591187,
+      "learning_rate": 0.0002,
+      "loss": 1.7397,
+      "step": 800
+    },
+    {
+      "epoch": 0.542713567839196,
+      "grad_norm": 0.313669890165329,
+      "learning_rate": 0.0002,
+      "loss": 1.8251,
+      "step": 810
+    },
+    {
+      "epoch": 0.5494137353433836,
+      "grad_norm": 0.3355236053466797,
+      "learning_rate": 0.0002,
+      "loss": 1.8369,
+      "step": 820
+    },
+    {
+      "epoch": 0.5561139028475712,
+      "grad_norm": 0.3186608552932739,
+      "learning_rate": 0.0002,
+      "loss": 1.7381,
+      "step": 830
+    },
+    {
+      "epoch": 0.5628140703517588,
+      "grad_norm": 0.30357518792152405,
+      "learning_rate": 0.0002,
+      "loss": 1.8034,
+      "step": 840
+    },
+    {
+      "epoch": 0.5695142378559463,
+      "grad_norm": 0.3990040123462677,
+      "learning_rate": 0.0002,
+      "loss": 1.769,
+      "step": 850
+    },
+    {
+      "epoch": 0.576214405360134,
+      "grad_norm": 0.34363803267478943,
+      "learning_rate": 0.0002,
+      "loss": 1.7482,
+      "step": 860
+    },
+    {
+      "epoch": 0.5829145728643216,
+      "grad_norm": 0.3757908046245575,
+      "learning_rate": 0.0002,
+      "loss": 1.8106,
+      "step": 870
+    },
+    {
+      "epoch": 0.5896147403685092,
+      "grad_norm": 0.3359757661819458,
+      "learning_rate": 0.0002,
+      "loss": 1.8104,
+      "step": 880
+    },
+    {
+      "epoch": 0.5963149078726968,
+      "grad_norm": 0.5555329918861389,
+      "learning_rate": 0.0002,
+      "loss": 1.7591,
+      "step": 890
+    },
+    {
+      "epoch": 0.6030150753768844,
+      "grad_norm": 0.4046323895454407,
+      "learning_rate": 0.0002,
+      "loss": 1.7715,
+      "step": 900
+    },
+    {
+      "epoch": 0.609715242881072,
+      "grad_norm": 0.29834219813346863,
+      "learning_rate": 0.0002,
+      "loss": 1.7998,
+      "step": 910
+    },
+    {
+      "epoch": 0.6164154103852596,
+      "grad_norm": 0.3241238594055176,
+      "learning_rate": 0.0002,
+      "loss": 1.7826,
+      "step": 920
+    },
+    {
+      "epoch": 0.6231155778894473,
+      "grad_norm": 0.35154739022254944,
+      "learning_rate": 0.0002,
+      "loss": 1.8342,
+      "step": 930
+    },
+    {
+      "epoch": 0.6298157453936348,
+      "grad_norm": 0.3287706673145294,
+      "learning_rate": 0.0002,
+      "loss": 1.8076,
+      "step": 940
+    },
+    {
+      "epoch": 0.6365159128978225,
+      "grad_norm": 0.35670626163482666,
+      "learning_rate": 0.0002,
+      "loss": 1.8038,
+      "step": 950
+    },
+    {
+      "epoch": 0.6432160804020101,
+      "grad_norm": 0.6114104986190796,
+      "learning_rate": 0.0002,
+      "loss": 1.869,
+      "step": 960
+    },
+    {
+      "epoch": 0.6499162479061976,
+      "grad_norm": 0.3186565041542053,
+      "learning_rate": 0.0002,
+      "loss": 1.8297,
+      "step": 970
+    },
+    {
+      "epoch": 0.6566164154103853,
+      "grad_norm": 0.27164125442504883,
+      "learning_rate": 0.0002,
+      "loss": 1.7539,
+      "step": 980
+    },
+    {
+      "epoch": 0.6633165829145728,
+      "grad_norm": 0.34407344460487366,
+      "learning_rate": 0.0002,
+      "loss": 1.8339,
+      "step": 990
+    },
+    {
+      "epoch": 0.6700167504187605,
+      "grad_norm": 0.368415892124176,
+      "learning_rate": 0.0002,
+      "loss": 1.855,
+      "step": 1000
+    },
+    {
+      "epoch": 0.6767169179229481,
+      "grad_norm": 0.3306390643119812,
+      "learning_rate": 0.0002,
+      "loss": 1.7821,
+      "step": 1010
+    },
+    {
+      "epoch": 0.6834170854271356,
+      "grad_norm": 0.3198648989200592,
+      "learning_rate": 0.0002,
+      "loss": 1.786,
+      "step": 1020
+    },
+    {
+      "epoch": 0.6901172529313233,
+      "grad_norm": 0.3092987537384033,
+      "learning_rate": 0.0002,
+      "loss": 1.816,
+      "step": 1030
+    },
+    {
+      "epoch": 0.6968174204355109,
+      "grad_norm": 0.3090653419494629,
+      "learning_rate": 0.0002,
+      "loss": 1.7689,
+      "step": 1040
+    },
+    {
+      "epoch": 0.7035175879396985,
+      "grad_norm": 0.3485880196094513,
+      "learning_rate": 0.0002,
+      "loss": 1.7544,
+      "step": 1050
+    },
+    {
+      "epoch": 0.7102177554438861,
+      "grad_norm": 0.35782721638679504,
+      "learning_rate": 0.0002,
+      "loss": 1.8187,
+      "step": 1060
+    },
+    {
+      "epoch": 0.7169179229480737,
+      "grad_norm": 0.34256869554519653,
+      "learning_rate": 0.0002,
+      "loss": 1.806,
+      "step": 1070
+    },
+    {
+      "epoch": 0.7236180904522613,
+      "grad_norm": 0.30461037158966064,
+      "learning_rate": 0.0002,
+      "loss": 1.7873,
+      "step": 1080
+    },
+    {
+      "epoch": 0.7303182579564489,
+      "grad_norm": 0.3398691713809967,
+      "learning_rate": 0.0002,
+      "loss": 1.7367,
+      "step": 1090
+    },
+    {
+      "epoch": 0.7370184254606366,
+      "grad_norm": 0.3180808126926422,
+      "learning_rate": 0.0002,
+      "loss": 1.8756,
+      "step": 1100
+    },
+    {
+      "epoch": 0.7437185929648241,
+      "grad_norm": 0.34400665760040283,
+      "learning_rate": 0.0002,
+      "loss": 1.6988,
+      "step": 1110
+    },
+    {
+      "epoch": 0.7504187604690117,
+      "grad_norm": 0.34244877099990845,
+      "learning_rate": 0.0002,
+      "loss": 1.7851,
+      "step": 1120
+    },
+    {
+      "epoch": 0.7571189279731994,
+      "grad_norm": 0.29946693778038025,
+      "learning_rate": 0.0002,
+      "loss": 1.7841,
+      "step": 1130
+    },
+    {
+      "epoch": 0.7638190954773869,
+      "grad_norm": 0.37547236680984497,
+      "learning_rate": 0.0002,
+      "loss": 1.7456,
+      "step": 1140
+    },
+    {
+      "epoch": 0.7705192629815746,
+      "grad_norm": 0.3263005018234253,
+      "learning_rate": 0.0002,
+      "loss": 1.8425,
+      "step": 1150
+    },
+    {
+      "epoch": 0.7772194304857621,
+      "grad_norm": 0.41363608837127686,
+      "learning_rate": 0.0002,
+      "loss": 1.7222,
+      "step": 1160
+    },
+    {
+      "epoch": 0.7839195979899497,
+      "grad_norm": 0.36267954111099243,
+      "learning_rate": 0.0002,
+      "loss": 1.7836,
+      "step": 1170
+    },
+    {
+      "epoch": 0.7906197654941374,
+      "grad_norm": 0.31789499521255493,
+      "learning_rate": 0.0002,
+      "loss": 1.9183,
+      "step": 1180
+    },
+    {
+      "epoch": 0.7973199329983249,
+      "grad_norm": 0.5708149075508118,
+      "learning_rate": 0.0002,
+      "loss": 1.78,
+      "step": 1190
+    },
+    {
+      "epoch": 0.8040201005025126,
+      "grad_norm": 0.322099506855011,
+      "learning_rate": 0.0002,
+      "loss": 1.6908,
+      "step": 1200
+    },
+    {
+      "epoch": 0.8107202680067002,
+      "grad_norm": 0.3419909179210663,
+      "learning_rate": 0.0002,
+      "loss": 1.7639,
+      "step": 1210
+    },
+    {
+      "epoch": 0.8174204355108877,
+      "grad_norm": 0.36286255717277527,
+      "learning_rate": 0.0002,
+      "loss": 1.7428,
+      "step": 1220
+    },
+    {
+      "epoch": 0.8241206030150754,
+      "grad_norm": 0.33992862701416016,
+      "learning_rate": 0.0002,
+      "loss": 1.8409,
+      "step": 1230
+    },
+    {
+      "epoch": 0.830820770519263,
+      "grad_norm": 0.32622793316841125,
+      "learning_rate": 0.0002,
+      "loss": 1.7507,
+      "step": 1240
+    },
+    {
+      "epoch": 0.8375209380234506,
+      "grad_norm": 0.3036167621612549,
+      "learning_rate": 0.0002,
+      "loss": 1.8098,
+      "step": 1250
+    },
+    {
+      "epoch": 0.8442211055276382,
+      "grad_norm": 0.3182215392589569,
+      "learning_rate": 0.0002,
+      "loss": 1.8094,
+      "step": 1260
+    },
+    {
+      "epoch": 0.8509212730318257,
+      "grad_norm": 0.3270018696784973,
+      "learning_rate": 0.0002,
+      "loss": 1.8017,
+      "step": 1270
+    },
+    {
+      "epoch": 0.8576214405360134,
+      "grad_norm": 0.32652342319488525,
+      "learning_rate": 0.0002,
+      "loss": 1.798,
+      "step": 1280
+    },
+    {
+      "epoch": 0.864321608040201,
+      "grad_norm": 0.3631329834461212,
+      "learning_rate": 0.0002,
+      "loss": 1.7448,
+      "step": 1290
+    },
+    {
+      "epoch": 0.8710217755443886,
+      "grad_norm": 0.36706018447875977,
+      "learning_rate": 0.0002,
+      "loss": 1.7,
+      "step": 1300
+    },
+    {
+      "epoch": 0.8777219430485762,
+      "grad_norm": 0.3347418010234833,
+      "learning_rate": 0.0002,
+      "loss": 1.8178,
+      "step": 1310
+    },
+    {
+      "epoch": 0.8844221105527639,
+      "grad_norm": 0.34371060132980347,
+      "learning_rate": 0.0002,
+      "loss": 1.7824,
+      "step": 1320
+    },
+    {
+      "epoch": 0.8911222780569514,
+      "grad_norm": 0.3029090166091919,
+      "learning_rate": 0.0002,
+      "loss": 1.783,
+      "step": 1330
+    },
+    {
+      "epoch": 0.897822445561139,
+      "grad_norm": 0.34700682759284973,
+      "learning_rate": 0.0002,
+      "loss": 1.8017,
+      "step": 1340
+    },
+    {
+      "epoch": 0.9045226130653267,
+      "grad_norm": 0.35574328899383545,
+      "learning_rate": 0.0002,
+      "loss": 1.7998,
+      "step": 1350
+    },
+    {
+      "epoch": 0.9112227805695142,
+      "grad_norm": 0.30928221344947815,
+      "learning_rate": 0.0002,
+      "loss": 1.7339,
+      "step": 1360
+    },
+    {
+      "epoch": 0.9179229480737019,
+      "grad_norm": 0.30652928352355957,
+      "learning_rate": 0.0002,
+      "loss": 1.7479,
+      "step": 1370
+    },
+    {
+      "epoch": 0.9246231155778895,
+      "grad_norm": 0.3838157653808594,
+      "learning_rate": 0.0002,
+      "loss": 1.7491,
+      "step": 1380
+    },
+    {
+      "epoch": 0.931323283082077,
+      "grad_norm": 0.31655240058898926,
+      "learning_rate": 0.0002,
+      "loss": 1.7977,
+      "step": 1390
+    },
+    {
+      "epoch": 0.9380234505862647,
+      "grad_norm": 0.41737303137779236,
+      "learning_rate": 0.0002,
+      "loss": 1.8175,
+      "step": 1400
+    },
+    {
+      "epoch": 0.9447236180904522,
+      "grad_norm": 0.3227267861366272,
+      "learning_rate": 0.0002,
+      "loss": 1.6811,
+      "step": 1410
+    },
+    {
+      "epoch": 0.9514237855946399,
+      "grad_norm": 0.3729925751686096,
+      "learning_rate": 0.0002,
+      "loss": 1.7343,
+      "step": 1420
+    },
+    {
+      "epoch": 0.9581239530988275,
+      "grad_norm": 0.30779409408569336,
+      "learning_rate": 0.0002,
+      "loss": 1.8221,
+      "step": 1430
+    },
+    {
+      "epoch": 0.964824120603015,
+      "grad_norm": 0.334379643201828,
+      "learning_rate": 0.0002,
+      "loss": 1.7972,
+      "step": 1440
+    },
+    {
+      "epoch": 0.9715242881072027,
+      "grad_norm": 0.3568236231803894,
+      "learning_rate": 0.0002,
+      "loss": 1.7141,
+      "step": 1450
+    },
+    {
+      "epoch": 0.9782244556113903,
+      "grad_norm": 0.33310577273368835,
+      "learning_rate": 0.0002,
+      "loss": 1.7541,
+      "step": 1460
+    },
+    {
+      "epoch": 0.9849246231155779,
+      "grad_norm": 0.2972261905670166,
+      "learning_rate": 0.0002,
+      "loss": 1.8511,
+      "step": 1470
+    },
+    {
+      "epoch": 0.9916247906197655,
+      "grad_norm": 0.3322717845439911,
+      "learning_rate": 0.0002,
+      "loss": 1.7654,
+      "step": 1480
+    },
+    {
+      "epoch": 0.998324958123953,
+      "grad_norm": 0.3276330828666687,
+      "learning_rate": 0.0002,
+      "loss": 1.8033,
+      "step": 1490
+    },
+    {
+      "epoch": 0.9996649916247906,
+      "eval_loss": 1.8036354780197144,
+      "eval_runtime": 37.8949,
+      "eval_samples_per_second": 13.59,
+      "eval_steps_per_second": 1.715,
+      "step": 1492
+    },
+    {
+      "epoch": 1.0050251256281406,
+      "grad_norm": 0.29252371191978455,
+      "learning_rate": 0.0002,
+      "loss": 1.7138,
+      "step": 1500
+    },
+    {
+      "epoch": 1.0117252931323284,
+      "grad_norm": 0.31607162952423096,
+      "learning_rate": 0.0002,
+      "loss": 1.8198,
+      "step": 1510
+    },
+    {
+      "epoch": 1.018425460636516,
+      "grad_norm": 0.32294467091560364,
+      "learning_rate": 0.0002,
+      "loss": 1.6779,
+      "step": 1520
+    },
+    {
+      "epoch": 1.0251256281407035,
+      "grad_norm": 0.3868017792701721,
+      "learning_rate": 0.0002,
+      "loss": 1.7919,
+      "step": 1530
+    },
+    {
+      "epoch": 1.031825795644891,
+      "grad_norm": 0.3178282082080841,
+      "learning_rate": 0.0002,
+      "loss": 1.7954,
+      "step": 1540
+    },
+    {
+      "epoch": 1.0385259631490786,
+      "grad_norm": 0.3706750273704529,
+      "learning_rate": 0.0002,
+      "loss": 1.7136,
+      "step": 1550
+    },
+    {
+      "epoch": 1.0452261306532664,
+      "grad_norm": 0.33930912613868713,
+      "learning_rate": 0.0002,
+      "loss": 1.7382,
+      "step": 1560
+    },
+    {
+      "epoch": 1.051926298157454,
+      "grad_norm": 0.33970504999160767,
+      "learning_rate": 0.0002,
+      "loss": 1.7602,
+      "step": 1570
+    },
+    {
+      "epoch": 1.0586264656616415,
+      "grad_norm": 0.42553383111953735,
+      "learning_rate": 0.0002,
+      "loss": 1.6573,
+      "step": 1580
+    },
+    {
+      "epoch": 1.065326633165829,
+      "grad_norm": 0.3772421181201935,
+      "learning_rate": 0.0002,
+      "loss": 1.645,
+      "step": 1590
+    },
+    {
+      "epoch": 1.0720268006700167,
+      "grad_norm": 0.34212902188301086,
+      "learning_rate": 0.0002,
+      "loss": 1.7362,
+      "step": 1600
+    },
+    {
+      "epoch": 1.0787269681742044,
+      "grad_norm": 0.3798283338546753,
+      "learning_rate": 0.0002,
+      "loss": 1.7057,
+      "step": 1610
+    },
+    {
+      "epoch": 1.085427135678392,
+      "grad_norm": 0.36909598112106323,
+      "learning_rate": 0.0002,
+      "loss": 1.7468,
+      "step": 1620
+    },
+    {
+      "epoch": 1.0921273031825796,
+      "grad_norm": 0.3344230651855469,
+      "learning_rate": 0.0002,
+      "loss": 1.7807,
+      "step": 1630
+    },
+    {
+      "epoch": 1.0988274706867671,
+      "grad_norm": 0.3862569332122803,
+      "learning_rate": 0.0002,
+      "loss": 1.7111,
+      "step": 1640
+    },
+    {
+      "epoch": 1.1055276381909547,
+      "grad_norm": 0.31188511848449707,
+      "learning_rate": 0.0002,
+      "loss": 1.7163,
+      "step": 1650
+    },
+    {
+      "epoch": 1.1122278056951425,
+      "grad_norm": 0.3563670814037323,
+      "learning_rate": 0.0002,
+      "loss": 1.7263,
+      "step": 1660
+    },
+    {
+      "epoch": 1.11892797319933,
+      "grad_norm": 0.35052165389060974,
+      "learning_rate": 0.0002,
+      "loss": 1.7718,
+      "step": 1670
+    },
+    {
+      "epoch": 1.1256281407035176,
+      "grad_norm": 0.3285699188709259,
+      "learning_rate": 0.0002,
+      "loss": 1.7601,
+      "step": 1680
+    },
+    {
+      "epoch": 1.1323283082077051,
+      "grad_norm": 0.3639393746852875,
+      "learning_rate": 0.0002,
+      "loss": 1.6877,
+      "step": 1690
+    },
+    {
+      "epoch": 1.1390284757118927,
+      "grad_norm": 0.3842753767967224,
+      "learning_rate": 0.0002,
+      "loss": 1.7719,
+      "step": 1700
+    },
+    {
+      "epoch": 1.1457286432160805,
+      "grad_norm": 0.3624933063983917,
+      "learning_rate": 0.0002,
+      "loss": 1.7002,
+      "step": 1710
+    },
+    {
+      "epoch": 1.152428810720268,
+      "grad_norm": 0.3641220033168793,
+      "learning_rate": 0.0002,
+      "loss": 1.7243,
+      "step": 1720
+    },
+    {
+      "epoch": 1.1591289782244556,
+      "grad_norm": 0.32765355706214905,
+      "learning_rate": 0.0002,
+      "loss": 1.752,
+      "step": 1730
+    },
+    {
+      "epoch": 1.1658291457286432,
+      "grad_norm": 0.34974896907806396,
+      "learning_rate": 0.0002,
+      "loss": 1.6556,
+      "step": 1740
+    },
+    {
+      "epoch": 1.1725293132328307,
+      "grad_norm": 0.3910926580429077,
+      "learning_rate": 0.0002,
+      "loss": 1.7273,
+      "step": 1750
+    },
+    {
+      "epoch": 1.1792294807370185,
+      "grad_norm": 0.3564300537109375,
+      "learning_rate": 0.0002,
+      "loss": 1.7204,
+      "step": 1760
+    },
+    {
+      "epoch": 1.185929648241206,
+      "grad_norm": 0.34822574257850647,
+      "learning_rate": 0.0002,
+      "loss": 1.746,
+      "step": 1770
+    },
+    {
+      "epoch": 1.1926298157453936,
+      "grad_norm": 0.36185044050216675,
+      "learning_rate": 0.0002,
+      "loss": 1.7256,
+      "step": 1780
+    },
+    {
+      "epoch": 1.1993299832495812,
+      "grad_norm": 0.34866711497306824,
+      "learning_rate": 0.0002,
+      "loss": 1.6431,
+      "step": 1790
+    },
+    {
+      "epoch": 1.2060301507537687,
+      "grad_norm": 0.4017769992351532,
+      "learning_rate": 0.0002,
+      "loss": 1.8084,
+      "step": 1800
+    },
+    {
+      "epoch": 1.2127303182579565,
+      "grad_norm": 0.32930681109428406,
+      "learning_rate": 0.0002,
+      "loss": 1.6985,
+      "step": 1810
+    },
+    {
+      "epoch": 1.219430485762144,
+      "grad_norm": 0.35951921343803406,
+      "learning_rate": 0.0002,
+      "loss": 1.7606,
+      "step": 1820
+    },
+    {
+      "epoch": 1.2261306532663316,
+      "grad_norm": 0.37366992235183716,
+      "learning_rate": 0.0002,
+      "loss": 1.6933,
+      "step": 1830
+    },
+    {
+      "epoch": 1.2328308207705192,
+      "grad_norm": 0.3565689027309418,
+      "learning_rate": 0.0002,
+      "loss": 1.6737,
+      "step": 1840
+    },
+    {
+      "epoch": 1.2395309882747068,
+      "grad_norm": 0.3692343533039093,
+      "learning_rate": 0.0002,
+      "loss": 1.8013,
+      "step": 1850
+    },
+    {
+      "epoch": 1.2462311557788945,
+      "grad_norm": 0.38426971435546875,
+      "learning_rate": 0.0002,
+      "loss": 1.736,
+      "step": 1860
+    },
+    {
+      "epoch": 1.252931323283082,
+      "grad_norm": 0.33559855818748474,
+      "learning_rate": 0.0002,
+      "loss": 1.7031,
+      "step": 1870
+    },
+    {
+      "epoch": 1.2596314907872697,
+      "grad_norm": 0.34181106090545654,
+      "learning_rate": 0.0002,
+      "loss": 1.7033,
+      "step": 1880
+    },
+    {
+      "epoch": 1.2663316582914572,
+      "grad_norm": 0.3916318416595459,
+      "learning_rate": 0.0002,
+      "loss": 1.7707,
+      "step": 1890
+    },
+    {
+      "epoch": 1.2730318257956448,
+      "grad_norm": 0.3887825012207031,
+      "learning_rate": 0.0002,
+      "loss": 1.6686,
+      "step": 1900
+    },
+    {
+      "epoch": 1.2797319932998326,
+      "grad_norm": 0.33583927154541016,
+      "learning_rate": 0.0002,
+      "loss": 1.7062,
+      "step": 1910
+    },
+    {
+      "epoch": 1.2864321608040201,
+      "grad_norm": 0.37639349699020386,
+      "learning_rate": 0.0002,
+      "loss": 1.717,
+      "step": 1920
+    },
+    {
+      "epoch": 1.2931323283082077,
+      "grad_norm": 0.38059428334236145,
+      "learning_rate": 0.0002,
+      "loss": 1.777,
+      "step": 1930
+    },
+    {
+      "epoch": 1.2998324958123952,
+      "grad_norm": 0.37253183126449585,
+      "learning_rate": 0.0002,
+      "loss": 1.6126,
+      "step": 1940
+    },
+    {
+      "epoch": 1.3065326633165828,
+      "grad_norm": 0.37371566891670227,
+      "learning_rate": 0.0002,
+      "loss": 1.6758,
+      "step": 1950
+    },
+    {
+      "epoch": 1.3132328308207706,
+      "grad_norm": 0.4080910086631775,
+      "learning_rate": 0.0002,
+      "loss": 1.6788,
+      "step": 1960
+    },
+    {
+      "epoch": 1.3199329983249581,
+      "grad_norm": 0.3174354135990143,
+      "learning_rate": 0.0002,
+      "loss": 1.6518,
+      "step": 1970
+    },
+    {
+      "epoch": 1.3266331658291457,
+      "grad_norm": 0.4518888294696808,
+      "learning_rate": 0.0002,
+      "loss": 1.7925,
+      "step": 1980
+    },
+    {
+      "epoch": 1.3333333333333333,
+      "grad_norm": 0.3627921938896179,
+      "learning_rate": 0.0002,
+      "loss": 1.7085,
+      "step": 1990
+    },
+    {
+      "epoch": 1.3400335008375208,
+      "grad_norm": 0.3655930161476135,
+      "learning_rate": 0.0002,
+      "loss": 1.7676,
+      "step": 2000
+    },
+    {
+      "epoch": 1.3467336683417086,
+      "grad_norm": 0.3509993255138397,
+      "learning_rate": 0.0002,
+      "loss": 1.7016,
+      "step": 2010
+    },
+    {
+      "epoch": 1.3534338358458962,
+      "grad_norm": 0.4281129240989685,
+      "learning_rate": 0.0002,
+      "loss": 1.7359,
+      "step": 2020
+    },
+    {
+      "epoch": 1.3601340033500837,
+      "grad_norm": 0.3821414113044739,
+      "learning_rate": 0.0002,
+      "loss": 1.6884,
+      "step": 2030
+    },
+    {
+      "epoch": 1.3668341708542713,
+      "grad_norm": 0.3907586336135864,
+      "learning_rate": 0.0002,
+      "loss": 1.7075,
+      "step": 2040
+    },
+    {
+      "epoch": 1.3735343383584588,
+      "grad_norm": 0.37792932987213135,
+      "learning_rate": 0.0002,
+      "loss": 1.7424,
+      "step": 2050
+    },
+    {
+      "epoch": 1.3802345058626466,
+      "grad_norm": 0.3693985641002655,
+      "learning_rate": 0.0002,
+      "loss": 1.7305,
+      "step": 2060
+    },
+    {
+      "epoch": 1.3869346733668342,
+      "grad_norm": 0.32275936007499695,
+      "learning_rate": 0.0002,
+      "loss": 1.7434,
+      "step": 2070
+    },
+    {
+      "epoch": 1.3936348408710217,
+      "grad_norm": 0.3789440095424652,
+      "learning_rate": 0.0002,
+      "loss": 1.6677,
+      "step": 2080
+    },
+    {
+      "epoch": 1.4003350083752093,
+      "grad_norm": 0.3638380467891693,
+      "learning_rate": 0.0002,
+      "loss": 1.6825,
+      "step": 2090
+    },
+    {
+      "epoch": 1.4070351758793969,
+      "grad_norm": 0.3495481610298157,
+      "learning_rate": 0.0002,
+      "loss": 1.6542,
+      "step": 2100
+    },
+    {
+      "epoch": 1.4137353433835846,
+      "grad_norm": 0.37920597195625305,
+      "learning_rate": 0.0002,
+      "loss": 1.7225,
+      "step": 2110
+    },
+    {
+      "epoch": 1.4204355108877722,
+      "grad_norm": 0.37218064069747925,
+      "learning_rate": 0.0002,
+      "loss": 1.7329,
+      "step": 2120
+    },
+    {
+      "epoch": 1.4271356783919598,
+      "grad_norm": 0.38074082136154175,
+      "learning_rate": 0.0002,
+      "loss": 1.799,
+      "step": 2130
+    },
+    {
+      "epoch": 1.4338358458961473,
+      "grad_norm": 0.3455527126789093,
+      "learning_rate": 0.0002,
+      "loss": 1.7403,
+      "step": 2140
+    },
+    {
+      "epoch": 1.4405360134003349,
+      "grad_norm": 0.3712003529071808,
+      "learning_rate": 0.0002,
+      "loss": 1.776,
+      "step": 2150
+    },
+    {
+      "epoch": 1.4472361809045227,
+      "grad_norm": 0.3786754906177521,
+      "learning_rate": 0.0002,
+      "loss": 1.7619,
+      "step": 2160
+    },
+    {
+      "epoch": 1.4539363484087102,
+      "grad_norm": 0.3879223167896271,
+      "learning_rate": 0.0002,
+      "loss": 1.68,
+      "step": 2170
+    },
+    {
+      "epoch": 1.4606365159128978,
+      "grad_norm": 0.38738805055618286,
+      "learning_rate": 0.0002,
+      "loss": 1.7,
+      "step": 2180
+    },
+    {
+      "epoch": 1.4673366834170856,
+      "grad_norm": 0.39768800139427185,
+      "learning_rate": 0.0002,
+      "loss": 1.7581,
+      "step": 2190
+    },
+    {
+      "epoch": 1.474036850921273,
+      "grad_norm": 0.4172441065311432,
+      "learning_rate": 0.0002,
+      "loss": 1.7671,
+      "step": 2200
+    },
+    {
+      "epoch": 1.4807370184254607,
+      "grad_norm": 0.4043174982070923,
+      "learning_rate": 0.0002,
+      "loss": 1.6736,
+      "step": 2210
+    },
+    {
+      "epoch": 1.4874371859296482,
+      "grad_norm": 0.3750883936882019,
+      "learning_rate": 0.0002,
+      "loss": 1.7444,
+      "step": 2220
+    },
+    {
+      "epoch": 1.4941373534338358,
+      "grad_norm": 0.3552253246307373,
+      "learning_rate": 0.0002,
+      "loss": 1.6861,
+      "step": 2230
+    },
+    {
+      "epoch": 1.5008375209380236,
+      "grad_norm": 0.34607139229774475,
+      "learning_rate": 0.0002,
+      "loss": 1.6471,
+      "step": 2240
+    },
+    {
+      "epoch": 1.507537688442211,
+      "grad_norm": 0.3406706750392914,
+      "learning_rate": 0.0002,
+      "loss": 1.6962,
+      "step": 2250
+    },
+    {
+      "epoch": 1.5142378559463987,
+      "grad_norm": 0.36654895544052124,
+      "learning_rate": 0.0002,
+      "loss": 1.7694,
+      "step": 2260
+    },
+    {
+      "epoch": 1.5209380234505863,
+      "grad_norm": 0.3914054334163666,
+      "learning_rate": 0.0002,
+      "loss": 1.6812,
+      "step": 2270
+    },
+    {
+      "epoch": 1.5276381909547738,
+      "grad_norm": 0.42012137174606323,
+      "learning_rate": 0.0002,
+      "loss": 1.6822,
+      "step": 2280
+    },
+    {
+      "epoch": 1.5343383584589616,
+      "grad_norm": 0.39563435316085815,
+      "learning_rate": 0.0002,
+      "loss": 1.697,
+      "step": 2290
+    },
+    {
+      "epoch": 1.541038525963149,
+      "grad_norm": 0.3508438766002655,
+      "learning_rate": 0.0002,
+      "loss": 1.7491,
+      "step": 2300
+    },
+    {
+      "epoch": 1.5477386934673367,
+      "grad_norm": 0.3785218596458435,
+      "learning_rate": 0.0002,
+      "loss": 1.7727,
+      "step": 2310
+    },
+    {
+      "epoch": 1.5544388609715243,
+      "grad_norm": 0.39377647638320923,
+      "learning_rate": 0.0002,
+      "loss": 1.6963,
+      "step": 2320
+    },
+    {
+      "epoch": 1.5611390284757118,
+      "grad_norm": 0.3391438126564026,
+      "learning_rate": 0.0002,
+      "loss": 1.7263,
+      "step": 2330
+    },
+    {
+      "epoch": 1.5678391959798996,
+      "grad_norm": 0.37944263219833374,
+      "learning_rate": 0.0002,
+      "loss": 1.7722,
+      "step": 2340
+    },
+    {
+      "epoch": 1.574539363484087,
+      "grad_norm": 0.3523491322994232,
+      "learning_rate": 0.0002,
+      "loss": 1.6371,
+      "step": 2350
+    },
+    {
+      "epoch": 1.5812395309882747,
+      "grad_norm": 0.3911575973033905,
+      "learning_rate": 0.0002,
+      "loss": 1.7583,
+      "step": 2360
+    },
+    {
+      "epoch": 1.5879396984924623,
+      "grad_norm": 0.33832186460494995,
+      "learning_rate": 0.0002,
+      "loss": 1.7117,
+      "step": 2370
+    },
+    {
+      "epoch": 1.5946398659966499,
+      "grad_norm": 0.3665979206562042,
+      "learning_rate": 0.0002,
+      "loss": 1.7701,
+      "step": 2380
+    },
+    {
+      "epoch": 1.6013400335008376,
+      "grad_norm": 0.3871748149394989,
+      "learning_rate": 0.0002,
+      "loss": 1.779,
+      "step": 2390
+    },
+    {
+      "epoch": 1.608040201005025,
+      "grad_norm": 0.3586967885494232,
+      "learning_rate": 0.0002,
+      "loss": 1.7109,
+      "step": 2400
+    },
+    {
+      "epoch": 1.6147403685092128,
+      "grad_norm": 0.3563673198223114,
+      "learning_rate": 0.0002,
+      "loss": 1.7096,
+      "step": 2410
+    },
+    {
+      "epoch": 1.6214405360134003,
+      "grad_norm": 0.37588971853256226,
+      "learning_rate": 0.0002,
+      "loss": 1.745,
+      "step": 2420
+    },
+    {
+      "epoch": 1.6281407035175879,
+      "grad_norm": 0.352556437253952,
+      "learning_rate": 0.0002,
+      "loss": 1.7086,
+      "step": 2430
+    },
+    {
+      "epoch": 1.6348408710217757,
+      "grad_norm": 0.3716259300708771,
+      "learning_rate": 0.0002,
+      "loss": 1.6547,
+      "step": 2440
+    },
+    {
+      "epoch": 1.641541038525963,
+      "grad_norm": 0.372001975774765,
+      "learning_rate": 0.0002,
+      "loss": 1.7033,
+      "step": 2450
+    },
+    {
+      "epoch": 1.6482412060301508,
+      "grad_norm": 0.3430042862892151,
+      "learning_rate": 0.0002,
+      "loss": 1.6584,
+      "step": 2460
+    },
+    {
+      "epoch": 1.6549413735343383,
+      "grad_norm": 0.3741483688354492,
+      "learning_rate": 0.0002,
+      "loss": 1.7217,
+      "step": 2470
+    },
+    {
+      "epoch": 1.661641541038526,
+      "grad_norm": 0.3610571324825287,
+      "learning_rate": 0.0002,
+      "loss": 1.7701,
+      "step": 2480
+    },
+    {
+      "epoch": 1.6683417085427137,
+      "grad_norm": 0.4204719066619873,
+      "learning_rate": 0.0002,
+      "loss": 1.7057,
+      "step": 2490
+    },
+    {
+      "epoch": 1.675041876046901,
+      "grad_norm": 0.3938186466693878,
+      "learning_rate": 0.0002,
+      "loss": 1.7954,
+      "step": 2500
+    },
+    {
+      "epoch": 1.6817420435510888,
+      "grad_norm": 0.3421435058116913,
+      "learning_rate": 0.0002,
+      "loss": 1.6633,
+      "step": 2510
+    },
+    {
+      "epoch": 1.6884422110552764,
+      "grad_norm": 0.42441412806510925,
+      "learning_rate": 0.0002,
+      "loss": 1.7996,
+      "step": 2520
+    },
+    {
+      "epoch": 1.695142378559464,
+      "grad_norm": 0.38071519136428833,
+      "learning_rate": 0.0002,
+      "loss": 1.7142,
+      "step": 2530
+    },
+    {
+      "epoch": 1.7018425460636517,
+      "grad_norm": 0.34078919887542725,
+      "learning_rate": 0.0002,
+      "loss": 1.7232,
+      "step": 2540
+    },
+    {
+      "epoch": 1.708542713567839,
+      "grad_norm": 0.412844181060791,
+      "learning_rate": 0.0002,
+      "loss": 1.7126,
+      "step": 2550
+    },
+    {
+      "epoch": 1.7152428810720268,
+      "grad_norm": 0.3753604292869568,
+      "learning_rate": 0.0002,
+      "loss": 1.7149,
+      "step": 2560
+    },
+    {
+      "epoch": 1.7219430485762144,
+      "grad_norm": 0.41588476300239563,
+      "learning_rate": 0.0002,
+      "loss": 1.7011,
+      "step": 2570
+    },
+    {
+      "epoch": 1.728643216080402,
+      "grad_norm": 0.35504111647605896,
+      "learning_rate": 0.0002,
+      "loss": 1.6427,
+      "step": 2580
+    },
+    {
+      "epoch": 1.7353433835845897,
+      "grad_norm": 0.36909720301628113,
+      "learning_rate": 0.0002,
+      "loss": 1.7296,
+      "step": 2590
+    },
+    {
+      "epoch": 1.742043551088777,
+      "grad_norm": 0.4149979054927826,
+      "learning_rate": 0.0002,
+      "loss": 1.7022,
+      "step": 2600
+    },
+    {
+      "epoch": 1.7487437185929648,
+      "grad_norm": 0.38859328627586365,
+      "learning_rate": 0.0002,
+      "loss": 1.77,
+      "step": 2610
+    },
+    {
+      "epoch": 1.7554438860971524,
+      "grad_norm": 0.36738792061805725,
+      "learning_rate": 0.0002,
+      "loss": 1.7036,
+      "step": 2620
+    },
+    {
+      "epoch": 1.76214405360134,
+      "grad_norm": 0.3968178927898407,
+      "learning_rate": 0.0002,
+      "loss": 1.764,
+      "step": 2630
+    },
+    {
+      "epoch": 1.7688442211055277,
+      "grad_norm": 0.3972901999950409,
+      "learning_rate": 0.0002,
+      "loss": 1.7687,
+      "step": 2640
+    },
+    {
+      "epoch": 1.775544388609715,
+      "grad_norm": 0.3949959874153137,
+      "learning_rate": 0.0002,
+      "loss": 1.6774,
+      "step": 2650
+    },
+    {
+      "epoch": 1.7822445561139029,
+      "grad_norm": 0.44074657559394836,
+      "learning_rate": 0.0002,
+      "loss": 1.7247,
+      "step": 2660
+    },
+    {
+      "epoch": 1.7889447236180904,
+      "grad_norm": 0.39743664860725403,
+      "learning_rate": 0.0002,
+      "loss": 1.7188,
+      "step": 2670
+    },
+    {
+      "epoch": 1.795644891122278,
+      "grad_norm": 0.3950406610965729,
+      "learning_rate": 0.0002,
+      "loss": 1.7258,
+      "step": 2680
+    },
+    {
+      "epoch": 1.8023450586264658,
+      "grad_norm": 0.3568263649940491,
+      "learning_rate": 0.0002,
+      "loss": 1.6906,
+      "step": 2690
+    },
+    {
+      "epoch": 1.809045226130653,
+      "grad_norm": 0.3819476366043091,
+      "learning_rate": 0.0002,
+      "loss": 1.6735,
+      "step": 2700
+    },
+    {
+      "epoch": 1.8157453936348409,
+      "grad_norm": 0.3480634391307831,
+      "learning_rate": 0.0002,
+      "loss": 1.7198,
+      "step": 2710
+    },
+    {
+      "epoch": 1.8224455611390284,
+      "grad_norm": 0.3875853419303894,
+      "learning_rate": 0.0002,
+      "loss": 1.7042,
+      "step": 2720
+    },
+    {
+      "epoch": 1.829145728643216,
+      "grad_norm": 0.3441337049007416,
+      "learning_rate": 0.0002,
+      "loss": 1.6988,
+      "step": 2730
+    },
+    {
+      "epoch": 1.8358458961474038,
+      "grad_norm": 0.35692882537841797,
+      "learning_rate": 0.0002,
+      "loss": 1.7647,
+      "step": 2740
+    },
+    {
+      "epoch": 1.8425460636515911,
+      "grad_norm": 0.36959215998649597,
+      "learning_rate": 0.0002,
+      "loss": 1.7033,
+      "step": 2750
+    },
+    {
+      "epoch": 1.849246231155779,
+      "grad_norm": 0.3893393278121948,
+      "learning_rate": 0.0002,
+      "loss": 1.7657,
+      "step": 2760
+    },
+    {
+      "epoch": 1.8559463986599665,
+      "grad_norm": 0.37817293405532837,
+      "learning_rate": 0.0002,
+      "loss": 1.7068,
+      "step": 2770
+    },
+    {
+      "epoch": 1.862646566164154,
+      "grad_norm": 0.36071285605430603,
+      "learning_rate": 0.0002,
+      "loss": 1.761,
+      "step": 2780
+    },
+    {
+      "epoch": 1.8693467336683418,
+      "grad_norm": 0.3758420944213867,
+      "learning_rate": 0.0002,
+      "loss": 1.7623,
+      "step": 2790
+    },
+    {
+      "epoch": 1.8760469011725294,
+      "grad_norm": 0.3889938294887543,
+      "learning_rate": 0.0002,
+      "loss": 1.6743,
+      "step": 2800
+    },
+    {
+      "epoch": 1.882747068676717,
+      "grad_norm": 0.34361857175827026,
+      "learning_rate": 0.0002,
+      "loss": 1.6151,
+      "step": 2810
+    },
+    {
+      "epoch": 1.8894472361809045,
+      "grad_norm": 0.39283323287963867,
+      "learning_rate": 0.0002,
+      "loss": 1.6038,
+      "step": 2820
+    },
+    {
+      "epoch": 1.896147403685092,
+      "grad_norm": 0.3919452726840973,
+      "learning_rate": 0.0002,
+      "loss": 1.7555,
+      "step": 2830
+    },
+    {
+      "epoch": 1.9028475711892798,
+      "grad_norm": 0.38215070962905884,
+      "learning_rate": 0.0002,
+      "loss": 1.673,
+      "step": 2840
+    },
+    {
+      "epoch": 1.9095477386934674,
+      "grad_norm": 0.4235064387321472,
+      "learning_rate": 0.0002,
+      "loss": 1.7044,
+      "step": 2850
+    },
+    {
+      "epoch": 1.916247906197655,
+      "grad_norm": 0.35694634914398193,
+      "learning_rate": 0.0002,
+      "loss": 1.7123,
+      "step": 2860
+    },
+    {
+      "epoch": 1.9229480737018425,
+      "grad_norm": 0.383492112159729,
+      "learning_rate": 0.0002,
+      "loss": 1.8128,
+      "step": 2870
+    },
+    {
+      "epoch": 1.92964824120603,
+      "grad_norm": 0.5945147275924683,
+      "learning_rate": 0.0002,
+      "loss": 1.7581,
+      "step": 2880
+    },
+    {
+      "epoch": 1.9363484087102178,
+      "grad_norm": 0.3367522358894348,
+      "learning_rate": 0.0002,
+      "loss": 1.7421,
+      "step": 2890
+    },
+    {
+      "epoch": 1.9430485762144054,
+      "grad_norm": 0.35300394892692566,
+      "learning_rate": 0.0002,
+      "loss": 1.6561,
+      "step": 2900
+    },
+    {
+      "epoch": 1.949748743718593,
+      "grad_norm": 0.38084495067596436,
+      "learning_rate": 0.0002,
+      "loss": 1.7033,
+      "step": 2910
+    },
+    {
+      "epoch": 1.9564489112227805,
+      "grad_norm": 0.37559160590171814,
+      "learning_rate": 0.0002,
+      "loss": 1.7132,
+      "step": 2920
+    },
+    {
+      "epoch": 1.963149078726968,
+      "grad_norm": 0.3661738336086273,
+      "learning_rate": 0.0002,
+      "loss": 1.6759,
+      "step": 2930
+    },
+    {
+      "epoch": 1.9698492462311559,
+      "grad_norm": 0.4073849320411682,
+      "learning_rate": 0.0002,
+      "loss": 1.7643,
+      "step": 2940
+    },
+    {
+      "epoch": 1.9765494137353434,
+      "grad_norm": 0.3723304271697998,
+      "learning_rate": 0.0002,
+      "loss": 1.6806,
+      "step": 2950
+    },
+    {
+      "epoch": 1.983249581239531,
+      "grad_norm": 0.3991098999977112,
+      "learning_rate": 0.0002,
+      "loss": 1.7611,
+      "step": 2960
+    },
+    {
+      "epoch": 1.9899497487437185,
+      "grad_norm": 0.3947085440158844,
+      "learning_rate": 0.0002,
+      "loss": 1.7263,
+      "step": 2970
+    },
+    {
+      "epoch": 1.996649916247906,
+      "grad_norm": 0.3786258399486542,
+      "learning_rate": 0.0002,
+      "loss": 1.7217,
+      "step": 2980
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 1.8028968572616577,
+      "eval_runtime": 37.8985,
+      "eval_samples_per_second": 13.589,
+      "eval_steps_per_second": 1.715,
+      "step": 2985
+    },
+    {
+      "epoch": 2.003350083752094,
+      "grad_norm": 0.34824079275131226,
+      "learning_rate": 0.0002,
+      "loss": 1.695,
+      "step": 2990
+    },
+    {
+      "epoch": 2.0100502512562812,
+      "grad_norm": 0.3394894003868103,
+      "learning_rate": 0.0002,
+      "loss": 1.5853,
+      "step": 3000
+    },
+    {
+      "epoch": 2.016750418760469,
+      "grad_norm": 0.36910977959632874,
+      "learning_rate": 0.0002,
+      "loss": 1.5783,
+      "step": 3010
+    },
+    {
+      "epoch": 2.023450586264657,
+      "grad_norm": 0.45000967383384705,
+      "learning_rate": 0.0002,
+      "loss": 1.6105,
+      "step": 3020
+    },
+    {
+      "epoch": 2.030150753768844,
+      "grad_norm": 0.3791407346725464,
+      "learning_rate": 0.0002,
+      "loss": 1.6019,
+      "step": 3030
+    },
+    {
+      "epoch": 2.036850921273032,
+      "grad_norm": 0.387321799993515,
+      "learning_rate": 0.0002,
+      "loss": 1.5832,
+      "step": 3040
+    },
+    {
+      "epoch": 2.0435510887772192,
+      "grad_norm": 0.4185757040977478,
+      "learning_rate": 0.0002,
+      "loss": 1.6834,
+      "step": 3050
+    },
+    {
+      "epoch": 2.050251256281407,
+      "grad_norm": 0.45110777020454407,
+      "learning_rate": 0.0002,
+      "loss": 1.5696,
+      "step": 3060
+    },
+    {
+      "epoch": 2.056951423785595,
+      "grad_norm": 0.42663660645484924,
+      "learning_rate": 0.0002,
+      "loss": 1.6231,
+      "step": 3070
+    },
+    {
+      "epoch": 2.063651591289782,
+      "grad_norm": 0.4546292722225189,
+      "learning_rate": 0.0002,
+      "loss": 1.6279,
+      "step": 3080
+    },
+    {
+      "epoch": 2.07035175879397,
+      "grad_norm": 0.3979759216308594,
+      "learning_rate": 0.0002,
+      "loss": 1.6141,
+      "step": 3090
+    },
+    {
+      "epoch": 2.0770519262981573,
+      "grad_norm": 0.43596673011779785,
+      "learning_rate": 0.0002,
+      "loss": 1.6343,
+      "step": 3100
+    },
+    {
+      "epoch": 2.083752093802345,
+      "grad_norm": 0.40120232105255127,
+      "learning_rate": 0.0002,
+      "loss": 1.5441,
+      "step": 3110
+    },
+    {
+      "epoch": 2.090452261306533,
+      "grad_norm": 0.44449281692504883,
+      "learning_rate": 0.0002,
+      "loss": 1.6309,
+      "step": 3120
+    },
+    {
+      "epoch": 2.09715242881072,
+      "grad_norm": 0.42672568559646606,
+      "learning_rate": 0.0002,
+      "loss": 1.5652,
+      "step": 3130
+    },
+    {
+      "epoch": 2.103852596314908,
+      "grad_norm": 0.4232690930366516,
+      "learning_rate": 0.0002,
+      "loss": 1.682,
+      "step": 3140
+    },
+    {
+      "epoch": 2.1105527638190953,
+      "grad_norm": 0.4299317002296448,
+      "learning_rate": 0.0002,
+      "loss": 1.624,
+      "step": 3150
+    },
+    {
+      "epoch": 2.117252931323283,
+      "grad_norm": 0.4067758023738861,
+      "learning_rate": 0.0002,
+      "loss": 1.6766,
+      "step": 3160
+    },
+    {
+      "epoch": 2.123953098827471,
+      "grad_norm": 0.4918815791606903,
+      "learning_rate": 0.0002,
+      "loss": 1.6759,
+      "step": 3170
+    },
+    {
+      "epoch": 2.130653266331658,
+      "grad_norm": 0.4140559732913971,
+      "learning_rate": 0.0002,
+      "loss": 1.6478,
+      "step": 3180
+    },
+    {
+      "epoch": 2.137353433835846,
+      "grad_norm": 0.4555995464324951,
+      "learning_rate": 0.0002,
+      "loss": 1.6641,
+      "step": 3190
+    },
+    {
+      "epoch": 2.1440536013400333,
+      "grad_norm": 0.42943915724754333,
+      "learning_rate": 0.0002,
+      "loss": 1.5888,
+      "step": 3200
+    },
+    {
+      "epoch": 2.150753768844221,
+      "grad_norm": 0.4730435013771057,
+      "learning_rate": 0.0002,
+      "loss": 1.5886,
+      "step": 3210
+    },
+    {
+      "epoch": 2.157453936348409,
+      "grad_norm": 0.43310216069221497,
+      "learning_rate": 0.0002,
+      "loss": 1.6022,
+      "step": 3220
+    },
+    {
+      "epoch": 2.164154103852596,
+      "grad_norm": 0.42054110765457153,
+      "learning_rate": 0.0002,
+      "loss": 1.6058,
+      "step": 3230
+    },
+    {
+      "epoch": 2.170854271356784,
+      "grad_norm": 0.4897233247756958,
+      "learning_rate": 0.0002,
+      "loss": 1.6749,
+      "step": 3240
+    },
+    {
+      "epoch": 2.1775544388609713,
+      "grad_norm": 0.42194533348083496,
+      "learning_rate": 0.0002,
+      "loss": 1.6983,
+      "step": 3250
+    },
+    {
+      "epoch": 2.184254606365159,
+      "grad_norm": 0.44494450092315674,
+      "learning_rate": 0.0002,
+      "loss": 1.6159,
+      "step": 3260
+    },
+    {
+      "epoch": 2.190954773869347,
+      "grad_norm": 0.43524879217147827,
+      "learning_rate": 0.0002,
+      "loss": 1.6977,
+      "step": 3270
+    },
+    {
+      "epoch": 2.1976549413735342,
+      "grad_norm": 0.4621117413043976,
+      "learning_rate": 0.0002,
+      "loss": 1.528,
+      "step": 3280
+    },
+    {
+      "epoch": 2.204355108877722,
+      "grad_norm": 0.4073285460472107,
+      "learning_rate": 0.0002,
+      "loss": 1.632,
+      "step": 3290
+    },
+    {
+      "epoch": 2.2110552763819094,
+      "grad_norm": 0.47868335247039795,
+      "learning_rate": 0.0002,
+      "loss": 1.6141,
+      "step": 3300
+    },
+    {
+      "epoch": 2.217755443886097,
+      "grad_norm": 0.4264970123767853,
+      "learning_rate": 0.0002,
+      "loss": 1.6857,
+      "step": 3310
+    },
+    {
+      "epoch": 2.224455611390285,
+      "grad_norm": 0.4491245150566101,
+      "learning_rate": 0.0002,
+      "loss": 1.5653,
+      "step": 3320
+    },
+    {
+      "epoch": 2.2311557788944723,
+      "grad_norm": 0.4010344445705414,
+      "learning_rate": 0.0002,
+      "loss": 1.5881,
+      "step": 3330
+    },
+    {
+      "epoch": 2.23785594639866,
+      "grad_norm": 0.4232759177684784,
+      "learning_rate": 0.0002,
+      "loss": 1.6684,
+      "step": 3340
+    },
+    {
+      "epoch": 2.2445561139028474,
+      "grad_norm": 0.5099776983261108,
+      "learning_rate": 0.0002,
+      "loss": 1.6336,
+      "step": 3350
+    },
+    {
+      "epoch": 2.251256281407035,
+      "grad_norm": 0.5223407745361328,
+      "learning_rate": 0.0002,
+      "loss": 1.6764,
+      "step": 3360
+    },
+    {
+      "epoch": 2.257956448911223,
+      "grad_norm": 0.47818470001220703,
+      "learning_rate": 0.0002,
+      "loss": 1.6625,
+      "step": 3370
+    },
+    {
+      "epoch": 2.2646566164154103,
+      "grad_norm": 0.4721255898475647,
+      "learning_rate": 0.0002,
+      "loss": 1.5946,
+      "step": 3380
+    },
+    {
+      "epoch": 2.271356783919598,
+      "grad_norm": 0.4113229513168335,
+      "learning_rate": 0.0002,
+      "loss": 1.5568,
+      "step": 3390
+    },
+    {
+      "epoch": 2.2780569514237854,
+      "grad_norm": 0.507080078125,
+      "learning_rate": 0.0002,
+      "loss": 1.6494,
+      "step": 3400
+    },
+    {
+      "epoch": 2.284757118927973,
+      "grad_norm": 0.4852292239665985,
+      "learning_rate": 0.0002,
+      "loss": 1.6183,
+      "step": 3410
+    },
+    {
+      "epoch": 2.291457286432161,
+      "grad_norm": 0.4503684341907501,
+      "learning_rate": 0.0002,
+      "loss": 1.6132,
+      "step": 3420
+    },
+    {
+      "epoch": 2.2981574539363483,
+      "grad_norm": 0.8359600305557251,
+      "learning_rate": 0.0002,
+      "loss": 1.6649,
+      "step": 3430
+    },
+    {
+      "epoch": 2.304857621440536,
+      "grad_norm": 0.44604045152664185,
+      "learning_rate": 0.0002,
+      "loss": 1.6644,
+      "step": 3440
+    },
+    {
+      "epoch": 2.3115577889447234,
+      "grad_norm": 0.45667049288749695,
+      "learning_rate": 0.0002,
+      "loss": 1.5972,
+      "step": 3450
+    },
+    {
+      "epoch": 2.318257956448911,
+      "grad_norm": 0.4879349172115326,
+      "learning_rate": 0.0002,
+      "loss": 1.6667,
+      "step": 3460
+    },
+    {
+      "epoch": 2.324958123953099,
+      "grad_norm": 0.4033963084220886,
+      "learning_rate": 0.0002,
+      "loss": 1.5804,
+      "step": 3470
+    },
+    {
+      "epoch": 2.3316582914572863,
+      "grad_norm": 0.44494301080703735,
+      "learning_rate": 0.0002,
+      "loss": 1.5838,
+      "step": 3480
+    },
+    {
+      "epoch": 2.338358458961474,
+      "grad_norm": 0.4794621765613556,
+      "learning_rate": 0.0002,
+      "loss": 1.5951,
+      "step": 3490
+    },
+    {
+      "epoch": 2.3450586264656614,
+      "grad_norm": 0.41404327750205994,
+      "learning_rate": 0.0002,
+      "loss": 1.6807,
+      "step": 3500
+    },
+    {
+      "epoch": 2.351758793969849,
+      "grad_norm": 0.4664851725101471,
+      "learning_rate": 0.0002,
+      "loss": 1.714,
+      "step": 3510
+    },
+    {
+      "epoch": 2.358458961474037,
+      "grad_norm": 0.4263697564601898,
+      "learning_rate": 0.0002,
+      "loss": 1.6537,
+      "step": 3520
+    },
+    {
+      "epoch": 2.3651591289782243,
+      "grad_norm": 0.5035167336463928,
+      "learning_rate": 0.0002,
+      "loss": 1.6551,
+      "step": 3530
+    },
+    {
+      "epoch": 2.371859296482412,
+      "grad_norm": 0.4380664527416229,
+      "learning_rate": 0.0002,
+      "loss": 1.6208,
+      "step": 3540
+    },
+    {
+      "epoch": 2.3785594639865995,
+      "grad_norm": 0.5227681994438171,
+      "learning_rate": 0.0002,
+      "loss": 1.634,
+      "step": 3550
+    },
+    {
+      "epoch": 2.3852596314907872,
+      "grad_norm": 0.4382302761077881,
+      "learning_rate": 0.0002,
+      "loss": 1.6146,
+      "step": 3560
+    },
+    {
+      "epoch": 2.391959798994975,
+      "grad_norm": 0.4392451047897339,
+      "learning_rate": 0.0002,
+      "loss": 1.5653,
+      "step": 3570
+    },
+    {
+      "epoch": 2.3986599664991624,
+      "grad_norm": 0.4372786581516266,
+      "learning_rate": 0.0002,
+      "loss": 1.6626,
+      "step": 3580
+    },
+    {
+      "epoch": 2.40536013400335,
+      "grad_norm": 0.5015502572059631,
+      "learning_rate": 0.0002,
+      "loss": 1.519,
+      "step": 3590
+    },
+    {
+      "epoch": 2.4120603015075375,
+      "grad_norm": 0.5653210878372192,
+      "learning_rate": 0.0002,
+      "loss": 1.588,
+      "step": 3600
+    },
+    {
+      "epoch": 2.4187604690117253,
+      "grad_norm": 0.53007972240448,
+      "learning_rate": 0.0002,
+      "loss": 1.6075,
+      "step": 3610
+    },
+    {
+      "epoch": 2.425460636515913,
+      "grad_norm": 0.4659176766872406,
+      "learning_rate": 0.0002,
+      "loss": 1.6421,
+      "step": 3620
+    },
+    {
+      "epoch": 2.4321608040201004,
+      "grad_norm": 0.5637837052345276,
+      "learning_rate": 0.0002,
+      "loss": 1.625,
+      "step": 3630
+    },
+    {
+      "epoch": 2.438860971524288,
+      "grad_norm": 0.4248391389846802,
+      "learning_rate": 0.0002,
+      "loss": 1.6168,
+      "step": 3640
+    },
+    {
+      "epoch": 2.4455611390284755,
+      "grad_norm": 0.44668248295783997,
+      "learning_rate": 0.0002,
+      "loss": 1.6822,
+      "step": 3650
+    },
+    {
+      "epoch": 2.4522613065326633,
+      "grad_norm": 0.43990179896354675,
+      "learning_rate": 0.0002,
+      "loss": 1.6417,
+      "step": 3660
+    },
+    {
+      "epoch": 2.458961474036851,
+      "grad_norm": 0.4532523453235626,
+      "learning_rate": 0.0002,
+      "loss": 1.6723,
+      "step": 3670
+    },
+    {
+      "epoch": 2.4656616415410384,
+      "grad_norm": 0.6605591773986816,
+      "learning_rate": 0.0002,
+      "loss": 1.6957,
+      "step": 3680
+    },
+    {
+      "epoch": 2.472361809045226,
+      "grad_norm": 0.4694533348083496,
+      "learning_rate": 0.0002,
+      "loss": 1.6159,
+      "step": 3690
+    },
+    {
+      "epoch": 2.4790619765494135,
+      "grad_norm": 0.4485011100769043,
+      "learning_rate": 0.0002,
+      "loss": 1.6239,
+      "step": 3700
+    },
+    {
+      "epoch": 2.4857621440536013,
+      "grad_norm": 0.4761785864830017,
+      "learning_rate": 0.0002,
+      "loss": 1.6834,
+      "step": 3710
+    },
+    {
+      "epoch": 2.492462311557789,
+      "grad_norm": 0.5116432309150696,
+      "learning_rate": 0.0002,
+      "loss": 1.6313,
+      "step": 3720
+    },
+    {
+      "epoch": 2.4991624790619764,
+      "grad_norm": 0.49523618817329407,
+      "learning_rate": 0.0002,
+      "loss": 1.5054,
+      "step": 3730
+    },
+    {
+      "epoch": 2.505862646566164,
+      "grad_norm": 0.43826380372047424,
+      "learning_rate": 0.0002,
+      "loss": 1.6249,
+      "step": 3740
+    },
+    {
+      "epoch": 2.5125628140703515,
+      "grad_norm": 0.4916154146194458,
+      "learning_rate": 0.0002,
+      "loss": 1.5762,
+      "step": 3750
+    },
+    {
+      "epoch": 2.5192629815745393,
+      "grad_norm": 0.5381299257278442,
+      "learning_rate": 0.0002,
+      "loss": 1.5157,
+      "step": 3760
+    },
+    {
+      "epoch": 2.525963149078727,
+      "grad_norm": 0.44947415590286255,
+      "learning_rate": 0.0002,
+      "loss": 1.6467,
+      "step": 3770
+    },
+    {
+      "epoch": 2.5326633165829144,
+      "grad_norm": 0.49979084730148315,
+      "learning_rate": 0.0002,
+      "loss": 1.67,
+      "step": 3780
+    },
+    {
+      "epoch": 2.539363484087102,
+      "grad_norm": 0.43046900629997253,
+      "learning_rate": 0.0002,
+      "loss": 1.622,
+      "step": 3790
+    },
+    {
+      "epoch": 2.5460636515912896,
+      "grad_norm": 0.4513470530509949,
+      "learning_rate": 0.0002,
+      "loss": 1.6789,
+      "step": 3800
+    },
+    {
+      "epoch": 2.5527638190954773,
+      "grad_norm": 0.49900051951408386,
+      "learning_rate": 0.0002,
+      "loss": 1.6335,
+      "step": 3810
+    },
+    {
+      "epoch": 2.559463986599665,
+      "grad_norm": 0.4348420202732086,
+      "learning_rate": 0.0002,
+      "loss": 1.6322,
+      "step": 3820
+    },
+    {
+      "epoch": 2.5661641541038525,
+      "grad_norm": 0.4684867560863495,
+      "learning_rate": 0.0002,
+      "loss": 1.6218,
+      "step": 3830
+    },
+    {
+      "epoch": 2.5728643216080402,
+      "grad_norm": 0.44430989027023315,
+      "learning_rate": 0.0002,
+      "loss": 1.6535,
+      "step": 3840
+    },
+    {
+      "epoch": 2.5795644891122276,
+      "grad_norm": 0.47375255823135376,
+      "learning_rate": 0.0002,
+      "loss": 1.5909,
+      "step": 3850
+    },
+    {
+      "epoch": 2.5862646566164154,
+      "grad_norm": 0.45493075251579285,
+      "learning_rate": 0.0002,
+      "loss": 1.6269,
+      "step": 3860
+    },
+    {
+      "epoch": 2.592964824120603,
+      "grad_norm": 0.4563275873661041,
+      "learning_rate": 0.0002,
+      "loss": 1.604,
+      "step": 3870
+    },
+    {
+      "epoch": 2.5996649916247905,
+      "grad_norm": 0.46060335636138916,
+      "learning_rate": 0.0002,
+      "loss": 1.642,
+      "step": 3880
+    },
+    {
+      "epoch": 2.6063651591289783,
+      "grad_norm": 0.4718867540359497,
+      "learning_rate": 0.0002,
+      "loss": 1.6302,
+      "step": 3890
+    },
+    {
+      "epoch": 2.6130653266331656,
+      "grad_norm": 0.41570305824279785,
+      "learning_rate": 0.0002,
+      "loss": 1.6242,
+      "step": 3900
+    },
+    {
+      "epoch": 2.6197654941373534,
+      "grad_norm": 0.4603121876716614,
+      "learning_rate": 0.0002,
+      "loss": 1.6401,
+      "step": 3910
+    },
+    {
+      "epoch": 2.626465661641541,
+      "grad_norm": 0.4734652638435364,
+      "learning_rate": 0.0002,
+      "loss": 1.6839,
+      "step": 3920
+    },
+    {
+      "epoch": 2.6331658291457285,
+      "grad_norm": 0.45348483324050903,
+      "learning_rate": 0.0002,
+      "loss": 1.5448,
+      "step": 3930
+    },
+    {
+      "epoch": 2.6398659966499163,
+      "grad_norm": 0.46559447050094604,
+      "learning_rate": 0.0002,
+      "loss": 1.6157,
+      "step": 3940
+    },
+    {
+      "epoch": 2.6465661641541036,
+      "grad_norm": 0.44113144278526306,
+      "learning_rate": 0.0002,
+      "loss": 1.7052,
+      "step": 3950
+    },
+    {
+      "epoch": 2.6532663316582914,
+      "grad_norm": 0.41415104269981384,
+      "learning_rate": 0.0002,
+      "loss": 1.6315,
+      "step": 3960
+    },
+    {
+      "epoch": 2.659966499162479,
+      "grad_norm": 0.48868080973625183,
+      "learning_rate": 0.0002,
+      "loss": 1.6589,
+      "step": 3970
+    },
+    {
+      "epoch": 2.6666666666666665,
+      "grad_norm": 0.49610549211502075,
+      "learning_rate": 0.0002,
+      "loss": 1.6211,
+      "step": 3980
+    },
+    {
+      "epoch": 2.6733668341708543,
+      "grad_norm": 0.4309130907058716,
+      "learning_rate": 0.0002,
+      "loss": 1.6235,
+      "step": 3990
+    },
+    {
+      "epoch": 2.6800670016750416,
+      "grad_norm": 0.4489327669143677,
+      "learning_rate": 0.0002,
+      "loss": 1.6452,
+      "step": 4000
+    },
+    {
+      "epoch": 2.6867671691792294,
+      "grad_norm": 0.5380139946937561,
+      "learning_rate": 0.0002,
+      "loss": 1.5744,
+      "step": 4010
+    },
+    {
+      "epoch": 2.693467336683417,
+      "grad_norm": 0.5076672434806824,
+      "learning_rate": 0.0002,
+      "loss": 1.6524,
+      "step": 4020
+    },
+    {
+      "epoch": 2.7001675041876045,
+      "grad_norm": 0.47620031237602234,
+      "learning_rate": 0.0002,
+      "loss": 1.636,
+      "step": 4030
+    },
+    {
+      "epoch": 2.7068676716917923,
+      "grad_norm": 0.48089155554771423,
+      "learning_rate": 0.0002,
+      "loss": 1.5543,
+      "step": 4040
+    },
+    {
+      "epoch": 2.7135678391959797,
+      "grad_norm": 0.5108814239501953,
+      "learning_rate": 0.0002,
+      "loss": 1.6396,
+      "step": 4050
+    },
+    {
+      "epoch": 2.7202680067001674,
+      "grad_norm": 0.4196513295173645,
+      "learning_rate": 0.0002,
+      "loss": 1.5905,
+      "step": 4060
+    },
+    {
+      "epoch": 2.726968174204355,
+      "grad_norm": 0.4574664831161499,
+      "learning_rate": 0.0002,
+      "loss": 1.686,
+      "step": 4070
+    },
+    {
+      "epoch": 2.7336683417085426,
+      "grad_norm": 0.4671640992164612,
+      "learning_rate": 0.0002,
+      "loss": 1.6234,
+      "step": 4080
+    },
+    {
+      "epoch": 2.7403685092127303,
+      "grad_norm": 0.49355530738830566,
+      "learning_rate": 0.0002,
+      "loss": 1.6827,
+      "step": 4090
+    },
+    {
+      "epoch": 2.7470686767169177,
+      "grad_norm": 0.46716663241386414,
+      "learning_rate": 0.0002,
+      "loss": 1.6999,
+      "step": 4100
+    },
+    {
+      "epoch": 2.7537688442211055,
+      "grad_norm": 0.45420581102371216,
+      "learning_rate": 0.0002,
+      "loss": 1.6463,
+      "step": 4110
+    },
+    {
+      "epoch": 2.7604690117252932,
+      "grad_norm": 0.4680487811565399,
+      "learning_rate": 0.0002,
+      "loss": 1.5718,
+      "step": 4120
+    },
+    {
+      "epoch": 2.7671691792294806,
+      "grad_norm": 0.5375032424926758,
+      "learning_rate": 0.0002,
+      "loss": 1.5968,
+      "step": 4130
+    },
+    {
+      "epoch": 2.7738693467336684,
+      "grad_norm": 0.46026280522346497,
+      "learning_rate": 0.0002,
+      "loss": 1.5254,
+      "step": 4140
+    },
+    {
+      "epoch": 2.7805695142378557,
+      "grad_norm": 0.43658447265625,
+      "learning_rate": 0.0002,
+      "loss": 1.6613,
+      "step": 4150
+    },
+    {
+      "epoch": 2.7872696817420435,
+      "grad_norm": 0.4935547113418579,
+      "learning_rate": 0.0002,
+      "loss": 1.6546,
+      "step": 4160
+    },
+    {
+      "epoch": 2.7939698492462313,
+      "grad_norm": 0.8167962431907654,
+      "learning_rate": 0.0002,
+      "loss": 1.5961,
+      "step": 4170
+    },
+    {
+      "epoch": 2.8006700167504186,
+      "grad_norm": 0.4289683997631073,
+      "learning_rate": 0.0002,
+      "loss": 1.6907,
+      "step": 4180
+    },
+    {
+      "epoch": 2.8073701842546064,
+      "grad_norm": 0.4569324254989624,
+      "learning_rate": 0.0002,
+      "loss": 1.6385,
+      "step": 4190
+    },
+    {
+      "epoch": 2.8140703517587937,
+      "grad_norm": 0.474795937538147,
+      "learning_rate": 0.0002,
+      "loss": 1.6077,
+      "step": 4200
+    },
+    {
+      "epoch": 2.8207705192629815,
+      "grad_norm": 0.44272229075431824,
+      "learning_rate": 0.0002,
+      "loss": 1.6223,
+      "step": 4210
+    },
+    {
+      "epoch": 2.8274706867671693,
+      "grad_norm": 0.525240957736969,
+      "learning_rate": 0.0002,
+      "loss": 1.6706,
+      "step": 4220
+    },
+    {
+      "epoch": 2.8341708542713566,
+      "grad_norm": 0.4802303910255432,
+      "learning_rate": 0.0002,
+      "loss": 1.7196,
+      "step": 4230
+    },
+    {
+      "epoch": 2.8408710217755444,
+      "grad_norm": 0.46400442719459534,
+      "learning_rate": 0.0002,
+      "loss": 1.6002,
+      "step": 4240
+    },
+    {
+      "epoch": 2.8475711892797317,
+      "grad_norm": 0.49884888529777527,
+      "learning_rate": 0.0002,
+      "loss": 1.6052,
+      "step": 4250
+    },
+    {
+      "epoch": 2.8542713567839195,
+      "grad_norm": 0.5015072226524353,
+      "learning_rate": 0.0002,
+      "loss": 1.6919,
+      "step": 4260
+    },
+    {
+      "epoch": 2.8609715242881073,
+      "grad_norm": 0.4335440695285797,
+      "learning_rate": 0.0002,
+      "loss": 1.6335,
+      "step": 4270
+    },
+    {
+      "epoch": 2.8676716917922946,
+      "grad_norm": 0.5131644606590271,
+      "learning_rate": 0.0002,
+      "loss": 1.5664,
+      "step": 4280
+    },
+    {
+      "epoch": 2.8743718592964824,
+      "grad_norm": 0.6977195739746094,
+      "learning_rate": 0.0002,
+      "loss": 1.6409,
+      "step": 4290
+    },
+    {
+      "epoch": 2.8810720268006698,
+      "grad_norm": 0.5133762955665588,
+      "learning_rate": 0.0002,
+      "loss": 1.7192,
+      "step": 4300
+    },
+    {
+      "epoch": 2.8877721943048575,
+      "grad_norm": 0.4737614393234253,
+      "learning_rate": 0.0002,
+      "loss": 1.6257,
+      "step": 4310
+    },
+    {
+      "epoch": 2.8944723618090453,
+      "grad_norm": 0.4580535590648651,
+      "learning_rate": 0.0002,
+      "loss": 1.6076,
+      "step": 4320
+    },
+    {
+      "epoch": 2.901172529313233,
+      "grad_norm": 0.43863341212272644,
+      "learning_rate": 0.0002,
+      "loss": 1.6538,
+      "step": 4330
+    },
+    {
+      "epoch": 2.9078726968174204,
+      "grad_norm": 0.4103737473487854,
+      "learning_rate": 0.0002,
+      "loss": 1.6091,
+      "step": 4340
+    },
+    {
+      "epoch": 2.914572864321608,
+      "grad_norm": 0.438014417886734,
+      "learning_rate": 0.0002,
+      "loss": 1.7106,
+      "step": 4350
+    },
+    {
+      "epoch": 2.9212730318257956,
+      "grad_norm": 0.5068213939666748,
+      "learning_rate": 0.0002,
+      "loss": 1.6025,
+      "step": 4360
+    },
+    {
+      "epoch": 2.9279731993299833,
+      "grad_norm": 0.45305484533309937,
+      "learning_rate": 0.0002,
+      "loss": 1.6426,
+      "step": 4370
+    },
+    {
+      "epoch": 2.934673366834171,
+      "grad_norm": 0.4612090289592743,
+      "learning_rate": 0.0002,
+      "loss": 1.5726,
+      "step": 4380
+    },
+    {
+      "epoch": 2.9413735343383585,
+      "grad_norm": 0.508736789226532,
+      "learning_rate": 0.0002,
+      "loss": 1.6536,
+      "step": 4390
+    },
+    {
+      "epoch": 2.948073701842546,
+      "grad_norm": 0.4924427270889282,
+      "learning_rate": 0.0002,
+      "loss": 1.6132,
+      "step": 4400
+    },
+    {
+      "epoch": 2.9547738693467336,
+      "grad_norm": 0.5707460641860962,
+      "learning_rate": 0.0002,
+      "loss": 1.7007,
+      "step": 4410
+    },
+    {
+      "epoch": 2.9614740368509214,
+      "grad_norm": 0.42270299792289734,
+      "learning_rate": 0.0002,
+      "loss": 1.6814,
+      "step": 4420
+    },
+    {
+      "epoch": 2.968174204355109,
+      "grad_norm": 0.4429931044578552,
+      "learning_rate": 0.0002,
+      "loss": 1.6644,
+      "step": 4430
+    },
+    {
+      "epoch": 2.9748743718592965,
+      "grad_norm": 0.49760574102401733,
+      "learning_rate": 0.0002,
+      "loss": 1.6251,
+      "step": 4440
+    },
+    {
+      "epoch": 2.981574539363484,
+      "grad_norm": 0.4558229148387909,
+      "learning_rate": 0.0002,
+      "loss": 1.6169,
+      "step": 4450
+    },
+    {
+      "epoch": 2.9882747068676716,
+      "grad_norm": 0.39848530292510986,
+      "learning_rate": 0.0002,
+      "loss": 1.6055,
+      "step": 4460
+    },
+    {
+      "epoch": 2.9949748743718594,
+      "grad_norm": 0.5224862098693848,
+      "learning_rate": 0.0002,
+      "loss": 1.6705,
+      "step": 4470
+    },
+    {
+      "epoch": 2.9996649916247904,
+      "eval_loss": 1.8228833675384521,
+      "eval_runtime": 37.9049,
+      "eval_samples_per_second": 13.587,
+      "eval_steps_per_second": 1.715,
+      "step": 4477
+    },
+    {
+      "epoch": 3.0016750418760467,
+      "grad_norm": 0.41169142723083496,
+      "learning_rate": 0.0002,
+      "loss": 1.6637,
+      "step": 4480
+    },
+    {
+      "epoch": 3.0083752093802345,
+      "grad_norm": 0.4865207374095917,
+      "learning_rate": 0.0002,
+      "loss": 1.5974,
+      "step": 4490
+    },
+    {
+      "epoch": 3.0150753768844223,
+      "grad_norm": 0.5462028384208679,
+      "learning_rate": 0.0002,
+      "loss": 1.5297,
+      "step": 4500
+    },
+    {
+      "epoch": 3.0217755443886096,
+      "grad_norm": 0.6169732809066772,
+      "learning_rate": 0.0002,
+      "loss": 1.5251,
+      "step": 4510
+    },
+    {
+      "epoch": 3.0284757118927974,
+      "grad_norm": 0.5667954087257385,
+      "learning_rate": 0.0002,
+      "loss": 1.5559,
+      "step": 4520
+    },
+    {
+      "epoch": 3.0351758793969847,
+      "grad_norm": 0.5758325457572937,
+      "learning_rate": 0.0002,
+      "loss": 1.5037,
+      "step": 4530
+    },
+    {
+      "epoch": 3.0418760469011725,
+      "grad_norm": 0.5220064520835876,
+      "learning_rate": 0.0002,
+      "loss": 1.4873,
+      "step": 4540
+    },
+    {
+      "epoch": 3.0485762144053603,
+      "grad_norm": 0.5469558835029602,
+      "learning_rate": 0.0002,
+      "loss": 1.5126,
+      "step": 4550
+    },
+    {
+      "epoch": 3.0552763819095476,
+      "grad_norm": 0.5680848956108093,
+      "learning_rate": 0.0002,
+      "loss": 1.4275,
+      "step": 4560
+    },
+    {
+      "epoch": 3.0619765494137354,
+      "grad_norm": 0.5906574726104736,
+      "learning_rate": 0.0002,
+      "loss": 1.5187,
+      "step": 4570
+    },
+    {
+      "epoch": 3.0686767169179228,
+      "grad_norm": 0.4725631773471832,
+      "learning_rate": 0.0002,
+      "loss": 1.4551,
+      "step": 4580
+    },
+    {
+      "epoch": 3.0753768844221105,
+      "grad_norm": 0.5273477435112,
+      "learning_rate": 0.0002,
+      "loss": 1.5083,
+      "step": 4590
+    },
+    {
+      "epoch": 3.0820770519262983,
+      "grad_norm": 0.5861203074455261,
+      "learning_rate": 0.0002,
+      "loss": 1.5154,
+      "step": 4600
+    },
+    {
+      "epoch": 3.0887772194304857,
+      "grad_norm": 0.5343965291976929,
+      "learning_rate": 0.0002,
+      "loss": 1.4924,
+      "step": 4610
+    },
+    {
+      "epoch": 3.0954773869346734,
+      "grad_norm": 0.5348150730133057,
+      "learning_rate": 0.0002,
+      "loss": 1.5608,
+      "step": 4620
+    },
+    {
+      "epoch": 3.102177554438861,
+      "grad_norm": 0.5971846580505371,
+      "learning_rate": 0.0002,
+      "loss": 1.5399,
+      "step": 4630
+    },
+    {
+      "epoch": 3.1088777219430486,
+      "grad_norm": 0.5203177332878113,
+      "learning_rate": 0.0002,
+      "loss": 1.4662,
+      "step": 4640
+    },
+    {
+      "epoch": 3.1155778894472363,
+      "grad_norm": 0.55289226770401,
+      "learning_rate": 0.0002,
+      "loss": 1.5805,
+      "step": 4650
+    },
+    {
+      "epoch": 3.1222780569514237,
+      "grad_norm": 0.6878530979156494,
+      "learning_rate": 0.0002,
+      "loss": 1.4745,
+      "step": 4660
+    },
+    {
+      "epoch": 3.1289782244556115,
+      "grad_norm": 0.6173256635665894,
+      "learning_rate": 0.0002,
+      "loss": 1.5335,
+      "step": 4670
+    },
+    {
+      "epoch": 3.135678391959799,
+      "grad_norm": 0.536796510219574,
+      "learning_rate": 0.0002,
+      "loss": 1.51,
+      "step": 4680
+    },
+    {
+      "epoch": 3.1423785594639866,
+      "grad_norm": 0.58846116065979,
+      "learning_rate": 0.0002,
+      "loss": 1.4713,
+      "step": 4690
+    },
+    {
+      "epoch": 3.1490787269681744,
+      "grad_norm": 0.645889401435852,
+      "learning_rate": 0.0002,
+      "loss": 1.5114,
+      "step": 4700
+    },
+    {
+      "epoch": 3.1557788944723617,
+      "grad_norm": 0.6118691563606262,
+      "learning_rate": 0.0002,
+      "loss": 1.4705,
+      "step": 4710
+    },
+    {
+      "epoch": 3.1624790619765495,
+      "grad_norm": 0.5189669132232666,
+      "learning_rate": 0.0002,
+      "loss": 1.5533,
+      "step": 4720
+    },
+    {
+      "epoch": 3.169179229480737,
+      "grad_norm": 0.5794713497161865,
+      "learning_rate": 0.0002,
+      "loss": 1.4769,
+      "step": 4730
+    },
+    {
+      "epoch": 3.1758793969849246,
+      "grad_norm": 0.6579326391220093,
+      "learning_rate": 0.0002,
+      "loss": 1.4849,
+      "step": 4740
+    },
+    {
+      "epoch": 3.1825795644891124,
+      "grad_norm": 0.5822742581367493,
+      "learning_rate": 0.0002,
+      "loss": 1.545,
+      "step": 4750
+    },
+    {
+      "epoch": 3.1892797319932997,
+      "grad_norm": 0.5475956201553345,
+      "learning_rate": 0.0002,
+      "loss": 1.4358,
+      "step": 4760
+    },
+    {
+      "epoch": 3.1959798994974875,
+      "grad_norm": 0.6743834018707275,
+      "learning_rate": 0.0002,
+      "loss": 1.4723,
+      "step": 4770
+    },
+    {
+      "epoch": 3.202680067001675,
+      "grad_norm": 0.6110585927963257,
+      "learning_rate": 0.0002,
+      "loss": 1.5161,
+      "step": 4780
+    },
+    {
+      "epoch": 3.2093802345058626,
+      "grad_norm": 0.5426181554794312,
+      "learning_rate": 0.0002,
+      "loss": 1.5455,
+      "step": 4790
+    },
+    {
+      "epoch": 3.2160804020100504,
+      "grad_norm": 0.6077824234962463,
+      "learning_rate": 0.0002,
+      "loss": 1.5315,
+      "step": 4800
+    },
+    {
+      "epoch": 3.2227805695142377,
+      "grad_norm": 0.5785858631134033,
+      "learning_rate": 0.0002,
+      "loss": 1.5314,
+      "step": 4810
+    },
+    {
+      "epoch": 3.2294807370184255,
+      "grad_norm": 0.6425958275794983,
+      "learning_rate": 0.0002,
+      "loss": 1.4041,
+      "step": 4820
+    },
+    {
+      "epoch": 3.236180904522613,
+      "grad_norm": 0.6607080698013306,
+      "learning_rate": 0.0002,
+      "loss": 1.4751,
+      "step": 4830
+    },
+    {
+      "epoch": 3.2428810720268006,
+      "grad_norm": 0.5385788679122925,
+      "learning_rate": 0.0002,
+      "loss": 1.5267,
+      "step": 4840
+    },
+    {
+      "epoch": 3.2495812395309884,
+      "grad_norm": 0.5630403757095337,
+      "learning_rate": 0.0002,
+      "loss": 1.4673,
+      "step": 4850
+    },
+    {
+      "epoch": 3.2562814070351758,
+      "grad_norm": 0.6340779662132263,
+      "learning_rate": 0.0002,
+      "loss": 1.5257,
+      "step": 4860
+    },
+    {
+      "epoch": 3.2629815745393635,
+      "grad_norm": 0.5305342674255371,
+      "learning_rate": 0.0002,
+      "loss": 1.5148,
+      "step": 4870
+    },
+    {
+      "epoch": 3.2696817420435513,
+      "grad_norm": 0.597670316696167,
+      "learning_rate": 0.0002,
+      "loss": 1.5162,
+      "step": 4880
+    },
+    {
+      "epoch": 3.2763819095477387,
+      "grad_norm": 0.665553867816925,
+      "learning_rate": 0.0002,
+      "loss": 1.5429,
+      "step": 4890
+    },
+    {
+      "epoch": 3.2830820770519265,
+      "grad_norm": 0.579767644405365,
+      "learning_rate": 0.0002,
+      "loss": 1.4607,
+      "step": 4900
+    },
+    {
+      "epoch": 3.289782244556114,
+      "grad_norm": 0.5512481331825256,
+      "learning_rate": 0.0002,
+      "loss": 1.4999,
+      "step": 4910
+    },
+    {
+      "epoch": 3.2964824120603016,
+      "grad_norm": 0.5916532278060913,
+      "learning_rate": 0.0002,
+      "loss": 1.5022,
+      "step": 4920
+    },
+    {
+      "epoch": 3.3031825795644894,
+      "grad_norm": 0.7521726489067078,
+      "learning_rate": 0.0002,
+      "loss": 1.4889,
+      "step": 4930
+    },
+    {
+      "epoch": 3.3098827470686767,
+      "grad_norm": 0.5352797508239746,
+      "learning_rate": 0.0002,
+      "loss": 1.4223,
+      "step": 4940
+    },
+    {
+      "epoch": 3.3165829145728645,
+      "grad_norm": 0.5950371623039246,
+      "learning_rate": 0.0002,
+      "loss": 1.5122,
+      "step": 4950
+    },
+    {
+      "epoch": 3.323283082077052,
+      "grad_norm": 0.8020477890968323,
+      "learning_rate": 0.0002,
+      "loss": 1.5072,
+      "step": 4960
+    },
+    {
+      "epoch": 3.3299832495812396,
+      "grad_norm": 0.6790024638175964,
+      "learning_rate": 0.0002,
+      "loss": 1.5422,
+      "step": 4970
+    },
+    {
+      "epoch": 3.3366834170854274,
+      "grad_norm": 0.687627375125885,
+      "learning_rate": 0.0002,
+      "loss": 1.5363,
+      "step": 4980
+    },
+    {
+      "epoch": 3.3433835845896147,
+      "grad_norm": 0.6094385385513306,
+      "learning_rate": 0.0002,
+      "loss": 1.5276,
+      "step": 4990
+    },
+    {
+      "epoch": 3.3500837520938025,
+      "grad_norm": 0.6541242003440857,
+      "learning_rate": 0.0002,
+      "loss": 1.549,
+      "step": 5000
+    },
+    {
+      "epoch": 3.35678391959799,
+      "grad_norm": 0.5560880303382874,
+      "learning_rate": 0.0002,
+      "loss": 1.6067,
+      "step": 5010
+    },
+    {
+      "epoch": 3.3634840871021776,
+      "grad_norm": 0.5440094470977783,
+      "learning_rate": 0.0002,
+      "loss": 1.5769,
+      "step": 5020
+    },
+    {
+      "epoch": 3.3701842546063654,
+      "grad_norm": 0.5749301314353943,
+      "learning_rate": 0.0002,
+      "loss": 1.6183,
+      "step": 5030
+    },
+    {
+      "epoch": 3.3768844221105527,
+      "grad_norm": 0.5919716954231262,
+      "learning_rate": 0.0002,
+      "loss": 1.4801,
+      "step": 5040
+    },
+    {
+      "epoch": 3.3835845896147405,
+      "grad_norm": 0.6331481337547302,
+      "learning_rate": 0.0002,
+      "loss": 1.5957,
+      "step": 5050
+    },
+    {
+      "epoch": 3.390284757118928,
+      "grad_norm": 0.5687161684036255,
+      "learning_rate": 0.0002,
+      "loss": 1.5188,
+      "step": 5060
+    },
+    {
+      "epoch": 3.3969849246231156,
+      "grad_norm": 0.6718577742576599,
+      "learning_rate": 0.0002,
+      "loss": 1.5702,
+      "step": 5070
+    },
+    {
+      "epoch": 3.4036850921273034,
+      "grad_norm": 0.5089324116706848,
+      "learning_rate": 0.0002,
+      "loss": 1.5577,
+      "step": 5080
+    },
+    {
+      "epoch": 3.4103852596314908,
+      "grad_norm": 0.5710174441337585,
+      "learning_rate": 0.0002,
+      "loss": 1.512,
+      "step": 5090
+    },
+    {
+      "epoch": 3.4170854271356785,
+      "grad_norm": 0.6670721173286438,
+      "learning_rate": 0.0002,
+      "loss": 1.5492,
+      "step": 5100
+    },
+    {
+      "epoch": 3.423785594639866,
+      "grad_norm": 0.6875665187835693,
+      "learning_rate": 0.0002,
+      "loss": 1.5227,
+      "step": 5110
+    },
+    {
+      "epoch": 3.4304857621440537,
+      "grad_norm": 0.5375880599021912,
+      "learning_rate": 0.0002,
+      "loss": 1.4496,
+      "step": 5120
+    },
+    {
+      "epoch": 3.4371859296482414,
+      "grad_norm": 0.6550399661064148,
+      "learning_rate": 0.0002,
+      "loss": 1.5527,
+      "step": 5130
+    },
+    {
+      "epoch": 3.4438860971524288,
+      "grad_norm": 0.5948067903518677,
+      "learning_rate": 0.0002,
+      "loss": 1.5687,
+      "step": 5140
+    },
+    {
+      "epoch": 3.4505862646566166,
+      "grad_norm": 0.6134477257728577,
+      "learning_rate": 0.0002,
+      "loss": 1.4813,
+      "step": 5150
+    },
+    {
+      "epoch": 3.457286432160804,
+      "grad_norm": 0.6506398320198059,
+      "learning_rate": 0.0002,
+      "loss": 1.5069,
+      "step": 5160
+    },
+    {
+      "epoch": 3.4639865996649917,
+      "grad_norm": 0.6060147881507874,
+      "learning_rate": 0.0002,
+      "loss": 1.4422,
+      "step": 5170
+    },
+    {
+      "epoch": 3.4706867671691795,
+      "grad_norm": 0.6173806190490723,
+      "learning_rate": 0.0002,
+      "loss": 1.5093,
+      "step": 5180
+    },
+    {
+      "epoch": 3.477386934673367,
+      "grad_norm": 0.6032607555389404,
+      "learning_rate": 0.0002,
+      "loss": 1.4975,
+      "step": 5190
+    },
+    {
+      "epoch": 3.4840871021775546,
+      "grad_norm": 0.5652492046356201,
+      "learning_rate": 0.0002,
+      "loss": 1.4979,
+      "step": 5200
+    },
+    {
+      "epoch": 3.490787269681742,
+      "grad_norm": 0.6168607473373413,
+      "learning_rate": 0.0002,
+      "loss": 1.4883,
+      "step": 5210
+    },
+    {
+      "epoch": 3.4974874371859297,
+      "grad_norm": 0.6170629262924194,
+      "learning_rate": 0.0002,
+      "loss": 1.5164,
+      "step": 5220
+    },
+    {
+      "epoch": 3.5041876046901175,
+      "grad_norm": 0.6926297545433044,
+      "learning_rate": 0.0002,
+      "loss": 1.4879,
+      "step": 5230
+    },
+    {
+      "epoch": 3.510887772194305,
+      "grad_norm": 0.6702437996864319,
+      "learning_rate": 0.0002,
+      "loss": 1.4982,
+      "step": 5240
+    },
+    {
+      "epoch": 3.5175879396984926,
+      "grad_norm": 0.5421436429023743,
+      "learning_rate": 0.0002,
+      "loss": 1.4986,
+      "step": 5250
+    },
+    {
+      "epoch": 3.52428810720268,
+      "grad_norm": 0.5726765990257263,
+      "learning_rate": 0.0002,
+      "loss": 1.4673,
+      "step": 5260
+    },
+    {
+      "epoch": 3.5309882747068677,
+      "grad_norm": 0.5685455203056335,
+      "learning_rate": 0.0002,
+      "loss": 1.5423,
+      "step": 5270
+    },
+    {
+      "epoch": 3.5376884422110555,
+      "grad_norm": 0.6018396019935608,
+      "learning_rate": 0.0002,
+      "loss": 1.4715,
+      "step": 5280
+    },
+    {
+      "epoch": 3.544388609715243,
+      "grad_norm": 0.5731932520866394,
+      "learning_rate": 0.0002,
+      "loss": 1.5451,
+      "step": 5290
+    },
+    {
+      "epoch": 3.5510887772194306,
+      "grad_norm": 0.6601519584655762,
+      "learning_rate": 0.0002,
+      "loss": 1.4752,
+      "step": 5300
+    },
+    {
+      "epoch": 3.557788944723618,
+      "grad_norm": 0.5545530319213867,
+      "learning_rate": 0.0002,
+      "loss": 1.5434,
+      "step": 5310
+    },
+    {
+      "epoch": 3.5644891122278057,
+      "grad_norm": 0.5998541116714478,
+      "learning_rate": 0.0002,
+      "loss": 1.5438,
+      "step": 5320
+    },
+    {
+      "epoch": 3.5711892797319935,
+      "grad_norm": 0.5651767253875732,
+      "learning_rate": 0.0002,
+      "loss": 1.56,
+      "step": 5330
+    },
+    {
+      "epoch": 3.577889447236181,
+      "grad_norm": 0.7425084114074707,
+      "learning_rate": 0.0002,
+      "loss": 1.4829,
+      "step": 5340
+    },
+    {
+      "epoch": 3.5845896147403686,
+      "grad_norm": 0.5770602226257324,
+      "learning_rate": 0.0002,
+      "loss": 1.5571,
+      "step": 5350
+    },
+    {
+      "epoch": 3.591289782244556,
+      "grad_norm": 0.54723060131073,
+      "learning_rate": 0.0002,
+      "loss": 1.458,
+      "step": 5360
+    },
+    {
+      "epoch": 3.5979899497487438,
+      "grad_norm": 0.6658238172531128,
+      "learning_rate": 0.0002,
+      "loss": 1.497,
+      "step": 5370
+    },
+    {
+      "epoch": 3.6046901172529315,
+      "grad_norm": 0.5787645578384399,
+      "learning_rate": 0.0002,
+      "loss": 1.5456,
+      "step": 5380
+    },
+    {
+      "epoch": 3.611390284757119,
+      "grad_norm": 0.594913125038147,
+      "learning_rate": 0.0002,
+      "loss": 1.5343,
+      "step": 5390
+    },
+    {
+      "epoch": 3.6180904522613067,
+      "grad_norm": 0.4964977502822876,
+      "learning_rate": 0.0002,
+      "loss": 1.4727,
+      "step": 5400
+    },
+    {
+      "epoch": 3.624790619765494,
+      "grad_norm": 0.6087527275085449,
+      "learning_rate": 0.0002,
+      "loss": 1.5062,
+      "step": 5410
+    },
+    {
+      "epoch": 3.6314907872696818,
+      "grad_norm": 0.6315323710441589,
+      "learning_rate": 0.0002,
+      "loss": 1.5098,
+      "step": 5420
+    },
+    {
+      "epoch": 3.6381909547738696,
+      "grad_norm": 0.574799120426178,
+      "learning_rate": 0.0002,
+      "loss": 1.4855,
+      "step": 5430
+    },
+    {
+      "epoch": 3.644891122278057,
+      "grad_norm": 0.5949277877807617,
+      "learning_rate": 0.0002,
+      "loss": 1.4595,
+      "step": 5440
+    },
+    {
+      "epoch": 3.6515912897822447,
+      "grad_norm": 0.5640677213668823,
+      "learning_rate": 0.0002,
+      "loss": 1.4816,
+      "step": 5450
+    },
+    {
+      "epoch": 3.658291457286432,
+      "grad_norm": 0.6198237538337708,
+      "learning_rate": 0.0002,
+      "loss": 1.525,
+      "step": 5460
+    },
+    {
+      "epoch": 3.66499162479062,
+      "grad_norm": 0.6902034878730774,
+      "learning_rate": 0.0002,
+      "loss": 1.5676,
+      "step": 5470
+    },
+    {
+      "epoch": 3.6716917922948076,
+      "grad_norm": 0.5686674118041992,
+      "learning_rate": 0.0002,
+      "loss": 1.544,
+      "step": 5480
+    },
+    {
+      "epoch": 3.678391959798995,
+      "grad_norm": 0.6532107591629028,
+      "learning_rate": 0.0002,
+      "loss": 1.5255,
+      "step": 5490
+    },
+    {
+      "epoch": 3.6850921273031827,
+      "grad_norm": 0.5790849924087524,
+      "learning_rate": 0.0002,
+      "loss": 1.5767,
+      "step": 5500
+    },
+    {
+      "epoch": 3.69179229480737,
+      "grad_norm": 0.6055065393447876,
+      "learning_rate": 0.0002,
+      "loss": 1.5507,
+      "step": 5510
+    },
+    {
+      "epoch": 3.698492462311558,
+      "grad_norm": 0.5630605816841125,
+      "learning_rate": 0.0002,
+      "loss": 1.4656,
+      "step": 5520
+    },
+    {
+      "epoch": 3.7051926298157456,
+      "grad_norm": 0.6005825996398926,
+      "learning_rate": 0.0002,
+      "loss": 1.537,
+      "step": 5530
+    },
+    {
+      "epoch": 3.711892797319933,
+      "grad_norm": 0.6553038954734802,
+      "learning_rate": 0.0002,
+      "loss": 1.5313,
+      "step": 5540
+    },
+    {
+      "epoch": 3.7185929648241207,
+      "grad_norm": 0.5601094961166382,
+      "learning_rate": 0.0002,
+      "loss": 1.4943,
+      "step": 5550
+    },
+    {
+      "epoch": 3.725293132328308,
+      "grad_norm": 0.6598808169364929,
+      "learning_rate": 0.0002,
+      "loss": 1.5161,
+      "step": 5560
+    },
+    {
+      "epoch": 3.731993299832496,
+      "grad_norm": 0.5506255626678467,
+      "learning_rate": 0.0002,
+      "loss": 1.5345,
+      "step": 5570
+    },
+    {
+      "epoch": 3.7386934673366836,
+      "grad_norm": 0.6001223921775818,
+      "learning_rate": 0.0002,
+      "loss": 1.4805,
+      "step": 5580
+    },
+    {
+      "epoch": 3.745393634840871,
+      "grad_norm": 0.6287297606468201,
+      "learning_rate": 0.0002,
+      "loss": 1.4652,
+      "step": 5590
+    },
+    {
+      "epoch": 3.7520938023450587,
+      "grad_norm": 0.6253238916397095,
+      "learning_rate": 0.0002,
+      "loss": 1.5246,
+      "step": 5600
+    },
+    {
+      "epoch": 3.758793969849246,
+      "grad_norm": 0.5713174939155579,
+      "learning_rate": 0.0002,
+      "loss": 1.5691,
+      "step": 5610
+    },
+    {
+      "epoch": 3.765494137353434,
+      "grad_norm": 0.6198310852050781,
+      "learning_rate": 0.0002,
+      "loss": 1.5661,
+      "step": 5620
+    },
+    {
+      "epoch": 3.7721943048576216,
+      "grad_norm": 0.5941224098205566,
+      "learning_rate": 0.0002,
+      "loss": 1.5448,
+      "step": 5630
+    },
+    {
+      "epoch": 3.778894472361809,
+      "grad_norm": 0.606002151966095,
+      "learning_rate": 0.0002,
+      "loss": 1.4925,
+      "step": 5640
+    },
+    {
+      "epoch": 3.7855946398659968,
+      "grad_norm": 0.6540704965591431,
+      "learning_rate": 0.0002,
+      "loss": 1.5182,
+      "step": 5650
+    },
+    {
+      "epoch": 3.792294807370184,
+      "grad_norm": 0.6147415041923523,
+      "learning_rate": 0.0002,
+      "loss": 1.5903,
+      "step": 5660
+    },
+    {
+      "epoch": 3.798994974874372,
+      "grad_norm": 0.5649605393409729,
+      "learning_rate": 0.0002,
+      "loss": 1.5329,
+      "step": 5670
+    },
+    {
+      "epoch": 3.8056951423785597,
+      "grad_norm": 0.6788773536682129,
+      "learning_rate": 0.0002,
+      "loss": 1.5747,
+      "step": 5680
+    },
+    {
+      "epoch": 3.812395309882747,
+      "grad_norm": 0.6581860780715942,
+      "learning_rate": 0.0002,
+      "loss": 1.535,
+      "step": 5690
+    },
+    {
+      "epoch": 3.819095477386935,
+      "grad_norm": 0.5529348850250244,
+      "learning_rate": 0.0002,
+      "loss": 1.4587,
+      "step": 5700
+    },
+    {
+      "epoch": 3.825795644891122,
+      "grad_norm": 0.6320232152938843,
+      "learning_rate": 0.0002,
+      "loss": 1.5594,
+      "step": 5710
+    },
+    {
+      "epoch": 3.83249581239531,
+      "grad_norm": 0.6529698371887207,
+      "learning_rate": 0.0002,
+      "loss": 1.4696,
+      "step": 5720
+    },
+    {
+      "epoch": 3.8391959798994977,
+      "grad_norm": 0.5983362793922424,
+      "learning_rate": 0.0002,
+      "loss": 1.5854,
+      "step": 5730
+    },
+    {
+      "epoch": 3.845896147403685,
+      "grad_norm": 0.6335684061050415,
+      "learning_rate": 0.0002,
+      "loss": 1.465,
+      "step": 5740
+    },
+    {
+      "epoch": 3.852596314907873,
+      "grad_norm": 0.700446605682373,
+      "learning_rate": 0.0002,
+      "loss": 1.5545,
+      "step": 5750
+    },
+    {
+      "epoch": 3.85929648241206,
+      "grad_norm": 0.6092597842216492,
+      "learning_rate": 0.0002,
+      "loss": 1.5707,
+      "step": 5760
+    },
+    {
+      "epoch": 3.865996649916248,
+      "grad_norm": 0.564146101474762,
+      "learning_rate": 0.0002,
+      "loss": 1.5729,
+      "step": 5770
+    },
+    {
+      "epoch": 3.8726968174204357,
+      "grad_norm": 0.615275502204895,
+      "learning_rate": 0.0002,
+      "loss": 1.5872,
+      "step": 5780
+    },
+    {
+      "epoch": 3.879396984924623,
+      "grad_norm": 0.6685376763343811,
+      "learning_rate": 0.0002,
+      "loss": 1.5142,
+      "step": 5790
+    },
+    {
+      "epoch": 3.886097152428811,
+      "grad_norm": 0.6116922497749329,
+      "learning_rate": 0.0002,
+      "loss": 1.4752,
+      "step": 5800
+    },
+    {
+      "epoch": 3.892797319932998,
+      "grad_norm": 0.5486813187599182,
+      "learning_rate": 0.0002,
+      "loss": 1.5179,
+      "step": 5810
+    },
+    {
+      "epoch": 3.899497487437186,
+      "grad_norm": 0.6208204030990601,
+      "learning_rate": 0.0002,
+      "loss": 1.5167,
+      "step": 5820
+    },
+    {
+      "epoch": 3.9061976549413737,
+      "grad_norm": 0.6500625014305115,
+      "learning_rate": 0.0002,
+      "loss": 1.5334,
+      "step": 5830
+    },
+    {
+      "epoch": 3.912897822445561,
+      "grad_norm": 0.5948089361190796,
+      "learning_rate": 0.0002,
+      "loss": 1.4716,
+      "step": 5840
+    },
+    {
+      "epoch": 3.919597989949749,
+      "grad_norm": 0.7210732698440552,
+      "learning_rate": 0.0002,
+      "loss": 1.6011,
+      "step": 5850
+    },
+    {
+      "epoch": 3.926298157453936,
+      "grad_norm": 0.6662322878837585,
+      "learning_rate": 0.0002,
+      "loss": 1.5519,
+      "step": 5860
+    },
+    {
+      "epoch": 3.932998324958124,
+      "grad_norm": 0.5613839626312256,
+      "learning_rate": 0.0002,
+      "loss": 1.5656,
+      "step": 5870
+    },
+    {
+      "epoch": 3.9396984924623117,
+      "grad_norm": 0.6069002151489258,
+      "learning_rate": 0.0002,
+      "loss": 1.544,
+      "step": 5880
+    },
+    {
+      "epoch": 3.946398659966499,
+      "grad_norm": 0.7075562477111816,
+      "learning_rate": 0.0002,
+      "loss": 1.6745,
+      "step": 5890
+    },
+    {
+      "epoch": 3.953098827470687,
+      "grad_norm": 0.6316173076629639,
+      "learning_rate": 0.0002,
+      "loss": 1.5391,
+      "step": 5900
+    },
+    {
+      "epoch": 3.959798994974874,
+      "grad_norm": 0.5716308355331421,
+      "learning_rate": 0.0002,
+      "loss": 1.6314,
+      "step": 5910
+    },
+    {
+      "epoch": 3.966499162479062,
+      "grad_norm": 0.6800096035003662,
+      "learning_rate": 0.0002,
+      "loss": 1.5947,
+      "step": 5920
+    },
+    {
+      "epoch": 3.9731993299832498,
+      "grad_norm": 0.6057983040809631,
+      "learning_rate": 0.0002,
+      "loss": 1.5189,
+      "step": 5930
+    },
+    {
+      "epoch": 3.979899497487437,
+      "grad_norm": 0.5938987731933594,
+      "learning_rate": 0.0002,
+      "loss": 1.5431,
+      "step": 5940
+    },
+    {
+      "epoch": 3.986599664991625,
+      "grad_norm": 0.6963576674461365,
+      "learning_rate": 0.0002,
+      "loss": 1.5111,
+      "step": 5950
+    },
+    {
+      "epoch": 3.993299832495812,
+      "grad_norm": 0.6279940009117126,
+      "learning_rate": 0.0002,
+      "loss": 1.5521,
+      "step": 5960
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 0.7161159515380859,
+      "learning_rate": 0.0002,
+      "loss": 1.5974,
+      "step": 5970
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 1.8655421733856201,
+      "eval_runtime": 37.9276,
+      "eval_samples_per_second": 13.579,
+      "eval_steps_per_second": 1.714,
+      "step": 5970
+    },
+    {
+      "epoch": 4.006700167504188,
+      "grad_norm": 0.7380476593971252,
+      "learning_rate": 0.0002,
+      "loss": 1.3666,
+      "step": 5980
+    },
+    {
+      "epoch": 4.013400335008376,
+      "grad_norm": 0.7148947715759277,
+      "learning_rate": 0.0002,
+      "loss": 1.3913,
+      "step": 5990
+    },
+    {
+      "epoch": 4.0201005025125625,
+      "grad_norm": 0.6177082657814026,
+      "learning_rate": 0.0002,
+      "loss": 1.4204,
+      "step": 6000
+    },
+    {
+      "epoch": 4.02680067001675,
+      "grad_norm": 0.8552946448326111,
+      "learning_rate": 0.0002,
+      "loss": 1.4421,
+      "step": 6010
+    },
+    {
+      "epoch": 4.033500837520938,
+      "grad_norm": 0.8033416271209717,
+      "learning_rate": 0.0002,
+      "loss": 1.4342,
+      "step": 6020
+    },
+    {
+      "epoch": 4.040201005025126,
+      "grad_norm": 0.8501318097114563,
+      "learning_rate": 0.0002,
+      "loss": 1.4092,
+      "step": 6030
+    },
+    {
+      "epoch": 4.046901172529314,
+      "grad_norm": 0.6981393098831177,
+      "learning_rate": 0.0002,
+      "loss": 1.3367,
+      "step": 6040
+    },
+    {
+      "epoch": 4.0536013400335005,
+      "grad_norm": 0.7227180600166321,
+      "learning_rate": 0.0002,
+      "loss": 1.3925,
+      "step": 6050
+    },
+    {
+      "epoch": 4.060301507537688,
+      "grad_norm": 0.6923989653587341,
+      "learning_rate": 0.0002,
+      "loss": 1.4007,
+      "step": 6060
+    },
+    {
+      "epoch": 4.067001675041876,
+      "grad_norm": 0.879779040813446,
+      "learning_rate": 0.0002,
+      "loss": 1.3837,
+      "step": 6070
+    },
+    {
+      "epoch": 4.073701842546064,
+      "grad_norm": 0.8184754848480225,
+      "learning_rate": 0.0002,
+      "loss": 1.4383,
+      "step": 6080
+    },
+    {
+      "epoch": 4.080402010050252,
+      "grad_norm": 0.8211342692375183,
+      "learning_rate": 0.0002,
+      "loss": 1.3128,
+      "step": 6090
+    },
+    {
+      "epoch": 4.0871021775544385,
+      "grad_norm": 0.7542396783828735,
+      "learning_rate": 0.0002,
+      "loss": 1.3892,
+      "step": 6100
+    },
+    {
+      "epoch": 4.093802345058626,
+      "grad_norm": 0.6631066799163818,
+      "learning_rate": 0.0002,
+      "loss": 1.3607,
+      "step": 6110
+    },
+    {
+      "epoch": 4.100502512562814,
+      "grad_norm": 0.6728386282920837,
+      "learning_rate": 0.0002,
+      "loss": 1.3275,
+      "step": 6120
+    },
+    {
+      "epoch": 4.107202680067002,
+      "grad_norm": 0.681851863861084,
+      "learning_rate": 0.0002,
+      "loss": 1.3443,
+      "step": 6130
+    },
+    {
+      "epoch": 4.11390284757119,
+      "grad_norm": 0.8757794499397278,
+      "learning_rate": 0.0002,
+      "loss": 1.3486,
+      "step": 6140
+    },
+    {
+      "epoch": 4.1206030150753765,
+      "grad_norm": 0.6567301750183105,
+      "learning_rate": 0.0002,
+      "loss": 1.351,
+      "step": 6150
+    },
+    {
+      "epoch": 4.127303182579564,
+      "grad_norm": 0.7950329184532166,
+      "learning_rate": 0.0002,
+      "loss": 1.3824,
+      "step": 6160
+    },
+    {
+      "epoch": 4.134003350083752,
+      "grad_norm": 0.7545644044876099,
+      "learning_rate": 0.0002,
+      "loss": 1.3738,
+      "step": 6170
+    },
+    {
+      "epoch": 4.14070351758794,
+      "grad_norm": 0.7172710299491882,
+      "learning_rate": 0.0002,
+      "loss": 1.4214,
+      "step": 6180
+    },
+    {
+      "epoch": 4.147403685092128,
+      "grad_norm": 0.7040584087371826,
+      "learning_rate": 0.0002,
+      "loss": 1.4091,
+      "step": 6190
+    },
+    {
+      "epoch": 4.1541038525963145,
+      "grad_norm": 0.7482913732528687,
+      "learning_rate": 0.0002,
+      "loss": 1.4149,
+      "step": 6200
+    },
+    {
+      "epoch": 4.160804020100502,
+      "grad_norm": 0.8523276448249817,
+      "learning_rate": 0.0002,
+      "loss": 1.3227,
+      "step": 6210
+    },
+    {
+      "epoch": 4.16750418760469,
+      "grad_norm": 0.6672041416168213,
+      "learning_rate": 0.0002,
+      "loss": 1.4194,
+      "step": 6220
+    },
+    {
+      "epoch": 4.174204355108878,
+      "grad_norm": 0.7523500919342041,
+      "learning_rate": 0.0002,
+      "loss": 1.3953,
+      "step": 6230
+    },
+    {
+      "epoch": 4.180904522613066,
+      "grad_norm": 0.8085253834724426,
+      "learning_rate": 0.0002,
+      "loss": 1.371,
+      "step": 6240
+    },
+    {
+      "epoch": 4.187604690117253,
+      "grad_norm": 0.789450466632843,
+      "learning_rate": 0.0002,
+      "loss": 1.3293,
+      "step": 6250
+    },
+    {
+      "epoch": 4.19430485762144,
+      "grad_norm": 0.7502310872077942,
+      "learning_rate": 0.0002,
+      "loss": 1.3539,
+      "step": 6260
+    },
+    {
+      "epoch": 4.201005025125628,
+      "grad_norm": 0.7397456765174866,
+      "learning_rate": 0.0002,
+      "loss": 1.3415,
+      "step": 6270
+    },
+    {
+      "epoch": 4.207705192629816,
+      "grad_norm": 0.6921947002410889,
+      "learning_rate": 0.0002,
+      "loss": 1.3963,
+      "step": 6280
+    },
+    {
+      "epoch": 4.214405360134004,
+      "grad_norm": 0.9334571957588196,
+      "learning_rate": 0.0002,
+      "loss": 1.3125,
+      "step": 6290
+    },
+    {
+      "epoch": 4.221105527638191,
+      "grad_norm": 0.725799024105072,
+      "learning_rate": 0.0002,
+      "loss": 1.3612,
+      "step": 6300
+    },
+    {
+      "epoch": 4.227805695142378,
+      "grad_norm": 0.8290495872497559,
+      "learning_rate": 0.0002,
+      "loss": 1.4217,
+      "step": 6310
+    },
+    {
+      "epoch": 4.234505862646566,
+      "grad_norm": 0.688983678817749,
+      "learning_rate": 0.0002,
+      "loss": 1.4135,
+      "step": 6320
+    },
+    {
+      "epoch": 4.241206030150754,
+      "grad_norm": 0.8620913028717041,
+      "learning_rate": 0.0002,
+      "loss": 1.3807,
+      "step": 6330
+    },
+    {
+      "epoch": 4.247906197654942,
+      "grad_norm": 0.8008657693862915,
+      "learning_rate": 0.0002,
+      "loss": 1.3738,
+      "step": 6340
+    },
+    {
+      "epoch": 4.254606365159129,
+      "grad_norm": 0.7379199266433716,
+      "learning_rate": 0.0002,
+      "loss": 1.4005,
+      "step": 6350
+    },
+    {
+      "epoch": 4.261306532663316,
+      "grad_norm": 0.7842815518379211,
+      "learning_rate": 0.0002,
+      "loss": 1.426,
+      "step": 6360
+    },
+    {
+      "epoch": 4.268006700167504,
+      "grad_norm": 0.812600314617157,
+      "learning_rate": 0.0002,
+      "loss": 1.4262,
+      "step": 6370
+    },
+    {
+      "epoch": 4.274706867671692,
+      "grad_norm": 0.7852841019630432,
+      "learning_rate": 0.0002,
+      "loss": 1.4028,
+      "step": 6380
+    },
+    {
+      "epoch": 4.28140703517588,
+      "grad_norm": 1.0377534627914429,
+      "learning_rate": 0.0002,
+      "loss": 1.3722,
+      "step": 6390
+    },
+    {
+      "epoch": 4.288107202680067,
+      "grad_norm": 1.03935706615448,
+      "learning_rate": 0.0002,
+      "loss": 1.3755,
+      "step": 6400
+    },
+    {
+      "epoch": 4.294807370184254,
+      "grad_norm": 0.7244732975959778,
+      "learning_rate": 0.0002,
+      "loss": 1.3961,
+      "step": 6410
+    },
+    {
+      "epoch": 4.301507537688442,
+      "grad_norm": 0.7137406468391418,
+      "learning_rate": 0.0002,
+      "loss": 1.4608,
+      "step": 6420
+    },
+    {
+      "epoch": 4.30820770519263,
+      "grad_norm": 0.7492543458938599,
+      "learning_rate": 0.0002,
+      "loss": 1.4461,
+      "step": 6430
+    },
+    {
+      "epoch": 4.314907872696818,
+      "grad_norm": 0.7065439224243164,
+      "learning_rate": 0.0002,
+      "loss": 1.4562,
+      "step": 6440
+    },
+    {
+      "epoch": 4.321608040201005,
+      "grad_norm": 0.7786989808082581,
+      "learning_rate": 0.0002,
+      "loss": 1.4246,
+      "step": 6450
+    },
+    {
+      "epoch": 4.328308207705192,
+      "grad_norm": 0.7369208335876465,
+      "learning_rate": 0.0002,
+      "loss": 1.3098,
+      "step": 6460
+    },
+    {
+      "epoch": 4.33500837520938,
+      "grad_norm": 0.7412346005439758,
+      "learning_rate": 0.0002,
+      "loss": 1.3686,
+      "step": 6470
+    },
+    {
+      "epoch": 4.341708542713568,
+      "grad_norm": 0.780927300453186,
+      "learning_rate": 0.0002,
+      "loss": 1.4087,
+      "step": 6480
+    },
+    {
+      "epoch": 4.348408710217756,
+      "grad_norm": 0.8320930600166321,
+      "learning_rate": 0.0002,
+      "loss": 1.3628,
+      "step": 6490
+    },
+    {
+      "epoch": 4.355108877721943,
+      "grad_norm": 0.6871094703674316,
+      "learning_rate": 0.0002,
+      "loss": 1.3715,
+      "step": 6500
+    },
+    {
+      "epoch": 4.36180904522613,
+      "grad_norm": 0.6751559972763062,
+      "learning_rate": 0.0002,
+      "loss": 1.3257,
+      "step": 6510
+    },
+    {
+      "epoch": 4.368509212730318,
+      "grad_norm": 0.7723976969718933,
+      "learning_rate": 0.0002,
+      "loss": 1.4311,
+      "step": 6520
+    },
+    {
+      "epoch": 4.375209380234506,
+      "grad_norm": 0.7915401458740234,
+      "learning_rate": 0.0002,
+      "loss": 1.4086,
+      "step": 6530
+    },
+    {
+      "epoch": 4.381909547738694,
+      "grad_norm": 0.7329102754592896,
+      "learning_rate": 0.0002,
+      "loss": 1.3973,
+      "step": 6540
+    },
+    {
+      "epoch": 4.388609715242881,
+      "grad_norm": 0.7388760447502136,
+      "learning_rate": 0.0002,
+      "loss": 1.447,
+      "step": 6550
+    },
+    {
+      "epoch": 4.3953098827470685,
+      "grad_norm": 0.8282579183578491,
+      "learning_rate": 0.0002,
+      "loss": 1.4378,
+      "step": 6560
+    },
+    {
+      "epoch": 4.402010050251256,
+      "grad_norm": 0.7192724347114563,
+      "learning_rate": 0.0002,
+      "loss": 1.3923,
+      "step": 6570
+    },
+    {
+      "epoch": 4.408710217755444,
+      "grad_norm": 0.746526837348938,
+      "learning_rate": 0.0002,
+      "loss": 1.4141,
+      "step": 6580
+    },
+    {
+      "epoch": 4.415410385259632,
+      "grad_norm": 0.8738046288490295,
+      "learning_rate": 0.0002,
+      "loss": 1.33,
+      "step": 6590
+    },
+    {
+      "epoch": 4.422110552763819,
+      "grad_norm": 0.8408458828926086,
+      "learning_rate": 0.0002,
+      "loss": 1.3995,
+      "step": 6600
+    },
+    {
+      "epoch": 4.4288107202680065,
+      "grad_norm": 0.8110666275024414,
+      "learning_rate": 0.0002,
+      "loss": 1.4148,
+      "step": 6610
+    },
+    {
+      "epoch": 4.435510887772194,
+      "grad_norm": 0.8602406978607178,
+      "learning_rate": 0.0002,
+      "loss": 1.441,
+      "step": 6620
+    },
+    {
+      "epoch": 4.442211055276382,
+      "grad_norm": 0.7549102902412415,
+      "learning_rate": 0.0002,
+      "loss": 1.4319,
+      "step": 6630
+    },
+    {
+      "epoch": 4.44891122278057,
+      "grad_norm": 0.7831804156303406,
+      "learning_rate": 0.0002,
+      "loss": 1.388,
+      "step": 6640
+    },
+    {
+      "epoch": 4.455611390284757,
+      "grad_norm": 0.7269673943519592,
+      "learning_rate": 0.0002,
+      "loss": 1.4283,
+      "step": 6650
+    },
+    {
+      "epoch": 4.4623115577889445,
+      "grad_norm": 0.7397838830947876,
+      "learning_rate": 0.0002,
+      "loss": 1.4132,
+      "step": 6660
+    },
+    {
+      "epoch": 4.469011725293132,
+      "grad_norm": 0.713707447052002,
+      "learning_rate": 0.0002,
+      "loss": 1.3174,
+      "step": 6670
+    },
+    {
+      "epoch": 4.47571189279732,
+      "grad_norm": 0.7525581121444702,
+      "learning_rate": 0.0002,
+      "loss": 1.3406,
+      "step": 6680
+    },
+    {
+      "epoch": 4.482412060301508,
+      "grad_norm": 0.8030191659927368,
+      "learning_rate": 0.0002,
+      "loss": 1.4283,
+      "step": 6690
+    },
+    {
+      "epoch": 4.489112227805695,
+      "grad_norm": 0.7469439506530762,
+      "learning_rate": 0.0002,
+      "loss": 1.4586,
+      "step": 6700
+    },
+    {
+      "epoch": 4.4958123953098825,
+      "grad_norm": 0.7743868231773376,
+      "learning_rate": 0.0002,
+      "loss": 1.367,
+      "step": 6710
+    },
+    {
+      "epoch": 4.50251256281407,
+      "grad_norm": 0.6539737582206726,
+      "learning_rate": 0.0002,
+      "loss": 1.3439,
+      "step": 6720
+    },
+    {
+      "epoch": 4.509212730318258,
+      "grad_norm": 0.825818657875061,
+      "learning_rate": 0.0002,
+      "loss": 1.4513,
+      "step": 6730
+    },
+    {
+      "epoch": 4.515912897822446,
+      "grad_norm": 0.8048575520515442,
+      "learning_rate": 0.0002,
+      "loss": 1.3984,
+      "step": 6740
+    },
+    {
+      "epoch": 4.522613065326633,
+      "grad_norm": 0.7828766107559204,
+      "learning_rate": 0.0002,
+      "loss": 1.3923,
+      "step": 6750
+    },
+    {
+      "epoch": 4.5293132328308205,
+      "grad_norm": 0.7406010031700134,
+      "learning_rate": 0.0002,
+      "loss": 1.3886,
+      "step": 6760
+    },
+    {
+      "epoch": 4.536013400335008,
+      "grad_norm": 0.840345561504364,
+      "learning_rate": 0.0002,
+      "loss": 1.3109,
+      "step": 6770
+    },
+    {
+      "epoch": 4.542713567839196,
+      "grad_norm": 0.8492622971534729,
+      "learning_rate": 0.0002,
+      "loss": 1.4808,
+      "step": 6780
+    },
+    {
+      "epoch": 4.549413735343384,
+      "grad_norm": 0.7130163908004761,
+      "learning_rate": 0.0002,
+      "loss": 1.4384,
+      "step": 6790
+    },
+    {
+      "epoch": 4.556113902847571,
+      "grad_norm": 0.8454728126525879,
+      "learning_rate": 0.0002,
+      "loss": 1.4531,
+      "step": 6800
+    },
+    {
+      "epoch": 4.562814070351759,
+      "grad_norm": 0.7847645282745361,
+      "learning_rate": 0.0002,
+      "loss": 1.3239,
+      "step": 6810
+    },
+    {
+      "epoch": 4.569514237855946,
+      "grad_norm": 0.7245864272117615,
+      "learning_rate": 0.0002,
+      "loss": 1.4181,
+      "step": 6820
+    },
+    {
+      "epoch": 4.576214405360134,
+      "grad_norm": 0.768893301486969,
+      "learning_rate": 0.0002,
+      "loss": 1.3233,
+      "step": 6830
+    },
+    {
+      "epoch": 4.582914572864322,
+      "grad_norm": 0.8028400540351868,
+      "learning_rate": 0.0002,
+      "loss": 1.3932,
+      "step": 6840
+    },
+    {
+      "epoch": 4.589614740368509,
+      "grad_norm": 0.763945460319519,
+      "learning_rate": 0.0002,
+      "loss": 1.3745,
+      "step": 6850
+    },
+    {
+      "epoch": 4.596314907872697,
+      "grad_norm": 0.7417685389518738,
+      "learning_rate": 0.0002,
+      "loss": 1.4797,
+      "step": 6860
+    },
+    {
+      "epoch": 4.603015075376884,
+      "grad_norm": 0.7603038549423218,
+      "learning_rate": 0.0002,
+      "loss": 1.4468,
+      "step": 6870
+    },
+    {
+      "epoch": 4.609715242881072,
+      "grad_norm": 0.7981528043746948,
+      "learning_rate": 0.0002,
+      "loss": 1.4095,
+      "step": 6880
+    },
+    {
+      "epoch": 4.61641541038526,
+      "grad_norm": 0.8077111840248108,
+      "learning_rate": 0.0002,
+      "loss": 1.3963,
+      "step": 6890
+    },
+    {
+      "epoch": 4.623115577889447,
+      "grad_norm": 0.8778454065322876,
+      "learning_rate": 0.0002,
+      "loss": 1.4721,
+      "step": 6900
+    },
+    {
+      "epoch": 4.629815745393635,
+      "grad_norm": 0.8620710372924805,
+      "learning_rate": 0.0002,
+      "loss": 1.3758,
+      "step": 6910
+    },
+    {
+      "epoch": 4.636515912897822,
+      "grad_norm": 0.7486072778701782,
+      "learning_rate": 0.0002,
+      "loss": 1.344,
+      "step": 6920
+    },
+    {
+      "epoch": 4.64321608040201,
+      "grad_norm": 0.7493042945861816,
+      "learning_rate": 0.0002,
+      "loss": 1.3913,
+      "step": 6930
+    },
+    {
+      "epoch": 4.649916247906198,
+      "grad_norm": 0.7388978600502014,
+      "learning_rate": 0.0002,
+      "loss": 1.397,
+      "step": 6940
+    },
+    {
+      "epoch": 4.656616415410385,
+      "grad_norm": 0.798530638217926,
+      "learning_rate": 0.0002,
+      "loss": 1.3593,
+      "step": 6950
+    },
+    {
+      "epoch": 4.663316582914573,
+      "grad_norm": 0.7929500937461853,
+      "learning_rate": 0.0002,
+      "loss": 1.3982,
+      "step": 6960
+    },
+    {
+      "epoch": 4.67001675041876,
+      "grad_norm": 0.9186785221099854,
+      "learning_rate": 0.0002,
+      "loss": 1.4183,
+      "step": 6970
+    },
+    {
+      "epoch": 4.676716917922948,
+      "grad_norm": 1.1103485822677612,
+      "learning_rate": 0.0002,
+      "loss": 1.3955,
+      "step": 6980
+    },
+    {
+      "epoch": 4.683417085427136,
+      "grad_norm": 0.8000466823577881,
+      "learning_rate": 0.0002,
+      "loss": 1.3941,
+      "step": 6990
+    },
+    {
+      "epoch": 4.690117252931323,
+      "grad_norm": 0.7520599961280823,
+      "learning_rate": 0.0002,
+      "loss": 1.371,
+      "step": 7000
+    },
+    {
+      "epoch": 4.696817420435511,
+      "grad_norm": 0.7971973419189453,
+      "learning_rate": 0.0002,
+      "loss": 1.4582,
+      "step": 7010
+    },
+    {
+      "epoch": 4.703517587939698,
+      "grad_norm": 0.7363343834877014,
+      "learning_rate": 0.0002,
+      "loss": 1.3682,
+      "step": 7020
+    },
+    {
+      "epoch": 4.710217755443886,
+      "grad_norm": 0.8268865942955017,
+      "learning_rate": 0.0002,
+      "loss": 1.3889,
+      "step": 7030
+    },
+    {
+      "epoch": 4.716917922948074,
+      "grad_norm": 0.7054963111877441,
+      "learning_rate": 0.0002,
+      "loss": 1.4382,
+      "step": 7040
+    },
+    {
+      "epoch": 4.723618090452261,
+      "grad_norm": 0.8196262121200562,
+      "learning_rate": 0.0002,
+      "loss": 1.4578,
+      "step": 7050
+    },
+    {
+      "epoch": 4.730318257956449,
+      "grad_norm": 0.8276031017303467,
+      "learning_rate": 0.0002,
+      "loss": 1.365,
+      "step": 7060
+    },
+    {
+      "epoch": 4.7370184254606365,
+      "grad_norm": 0.8248157501220703,
+      "learning_rate": 0.0002,
+      "loss": 1.3887,
+      "step": 7070
+    },
+    {
+      "epoch": 4.743718592964824,
+      "grad_norm": 0.8937979936599731,
+      "learning_rate": 0.0002,
+      "loss": 1.4193,
+      "step": 7080
+    },
+    {
+      "epoch": 4.750418760469012,
+      "grad_norm": 0.7984238862991333,
+      "learning_rate": 0.0002,
+      "loss": 1.4334,
+      "step": 7090
+    },
+    {
+      "epoch": 4.757118927973199,
+      "grad_norm": 0.9495313763618469,
+      "learning_rate": 0.0002,
+      "loss": 1.4385,
+      "step": 7100
+    },
+    {
+      "epoch": 4.763819095477387,
+      "grad_norm": 0.8598204255104065,
+      "learning_rate": 0.0002,
+      "loss": 1.4504,
+      "step": 7110
+    },
+    {
+      "epoch": 4.7705192629815745,
+      "grad_norm": 0.8951472640037537,
+      "learning_rate": 0.0002,
+      "loss": 1.3969,
+      "step": 7120
+    },
+    {
+      "epoch": 4.777219430485762,
+      "grad_norm": 0.9110309481620789,
+      "learning_rate": 0.0002,
+      "loss": 1.4339,
+      "step": 7130
+    },
+    {
+      "epoch": 4.78391959798995,
+      "grad_norm": 0.7929584980010986,
+      "learning_rate": 0.0002,
+      "loss": 1.4001,
+      "step": 7140
+    },
+    {
+      "epoch": 4.790619765494137,
+      "grad_norm": 0.7415322661399841,
+      "learning_rate": 0.0002,
+      "loss": 1.467,
+      "step": 7150
+    },
+    {
+      "epoch": 4.797319932998325,
+      "grad_norm": 0.7504757046699524,
+      "learning_rate": 0.0002,
+      "loss": 1.5107,
+      "step": 7160
+    },
+    {
+      "epoch": 4.8040201005025125,
+      "grad_norm": 0.7166924476623535,
+      "learning_rate": 0.0002,
+      "loss": 1.3736,
+      "step": 7170
+    },
+    {
+      "epoch": 4.8107202680067,
+      "grad_norm": 0.7728400826454163,
+      "learning_rate": 0.0002,
+      "loss": 1.4088,
+      "step": 7180
+    },
+    {
+      "epoch": 4.817420435510888,
+      "grad_norm": 0.7992154955863953,
+      "learning_rate": 0.0002,
+      "loss": 1.3814,
+      "step": 7190
+    },
+    {
+      "epoch": 4.824120603015075,
+      "grad_norm": 0.8655321002006531,
+      "learning_rate": 0.0002,
+      "loss": 1.3958,
+      "step": 7200
+    },
+    {
+      "epoch": 4.830820770519263,
+      "grad_norm": 0.7672632336616516,
+      "learning_rate": 0.0002,
+      "loss": 1.3837,
+      "step": 7210
+    },
+    {
+      "epoch": 4.8375209380234505,
+      "grad_norm": 0.708416223526001,
+      "learning_rate": 0.0002,
+      "loss": 1.4578,
+      "step": 7220
+    },
+    {
+      "epoch": 4.844221105527638,
+      "grad_norm": 0.8914081454277039,
+      "learning_rate": 0.0002,
+      "loss": 1.5413,
+      "step": 7230
+    },
+    {
+      "epoch": 4.850921273031826,
+      "grad_norm": 0.7141931653022766,
+      "learning_rate": 0.0002,
+      "loss": 1.3569,
+      "step": 7240
+    },
+    {
+      "epoch": 4.857621440536013,
+      "grad_norm": 0.6913040280342102,
+      "learning_rate": 0.0002,
+      "loss": 1.4532,
+      "step": 7250
+    },
+    {
+      "epoch": 4.864321608040201,
+      "grad_norm": 0.7871233820915222,
+      "learning_rate": 0.0002,
+      "loss": 1.3912,
+      "step": 7260
+    },
+    {
+      "epoch": 4.8710217755443885,
+      "grad_norm": 0.8466277122497559,
+      "learning_rate": 0.0002,
+      "loss": 1.3688,
+      "step": 7270
+    },
+    {
+      "epoch": 4.877721943048576,
+      "grad_norm": 0.8492183685302734,
+      "learning_rate": 0.0002,
+      "loss": 1.33,
+      "step": 7280
+    },
+    {
+      "epoch": 4.884422110552764,
+      "grad_norm": 0.8339574933052063,
+      "learning_rate": 0.0002,
+      "loss": 1.3744,
+      "step": 7290
+    },
+    {
+      "epoch": 4.891122278056951,
+      "grad_norm": 0.787022590637207,
+      "learning_rate": 0.0002,
+      "loss": 1.4157,
+      "step": 7300
+    },
+    {
+      "epoch": 4.897822445561139,
+      "grad_norm": 0.8877332806587219,
+      "learning_rate": 0.0002,
+      "loss": 1.3725,
+      "step": 7310
+    },
+    {
+      "epoch": 4.9045226130653266,
+      "grad_norm": 0.744989812374115,
+      "learning_rate": 0.0002,
+      "loss": 1.3968,
+      "step": 7320
+    },
+    {
+      "epoch": 4.911222780569514,
+      "grad_norm": 0.8027268648147583,
+      "learning_rate": 0.0002,
+      "loss": 1.4421,
+      "step": 7330
+    },
+    {
+      "epoch": 4.917922948073702,
+      "grad_norm": 0.6437455415725708,
+      "learning_rate": 0.0002,
+      "loss": 1.425,
+      "step": 7340
+    },
+    {
+      "epoch": 4.924623115577889,
+      "grad_norm": 0.685999870300293,
+      "learning_rate": 0.0002,
+      "loss": 1.4829,
+      "step": 7350
+    },
+    {
+      "epoch": 4.931323283082077,
+      "grad_norm": 0.9086187481880188,
+      "learning_rate": 0.0002,
+      "loss": 1.4352,
+      "step": 7360
+    },
+    {
+      "epoch": 4.938023450586265,
+      "grad_norm": 0.8272411227226257,
+      "learning_rate": 0.0002,
+      "loss": 1.4245,
+      "step": 7370
+    },
+    {
+      "epoch": 4.944723618090452,
+      "grad_norm": 0.9227852821350098,
+      "learning_rate": 0.0002,
+      "loss": 1.4226,
+      "step": 7380
+    },
+    {
+      "epoch": 4.95142378559464,
+      "grad_norm": 0.7688441276550293,
+      "learning_rate": 0.0002,
+      "loss": 1.3643,
+      "step": 7390
+    },
+    {
+      "epoch": 4.958123953098827,
+      "grad_norm": 0.8662643432617188,
+      "learning_rate": 0.0002,
+      "loss": 1.4491,
+      "step": 7400
+    },
+    {
+      "epoch": 4.964824120603015,
+      "grad_norm": 0.9234127998352051,
+      "learning_rate": 0.0002,
+      "loss": 1.4194,
+      "step": 7410
+    },
+    {
+      "epoch": 4.971524288107203,
+      "grad_norm": 0.9131470918655396,
+      "learning_rate": 0.0002,
+      "loss": 1.4009,
+      "step": 7420
+    },
+    {
+      "epoch": 4.97822445561139,
+      "grad_norm": 0.7377504110336304,
+      "learning_rate": 0.0002,
+      "loss": 1.4544,
+      "step": 7430
+    },
+    {
+      "epoch": 4.984924623115578,
+      "grad_norm": 0.8762801289558411,
+      "learning_rate": 0.0002,
+      "loss": 1.4008,
+      "step": 7440
+    },
+    {
+      "epoch": 4.991624790619765,
+      "grad_norm": 0.7919872999191284,
+      "learning_rate": 0.0002,
+      "loss": 1.4304,
+      "step": 7450
+    },
+    {
+      "epoch": 4.998324958123953,
+      "grad_norm": 0.7144299149513245,
+      "learning_rate": 0.0002,
+      "loss": 1.3817,
+      "step": 7460
+    },
+    {
+      "epoch": 4.99966499162479,
+      "eval_loss": 1.9291157722473145,
+      "eval_runtime": 37.9831,
+      "eval_samples_per_second": 13.559,
+      "eval_steps_per_second": 1.711,
+      "step": 7462
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 11936,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 8,
+  "save_steps": 200,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 3.453477907857408e+17,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-7462/training_args.bin b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-7462/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..064f299b0f3d2a28f5b1f5c68ef32caab3e2dd49
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-7462/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aa7b586fc440d1e22156593e2dd4267d2bdcb8920a02fdf352ea29a9bec3dd94
+size 5560
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-8955/README.md b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-8955/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..835e31ab8469ee39ddc8b2b6b2143a8c66dad510
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-8955/README.md
@@ -0,0 +1,202 @@
+---
+base_model: meta-llama/Meta-Llama-3-8B-Instruct
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.13.1
\ No newline at end of file
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-8955/adapter_config.json b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-8955/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..406c5a08dc4a2a33b52c62a482f98c217c417215
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-8955/adapter_config.json
@@ -0,0 +1,29 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "meta-llama/Meta-Llama-3-8B-Instruct",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}
\ No newline at end of file
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-8955/adapter_model.safetensors b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-8955/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..5b40d70adec9032ad731a772b6872ee7ddd87f63
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-8955/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e4c763944653cdd95b76a02dcbf86e28f4756a6d4cc892f064a27d4ddc8f9133
+size 109069176
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-8955/optimizer.pt b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-8955/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..b9ef194f0a312c74510403471b8a43ca27a9ac3e
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-8955/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5d79583f48db1e65beafca5194fdbbcf716952c7ca105d9d491cbfac04d6e093
+size 55532666
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-8955/rng_state.pth b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-8955/rng_state.pth
new file mode 100644
index 0000000000000000000000000000000000000000..6263b5766bc70d4dcd4ab4a3290016f260d99227
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-8955/rng_state.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9bb6a4d75a6ec8bc5a512db762e8603fc7707c39c1e931dd5c514a1b914b1314
+size 14244
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-8955/scheduler.pt b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-8955/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..145d0f09e381e5a002d1f525781535e205bde3a8
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-8955/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ea89e9fb267ad0bf2939fed06eb40079c33ce20fc2f9224d95bbb89c5510bab1
+size 1064
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-8955/special_tokens_map.json b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-8955/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..b43be96621d147110fb8a18b5776ec6e38516127
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-8955/special_tokens_map.json
@@ -0,0 +1,17 @@
+{
+  "bos_token": {
+    "content": "<|begin_of_text|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<|eot_id|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<|eot_id|>"
+}
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-8955/tokenizer.json b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-8955/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..86a33946b0c77216d2cce91bb28c8fada4a5e80b
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-8955/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3c5cf44023714fb39b05e71e425f8d7b92805ff73f7988b083b8c87f0bf87393
+size 17209961
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-8955/tokenizer_config.json b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-8955/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..061e40d9db3253624f86e8e364c15ef546527c9d
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-8955/tokenizer_config.json
@@ -0,0 +1,2063 @@
+{
+  "added_tokens_decoder": {
+    "128000": {
+      "content": "<|begin_of_text|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128001": {
+      "content": "<|end_of_text|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128002": {
+      "content": "<|reserved_special_token_0|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128003": {
+      "content": "<|reserved_special_token_1|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128004": {
+      "content": "<|reserved_special_token_2|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128005": {
+      "content": "<|reserved_special_token_3|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128006": {
+      "content": "<|start_header_id|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128007": {
+      "content": "<|end_header_id|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128008": {
+      "content": "<|reserved_special_token_4|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128009": {
+      "content": "<|eot_id|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128010": {
+      "content": "<|reserved_special_token_5|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128011": {
+      "content": "<|reserved_special_token_6|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128012": {
+      "content": "<|reserved_special_token_7|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128013": {
+      "content": "<|reserved_special_token_8|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128014": {
+      "content": "<|reserved_special_token_9|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128015": {
+      "content": "<|reserved_special_token_10|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128016": {
+      "content": "<|reserved_special_token_11|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128017": {
+      "content": "<|reserved_special_token_12|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128018": {
+      "content": "<|reserved_special_token_13|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128019": {
+      "content": "<|reserved_special_token_14|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128020": {
+      "content": "<|reserved_special_token_15|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128021": {
+      "content": "<|reserved_special_token_16|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128022": {
+      "content": "<|reserved_special_token_17|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128023": {
+      "content": "<|reserved_special_token_18|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128024": {
+      "content": "<|reserved_special_token_19|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128025": {
+      "content": "<|reserved_special_token_20|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128026": {
+      "content": "<|reserved_special_token_21|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128027": {
+      "content": "<|reserved_special_token_22|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128028": {
+      "content": "<|reserved_special_token_23|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128029": {
+      "content": "<|reserved_special_token_24|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128030": {
+      "content": "<|reserved_special_token_25|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128031": {
+      "content": "<|reserved_special_token_26|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128032": {
+      "content": "<|reserved_special_token_27|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128033": {
+      "content": "<|reserved_special_token_28|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128034": {
+      "content": "<|reserved_special_token_29|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128035": {
+      "content": "<|reserved_special_token_30|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128036": {
+      "content": "<|reserved_special_token_31|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128037": {
+      "content": "<|reserved_special_token_32|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128038": {
+      "content": "<|reserved_special_token_33|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128039": {
+      "content": "<|reserved_special_token_34|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128040": {
+      "content": "<|reserved_special_token_35|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128041": {
+      "content": "<|reserved_special_token_36|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128042": {
+      "content": "<|reserved_special_token_37|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128043": {
+      "content": "<|reserved_special_token_38|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128044": {
+      "content": "<|reserved_special_token_39|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128045": {
+      "content": "<|reserved_special_token_40|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128046": {
+      "content": "<|reserved_special_token_41|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128047": {
+      "content": "<|reserved_special_token_42|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128048": {
+      "content": "<|reserved_special_token_43|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128049": {
+      "content": "<|reserved_special_token_44|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128050": {
+      "content": "<|reserved_special_token_45|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128051": {
+      "content": "<|reserved_special_token_46|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128052": {
+      "content": "<|reserved_special_token_47|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128053": {
+      "content": "<|reserved_special_token_48|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128054": {
+      "content": "<|reserved_special_token_49|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128055": {
+      "content": "<|reserved_special_token_50|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128056": {
+      "content": "<|reserved_special_token_51|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128057": {
+      "content": "<|reserved_special_token_52|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128058": {
+      "content": "<|reserved_special_token_53|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128059": {
+      "content": "<|reserved_special_token_54|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128060": {
+      "content": "<|reserved_special_token_55|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128061": {
+      "content": "<|reserved_special_token_56|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128062": {
+      "content": "<|reserved_special_token_57|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128063": {
+      "content": "<|reserved_special_token_58|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128064": {
+      "content": "<|reserved_special_token_59|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128065": {
+      "content": "<|reserved_special_token_60|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128066": {
+      "content": "<|reserved_special_token_61|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128067": {
+      "content": "<|reserved_special_token_62|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128068": {
+      "content": "<|reserved_special_token_63|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128069": {
+      "content": "<|reserved_special_token_64|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128070": {
+      "content": "<|reserved_special_token_65|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128071": {
+      "content": "<|reserved_special_token_66|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128072": {
+      "content": "<|reserved_special_token_67|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128073": {
+      "content": "<|reserved_special_token_68|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128074": {
+      "content": "<|reserved_special_token_69|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128075": {
+      "content": "<|reserved_special_token_70|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128076": {
+      "content": "<|reserved_special_token_71|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128077": {
+      "content": "<|reserved_special_token_72|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128078": {
+      "content": "<|reserved_special_token_73|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128079": {
+      "content": "<|reserved_special_token_74|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128080": {
+      "content": "<|reserved_special_token_75|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128081": {
+      "content": "<|reserved_special_token_76|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128082": {
+      "content": "<|reserved_special_token_77|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128083": {
+      "content": "<|reserved_special_token_78|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128084": {
+      "content": "<|reserved_special_token_79|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128085": {
+      "content": "<|reserved_special_token_80|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128086": {
+      "content": "<|reserved_special_token_81|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128087": {
+      "content": "<|reserved_special_token_82|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128088": {
+      "content": "<|reserved_special_token_83|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128089": {
+      "content": "<|reserved_special_token_84|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128090": {
+      "content": "<|reserved_special_token_85|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128091": {
+      "content": "<|reserved_special_token_86|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128092": {
+      "content": "<|reserved_special_token_87|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128093": {
+      "content": "<|reserved_special_token_88|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128094": {
+      "content": "<|reserved_special_token_89|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128095": {
+      "content": "<|reserved_special_token_90|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128096": {
+      "content": "<|reserved_special_token_91|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128097": {
+      "content": "<|reserved_special_token_92|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128098": {
+      "content": "<|reserved_special_token_93|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128099": {
+      "content": "<|reserved_special_token_94|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128100": {
+      "content": "<|reserved_special_token_95|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128101": {
+      "content": "<|reserved_special_token_96|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128102": {
+      "content": "<|reserved_special_token_97|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128103": {
+      "content": "<|reserved_special_token_98|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128104": {
+      "content": "<|reserved_special_token_99|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128105": {
+      "content": "<|reserved_special_token_100|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128106": {
+      "content": "<|reserved_special_token_101|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128107": {
+      "content": "<|reserved_special_token_102|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128108": {
+      "content": "<|reserved_special_token_103|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128109": {
+      "content": "<|reserved_special_token_104|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128110": {
+      "content": "<|reserved_special_token_105|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128111": {
+      "content": "<|reserved_special_token_106|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128112": {
+      "content": "<|reserved_special_token_107|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128113": {
+      "content": "<|reserved_special_token_108|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128114": {
+      "content": "<|reserved_special_token_109|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128115": {
+      "content": "<|reserved_special_token_110|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128116": {
+      "content": "<|reserved_special_token_111|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128117": {
+      "content": "<|reserved_special_token_112|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128118": {
+      "content": "<|reserved_special_token_113|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128119": {
+      "content": "<|reserved_special_token_114|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128120": {
+      "content": "<|reserved_special_token_115|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128121": {
+      "content": "<|reserved_special_token_116|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128122": {
+      "content": "<|reserved_special_token_117|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128123": {
+      "content": "<|reserved_special_token_118|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128124": {
+      "content": "<|reserved_special_token_119|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128125": {
+      "content": "<|reserved_special_token_120|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128126": {
+      "content": "<|reserved_special_token_121|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128127": {
+      "content": "<|reserved_special_token_122|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128128": {
+      "content": "<|reserved_special_token_123|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128129": {
+      "content": "<|reserved_special_token_124|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128130": {
+      "content": "<|reserved_special_token_125|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128131": {
+      "content": "<|reserved_special_token_126|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128132": {
+      "content": "<|reserved_special_token_127|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128133": {
+      "content": "<|reserved_special_token_128|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128134": {
+      "content": "<|reserved_special_token_129|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128135": {
+      "content": "<|reserved_special_token_130|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128136": {
+      "content": "<|reserved_special_token_131|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128137": {
+      "content": "<|reserved_special_token_132|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128138": {
+      "content": "<|reserved_special_token_133|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128139": {
+      "content": "<|reserved_special_token_134|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128140": {
+      "content": "<|reserved_special_token_135|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128141": {
+      "content": "<|reserved_special_token_136|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128142": {
+      "content": "<|reserved_special_token_137|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128143": {
+      "content": "<|reserved_special_token_138|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128144": {
+      "content": "<|reserved_special_token_139|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128145": {
+      "content": "<|reserved_special_token_140|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128146": {
+      "content": "<|reserved_special_token_141|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128147": {
+      "content": "<|reserved_special_token_142|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128148": {
+      "content": "<|reserved_special_token_143|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128149": {
+      "content": "<|reserved_special_token_144|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128150": {
+      "content": "<|reserved_special_token_145|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128151": {
+      "content": "<|reserved_special_token_146|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128152": {
+      "content": "<|reserved_special_token_147|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128153": {
+      "content": "<|reserved_special_token_148|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128154": {
+      "content": "<|reserved_special_token_149|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128155": {
+      "content": "<|reserved_special_token_150|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128156": {
+      "content": "<|reserved_special_token_151|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128157": {
+      "content": "<|reserved_special_token_152|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128158": {
+      "content": "<|reserved_special_token_153|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128159": {
+      "content": "<|reserved_special_token_154|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128160": {
+      "content": "<|reserved_special_token_155|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128161": {
+      "content": "<|reserved_special_token_156|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128162": {
+      "content": "<|reserved_special_token_157|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128163": {
+      "content": "<|reserved_special_token_158|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128164": {
+      "content": "<|reserved_special_token_159|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128165": {
+      "content": "<|reserved_special_token_160|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128166": {
+      "content": "<|reserved_special_token_161|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128167": {
+      "content": "<|reserved_special_token_162|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128168": {
+      "content": "<|reserved_special_token_163|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128169": {
+      "content": "<|reserved_special_token_164|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128170": {
+      "content": "<|reserved_special_token_165|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128171": {
+      "content": "<|reserved_special_token_166|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128172": {
+      "content": "<|reserved_special_token_167|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128173": {
+      "content": "<|reserved_special_token_168|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128174": {
+      "content": "<|reserved_special_token_169|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128175": {
+      "content": "<|reserved_special_token_170|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128176": {
+      "content": "<|reserved_special_token_171|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128177": {
+      "content": "<|reserved_special_token_172|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128178": {
+      "content": "<|reserved_special_token_173|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128179": {
+      "content": "<|reserved_special_token_174|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128180": {
+      "content": "<|reserved_special_token_175|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128181": {
+      "content": "<|reserved_special_token_176|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128182": {
+      "content": "<|reserved_special_token_177|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128183": {
+      "content": "<|reserved_special_token_178|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128184": {
+      "content": "<|reserved_special_token_179|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128185": {
+      "content": "<|reserved_special_token_180|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128186": {
+      "content": "<|reserved_special_token_181|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128187": {
+      "content": "<|reserved_special_token_182|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128188": {
+      "content": "<|reserved_special_token_183|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128189": {
+      "content": "<|reserved_special_token_184|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128190": {
+      "content": "<|reserved_special_token_185|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128191": {
+      "content": "<|reserved_special_token_186|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128192": {
+      "content": "<|reserved_special_token_187|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128193": {
+      "content": "<|reserved_special_token_188|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128194": {
+      "content": "<|reserved_special_token_189|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128195": {
+      "content": "<|reserved_special_token_190|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128196": {
+      "content": "<|reserved_special_token_191|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128197": {
+      "content": "<|reserved_special_token_192|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128198": {
+      "content": "<|reserved_special_token_193|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128199": {
+      "content": "<|reserved_special_token_194|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128200": {
+      "content": "<|reserved_special_token_195|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128201": {
+      "content": "<|reserved_special_token_196|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128202": {
+      "content": "<|reserved_special_token_197|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128203": {
+      "content": "<|reserved_special_token_198|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128204": {
+      "content": "<|reserved_special_token_199|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128205": {
+      "content": "<|reserved_special_token_200|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128206": {
+      "content": "<|reserved_special_token_201|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128207": {
+      "content": "<|reserved_special_token_202|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128208": {
+      "content": "<|reserved_special_token_203|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128209": {
+      "content": "<|reserved_special_token_204|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128210": {
+      "content": "<|reserved_special_token_205|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128211": {
+      "content": "<|reserved_special_token_206|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128212": {
+      "content": "<|reserved_special_token_207|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128213": {
+      "content": "<|reserved_special_token_208|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128214": {
+      "content": "<|reserved_special_token_209|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128215": {
+      "content": "<|reserved_special_token_210|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128216": {
+      "content": "<|reserved_special_token_211|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128217": {
+      "content": "<|reserved_special_token_212|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128218": {
+      "content": "<|reserved_special_token_213|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128219": {
+      "content": "<|reserved_special_token_214|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128220": {
+      "content": "<|reserved_special_token_215|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128221": {
+      "content": "<|reserved_special_token_216|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128222": {
+      "content": "<|reserved_special_token_217|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128223": {
+      "content": "<|reserved_special_token_218|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128224": {
+      "content": "<|reserved_special_token_219|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128225": {
+      "content": "<|reserved_special_token_220|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128226": {
+      "content": "<|reserved_special_token_221|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128227": {
+      "content": "<|reserved_special_token_222|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128228": {
+      "content": "<|reserved_special_token_223|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128229": {
+      "content": "<|reserved_special_token_224|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128230": {
+      "content": "<|reserved_special_token_225|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128231": {
+      "content": "<|reserved_special_token_226|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128232": {
+      "content": "<|reserved_special_token_227|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128233": {
+      "content": "<|reserved_special_token_228|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128234": {
+      "content": "<|reserved_special_token_229|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128235": {
+      "content": "<|reserved_special_token_230|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128236": {
+      "content": "<|reserved_special_token_231|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128237": {
+      "content": "<|reserved_special_token_232|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128238": {
+      "content": "<|reserved_special_token_233|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128239": {
+      "content": "<|reserved_special_token_234|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128240": {
+      "content": "<|reserved_special_token_235|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128241": {
+      "content": "<|reserved_special_token_236|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128242": {
+      "content": "<|reserved_special_token_237|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128243": {
+      "content": "<|reserved_special_token_238|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128244": {
+      "content": "<|reserved_special_token_239|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128245": {
+      "content": "<|reserved_special_token_240|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128246": {
+      "content": "<|reserved_special_token_241|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128247": {
+      "content": "<|reserved_special_token_242|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128248": {
+      "content": "<|reserved_special_token_243|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128249": {
+      "content": "<|reserved_special_token_244|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128250": {
+      "content": "<|reserved_special_token_245|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128251": {
+      "content": "<|reserved_special_token_246|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128252": {
+      "content": "<|reserved_special_token_247|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128253": {
+      "content": "<|reserved_special_token_248|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128254": {
+      "content": "<|reserved_special_token_249|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128255": {
+      "content": "<|reserved_special_token_250|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<|begin_of_text|>",
+  "chat_template": "{% set loop_messages = messages %}{% for message in loop_messages %}{% set content = '<|start_header_id|>' + message['role'] + '<|end_header_id|>\n\n'+ message['content'] | trim + '<|eot_id|>' %}{% if loop.index0 == 0 %}{% set content = bos_token + content %}{% endif %}{{ content }}{% endfor %}{% if add_generation_prompt %}{{ '<|start_header_id|>assistant<|end_header_id|>\n\n' }}{% endif %}",
+  "clean_up_tokenization_spaces": true,
+  "eos_token": "<|eot_id|>",
+  "model_input_names": [
+    "input_ids",
+    "attention_mask"
+  ],
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<|eot_id|>",
+  "tokenizer_class": "PreTrainedTokenizerFast"
+}
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-8955/trainer_state.json b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-8955/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..16edd060a824a69118f566a12abeebbab3325edf
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-8955/trainer_state.json
@@ -0,0 +1,6346 @@
+{
+  "best_metric": 1.8028968572616577,
+  "best_model_checkpoint": "outputs-001/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-2985",
+  "epoch": 6.0,
+  "eval_steps": 10,
+  "global_step": 8955,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.006700167504187605,
+      "grad_norm": 0.565915048122406,
+      "learning_rate": 0.0002,
+      "loss": 2.6189,
+      "step": 10
+    },
+    {
+      "epoch": 0.01340033500837521,
+      "grad_norm": 0.5004463791847229,
+      "learning_rate": 0.0002,
+      "loss": 2.3162,
+      "step": 20
+    },
+    {
+      "epoch": 0.020100502512562814,
+      "grad_norm": 0.511043906211853,
+      "learning_rate": 0.0002,
+      "loss": 2.0576,
+      "step": 30
+    },
+    {
+      "epoch": 0.02680067001675042,
+      "grad_norm": 0.47327178716659546,
+      "learning_rate": 0.0002,
+      "loss": 2.0085,
+      "step": 40
+    },
+    {
+      "epoch": 0.03350083752093802,
+      "grad_norm": 0.5511676669120789,
+      "learning_rate": 0.0002,
+      "loss": 2.0276,
+      "step": 50
+    },
+    {
+      "epoch": 0.04020100502512563,
+      "grad_norm": 0.4666278064250946,
+      "learning_rate": 0.0002,
+      "loss": 1.9075,
+      "step": 60
+    },
+    {
+      "epoch": 0.04690117252931323,
+      "grad_norm": 0.5310961008071899,
+      "learning_rate": 0.0002,
+      "loss": 1.8413,
+      "step": 70
+    },
+    {
+      "epoch": 0.05360134003350084,
+      "grad_norm": 0.5606027245521545,
+      "learning_rate": 0.0002,
+      "loss": 1.8711,
+      "step": 80
+    },
+    {
+      "epoch": 0.06030150753768844,
+      "grad_norm": 0.4934779703617096,
+      "learning_rate": 0.0002,
+      "loss": 1.9282,
+      "step": 90
+    },
+    {
+      "epoch": 0.06700167504187604,
+      "grad_norm": 0.4821869730949402,
+      "learning_rate": 0.0002,
+      "loss": 1.8925,
+      "step": 100
+    },
+    {
+      "epoch": 0.07370184254606366,
+      "grad_norm": 0.5262084603309631,
+      "learning_rate": 0.0002,
+      "loss": 1.8628,
+      "step": 110
+    },
+    {
+      "epoch": 0.08040201005025126,
+      "grad_norm": 0.3774230182170868,
+      "learning_rate": 0.0002,
+      "loss": 1.8347,
+      "step": 120
+    },
+    {
+      "epoch": 0.08710217755443886,
+      "grad_norm": 0.34137430787086487,
+      "learning_rate": 0.0002,
+      "loss": 1.8386,
+      "step": 130
+    },
+    {
+      "epoch": 0.09380234505862646,
+      "grad_norm": 0.407272070646286,
+      "learning_rate": 0.0002,
+      "loss": 1.861,
+      "step": 140
+    },
+    {
+      "epoch": 0.10050251256281408,
+      "grad_norm": 0.4011937975883484,
+      "learning_rate": 0.0002,
+      "loss": 1.8279,
+      "step": 150
+    },
+    {
+      "epoch": 0.10720268006700168,
+      "grad_norm": 0.4432467222213745,
+      "learning_rate": 0.0002,
+      "loss": 1.9317,
+      "step": 160
+    },
+    {
+      "epoch": 0.11390284757118928,
+      "grad_norm": 0.44030463695526123,
+      "learning_rate": 0.0002,
+      "loss": 1.8157,
+      "step": 170
+    },
+    {
+      "epoch": 0.12060301507537688,
+      "grad_norm": 0.3799569308757782,
+      "learning_rate": 0.0002,
+      "loss": 1.8534,
+      "step": 180
+    },
+    {
+      "epoch": 0.1273031825795645,
+      "grad_norm": 0.33721521496772766,
+      "learning_rate": 0.0002,
+      "loss": 1.7658,
+      "step": 190
+    },
+    {
+      "epoch": 0.13400335008375208,
+      "grad_norm": 0.4096226692199707,
+      "learning_rate": 0.0002,
+      "loss": 1.8269,
+      "step": 200
+    },
+    {
+      "epoch": 0.1407035175879397,
+      "grad_norm": 0.37374693155288696,
+      "learning_rate": 0.0002,
+      "loss": 1.802,
+      "step": 210
+    },
+    {
+      "epoch": 0.1474036850921273,
+      "grad_norm": 0.3249480128288269,
+      "learning_rate": 0.0002,
+      "loss": 1.8901,
+      "step": 220
+    },
+    {
+      "epoch": 0.1541038525963149,
+      "grad_norm": 0.3612042963504791,
+      "learning_rate": 0.0002,
+      "loss": 1.8163,
+      "step": 230
+    },
+    {
+      "epoch": 0.16080402010050251,
+      "grad_norm": 0.3686671257019043,
+      "learning_rate": 0.0002,
+      "loss": 1.7585,
+      "step": 240
+    },
+    {
+      "epoch": 0.16750418760469013,
+      "grad_norm": 0.3521044850349426,
+      "learning_rate": 0.0002,
+      "loss": 1.8365,
+      "step": 250
+    },
+    {
+      "epoch": 0.17420435510887772,
+      "grad_norm": 0.4073677361011505,
+      "learning_rate": 0.0002,
+      "loss": 1.8623,
+      "step": 260
+    },
+    {
+      "epoch": 0.18090452261306533,
+      "grad_norm": 0.34522193670272827,
+      "learning_rate": 0.0002,
+      "loss": 1.8026,
+      "step": 270
+    },
+    {
+      "epoch": 0.18760469011725292,
+      "grad_norm": 0.4121900498867035,
+      "learning_rate": 0.0002,
+      "loss": 1.8162,
+      "step": 280
+    },
+    {
+      "epoch": 0.19430485762144054,
+      "grad_norm": 0.3544778525829315,
+      "learning_rate": 0.0002,
+      "loss": 1.7976,
+      "step": 290
+    },
+    {
+      "epoch": 0.20100502512562815,
+      "grad_norm": 0.3482133448123932,
+      "learning_rate": 0.0002,
+      "loss": 1.8787,
+      "step": 300
+    },
+    {
+      "epoch": 0.20770519262981574,
+      "grad_norm": 0.3421826660633087,
+      "learning_rate": 0.0002,
+      "loss": 1.8578,
+      "step": 310
+    },
+    {
+      "epoch": 0.21440536013400335,
+      "grad_norm": 0.5024696588516235,
+      "learning_rate": 0.0002,
+      "loss": 1.8013,
+      "step": 320
+    },
+    {
+      "epoch": 0.22110552763819097,
+      "grad_norm": 0.36013063788414,
+      "learning_rate": 0.0002,
+      "loss": 1.8607,
+      "step": 330
+    },
+    {
+      "epoch": 0.22780569514237856,
+      "grad_norm": 0.3611244857311249,
+      "learning_rate": 0.0002,
+      "loss": 1.9075,
+      "step": 340
+    },
+    {
+      "epoch": 0.23450586264656617,
+      "grad_norm": 0.39244529604911804,
+      "learning_rate": 0.0002,
+      "loss": 1.8128,
+      "step": 350
+    },
+    {
+      "epoch": 0.24120603015075376,
+      "grad_norm": 0.3299325704574585,
+      "learning_rate": 0.0002,
+      "loss": 1.7885,
+      "step": 360
+    },
+    {
+      "epoch": 0.24790619765494137,
+      "grad_norm": 0.3994322419166565,
+      "learning_rate": 0.0002,
+      "loss": 1.8028,
+      "step": 370
+    },
+    {
+      "epoch": 0.254606365159129,
+      "grad_norm": 0.3559151887893677,
+      "learning_rate": 0.0002,
+      "loss": 1.8321,
+      "step": 380
+    },
+    {
+      "epoch": 0.2613065326633166,
+      "grad_norm": 0.3873756229877472,
+      "learning_rate": 0.0002,
+      "loss": 1.7802,
+      "step": 390
+    },
+    {
+      "epoch": 0.26800670016750416,
+      "grad_norm": 0.3710744082927704,
+      "learning_rate": 0.0002,
+      "loss": 1.7844,
+      "step": 400
+    },
+    {
+      "epoch": 0.2747068676716918,
+      "grad_norm": 0.3618465065956116,
+      "learning_rate": 0.0002,
+      "loss": 1.7789,
+      "step": 410
+    },
+    {
+      "epoch": 0.2814070351758794,
+      "grad_norm": 0.30063769221305847,
+      "learning_rate": 0.0002,
+      "loss": 1.8529,
+      "step": 420
+    },
+    {
+      "epoch": 0.288107202680067,
+      "grad_norm": 0.3695628345012665,
+      "learning_rate": 0.0002,
+      "loss": 1.7765,
+      "step": 430
+    },
+    {
+      "epoch": 0.2948073701842546,
+      "grad_norm": 0.31451135873794556,
+      "learning_rate": 0.0002,
+      "loss": 1.7982,
+      "step": 440
+    },
+    {
+      "epoch": 0.3015075376884422,
+      "grad_norm": 0.3959707021713257,
+      "learning_rate": 0.0002,
+      "loss": 1.7517,
+      "step": 450
+    },
+    {
+      "epoch": 0.3082077051926298,
+      "grad_norm": 0.33475354313850403,
+      "learning_rate": 0.0002,
+      "loss": 1.8142,
+      "step": 460
+    },
+    {
+      "epoch": 0.3149078726968174,
+      "grad_norm": 0.33933115005493164,
+      "learning_rate": 0.0002,
+      "loss": 1.8805,
+      "step": 470
+    },
+    {
+      "epoch": 0.32160804020100503,
+      "grad_norm": 0.3264943063259125,
+      "learning_rate": 0.0002,
+      "loss": 1.7564,
+      "step": 480
+    },
+    {
+      "epoch": 0.32830820770519265,
+      "grad_norm": 0.40188100934028625,
+      "learning_rate": 0.0002,
+      "loss": 1.8428,
+      "step": 490
+    },
+    {
+      "epoch": 0.33500837520938026,
+      "grad_norm": 0.37408649921417236,
+      "learning_rate": 0.0002,
+      "loss": 1.7624,
+      "step": 500
+    },
+    {
+      "epoch": 0.3417085427135678,
+      "grad_norm": 0.33925938606262207,
+      "learning_rate": 0.0002,
+      "loss": 1.7745,
+      "step": 510
+    },
+    {
+      "epoch": 0.34840871021775544,
+      "grad_norm": 0.36836713552474976,
+      "learning_rate": 0.0002,
+      "loss": 1.814,
+      "step": 520
+    },
+    {
+      "epoch": 0.35510887772194305,
+      "grad_norm": 0.37284499406814575,
+      "learning_rate": 0.0002,
+      "loss": 1.8037,
+      "step": 530
+    },
+    {
+      "epoch": 0.36180904522613067,
+      "grad_norm": 0.3192278742790222,
+      "learning_rate": 0.0002,
+      "loss": 1.8379,
+      "step": 540
+    },
+    {
+      "epoch": 0.3685092127303183,
+      "grad_norm": 0.30233290791511536,
+      "learning_rate": 0.0002,
+      "loss": 1.8702,
+      "step": 550
+    },
+    {
+      "epoch": 0.37520938023450584,
+      "grad_norm": 0.3340817391872406,
+      "learning_rate": 0.0002,
+      "loss": 1.8799,
+      "step": 560
+    },
+    {
+      "epoch": 0.38190954773869346,
+      "grad_norm": 0.32600095868110657,
+      "learning_rate": 0.0002,
+      "loss": 1.8404,
+      "step": 570
+    },
+    {
+      "epoch": 0.38860971524288107,
+      "grad_norm": 0.33711278438568115,
+      "learning_rate": 0.0002,
+      "loss": 1.7804,
+      "step": 580
+    },
+    {
+      "epoch": 0.3953098827470687,
+      "grad_norm": 0.34890690445899963,
+      "learning_rate": 0.0002,
+      "loss": 1.8445,
+      "step": 590
+    },
+    {
+      "epoch": 0.4020100502512563,
+      "grad_norm": 0.38238924741744995,
+      "learning_rate": 0.0002,
+      "loss": 1.8187,
+      "step": 600
+    },
+    {
+      "epoch": 0.40871021775544386,
+      "grad_norm": 0.34399354457855225,
+      "learning_rate": 0.0002,
+      "loss": 1.8111,
+      "step": 610
+    },
+    {
+      "epoch": 0.4154103852596315,
+      "grad_norm": 0.3346073627471924,
+      "learning_rate": 0.0002,
+      "loss": 1.8006,
+      "step": 620
+    },
+    {
+      "epoch": 0.4221105527638191,
+      "grad_norm": 0.3545648157596588,
+      "learning_rate": 0.0002,
+      "loss": 1.7705,
+      "step": 630
+    },
+    {
+      "epoch": 0.4288107202680067,
+      "grad_norm": 0.3378899097442627,
+      "learning_rate": 0.0002,
+      "loss": 1.8445,
+      "step": 640
+    },
+    {
+      "epoch": 0.4355108877721943,
+      "grad_norm": 0.3255569040775299,
+      "learning_rate": 0.0002,
+      "loss": 1.804,
+      "step": 650
+    },
+    {
+      "epoch": 0.44221105527638194,
+      "grad_norm": 0.34880587458610535,
+      "learning_rate": 0.0002,
+      "loss": 1.7679,
+      "step": 660
+    },
+    {
+      "epoch": 0.4489112227805695,
+      "grad_norm": 0.3402383625507355,
+      "learning_rate": 0.0002,
+      "loss": 1.7861,
+      "step": 670
+    },
+    {
+      "epoch": 0.4556113902847571,
+      "grad_norm": 0.3594033718109131,
+      "learning_rate": 0.0002,
+      "loss": 1.8131,
+      "step": 680
+    },
+    {
+      "epoch": 0.4623115577889447,
+      "grad_norm": 0.31000566482543945,
+      "learning_rate": 0.0002,
+      "loss": 1.8399,
+      "step": 690
+    },
+    {
+      "epoch": 0.46901172529313234,
+      "grad_norm": 0.37229061126708984,
+      "learning_rate": 0.0002,
+      "loss": 1.7521,
+      "step": 700
+    },
+    {
+      "epoch": 0.47571189279731996,
+      "grad_norm": 0.315801739692688,
+      "learning_rate": 0.0002,
+      "loss": 1.7779,
+      "step": 710
+    },
+    {
+      "epoch": 0.4824120603015075,
+      "grad_norm": 0.3220832645893097,
+      "learning_rate": 0.0002,
+      "loss": 1.7515,
+      "step": 720
+    },
+    {
+      "epoch": 0.48911222780569513,
+      "grad_norm": 0.3435456156730652,
+      "learning_rate": 0.0002,
+      "loss": 1.7181,
+      "step": 730
+    },
+    {
+      "epoch": 0.49581239530988275,
+      "grad_norm": 0.30380892753601074,
+      "learning_rate": 0.0002,
+      "loss": 1.8844,
+      "step": 740
+    },
+    {
+      "epoch": 0.5025125628140703,
+      "grad_norm": 0.3555026054382324,
+      "learning_rate": 0.0002,
+      "loss": 1.7792,
+      "step": 750
+    },
+    {
+      "epoch": 0.509212730318258,
+      "grad_norm": 0.3019855320453644,
+      "learning_rate": 0.0002,
+      "loss": 1.7714,
+      "step": 760
+    },
+    {
+      "epoch": 0.5159128978224455,
+      "grad_norm": 0.309111088514328,
+      "learning_rate": 0.0002,
+      "loss": 1.7962,
+      "step": 770
+    },
+    {
+      "epoch": 0.5226130653266332,
+      "grad_norm": 0.366020530462265,
+      "learning_rate": 0.0002,
+      "loss": 1.7913,
+      "step": 780
+    },
+    {
+      "epoch": 0.5293132328308208,
+      "grad_norm": 0.3267050087451935,
+      "learning_rate": 0.0002,
+      "loss": 1.8008,
+      "step": 790
+    },
+    {
+      "epoch": 0.5360134003350083,
+      "grad_norm": 0.34265750646591187,
+      "learning_rate": 0.0002,
+      "loss": 1.7397,
+      "step": 800
+    },
+    {
+      "epoch": 0.542713567839196,
+      "grad_norm": 0.313669890165329,
+      "learning_rate": 0.0002,
+      "loss": 1.8251,
+      "step": 810
+    },
+    {
+      "epoch": 0.5494137353433836,
+      "grad_norm": 0.3355236053466797,
+      "learning_rate": 0.0002,
+      "loss": 1.8369,
+      "step": 820
+    },
+    {
+      "epoch": 0.5561139028475712,
+      "grad_norm": 0.3186608552932739,
+      "learning_rate": 0.0002,
+      "loss": 1.7381,
+      "step": 830
+    },
+    {
+      "epoch": 0.5628140703517588,
+      "grad_norm": 0.30357518792152405,
+      "learning_rate": 0.0002,
+      "loss": 1.8034,
+      "step": 840
+    },
+    {
+      "epoch": 0.5695142378559463,
+      "grad_norm": 0.3990040123462677,
+      "learning_rate": 0.0002,
+      "loss": 1.769,
+      "step": 850
+    },
+    {
+      "epoch": 0.576214405360134,
+      "grad_norm": 0.34363803267478943,
+      "learning_rate": 0.0002,
+      "loss": 1.7482,
+      "step": 860
+    },
+    {
+      "epoch": 0.5829145728643216,
+      "grad_norm": 0.3757908046245575,
+      "learning_rate": 0.0002,
+      "loss": 1.8106,
+      "step": 870
+    },
+    {
+      "epoch": 0.5896147403685092,
+      "grad_norm": 0.3359757661819458,
+      "learning_rate": 0.0002,
+      "loss": 1.8104,
+      "step": 880
+    },
+    {
+      "epoch": 0.5963149078726968,
+      "grad_norm": 0.5555329918861389,
+      "learning_rate": 0.0002,
+      "loss": 1.7591,
+      "step": 890
+    },
+    {
+      "epoch": 0.6030150753768844,
+      "grad_norm": 0.4046323895454407,
+      "learning_rate": 0.0002,
+      "loss": 1.7715,
+      "step": 900
+    },
+    {
+      "epoch": 0.609715242881072,
+      "grad_norm": 0.29834219813346863,
+      "learning_rate": 0.0002,
+      "loss": 1.7998,
+      "step": 910
+    },
+    {
+      "epoch": 0.6164154103852596,
+      "grad_norm": 0.3241238594055176,
+      "learning_rate": 0.0002,
+      "loss": 1.7826,
+      "step": 920
+    },
+    {
+      "epoch": 0.6231155778894473,
+      "grad_norm": 0.35154739022254944,
+      "learning_rate": 0.0002,
+      "loss": 1.8342,
+      "step": 930
+    },
+    {
+      "epoch": 0.6298157453936348,
+      "grad_norm": 0.3287706673145294,
+      "learning_rate": 0.0002,
+      "loss": 1.8076,
+      "step": 940
+    },
+    {
+      "epoch": 0.6365159128978225,
+      "grad_norm": 0.35670626163482666,
+      "learning_rate": 0.0002,
+      "loss": 1.8038,
+      "step": 950
+    },
+    {
+      "epoch": 0.6432160804020101,
+      "grad_norm": 0.6114104986190796,
+      "learning_rate": 0.0002,
+      "loss": 1.869,
+      "step": 960
+    },
+    {
+      "epoch": 0.6499162479061976,
+      "grad_norm": 0.3186565041542053,
+      "learning_rate": 0.0002,
+      "loss": 1.8297,
+      "step": 970
+    },
+    {
+      "epoch": 0.6566164154103853,
+      "grad_norm": 0.27164125442504883,
+      "learning_rate": 0.0002,
+      "loss": 1.7539,
+      "step": 980
+    },
+    {
+      "epoch": 0.6633165829145728,
+      "grad_norm": 0.34407344460487366,
+      "learning_rate": 0.0002,
+      "loss": 1.8339,
+      "step": 990
+    },
+    {
+      "epoch": 0.6700167504187605,
+      "grad_norm": 0.368415892124176,
+      "learning_rate": 0.0002,
+      "loss": 1.855,
+      "step": 1000
+    },
+    {
+      "epoch": 0.6767169179229481,
+      "grad_norm": 0.3306390643119812,
+      "learning_rate": 0.0002,
+      "loss": 1.7821,
+      "step": 1010
+    },
+    {
+      "epoch": 0.6834170854271356,
+      "grad_norm": 0.3198648989200592,
+      "learning_rate": 0.0002,
+      "loss": 1.786,
+      "step": 1020
+    },
+    {
+      "epoch": 0.6901172529313233,
+      "grad_norm": 0.3092987537384033,
+      "learning_rate": 0.0002,
+      "loss": 1.816,
+      "step": 1030
+    },
+    {
+      "epoch": 0.6968174204355109,
+      "grad_norm": 0.3090653419494629,
+      "learning_rate": 0.0002,
+      "loss": 1.7689,
+      "step": 1040
+    },
+    {
+      "epoch": 0.7035175879396985,
+      "grad_norm": 0.3485880196094513,
+      "learning_rate": 0.0002,
+      "loss": 1.7544,
+      "step": 1050
+    },
+    {
+      "epoch": 0.7102177554438861,
+      "grad_norm": 0.35782721638679504,
+      "learning_rate": 0.0002,
+      "loss": 1.8187,
+      "step": 1060
+    },
+    {
+      "epoch": 0.7169179229480737,
+      "grad_norm": 0.34256869554519653,
+      "learning_rate": 0.0002,
+      "loss": 1.806,
+      "step": 1070
+    },
+    {
+      "epoch": 0.7236180904522613,
+      "grad_norm": 0.30461037158966064,
+      "learning_rate": 0.0002,
+      "loss": 1.7873,
+      "step": 1080
+    },
+    {
+      "epoch": 0.7303182579564489,
+      "grad_norm": 0.3398691713809967,
+      "learning_rate": 0.0002,
+      "loss": 1.7367,
+      "step": 1090
+    },
+    {
+      "epoch": 0.7370184254606366,
+      "grad_norm": 0.3180808126926422,
+      "learning_rate": 0.0002,
+      "loss": 1.8756,
+      "step": 1100
+    },
+    {
+      "epoch": 0.7437185929648241,
+      "grad_norm": 0.34400665760040283,
+      "learning_rate": 0.0002,
+      "loss": 1.6988,
+      "step": 1110
+    },
+    {
+      "epoch": 0.7504187604690117,
+      "grad_norm": 0.34244877099990845,
+      "learning_rate": 0.0002,
+      "loss": 1.7851,
+      "step": 1120
+    },
+    {
+      "epoch": 0.7571189279731994,
+      "grad_norm": 0.29946693778038025,
+      "learning_rate": 0.0002,
+      "loss": 1.7841,
+      "step": 1130
+    },
+    {
+      "epoch": 0.7638190954773869,
+      "grad_norm": 0.37547236680984497,
+      "learning_rate": 0.0002,
+      "loss": 1.7456,
+      "step": 1140
+    },
+    {
+      "epoch": 0.7705192629815746,
+      "grad_norm": 0.3263005018234253,
+      "learning_rate": 0.0002,
+      "loss": 1.8425,
+      "step": 1150
+    },
+    {
+      "epoch": 0.7772194304857621,
+      "grad_norm": 0.41363608837127686,
+      "learning_rate": 0.0002,
+      "loss": 1.7222,
+      "step": 1160
+    },
+    {
+      "epoch": 0.7839195979899497,
+      "grad_norm": 0.36267954111099243,
+      "learning_rate": 0.0002,
+      "loss": 1.7836,
+      "step": 1170
+    },
+    {
+      "epoch": 0.7906197654941374,
+      "grad_norm": 0.31789499521255493,
+      "learning_rate": 0.0002,
+      "loss": 1.9183,
+      "step": 1180
+    },
+    {
+      "epoch": 0.7973199329983249,
+      "grad_norm": 0.5708149075508118,
+      "learning_rate": 0.0002,
+      "loss": 1.78,
+      "step": 1190
+    },
+    {
+      "epoch": 0.8040201005025126,
+      "grad_norm": 0.322099506855011,
+      "learning_rate": 0.0002,
+      "loss": 1.6908,
+      "step": 1200
+    },
+    {
+      "epoch": 0.8107202680067002,
+      "grad_norm": 0.3419909179210663,
+      "learning_rate": 0.0002,
+      "loss": 1.7639,
+      "step": 1210
+    },
+    {
+      "epoch": 0.8174204355108877,
+      "grad_norm": 0.36286255717277527,
+      "learning_rate": 0.0002,
+      "loss": 1.7428,
+      "step": 1220
+    },
+    {
+      "epoch": 0.8241206030150754,
+      "grad_norm": 0.33992862701416016,
+      "learning_rate": 0.0002,
+      "loss": 1.8409,
+      "step": 1230
+    },
+    {
+      "epoch": 0.830820770519263,
+      "grad_norm": 0.32622793316841125,
+      "learning_rate": 0.0002,
+      "loss": 1.7507,
+      "step": 1240
+    },
+    {
+      "epoch": 0.8375209380234506,
+      "grad_norm": 0.3036167621612549,
+      "learning_rate": 0.0002,
+      "loss": 1.8098,
+      "step": 1250
+    },
+    {
+      "epoch": 0.8442211055276382,
+      "grad_norm": 0.3182215392589569,
+      "learning_rate": 0.0002,
+      "loss": 1.8094,
+      "step": 1260
+    },
+    {
+      "epoch": 0.8509212730318257,
+      "grad_norm": 0.3270018696784973,
+      "learning_rate": 0.0002,
+      "loss": 1.8017,
+      "step": 1270
+    },
+    {
+      "epoch": 0.8576214405360134,
+      "grad_norm": 0.32652342319488525,
+      "learning_rate": 0.0002,
+      "loss": 1.798,
+      "step": 1280
+    },
+    {
+      "epoch": 0.864321608040201,
+      "grad_norm": 0.3631329834461212,
+      "learning_rate": 0.0002,
+      "loss": 1.7448,
+      "step": 1290
+    },
+    {
+      "epoch": 0.8710217755443886,
+      "grad_norm": 0.36706018447875977,
+      "learning_rate": 0.0002,
+      "loss": 1.7,
+      "step": 1300
+    },
+    {
+      "epoch": 0.8777219430485762,
+      "grad_norm": 0.3347418010234833,
+      "learning_rate": 0.0002,
+      "loss": 1.8178,
+      "step": 1310
+    },
+    {
+      "epoch": 0.8844221105527639,
+      "grad_norm": 0.34371060132980347,
+      "learning_rate": 0.0002,
+      "loss": 1.7824,
+      "step": 1320
+    },
+    {
+      "epoch": 0.8911222780569514,
+      "grad_norm": 0.3029090166091919,
+      "learning_rate": 0.0002,
+      "loss": 1.783,
+      "step": 1330
+    },
+    {
+      "epoch": 0.897822445561139,
+      "grad_norm": 0.34700682759284973,
+      "learning_rate": 0.0002,
+      "loss": 1.8017,
+      "step": 1340
+    },
+    {
+      "epoch": 0.9045226130653267,
+      "grad_norm": 0.35574328899383545,
+      "learning_rate": 0.0002,
+      "loss": 1.7998,
+      "step": 1350
+    },
+    {
+      "epoch": 0.9112227805695142,
+      "grad_norm": 0.30928221344947815,
+      "learning_rate": 0.0002,
+      "loss": 1.7339,
+      "step": 1360
+    },
+    {
+      "epoch": 0.9179229480737019,
+      "grad_norm": 0.30652928352355957,
+      "learning_rate": 0.0002,
+      "loss": 1.7479,
+      "step": 1370
+    },
+    {
+      "epoch": 0.9246231155778895,
+      "grad_norm": 0.3838157653808594,
+      "learning_rate": 0.0002,
+      "loss": 1.7491,
+      "step": 1380
+    },
+    {
+      "epoch": 0.931323283082077,
+      "grad_norm": 0.31655240058898926,
+      "learning_rate": 0.0002,
+      "loss": 1.7977,
+      "step": 1390
+    },
+    {
+      "epoch": 0.9380234505862647,
+      "grad_norm": 0.41737303137779236,
+      "learning_rate": 0.0002,
+      "loss": 1.8175,
+      "step": 1400
+    },
+    {
+      "epoch": 0.9447236180904522,
+      "grad_norm": 0.3227267861366272,
+      "learning_rate": 0.0002,
+      "loss": 1.6811,
+      "step": 1410
+    },
+    {
+      "epoch": 0.9514237855946399,
+      "grad_norm": 0.3729925751686096,
+      "learning_rate": 0.0002,
+      "loss": 1.7343,
+      "step": 1420
+    },
+    {
+      "epoch": 0.9581239530988275,
+      "grad_norm": 0.30779409408569336,
+      "learning_rate": 0.0002,
+      "loss": 1.8221,
+      "step": 1430
+    },
+    {
+      "epoch": 0.964824120603015,
+      "grad_norm": 0.334379643201828,
+      "learning_rate": 0.0002,
+      "loss": 1.7972,
+      "step": 1440
+    },
+    {
+      "epoch": 0.9715242881072027,
+      "grad_norm": 0.3568236231803894,
+      "learning_rate": 0.0002,
+      "loss": 1.7141,
+      "step": 1450
+    },
+    {
+      "epoch": 0.9782244556113903,
+      "grad_norm": 0.33310577273368835,
+      "learning_rate": 0.0002,
+      "loss": 1.7541,
+      "step": 1460
+    },
+    {
+      "epoch": 0.9849246231155779,
+      "grad_norm": 0.2972261905670166,
+      "learning_rate": 0.0002,
+      "loss": 1.8511,
+      "step": 1470
+    },
+    {
+      "epoch": 0.9916247906197655,
+      "grad_norm": 0.3322717845439911,
+      "learning_rate": 0.0002,
+      "loss": 1.7654,
+      "step": 1480
+    },
+    {
+      "epoch": 0.998324958123953,
+      "grad_norm": 0.3276330828666687,
+      "learning_rate": 0.0002,
+      "loss": 1.8033,
+      "step": 1490
+    },
+    {
+      "epoch": 0.9996649916247906,
+      "eval_loss": 1.8036354780197144,
+      "eval_runtime": 37.8949,
+      "eval_samples_per_second": 13.59,
+      "eval_steps_per_second": 1.715,
+      "step": 1492
+    },
+    {
+      "epoch": 1.0050251256281406,
+      "grad_norm": 0.29252371191978455,
+      "learning_rate": 0.0002,
+      "loss": 1.7138,
+      "step": 1500
+    },
+    {
+      "epoch": 1.0117252931323284,
+      "grad_norm": 0.31607162952423096,
+      "learning_rate": 0.0002,
+      "loss": 1.8198,
+      "step": 1510
+    },
+    {
+      "epoch": 1.018425460636516,
+      "grad_norm": 0.32294467091560364,
+      "learning_rate": 0.0002,
+      "loss": 1.6779,
+      "step": 1520
+    },
+    {
+      "epoch": 1.0251256281407035,
+      "grad_norm": 0.3868017792701721,
+      "learning_rate": 0.0002,
+      "loss": 1.7919,
+      "step": 1530
+    },
+    {
+      "epoch": 1.031825795644891,
+      "grad_norm": 0.3178282082080841,
+      "learning_rate": 0.0002,
+      "loss": 1.7954,
+      "step": 1540
+    },
+    {
+      "epoch": 1.0385259631490786,
+      "grad_norm": 0.3706750273704529,
+      "learning_rate": 0.0002,
+      "loss": 1.7136,
+      "step": 1550
+    },
+    {
+      "epoch": 1.0452261306532664,
+      "grad_norm": 0.33930912613868713,
+      "learning_rate": 0.0002,
+      "loss": 1.7382,
+      "step": 1560
+    },
+    {
+      "epoch": 1.051926298157454,
+      "grad_norm": 0.33970504999160767,
+      "learning_rate": 0.0002,
+      "loss": 1.7602,
+      "step": 1570
+    },
+    {
+      "epoch": 1.0586264656616415,
+      "grad_norm": 0.42553383111953735,
+      "learning_rate": 0.0002,
+      "loss": 1.6573,
+      "step": 1580
+    },
+    {
+      "epoch": 1.065326633165829,
+      "grad_norm": 0.3772421181201935,
+      "learning_rate": 0.0002,
+      "loss": 1.645,
+      "step": 1590
+    },
+    {
+      "epoch": 1.0720268006700167,
+      "grad_norm": 0.34212902188301086,
+      "learning_rate": 0.0002,
+      "loss": 1.7362,
+      "step": 1600
+    },
+    {
+      "epoch": 1.0787269681742044,
+      "grad_norm": 0.3798283338546753,
+      "learning_rate": 0.0002,
+      "loss": 1.7057,
+      "step": 1610
+    },
+    {
+      "epoch": 1.085427135678392,
+      "grad_norm": 0.36909598112106323,
+      "learning_rate": 0.0002,
+      "loss": 1.7468,
+      "step": 1620
+    },
+    {
+      "epoch": 1.0921273031825796,
+      "grad_norm": 0.3344230651855469,
+      "learning_rate": 0.0002,
+      "loss": 1.7807,
+      "step": 1630
+    },
+    {
+      "epoch": 1.0988274706867671,
+      "grad_norm": 0.3862569332122803,
+      "learning_rate": 0.0002,
+      "loss": 1.7111,
+      "step": 1640
+    },
+    {
+      "epoch": 1.1055276381909547,
+      "grad_norm": 0.31188511848449707,
+      "learning_rate": 0.0002,
+      "loss": 1.7163,
+      "step": 1650
+    },
+    {
+      "epoch": 1.1122278056951425,
+      "grad_norm": 0.3563670814037323,
+      "learning_rate": 0.0002,
+      "loss": 1.7263,
+      "step": 1660
+    },
+    {
+      "epoch": 1.11892797319933,
+      "grad_norm": 0.35052165389060974,
+      "learning_rate": 0.0002,
+      "loss": 1.7718,
+      "step": 1670
+    },
+    {
+      "epoch": 1.1256281407035176,
+      "grad_norm": 0.3285699188709259,
+      "learning_rate": 0.0002,
+      "loss": 1.7601,
+      "step": 1680
+    },
+    {
+      "epoch": 1.1323283082077051,
+      "grad_norm": 0.3639393746852875,
+      "learning_rate": 0.0002,
+      "loss": 1.6877,
+      "step": 1690
+    },
+    {
+      "epoch": 1.1390284757118927,
+      "grad_norm": 0.3842753767967224,
+      "learning_rate": 0.0002,
+      "loss": 1.7719,
+      "step": 1700
+    },
+    {
+      "epoch": 1.1457286432160805,
+      "grad_norm": 0.3624933063983917,
+      "learning_rate": 0.0002,
+      "loss": 1.7002,
+      "step": 1710
+    },
+    {
+      "epoch": 1.152428810720268,
+      "grad_norm": 0.3641220033168793,
+      "learning_rate": 0.0002,
+      "loss": 1.7243,
+      "step": 1720
+    },
+    {
+      "epoch": 1.1591289782244556,
+      "grad_norm": 0.32765355706214905,
+      "learning_rate": 0.0002,
+      "loss": 1.752,
+      "step": 1730
+    },
+    {
+      "epoch": 1.1658291457286432,
+      "grad_norm": 0.34974896907806396,
+      "learning_rate": 0.0002,
+      "loss": 1.6556,
+      "step": 1740
+    },
+    {
+      "epoch": 1.1725293132328307,
+      "grad_norm": 0.3910926580429077,
+      "learning_rate": 0.0002,
+      "loss": 1.7273,
+      "step": 1750
+    },
+    {
+      "epoch": 1.1792294807370185,
+      "grad_norm": 0.3564300537109375,
+      "learning_rate": 0.0002,
+      "loss": 1.7204,
+      "step": 1760
+    },
+    {
+      "epoch": 1.185929648241206,
+      "grad_norm": 0.34822574257850647,
+      "learning_rate": 0.0002,
+      "loss": 1.746,
+      "step": 1770
+    },
+    {
+      "epoch": 1.1926298157453936,
+      "grad_norm": 0.36185044050216675,
+      "learning_rate": 0.0002,
+      "loss": 1.7256,
+      "step": 1780
+    },
+    {
+      "epoch": 1.1993299832495812,
+      "grad_norm": 0.34866711497306824,
+      "learning_rate": 0.0002,
+      "loss": 1.6431,
+      "step": 1790
+    },
+    {
+      "epoch": 1.2060301507537687,
+      "grad_norm": 0.4017769992351532,
+      "learning_rate": 0.0002,
+      "loss": 1.8084,
+      "step": 1800
+    },
+    {
+      "epoch": 1.2127303182579565,
+      "grad_norm": 0.32930681109428406,
+      "learning_rate": 0.0002,
+      "loss": 1.6985,
+      "step": 1810
+    },
+    {
+      "epoch": 1.219430485762144,
+      "grad_norm": 0.35951921343803406,
+      "learning_rate": 0.0002,
+      "loss": 1.7606,
+      "step": 1820
+    },
+    {
+      "epoch": 1.2261306532663316,
+      "grad_norm": 0.37366992235183716,
+      "learning_rate": 0.0002,
+      "loss": 1.6933,
+      "step": 1830
+    },
+    {
+      "epoch": 1.2328308207705192,
+      "grad_norm": 0.3565689027309418,
+      "learning_rate": 0.0002,
+      "loss": 1.6737,
+      "step": 1840
+    },
+    {
+      "epoch": 1.2395309882747068,
+      "grad_norm": 0.3692343533039093,
+      "learning_rate": 0.0002,
+      "loss": 1.8013,
+      "step": 1850
+    },
+    {
+      "epoch": 1.2462311557788945,
+      "grad_norm": 0.38426971435546875,
+      "learning_rate": 0.0002,
+      "loss": 1.736,
+      "step": 1860
+    },
+    {
+      "epoch": 1.252931323283082,
+      "grad_norm": 0.33559855818748474,
+      "learning_rate": 0.0002,
+      "loss": 1.7031,
+      "step": 1870
+    },
+    {
+      "epoch": 1.2596314907872697,
+      "grad_norm": 0.34181106090545654,
+      "learning_rate": 0.0002,
+      "loss": 1.7033,
+      "step": 1880
+    },
+    {
+      "epoch": 1.2663316582914572,
+      "grad_norm": 0.3916318416595459,
+      "learning_rate": 0.0002,
+      "loss": 1.7707,
+      "step": 1890
+    },
+    {
+      "epoch": 1.2730318257956448,
+      "grad_norm": 0.3887825012207031,
+      "learning_rate": 0.0002,
+      "loss": 1.6686,
+      "step": 1900
+    },
+    {
+      "epoch": 1.2797319932998326,
+      "grad_norm": 0.33583927154541016,
+      "learning_rate": 0.0002,
+      "loss": 1.7062,
+      "step": 1910
+    },
+    {
+      "epoch": 1.2864321608040201,
+      "grad_norm": 0.37639349699020386,
+      "learning_rate": 0.0002,
+      "loss": 1.717,
+      "step": 1920
+    },
+    {
+      "epoch": 1.2931323283082077,
+      "grad_norm": 0.38059428334236145,
+      "learning_rate": 0.0002,
+      "loss": 1.777,
+      "step": 1930
+    },
+    {
+      "epoch": 1.2998324958123952,
+      "grad_norm": 0.37253183126449585,
+      "learning_rate": 0.0002,
+      "loss": 1.6126,
+      "step": 1940
+    },
+    {
+      "epoch": 1.3065326633165828,
+      "grad_norm": 0.37371566891670227,
+      "learning_rate": 0.0002,
+      "loss": 1.6758,
+      "step": 1950
+    },
+    {
+      "epoch": 1.3132328308207706,
+      "grad_norm": 0.4080910086631775,
+      "learning_rate": 0.0002,
+      "loss": 1.6788,
+      "step": 1960
+    },
+    {
+      "epoch": 1.3199329983249581,
+      "grad_norm": 0.3174354135990143,
+      "learning_rate": 0.0002,
+      "loss": 1.6518,
+      "step": 1970
+    },
+    {
+      "epoch": 1.3266331658291457,
+      "grad_norm": 0.4518888294696808,
+      "learning_rate": 0.0002,
+      "loss": 1.7925,
+      "step": 1980
+    },
+    {
+      "epoch": 1.3333333333333333,
+      "grad_norm": 0.3627921938896179,
+      "learning_rate": 0.0002,
+      "loss": 1.7085,
+      "step": 1990
+    },
+    {
+      "epoch": 1.3400335008375208,
+      "grad_norm": 0.3655930161476135,
+      "learning_rate": 0.0002,
+      "loss": 1.7676,
+      "step": 2000
+    },
+    {
+      "epoch": 1.3467336683417086,
+      "grad_norm": 0.3509993255138397,
+      "learning_rate": 0.0002,
+      "loss": 1.7016,
+      "step": 2010
+    },
+    {
+      "epoch": 1.3534338358458962,
+      "grad_norm": 0.4281129240989685,
+      "learning_rate": 0.0002,
+      "loss": 1.7359,
+      "step": 2020
+    },
+    {
+      "epoch": 1.3601340033500837,
+      "grad_norm": 0.3821414113044739,
+      "learning_rate": 0.0002,
+      "loss": 1.6884,
+      "step": 2030
+    },
+    {
+      "epoch": 1.3668341708542713,
+      "grad_norm": 0.3907586336135864,
+      "learning_rate": 0.0002,
+      "loss": 1.7075,
+      "step": 2040
+    },
+    {
+      "epoch": 1.3735343383584588,
+      "grad_norm": 0.37792932987213135,
+      "learning_rate": 0.0002,
+      "loss": 1.7424,
+      "step": 2050
+    },
+    {
+      "epoch": 1.3802345058626466,
+      "grad_norm": 0.3693985641002655,
+      "learning_rate": 0.0002,
+      "loss": 1.7305,
+      "step": 2060
+    },
+    {
+      "epoch": 1.3869346733668342,
+      "grad_norm": 0.32275936007499695,
+      "learning_rate": 0.0002,
+      "loss": 1.7434,
+      "step": 2070
+    },
+    {
+      "epoch": 1.3936348408710217,
+      "grad_norm": 0.3789440095424652,
+      "learning_rate": 0.0002,
+      "loss": 1.6677,
+      "step": 2080
+    },
+    {
+      "epoch": 1.4003350083752093,
+      "grad_norm": 0.3638380467891693,
+      "learning_rate": 0.0002,
+      "loss": 1.6825,
+      "step": 2090
+    },
+    {
+      "epoch": 1.4070351758793969,
+      "grad_norm": 0.3495481610298157,
+      "learning_rate": 0.0002,
+      "loss": 1.6542,
+      "step": 2100
+    },
+    {
+      "epoch": 1.4137353433835846,
+      "grad_norm": 0.37920597195625305,
+      "learning_rate": 0.0002,
+      "loss": 1.7225,
+      "step": 2110
+    },
+    {
+      "epoch": 1.4204355108877722,
+      "grad_norm": 0.37218064069747925,
+      "learning_rate": 0.0002,
+      "loss": 1.7329,
+      "step": 2120
+    },
+    {
+      "epoch": 1.4271356783919598,
+      "grad_norm": 0.38074082136154175,
+      "learning_rate": 0.0002,
+      "loss": 1.799,
+      "step": 2130
+    },
+    {
+      "epoch": 1.4338358458961473,
+      "grad_norm": 0.3455527126789093,
+      "learning_rate": 0.0002,
+      "loss": 1.7403,
+      "step": 2140
+    },
+    {
+      "epoch": 1.4405360134003349,
+      "grad_norm": 0.3712003529071808,
+      "learning_rate": 0.0002,
+      "loss": 1.776,
+      "step": 2150
+    },
+    {
+      "epoch": 1.4472361809045227,
+      "grad_norm": 0.3786754906177521,
+      "learning_rate": 0.0002,
+      "loss": 1.7619,
+      "step": 2160
+    },
+    {
+      "epoch": 1.4539363484087102,
+      "grad_norm": 0.3879223167896271,
+      "learning_rate": 0.0002,
+      "loss": 1.68,
+      "step": 2170
+    },
+    {
+      "epoch": 1.4606365159128978,
+      "grad_norm": 0.38738805055618286,
+      "learning_rate": 0.0002,
+      "loss": 1.7,
+      "step": 2180
+    },
+    {
+      "epoch": 1.4673366834170856,
+      "grad_norm": 0.39768800139427185,
+      "learning_rate": 0.0002,
+      "loss": 1.7581,
+      "step": 2190
+    },
+    {
+      "epoch": 1.474036850921273,
+      "grad_norm": 0.4172441065311432,
+      "learning_rate": 0.0002,
+      "loss": 1.7671,
+      "step": 2200
+    },
+    {
+      "epoch": 1.4807370184254607,
+      "grad_norm": 0.4043174982070923,
+      "learning_rate": 0.0002,
+      "loss": 1.6736,
+      "step": 2210
+    },
+    {
+      "epoch": 1.4874371859296482,
+      "grad_norm": 0.3750883936882019,
+      "learning_rate": 0.0002,
+      "loss": 1.7444,
+      "step": 2220
+    },
+    {
+      "epoch": 1.4941373534338358,
+      "grad_norm": 0.3552253246307373,
+      "learning_rate": 0.0002,
+      "loss": 1.6861,
+      "step": 2230
+    },
+    {
+      "epoch": 1.5008375209380236,
+      "grad_norm": 0.34607139229774475,
+      "learning_rate": 0.0002,
+      "loss": 1.6471,
+      "step": 2240
+    },
+    {
+      "epoch": 1.507537688442211,
+      "grad_norm": 0.3406706750392914,
+      "learning_rate": 0.0002,
+      "loss": 1.6962,
+      "step": 2250
+    },
+    {
+      "epoch": 1.5142378559463987,
+      "grad_norm": 0.36654895544052124,
+      "learning_rate": 0.0002,
+      "loss": 1.7694,
+      "step": 2260
+    },
+    {
+      "epoch": 1.5209380234505863,
+      "grad_norm": 0.3914054334163666,
+      "learning_rate": 0.0002,
+      "loss": 1.6812,
+      "step": 2270
+    },
+    {
+      "epoch": 1.5276381909547738,
+      "grad_norm": 0.42012137174606323,
+      "learning_rate": 0.0002,
+      "loss": 1.6822,
+      "step": 2280
+    },
+    {
+      "epoch": 1.5343383584589616,
+      "grad_norm": 0.39563435316085815,
+      "learning_rate": 0.0002,
+      "loss": 1.697,
+      "step": 2290
+    },
+    {
+      "epoch": 1.541038525963149,
+      "grad_norm": 0.3508438766002655,
+      "learning_rate": 0.0002,
+      "loss": 1.7491,
+      "step": 2300
+    },
+    {
+      "epoch": 1.5477386934673367,
+      "grad_norm": 0.3785218596458435,
+      "learning_rate": 0.0002,
+      "loss": 1.7727,
+      "step": 2310
+    },
+    {
+      "epoch": 1.5544388609715243,
+      "grad_norm": 0.39377647638320923,
+      "learning_rate": 0.0002,
+      "loss": 1.6963,
+      "step": 2320
+    },
+    {
+      "epoch": 1.5611390284757118,
+      "grad_norm": 0.3391438126564026,
+      "learning_rate": 0.0002,
+      "loss": 1.7263,
+      "step": 2330
+    },
+    {
+      "epoch": 1.5678391959798996,
+      "grad_norm": 0.37944263219833374,
+      "learning_rate": 0.0002,
+      "loss": 1.7722,
+      "step": 2340
+    },
+    {
+      "epoch": 1.574539363484087,
+      "grad_norm": 0.3523491322994232,
+      "learning_rate": 0.0002,
+      "loss": 1.6371,
+      "step": 2350
+    },
+    {
+      "epoch": 1.5812395309882747,
+      "grad_norm": 0.3911575973033905,
+      "learning_rate": 0.0002,
+      "loss": 1.7583,
+      "step": 2360
+    },
+    {
+      "epoch": 1.5879396984924623,
+      "grad_norm": 0.33832186460494995,
+      "learning_rate": 0.0002,
+      "loss": 1.7117,
+      "step": 2370
+    },
+    {
+      "epoch": 1.5946398659966499,
+      "grad_norm": 0.3665979206562042,
+      "learning_rate": 0.0002,
+      "loss": 1.7701,
+      "step": 2380
+    },
+    {
+      "epoch": 1.6013400335008376,
+      "grad_norm": 0.3871748149394989,
+      "learning_rate": 0.0002,
+      "loss": 1.779,
+      "step": 2390
+    },
+    {
+      "epoch": 1.608040201005025,
+      "grad_norm": 0.3586967885494232,
+      "learning_rate": 0.0002,
+      "loss": 1.7109,
+      "step": 2400
+    },
+    {
+      "epoch": 1.6147403685092128,
+      "grad_norm": 0.3563673198223114,
+      "learning_rate": 0.0002,
+      "loss": 1.7096,
+      "step": 2410
+    },
+    {
+      "epoch": 1.6214405360134003,
+      "grad_norm": 0.37588971853256226,
+      "learning_rate": 0.0002,
+      "loss": 1.745,
+      "step": 2420
+    },
+    {
+      "epoch": 1.6281407035175879,
+      "grad_norm": 0.352556437253952,
+      "learning_rate": 0.0002,
+      "loss": 1.7086,
+      "step": 2430
+    },
+    {
+      "epoch": 1.6348408710217757,
+      "grad_norm": 0.3716259300708771,
+      "learning_rate": 0.0002,
+      "loss": 1.6547,
+      "step": 2440
+    },
+    {
+      "epoch": 1.641541038525963,
+      "grad_norm": 0.372001975774765,
+      "learning_rate": 0.0002,
+      "loss": 1.7033,
+      "step": 2450
+    },
+    {
+      "epoch": 1.6482412060301508,
+      "grad_norm": 0.3430042862892151,
+      "learning_rate": 0.0002,
+      "loss": 1.6584,
+      "step": 2460
+    },
+    {
+      "epoch": 1.6549413735343383,
+      "grad_norm": 0.3741483688354492,
+      "learning_rate": 0.0002,
+      "loss": 1.7217,
+      "step": 2470
+    },
+    {
+      "epoch": 1.661641541038526,
+      "grad_norm": 0.3610571324825287,
+      "learning_rate": 0.0002,
+      "loss": 1.7701,
+      "step": 2480
+    },
+    {
+      "epoch": 1.6683417085427137,
+      "grad_norm": 0.4204719066619873,
+      "learning_rate": 0.0002,
+      "loss": 1.7057,
+      "step": 2490
+    },
+    {
+      "epoch": 1.675041876046901,
+      "grad_norm": 0.3938186466693878,
+      "learning_rate": 0.0002,
+      "loss": 1.7954,
+      "step": 2500
+    },
+    {
+      "epoch": 1.6817420435510888,
+      "grad_norm": 0.3421435058116913,
+      "learning_rate": 0.0002,
+      "loss": 1.6633,
+      "step": 2510
+    },
+    {
+      "epoch": 1.6884422110552764,
+      "grad_norm": 0.42441412806510925,
+      "learning_rate": 0.0002,
+      "loss": 1.7996,
+      "step": 2520
+    },
+    {
+      "epoch": 1.695142378559464,
+      "grad_norm": 0.38071519136428833,
+      "learning_rate": 0.0002,
+      "loss": 1.7142,
+      "step": 2530
+    },
+    {
+      "epoch": 1.7018425460636517,
+      "grad_norm": 0.34078919887542725,
+      "learning_rate": 0.0002,
+      "loss": 1.7232,
+      "step": 2540
+    },
+    {
+      "epoch": 1.708542713567839,
+      "grad_norm": 0.412844181060791,
+      "learning_rate": 0.0002,
+      "loss": 1.7126,
+      "step": 2550
+    },
+    {
+      "epoch": 1.7152428810720268,
+      "grad_norm": 0.3753604292869568,
+      "learning_rate": 0.0002,
+      "loss": 1.7149,
+      "step": 2560
+    },
+    {
+      "epoch": 1.7219430485762144,
+      "grad_norm": 0.41588476300239563,
+      "learning_rate": 0.0002,
+      "loss": 1.7011,
+      "step": 2570
+    },
+    {
+      "epoch": 1.728643216080402,
+      "grad_norm": 0.35504111647605896,
+      "learning_rate": 0.0002,
+      "loss": 1.6427,
+      "step": 2580
+    },
+    {
+      "epoch": 1.7353433835845897,
+      "grad_norm": 0.36909720301628113,
+      "learning_rate": 0.0002,
+      "loss": 1.7296,
+      "step": 2590
+    },
+    {
+      "epoch": 1.742043551088777,
+      "grad_norm": 0.4149979054927826,
+      "learning_rate": 0.0002,
+      "loss": 1.7022,
+      "step": 2600
+    },
+    {
+      "epoch": 1.7487437185929648,
+      "grad_norm": 0.38859328627586365,
+      "learning_rate": 0.0002,
+      "loss": 1.77,
+      "step": 2610
+    },
+    {
+      "epoch": 1.7554438860971524,
+      "grad_norm": 0.36738792061805725,
+      "learning_rate": 0.0002,
+      "loss": 1.7036,
+      "step": 2620
+    },
+    {
+      "epoch": 1.76214405360134,
+      "grad_norm": 0.3968178927898407,
+      "learning_rate": 0.0002,
+      "loss": 1.764,
+      "step": 2630
+    },
+    {
+      "epoch": 1.7688442211055277,
+      "grad_norm": 0.3972901999950409,
+      "learning_rate": 0.0002,
+      "loss": 1.7687,
+      "step": 2640
+    },
+    {
+      "epoch": 1.775544388609715,
+      "grad_norm": 0.3949959874153137,
+      "learning_rate": 0.0002,
+      "loss": 1.6774,
+      "step": 2650
+    },
+    {
+      "epoch": 1.7822445561139029,
+      "grad_norm": 0.44074657559394836,
+      "learning_rate": 0.0002,
+      "loss": 1.7247,
+      "step": 2660
+    },
+    {
+      "epoch": 1.7889447236180904,
+      "grad_norm": 0.39743664860725403,
+      "learning_rate": 0.0002,
+      "loss": 1.7188,
+      "step": 2670
+    },
+    {
+      "epoch": 1.795644891122278,
+      "grad_norm": 0.3950406610965729,
+      "learning_rate": 0.0002,
+      "loss": 1.7258,
+      "step": 2680
+    },
+    {
+      "epoch": 1.8023450586264658,
+      "grad_norm": 0.3568263649940491,
+      "learning_rate": 0.0002,
+      "loss": 1.6906,
+      "step": 2690
+    },
+    {
+      "epoch": 1.809045226130653,
+      "grad_norm": 0.3819476366043091,
+      "learning_rate": 0.0002,
+      "loss": 1.6735,
+      "step": 2700
+    },
+    {
+      "epoch": 1.8157453936348409,
+      "grad_norm": 0.3480634391307831,
+      "learning_rate": 0.0002,
+      "loss": 1.7198,
+      "step": 2710
+    },
+    {
+      "epoch": 1.8224455611390284,
+      "grad_norm": 0.3875853419303894,
+      "learning_rate": 0.0002,
+      "loss": 1.7042,
+      "step": 2720
+    },
+    {
+      "epoch": 1.829145728643216,
+      "grad_norm": 0.3441337049007416,
+      "learning_rate": 0.0002,
+      "loss": 1.6988,
+      "step": 2730
+    },
+    {
+      "epoch": 1.8358458961474038,
+      "grad_norm": 0.35692882537841797,
+      "learning_rate": 0.0002,
+      "loss": 1.7647,
+      "step": 2740
+    },
+    {
+      "epoch": 1.8425460636515911,
+      "grad_norm": 0.36959215998649597,
+      "learning_rate": 0.0002,
+      "loss": 1.7033,
+      "step": 2750
+    },
+    {
+      "epoch": 1.849246231155779,
+      "grad_norm": 0.3893393278121948,
+      "learning_rate": 0.0002,
+      "loss": 1.7657,
+      "step": 2760
+    },
+    {
+      "epoch": 1.8559463986599665,
+      "grad_norm": 0.37817293405532837,
+      "learning_rate": 0.0002,
+      "loss": 1.7068,
+      "step": 2770
+    },
+    {
+      "epoch": 1.862646566164154,
+      "grad_norm": 0.36071285605430603,
+      "learning_rate": 0.0002,
+      "loss": 1.761,
+      "step": 2780
+    },
+    {
+      "epoch": 1.8693467336683418,
+      "grad_norm": 0.3758420944213867,
+      "learning_rate": 0.0002,
+      "loss": 1.7623,
+      "step": 2790
+    },
+    {
+      "epoch": 1.8760469011725294,
+      "grad_norm": 0.3889938294887543,
+      "learning_rate": 0.0002,
+      "loss": 1.6743,
+      "step": 2800
+    },
+    {
+      "epoch": 1.882747068676717,
+      "grad_norm": 0.34361857175827026,
+      "learning_rate": 0.0002,
+      "loss": 1.6151,
+      "step": 2810
+    },
+    {
+      "epoch": 1.8894472361809045,
+      "grad_norm": 0.39283323287963867,
+      "learning_rate": 0.0002,
+      "loss": 1.6038,
+      "step": 2820
+    },
+    {
+      "epoch": 1.896147403685092,
+      "grad_norm": 0.3919452726840973,
+      "learning_rate": 0.0002,
+      "loss": 1.7555,
+      "step": 2830
+    },
+    {
+      "epoch": 1.9028475711892798,
+      "grad_norm": 0.38215070962905884,
+      "learning_rate": 0.0002,
+      "loss": 1.673,
+      "step": 2840
+    },
+    {
+      "epoch": 1.9095477386934674,
+      "grad_norm": 0.4235064387321472,
+      "learning_rate": 0.0002,
+      "loss": 1.7044,
+      "step": 2850
+    },
+    {
+      "epoch": 1.916247906197655,
+      "grad_norm": 0.35694634914398193,
+      "learning_rate": 0.0002,
+      "loss": 1.7123,
+      "step": 2860
+    },
+    {
+      "epoch": 1.9229480737018425,
+      "grad_norm": 0.383492112159729,
+      "learning_rate": 0.0002,
+      "loss": 1.8128,
+      "step": 2870
+    },
+    {
+      "epoch": 1.92964824120603,
+      "grad_norm": 0.5945147275924683,
+      "learning_rate": 0.0002,
+      "loss": 1.7581,
+      "step": 2880
+    },
+    {
+      "epoch": 1.9363484087102178,
+      "grad_norm": 0.3367522358894348,
+      "learning_rate": 0.0002,
+      "loss": 1.7421,
+      "step": 2890
+    },
+    {
+      "epoch": 1.9430485762144054,
+      "grad_norm": 0.35300394892692566,
+      "learning_rate": 0.0002,
+      "loss": 1.6561,
+      "step": 2900
+    },
+    {
+      "epoch": 1.949748743718593,
+      "grad_norm": 0.38084495067596436,
+      "learning_rate": 0.0002,
+      "loss": 1.7033,
+      "step": 2910
+    },
+    {
+      "epoch": 1.9564489112227805,
+      "grad_norm": 0.37559160590171814,
+      "learning_rate": 0.0002,
+      "loss": 1.7132,
+      "step": 2920
+    },
+    {
+      "epoch": 1.963149078726968,
+      "grad_norm": 0.3661738336086273,
+      "learning_rate": 0.0002,
+      "loss": 1.6759,
+      "step": 2930
+    },
+    {
+      "epoch": 1.9698492462311559,
+      "grad_norm": 0.4073849320411682,
+      "learning_rate": 0.0002,
+      "loss": 1.7643,
+      "step": 2940
+    },
+    {
+      "epoch": 1.9765494137353434,
+      "grad_norm": 0.3723304271697998,
+      "learning_rate": 0.0002,
+      "loss": 1.6806,
+      "step": 2950
+    },
+    {
+      "epoch": 1.983249581239531,
+      "grad_norm": 0.3991098999977112,
+      "learning_rate": 0.0002,
+      "loss": 1.7611,
+      "step": 2960
+    },
+    {
+      "epoch": 1.9899497487437185,
+      "grad_norm": 0.3947085440158844,
+      "learning_rate": 0.0002,
+      "loss": 1.7263,
+      "step": 2970
+    },
+    {
+      "epoch": 1.996649916247906,
+      "grad_norm": 0.3786258399486542,
+      "learning_rate": 0.0002,
+      "loss": 1.7217,
+      "step": 2980
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 1.8028968572616577,
+      "eval_runtime": 37.8985,
+      "eval_samples_per_second": 13.589,
+      "eval_steps_per_second": 1.715,
+      "step": 2985
+    },
+    {
+      "epoch": 2.003350083752094,
+      "grad_norm": 0.34824079275131226,
+      "learning_rate": 0.0002,
+      "loss": 1.695,
+      "step": 2990
+    },
+    {
+      "epoch": 2.0100502512562812,
+      "grad_norm": 0.3394894003868103,
+      "learning_rate": 0.0002,
+      "loss": 1.5853,
+      "step": 3000
+    },
+    {
+      "epoch": 2.016750418760469,
+      "grad_norm": 0.36910977959632874,
+      "learning_rate": 0.0002,
+      "loss": 1.5783,
+      "step": 3010
+    },
+    {
+      "epoch": 2.023450586264657,
+      "grad_norm": 0.45000967383384705,
+      "learning_rate": 0.0002,
+      "loss": 1.6105,
+      "step": 3020
+    },
+    {
+      "epoch": 2.030150753768844,
+      "grad_norm": 0.3791407346725464,
+      "learning_rate": 0.0002,
+      "loss": 1.6019,
+      "step": 3030
+    },
+    {
+      "epoch": 2.036850921273032,
+      "grad_norm": 0.387321799993515,
+      "learning_rate": 0.0002,
+      "loss": 1.5832,
+      "step": 3040
+    },
+    {
+      "epoch": 2.0435510887772192,
+      "grad_norm": 0.4185757040977478,
+      "learning_rate": 0.0002,
+      "loss": 1.6834,
+      "step": 3050
+    },
+    {
+      "epoch": 2.050251256281407,
+      "grad_norm": 0.45110777020454407,
+      "learning_rate": 0.0002,
+      "loss": 1.5696,
+      "step": 3060
+    },
+    {
+      "epoch": 2.056951423785595,
+      "grad_norm": 0.42663660645484924,
+      "learning_rate": 0.0002,
+      "loss": 1.6231,
+      "step": 3070
+    },
+    {
+      "epoch": 2.063651591289782,
+      "grad_norm": 0.4546292722225189,
+      "learning_rate": 0.0002,
+      "loss": 1.6279,
+      "step": 3080
+    },
+    {
+      "epoch": 2.07035175879397,
+      "grad_norm": 0.3979759216308594,
+      "learning_rate": 0.0002,
+      "loss": 1.6141,
+      "step": 3090
+    },
+    {
+      "epoch": 2.0770519262981573,
+      "grad_norm": 0.43596673011779785,
+      "learning_rate": 0.0002,
+      "loss": 1.6343,
+      "step": 3100
+    },
+    {
+      "epoch": 2.083752093802345,
+      "grad_norm": 0.40120232105255127,
+      "learning_rate": 0.0002,
+      "loss": 1.5441,
+      "step": 3110
+    },
+    {
+      "epoch": 2.090452261306533,
+      "grad_norm": 0.44449281692504883,
+      "learning_rate": 0.0002,
+      "loss": 1.6309,
+      "step": 3120
+    },
+    {
+      "epoch": 2.09715242881072,
+      "grad_norm": 0.42672568559646606,
+      "learning_rate": 0.0002,
+      "loss": 1.5652,
+      "step": 3130
+    },
+    {
+      "epoch": 2.103852596314908,
+      "grad_norm": 0.4232690930366516,
+      "learning_rate": 0.0002,
+      "loss": 1.682,
+      "step": 3140
+    },
+    {
+      "epoch": 2.1105527638190953,
+      "grad_norm": 0.4299317002296448,
+      "learning_rate": 0.0002,
+      "loss": 1.624,
+      "step": 3150
+    },
+    {
+      "epoch": 2.117252931323283,
+      "grad_norm": 0.4067758023738861,
+      "learning_rate": 0.0002,
+      "loss": 1.6766,
+      "step": 3160
+    },
+    {
+      "epoch": 2.123953098827471,
+      "grad_norm": 0.4918815791606903,
+      "learning_rate": 0.0002,
+      "loss": 1.6759,
+      "step": 3170
+    },
+    {
+      "epoch": 2.130653266331658,
+      "grad_norm": 0.4140559732913971,
+      "learning_rate": 0.0002,
+      "loss": 1.6478,
+      "step": 3180
+    },
+    {
+      "epoch": 2.137353433835846,
+      "grad_norm": 0.4555995464324951,
+      "learning_rate": 0.0002,
+      "loss": 1.6641,
+      "step": 3190
+    },
+    {
+      "epoch": 2.1440536013400333,
+      "grad_norm": 0.42943915724754333,
+      "learning_rate": 0.0002,
+      "loss": 1.5888,
+      "step": 3200
+    },
+    {
+      "epoch": 2.150753768844221,
+      "grad_norm": 0.4730435013771057,
+      "learning_rate": 0.0002,
+      "loss": 1.5886,
+      "step": 3210
+    },
+    {
+      "epoch": 2.157453936348409,
+      "grad_norm": 0.43310216069221497,
+      "learning_rate": 0.0002,
+      "loss": 1.6022,
+      "step": 3220
+    },
+    {
+      "epoch": 2.164154103852596,
+      "grad_norm": 0.42054110765457153,
+      "learning_rate": 0.0002,
+      "loss": 1.6058,
+      "step": 3230
+    },
+    {
+      "epoch": 2.170854271356784,
+      "grad_norm": 0.4897233247756958,
+      "learning_rate": 0.0002,
+      "loss": 1.6749,
+      "step": 3240
+    },
+    {
+      "epoch": 2.1775544388609713,
+      "grad_norm": 0.42194533348083496,
+      "learning_rate": 0.0002,
+      "loss": 1.6983,
+      "step": 3250
+    },
+    {
+      "epoch": 2.184254606365159,
+      "grad_norm": 0.44494450092315674,
+      "learning_rate": 0.0002,
+      "loss": 1.6159,
+      "step": 3260
+    },
+    {
+      "epoch": 2.190954773869347,
+      "grad_norm": 0.43524879217147827,
+      "learning_rate": 0.0002,
+      "loss": 1.6977,
+      "step": 3270
+    },
+    {
+      "epoch": 2.1976549413735342,
+      "grad_norm": 0.4621117413043976,
+      "learning_rate": 0.0002,
+      "loss": 1.528,
+      "step": 3280
+    },
+    {
+      "epoch": 2.204355108877722,
+      "grad_norm": 0.4073285460472107,
+      "learning_rate": 0.0002,
+      "loss": 1.632,
+      "step": 3290
+    },
+    {
+      "epoch": 2.2110552763819094,
+      "grad_norm": 0.47868335247039795,
+      "learning_rate": 0.0002,
+      "loss": 1.6141,
+      "step": 3300
+    },
+    {
+      "epoch": 2.217755443886097,
+      "grad_norm": 0.4264970123767853,
+      "learning_rate": 0.0002,
+      "loss": 1.6857,
+      "step": 3310
+    },
+    {
+      "epoch": 2.224455611390285,
+      "grad_norm": 0.4491245150566101,
+      "learning_rate": 0.0002,
+      "loss": 1.5653,
+      "step": 3320
+    },
+    {
+      "epoch": 2.2311557788944723,
+      "grad_norm": 0.4010344445705414,
+      "learning_rate": 0.0002,
+      "loss": 1.5881,
+      "step": 3330
+    },
+    {
+      "epoch": 2.23785594639866,
+      "grad_norm": 0.4232759177684784,
+      "learning_rate": 0.0002,
+      "loss": 1.6684,
+      "step": 3340
+    },
+    {
+      "epoch": 2.2445561139028474,
+      "grad_norm": 0.5099776983261108,
+      "learning_rate": 0.0002,
+      "loss": 1.6336,
+      "step": 3350
+    },
+    {
+      "epoch": 2.251256281407035,
+      "grad_norm": 0.5223407745361328,
+      "learning_rate": 0.0002,
+      "loss": 1.6764,
+      "step": 3360
+    },
+    {
+      "epoch": 2.257956448911223,
+      "grad_norm": 0.47818470001220703,
+      "learning_rate": 0.0002,
+      "loss": 1.6625,
+      "step": 3370
+    },
+    {
+      "epoch": 2.2646566164154103,
+      "grad_norm": 0.4721255898475647,
+      "learning_rate": 0.0002,
+      "loss": 1.5946,
+      "step": 3380
+    },
+    {
+      "epoch": 2.271356783919598,
+      "grad_norm": 0.4113229513168335,
+      "learning_rate": 0.0002,
+      "loss": 1.5568,
+      "step": 3390
+    },
+    {
+      "epoch": 2.2780569514237854,
+      "grad_norm": 0.507080078125,
+      "learning_rate": 0.0002,
+      "loss": 1.6494,
+      "step": 3400
+    },
+    {
+      "epoch": 2.284757118927973,
+      "grad_norm": 0.4852292239665985,
+      "learning_rate": 0.0002,
+      "loss": 1.6183,
+      "step": 3410
+    },
+    {
+      "epoch": 2.291457286432161,
+      "grad_norm": 0.4503684341907501,
+      "learning_rate": 0.0002,
+      "loss": 1.6132,
+      "step": 3420
+    },
+    {
+      "epoch": 2.2981574539363483,
+      "grad_norm": 0.8359600305557251,
+      "learning_rate": 0.0002,
+      "loss": 1.6649,
+      "step": 3430
+    },
+    {
+      "epoch": 2.304857621440536,
+      "grad_norm": 0.44604045152664185,
+      "learning_rate": 0.0002,
+      "loss": 1.6644,
+      "step": 3440
+    },
+    {
+      "epoch": 2.3115577889447234,
+      "grad_norm": 0.45667049288749695,
+      "learning_rate": 0.0002,
+      "loss": 1.5972,
+      "step": 3450
+    },
+    {
+      "epoch": 2.318257956448911,
+      "grad_norm": 0.4879349172115326,
+      "learning_rate": 0.0002,
+      "loss": 1.6667,
+      "step": 3460
+    },
+    {
+      "epoch": 2.324958123953099,
+      "grad_norm": 0.4033963084220886,
+      "learning_rate": 0.0002,
+      "loss": 1.5804,
+      "step": 3470
+    },
+    {
+      "epoch": 2.3316582914572863,
+      "grad_norm": 0.44494301080703735,
+      "learning_rate": 0.0002,
+      "loss": 1.5838,
+      "step": 3480
+    },
+    {
+      "epoch": 2.338358458961474,
+      "grad_norm": 0.4794621765613556,
+      "learning_rate": 0.0002,
+      "loss": 1.5951,
+      "step": 3490
+    },
+    {
+      "epoch": 2.3450586264656614,
+      "grad_norm": 0.41404327750205994,
+      "learning_rate": 0.0002,
+      "loss": 1.6807,
+      "step": 3500
+    },
+    {
+      "epoch": 2.351758793969849,
+      "grad_norm": 0.4664851725101471,
+      "learning_rate": 0.0002,
+      "loss": 1.714,
+      "step": 3510
+    },
+    {
+      "epoch": 2.358458961474037,
+      "grad_norm": 0.4263697564601898,
+      "learning_rate": 0.0002,
+      "loss": 1.6537,
+      "step": 3520
+    },
+    {
+      "epoch": 2.3651591289782243,
+      "grad_norm": 0.5035167336463928,
+      "learning_rate": 0.0002,
+      "loss": 1.6551,
+      "step": 3530
+    },
+    {
+      "epoch": 2.371859296482412,
+      "grad_norm": 0.4380664527416229,
+      "learning_rate": 0.0002,
+      "loss": 1.6208,
+      "step": 3540
+    },
+    {
+      "epoch": 2.3785594639865995,
+      "grad_norm": 0.5227681994438171,
+      "learning_rate": 0.0002,
+      "loss": 1.634,
+      "step": 3550
+    },
+    {
+      "epoch": 2.3852596314907872,
+      "grad_norm": 0.4382302761077881,
+      "learning_rate": 0.0002,
+      "loss": 1.6146,
+      "step": 3560
+    },
+    {
+      "epoch": 2.391959798994975,
+      "grad_norm": 0.4392451047897339,
+      "learning_rate": 0.0002,
+      "loss": 1.5653,
+      "step": 3570
+    },
+    {
+      "epoch": 2.3986599664991624,
+      "grad_norm": 0.4372786581516266,
+      "learning_rate": 0.0002,
+      "loss": 1.6626,
+      "step": 3580
+    },
+    {
+      "epoch": 2.40536013400335,
+      "grad_norm": 0.5015502572059631,
+      "learning_rate": 0.0002,
+      "loss": 1.519,
+      "step": 3590
+    },
+    {
+      "epoch": 2.4120603015075375,
+      "grad_norm": 0.5653210878372192,
+      "learning_rate": 0.0002,
+      "loss": 1.588,
+      "step": 3600
+    },
+    {
+      "epoch": 2.4187604690117253,
+      "grad_norm": 0.53007972240448,
+      "learning_rate": 0.0002,
+      "loss": 1.6075,
+      "step": 3610
+    },
+    {
+      "epoch": 2.425460636515913,
+      "grad_norm": 0.4659176766872406,
+      "learning_rate": 0.0002,
+      "loss": 1.6421,
+      "step": 3620
+    },
+    {
+      "epoch": 2.4321608040201004,
+      "grad_norm": 0.5637837052345276,
+      "learning_rate": 0.0002,
+      "loss": 1.625,
+      "step": 3630
+    },
+    {
+      "epoch": 2.438860971524288,
+      "grad_norm": 0.4248391389846802,
+      "learning_rate": 0.0002,
+      "loss": 1.6168,
+      "step": 3640
+    },
+    {
+      "epoch": 2.4455611390284755,
+      "grad_norm": 0.44668248295783997,
+      "learning_rate": 0.0002,
+      "loss": 1.6822,
+      "step": 3650
+    },
+    {
+      "epoch": 2.4522613065326633,
+      "grad_norm": 0.43990179896354675,
+      "learning_rate": 0.0002,
+      "loss": 1.6417,
+      "step": 3660
+    },
+    {
+      "epoch": 2.458961474036851,
+      "grad_norm": 0.4532523453235626,
+      "learning_rate": 0.0002,
+      "loss": 1.6723,
+      "step": 3670
+    },
+    {
+      "epoch": 2.4656616415410384,
+      "grad_norm": 0.6605591773986816,
+      "learning_rate": 0.0002,
+      "loss": 1.6957,
+      "step": 3680
+    },
+    {
+      "epoch": 2.472361809045226,
+      "grad_norm": 0.4694533348083496,
+      "learning_rate": 0.0002,
+      "loss": 1.6159,
+      "step": 3690
+    },
+    {
+      "epoch": 2.4790619765494135,
+      "grad_norm": 0.4485011100769043,
+      "learning_rate": 0.0002,
+      "loss": 1.6239,
+      "step": 3700
+    },
+    {
+      "epoch": 2.4857621440536013,
+      "grad_norm": 0.4761785864830017,
+      "learning_rate": 0.0002,
+      "loss": 1.6834,
+      "step": 3710
+    },
+    {
+      "epoch": 2.492462311557789,
+      "grad_norm": 0.5116432309150696,
+      "learning_rate": 0.0002,
+      "loss": 1.6313,
+      "step": 3720
+    },
+    {
+      "epoch": 2.4991624790619764,
+      "grad_norm": 0.49523618817329407,
+      "learning_rate": 0.0002,
+      "loss": 1.5054,
+      "step": 3730
+    },
+    {
+      "epoch": 2.505862646566164,
+      "grad_norm": 0.43826380372047424,
+      "learning_rate": 0.0002,
+      "loss": 1.6249,
+      "step": 3740
+    },
+    {
+      "epoch": 2.5125628140703515,
+      "grad_norm": 0.4916154146194458,
+      "learning_rate": 0.0002,
+      "loss": 1.5762,
+      "step": 3750
+    },
+    {
+      "epoch": 2.5192629815745393,
+      "grad_norm": 0.5381299257278442,
+      "learning_rate": 0.0002,
+      "loss": 1.5157,
+      "step": 3760
+    },
+    {
+      "epoch": 2.525963149078727,
+      "grad_norm": 0.44947415590286255,
+      "learning_rate": 0.0002,
+      "loss": 1.6467,
+      "step": 3770
+    },
+    {
+      "epoch": 2.5326633165829144,
+      "grad_norm": 0.49979084730148315,
+      "learning_rate": 0.0002,
+      "loss": 1.67,
+      "step": 3780
+    },
+    {
+      "epoch": 2.539363484087102,
+      "grad_norm": 0.43046900629997253,
+      "learning_rate": 0.0002,
+      "loss": 1.622,
+      "step": 3790
+    },
+    {
+      "epoch": 2.5460636515912896,
+      "grad_norm": 0.4513470530509949,
+      "learning_rate": 0.0002,
+      "loss": 1.6789,
+      "step": 3800
+    },
+    {
+      "epoch": 2.5527638190954773,
+      "grad_norm": 0.49900051951408386,
+      "learning_rate": 0.0002,
+      "loss": 1.6335,
+      "step": 3810
+    },
+    {
+      "epoch": 2.559463986599665,
+      "grad_norm": 0.4348420202732086,
+      "learning_rate": 0.0002,
+      "loss": 1.6322,
+      "step": 3820
+    },
+    {
+      "epoch": 2.5661641541038525,
+      "grad_norm": 0.4684867560863495,
+      "learning_rate": 0.0002,
+      "loss": 1.6218,
+      "step": 3830
+    },
+    {
+      "epoch": 2.5728643216080402,
+      "grad_norm": 0.44430989027023315,
+      "learning_rate": 0.0002,
+      "loss": 1.6535,
+      "step": 3840
+    },
+    {
+      "epoch": 2.5795644891122276,
+      "grad_norm": 0.47375255823135376,
+      "learning_rate": 0.0002,
+      "loss": 1.5909,
+      "step": 3850
+    },
+    {
+      "epoch": 2.5862646566164154,
+      "grad_norm": 0.45493075251579285,
+      "learning_rate": 0.0002,
+      "loss": 1.6269,
+      "step": 3860
+    },
+    {
+      "epoch": 2.592964824120603,
+      "grad_norm": 0.4563275873661041,
+      "learning_rate": 0.0002,
+      "loss": 1.604,
+      "step": 3870
+    },
+    {
+      "epoch": 2.5996649916247905,
+      "grad_norm": 0.46060335636138916,
+      "learning_rate": 0.0002,
+      "loss": 1.642,
+      "step": 3880
+    },
+    {
+      "epoch": 2.6063651591289783,
+      "grad_norm": 0.4718867540359497,
+      "learning_rate": 0.0002,
+      "loss": 1.6302,
+      "step": 3890
+    },
+    {
+      "epoch": 2.6130653266331656,
+      "grad_norm": 0.41570305824279785,
+      "learning_rate": 0.0002,
+      "loss": 1.6242,
+      "step": 3900
+    },
+    {
+      "epoch": 2.6197654941373534,
+      "grad_norm": 0.4603121876716614,
+      "learning_rate": 0.0002,
+      "loss": 1.6401,
+      "step": 3910
+    },
+    {
+      "epoch": 2.626465661641541,
+      "grad_norm": 0.4734652638435364,
+      "learning_rate": 0.0002,
+      "loss": 1.6839,
+      "step": 3920
+    },
+    {
+      "epoch": 2.6331658291457285,
+      "grad_norm": 0.45348483324050903,
+      "learning_rate": 0.0002,
+      "loss": 1.5448,
+      "step": 3930
+    },
+    {
+      "epoch": 2.6398659966499163,
+      "grad_norm": 0.46559447050094604,
+      "learning_rate": 0.0002,
+      "loss": 1.6157,
+      "step": 3940
+    },
+    {
+      "epoch": 2.6465661641541036,
+      "grad_norm": 0.44113144278526306,
+      "learning_rate": 0.0002,
+      "loss": 1.7052,
+      "step": 3950
+    },
+    {
+      "epoch": 2.6532663316582914,
+      "grad_norm": 0.41415104269981384,
+      "learning_rate": 0.0002,
+      "loss": 1.6315,
+      "step": 3960
+    },
+    {
+      "epoch": 2.659966499162479,
+      "grad_norm": 0.48868080973625183,
+      "learning_rate": 0.0002,
+      "loss": 1.6589,
+      "step": 3970
+    },
+    {
+      "epoch": 2.6666666666666665,
+      "grad_norm": 0.49610549211502075,
+      "learning_rate": 0.0002,
+      "loss": 1.6211,
+      "step": 3980
+    },
+    {
+      "epoch": 2.6733668341708543,
+      "grad_norm": 0.4309130907058716,
+      "learning_rate": 0.0002,
+      "loss": 1.6235,
+      "step": 3990
+    },
+    {
+      "epoch": 2.6800670016750416,
+      "grad_norm": 0.4489327669143677,
+      "learning_rate": 0.0002,
+      "loss": 1.6452,
+      "step": 4000
+    },
+    {
+      "epoch": 2.6867671691792294,
+      "grad_norm": 0.5380139946937561,
+      "learning_rate": 0.0002,
+      "loss": 1.5744,
+      "step": 4010
+    },
+    {
+      "epoch": 2.693467336683417,
+      "grad_norm": 0.5076672434806824,
+      "learning_rate": 0.0002,
+      "loss": 1.6524,
+      "step": 4020
+    },
+    {
+      "epoch": 2.7001675041876045,
+      "grad_norm": 0.47620031237602234,
+      "learning_rate": 0.0002,
+      "loss": 1.636,
+      "step": 4030
+    },
+    {
+      "epoch": 2.7068676716917923,
+      "grad_norm": 0.48089155554771423,
+      "learning_rate": 0.0002,
+      "loss": 1.5543,
+      "step": 4040
+    },
+    {
+      "epoch": 2.7135678391959797,
+      "grad_norm": 0.5108814239501953,
+      "learning_rate": 0.0002,
+      "loss": 1.6396,
+      "step": 4050
+    },
+    {
+      "epoch": 2.7202680067001674,
+      "grad_norm": 0.4196513295173645,
+      "learning_rate": 0.0002,
+      "loss": 1.5905,
+      "step": 4060
+    },
+    {
+      "epoch": 2.726968174204355,
+      "grad_norm": 0.4574664831161499,
+      "learning_rate": 0.0002,
+      "loss": 1.686,
+      "step": 4070
+    },
+    {
+      "epoch": 2.7336683417085426,
+      "grad_norm": 0.4671640992164612,
+      "learning_rate": 0.0002,
+      "loss": 1.6234,
+      "step": 4080
+    },
+    {
+      "epoch": 2.7403685092127303,
+      "grad_norm": 0.49355530738830566,
+      "learning_rate": 0.0002,
+      "loss": 1.6827,
+      "step": 4090
+    },
+    {
+      "epoch": 2.7470686767169177,
+      "grad_norm": 0.46716663241386414,
+      "learning_rate": 0.0002,
+      "loss": 1.6999,
+      "step": 4100
+    },
+    {
+      "epoch": 2.7537688442211055,
+      "grad_norm": 0.45420581102371216,
+      "learning_rate": 0.0002,
+      "loss": 1.6463,
+      "step": 4110
+    },
+    {
+      "epoch": 2.7604690117252932,
+      "grad_norm": 0.4680487811565399,
+      "learning_rate": 0.0002,
+      "loss": 1.5718,
+      "step": 4120
+    },
+    {
+      "epoch": 2.7671691792294806,
+      "grad_norm": 0.5375032424926758,
+      "learning_rate": 0.0002,
+      "loss": 1.5968,
+      "step": 4130
+    },
+    {
+      "epoch": 2.7738693467336684,
+      "grad_norm": 0.46026280522346497,
+      "learning_rate": 0.0002,
+      "loss": 1.5254,
+      "step": 4140
+    },
+    {
+      "epoch": 2.7805695142378557,
+      "grad_norm": 0.43658447265625,
+      "learning_rate": 0.0002,
+      "loss": 1.6613,
+      "step": 4150
+    },
+    {
+      "epoch": 2.7872696817420435,
+      "grad_norm": 0.4935547113418579,
+      "learning_rate": 0.0002,
+      "loss": 1.6546,
+      "step": 4160
+    },
+    {
+      "epoch": 2.7939698492462313,
+      "grad_norm": 0.8167962431907654,
+      "learning_rate": 0.0002,
+      "loss": 1.5961,
+      "step": 4170
+    },
+    {
+      "epoch": 2.8006700167504186,
+      "grad_norm": 0.4289683997631073,
+      "learning_rate": 0.0002,
+      "loss": 1.6907,
+      "step": 4180
+    },
+    {
+      "epoch": 2.8073701842546064,
+      "grad_norm": 0.4569324254989624,
+      "learning_rate": 0.0002,
+      "loss": 1.6385,
+      "step": 4190
+    },
+    {
+      "epoch": 2.8140703517587937,
+      "grad_norm": 0.474795937538147,
+      "learning_rate": 0.0002,
+      "loss": 1.6077,
+      "step": 4200
+    },
+    {
+      "epoch": 2.8207705192629815,
+      "grad_norm": 0.44272229075431824,
+      "learning_rate": 0.0002,
+      "loss": 1.6223,
+      "step": 4210
+    },
+    {
+      "epoch": 2.8274706867671693,
+      "grad_norm": 0.525240957736969,
+      "learning_rate": 0.0002,
+      "loss": 1.6706,
+      "step": 4220
+    },
+    {
+      "epoch": 2.8341708542713566,
+      "grad_norm": 0.4802303910255432,
+      "learning_rate": 0.0002,
+      "loss": 1.7196,
+      "step": 4230
+    },
+    {
+      "epoch": 2.8408710217755444,
+      "grad_norm": 0.46400442719459534,
+      "learning_rate": 0.0002,
+      "loss": 1.6002,
+      "step": 4240
+    },
+    {
+      "epoch": 2.8475711892797317,
+      "grad_norm": 0.49884888529777527,
+      "learning_rate": 0.0002,
+      "loss": 1.6052,
+      "step": 4250
+    },
+    {
+      "epoch": 2.8542713567839195,
+      "grad_norm": 0.5015072226524353,
+      "learning_rate": 0.0002,
+      "loss": 1.6919,
+      "step": 4260
+    },
+    {
+      "epoch": 2.8609715242881073,
+      "grad_norm": 0.4335440695285797,
+      "learning_rate": 0.0002,
+      "loss": 1.6335,
+      "step": 4270
+    },
+    {
+      "epoch": 2.8676716917922946,
+      "grad_norm": 0.5131644606590271,
+      "learning_rate": 0.0002,
+      "loss": 1.5664,
+      "step": 4280
+    },
+    {
+      "epoch": 2.8743718592964824,
+      "grad_norm": 0.6977195739746094,
+      "learning_rate": 0.0002,
+      "loss": 1.6409,
+      "step": 4290
+    },
+    {
+      "epoch": 2.8810720268006698,
+      "grad_norm": 0.5133762955665588,
+      "learning_rate": 0.0002,
+      "loss": 1.7192,
+      "step": 4300
+    },
+    {
+      "epoch": 2.8877721943048575,
+      "grad_norm": 0.4737614393234253,
+      "learning_rate": 0.0002,
+      "loss": 1.6257,
+      "step": 4310
+    },
+    {
+      "epoch": 2.8944723618090453,
+      "grad_norm": 0.4580535590648651,
+      "learning_rate": 0.0002,
+      "loss": 1.6076,
+      "step": 4320
+    },
+    {
+      "epoch": 2.901172529313233,
+      "grad_norm": 0.43863341212272644,
+      "learning_rate": 0.0002,
+      "loss": 1.6538,
+      "step": 4330
+    },
+    {
+      "epoch": 2.9078726968174204,
+      "grad_norm": 0.4103737473487854,
+      "learning_rate": 0.0002,
+      "loss": 1.6091,
+      "step": 4340
+    },
+    {
+      "epoch": 2.914572864321608,
+      "grad_norm": 0.438014417886734,
+      "learning_rate": 0.0002,
+      "loss": 1.7106,
+      "step": 4350
+    },
+    {
+      "epoch": 2.9212730318257956,
+      "grad_norm": 0.5068213939666748,
+      "learning_rate": 0.0002,
+      "loss": 1.6025,
+      "step": 4360
+    },
+    {
+      "epoch": 2.9279731993299833,
+      "grad_norm": 0.45305484533309937,
+      "learning_rate": 0.0002,
+      "loss": 1.6426,
+      "step": 4370
+    },
+    {
+      "epoch": 2.934673366834171,
+      "grad_norm": 0.4612090289592743,
+      "learning_rate": 0.0002,
+      "loss": 1.5726,
+      "step": 4380
+    },
+    {
+      "epoch": 2.9413735343383585,
+      "grad_norm": 0.508736789226532,
+      "learning_rate": 0.0002,
+      "loss": 1.6536,
+      "step": 4390
+    },
+    {
+      "epoch": 2.948073701842546,
+      "grad_norm": 0.4924427270889282,
+      "learning_rate": 0.0002,
+      "loss": 1.6132,
+      "step": 4400
+    },
+    {
+      "epoch": 2.9547738693467336,
+      "grad_norm": 0.5707460641860962,
+      "learning_rate": 0.0002,
+      "loss": 1.7007,
+      "step": 4410
+    },
+    {
+      "epoch": 2.9614740368509214,
+      "grad_norm": 0.42270299792289734,
+      "learning_rate": 0.0002,
+      "loss": 1.6814,
+      "step": 4420
+    },
+    {
+      "epoch": 2.968174204355109,
+      "grad_norm": 0.4429931044578552,
+      "learning_rate": 0.0002,
+      "loss": 1.6644,
+      "step": 4430
+    },
+    {
+      "epoch": 2.9748743718592965,
+      "grad_norm": 0.49760574102401733,
+      "learning_rate": 0.0002,
+      "loss": 1.6251,
+      "step": 4440
+    },
+    {
+      "epoch": 2.981574539363484,
+      "grad_norm": 0.4558229148387909,
+      "learning_rate": 0.0002,
+      "loss": 1.6169,
+      "step": 4450
+    },
+    {
+      "epoch": 2.9882747068676716,
+      "grad_norm": 0.39848530292510986,
+      "learning_rate": 0.0002,
+      "loss": 1.6055,
+      "step": 4460
+    },
+    {
+      "epoch": 2.9949748743718594,
+      "grad_norm": 0.5224862098693848,
+      "learning_rate": 0.0002,
+      "loss": 1.6705,
+      "step": 4470
+    },
+    {
+      "epoch": 2.9996649916247904,
+      "eval_loss": 1.8228833675384521,
+      "eval_runtime": 37.9049,
+      "eval_samples_per_second": 13.587,
+      "eval_steps_per_second": 1.715,
+      "step": 4477
+    },
+    {
+      "epoch": 3.0016750418760467,
+      "grad_norm": 0.41169142723083496,
+      "learning_rate": 0.0002,
+      "loss": 1.6637,
+      "step": 4480
+    },
+    {
+      "epoch": 3.0083752093802345,
+      "grad_norm": 0.4865207374095917,
+      "learning_rate": 0.0002,
+      "loss": 1.5974,
+      "step": 4490
+    },
+    {
+      "epoch": 3.0150753768844223,
+      "grad_norm": 0.5462028384208679,
+      "learning_rate": 0.0002,
+      "loss": 1.5297,
+      "step": 4500
+    },
+    {
+      "epoch": 3.0217755443886096,
+      "grad_norm": 0.6169732809066772,
+      "learning_rate": 0.0002,
+      "loss": 1.5251,
+      "step": 4510
+    },
+    {
+      "epoch": 3.0284757118927974,
+      "grad_norm": 0.5667954087257385,
+      "learning_rate": 0.0002,
+      "loss": 1.5559,
+      "step": 4520
+    },
+    {
+      "epoch": 3.0351758793969847,
+      "grad_norm": 0.5758325457572937,
+      "learning_rate": 0.0002,
+      "loss": 1.5037,
+      "step": 4530
+    },
+    {
+      "epoch": 3.0418760469011725,
+      "grad_norm": 0.5220064520835876,
+      "learning_rate": 0.0002,
+      "loss": 1.4873,
+      "step": 4540
+    },
+    {
+      "epoch": 3.0485762144053603,
+      "grad_norm": 0.5469558835029602,
+      "learning_rate": 0.0002,
+      "loss": 1.5126,
+      "step": 4550
+    },
+    {
+      "epoch": 3.0552763819095476,
+      "grad_norm": 0.5680848956108093,
+      "learning_rate": 0.0002,
+      "loss": 1.4275,
+      "step": 4560
+    },
+    {
+      "epoch": 3.0619765494137354,
+      "grad_norm": 0.5906574726104736,
+      "learning_rate": 0.0002,
+      "loss": 1.5187,
+      "step": 4570
+    },
+    {
+      "epoch": 3.0686767169179228,
+      "grad_norm": 0.4725631773471832,
+      "learning_rate": 0.0002,
+      "loss": 1.4551,
+      "step": 4580
+    },
+    {
+      "epoch": 3.0753768844221105,
+      "grad_norm": 0.5273477435112,
+      "learning_rate": 0.0002,
+      "loss": 1.5083,
+      "step": 4590
+    },
+    {
+      "epoch": 3.0820770519262983,
+      "grad_norm": 0.5861203074455261,
+      "learning_rate": 0.0002,
+      "loss": 1.5154,
+      "step": 4600
+    },
+    {
+      "epoch": 3.0887772194304857,
+      "grad_norm": 0.5343965291976929,
+      "learning_rate": 0.0002,
+      "loss": 1.4924,
+      "step": 4610
+    },
+    {
+      "epoch": 3.0954773869346734,
+      "grad_norm": 0.5348150730133057,
+      "learning_rate": 0.0002,
+      "loss": 1.5608,
+      "step": 4620
+    },
+    {
+      "epoch": 3.102177554438861,
+      "grad_norm": 0.5971846580505371,
+      "learning_rate": 0.0002,
+      "loss": 1.5399,
+      "step": 4630
+    },
+    {
+      "epoch": 3.1088777219430486,
+      "grad_norm": 0.5203177332878113,
+      "learning_rate": 0.0002,
+      "loss": 1.4662,
+      "step": 4640
+    },
+    {
+      "epoch": 3.1155778894472363,
+      "grad_norm": 0.55289226770401,
+      "learning_rate": 0.0002,
+      "loss": 1.5805,
+      "step": 4650
+    },
+    {
+      "epoch": 3.1222780569514237,
+      "grad_norm": 0.6878530979156494,
+      "learning_rate": 0.0002,
+      "loss": 1.4745,
+      "step": 4660
+    },
+    {
+      "epoch": 3.1289782244556115,
+      "grad_norm": 0.6173256635665894,
+      "learning_rate": 0.0002,
+      "loss": 1.5335,
+      "step": 4670
+    },
+    {
+      "epoch": 3.135678391959799,
+      "grad_norm": 0.536796510219574,
+      "learning_rate": 0.0002,
+      "loss": 1.51,
+      "step": 4680
+    },
+    {
+      "epoch": 3.1423785594639866,
+      "grad_norm": 0.58846116065979,
+      "learning_rate": 0.0002,
+      "loss": 1.4713,
+      "step": 4690
+    },
+    {
+      "epoch": 3.1490787269681744,
+      "grad_norm": 0.645889401435852,
+      "learning_rate": 0.0002,
+      "loss": 1.5114,
+      "step": 4700
+    },
+    {
+      "epoch": 3.1557788944723617,
+      "grad_norm": 0.6118691563606262,
+      "learning_rate": 0.0002,
+      "loss": 1.4705,
+      "step": 4710
+    },
+    {
+      "epoch": 3.1624790619765495,
+      "grad_norm": 0.5189669132232666,
+      "learning_rate": 0.0002,
+      "loss": 1.5533,
+      "step": 4720
+    },
+    {
+      "epoch": 3.169179229480737,
+      "grad_norm": 0.5794713497161865,
+      "learning_rate": 0.0002,
+      "loss": 1.4769,
+      "step": 4730
+    },
+    {
+      "epoch": 3.1758793969849246,
+      "grad_norm": 0.6579326391220093,
+      "learning_rate": 0.0002,
+      "loss": 1.4849,
+      "step": 4740
+    },
+    {
+      "epoch": 3.1825795644891124,
+      "grad_norm": 0.5822742581367493,
+      "learning_rate": 0.0002,
+      "loss": 1.545,
+      "step": 4750
+    },
+    {
+      "epoch": 3.1892797319932997,
+      "grad_norm": 0.5475956201553345,
+      "learning_rate": 0.0002,
+      "loss": 1.4358,
+      "step": 4760
+    },
+    {
+      "epoch": 3.1959798994974875,
+      "grad_norm": 0.6743834018707275,
+      "learning_rate": 0.0002,
+      "loss": 1.4723,
+      "step": 4770
+    },
+    {
+      "epoch": 3.202680067001675,
+      "grad_norm": 0.6110585927963257,
+      "learning_rate": 0.0002,
+      "loss": 1.5161,
+      "step": 4780
+    },
+    {
+      "epoch": 3.2093802345058626,
+      "grad_norm": 0.5426181554794312,
+      "learning_rate": 0.0002,
+      "loss": 1.5455,
+      "step": 4790
+    },
+    {
+      "epoch": 3.2160804020100504,
+      "grad_norm": 0.6077824234962463,
+      "learning_rate": 0.0002,
+      "loss": 1.5315,
+      "step": 4800
+    },
+    {
+      "epoch": 3.2227805695142377,
+      "grad_norm": 0.5785858631134033,
+      "learning_rate": 0.0002,
+      "loss": 1.5314,
+      "step": 4810
+    },
+    {
+      "epoch": 3.2294807370184255,
+      "grad_norm": 0.6425958275794983,
+      "learning_rate": 0.0002,
+      "loss": 1.4041,
+      "step": 4820
+    },
+    {
+      "epoch": 3.236180904522613,
+      "grad_norm": 0.6607080698013306,
+      "learning_rate": 0.0002,
+      "loss": 1.4751,
+      "step": 4830
+    },
+    {
+      "epoch": 3.2428810720268006,
+      "grad_norm": 0.5385788679122925,
+      "learning_rate": 0.0002,
+      "loss": 1.5267,
+      "step": 4840
+    },
+    {
+      "epoch": 3.2495812395309884,
+      "grad_norm": 0.5630403757095337,
+      "learning_rate": 0.0002,
+      "loss": 1.4673,
+      "step": 4850
+    },
+    {
+      "epoch": 3.2562814070351758,
+      "grad_norm": 0.6340779662132263,
+      "learning_rate": 0.0002,
+      "loss": 1.5257,
+      "step": 4860
+    },
+    {
+      "epoch": 3.2629815745393635,
+      "grad_norm": 0.5305342674255371,
+      "learning_rate": 0.0002,
+      "loss": 1.5148,
+      "step": 4870
+    },
+    {
+      "epoch": 3.2696817420435513,
+      "grad_norm": 0.597670316696167,
+      "learning_rate": 0.0002,
+      "loss": 1.5162,
+      "step": 4880
+    },
+    {
+      "epoch": 3.2763819095477387,
+      "grad_norm": 0.665553867816925,
+      "learning_rate": 0.0002,
+      "loss": 1.5429,
+      "step": 4890
+    },
+    {
+      "epoch": 3.2830820770519265,
+      "grad_norm": 0.579767644405365,
+      "learning_rate": 0.0002,
+      "loss": 1.4607,
+      "step": 4900
+    },
+    {
+      "epoch": 3.289782244556114,
+      "grad_norm": 0.5512481331825256,
+      "learning_rate": 0.0002,
+      "loss": 1.4999,
+      "step": 4910
+    },
+    {
+      "epoch": 3.2964824120603016,
+      "grad_norm": 0.5916532278060913,
+      "learning_rate": 0.0002,
+      "loss": 1.5022,
+      "step": 4920
+    },
+    {
+      "epoch": 3.3031825795644894,
+      "grad_norm": 0.7521726489067078,
+      "learning_rate": 0.0002,
+      "loss": 1.4889,
+      "step": 4930
+    },
+    {
+      "epoch": 3.3098827470686767,
+      "grad_norm": 0.5352797508239746,
+      "learning_rate": 0.0002,
+      "loss": 1.4223,
+      "step": 4940
+    },
+    {
+      "epoch": 3.3165829145728645,
+      "grad_norm": 0.5950371623039246,
+      "learning_rate": 0.0002,
+      "loss": 1.5122,
+      "step": 4950
+    },
+    {
+      "epoch": 3.323283082077052,
+      "grad_norm": 0.8020477890968323,
+      "learning_rate": 0.0002,
+      "loss": 1.5072,
+      "step": 4960
+    },
+    {
+      "epoch": 3.3299832495812396,
+      "grad_norm": 0.6790024638175964,
+      "learning_rate": 0.0002,
+      "loss": 1.5422,
+      "step": 4970
+    },
+    {
+      "epoch": 3.3366834170854274,
+      "grad_norm": 0.687627375125885,
+      "learning_rate": 0.0002,
+      "loss": 1.5363,
+      "step": 4980
+    },
+    {
+      "epoch": 3.3433835845896147,
+      "grad_norm": 0.6094385385513306,
+      "learning_rate": 0.0002,
+      "loss": 1.5276,
+      "step": 4990
+    },
+    {
+      "epoch": 3.3500837520938025,
+      "grad_norm": 0.6541242003440857,
+      "learning_rate": 0.0002,
+      "loss": 1.549,
+      "step": 5000
+    },
+    {
+      "epoch": 3.35678391959799,
+      "grad_norm": 0.5560880303382874,
+      "learning_rate": 0.0002,
+      "loss": 1.6067,
+      "step": 5010
+    },
+    {
+      "epoch": 3.3634840871021776,
+      "grad_norm": 0.5440094470977783,
+      "learning_rate": 0.0002,
+      "loss": 1.5769,
+      "step": 5020
+    },
+    {
+      "epoch": 3.3701842546063654,
+      "grad_norm": 0.5749301314353943,
+      "learning_rate": 0.0002,
+      "loss": 1.6183,
+      "step": 5030
+    },
+    {
+      "epoch": 3.3768844221105527,
+      "grad_norm": 0.5919716954231262,
+      "learning_rate": 0.0002,
+      "loss": 1.4801,
+      "step": 5040
+    },
+    {
+      "epoch": 3.3835845896147405,
+      "grad_norm": 0.6331481337547302,
+      "learning_rate": 0.0002,
+      "loss": 1.5957,
+      "step": 5050
+    },
+    {
+      "epoch": 3.390284757118928,
+      "grad_norm": 0.5687161684036255,
+      "learning_rate": 0.0002,
+      "loss": 1.5188,
+      "step": 5060
+    },
+    {
+      "epoch": 3.3969849246231156,
+      "grad_norm": 0.6718577742576599,
+      "learning_rate": 0.0002,
+      "loss": 1.5702,
+      "step": 5070
+    },
+    {
+      "epoch": 3.4036850921273034,
+      "grad_norm": 0.5089324116706848,
+      "learning_rate": 0.0002,
+      "loss": 1.5577,
+      "step": 5080
+    },
+    {
+      "epoch": 3.4103852596314908,
+      "grad_norm": 0.5710174441337585,
+      "learning_rate": 0.0002,
+      "loss": 1.512,
+      "step": 5090
+    },
+    {
+      "epoch": 3.4170854271356785,
+      "grad_norm": 0.6670721173286438,
+      "learning_rate": 0.0002,
+      "loss": 1.5492,
+      "step": 5100
+    },
+    {
+      "epoch": 3.423785594639866,
+      "grad_norm": 0.6875665187835693,
+      "learning_rate": 0.0002,
+      "loss": 1.5227,
+      "step": 5110
+    },
+    {
+      "epoch": 3.4304857621440537,
+      "grad_norm": 0.5375880599021912,
+      "learning_rate": 0.0002,
+      "loss": 1.4496,
+      "step": 5120
+    },
+    {
+      "epoch": 3.4371859296482414,
+      "grad_norm": 0.6550399661064148,
+      "learning_rate": 0.0002,
+      "loss": 1.5527,
+      "step": 5130
+    },
+    {
+      "epoch": 3.4438860971524288,
+      "grad_norm": 0.5948067903518677,
+      "learning_rate": 0.0002,
+      "loss": 1.5687,
+      "step": 5140
+    },
+    {
+      "epoch": 3.4505862646566166,
+      "grad_norm": 0.6134477257728577,
+      "learning_rate": 0.0002,
+      "loss": 1.4813,
+      "step": 5150
+    },
+    {
+      "epoch": 3.457286432160804,
+      "grad_norm": 0.6506398320198059,
+      "learning_rate": 0.0002,
+      "loss": 1.5069,
+      "step": 5160
+    },
+    {
+      "epoch": 3.4639865996649917,
+      "grad_norm": 0.6060147881507874,
+      "learning_rate": 0.0002,
+      "loss": 1.4422,
+      "step": 5170
+    },
+    {
+      "epoch": 3.4706867671691795,
+      "grad_norm": 0.6173806190490723,
+      "learning_rate": 0.0002,
+      "loss": 1.5093,
+      "step": 5180
+    },
+    {
+      "epoch": 3.477386934673367,
+      "grad_norm": 0.6032607555389404,
+      "learning_rate": 0.0002,
+      "loss": 1.4975,
+      "step": 5190
+    },
+    {
+      "epoch": 3.4840871021775546,
+      "grad_norm": 0.5652492046356201,
+      "learning_rate": 0.0002,
+      "loss": 1.4979,
+      "step": 5200
+    },
+    {
+      "epoch": 3.490787269681742,
+      "grad_norm": 0.6168607473373413,
+      "learning_rate": 0.0002,
+      "loss": 1.4883,
+      "step": 5210
+    },
+    {
+      "epoch": 3.4974874371859297,
+      "grad_norm": 0.6170629262924194,
+      "learning_rate": 0.0002,
+      "loss": 1.5164,
+      "step": 5220
+    },
+    {
+      "epoch": 3.5041876046901175,
+      "grad_norm": 0.6926297545433044,
+      "learning_rate": 0.0002,
+      "loss": 1.4879,
+      "step": 5230
+    },
+    {
+      "epoch": 3.510887772194305,
+      "grad_norm": 0.6702437996864319,
+      "learning_rate": 0.0002,
+      "loss": 1.4982,
+      "step": 5240
+    },
+    {
+      "epoch": 3.5175879396984926,
+      "grad_norm": 0.5421436429023743,
+      "learning_rate": 0.0002,
+      "loss": 1.4986,
+      "step": 5250
+    },
+    {
+      "epoch": 3.52428810720268,
+      "grad_norm": 0.5726765990257263,
+      "learning_rate": 0.0002,
+      "loss": 1.4673,
+      "step": 5260
+    },
+    {
+      "epoch": 3.5309882747068677,
+      "grad_norm": 0.5685455203056335,
+      "learning_rate": 0.0002,
+      "loss": 1.5423,
+      "step": 5270
+    },
+    {
+      "epoch": 3.5376884422110555,
+      "grad_norm": 0.6018396019935608,
+      "learning_rate": 0.0002,
+      "loss": 1.4715,
+      "step": 5280
+    },
+    {
+      "epoch": 3.544388609715243,
+      "grad_norm": 0.5731932520866394,
+      "learning_rate": 0.0002,
+      "loss": 1.5451,
+      "step": 5290
+    },
+    {
+      "epoch": 3.5510887772194306,
+      "grad_norm": 0.6601519584655762,
+      "learning_rate": 0.0002,
+      "loss": 1.4752,
+      "step": 5300
+    },
+    {
+      "epoch": 3.557788944723618,
+      "grad_norm": 0.5545530319213867,
+      "learning_rate": 0.0002,
+      "loss": 1.5434,
+      "step": 5310
+    },
+    {
+      "epoch": 3.5644891122278057,
+      "grad_norm": 0.5998541116714478,
+      "learning_rate": 0.0002,
+      "loss": 1.5438,
+      "step": 5320
+    },
+    {
+      "epoch": 3.5711892797319935,
+      "grad_norm": 0.5651767253875732,
+      "learning_rate": 0.0002,
+      "loss": 1.56,
+      "step": 5330
+    },
+    {
+      "epoch": 3.577889447236181,
+      "grad_norm": 0.7425084114074707,
+      "learning_rate": 0.0002,
+      "loss": 1.4829,
+      "step": 5340
+    },
+    {
+      "epoch": 3.5845896147403686,
+      "grad_norm": 0.5770602226257324,
+      "learning_rate": 0.0002,
+      "loss": 1.5571,
+      "step": 5350
+    },
+    {
+      "epoch": 3.591289782244556,
+      "grad_norm": 0.54723060131073,
+      "learning_rate": 0.0002,
+      "loss": 1.458,
+      "step": 5360
+    },
+    {
+      "epoch": 3.5979899497487438,
+      "grad_norm": 0.6658238172531128,
+      "learning_rate": 0.0002,
+      "loss": 1.497,
+      "step": 5370
+    },
+    {
+      "epoch": 3.6046901172529315,
+      "grad_norm": 0.5787645578384399,
+      "learning_rate": 0.0002,
+      "loss": 1.5456,
+      "step": 5380
+    },
+    {
+      "epoch": 3.611390284757119,
+      "grad_norm": 0.594913125038147,
+      "learning_rate": 0.0002,
+      "loss": 1.5343,
+      "step": 5390
+    },
+    {
+      "epoch": 3.6180904522613067,
+      "grad_norm": 0.4964977502822876,
+      "learning_rate": 0.0002,
+      "loss": 1.4727,
+      "step": 5400
+    },
+    {
+      "epoch": 3.624790619765494,
+      "grad_norm": 0.6087527275085449,
+      "learning_rate": 0.0002,
+      "loss": 1.5062,
+      "step": 5410
+    },
+    {
+      "epoch": 3.6314907872696818,
+      "grad_norm": 0.6315323710441589,
+      "learning_rate": 0.0002,
+      "loss": 1.5098,
+      "step": 5420
+    },
+    {
+      "epoch": 3.6381909547738696,
+      "grad_norm": 0.574799120426178,
+      "learning_rate": 0.0002,
+      "loss": 1.4855,
+      "step": 5430
+    },
+    {
+      "epoch": 3.644891122278057,
+      "grad_norm": 0.5949277877807617,
+      "learning_rate": 0.0002,
+      "loss": 1.4595,
+      "step": 5440
+    },
+    {
+      "epoch": 3.6515912897822447,
+      "grad_norm": 0.5640677213668823,
+      "learning_rate": 0.0002,
+      "loss": 1.4816,
+      "step": 5450
+    },
+    {
+      "epoch": 3.658291457286432,
+      "grad_norm": 0.6198237538337708,
+      "learning_rate": 0.0002,
+      "loss": 1.525,
+      "step": 5460
+    },
+    {
+      "epoch": 3.66499162479062,
+      "grad_norm": 0.6902034878730774,
+      "learning_rate": 0.0002,
+      "loss": 1.5676,
+      "step": 5470
+    },
+    {
+      "epoch": 3.6716917922948076,
+      "grad_norm": 0.5686674118041992,
+      "learning_rate": 0.0002,
+      "loss": 1.544,
+      "step": 5480
+    },
+    {
+      "epoch": 3.678391959798995,
+      "grad_norm": 0.6532107591629028,
+      "learning_rate": 0.0002,
+      "loss": 1.5255,
+      "step": 5490
+    },
+    {
+      "epoch": 3.6850921273031827,
+      "grad_norm": 0.5790849924087524,
+      "learning_rate": 0.0002,
+      "loss": 1.5767,
+      "step": 5500
+    },
+    {
+      "epoch": 3.69179229480737,
+      "grad_norm": 0.6055065393447876,
+      "learning_rate": 0.0002,
+      "loss": 1.5507,
+      "step": 5510
+    },
+    {
+      "epoch": 3.698492462311558,
+      "grad_norm": 0.5630605816841125,
+      "learning_rate": 0.0002,
+      "loss": 1.4656,
+      "step": 5520
+    },
+    {
+      "epoch": 3.7051926298157456,
+      "grad_norm": 0.6005825996398926,
+      "learning_rate": 0.0002,
+      "loss": 1.537,
+      "step": 5530
+    },
+    {
+      "epoch": 3.711892797319933,
+      "grad_norm": 0.6553038954734802,
+      "learning_rate": 0.0002,
+      "loss": 1.5313,
+      "step": 5540
+    },
+    {
+      "epoch": 3.7185929648241207,
+      "grad_norm": 0.5601094961166382,
+      "learning_rate": 0.0002,
+      "loss": 1.4943,
+      "step": 5550
+    },
+    {
+      "epoch": 3.725293132328308,
+      "grad_norm": 0.6598808169364929,
+      "learning_rate": 0.0002,
+      "loss": 1.5161,
+      "step": 5560
+    },
+    {
+      "epoch": 3.731993299832496,
+      "grad_norm": 0.5506255626678467,
+      "learning_rate": 0.0002,
+      "loss": 1.5345,
+      "step": 5570
+    },
+    {
+      "epoch": 3.7386934673366836,
+      "grad_norm": 0.6001223921775818,
+      "learning_rate": 0.0002,
+      "loss": 1.4805,
+      "step": 5580
+    },
+    {
+      "epoch": 3.745393634840871,
+      "grad_norm": 0.6287297606468201,
+      "learning_rate": 0.0002,
+      "loss": 1.4652,
+      "step": 5590
+    },
+    {
+      "epoch": 3.7520938023450587,
+      "grad_norm": 0.6253238916397095,
+      "learning_rate": 0.0002,
+      "loss": 1.5246,
+      "step": 5600
+    },
+    {
+      "epoch": 3.758793969849246,
+      "grad_norm": 0.5713174939155579,
+      "learning_rate": 0.0002,
+      "loss": 1.5691,
+      "step": 5610
+    },
+    {
+      "epoch": 3.765494137353434,
+      "grad_norm": 0.6198310852050781,
+      "learning_rate": 0.0002,
+      "loss": 1.5661,
+      "step": 5620
+    },
+    {
+      "epoch": 3.7721943048576216,
+      "grad_norm": 0.5941224098205566,
+      "learning_rate": 0.0002,
+      "loss": 1.5448,
+      "step": 5630
+    },
+    {
+      "epoch": 3.778894472361809,
+      "grad_norm": 0.606002151966095,
+      "learning_rate": 0.0002,
+      "loss": 1.4925,
+      "step": 5640
+    },
+    {
+      "epoch": 3.7855946398659968,
+      "grad_norm": 0.6540704965591431,
+      "learning_rate": 0.0002,
+      "loss": 1.5182,
+      "step": 5650
+    },
+    {
+      "epoch": 3.792294807370184,
+      "grad_norm": 0.6147415041923523,
+      "learning_rate": 0.0002,
+      "loss": 1.5903,
+      "step": 5660
+    },
+    {
+      "epoch": 3.798994974874372,
+      "grad_norm": 0.5649605393409729,
+      "learning_rate": 0.0002,
+      "loss": 1.5329,
+      "step": 5670
+    },
+    {
+      "epoch": 3.8056951423785597,
+      "grad_norm": 0.6788773536682129,
+      "learning_rate": 0.0002,
+      "loss": 1.5747,
+      "step": 5680
+    },
+    {
+      "epoch": 3.812395309882747,
+      "grad_norm": 0.6581860780715942,
+      "learning_rate": 0.0002,
+      "loss": 1.535,
+      "step": 5690
+    },
+    {
+      "epoch": 3.819095477386935,
+      "grad_norm": 0.5529348850250244,
+      "learning_rate": 0.0002,
+      "loss": 1.4587,
+      "step": 5700
+    },
+    {
+      "epoch": 3.825795644891122,
+      "grad_norm": 0.6320232152938843,
+      "learning_rate": 0.0002,
+      "loss": 1.5594,
+      "step": 5710
+    },
+    {
+      "epoch": 3.83249581239531,
+      "grad_norm": 0.6529698371887207,
+      "learning_rate": 0.0002,
+      "loss": 1.4696,
+      "step": 5720
+    },
+    {
+      "epoch": 3.8391959798994977,
+      "grad_norm": 0.5983362793922424,
+      "learning_rate": 0.0002,
+      "loss": 1.5854,
+      "step": 5730
+    },
+    {
+      "epoch": 3.845896147403685,
+      "grad_norm": 0.6335684061050415,
+      "learning_rate": 0.0002,
+      "loss": 1.465,
+      "step": 5740
+    },
+    {
+      "epoch": 3.852596314907873,
+      "grad_norm": 0.700446605682373,
+      "learning_rate": 0.0002,
+      "loss": 1.5545,
+      "step": 5750
+    },
+    {
+      "epoch": 3.85929648241206,
+      "grad_norm": 0.6092597842216492,
+      "learning_rate": 0.0002,
+      "loss": 1.5707,
+      "step": 5760
+    },
+    {
+      "epoch": 3.865996649916248,
+      "grad_norm": 0.564146101474762,
+      "learning_rate": 0.0002,
+      "loss": 1.5729,
+      "step": 5770
+    },
+    {
+      "epoch": 3.8726968174204357,
+      "grad_norm": 0.615275502204895,
+      "learning_rate": 0.0002,
+      "loss": 1.5872,
+      "step": 5780
+    },
+    {
+      "epoch": 3.879396984924623,
+      "grad_norm": 0.6685376763343811,
+      "learning_rate": 0.0002,
+      "loss": 1.5142,
+      "step": 5790
+    },
+    {
+      "epoch": 3.886097152428811,
+      "grad_norm": 0.6116922497749329,
+      "learning_rate": 0.0002,
+      "loss": 1.4752,
+      "step": 5800
+    },
+    {
+      "epoch": 3.892797319932998,
+      "grad_norm": 0.5486813187599182,
+      "learning_rate": 0.0002,
+      "loss": 1.5179,
+      "step": 5810
+    },
+    {
+      "epoch": 3.899497487437186,
+      "grad_norm": 0.6208204030990601,
+      "learning_rate": 0.0002,
+      "loss": 1.5167,
+      "step": 5820
+    },
+    {
+      "epoch": 3.9061976549413737,
+      "grad_norm": 0.6500625014305115,
+      "learning_rate": 0.0002,
+      "loss": 1.5334,
+      "step": 5830
+    },
+    {
+      "epoch": 3.912897822445561,
+      "grad_norm": 0.5948089361190796,
+      "learning_rate": 0.0002,
+      "loss": 1.4716,
+      "step": 5840
+    },
+    {
+      "epoch": 3.919597989949749,
+      "grad_norm": 0.7210732698440552,
+      "learning_rate": 0.0002,
+      "loss": 1.6011,
+      "step": 5850
+    },
+    {
+      "epoch": 3.926298157453936,
+      "grad_norm": 0.6662322878837585,
+      "learning_rate": 0.0002,
+      "loss": 1.5519,
+      "step": 5860
+    },
+    {
+      "epoch": 3.932998324958124,
+      "grad_norm": 0.5613839626312256,
+      "learning_rate": 0.0002,
+      "loss": 1.5656,
+      "step": 5870
+    },
+    {
+      "epoch": 3.9396984924623117,
+      "grad_norm": 0.6069002151489258,
+      "learning_rate": 0.0002,
+      "loss": 1.544,
+      "step": 5880
+    },
+    {
+      "epoch": 3.946398659966499,
+      "grad_norm": 0.7075562477111816,
+      "learning_rate": 0.0002,
+      "loss": 1.6745,
+      "step": 5890
+    },
+    {
+      "epoch": 3.953098827470687,
+      "grad_norm": 0.6316173076629639,
+      "learning_rate": 0.0002,
+      "loss": 1.5391,
+      "step": 5900
+    },
+    {
+      "epoch": 3.959798994974874,
+      "grad_norm": 0.5716308355331421,
+      "learning_rate": 0.0002,
+      "loss": 1.6314,
+      "step": 5910
+    },
+    {
+      "epoch": 3.966499162479062,
+      "grad_norm": 0.6800096035003662,
+      "learning_rate": 0.0002,
+      "loss": 1.5947,
+      "step": 5920
+    },
+    {
+      "epoch": 3.9731993299832498,
+      "grad_norm": 0.6057983040809631,
+      "learning_rate": 0.0002,
+      "loss": 1.5189,
+      "step": 5930
+    },
+    {
+      "epoch": 3.979899497487437,
+      "grad_norm": 0.5938987731933594,
+      "learning_rate": 0.0002,
+      "loss": 1.5431,
+      "step": 5940
+    },
+    {
+      "epoch": 3.986599664991625,
+      "grad_norm": 0.6963576674461365,
+      "learning_rate": 0.0002,
+      "loss": 1.5111,
+      "step": 5950
+    },
+    {
+      "epoch": 3.993299832495812,
+      "grad_norm": 0.6279940009117126,
+      "learning_rate": 0.0002,
+      "loss": 1.5521,
+      "step": 5960
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 0.7161159515380859,
+      "learning_rate": 0.0002,
+      "loss": 1.5974,
+      "step": 5970
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 1.8655421733856201,
+      "eval_runtime": 37.9276,
+      "eval_samples_per_second": 13.579,
+      "eval_steps_per_second": 1.714,
+      "step": 5970
+    },
+    {
+      "epoch": 4.006700167504188,
+      "grad_norm": 0.7380476593971252,
+      "learning_rate": 0.0002,
+      "loss": 1.3666,
+      "step": 5980
+    },
+    {
+      "epoch": 4.013400335008376,
+      "grad_norm": 0.7148947715759277,
+      "learning_rate": 0.0002,
+      "loss": 1.3913,
+      "step": 5990
+    },
+    {
+      "epoch": 4.0201005025125625,
+      "grad_norm": 0.6177082657814026,
+      "learning_rate": 0.0002,
+      "loss": 1.4204,
+      "step": 6000
+    },
+    {
+      "epoch": 4.02680067001675,
+      "grad_norm": 0.8552946448326111,
+      "learning_rate": 0.0002,
+      "loss": 1.4421,
+      "step": 6010
+    },
+    {
+      "epoch": 4.033500837520938,
+      "grad_norm": 0.8033416271209717,
+      "learning_rate": 0.0002,
+      "loss": 1.4342,
+      "step": 6020
+    },
+    {
+      "epoch": 4.040201005025126,
+      "grad_norm": 0.8501318097114563,
+      "learning_rate": 0.0002,
+      "loss": 1.4092,
+      "step": 6030
+    },
+    {
+      "epoch": 4.046901172529314,
+      "grad_norm": 0.6981393098831177,
+      "learning_rate": 0.0002,
+      "loss": 1.3367,
+      "step": 6040
+    },
+    {
+      "epoch": 4.0536013400335005,
+      "grad_norm": 0.7227180600166321,
+      "learning_rate": 0.0002,
+      "loss": 1.3925,
+      "step": 6050
+    },
+    {
+      "epoch": 4.060301507537688,
+      "grad_norm": 0.6923989653587341,
+      "learning_rate": 0.0002,
+      "loss": 1.4007,
+      "step": 6060
+    },
+    {
+      "epoch": 4.067001675041876,
+      "grad_norm": 0.879779040813446,
+      "learning_rate": 0.0002,
+      "loss": 1.3837,
+      "step": 6070
+    },
+    {
+      "epoch": 4.073701842546064,
+      "grad_norm": 0.8184754848480225,
+      "learning_rate": 0.0002,
+      "loss": 1.4383,
+      "step": 6080
+    },
+    {
+      "epoch": 4.080402010050252,
+      "grad_norm": 0.8211342692375183,
+      "learning_rate": 0.0002,
+      "loss": 1.3128,
+      "step": 6090
+    },
+    {
+      "epoch": 4.0871021775544385,
+      "grad_norm": 0.7542396783828735,
+      "learning_rate": 0.0002,
+      "loss": 1.3892,
+      "step": 6100
+    },
+    {
+      "epoch": 4.093802345058626,
+      "grad_norm": 0.6631066799163818,
+      "learning_rate": 0.0002,
+      "loss": 1.3607,
+      "step": 6110
+    },
+    {
+      "epoch": 4.100502512562814,
+      "grad_norm": 0.6728386282920837,
+      "learning_rate": 0.0002,
+      "loss": 1.3275,
+      "step": 6120
+    },
+    {
+      "epoch": 4.107202680067002,
+      "grad_norm": 0.681851863861084,
+      "learning_rate": 0.0002,
+      "loss": 1.3443,
+      "step": 6130
+    },
+    {
+      "epoch": 4.11390284757119,
+      "grad_norm": 0.8757794499397278,
+      "learning_rate": 0.0002,
+      "loss": 1.3486,
+      "step": 6140
+    },
+    {
+      "epoch": 4.1206030150753765,
+      "grad_norm": 0.6567301750183105,
+      "learning_rate": 0.0002,
+      "loss": 1.351,
+      "step": 6150
+    },
+    {
+      "epoch": 4.127303182579564,
+      "grad_norm": 0.7950329184532166,
+      "learning_rate": 0.0002,
+      "loss": 1.3824,
+      "step": 6160
+    },
+    {
+      "epoch": 4.134003350083752,
+      "grad_norm": 0.7545644044876099,
+      "learning_rate": 0.0002,
+      "loss": 1.3738,
+      "step": 6170
+    },
+    {
+      "epoch": 4.14070351758794,
+      "grad_norm": 0.7172710299491882,
+      "learning_rate": 0.0002,
+      "loss": 1.4214,
+      "step": 6180
+    },
+    {
+      "epoch": 4.147403685092128,
+      "grad_norm": 0.7040584087371826,
+      "learning_rate": 0.0002,
+      "loss": 1.4091,
+      "step": 6190
+    },
+    {
+      "epoch": 4.1541038525963145,
+      "grad_norm": 0.7482913732528687,
+      "learning_rate": 0.0002,
+      "loss": 1.4149,
+      "step": 6200
+    },
+    {
+      "epoch": 4.160804020100502,
+      "grad_norm": 0.8523276448249817,
+      "learning_rate": 0.0002,
+      "loss": 1.3227,
+      "step": 6210
+    },
+    {
+      "epoch": 4.16750418760469,
+      "grad_norm": 0.6672041416168213,
+      "learning_rate": 0.0002,
+      "loss": 1.4194,
+      "step": 6220
+    },
+    {
+      "epoch": 4.174204355108878,
+      "grad_norm": 0.7523500919342041,
+      "learning_rate": 0.0002,
+      "loss": 1.3953,
+      "step": 6230
+    },
+    {
+      "epoch": 4.180904522613066,
+      "grad_norm": 0.8085253834724426,
+      "learning_rate": 0.0002,
+      "loss": 1.371,
+      "step": 6240
+    },
+    {
+      "epoch": 4.187604690117253,
+      "grad_norm": 0.789450466632843,
+      "learning_rate": 0.0002,
+      "loss": 1.3293,
+      "step": 6250
+    },
+    {
+      "epoch": 4.19430485762144,
+      "grad_norm": 0.7502310872077942,
+      "learning_rate": 0.0002,
+      "loss": 1.3539,
+      "step": 6260
+    },
+    {
+      "epoch": 4.201005025125628,
+      "grad_norm": 0.7397456765174866,
+      "learning_rate": 0.0002,
+      "loss": 1.3415,
+      "step": 6270
+    },
+    {
+      "epoch": 4.207705192629816,
+      "grad_norm": 0.6921947002410889,
+      "learning_rate": 0.0002,
+      "loss": 1.3963,
+      "step": 6280
+    },
+    {
+      "epoch": 4.214405360134004,
+      "grad_norm": 0.9334571957588196,
+      "learning_rate": 0.0002,
+      "loss": 1.3125,
+      "step": 6290
+    },
+    {
+      "epoch": 4.221105527638191,
+      "grad_norm": 0.725799024105072,
+      "learning_rate": 0.0002,
+      "loss": 1.3612,
+      "step": 6300
+    },
+    {
+      "epoch": 4.227805695142378,
+      "grad_norm": 0.8290495872497559,
+      "learning_rate": 0.0002,
+      "loss": 1.4217,
+      "step": 6310
+    },
+    {
+      "epoch": 4.234505862646566,
+      "grad_norm": 0.688983678817749,
+      "learning_rate": 0.0002,
+      "loss": 1.4135,
+      "step": 6320
+    },
+    {
+      "epoch": 4.241206030150754,
+      "grad_norm": 0.8620913028717041,
+      "learning_rate": 0.0002,
+      "loss": 1.3807,
+      "step": 6330
+    },
+    {
+      "epoch": 4.247906197654942,
+      "grad_norm": 0.8008657693862915,
+      "learning_rate": 0.0002,
+      "loss": 1.3738,
+      "step": 6340
+    },
+    {
+      "epoch": 4.254606365159129,
+      "grad_norm": 0.7379199266433716,
+      "learning_rate": 0.0002,
+      "loss": 1.4005,
+      "step": 6350
+    },
+    {
+      "epoch": 4.261306532663316,
+      "grad_norm": 0.7842815518379211,
+      "learning_rate": 0.0002,
+      "loss": 1.426,
+      "step": 6360
+    },
+    {
+      "epoch": 4.268006700167504,
+      "grad_norm": 0.812600314617157,
+      "learning_rate": 0.0002,
+      "loss": 1.4262,
+      "step": 6370
+    },
+    {
+      "epoch": 4.274706867671692,
+      "grad_norm": 0.7852841019630432,
+      "learning_rate": 0.0002,
+      "loss": 1.4028,
+      "step": 6380
+    },
+    {
+      "epoch": 4.28140703517588,
+      "grad_norm": 1.0377534627914429,
+      "learning_rate": 0.0002,
+      "loss": 1.3722,
+      "step": 6390
+    },
+    {
+      "epoch": 4.288107202680067,
+      "grad_norm": 1.03935706615448,
+      "learning_rate": 0.0002,
+      "loss": 1.3755,
+      "step": 6400
+    },
+    {
+      "epoch": 4.294807370184254,
+      "grad_norm": 0.7244732975959778,
+      "learning_rate": 0.0002,
+      "loss": 1.3961,
+      "step": 6410
+    },
+    {
+      "epoch": 4.301507537688442,
+      "grad_norm": 0.7137406468391418,
+      "learning_rate": 0.0002,
+      "loss": 1.4608,
+      "step": 6420
+    },
+    {
+      "epoch": 4.30820770519263,
+      "grad_norm": 0.7492543458938599,
+      "learning_rate": 0.0002,
+      "loss": 1.4461,
+      "step": 6430
+    },
+    {
+      "epoch": 4.314907872696818,
+      "grad_norm": 0.7065439224243164,
+      "learning_rate": 0.0002,
+      "loss": 1.4562,
+      "step": 6440
+    },
+    {
+      "epoch": 4.321608040201005,
+      "grad_norm": 0.7786989808082581,
+      "learning_rate": 0.0002,
+      "loss": 1.4246,
+      "step": 6450
+    },
+    {
+      "epoch": 4.328308207705192,
+      "grad_norm": 0.7369208335876465,
+      "learning_rate": 0.0002,
+      "loss": 1.3098,
+      "step": 6460
+    },
+    {
+      "epoch": 4.33500837520938,
+      "grad_norm": 0.7412346005439758,
+      "learning_rate": 0.0002,
+      "loss": 1.3686,
+      "step": 6470
+    },
+    {
+      "epoch": 4.341708542713568,
+      "grad_norm": 0.780927300453186,
+      "learning_rate": 0.0002,
+      "loss": 1.4087,
+      "step": 6480
+    },
+    {
+      "epoch": 4.348408710217756,
+      "grad_norm": 0.8320930600166321,
+      "learning_rate": 0.0002,
+      "loss": 1.3628,
+      "step": 6490
+    },
+    {
+      "epoch": 4.355108877721943,
+      "grad_norm": 0.6871094703674316,
+      "learning_rate": 0.0002,
+      "loss": 1.3715,
+      "step": 6500
+    },
+    {
+      "epoch": 4.36180904522613,
+      "grad_norm": 0.6751559972763062,
+      "learning_rate": 0.0002,
+      "loss": 1.3257,
+      "step": 6510
+    },
+    {
+      "epoch": 4.368509212730318,
+      "grad_norm": 0.7723976969718933,
+      "learning_rate": 0.0002,
+      "loss": 1.4311,
+      "step": 6520
+    },
+    {
+      "epoch": 4.375209380234506,
+      "grad_norm": 0.7915401458740234,
+      "learning_rate": 0.0002,
+      "loss": 1.4086,
+      "step": 6530
+    },
+    {
+      "epoch": 4.381909547738694,
+      "grad_norm": 0.7329102754592896,
+      "learning_rate": 0.0002,
+      "loss": 1.3973,
+      "step": 6540
+    },
+    {
+      "epoch": 4.388609715242881,
+      "grad_norm": 0.7388760447502136,
+      "learning_rate": 0.0002,
+      "loss": 1.447,
+      "step": 6550
+    },
+    {
+      "epoch": 4.3953098827470685,
+      "grad_norm": 0.8282579183578491,
+      "learning_rate": 0.0002,
+      "loss": 1.4378,
+      "step": 6560
+    },
+    {
+      "epoch": 4.402010050251256,
+      "grad_norm": 0.7192724347114563,
+      "learning_rate": 0.0002,
+      "loss": 1.3923,
+      "step": 6570
+    },
+    {
+      "epoch": 4.408710217755444,
+      "grad_norm": 0.746526837348938,
+      "learning_rate": 0.0002,
+      "loss": 1.4141,
+      "step": 6580
+    },
+    {
+      "epoch": 4.415410385259632,
+      "grad_norm": 0.8738046288490295,
+      "learning_rate": 0.0002,
+      "loss": 1.33,
+      "step": 6590
+    },
+    {
+      "epoch": 4.422110552763819,
+      "grad_norm": 0.8408458828926086,
+      "learning_rate": 0.0002,
+      "loss": 1.3995,
+      "step": 6600
+    },
+    {
+      "epoch": 4.4288107202680065,
+      "grad_norm": 0.8110666275024414,
+      "learning_rate": 0.0002,
+      "loss": 1.4148,
+      "step": 6610
+    },
+    {
+      "epoch": 4.435510887772194,
+      "grad_norm": 0.8602406978607178,
+      "learning_rate": 0.0002,
+      "loss": 1.441,
+      "step": 6620
+    },
+    {
+      "epoch": 4.442211055276382,
+      "grad_norm": 0.7549102902412415,
+      "learning_rate": 0.0002,
+      "loss": 1.4319,
+      "step": 6630
+    },
+    {
+      "epoch": 4.44891122278057,
+      "grad_norm": 0.7831804156303406,
+      "learning_rate": 0.0002,
+      "loss": 1.388,
+      "step": 6640
+    },
+    {
+      "epoch": 4.455611390284757,
+      "grad_norm": 0.7269673943519592,
+      "learning_rate": 0.0002,
+      "loss": 1.4283,
+      "step": 6650
+    },
+    {
+      "epoch": 4.4623115577889445,
+      "grad_norm": 0.7397838830947876,
+      "learning_rate": 0.0002,
+      "loss": 1.4132,
+      "step": 6660
+    },
+    {
+      "epoch": 4.469011725293132,
+      "grad_norm": 0.713707447052002,
+      "learning_rate": 0.0002,
+      "loss": 1.3174,
+      "step": 6670
+    },
+    {
+      "epoch": 4.47571189279732,
+      "grad_norm": 0.7525581121444702,
+      "learning_rate": 0.0002,
+      "loss": 1.3406,
+      "step": 6680
+    },
+    {
+      "epoch": 4.482412060301508,
+      "grad_norm": 0.8030191659927368,
+      "learning_rate": 0.0002,
+      "loss": 1.4283,
+      "step": 6690
+    },
+    {
+      "epoch": 4.489112227805695,
+      "grad_norm": 0.7469439506530762,
+      "learning_rate": 0.0002,
+      "loss": 1.4586,
+      "step": 6700
+    },
+    {
+      "epoch": 4.4958123953098825,
+      "grad_norm": 0.7743868231773376,
+      "learning_rate": 0.0002,
+      "loss": 1.367,
+      "step": 6710
+    },
+    {
+      "epoch": 4.50251256281407,
+      "grad_norm": 0.6539737582206726,
+      "learning_rate": 0.0002,
+      "loss": 1.3439,
+      "step": 6720
+    },
+    {
+      "epoch": 4.509212730318258,
+      "grad_norm": 0.825818657875061,
+      "learning_rate": 0.0002,
+      "loss": 1.4513,
+      "step": 6730
+    },
+    {
+      "epoch": 4.515912897822446,
+      "grad_norm": 0.8048575520515442,
+      "learning_rate": 0.0002,
+      "loss": 1.3984,
+      "step": 6740
+    },
+    {
+      "epoch": 4.522613065326633,
+      "grad_norm": 0.7828766107559204,
+      "learning_rate": 0.0002,
+      "loss": 1.3923,
+      "step": 6750
+    },
+    {
+      "epoch": 4.5293132328308205,
+      "grad_norm": 0.7406010031700134,
+      "learning_rate": 0.0002,
+      "loss": 1.3886,
+      "step": 6760
+    },
+    {
+      "epoch": 4.536013400335008,
+      "grad_norm": 0.840345561504364,
+      "learning_rate": 0.0002,
+      "loss": 1.3109,
+      "step": 6770
+    },
+    {
+      "epoch": 4.542713567839196,
+      "grad_norm": 0.8492622971534729,
+      "learning_rate": 0.0002,
+      "loss": 1.4808,
+      "step": 6780
+    },
+    {
+      "epoch": 4.549413735343384,
+      "grad_norm": 0.7130163908004761,
+      "learning_rate": 0.0002,
+      "loss": 1.4384,
+      "step": 6790
+    },
+    {
+      "epoch": 4.556113902847571,
+      "grad_norm": 0.8454728126525879,
+      "learning_rate": 0.0002,
+      "loss": 1.4531,
+      "step": 6800
+    },
+    {
+      "epoch": 4.562814070351759,
+      "grad_norm": 0.7847645282745361,
+      "learning_rate": 0.0002,
+      "loss": 1.3239,
+      "step": 6810
+    },
+    {
+      "epoch": 4.569514237855946,
+      "grad_norm": 0.7245864272117615,
+      "learning_rate": 0.0002,
+      "loss": 1.4181,
+      "step": 6820
+    },
+    {
+      "epoch": 4.576214405360134,
+      "grad_norm": 0.768893301486969,
+      "learning_rate": 0.0002,
+      "loss": 1.3233,
+      "step": 6830
+    },
+    {
+      "epoch": 4.582914572864322,
+      "grad_norm": 0.8028400540351868,
+      "learning_rate": 0.0002,
+      "loss": 1.3932,
+      "step": 6840
+    },
+    {
+      "epoch": 4.589614740368509,
+      "grad_norm": 0.763945460319519,
+      "learning_rate": 0.0002,
+      "loss": 1.3745,
+      "step": 6850
+    },
+    {
+      "epoch": 4.596314907872697,
+      "grad_norm": 0.7417685389518738,
+      "learning_rate": 0.0002,
+      "loss": 1.4797,
+      "step": 6860
+    },
+    {
+      "epoch": 4.603015075376884,
+      "grad_norm": 0.7603038549423218,
+      "learning_rate": 0.0002,
+      "loss": 1.4468,
+      "step": 6870
+    },
+    {
+      "epoch": 4.609715242881072,
+      "grad_norm": 0.7981528043746948,
+      "learning_rate": 0.0002,
+      "loss": 1.4095,
+      "step": 6880
+    },
+    {
+      "epoch": 4.61641541038526,
+      "grad_norm": 0.8077111840248108,
+      "learning_rate": 0.0002,
+      "loss": 1.3963,
+      "step": 6890
+    },
+    {
+      "epoch": 4.623115577889447,
+      "grad_norm": 0.8778454065322876,
+      "learning_rate": 0.0002,
+      "loss": 1.4721,
+      "step": 6900
+    },
+    {
+      "epoch": 4.629815745393635,
+      "grad_norm": 0.8620710372924805,
+      "learning_rate": 0.0002,
+      "loss": 1.3758,
+      "step": 6910
+    },
+    {
+      "epoch": 4.636515912897822,
+      "grad_norm": 0.7486072778701782,
+      "learning_rate": 0.0002,
+      "loss": 1.344,
+      "step": 6920
+    },
+    {
+      "epoch": 4.64321608040201,
+      "grad_norm": 0.7493042945861816,
+      "learning_rate": 0.0002,
+      "loss": 1.3913,
+      "step": 6930
+    },
+    {
+      "epoch": 4.649916247906198,
+      "grad_norm": 0.7388978600502014,
+      "learning_rate": 0.0002,
+      "loss": 1.397,
+      "step": 6940
+    },
+    {
+      "epoch": 4.656616415410385,
+      "grad_norm": 0.798530638217926,
+      "learning_rate": 0.0002,
+      "loss": 1.3593,
+      "step": 6950
+    },
+    {
+      "epoch": 4.663316582914573,
+      "grad_norm": 0.7929500937461853,
+      "learning_rate": 0.0002,
+      "loss": 1.3982,
+      "step": 6960
+    },
+    {
+      "epoch": 4.67001675041876,
+      "grad_norm": 0.9186785221099854,
+      "learning_rate": 0.0002,
+      "loss": 1.4183,
+      "step": 6970
+    },
+    {
+      "epoch": 4.676716917922948,
+      "grad_norm": 1.1103485822677612,
+      "learning_rate": 0.0002,
+      "loss": 1.3955,
+      "step": 6980
+    },
+    {
+      "epoch": 4.683417085427136,
+      "grad_norm": 0.8000466823577881,
+      "learning_rate": 0.0002,
+      "loss": 1.3941,
+      "step": 6990
+    },
+    {
+      "epoch": 4.690117252931323,
+      "grad_norm": 0.7520599961280823,
+      "learning_rate": 0.0002,
+      "loss": 1.371,
+      "step": 7000
+    },
+    {
+      "epoch": 4.696817420435511,
+      "grad_norm": 0.7971973419189453,
+      "learning_rate": 0.0002,
+      "loss": 1.4582,
+      "step": 7010
+    },
+    {
+      "epoch": 4.703517587939698,
+      "grad_norm": 0.7363343834877014,
+      "learning_rate": 0.0002,
+      "loss": 1.3682,
+      "step": 7020
+    },
+    {
+      "epoch": 4.710217755443886,
+      "grad_norm": 0.8268865942955017,
+      "learning_rate": 0.0002,
+      "loss": 1.3889,
+      "step": 7030
+    },
+    {
+      "epoch": 4.716917922948074,
+      "grad_norm": 0.7054963111877441,
+      "learning_rate": 0.0002,
+      "loss": 1.4382,
+      "step": 7040
+    },
+    {
+      "epoch": 4.723618090452261,
+      "grad_norm": 0.8196262121200562,
+      "learning_rate": 0.0002,
+      "loss": 1.4578,
+      "step": 7050
+    },
+    {
+      "epoch": 4.730318257956449,
+      "grad_norm": 0.8276031017303467,
+      "learning_rate": 0.0002,
+      "loss": 1.365,
+      "step": 7060
+    },
+    {
+      "epoch": 4.7370184254606365,
+      "grad_norm": 0.8248157501220703,
+      "learning_rate": 0.0002,
+      "loss": 1.3887,
+      "step": 7070
+    },
+    {
+      "epoch": 4.743718592964824,
+      "grad_norm": 0.8937979936599731,
+      "learning_rate": 0.0002,
+      "loss": 1.4193,
+      "step": 7080
+    },
+    {
+      "epoch": 4.750418760469012,
+      "grad_norm": 0.7984238862991333,
+      "learning_rate": 0.0002,
+      "loss": 1.4334,
+      "step": 7090
+    },
+    {
+      "epoch": 4.757118927973199,
+      "grad_norm": 0.9495313763618469,
+      "learning_rate": 0.0002,
+      "loss": 1.4385,
+      "step": 7100
+    },
+    {
+      "epoch": 4.763819095477387,
+      "grad_norm": 0.8598204255104065,
+      "learning_rate": 0.0002,
+      "loss": 1.4504,
+      "step": 7110
+    },
+    {
+      "epoch": 4.7705192629815745,
+      "grad_norm": 0.8951472640037537,
+      "learning_rate": 0.0002,
+      "loss": 1.3969,
+      "step": 7120
+    },
+    {
+      "epoch": 4.777219430485762,
+      "grad_norm": 0.9110309481620789,
+      "learning_rate": 0.0002,
+      "loss": 1.4339,
+      "step": 7130
+    },
+    {
+      "epoch": 4.78391959798995,
+      "grad_norm": 0.7929584980010986,
+      "learning_rate": 0.0002,
+      "loss": 1.4001,
+      "step": 7140
+    },
+    {
+      "epoch": 4.790619765494137,
+      "grad_norm": 0.7415322661399841,
+      "learning_rate": 0.0002,
+      "loss": 1.467,
+      "step": 7150
+    },
+    {
+      "epoch": 4.797319932998325,
+      "grad_norm": 0.7504757046699524,
+      "learning_rate": 0.0002,
+      "loss": 1.5107,
+      "step": 7160
+    },
+    {
+      "epoch": 4.8040201005025125,
+      "grad_norm": 0.7166924476623535,
+      "learning_rate": 0.0002,
+      "loss": 1.3736,
+      "step": 7170
+    },
+    {
+      "epoch": 4.8107202680067,
+      "grad_norm": 0.7728400826454163,
+      "learning_rate": 0.0002,
+      "loss": 1.4088,
+      "step": 7180
+    },
+    {
+      "epoch": 4.817420435510888,
+      "grad_norm": 0.7992154955863953,
+      "learning_rate": 0.0002,
+      "loss": 1.3814,
+      "step": 7190
+    },
+    {
+      "epoch": 4.824120603015075,
+      "grad_norm": 0.8655321002006531,
+      "learning_rate": 0.0002,
+      "loss": 1.3958,
+      "step": 7200
+    },
+    {
+      "epoch": 4.830820770519263,
+      "grad_norm": 0.7672632336616516,
+      "learning_rate": 0.0002,
+      "loss": 1.3837,
+      "step": 7210
+    },
+    {
+      "epoch": 4.8375209380234505,
+      "grad_norm": 0.708416223526001,
+      "learning_rate": 0.0002,
+      "loss": 1.4578,
+      "step": 7220
+    },
+    {
+      "epoch": 4.844221105527638,
+      "grad_norm": 0.8914081454277039,
+      "learning_rate": 0.0002,
+      "loss": 1.5413,
+      "step": 7230
+    },
+    {
+      "epoch": 4.850921273031826,
+      "grad_norm": 0.7141931653022766,
+      "learning_rate": 0.0002,
+      "loss": 1.3569,
+      "step": 7240
+    },
+    {
+      "epoch": 4.857621440536013,
+      "grad_norm": 0.6913040280342102,
+      "learning_rate": 0.0002,
+      "loss": 1.4532,
+      "step": 7250
+    },
+    {
+      "epoch": 4.864321608040201,
+      "grad_norm": 0.7871233820915222,
+      "learning_rate": 0.0002,
+      "loss": 1.3912,
+      "step": 7260
+    },
+    {
+      "epoch": 4.8710217755443885,
+      "grad_norm": 0.8466277122497559,
+      "learning_rate": 0.0002,
+      "loss": 1.3688,
+      "step": 7270
+    },
+    {
+      "epoch": 4.877721943048576,
+      "grad_norm": 0.8492183685302734,
+      "learning_rate": 0.0002,
+      "loss": 1.33,
+      "step": 7280
+    },
+    {
+      "epoch": 4.884422110552764,
+      "grad_norm": 0.8339574933052063,
+      "learning_rate": 0.0002,
+      "loss": 1.3744,
+      "step": 7290
+    },
+    {
+      "epoch": 4.891122278056951,
+      "grad_norm": 0.787022590637207,
+      "learning_rate": 0.0002,
+      "loss": 1.4157,
+      "step": 7300
+    },
+    {
+      "epoch": 4.897822445561139,
+      "grad_norm": 0.8877332806587219,
+      "learning_rate": 0.0002,
+      "loss": 1.3725,
+      "step": 7310
+    },
+    {
+      "epoch": 4.9045226130653266,
+      "grad_norm": 0.744989812374115,
+      "learning_rate": 0.0002,
+      "loss": 1.3968,
+      "step": 7320
+    },
+    {
+      "epoch": 4.911222780569514,
+      "grad_norm": 0.8027268648147583,
+      "learning_rate": 0.0002,
+      "loss": 1.4421,
+      "step": 7330
+    },
+    {
+      "epoch": 4.917922948073702,
+      "grad_norm": 0.6437455415725708,
+      "learning_rate": 0.0002,
+      "loss": 1.425,
+      "step": 7340
+    },
+    {
+      "epoch": 4.924623115577889,
+      "grad_norm": 0.685999870300293,
+      "learning_rate": 0.0002,
+      "loss": 1.4829,
+      "step": 7350
+    },
+    {
+      "epoch": 4.931323283082077,
+      "grad_norm": 0.9086187481880188,
+      "learning_rate": 0.0002,
+      "loss": 1.4352,
+      "step": 7360
+    },
+    {
+      "epoch": 4.938023450586265,
+      "grad_norm": 0.8272411227226257,
+      "learning_rate": 0.0002,
+      "loss": 1.4245,
+      "step": 7370
+    },
+    {
+      "epoch": 4.944723618090452,
+      "grad_norm": 0.9227852821350098,
+      "learning_rate": 0.0002,
+      "loss": 1.4226,
+      "step": 7380
+    },
+    {
+      "epoch": 4.95142378559464,
+      "grad_norm": 0.7688441276550293,
+      "learning_rate": 0.0002,
+      "loss": 1.3643,
+      "step": 7390
+    },
+    {
+      "epoch": 4.958123953098827,
+      "grad_norm": 0.8662643432617188,
+      "learning_rate": 0.0002,
+      "loss": 1.4491,
+      "step": 7400
+    },
+    {
+      "epoch": 4.964824120603015,
+      "grad_norm": 0.9234127998352051,
+      "learning_rate": 0.0002,
+      "loss": 1.4194,
+      "step": 7410
+    },
+    {
+      "epoch": 4.971524288107203,
+      "grad_norm": 0.9131470918655396,
+      "learning_rate": 0.0002,
+      "loss": 1.4009,
+      "step": 7420
+    },
+    {
+      "epoch": 4.97822445561139,
+      "grad_norm": 0.7377504110336304,
+      "learning_rate": 0.0002,
+      "loss": 1.4544,
+      "step": 7430
+    },
+    {
+      "epoch": 4.984924623115578,
+      "grad_norm": 0.8762801289558411,
+      "learning_rate": 0.0002,
+      "loss": 1.4008,
+      "step": 7440
+    },
+    {
+      "epoch": 4.991624790619765,
+      "grad_norm": 0.7919872999191284,
+      "learning_rate": 0.0002,
+      "loss": 1.4304,
+      "step": 7450
+    },
+    {
+      "epoch": 4.998324958123953,
+      "grad_norm": 0.7144299149513245,
+      "learning_rate": 0.0002,
+      "loss": 1.3817,
+      "step": 7460
+    },
+    {
+      "epoch": 4.99966499162479,
+      "eval_loss": 1.9291157722473145,
+      "eval_runtime": 37.9831,
+      "eval_samples_per_second": 13.559,
+      "eval_steps_per_second": 1.711,
+      "step": 7462
+    },
+    {
+      "epoch": 5.005025125628141,
+      "grad_norm": 0.7860151529312134,
+      "learning_rate": 0.0002,
+      "loss": 1.2753,
+      "step": 7470
+    },
+    {
+      "epoch": 5.011725293132328,
+      "grad_norm": 0.9418314695358276,
+      "learning_rate": 0.0002,
+      "loss": 1.2149,
+      "step": 7480
+    },
+    {
+      "epoch": 5.018425460636516,
+      "grad_norm": 0.8474572896957397,
+      "learning_rate": 0.0002,
+      "loss": 1.1966,
+      "step": 7490
+    },
+    {
+      "epoch": 5.025125628140704,
+      "grad_norm": 1.0724040269851685,
+      "learning_rate": 0.0002,
+      "loss": 1.2111,
+      "step": 7500
+    },
+    {
+      "epoch": 5.031825795644891,
+      "grad_norm": 0.9109148979187012,
+      "learning_rate": 0.0002,
+      "loss": 1.2228,
+      "step": 7510
+    },
+    {
+      "epoch": 5.038525963149079,
+      "grad_norm": 1.0088659524917603,
+      "learning_rate": 0.0002,
+      "loss": 1.2239,
+      "step": 7520
+    },
+    {
+      "epoch": 5.045226130653266,
+      "grad_norm": 1.1421623229980469,
+      "learning_rate": 0.0002,
+      "loss": 1.2156,
+      "step": 7530
+    },
+    {
+      "epoch": 5.051926298157454,
+      "grad_norm": 0.9219902157783508,
+      "learning_rate": 0.0002,
+      "loss": 1.1739,
+      "step": 7540
+    },
+    {
+      "epoch": 5.058626465661642,
+      "grad_norm": 0.9150987863540649,
+      "learning_rate": 0.0002,
+      "loss": 1.2686,
+      "step": 7550
+    },
+    {
+      "epoch": 5.065326633165829,
+      "grad_norm": 0.8889328241348267,
+      "learning_rate": 0.0002,
+      "loss": 1.2068,
+      "step": 7560
+    },
+    {
+      "epoch": 5.072026800670017,
+      "grad_norm": 0.9751363396644592,
+      "learning_rate": 0.0002,
+      "loss": 1.276,
+      "step": 7570
+    },
+    {
+      "epoch": 5.078726968174204,
+      "grad_norm": 0.8603123426437378,
+      "learning_rate": 0.0002,
+      "loss": 1.2078,
+      "step": 7580
+    },
+    {
+      "epoch": 5.085427135678392,
+      "grad_norm": 0.8910616636276245,
+      "learning_rate": 0.0002,
+      "loss": 1.2175,
+      "step": 7590
+    },
+    {
+      "epoch": 5.09212730318258,
+      "grad_norm": 1.1128392219543457,
+      "learning_rate": 0.0002,
+      "loss": 1.2475,
+      "step": 7600
+    },
+    {
+      "epoch": 5.098827470686767,
+      "grad_norm": 0.9480258822441101,
+      "learning_rate": 0.0002,
+      "loss": 1.3065,
+      "step": 7610
+    },
+    {
+      "epoch": 5.105527638190955,
+      "grad_norm": 0.906958818435669,
+      "learning_rate": 0.0002,
+      "loss": 1.193,
+      "step": 7620
+    },
+    {
+      "epoch": 5.1122278056951425,
+      "grad_norm": 0.8741167187690735,
+      "learning_rate": 0.0002,
+      "loss": 1.2223,
+      "step": 7630
+    },
+    {
+      "epoch": 5.11892797319933,
+      "grad_norm": 0.966268002986908,
+      "learning_rate": 0.0002,
+      "loss": 1.2126,
+      "step": 7640
+    },
+    {
+      "epoch": 5.125628140703517,
+      "grad_norm": 0.9124358892440796,
+      "learning_rate": 0.0002,
+      "loss": 1.2782,
+      "step": 7650
+    },
+    {
+      "epoch": 5.132328308207705,
+      "grad_norm": 1.0436606407165527,
+      "learning_rate": 0.0002,
+      "loss": 1.3004,
+      "step": 7660
+    },
+    {
+      "epoch": 5.139028475711893,
+      "grad_norm": 0.9217309355735779,
+      "learning_rate": 0.0002,
+      "loss": 1.2675,
+      "step": 7670
+    },
+    {
+      "epoch": 5.1457286432160805,
+      "grad_norm": 1.344765543937683,
+      "learning_rate": 0.0002,
+      "loss": 1.2502,
+      "step": 7680
+    },
+    {
+      "epoch": 5.152428810720268,
+      "grad_norm": 1.0730723142623901,
+      "learning_rate": 0.0002,
+      "loss": 1.2416,
+      "step": 7690
+    },
+    {
+      "epoch": 5.159128978224456,
+      "grad_norm": 0.9321247339248657,
+      "learning_rate": 0.0002,
+      "loss": 1.1888,
+      "step": 7700
+    },
+    {
+      "epoch": 5.165829145728643,
+      "grad_norm": 0.8482614755630493,
+      "learning_rate": 0.0002,
+      "loss": 1.1941,
+      "step": 7710
+    },
+    {
+      "epoch": 5.172529313232831,
+      "grad_norm": 0.8274452686309814,
+      "learning_rate": 0.0002,
+      "loss": 1.2668,
+      "step": 7720
+    },
+    {
+      "epoch": 5.1792294807370185,
+      "grad_norm": 0.9120376706123352,
+      "learning_rate": 0.0002,
+      "loss": 1.1972,
+      "step": 7730
+    },
+    {
+      "epoch": 5.185929648241206,
+      "grad_norm": 1.0062892436981201,
+      "learning_rate": 0.0002,
+      "loss": 1.1648,
+      "step": 7740
+    },
+    {
+      "epoch": 5.192629815745394,
+      "grad_norm": 0.9521504640579224,
+      "learning_rate": 0.0002,
+      "loss": 1.2199,
+      "step": 7750
+    },
+    {
+      "epoch": 5.199329983249581,
+      "grad_norm": 0.8800198435783386,
+      "learning_rate": 0.0002,
+      "loss": 1.2855,
+      "step": 7760
+    },
+    {
+      "epoch": 5.206030150753769,
+      "grad_norm": 0.9749179482460022,
+      "learning_rate": 0.0002,
+      "loss": 1.2535,
+      "step": 7770
+    },
+    {
+      "epoch": 5.2127303182579565,
+      "grad_norm": 0.9441686868667603,
+      "learning_rate": 0.0002,
+      "loss": 1.2975,
+      "step": 7780
+    },
+    {
+      "epoch": 5.219430485762144,
+      "grad_norm": 0.9114066362380981,
+      "learning_rate": 0.0002,
+      "loss": 1.256,
+      "step": 7790
+    },
+    {
+      "epoch": 5.226130653266332,
+      "grad_norm": 0.9851446151733398,
+      "learning_rate": 0.0002,
+      "loss": 1.2621,
+      "step": 7800
+    },
+    {
+      "epoch": 5.232830820770519,
+      "grad_norm": 0.9526297450065613,
+      "learning_rate": 0.0002,
+      "loss": 1.2502,
+      "step": 7810
+    },
+    {
+      "epoch": 5.239530988274707,
+      "grad_norm": 1.05986487865448,
+      "learning_rate": 0.0002,
+      "loss": 1.1502,
+      "step": 7820
+    },
+    {
+      "epoch": 5.2462311557788945,
+      "grad_norm": 0.8956538438796997,
+      "learning_rate": 0.0002,
+      "loss": 1.2517,
+      "step": 7830
+    },
+    {
+      "epoch": 5.252931323283082,
+      "grad_norm": 0.9568153619766235,
+      "learning_rate": 0.0002,
+      "loss": 1.2556,
+      "step": 7840
+    },
+    {
+      "epoch": 5.259631490787269,
+      "grad_norm": 1.0035018920898438,
+      "learning_rate": 0.0002,
+      "loss": 1.2442,
+      "step": 7850
+    },
+    {
+      "epoch": 5.266331658291457,
+      "grad_norm": 0.8554368615150452,
+      "learning_rate": 0.0002,
+      "loss": 1.2605,
+      "step": 7860
+    },
+    {
+      "epoch": 5.273031825795645,
+      "grad_norm": 0.9677708148956299,
+      "learning_rate": 0.0002,
+      "loss": 1.2799,
+      "step": 7870
+    },
+    {
+      "epoch": 5.279731993299833,
+      "grad_norm": 0.943606436252594,
+      "learning_rate": 0.0002,
+      "loss": 1.275,
+      "step": 7880
+    },
+    {
+      "epoch": 5.28643216080402,
+      "grad_norm": 1.0029335021972656,
+      "learning_rate": 0.0002,
+      "loss": 1.2335,
+      "step": 7890
+    },
+    {
+      "epoch": 5.293132328308207,
+      "grad_norm": 1.0164015293121338,
+      "learning_rate": 0.0002,
+      "loss": 1.2494,
+      "step": 7900
+    },
+    {
+      "epoch": 5.299832495812395,
+      "grad_norm": 0.8908365368843079,
+      "learning_rate": 0.0002,
+      "loss": 1.3117,
+      "step": 7910
+    },
+    {
+      "epoch": 5.306532663316583,
+      "grad_norm": 0.9307826161384583,
+      "learning_rate": 0.0002,
+      "loss": 1.2832,
+      "step": 7920
+    },
+    {
+      "epoch": 5.313232830820771,
+      "grad_norm": 1.0730371475219727,
+      "learning_rate": 0.0002,
+      "loss": 1.242,
+      "step": 7930
+    },
+    {
+      "epoch": 5.319932998324958,
+      "grad_norm": 0.844739556312561,
+      "learning_rate": 0.0002,
+      "loss": 1.2003,
+      "step": 7940
+    },
+    {
+      "epoch": 5.326633165829146,
+      "grad_norm": 1.275833010673523,
+      "learning_rate": 0.0002,
+      "loss": 1.2688,
+      "step": 7950
+    },
+    {
+      "epoch": 5.333333333333333,
+      "grad_norm": 0.9042661190032959,
+      "learning_rate": 0.0002,
+      "loss": 1.2957,
+      "step": 7960
+    },
+    {
+      "epoch": 5.340033500837521,
+      "grad_norm": 0.9374269247055054,
+      "learning_rate": 0.0002,
+      "loss": 1.2912,
+      "step": 7970
+    },
+    {
+      "epoch": 5.346733668341709,
+      "grad_norm": 1.033098578453064,
+      "learning_rate": 0.0002,
+      "loss": 1.2721,
+      "step": 7980
+    },
+    {
+      "epoch": 5.353433835845896,
+      "grad_norm": 1.062775731086731,
+      "learning_rate": 0.0002,
+      "loss": 1.3208,
+      "step": 7990
+    },
+    {
+      "epoch": 5.360134003350084,
+      "grad_norm": 1.1064317226409912,
+      "learning_rate": 0.0002,
+      "loss": 1.3065,
+      "step": 8000
+    },
+    {
+      "epoch": 5.366834170854271,
+      "grad_norm": 1.1114039421081543,
+      "learning_rate": 0.0002,
+      "loss": 1.2341,
+      "step": 8010
+    },
+    {
+      "epoch": 5.373534338358459,
+      "grad_norm": 1.0198014974594116,
+      "learning_rate": 0.0002,
+      "loss": 1.2255,
+      "step": 8020
+    },
+    {
+      "epoch": 5.380234505862647,
+      "grad_norm": 0.8443173170089722,
+      "learning_rate": 0.0002,
+      "loss": 1.2433,
+      "step": 8030
+    },
+    {
+      "epoch": 5.386934673366834,
+      "grad_norm": 1.000881314277649,
+      "learning_rate": 0.0002,
+      "loss": 1.206,
+      "step": 8040
+    },
+    {
+      "epoch": 5.393634840871022,
+      "grad_norm": 0.9874443411827087,
+      "learning_rate": 0.0002,
+      "loss": 1.2982,
+      "step": 8050
+    },
+    {
+      "epoch": 5.400335008375209,
+      "grad_norm": 0.9895344972610474,
+      "learning_rate": 0.0002,
+      "loss": 1.2289,
+      "step": 8060
+    },
+    {
+      "epoch": 5.407035175879397,
+      "grad_norm": 0.8595236539840698,
+      "learning_rate": 0.0002,
+      "loss": 1.249,
+      "step": 8070
+    },
+    {
+      "epoch": 5.413735343383585,
+      "grad_norm": 0.9523849487304688,
+      "learning_rate": 0.0002,
+      "loss": 1.2308,
+      "step": 8080
+    },
+    {
+      "epoch": 5.420435510887772,
+      "grad_norm": 1.0560476779937744,
+      "learning_rate": 0.0002,
+      "loss": 1.2343,
+      "step": 8090
+    },
+    {
+      "epoch": 5.42713567839196,
+      "grad_norm": 1.0893689393997192,
+      "learning_rate": 0.0002,
+      "loss": 1.2956,
+      "step": 8100
+    },
+    {
+      "epoch": 5.433835845896147,
+      "grad_norm": 0.9395513534545898,
+      "learning_rate": 0.0002,
+      "loss": 1.2846,
+      "step": 8110
+    },
+    {
+      "epoch": 5.440536013400335,
+      "grad_norm": 0.9364215135574341,
+      "learning_rate": 0.0002,
+      "loss": 1.3444,
+      "step": 8120
+    },
+    {
+      "epoch": 5.447236180904523,
+      "grad_norm": 0.9502208232879639,
+      "learning_rate": 0.0002,
+      "loss": 1.2944,
+      "step": 8130
+    },
+    {
+      "epoch": 5.45393634840871,
+      "grad_norm": 0.9559208154678345,
+      "learning_rate": 0.0002,
+      "loss": 1.2971,
+      "step": 8140
+    },
+    {
+      "epoch": 5.460636515912898,
+      "grad_norm": 0.9261730313301086,
+      "learning_rate": 0.0002,
+      "loss": 1.2495,
+      "step": 8150
+    },
+    {
+      "epoch": 5.467336683417085,
+      "grad_norm": 0.9832326173782349,
+      "learning_rate": 0.0002,
+      "loss": 1.2599,
+      "step": 8160
+    },
+    {
+      "epoch": 5.474036850921273,
+      "grad_norm": 1.065953016281128,
+      "learning_rate": 0.0002,
+      "loss": 1.2771,
+      "step": 8170
+    },
+    {
+      "epoch": 5.480737018425461,
+      "grad_norm": 0.9139469861984253,
+      "learning_rate": 0.0002,
+      "loss": 1.3617,
+      "step": 8180
+    },
+    {
+      "epoch": 5.4874371859296485,
+      "grad_norm": 1.2322484254837036,
+      "learning_rate": 0.0002,
+      "loss": 1.2197,
+      "step": 8190
+    },
+    {
+      "epoch": 5.494137353433836,
+      "grad_norm": 0.9722974896430969,
+      "learning_rate": 0.0002,
+      "loss": 1.2879,
+      "step": 8200
+    },
+    {
+      "epoch": 5.500837520938023,
+      "grad_norm": 0.9338926076889038,
+      "learning_rate": 0.0002,
+      "loss": 1.2664,
+      "step": 8210
+    },
+    {
+      "epoch": 5.507537688442211,
+      "grad_norm": 0.9283728003501892,
+      "learning_rate": 0.0002,
+      "loss": 1.2128,
+      "step": 8220
+    },
+    {
+      "epoch": 5.514237855946399,
+      "grad_norm": 1.0489585399627686,
+      "learning_rate": 0.0002,
+      "loss": 1.2141,
+      "step": 8230
+    },
+    {
+      "epoch": 5.5209380234505865,
+      "grad_norm": 0.9881814122200012,
+      "learning_rate": 0.0002,
+      "loss": 1.3257,
+      "step": 8240
+    },
+    {
+      "epoch": 5.527638190954773,
+      "grad_norm": 0.9274460077285767,
+      "learning_rate": 0.0002,
+      "loss": 1.2221,
+      "step": 8250
+    },
+    {
+      "epoch": 5.534338358458961,
+      "grad_norm": 0.8650718331336975,
+      "learning_rate": 0.0002,
+      "loss": 1.2241,
+      "step": 8260
+    },
+    {
+      "epoch": 5.541038525963149,
+      "grad_norm": 1.014069676399231,
+      "learning_rate": 0.0002,
+      "loss": 1.2462,
+      "step": 8270
+    },
+    {
+      "epoch": 5.547738693467337,
+      "grad_norm": 0.9212974905967712,
+      "learning_rate": 0.0002,
+      "loss": 1.3502,
+      "step": 8280
+    },
+    {
+      "epoch": 5.5544388609715245,
+      "grad_norm": 1.1235398054122925,
+      "learning_rate": 0.0002,
+      "loss": 1.2779,
+      "step": 8290
+    },
+    {
+      "epoch": 5.561139028475711,
+      "grad_norm": 0.961954653263092,
+      "learning_rate": 0.0002,
+      "loss": 1.306,
+      "step": 8300
+    },
+    {
+      "epoch": 5.567839195979899,
+      "grad_norm": 0.9386700391769409,
+      "learning_rate": 0.0002,
+      "loss": 1.2946,
+      "step": 8310
+    },
+    {
+      "epoch": 5.574539363484087,
+      "grad_norm": 1.01912522315979,
+      "learning_rate": 0.0002,
+      "loss": 1.313,
+      "step": 8320
+    },
+    {
+      "epoch": 5.581239530988275,
+      "grad_norm": 0.9851216077804565,
+      "learning_rate": 0.0002,
+      "loss": 1.3121,
+      "step": 8330
+    },
+    {
+      "epoch": 5.5879396984924625,
+      "grad_norm": 1.0138001441955566,
+      "learning_rate": 0.0002,
+      "loss": 1.3071,
+      "step": 8340
+    },
+    {
+      "epoch": 5.594639865996649,
+      "grad_norm": 0.9262447357177734,
+      "learning_rate": 0.0002,
+      "loss": 1.2257,
+      "step": 8350
+    },
+    {
+      "epoch": 5.601340033500837,
+      "grad_norm": 1.1322970390319824,
+      "learning_rate": 0.0002,
+      "loss": 1.2473,
+      "step": 8360
+    },
+    {
+      "epoch": 5.608040201005025,
+      "grad_norm": 1.1429349184036255,
+      "learning_rate": 0.0002,
+      "loss": 1.3098,
+      "step": 8370
+    },
+    {
+      "epoch": 5.614740368509213,
+      "grad_norm": 0.9130118489265442,
+      "learning_rate": 0.0002,
+      "loss": 1.2686,
+      "step": 8380
+    },
+    {
+      "epoch": 5.6214405360134005,
+      "grad_norm": 0.9651545882225037,
+      "learning_rate": 0.0002,
+      "loss": 1.2541,
+      "step": 8390
+    },
+    {
+      "epoch": 5.628140703517588,
+      "grad_norm": 0.9595398306846619,
+      "learning_rate": 0.0002,
+      "loss": 1.2799,
+      "step": 8400
+    },
+    {
+      "epoch": 5.634840871021775,
+      "grad_norm": 1.0049372911453247,
+      "learning_rate": 0.0002,
+      "loss": 1.3429,
+      "step": 8410
+    },
+    {
+      "epoch": 5.641541038525963,
+      "grad_norm": 1.082804560661316,
+      "learning_rate": 0.0002,
+      "loss": 1.3224,
+      "step": 8420
+    },
+    {
+      "epoch": 5.648241206030151,
+      "grad_norm": 0.9489204287528992,
+      "learning_rate": 0.0002,
+      "loss": 1.297,
+      "step": 8430
+    },
+    {
+      "epoch": 5.654941373534339,
+      "grad_norm": 0.9470235109329224,
+      "learning_rate": 0.0002,
+      "loss": 1.3424,
+      "step": 8440
+    },
+    {
+      "epoch": 5.661641541038526,
+      "grad_norm": 1.0662927627563477,
+      "learning_rate": 0.0002,
+      "loss": 1.3358,
+      "step": 8450
+    },
+    {
+      "epoch": 5.668341708542713,
+      "grad_norm": 0.9097877740859985,
+      "learning_rate": 0.0002,
+      "loss": 1.2973,
+      "step": 8460
+    },
+    {
+      "epoch": 5.675041876046901,
+      "grad_norm": 0.9740368127822876,
+      "learning_rate": 0.0002,
+      "loss": 1.3072,
+      "step": 8470
+    },
+    {
+      "epoch": 5.681742043551089,
+      "grad_norm": 0.9878810048103333,
+      "learning_rate": 0.0002,
+      "loss": 1.286,
+      "step": 8480
+    },
+    {
+      "epoch": 5.688442211055277,
+      "grad_norm": 1.148260474205017,
+      "learning_rate": 0.0002,
+      "loss": 1.208,
+      "step": 8490
+    },
+    {
+      "epoch": 5.695142378559464,
+      "grad_norm": 0.9632558822631836,
+      "learning_rate": 0.0002,
+      "loss": 1.2842,
+      "step": 8500
+    },
+    {
+      "epoch": 5.701842546063651,
+      "grad_norm": 0.876812756061554,
+      "learning_rate": 0.0002,
+      "loss": 1.2787,
+      "step": 8510
+    },
+    {
+      "epoch": 5.708542713567839,
+      "grad_norm": 1.0730829238891602,
+      "learning_rate": 0.0002,
+      "loss": 1.3186,
+      "step": 8520
+    },
+    {
+      "epoch": 5.715242881072027,
+      "grad_norm": 1.2239218950271606,
+      "learning_rate": 0.0002,
+      "loss": 1.2856,
+      "step": 8530
+    },
+    {
+      "epoch": 5.721943048576215,
+      "grad_norm": 0.9460835456848145,
+      "learning_rate": 0.0002,
+      "loss": 1.2717,
+      "step": 8540
+    },
+    {
+      "epoch": 5.728643216080402,
+      "grad_norm": 0.9086270928382874,
+      "learning_rate": 0.0002,
+      "loss": 1.3509,
+      "step": 8550
+    },
+    {
+      "epoch": 5.735343383584589,
+      "grad_norm": 1.0258867740631104,
+      "learning_rate": 0.0002,
+      "loss": 1.2971,
+      "step": 8560
+    },
+    {
+      "epoch": 5.742043551088777,
+      "grad_norm": 1.0543923377990723,
+      "learning_rate": 0.0002,
+      "loss": 1.3581,
+      "step": 8570
+    },
+    {
+      "epoch": 5.748743718592965,
+      "grad_norm": 0.9063900113105774,
+      "learning_rate": 0.0002,
+      "loss": 1.2988,
+      "step": 8580
+    },
+    {
+      "epoch": 5.755443886097153,
+      "grad_norm": 1.1838830709457397,
+      "learning_rate": 0.0002,
+      "loss": 1.3535,
+      "step": 8590
+    },
+    {
+      "epoch": 5.76214405360134,
+      "grad_norm": 0.9631859064102173,
+      "learning_rate": 0.0002,
+      "loss": 1.2655,
+      "step": 8600
+    },
+    {
+      "epoch": 5.768844221105527,
+      "grad_norm": 0.9702655673027039,
+      "learning_rate": 0.0002,
+      "loss": 1.276,
+      "step": 8610
+    },
+    {
+      "epoch": 5.775544388609715,
+      "grad_norm": 1.0591435432434082,
+      "learning_rate": 0.0002,
+      "loss": 1.3196,
+      "step": 8620
+    },
+    {
+      "epoch": 5.782244556113903,
+      "grad_norm": 0.9989570379257202,
+      "learning_rate": 0.0002,
+      "loss": 1.267,
+      "step": 8630
+    },
+    {
+      "epoch": 5.788944723618091,
+      "grad_norm": 1.0836435556411743,
+      "learning_rate": 0.0002,
+      "loss": 1.3227,
+      "step": 8640
+    },
+    {
+      "epoch": 5.795644891122278,
+      "grad_norm": 0.8832896947860718,
+      "learning_rate": 0.0002,
+      "loss": 1.3334,
+      "step": 8650
+    },
+    {
+      "epoch": 5.802345058626465,
+      "grad_norm": 1.0104607343673706,
+      "learning_rate": 0.0002,
+      "loss": 1.3214,
+      "step": 8660
+    },
+    {
+      "epoch": 5.809045226130653,
+      "grad_norm": 0.8375084400177002,
+      "learning_rate": 0.0002,
+      "loss": 1.2703,
+      "step": 8670
+    },
+    {
+      "epoch": 5.815745393634841,
+      "grad_norm": 1.1300716400146484,
+      "learning_rate": 0.0002,
+      "loss": 1.3554,
+      "step": 8680
+    },
+    {
+      "epoch": 5.822445561139029,
+      "grad_norm": 0.9311910271644592,
+      "learning_rate": 0.0002,
+      "loss": 1.3468,
+      "step": 8690
+    },
+    {
+      "epoch": 5.8291457286432165,
+      "grad_norm": 0.9488391876220703,
+      "learning_rate": 0.0002,
+      "loss": 1.2749,
+      "step": 8700
+    },
+    {
+      "epoch": 5.835845896147403,
+      "grad_norm": 0.9747629761695862,
+      "learning_rate": 0.0002,
+      "loss": 1.2281,
+      "step": 8710
+    },
+    {
+      "epoch": 5.842546063651591,
+      "grad_norm": 1.1029598712921143,
+      "learning_rate": 0.0002,
+      "loss": 1.2923,
+      "step": 8720
+    },
+    {
+      "epoch": 5.849246231155779,
+      "grad_norm": 1.0396875143051147,
+      "learning_rate": 0.0002,
+      "loss": 1.3613,
+      "step": 8730
+    },
+    {
+      "epoch": 5.855946398659967,
+      "grad_norm": 0.9259780645370483,
+      "learning_rate": 0.0002,
+      "loss": 1.3272,
+      "step": 8740
+    },
+    {
+      "epoch": 5.8626465661641545,
+      "grad_norm": 1.020033597946167,
+      "learning_rate": 0.0002,
+      "loss": 1.3236,
+      "step": 8750
+    },
+    {
+      "epoch": 5.869346733668341,
+      "grad_norm": 0.9191218614578247,
+      "learning_rate": 0.0002,
+      "loss": 1.3453,
+      "step": 8760
+    },
+    {
+      "epoch": 5.876046901172529,
+      "grad_norm": 1.1093107461929321,
+      "learning_rate": 0.0002,
+      "loss": 1.3012,
+      "step": 8770
+    },
+    {
+      "epoch": 5.882747068676717,
+      "grad_norm": 1.1626793146133423,
+      "learning_rate": 0.0002,
+      "loss": 1.2718,
+      "step": 8780
+    },
+    {
+      "epoch": 5.889447236180905,
+      "grad_norm": 0.9542945027351379,
+      "learning_rate": 0.0002,
+      "loss": 1.2969,
+      "step": 8790
+    },
+    {
+      "epoch": 5.8961474036850925,
+      "grad_norm": 0.9086058139801025,
+      "learning_rate": 0.0002,
+      "loss": 1.3134,
+      "step": 8800
+    },
+    {
+      "epoch": 5.902847571189279,
+      "grad_norm": 0.9249639511108398,
+      "learning_rate": 0.0002,
+      "loss": 1.2731,
+      "step": 8810
+    },
+    {
+      "epoch": 5.909547738693467,
+      "grad_norm": 0.9414396286010742,
+      "learning_rate": 0.0002,
+      "loss": 1.337,
+      "step": 8820
+    },
+    {
+      "epoch": 5.916247906197655,
+      "grad_norm": 0.9086037874221802,
+      "learning_rate": 0.0002,
+      "loss": 1.2865,
+      "step": 8830
+    },
+    {
+      "epoch": 5.922948073701843,
+      "grad_norm": 0.8685907125473022,
+      "learning_rate": 0.0002,
+      "loss": 1.2756,
+      "step": 8840
+    },
+    {
+      "epoch": 5.9296482412060305,
+      "grad_norm": 1.036419153213501,
+      "learning_rate": 0.0002,
+      "loss": 1.297,
+      "step": 8850
+    },
+    {
+      "epoch": 5.936348408710217,
+      "grad_norm": 1.0183674097061157,
+      "learning_rate": 0.0002,
+      "loss": 1.3207,
+      "step": 8860
+    },
+    {
+      "epoch": 5.943048576214405,
+      "grad_norm": 0.966444194316864,
+      "learning_rate": 0.0002,
+      "loss": 1.3922,
+      "step": 8870
+    },
+    {
+      "epoch": 5.949748743718593,
+      "grad_norm": 1.125693917274475,
+      "learning_rate": 0.0002,
+      "loss": 1.333,
+      "step": 8880
+    },
+    {
+      "epoch": 5.956448911222781,
+      "grad_norm": 0.9857436418533325,
+      "learning_rate": 0.0002,
+      "loss": 1.3116,
+      "step": 8890
+    },
+    {
+      "epoch": 5.9631490787269685,
+      "grad_norm": 0.9377069473266602,
+      "learning_rate": 0.0002,
+      "loss": 1.2526,
+      "step": 8900
+    },
+    {
+      "epoch": 5.969849246231155,
+      "grad_norm": 0.9493814706802368,
+      "learning_rate": 0.0002,
+      "loss": 1.3221,
+      "step": 8910
+    },
+    {
+      "epoch": 5.976549413735343,
+      "grad_norm": 0.8806208372116089,
+      "learning_rate": 0.0002,
+      "loss": 1.2516,
+      "step": 8920
+    },
+    {
+      "epoch": 5.983249581239531,
+      "grad_norm": 0.8727600574493408,
+      "learning_rate": 0.0002,
+      "loss": 1.2558,
+      "step": 8930
+    },
+    {
+      "epoch": 5.989949748743719,
+      "grad_norm": 0.9799810647964478,
+      "learning_rate": 0.0002,
+      "loss": 1.3538,
+      "step": 8940
+    },
+    {
+      "epoch": 5.9966499162479066,
+      "grad_norm": 0.9866513609886169,
+      "learning_rate": 0.0002,
+      "loss": 1.3323,
+      "step": 8950
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 2.0282373428344727,
+      "eval_runtime": 38.0375,
+      "eval_samples_per_second": 13.539,
+      "eval_steps_per_second": 1.709,
+      "step": 8955
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 11936,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 8,
+  "save_steps": 200,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 4.1441734894288896e+17,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-8955/training_args.bin b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-8955/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..064f299b0f3d2a28f5b1f5c68ef32caab3e2dd49
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-8955/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aa7b586fc440d1e22156593e2dd4267d2bdcb8920a02fdf352ea29a9bec3dd94
+size 5560
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/special_tokens_map.json b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..b43be96621d147110fb8a18b5776ec6e38516127
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/special_tokens_map.json
@@ -0,0 +1,17 @@
+{
+  "bos_token": {
+    "content": "<|begin_of_text|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<|eot_id|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<|eot_id|>"
+}
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/tokenizer.json b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..86a33946b0c77216d2cce91bb28c8fada4a5e80b
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3c5cf44023714fb39b05e71e425f8d7b92805ff73f7988b083b8c87f0bf87393
+size 17209961
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/tokenizer_config.json b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..061e40d9db3253624f86e8e364c15ef546527c9d
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/tokenizer_config.json
@@ -0,0 +1,2063 @@
+{
+  "added_tokens_decoder": {
+    "128000": {
+      "content": "<|begin_of_text|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128001": {
+      "content": "<|end_of_text|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128002": {
+      "content": "<|reserved_special_token_0|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128003": {
+      "content": "<|reserved_special_token_1|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128004": {
+      "content": "<|reserved_special_token_2|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128005": {
+      "content": "<|reserved_special_token_3|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128006": {
+      "content": "<|start_header_id|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128007": {
+      "content": "<|end_header_id|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128008": {
+      "content": "<|reserved_special_token_4|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128009": {
+      "content": "<|eot_id|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128010": {
+      "content": "<|reserved_special_token_5|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128011": {
+      "content": "<|reserved_special_token_6|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128012": {
+      "content": "<|reserved_special_token_7|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128013": {
+      "content": "<|reserved_special_token_8|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128014": {
+      "content": "<|reserved_special_token_9|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128015": {
+      "content": "<|reserved_special_token_10|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128016": {
+      "content": "<|reserved_special_token_11|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128017": {
+      "content": "<|reserved_special_token_12|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128018": {
+      "content": "<|reserved_special_token_13|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128019": {
+      "content": "<|reserved_special_token_14|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128020": {
+      "content": "<|reserved_special_token_15|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128021": {
+      "content": "<|reserved_special_token_16|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128022": {
+      "content": "<|reserved_special_token_17|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128023": {
+      "content": "<|reserved_special_token_18|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128024": {
+      "content": "<|reserved_special_token_19|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128025": {
+      "content": "<|reserved_special_token_20|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128026": {
+      "content": "<|reserved_special_token_21|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128027": {
+      "content": "<|reserved_special_token_22|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128028": {
+      "content": "<|reserved_special_token_23|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128029": {
+      "content": "<|reserved_special_token_24|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128030": {
+      "content": "<|reserved_special_token_25|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128031": {
+      "content": "<|reserved_special_token_26|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128032": {
+      "content": "<|reserved_special_token_27|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128033": {
+      "content": "<|reserved_special_token_28|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128034": {
+      "content": "<|reserved_special_token_29|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128035": {
+      "content": "<|reserved_special_token_30|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128036": {
+      "content": "<|reserved_special_token_31|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128037": {
+      "content": "<|reserved_special_token_32|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128038": {
+      "content": "<|reserved_special_token_33|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128039": {
+      "content": "<|reserved_special_token_34|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128040": {
+      "content": "<|reserved_special_token_35|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128041": {
+      "content": "<|reserved_special_token_36|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128042": {
+      "content": "<|reserved_special_token_37|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128043": {
+      "content": "<|reserved_special_token_38|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128044": {
+      "content": "<|reserved_special_token_39|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128045": {
+      "content": "<|reserved_special_token_40|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128046": {
+      "content": "<|reserved_special_token_41|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128047": {
+      "content": "<|reserved_special_token_42|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128048": {
+      "content": "<|reserved_special_token_43|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128049": {
+      "content": "<|reserved_special_token_44|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128050": {
+      "content": "<|reserved_special_token_45|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128051": {
+      "content": "<|reserved_special_token_46|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128052": {
+      "content": "<|reserved_special_token_47|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128053": {
+      "content": "<|reserved_special_token_48|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128054": {
+      "content": "<|reserved_special_token_49|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128055": {
+      "content": "<|reserved_special_token_50|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128056": {
+      "content": "<|reserved_special_token_51|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128057": {
+      "content": "<|reserved_special_token_52|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128058": {
+      "content": "<|reserved_special_token_53|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128059": {
+      "content": "<|reserved_special_token_54|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128060": {
+      "content": "<|reserved_special_token_55|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128061": {
+      "content": "<|reserved_special_token_56|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128062": {
+      "content": "<|reserved_special_token_57|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128063": {
+      "content": "<|reserved_special_token_58|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128064": {
+      "content": "<|reserved_special_token_59|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128065": {
+      "content": "<|reserved_special_token_60|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128066": {
+      "content": "<|reserved_special_token_61|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128067": {
+      "content": "<|reserved_special_token_62|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128068": {
+      "content": "<|reserved_special_token_63|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128069": {
+      "content": "<|reserved_special_token_64|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128070": {
+      "content": "<|reserved_special_token_65|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128071": {
+      "content": "<|reserved_special_token_66|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128072": {
+      "content": "<|reserved_special_token_67|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128073": {
+      "content": "<|reserved_special_token_68|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128074": {
+      "content": "<|reserved_special_token_69|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128075": {
+      "content": "<|reserved_special_token_70|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128076": {
+      "content": "<|reserved_special_token_71|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128077": {
+      "content": "<|reserved_special_token_72|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128078": {
+      "content": "<|reserved_special_token_73|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128079": {
+      "content": "<|reserved_special_token_74|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128080": {
+      "content": "<|reserved_special_token_75|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128081": {
+      "content": "<|reserved_special_token_76|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128082": {
+      "content": "<|reserved_special_token_77|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128083": {
+      "content": "<|reserved_special_token_78|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128084": {
+      "content": "<|reserved_special_token_79|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128085": {
+      "content": "<|reserved_special_token_80|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128086": {
+      "content": "<|reserved_special_token_81|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128087": {
+      "content": "<|reserved_special_token_82|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128088": {
+      "content": "<|reserved_special_token_83|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128089": {
+      "content": "<|reserved_special_token_84|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128090": {
+      "content": "<|reserved_special_token_85|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128091": {
+      "content": "<|reserved_special_token_86|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128092": {
+      "content": "<|reserved_special_token_87|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128093": {
+      "content": "<|reserved_special_token_88|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128094": {
+      "content": "<|reserved_special_token_89|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128095": {
+      "content": "<|reserved_special_token_90|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128096": {
+      "content": "<|reserved_special_token_91|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128097": {
+      "content": "<|reserved_special_token_92|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128098": {
+      "content": "<|reserved_special_token_93|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128099": {
+      "content": "<|reserved_special_token_94|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128100": {
+      "content": "<|reserved_special_token_95|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128101": {
+      "content": "<|reserved_special_token_96|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128102": {
+      "content": "<|reserved_special_token_97|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128103": {
+      "content": "<|reserved_special_token_98|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128104": {
+      "content": "<|reserved_special_token_99|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128105": {
+      "content": "<|reserved_special_token_100|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128106": {
+      "content": "<|reserved_special_token_101|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128107": {
+      "content": "<|reserved_special_token_102|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128108": {
+      "content": "<|reserved_special_token_103|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128109": {
+      "content": "<|reserved_special_token_104|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128110": {
+      "content": "<|reserved_special_token_105|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128111": {
+      "content": "<|reserved_special_token_106|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128112": {
+      "content": "<|reserved_special_token_107|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128113": {
+      "content": "<|reserved_special_token_108|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128114": {
+      "content": "<|reserved_special_token_109|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128115": {
+      "content": "<|reserved_special_token_110|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128116": {
+      "content": "<|reserved_special_token_111|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128117": {
+      "content": "<|reserved_special_token_112|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128118": {
+      "content": "<|reserved_special_token_113|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128119": {
+      "content": "<|reserved_special_token_114|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128120": {
+      "content": "<|reserved_special_token_115|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128121": {
+      "content": "<|reserved_special_token_116|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128122": {
+      "content": "<|reserved_special_token_117|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128123": {
+      "content": "<|reserved_special_token_118|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128124": {
+      "content": "<|reserved_special_token_119|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128125": {
+      "content": "<|reserved_special_token_120|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128126": {
+      "content": "<|reserved_special_token_121|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128127": {
+      "content": "<|reserved_special_token_122|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128128": {
+      "content": "<|reserved_special_token_123|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128129": {
+      "content": "<|reserved_special_token_124|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128130": {
+      "content": "<|reserved_special_token_125|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128131": {
+      "content": "<|reserved_special_token_126|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128132": {
+      "content": "<|reserved_special_token_127|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128133": {
+      "content": "<|reserved_special_token_128|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128134": {
+      "content": "<|reserved_special_token_129|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128135": {
+      "content": "<|reserved_special_token_130|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128136": {
+      "content": "<|reserved_special_token_131|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128137": {
+      "content": "<|reserved_special_token_132|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128138": {
+      "content": "<|reserved_special_token_133|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128139": {
+      "content": "<|reserved_special_token_134|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128140": {
+      "content": "<|reserved_special_token_135|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128141": {
+      "content": "<|reserved_special_token_136|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128142": {
+      "content": "<|reserved_special_token_137|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128143": {
+      "content": "<|reserved_special_token_138|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128144": {
+      "content": "<|reserved_special_token_139|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128145": {
+      "content": "<|reserved_special_token_140|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128146": {
+      "content": "<|reserved_special_token_141|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128147": {
+      "content": "<|reserved_special_token_142|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128148": {
+      "content": "<|reserved_special_token_143|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128149": {
+      "content": "<|reserved_special_token_144|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128150": {
+      "content": "<|reserved_special_token_145|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128151": {
+      "content": "<|reserved_special_token_146|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128152": {
+      "content": "<|reserved_special_token_147|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128153": {
+      "content": "<|reserved_special_token_148|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128154": {
+      "content": "<|reserved_special_token_149|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128155": {
+      "content": "<|reserved_special_token_150|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128156": {
+      "content": "<|reserved_special_token_151|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128157": {
+      "content": "<|reserved_special_token_152|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128158": {
+      "content": "<|reserved_special_token_153|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128159": {
+      "content": "<|reserved_special_token_154|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128160": {
+      "content": "<|reserved_special_token_155|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128161": {
+      "content": "<|reserved_special_token_156|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128162": {
+      "content": "<|reserved_special_token_157|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128163": {
+      "content": "<|reserved_special_token_158|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128164": {
+      "content": "<|reserved_special_token_159|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128165": {
+      "content": "<|reserved_special_token_160|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128166": {
+      "content": "<|reserved_special_token_161|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128167": {
+      "content": "<|reserved_special_token_162|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128168": {
+      "content": "<|reserved_special_token_163|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128169": {
+      "content": "<|reserved_special_token_164|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128170": {
+      "content": "<|reserved_special_token_165|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128171": {
+      "content": "<|reserved_special_token_166|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128172": {
+      "content": "<|reserved_special_token_167|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128173": {
+      "content": "<|reserved_special_token_168|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128174": {
+      "content": "<|reserved_special_token_169|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128175": {
+      "content": "<|reserved_special_token_170|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128176": {
+      "content": "<|reserved_special_token_171|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128177": {
+      "content": "<|reserved_special_token_172|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128178": {
+      "content": "<|reserved_special_token_173|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128179": {
+      "content": "<|reserved_special_token_174|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128180": {
+      "content": "<|reserved_special_token_175|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128181": {
+      "content": "<|reserved_special_token_176|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128182": {
+      "content": "<|reserved_special_token_177|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128183": {
+      "content": "<|reserved_special_token_178|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128184": {
+      "content": "<|reserved_special_token_179|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128185": {
+      "content": "<|reserved_special_token_180|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128186": {
+      "content": "<|reserved_special_token_181|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128187": {
+      "content": "<|reserved_special_token_182|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128188": {
+      "content": "<|reserved_special_token_183|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128189": {
+      "content": "<|reserved_special_token_184|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128190": {
+      "content": "<|reserved_special_token_185|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128191": {
+      "content": "<|reserved_special_token_186|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128192": {
+      "content": "<|reserved_special_token_187|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128193": {
+      "content": "<|reserved_special_token_188|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128194": {
+      "content": "<|reserved_special_token_189|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128195": {
+      "content": "<|reserved_special_token_190|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128196": {
+      "content": "<|reserved_special_token_191|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128197": {
+      "content": "<|reserved_special_token_192|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128198": {
+      "content": "<|reserved_special_token_193|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128199": {
+      "content": "<|reserved_special_token_194|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128200": {
+      "content": "<|reserved_special_token_195|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128201": {
+      "content": "<|reserved_special_token_196|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128202": {
+      "content": "<|reserved_special_token_197|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128203": {
+      "content": "<|reserved_special_token_198|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128204": {
+      "content": "<|reserved_special_token_199|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128205": {
+      "content": "<|reserved_special_token_200|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128206": {
+      "content": "<|reserved_special_token_201|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128207": {
+      "content": "<|reserved_special_token_202|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128208": {
+      "content": "<|reserved_special_token_203|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128209": {
+      "content": "<|reserved_special_token_204|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128210": {
+      "content": "<|reserved_special_token_205|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128211": {
+      "content": "<|reserved_special_token_206|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128212": {
+      "content": "<|reserved_special_token_207|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128213": {
+      "content": "<|reserved_special_token_208|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128214": {
+      "content": "<|reserved_special_token_209|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128215": {
+      "content": "<|reserved_special_token_210|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128216": {
+      "content": "<|reserved_special_token_211|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128217": {
+      "content": "<|reserved_special_token_212|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128218": {
+      "content": "<|reserved_special_token_213|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128219": {
+      "content": "<|reserved_special_token_214|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128220": {
+      "content": "<|reserved_special_token_215|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128221": {
+      "content": "<|reserved_special_token_216|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128222": {
+      "content": "<|reserved_special_token_217|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128223": {
+      "content": "<|reserved_special_token_218|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128224": {
+      "content": "<|reserved_special_token_219|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128225": {
+      "content": "<|reserved_special_token_220|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128226": {
+      "content": "<|reserved_special_token_221|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128227": {
+      "content": "<|reserved_special_token_222|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128228": {
+      "content": "<|reserved_special_token_223|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128229": {
+      "content": "<|reserved_special_token_224|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128230": {
+      "content": "<|reserved_special_token_225|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128231": {
+      "content": "<|reserved_special_token_226|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128232": {
+      "content": "<|reserved_special_token_227|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128233": {
+      "content": "<|reserved_special_token_228|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128234": {
+      "content": "<|reserved_special_token_229|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128235": {
+      "content": "<|reserved_special_token_230|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128236": {
+      "content": "<|reserved_special_token_231|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128237": {
+      "content": "<|reserved_special_token_232|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128238": {
+      "content": "<|reserved_special_token_233|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128239": {
+      "content": "<|reserved_special_token_234|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128240": {
+      "content": "<|reserved_special_token_235|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128241": {
+      "content": "<|reserved_special_token_236|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128242": {
+      "content": "<|reserved_special_token_237|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128243": {
+      "content": "<|reserved_special_token_238|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128244": {
+      "content": "<|reserved_special_token_239|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128245": {
+      "content": "<|reserved_special_token_240|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128246": {
+      "content": "<|reserved_special_token_241|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128247": {
+      "content": "<|reserved_special_token_242|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128248": {
+      "content": "<|reserved_special_token_243|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128249": {
+      "content": "<|reserved_special_token_244|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128250": {
+      "content": "<|reserved_special_token_245|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128251": {
+      "content": "<|reserved_special_token_246|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128252": {
+      "content": "<|reserved_special_token_247|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128253": {
+      "content": "<|reserved_special_token_248|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128254": {
+      "content": "<|reserved_special_token_249|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128255": {
+      "content": "<|reserved_special_token_250|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<|begin_of_text|>",
+  "chat_template": "{% set loop_messages = messages %}{% for message in loop_messages %}{% set content = '<|start_header_id|>' + message['role'] + '<|end_header_id|>\n\n'+ message['content'] | trim + '<|eot_id|>' %}{% if loop.index0 == 0 %}{% set content = bos_token + content %}{% endif %}{{ content }}{% endfor %}{% if add_generation_prompt %}{{ '<|start_header_id|>assistant<|end_header_id|>\n\n' }}{% endif %}",
+  "clean_up_tokenization_spaces": true,
+  "eos_token": "<|eot_id|>",
+  "model_input_names": [
+    "input_ids",
+    "attention_mask"
+  ],
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<|eot_id|>",
+  "tokenizer_class": "PreTrainedTokenizerFast"
+}
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/training_args.bin b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..064f299b0f3d2a28f5b1f5c68ef32caab3e2dd49
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aa7b586fc440d1e22156593e2dd4267d2bdcb8920a02fdf352ea29a9bec3dd94
+size 5560
diff --git a/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/training_log.jsonl b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/training_log.jsonl
new file mode 100644
index 0000000000000000000000000000000000000000..3d158f7569c1d1db0d5a47a0d18863cef728106f
--- /dev/null
+++ b/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/training_log.jsonl
@@ -0,0 +1,8 @@
+{"epoch": 0.9996649916247906, "step": 1492, "epoch_duration": 1574.8106648921967, "total_accumulated_duration": 1574.8106648921967, "gpu_info": {"GPU_0": "NVIDIA A100-SXM4-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 7672.4541015625}, "peak_memory_usage": {"GPU_0": 9688.99365234375}, "avg_memory_reserved": {"GPU_0": 10406.0}, "peak_memory_reserved": {"GPU_0": 10406.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "N/A", "params": {"epochs": 8, "batch_size": 1, "learning_rate": 0.0002, "gradient_accumulation_steps": 2, "warmup_ratio": 0.03, "max_grad_norm": 0.3, "lora_alpha": 32, "lora_dropout": 0.05, "lora_r": 64, "tf32": true, "seed": 42}, "log_history": [{"loss": 2.6189, "grad_norm": 0.565915048122406, "learning_rate": 0.0002, "epoch": 0.006700167504187605, "step": 10}, {"loss": 2.3162, "grad_norm": 0.5004463791847229, "learning_rate": 0.0002, "epoch": 0.01340033500837521, "step": 20}, {"loss": 2.0576, "grad_norm": 0.511043906211853, "learning_rate": 0.0002, "epoch": 0.020100502512562814, "step": 30}, {"loss": 2.0085, "grad_norm": 0.47327178716659546, "learning_rate": 0.0002, "epoch": 0.02680067001675042, "step": 40}, {"loss": 2.0276, "grad_norm": 0.5511676669120789, "learning_rate": 0.0002, "epoch": 0.03350083752093802, "step": 50}, {"loss": 1.9075, "grad_norm": 0.4666278064250946, "learning_rate": 0.0002, "epoch": 0.04020100502512563, "step": 60}, {"loss": 1.8413, "grad_norm": 0.5310961008071899, "learning_rate": 0.0002, "epoch": 0.04690117252931323, "step": 70}, {"loss": 1.8711, "grad_norm": 0.5606027245521545, "learning_rate": 0.0002, "epoch": 0.05360134003350084, "step": 80}, {"loss": 1.9282, "grad_norm": 0.4934779703617096, "learning_rate": 0.0002, "epoch": 0.06030150753768844, "step": 90}, {"loss": 1.8925, "grad_norm": 0.4821869730949402, "learning_rate": 0.0002, "epoch": 0.06700167504187604, "step": 100}, {"loss": 1.8628, "grad_norm": 0.5262084603309631, "learning_rate": 0.0002, "epoch": 0.07370184254606366, "step": 110}, {"loss": 1.8347, "grad_norm": 0.3774230182170868, "learning_rate": 0.0002, "epoch": 0.08040201005025126, "step": 120}, {"loss": 1.8386, "grad_norm": 0.34137430787086487, "learning_rate": 0.0002, "epoch": 0.08710217755443886, "step": 130}, {"loss": 1.861, "grad_norm": 0.407272070646286, "learning_rate": 0.0002, "epoch": 0.09380234505862646, "step": 140}, {"loss": 1.8279, "grad_norm": 0.4011937975883484, "learning_rate": 0.0002, "epoch": 0.10050251256281408, "step": 150}, {"loss": 1.9317, "grad_norm": 0.4432467222213745, "learning_rate": 0.0002, "epoch": 0.10720268006700168, "step": 160}, {"loss": 1.8157, "grad_norm": 0.44030463695526123, "learning_rate": 0.0002, "epoch": 0.11390284757118928, "step": 170}, {"loss": 1.8534, "grad_norm": 0.3799569308757782, "learning_rate": 0.0002, "epoch": 0.12060301507537688, "step": 180}, {"loss": 1.7658, "grad_norm": 0.33721521496772766, "learning_rate": 0.0002, "epoch": 0.1273031825795645, "step": 190}, {"loss": 1.8269, "grad_norm": 0.4096226692199707, "learning_rate": 0.0002, "epoch": 0.13400335008375208, "step": 200}, {"loss": 1.802, "grad_norm": 0.37374693155288696, "learning_rate": 0.0002, "epoch": 0.1407035175879397, "step": 210}, {"loss": 1.8901, "grad_norm": 0.3249480128288269, "learning_rate": 0.0002, "epoch": 0.1474036850921273, "step": 220}, {"loss": 1.8163, "grad_norm": 0.3612042963504791, "learning_rate": 0.0002, "epoch": 0.1541038525963149, "step": 230}, {"loss": 1.7585, "grad_norm": 0.3686671257019043, "learning_rate": 0.0002, "epoch": 0.16080402010050251, "step": 240}, {"loss": 1.8365, "grad_norm": 0.3521044850349426, "learning_rate": 0.0002, "epoch": 0.16750418760469013, "step": 250}, {"loss": 1.8623, "grad_norm": 0.4073677361011505, "learning_rate": 0.0002, "epoch": 0.17420435510887772, "step": 260}, {"loss": 1.8026, "grad_norm": 0.34522193670272827, "learning_rate": 0.0002, "epoch": 0.18090452261306533, "step": 270}, {"loss": 1.8162, "grad_norm": 0.4121900498867035, "learning_rate": 0.0002, "epoch": 0.18760469011725292, "step": 280}, {"loss": 1.7976, "grad_norm": 0.3544778525829315, "learning_rate": 0.0002, "epoch": 0.19430485762144054, "step": 290}, {"loss": 1.8787, "grad_norm": 0.3482133448123932, "learning_rate": 0.0002, "epoch": 0.20100502512562815, "step": 300}, {"loss": 1.8578, "grad_norm": 0.3421826660633087, "learning_rate": 0.0002, "epoch": 0.20770519262981574, "step": 310}, {"loss": 1.8013, "grad_norm": 0.5024696588516235, "learning_rate": 0.0002, "epoch": 0.21440536013400335, "step": 320}, {"loss": 1.8607, "grad_norm": 0.36013063788414, "learning_rate": 0.0002, "epoch": 0.22110552763819097, "step": 330}, {"loss": 1.9075, "grad_norm": 0.3611244857311249, "learning_rate": 0.0002, "epoch": 0.22780569514237856, "step": 340}, {"loss": 1.8128, "grad_norm": 0.39244529604911804, "learning_rate": 0.0002, "epoch": 0.23450586264656617, "step": 350}, {"loss": 1.7885, "grad_norm": 0.3299325704574585, "learning_rate": 0.0002, "epoch": 0.24120603015075376, "step": 360}, {"loss": 1.8028, "grad_norm": 0.3994322419166565, "learning_rate": 0.0002, "epoch": 0.24790619765494137, "step": 370}, {"loss": 1.8321, "grad_norm": 0.3559151887893677, "learning_rate": 0.0002, "epoch": 0.254606365159129, "step": 380}, {"loss": 1.7802, "grad_norm": 0.3873756229877472, "learning_rate": 0.0002, "epoch": 0.2613065326633166, "step": 390}, {"loss": 1.7844, "grad_norm": 0.3710744082927704, "learning_rate": 0.0002, "epoch": 0.26800670016750416, "step": 400}, {"loss": 1.7789, "grad_norm": 0.3618465065956116, "learning_rate": 0.0002, "epoch": 0.2747068676716918, "step": 410}, {"loss": 1.8529, "grad_norm": 0.30063769221305847, "learning_rate": 0.0002, "epoch": 0.2814070351758794, "step": 420}, {"loss": 1.7765, "grad_norm": 0.3695628345012665, "learning_rate": 0.0002, "epoch": 0.288107202680067, "step": 430}, {"loss": 1.7982, "grad_norm": 0.31451135873794556, "learning_rate": 0.0002, "epoch": 0.2948073701842546, "step": 440}, {"loss": 1.7517, "grad_norm": 0.3959707021713257, "learning_rate": 0.0002, "epoch": 0.3015075376884422, "step": 450}, {"loss": 1.8142, "grad_norm": 0.33475354313850403, "learning_rate": 0.0002, "epoch": 0.3082077051926298, "step": 460}, {"loss": 1.8805, "grad_norm": 0.33933115005493164, "learning_rate": 0.0002, "epoch": 0.3149078726968174, "step": 470}, {"loss": 1.7564, "grad_norm": 0.3264943063259125, "learning_rate": 0.0002, "epoch": 0.32160804020100503, "step": 480}, {"loss": 1.8428, "grad_norm": 0.40188100934028625, "learning_rate": 0.0002, "epoch": 0.32830820770519265, "step": 490}, {"loss": 1.7624, "grad_norm": 0.37408649921417236, "learning_rate": 0.0002, "epoch": 0.33500837520938026, "step": 500}, {"loss": 1.7745, "grad_norm": 0.33925938606262207, "learning_rate": 0.0002, "epoch": 0.3417085427135678, "step": 510}, {"loss": 1.814, "grad_norm": 0.36836713552474976, "learning_rate": 0.0002, "epoch": 0.34840871021775544, "step": 520}, {"loss": 1.8037, "grad_norm": 0.37284499406814575, "learning_rate": 0.0002, "epoch": 0.35510887772194305, "step": 530}, {"loss": 1.8379, "grad_norm": 0.3192278742790222, "learning_rate": 0.0002, "epoch": 0.36180904522613067, "step": 540}, {"loss": 1.8702, "grad_norm": 0.30233290791511536, "learning_rate": 0.0002, "epoch": 0.3685092127303183, "step": 550}, {"loss": 1.8799, "grad_norm": 0.3340817391872406, "learning_rate": 0.0002, "epoch": 0.37520938023450584, "step": 560}, {"loss": 1.8404, "grad_norm": 0.32600095868110657, "learning_rate": 0.0002, "epoch": 0.38190954773869346, "step": 570}, {"loss": 1.7804, "grad_norm": 0.33711278438568115, "learning_rate": 0.0002, "epoch": 0.38860971524288107, "step": 580}, {"loss": 1.8445, "grad_norm": 0.34890690445899963, "learning_rate": 0.0002, "epoch": 0.3953098827470687, "step": 590}, {"loss": 1.8187, "grad_norm": 0.38238924741744995, "learning_rate": 0.0002, "epoch": 0.4020100502512563, "step": 600}, {"loss": 1.8111, "grad_norm": 0.34399354457855225, "learning_rate": 0.0002, "epoch": 0.40871021775544386, "step": 610}, {"loss": 1.8006, "grad_norm": 0.3346073627471924, "learning_rate": 0.0002, "epoch": 0.4154103852596315, "step": 620}, {"loss": 1.7705, "grad_norm": 0.3545648157596588, "learning_rate": 0.0002, "epoch": 0.4221105527638191, "step": 630}, {"loss": 1.8445, "grad_norm": 0.3378899097442627, "learning_rate": 0.0002, "epoch": 0.4288107202680067, "step": 640}, {"loss": 1.804, "grad_norm": 0.3255569040775299, "learning_rate": 0.0002, "epoch": 0.4355108877721943, "step": 650}, {"loss": 1.7679, "grad_norm": 0.34880587458610535, "learning_rate": 0.0002, "epoch": 0.44221105527638194, "step": 660}, {"loss": 1.7861, "grad_norm": 0.3402383625507355, "learning_rate": 0.0002, "epoch": 0.4489112227805695, "step": 670}, {"loss": 1.8131, "grad_norm": 0.3594033718109131, "learning_rate": 0.0002, "epoch": 0.4556113902847571, "step": 680}, {"loss": 1.8399, "grad_norm": 0.31000566482543945, "learning_rate": 0.0002, "epoch": 0.4623115577889447, "step": 690}, {"loss": 1.7521, "grad_norm": 0.37229061126708984, "learning_rate": 0.0002, "epoch": 0.46901172529313234, "step": 700}, {"loss": 1.7779, "grad_norm": 0.315801739692688, "learning_rate": 0.0002, "epoch": 0.47571189279731996, "step": 710}, {"loss": 1.7515, "grad_norm": 0.3220832645893097, "learning_rate": 0.0002, "epoch": 0.4824120603015075, "step": 720}, {"loss": 1.7181, "grad_norm": 0.3435456156730652, "learning_rate": 0.0002, "epoch": 0.48911222780569513, "step": 730}, {"loss": 1.8844, "grad_norm": 0.30380892753601074, "learning_rate": 0.0002, "epoch": 0.49581239530988275, "step": 740}, {"loss": 1.7792, "grad_norm": 0.3555026054382324, "learning_rate": 0.0002, "epoch": 0.5025125628140703, "step": 750}, {"loss": 1.7714, "grad_norm": 0.3019855320453644, "learning_rate": 0.0002, "epoch": 0.509212730318258, "step": 760}, {"loss": 1.7962, "grad_norm": 0.309111088514328, "learning_rate": 0.0002, "epoch": 0.5159128978224455, "step": 770}, {"loss": 1.7913, "grad_norm": 0.366020530462265, "learning_rate": 0.0002, "epoch": 0.5226130653266332, "step": 780}, {"loss": 1.8008, "grad_norm": 0.3267050087451935, "learning_rate": 0.0002, "epoch": 0.5293132328308208, "step": 790}, {"loss": 1.7397, "grad_norm": 0.34265750646591187, "learning_rate": 0.0002, "epoch": 0.5360134003350083, "step": 800}, {"loss": 1.8251, "grad_norm": 0.313669890165329, "learning_rate": 0.0002, "epoch": 0.542713567839196, "step": 810}, {"loss": 1.8369, "grad_norm": 0.3355236053466797, "learning_rate": 0.0002, "epoch": 0.5494137353433836, "step": 820}, {"loss": 1.7381, "grad_norm": 0.3186608552932739, "learning_rate": 0.0002, "epoch": 0.5561139028475712, "step": 830}, {"loss": 1.8034, "grad_norm": 0.30357518792152405, "learning_rate": 0.0002, "epoch": 0.5628140703517588, "step": 840}, {"loss": 1.769, "grad_norm": 0.3990040123462677, "learning_rate": 0.0002, "epoch": 0.5695142378559463, "step": 850}, {"loss": 1.7482, "grad_norm": 0.34363803267478943, "learning_rate": 0.0002, "epoch": 0.576214405360134, "step": 860}, {"loss": 1.8106, "grad_norm": 0.3757908046245575, "learning_rate": 0.0002, "epoch": 0.5829145728643216, "step": 870}, {"loss": 1.8104, "grad_norm": 0.3359757661819458, "learning_rate": 0.0002, "epoch": 0.5896147403685092, "step": 880}, {"loss": 1.7591, "grad_norm": 0.5555329918861389, "learning_rate": 0.0002, "epoch": 0.5963149078726968, "step": 890}, {"loss": 1.7715, "grad_norm": 0.4046323895454407, "learning_rate": 0.0002, "epoch": 0.6030150753768844, "step": 900}, {"loss": 1.7998, "grad_norm": 0.29834219813346863, "learning_rate": 0.0002, "epoch": 0.609715242881072, "step": 910}, {"loss": 1.7826, "grad_norm": 0.3241238594055176, "learning_rate": 0.0002, "epoch": 0.6164154103852596, "step": 920}, {"loss": 1.8342, "grad_norm": 0.35154739022254944, "learning_rate": 0.0002, "epoch": 0.6231155778894473, "step": 930}, {"loss": 1.8076, "grad_norm": 0.3287706673145294, "learning_rate": 0.0002, "epoch": 0.6298157453936348, "step": 940}, {"loss": 1.8038, "grad_norm": 0.35670626163482666, "learning_rate": 0.0002, "epoch": 0.6365159128978225, "step": 950}, {"loss": 1.869, "grad_norm": 0.6114104986190796, "learning_rate": 0.0002, "epoch": 0.6432160804020101, "step": 960}, {"loss": 1.8297, "grad_norm": 0.3186565041542053, "learning_rate": 0.0002, "epoch": 0.6499162479061976, "step": 970}, {"loss": 1.7539, "grad_norm": 0.27164125442504883, "learning_rate": 0.0002, "epoch": 0.6566164154103853, "step": 980}, {"loss": 1.8339, "grad_norm": 0.34407344460487366, "learning_rate": 0.0002, "epoch": 0.6633165829145728, "step": 990}, {"loss": 1.855, "grad_norm": 0.368415892124176, "learning_rate": 0.0002, "epoch": 0.6700167504187605, "step": 1000}, {"loss": 1.7821, "grad_norm": 0.3306390643119812, "learning_rate": 0.0002, "epoch": 0.6767169179229481, "step": 1010}, {"loss": 1.786, "grad_norm": 0.3198648989200592, "learning_rate": 0.0002, "epoch": 0.6834170854271356, "step": 1020}, {"loss": 1.816, "grad_norm": 0.3092987537384033, "learning_rate": 0.0002, "epoch": 0.6901172529313233, "step": 1030}, {"loss": 1.7689, "grad_norm": 0.3090653419494629, "learning_rate": 0.0002, "epoch": 0.6968174204355109, "step": 1040}, {"loss": 1.7544, "grad_norm": 0.3485880196094513, "learning_rate": 0.0002, "epoch": 0.7035175879396985, "step": 1050}, {"loss": 1.8187, "grad_norm": 0.35782721638679504, "learning_rate": 0.0002, "epoch": 0.7102177554438861, "step": 1060}, {"loss": 1.806, "grad_norm": 0.34256869554519653, "learning_rate": 0.0002, "epoch": 0.7169179229480737, "step": 1070}, {"loss": 1.7873, "grad_norm": 0.30461037158966064, "learning_rate": 0.0002, "epoch": 0.7236180904522613, "step": 1080}, {"loss": 1.7367, "grad_norm": 0.3398691713809967, "learning_rate": 0.0002, "epoch": 0.7303182579564489, "step": 1090}, {"loss": 1.8756, "grad_norm": 0.3180808126926422, "learning_rate": 0.0002, "epoch": 0.7370184254606366, "step": 1100}, {"loss": 1.6988, "grad_norm": 0.34400665760040283, "learning_rate": 0.0002, "epoch": 0.7437185929648241, "step": 1110}, {"loss": 1.7851, "grad_norm": 0.34244877099990845, "learning_rate": 0.0002, "epoch": 0.7504187604690117, "step": 1120}, {"loss": 1.7841, "grad_norm": 0.29946693778038025, "learning_rate": 0.0002, "epoch": 0.7571189279731994, "step": 1130}, {"loss": 1.7456, "grad_norm": 0.37547236680984497, "learning_rate": 0.0002, "epoch": 0.7638190954773869, "step": 1140}, {"loss": 1.8425, "grad_norm": 0.3263005018234253, "learning_rate": 0.0002, "epoch": 0.7705192629815746, "step": 1150}, {"loss": 1.7222, "grad_norm": 0.41363608837127686, "learning_rate": 0.0002, "epoch": 0.7772194304857621, "step": 1160}, {"loss": 1.7836, "grad_norm": 0.36267954111099243, "learning_rate": 0.0002, "epoch": 0.7839195979899497, "step": 1170}, {"loss": 1.9183, "grad_norm": 0.31789499521255493, "learning_rate": 0.0002, "epoch": 0.7906197654941374, "step": 1180}, {"loss": 1.78, "grad_norm": 0.5708149075508118, "learning_rate": 0.0002, "epoch": 0.7973199329983249, "step": 1190}, {"loss": 1.6908, "grad_norm": 0.322099506855011, "learning_rate": 0.0002, "epoch": 0.8040201005025126, "step": 1200}, {"loss": 1.7639, "grad_norm": 0.3419909179210663, "learning_rate": 0.0002, "epoch": 0.8107202680067002, "step": 1210}, {"loss": 1.7428, "grad_norm": 0.36286255717277527, "learning_rate": 0.0002, "epoch": 0.8174204355108877, "step": 1220}, {"loss": 1.8409, "grad_norm": 0.33992862701416016, "learning_rate": 0.0002, "epoch": 0.8241206030150754, "step": 1230}, {"loss": 1.7507, "grad_norm": 0.32622793316841125, "learning_rate": 0.0002, "epoch": 0.830820770519263, "step": 1240}, {"loss": 1.8098, "grad_norm": 0.3036167621612549, "learning_rate": 0.0002, "epoch": 0.8375209380234506, "step": 1250}, {"loss": 1.8094, "grad_norm": 0.3182215392589569, "learning_rate": 0.0002, "epoch": 0.8442211055276382, "step": 1260}, {"loss": 1.8017, "grad_norm": 0.3270018696784973, "learning_rate": 0.0002, "epoch": 0.8509212730318257, "step": 1270}, {"loss": 1.798, "grad_norm": 0.32652342319488525, "learning_rate": 0.0002, "epoch": 0.8576214405360134, "step": 1280}, {"loss": 1.7448, "grad_norm": 0.3631329834461212, "learning_rate": 0.0002, "epoch": 0.864321608040201, "step": 1290}, {"loss": 1.7, "grad_norm": 0.36706018447875977, "learning_rate": 0.0002, "epoch": 0.8710217755443886, "step": 1300}, {"loss": 1.8178, "grad_norm": 0.3347418010234833, "learning_rate": 0.0002, "epoch": 0.8777219430485762, "step": 1310}, {"loss": 1.7824, "grad_norm": 0.34371060132980347, "learning_rate": 0.0002, "epoch": 0.8844221105527639, "step": 1320}, {"loss": 1.783, "grad_norm": 0.3029090166091919, "learning_rate": 0.0002, "epoch": 0.8911222780569514, "step": 1330}, {"loss": 1.8017, "grad_norm": 0.34700682759284973, "learning_rate": 0.0002, "epoch": 0.897822445561139, "step": 1340}, {"loss": 1.7998, "grad_norm": 0.35574328899383545, "learning_rate": 0.0002, "epoch": 0.9045226130653267, "step": 1350}, {"loss": 1.7339, "grad_norm": 0.30928221344947815, "learning_rate": 0.0002, "epoch": 0.9112227805695142, "step": 1360}, {"loss": 1.7479, "grad_norm": 0.30652928352355957, "learning_rate": 0.0002, "epoch": 0.9179229480737019, "step": 1370}, {"loss": 1.7491, "grad_norm": 0.3838157653808594, "learning_rate": 0.0002, "epoch": 0.9246231155778895, "step": 1380}, {"loss": 1.7977, "grad_norm": 0.31655240058898926, "learning_rate": 0.0002, "epoch": 0.931323283082077, "step": 1390}, {"loss": 1.8175, "grad_norm": 0.41737303137779236, "learning_rate": 0.0002, "epoch": 0.9380234505862647, "step": 1400}, {"loss": 1.6811, "grad_norm": 0.3227267861366272, "learning_rate": 0.0002, "epoch": 0.9447236180904522, "step": 1410}, {"loss": 1.7343, "grad_norm": 0.3729925751686096, "learning_rate": 0.0002, "epoch": 0.9514237855946399, "step": 1420}, {"loss": 1.8221, "grad_norm": 0.30779409408569336, "learning_rate": 0.0002, "epoch": 0.9581239530988275, "step": 1430}, {"loss": 1.7972, "grad_norm": 0.334379643201828, "learning_rate": 0.0002, "epoch": 0.964824120603015, "step": 1440}, {"loss": 1.7141, "grad_norm": 0.3568236231803894, "learning_rate": 0.0002, "epoch": 0.9715242881072027, "step": 1450}, {"loss": 1.7541, "grad_norm": 0.33310577273368835, "learning_rate": 0.0002, "epoch": 0.9782244556113903, "step": 1460}, {"loss": 1.8511, "grad_norm": 0.2972261905670166, "learning_rate": 0.0002, "epoch": 0.9849246231155779, "step": 1470}, {"loss": 1.7654, "grad_norm": 0.3322717845439911, "learning_rate": 0.0002, "epoch": 0.9916247906197655, "step": 1480}, {"loss": 1.8033, "grad_norm": 0.3276330828666687, "learning_rate": 0.0002, "epoch": 0.998324958123953, "step": 1490}]}
+{"epoch": 2.0, "step": 2985, "epoch_duration": 1577.000766992569, "total_accumulated_duration": 3151.8114318847656, "gpu_info": {"GPU_0": "NVIDIA A100-SXM4-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 7568.4541015625}, "peak_memory_usage": {"GPU_0": 13792.75439453125}, "avg_memory_reserved": {"GPU_0": 17416.0}, "peak_memory_reserved": {"GPU_0": 17416.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "outputs-001/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-1492", "params": {"epochs": 8, "batch_size": 1, "learning_rate": 0.0002, "gradient_accumulation_steps": 2, "warmup_ratio": 0.03, "max_grad_norm": 0.3, "lora_alpha": 32, "lora_dropout": 0.05, "lora_r": 64, "tf32": true, "seed": 42}, "log_history": [{"loss": 2.6189, "grad_norm": 0.565915048122406, "learning_rate": 0.0002, "epoch": 0.006700167504187605, "step": 10}, {"loss": 2.3162, "grad_norm": 0.5004463791847229, "learning_rate": 0.0002, "epoch": 0.01340033500837521, "step": 20}, {"loss": 2.0576, "grad_norm": 0.511043906211853, "learning_rate": 0.0002, "epoch": 0.020100502512562814, "step": 30}, {"loss": 2.0085, "grad_norm": 0.47327178716659546, "learning_rate": 0.0002, "epoch": 0.02680067001675042, "step": 40}, {"loss": 2.0276, "grad_norm": 0.5511676669120789, "learning_rate": 0.0002, "epoch": 0.03350083752093802, "step": 50}, {"loss": 1.9075, "grad_norm": 0.4666278064250946, "learning_rate": 0.0002, "epoch": 0.04020100502512563, "step": 60}, {"loss": 1.8413, "grad_norm": 0.5310961008071899, "learning_rate": 0.0002, "epoch": 0.04690117252931323, "step": 70}, {"loss": 1.8711, "grad_norm": 0.5606027245521545, "learning_rate": 0.0002, "epoch": 0.05360134003350084, "step": 80}, {"loss": 1.9282, "grad_norm": 0.4934779703617096, "learning_rate": 0.0002, "epoch": 0.06030150753768844, "step": 90}, {"loss": 1.8925, "grad_norm": 0.4821869730949402, "learning_rate": 0.0002, "epoch": 0.06700167504187604, "step": 100}, {"loss": 1.8628, "grad_norm": 0.5262084603309631, "learning_rate": 0.0002, "epoch": 0.07370184254606366, "step": 110}, {"loss": 1.8347, "grad_norm": 0.3774230182170868, "learning_rate": 0.0002, "epoch": 0.08040201005025126, "step": 120}, {"loss": 1.8386, "grad_norm": 0.34137430787086487, "learning_rate": 0.0002, "epoch": 0.08710217755443886, "step": 130}, {"loss": 1.861, "grad_norm": 0.407272070646286, "learning_rate": 0.0002, "epoch": 0.09380234505862646, "step": 140}, {"loss": 1.8279, "grad_norm": 0.4011937975883484, "learning_rate": 0.0002, "epoch": 0.10050251256281408, "step": 150}, {"loss": 1.9317, "grad_norm": 0.4432467222213745, "learning_rate": 0.0002, "epoch": 0.10720268006700168, "step": 160}, {"loss": 1.8157, "grad_norm": 0.44030463695526123, "learning_rate": 0.0002, "epoch": 0.11390284757118928, "step": 170}, {"loss": 1.8534, "grad_norm": 0.3799569308757782, "learning_rate": 0.0002, "epoch": 0.12060301507537688, "step": 180}, {"loss": 1.7658, "grad_norm": 0.33721521496772766, "learning_rate": 0.0002, "epoch": 0.1273031825795645, "step": 190}, {"loss": 1.8269, "grad_norm": 0.4096226692199707, "learning_rate": 0.0002, "epoch": 0.13400335008375208, "step": 200}, {"loss": 1.802, "grad_norm": 0.37374693155288696, "learning_rate": 0.0002, "epoch": 0.1407035175879397, "step": 210}, {"loss": 1.8901, "grad_norm": 0.3249480128288269, "learning_rate": 0.0002, "epoch": 0.1474036850921273, "step": 220}, {"loss": 1.8163, "grad_norm": 0.3612042963504791, "learning_rate": 0.0002, "epoch": 0.1541038525963149, "step": 230}, {"loss": 1.7585, "grad_norm": 0.3686671257019043, "learning_rate": 0.0002, "epoch": 0.16080402010050251, "step": 240}, {"loss": 1.8365, "grad_norm": 0.3521044850349426, "learning_rate": 0.0002, "epoch": 0.16750418760469013, "step": 250}, {"loss": 1.8623, "grad_norm": 0.4073677361011505, "learning_rate": 0.0002, "epoch": 0.17420435510887772, "step": 260}, {"loss": 1.8026, "grad_norm": 0.34522193670272827, "learning_rate": 0.0002, "epoch": 0.18090452261306533, "step": 270}, {"loss": 1.8162, "grad_norm": 0.4121900498867035, "learning_rate": 0.0002, "epoch": 0.18760469011725292, "step": 280}, {"loss": 1.7976, "grad_norm": 0.3544778525829315, "learning_rate": 0.0002, "epoch": 0.19430485762144054, "step": 290}, {"loss": 1.8787, "grad_norm": 0.3482133448123932, "learning_rate": 0.0002, "epoch": 0.20100502512562815, "step": 300}, {"loss": 1.8578, "grad_norm": 0.3421826660633087, "learning_rate": 0.0002, "epoch": 0.20770519262981574, "step": 310}, {"loss": 1.8013, "grad_norm": 0.5024696588516235, "learning_rate": 0.0002, "epoch": 0.21440536013400335, "step": 320}, {"loss": 1.8607, "grad_norm": 0.36013063788414, "learning_rate": 0.0002, "epoch": 0.22110552763819097, "step": 330}, {"loss": 1.9075, "grad_norm": 0.3611244857311249, "learning_rate": 0.0002, "epoch": 0.22780569514237856, "step": 340}, {"loss": 1.8128, "grad_norm": 0.39244529604911804, "learning_rate": 0.0002, "epoch": 0.23450586264656617, "step": 350}, {"loss": 1.7885, "grad_norm": 0.3299325704574585, "learning_rate": 0.0002, "epoch": 0.24120603015075376, "step": 360}, {"loss": 1.8028, "grad_norm": 0.3994322419166565, "learning_rate": 0.0002, "epoch": 0.24790619765494137, "step": 370}, {"loss": 1.8321, "grad_norm": 0.3559151887893677, "learning_rate": 0.0002, "epoch": 0.254606365159129, "step": 380}, {"loss": 1.7802, "grad_norm": 0.3873756229877472, "learning_rate": 0.0002, "epoch": 0.2613065326633166, "step": 390}, {"loss": 1.7844, "grad_norm": 0.3710744082927704, "learning_rate": 0.0002, "epoch": 0.26800670016750416, "step": 400}, {"loss": 1.7789, "grad_norm": 0.3618465065956116, "learning_rate": 0.0002, "epoch": 0.2747068676716918, "step": 410}, {"loss": 1.8529, "grad_norm": 0.30063769221305847, "learning_rate": 0.0002, "epoch": 0.2814070351758794, "step": 420}, {"loss": 1.7765, "grad_norm": 0.3695628345012665, "learning_rate": 0.0002, "epoch": 0.288107202680067, "step": 430}, {"loss": 1.7982, "grad_norm": 0.31451135873794556, "learning_rate": 0.0002, "epoch": 0.2948073701842546, "step": 440}, {"loss": 1.7517, "grad_norm": 0.3959707021713257, "learning_rate": 0.0002, "epoch": 0.3015075376884422, "step": 450}, {"loss": 1.8142, "grad_norm": 0.33475354313850403, "learning_rate": 0.0002, "epoch": 0.3082077051926298, "step": 460}, {"loss": 1.8805, "grad_norm": 0.33933115005493164, "learning_rate": 0.0002, "epoch": 0.3149078726968174, "step": 470}, {"loss": 1.7564, "grad_norm": 0.3264943063259125, "learning_rate": 0.0002, "epoch": 0.32160804020100503, "step": 480}, {"loss": 1.8428, "grad_norm": 0.40188100934028625, "learning_rate": 0.0002, "epoch": 0.32830820770519265, "step": 490}, {"loss": 1.7624, "grad_norm": 0.37408649921417236, "learning_rate": 0.0002, "epoch": 0.33500837520938026, "step": 500}, {"loss": 1.7745, "grad_norm": 0.33925938606262207, "learning_rate": 0.0002, "epoch": 0.3417085427135678, "step": 510}, {"loss": 1.814, "grad_norm": 0.36836713552474976, "learning_rate": 0.0002, "epoch": 0.34840871021775544, "step": 520}, {"loss": 1.8037, "grad_norm": 0.37284499406814575, "learning_rate": 0.0002, "epoch": 0.35510887772194305, "step": 530}, {"loss": 1.8379, "grad_norm": 0.3192278742790222, "learning_rate": 0.0002, "epoch": 0.36180904522613067, "step": 540}, {"loss": 1.8702, "grad_norm": 0.30233290791511536, "learning_rate": 0.0002, "epoch": 0.3685092127303183, "step": 550}, {"loss": 1.8799, "grad_norm": 0.3340817391872406, "learning_rate": 0.0002, "epoch": 0.37520938023450584, "step": 560}, {"loss": 1.8404, "grad_norm": 0.32600095868110657, "learning_rate": 0.0002, "epoch": 0.38190954773869346, "step": 570}, {"loss": 1.7804, "grad_norm": 0.33711278438568115, "learning_rate": 0.0002, "epoch": 0.38860971524288107, "step": 580}, {"loss": 1.8445, "grad_norm": 0.34890690445899963, "learning_rate": 0.0002, "epoch": 0.3953098827470687, "step": 590}, {"loss": 1.8187, "grad_norm": 0.38238924741744995, "learning_rate": 0.0002, "epoch": 0.4020100502512563, "step": 600}, {"loss": 1.8111, "grad_norm": 0.34399354457855225, "learning_rate": 0.0002, "epoch": 0.40871021775544386, "step": 610}, {"loss": 1.8006, "grad_norm": 0.3346073627471924, "learning_rate": 0.0002, "epoch": 0.4154103852596315, "step": 620}, {"loss": 1.7705, "grad_norm": 0.3545648157596588, "learning_rate": 0.0002, "epoch": 0.4221105527638191, "step": 630}, {"loss": 1.8445, "grad_norm": 0.3378899097442627, "learning_rate": 0.0002, "epoch": 0.4288107202680067, "step": 640}, {"loss": 1.804, "grad_norm": 0.3255569040775299, "learning_rate": 0.0002, "epoch": 0.4355108877721943, "step": 650}, {"loss": 1.7679, "grad_norm": 0.34880587458610535, "learning_rate": 0.0002, "epoch": 0.44221105527638194, "step": 660}, {"loss": 1.7861, "grad_norm": 0.3402383625507355, "learning_rate": 0.0002, "epoch": 0.4489112227805695, "step": 670}, {"loss": 1.8131, "grad_norm": 0.3594033718109131, "learning_rate": 0.0002, "epoch": 0.4556113902847571, "step": 680}, {"loss": 1.8399, "grad_norm": 0.31000566482543945, "learning_rate": 0.0002, "epoch": 0.4623115577889447, "step": 690}, {"loss": 1.7521, "grad_norm": 0.37229061126708984, "learning_rate": 0.0002, "epoch": 0.46901172529313234, "step": 700}, {"loss": 1.7779, "grad_norm": 0.315801739692688, "learning_rate": 0.0002, "epoch": 0.47571189279731996, "step": 710}, {"loss": 1.7515, "grad_norm": 0.3220832645893097, "learning_rate": 0.0002, "epoch": 0.4824120603015075, "step": 720}, {"loss": 1.7181, "grad_norm": 0.3435456156730652, "learning_rate": 0.0002, "epoch": 0.48911222780569513, "step": 730}, {"loss": 1.8844, "grad_norm": 0.30380892753601074, "learning_rate": 0.0002, "epoch": 0.49581239530988275, "step": 740}, {"loss": 1.7792, "grad_norm": 0.3555026054382324, "learning_rate": 0.0002, "epoch": 0.5025125628140703, "step": 750}, {"loss": 1.7714, "grad_norm": 0.3019855320453644, "learning_rate": 0.0002, "epoch": 0.509212730318258, "step": 760}, {"loss": 1.7962, "grad_norm": 0.309111088514328, "learning_rate": 0.0002, "epoch": 0.5159128978224455, "step": 770}, {"loss": 1.7913, "grad_norm": 0.366020530462265, "learning_rate": 0.0002, "epoch": 0.5226130653266332, "step": 780}, {"loss": 1.8008, "grad_norm": 0.3267050087451935, "learning_rate": 0.0002, "epoch": 0.5293132328308208, "step": 790}, {"loss": 1.7397, "grad_norm": 0.34265750646591187, "learning_rate": 0.0002, "epoch": 0.5360134003350083, "step": 800}, {"loss": 1.8251, "grad_norm": 0.313669890165329, "learning_rate": 0.0002, "epoch": 0.542713567839196, "step": 810}, {"loss": 1.8369, "grad_norm": 0.3355236053466797, "learning_rate": 0.0002, "epoch": 0.5494137353433836, "step": 820}, {"loss": 1.7381, "grad_norm": 0.3186608552932739, "learning_rate": 0.0002, "epoch": 0.5561139028475712, "step": 830}, {"loss": 1.8034, "grad_norm": 0.30357518792152405, "learning_rate": 0.0002, "epoch": 0.5628140703517588, "step": 840}, {"loss": 1.769, "grad_norm": 0.3990040123462677, "learning_rate": 0.0002, "epoch": 0.5695142378559463, "step": 850}, {"loss": 1.7482, "grad_norm": 0.34363803267478943, "learning_rate": 0.0002, "epoch": 0.576214405360134, "step": 860}, {"loss": 1.8106, "grad_norm": 0.3757908046245575, "learning_rate": 0.0002, "epoch": 0.5829145728643216, "step": 870}, {"loss": 1.8104, "grad_norm": 0.3359757661819458, "learning_rate": 0.0002, "epoch": 0.5896147403685092, "step": 880}, {"loss": 1.7591, "grad_norm": 0.5555329918861389, "learning_rate": 0.0002, "epoch": 0.5963149078726968, "step": 890}, {"loss": 1.7715, "grad_norm": 0.4046323895454407, "learning_rate": 0.0002, "epoch": 0.6030150753768844, "step": 900}, {"loss": 1.7998, "grad_norm": 0.29834219813346863, "learning_rate": 0.0002, "epoch": 0.609715242881072, "step": 910}, {"loss": 1.7826, "grad_norm": 0.3241238594055176, "learning_rate": 0.0002, "epoch": 0.6164154103852596, "step": 920}, {"loss": 1.8342, "grad_norm": 0.35154739022254944, "learning_rate": 0.0002, "epoch": 0.6231155778894473, "step": 930}, {"loss": 1.8076, "grad_norm": 0.3287706673145294, "learning_rate": 0.0002, "epoch": 0.6298157453936348, "step": 940}, {"loss": 1.8038, "grad_norm": 0.35670626163482666, "learning_rate": 0.0002, "epoch": 0.6365159128978225, "step": 950}, {"loss": 1.869, "grad_norm": 0.6114104986190796, "learning_rate": 0.0002, "epoch": 0.6432160804020101, "step": 960}, {"loss": 1.8297, "grad_norm": 0.3186565041542053, "learning_rate": 0.0002, "epoch": 0.6499162479061976, "step": 970}, {"loss": 1.7539, "grad_norm": 0.27164125442504883, "learning_rate": 0.0002, "epoch": 0.6566164154103853, "step": 980}, {"loss": 1.8339, "grad_norm": 0.34407344460487366, "learning_rate": 0.0002, "epoch": 0.6633165829145728, "step": 990}, {"loss": 1.855, "grad_norm": 0.368415892124176, "learning_rate": 0.0002, "epoch": 0.6700167504187605, "step": 1000}, {"loss": 1.7821, "grad_norm": 0.3306390643119812, "learning_rate": 0.0002, "epoch": 0.6767169179229481, "step": 1010}, {"loss": 1.786, "grad_norm": 0.3198648989200592, "learning_rate": 0.0002, "epoch": 0.6834170854271356, "step": 1020}, {"loss": 1.816, "grad_norm": 0.3092987537384033, "learning_rate": 0.0002, "epoch": 0.6901172529313233, "step": 1030}, {"loss": 1.7689, "grad_norm": 0.3090653419494629, "learning_rate": 0.0002, "epoch": 0.6968174204355109, "step": 1040}, {"loss": 1.7544, "grad_norm": 0.3485880196094513, "learning_rate": 0.0002, "epoch": 0.7035175879396985, "step": 1050}, {"loss": 1.8187, "grad_norm": 0.35782721638679504, "learning_rate": 0.0002, "epoch": 0.7102177554438861, "step": 1060}, {"loss": 1.806, "grad_norm": 0.34256869554519653, "learning_rate": 0.0002, "epoch": 0.7169179229480737, "step": 1070}, {"loss": 1.7873, "grad_norm": 0.30461037158966064, "learning_rate": 0.0002, "epoch": 0.7236180904522613, "step": 1080}, {"loss": 1.7367, "grad_norm": 0.3398691713809967, "learning_rate": 0.0002, "epoch": 0.7303182579564489, "step": 1090}, {"loss": 1.8756, "grad_norm": 0.3180808126926422, "learning_rate": 0.0002, "epoch": 0.7370184254606366, "step": 1100}, {"loss": 1.6988, "grad_norm": 0.34400665760040283, "learning_rate": 0.0002, "epoch": 0.7437185929648241, "step": 1110}, {"loss": 1.7851, "grad_norm": 0.34244877099990845, "learning_rate": 0.0002, "epoch": 0.7504187604690117, "step": 1120}, {"loss": 1.7841, "grad_norm": 0.29946693778038025, "learning_rate": 0.0002, "epoch": 0.7571189279731994, "step": 1130}, {"loss": 1.7456, "grad_norm": 0.37547236680984497, "learning_rate": 0.0002, "epoch": 0.7638190954773869, "step": 1140}, {"loss": 1.8425, "grad_norm": 0.3263005018234253, "learning_rate": 0.0002, "epoch": 0.7705192629815746, "step": 1150}, {"loss": 1.7222, "grad_norm": 0.41363608837127686, "learning_rate": 0.0002, "epoch": 0.7772194304857621, "step": 1160}, {"loss": 1.7836, "grad_norm": 0.36267954111099243, "learning_rate": 0.0002, "epoch": 0.7839195979899497, "step": 1170}, {"loss": 1.9183, "grad_norm": 0.31789499521255493, "learning_rate": 0.0002, "epoch": 0.7906197654941374, "step": 1180}, {"loss": 1.78, "grad_norm": 0.5708149075508118, "learning_rate": 0.0002, "epoch": 0.7973199329983249, "step": 1190}, {"loss": 1.6908, "grad_norm": 0.322099506855011, "learning_rate": 0.0002, "epoch": 0.8040201005025126, "step": 1200}, {"loss": 1.7639, "grad_norm": 0.3419909179210663, "learning_rate": 0.0002, "epoch": 0.8107202680067002, "step": 1210}, {"loss": 1.7428, "grad_norm": 0.36286255717277527, "learning_rate": 0.0002, "epoch": 0.8174204355108877, "step": 1220}, {"loss": 1.8409, "grad_norm": 0.33992862701416016, "learning_rate": 0.0002, "epoch": 0.8241206030150754, "step": 1230}, {"loss": 1.7507, "grad_norm": 0.32622793316841125, "learning_rate": 0.0002, "epoch": 0.830820770519263, "step": 1240}, {"loss": 1.8098, "grad_norm": 0.3036167621612549, "learning_rate": 0.0002, "epoch": 0.8375209380234506, "step": 1250}, {"loss": 1.8094, "grad_norm": 0.3182215392589569, "learning_rate": 0.0002, "epoch": 0.8442211055276382, "step": 1260}, {"loss": 1.8017, "grad_norm": 0.3270018696784973, "learning_rate": 0.0002, "epoch": 0.8509212730318257, "step": 1270}, {"loss": 1.798, "grad_norm": 0.32652342319488525, "learning_rate": 0.0002, "epoch": 0.8576214405360134, "step": 1280}, {"loss": 1.7448, "grad_norm": 0.3631329834461212, "learning_rate": 0.0002, "epoch": 0.864321608040201, "step": 1290}, {"loss": 1.7, "grad_norm": 0.36706018447875977, "learning_rate": 0.0002, "epoch": 0.8710217755443886, "step": 1300}, {"loss": 1.8178, "grad_norm": 0.3347418010234833, "learning_rate": 0.0002, "epoch": 0.8777219430485762, "step": 1310}, {"loss": 1.7824, "grad_norm": 0.34371060132980347, "learning_rate": 0.0002, "epoch": 0.8844221105527639, "step": 1320}, {"loss": 1.783, "grad_norm": 0.3029090166091919, "learning_rate": 0.0002, "epoch": 0.8911222780569514, "step": 1330}, {"loss": 1.8017, "grad_norm": 0.34700682759284973, "learning_rate": 0.0002, "epoch": 0.897822445561139, "step": 1340}, {"loss": 1.7998, "grad_norm": 0.35574328899383545, "learning_rate": 0.0002, "epoch": 0.9045226130653267, "step": 1350}, {"loss": 1.7339, "grad_norm": 0.30928221344947815, "learning_rate": 0.0002, "epoch": 0.9112227805695142, "step": 1360}, {"loss": 1.7479, "grad_norm": 0.30652928352355957, "learning_rate": 0.0002, "epoch": 0.9179229480737019, "step": 1370}, {"loss": 1.7491, "grad_norm": 0.3838157653808594, "learning_rate": 0.0002, "epoch": 0.9246231155778895, "step": 1380}, {"loss": 1.7977, "grad_norm": 0.31655240058898926, "learning_rate": 0.0002, "epoch": 0.931323283082077, "step": 1390}, {"loss": 1.8175, "grad_norm": 0.41737303137779236, "learning_rate": 0.0002, "epoch": 0.9380234505862647, "step": 1400}, {"loss": 1.6811, "grad_norm": 0.3227267861366272, "learning_rate": 0.0002, "epoch": 0.9447236180904522, "step": 1410}, {"loss": 1.7343, "grad_norm": 0.3729925751686096, "learning_rate": 0.0002, "epoch": 0.9514237855946399, "step": 1420}, {"loss": 1.8221, "grad_norm": 0.30779409408569336, "learning_rate": 0.0002, "epoch": 0.9581239530988275, "step": 1430}, {"loss": 1.7972, "grad_norm": 0.334379643201828, "learning_rate": 0.0002, "epoch": 0.964824120603015, "step": 1440}, {"loss": 1.7141, "grad_norm": 0.3568236231803894, "learning_rate": 0.0002, "epoch": 0.9715242881072027, "step": 1450}, {"loss": 1.7541, "grad_norm": 0.33310577273368835, "learning_rate": 0.0002, "epoch": 0.9782244556113903, "step": 1460}, {"loss": 1.8511, "grad_norm": 0.2972261905670166, "learning_rate": 0.0002, "epoch": 0.9849246231155779, "step": 1470}, {"loss": 1.7654, "grad_norm": 0.3322717845439911, "learning_rate": 0.0002, "epoch": 0.9916247906197655, "step": 1480}, {"loss": 1.8033, "grad_norm": 0.3276330828666687, "learning_rate": 0.0002, "epoch": 0.998324958123953, "step": 1490}, {"eval_loss": 1.8036354780197144, "eval_runtime": 37.8949, "eval_samples_per_second": 13.59, "eval_steps_per_second": 1.715, "epoch": 0.9996649916247906, "step": 1492}, {"loss": 1.7138, "grad_norm": 0.29252371191978455, "learning_rate": 0.0002, "epoch": 1.0050251256281406, "step": 1500}, {"loss": 1.8198, "grad_norm": 0.31607162952423096, "learning_rate": 0.0002, "epoch": 1.0117252931323284, "step": 1510}, {"loss": 1.6779, "grad_norm": 0.32294467091560364, "learning_rate": 0.0002, "epoch": 1.018425460636516, "step": 1520}, {"loss": 1.7919, "grad_norm": 0.3868017792701721, "learning_rate": 0.0002, "epoch": 1.0251256281407035, "step": 1530}, {"loss": 1.7954, "grad_norm": 0.3178282082080841, "learning_rate": 0.0002, "epoch": 1.031825795644891, "step": 1540}, {"loss": 1.7136, "grad_norm": 0.3706750273704529, "learning_rate": 0.0002, "epoch": 1.0385259631490786, "step": 1550}, {"loss": 1.7382, "grad_norm": 0.33930912613868713, "learning_rate": 0.0002, "epoch": 1.0452261306532664, "step": 1560}, {"loss": 1.7602, "grad_norm": 0.33970504999160767, "learning_rate": 0.0002, "epoch": 1.051926298157454, "step": 1570}, {"loss": 1.6573, "grad_norm": 0.42553383111953735, "learning_rate": 0.0002, "epoch": 1.0586264656616415, "step": 1580}, {"loss": 1.645, "grad_norm": 0.3772421181201935, "learning_rate": 0.0002, "epoch": 1.065326633165829, "step": 1590}, {"loss": 1.7362, "grad_norm": 0.34212902188301086, "learning_rate": 0.0002, "epoch": 1.0720268006700167, "step": 1600}, {"loss": 1.7057, "grad_norm": 0.3798283338546753, "learning_rate": 0.0002, "epoch": 1.0787269681742044, "step": 1610}, {"loss": 1.7468, "grad_norm": 0.36909598112106323, "learning_rate": 0.0002, "epoch": 1.085427135678392, "step": 1620}, {"loss": 1.7807, "grad_norm": 0.3344230651855469, "learning_rate": 0.0002, "epoch": 1.0921273031825796, "step": 1630}, {"loss": 1.7111, "grad_norm": 0.3862569332122803, "learning_rate": 0.0002, "epoch": 1.0988274706867671, "step": 1640}, {"loss": 1.7163, "grad_norm": 0.31188511848449707, "learning_rate": 0.0002, "epoch": 1.1055276381909547, "step": 1650}, {"loss": 1.7263, "grad_norm": 0.3563670814037323, "learning_rate": 0.0002, "epoch": 1.1122278056951425, "step": 1660}, {"loss": 1.7718, "grad_norm": 0.35052165389060974, "learning_rate": 0.0002, "epoch": 1.11892797319933, "step": 1670}, {"loss": 1.7601, "grad_norm": 0.3285699188709259, "learning_rate": 0.0002, "epoch": 1.1256281407035176, "step": 1680}, {"loss": 1.6877, "grad_norm": 0.3639393746852875, "learning_rate": 0.0002, "epoch": 1.1323283082077051, "step": 1690}, {"loss": 1.7719, "grad_norm": 0.3842753767967224, "learning_rate": 0.0002, "epoch": 1.1390284757118927, "step": 1700}, {"loss": 1.7002, "grad_norm": 0.3624933063983917, "learning_rate": 0.0002, "epoch": 1.1457286432160805, "step": 1710}, {"loss": 1.7243, "grad_norm": 0.3641220033168793, "learning_rate": 0.0002, "epoch": 1.152428810720268, "step": 1720}, {"loss": 1.752, "grad_norm": 0.32765355706214905, "learning_rate": 0.0002, "epoch": 1.1591289782244556, "step": 1730}, {"loss": 1.6556, "grad_norm": 0.34974896907806396, "learning_rate": 0.0002, "epoch": 1.1658291457286432, "step": 1740}, {"loss": 1.7273, "grad_norm": 0.3910926580429077, "learning_rate": 0.0002, "epoch": 1.1725293132328307, "step": 1750}, {"loss": 1.7204, "grad_norm": 0.3564300537109375, "learning_rate": 0.0002, "epoch": 1.1792294807370185, "step": 1760}, {"loss": 1.746, "grad_norm": 0.34822574257850647, "learning_rate": 0.0002, "epoch": 1.185929648241206, "step": 1770}, {"loss": 1.7256, "grad_norm": 0.36185044050216675, "learning_rate": 0.0002, "epoch": 1.1926298157453936, "step": 1780}, {"loss": 1.6431, "grad_norm": 0.34866711497306824, "learning_rate": 0.0002, "epoch": 1.1993299832495812, "step": 1790}, {"loss": 1.8084, "grad_norm": 0.4017769992351532, "learning_rate": 0.0002, "epoch": 1.2060301507537687, "step": 1800}, {"loss": 1.6985, "grad_norm": 0.32930681109428406, "learning_rate": 0.0002, "epoch": 1.2127303182579565, "step": 1810}, {"loss": 1.7606, "grad_norm": 0.35951921343803406, "learning_rate": 0.0002, "epoch": 1.219430485762144, "step": 1820}, {"loss": 1.6933, "grad_norm": 0.37366992235183716, "learning_rate": 0.0002, "epoch": 1.2261306532663316, "step": 1830}, {"loss": 1.6737, "grad_norm": 0.3565689027309418, "learning_rate": 0.0002, "epoch": 1.2328308207705192, "step": 1840}, {"loss": 1.8013, "grad_norm": 0.3692343533039093, "learning_rate": 0.0002, "epoch": 1.2395309882747068, "step": 1850}, {"loss": 1.736, "grad_norm": 0.38426971435546875, "learning_rate": 0.0002, "epoch": 1.2462311557788945, "step": 1860}, {"loss": 1.7031, "grad_norm": 0.33559855818748474, "learning_rate": 0.0002, "epoch": 1.252931323283082, "step": 1870}, {"loss": 1.7033, "grad_norm": 0.34181106090545654, "learning_rate": 0.0002, "epoch": 1.2596314907872697, "step": 1880}, {"loss": 1.7707, "grad_norm": 0.3916318416595459, "learning_rate": 0.0002, "epoch": 1.2663316582914572, "step": 1890}, {"loss": 1.6686, "grad_norm": 0.3887825012207031, "learning_rate": 0.0002, "epoch": 1.2730318257956448, "step": 1900}, {"loss": 1.7062, "grad_norm": 0.33583927154541016, "learning_rate": 0.0002, "epoch": 1.2797319932998326, "step": 1910}, {"loss": 1.717, "grad_norm": 0.37639349699020386, "learning_rate": 0.0002, "epoch": 1.2864321608040201, "step": 1920}, {"loss": 1.777, "grad_norm": 0.38059428334236145, "learning_rate": 0.0002, "epoch": 1.2931323283082077, "step": 1930}, {"loss": 1.6126, "grad_norm": 0.37253183126449585, "learning_rate": 0.0002, "epoch": 1.2998324958123952, "step": 1940}, {"loss": 1.6758, "grad_norm": 0.37371566891670227, "learning_rate": 0.0002, "epoch": 1.3065326633165828, "step": 1950}, {"loss": 1.6788, "grad_norm": 0.4080910086631775, "learning_rate": 0.0002, "epoch": 1.3132328308207706, "step": 1960}, {"loss": 1.6518, "grad_norm": 0.3174354135990143, "learning_rate": 0.0002, "epoch": 1.3199329983249581, "step": 1970}, {"loss": 1.7925, "grad_norm": 0.4518888294696808, "learning_rate": 0.0002, "epoch": 1.3266331658291457, "step": 1980}, {"loss": 1.7085, "grad_norm": 0.3627921938896179, "learning_rate": 0.0002, "epoch": 1.3333333333333333, "step": 1990}, {"loss": 1.7676, "grad_norm": 0.3655930161476135, "learning_rate": 0.0002, "epoch": 1.3400335008375208, "step": 2000}, {"loss": 1.7016, "grad_norm": 0.3509993255138397, "learning_rate": 0.0002, "epoch": 1.3467336683417086, "step": 2010}, {"loss": 1.7359, "grad_norm": 0.4281129240989685, "learning_rate": 0.0002, "epoch": 1.3534338358458962, "step": 2020}, {"loss": 1.6884, "grad_norm": 0.3821414113044739, "learning_rate": 0.0002, "epoch": 1.3601340033500837, "step": 2030}, {"loss": 1.7075, "grad_norm": 0.3907586336135864, "learning_rate": 0.0002, "epoch": 1.3668341708542713, "step": 2040}, {"loss": 1.7424, "grad_norm": 0.37792932987213135, "learning_rate": 0.0002, "epoch": 1.3735343383584588, "step": 2050}, {"loss": 1.7305, "grad_norm": 0.3693985641002655, "learning_rate": 0.0002, "epoch": 1.3802345058626466, "step": 2060}, {"loss": 1.7434, "grad_norm": 0.32275936007499695, "learning_rate": 0.0002, "epoch": 1.3869346733668342, "step": 2070}, {"loss": 1.6677, "grad_norm": 0.3789440095424652, "learning_rate": 0.0002, "epoch": 1.3936348408710217, "step": 2080}, {"loss": 1.6825, "grad_norm": 0.3638380467891693, "learning_rate": 0.0002, "epoch": 1.4003350083752093, "step": 2090}, {"loss": 1.6542, "grad_norm": 0.3495481610298157, "learning_rate": 0.0002, "epoch": 1.4070351758793969, "step": 2100}, {"loss": 1.7225, "grad_norm": 0.37920597195625305, "learning_rate": 0.0002, "epoch": 1.4137353433835846, "step": 2110}, {"loss": 1.7329, "grad_norm": 0.37218064069747925, "learning_rate": 0.0002, "epoch": 1.4204355108877722, "step": 2120}, {"loss": 1.799, "grad_norm": 0.38074082136154175, "learning_rate": 0.0002, "epoch": 1.4271356783919598, "step": 2130}, {"loss": 1.7403, "grad_norm": 0.3455527126789093, "learning_rate": 0.0002, "epoch": 1.4338358458961473, "step": 2140}, {"loss": 1.776, "grad_norm": 0.3712003529071808, "learning_rate": 0.0002, "epoch": 1.4405360134003349, "step": 2150}, {"loss": 1.7619, "grad_norm": 0.3786754906177521, "learning_rate": 0.0002, "epoch": 1.4472361809045227, "step": 2160}, {"loss": 1.68, "grad_norm": 0.3879223167896271, "learning_rate": 0.0002, "epoch": 1.4539363484087102, "step": 2170}, {"loss": 1.7, "grad_norm": 0.38738805055618286, "learning_rate": 0.0002, "epoch": 1.4606365159128978, "step": 2180}, {"loss": 1.7581, "grad_norm": 0.39768800139427185, "learning_rate": 0.0002, "epoch": 1.4673366834170856, "step": 2190}, {"loss": 1.7671, "grad_norm": 0.4172441065311432, "learning_rate": 0.0002, "epoch": 1.474036850921273, "step": 2200}, {"loss": 1.6736, "grad_norm": 0.4043174982070923, "learning_rate": 0.0002, "epoch": 1.4807370184254607, "step": 2210}, {"loss": 1.7444, "grad_norm": 0.3750883936882019, "learning_rate": 0.0002, "epoch": 1.4874371859296482, "step": 2220}, {"loss": 1.6861, "grad_norm": 0.3552253246307373, "learning_rate": 0.0002, "epoch": 1.4941373534338358, "step": 2230}, {"loss": 1.6471, "grad_norm": 0.34607139229774475, "learning_rate": 0.0002, "epoch": 1.5008375209380236, "step": 2240}, {"loss": 1.6962, "grad_norm": 0.3406706750392914, "learning_rate": 0.0002, "epoch": 1.507537688442211, "step": 2250}, {"loss": 1.7694, "grad_norm": 0.36654895544052124, "learning_rate": 0.0002, "epoch": 1.5142378559463987, "step": 2260}, {"loss": 1.6812, "grad_norm": 0.3914054334163666, "learning_rate": 0.0002, "epoch": 1.5209380234505863, "step": 2270}, {"loss": 1.6822, "grad_norm": 0.42012137174606323, "learning_rate": 0.0002, "epoch": 1.5276381909547738, "step": 2280}, {"loss": 1.697, "grad_norm": 0.39563435316085815, "learning_rate": 0.0002, "epoch": 1.5343383584589616, "step": 2290}, {"loss": 1.7491, "grad_norm": 0.3508438766002655, "learning_rate": 0.0002, "epoch": 1.541038525963149, "step": 2300}, {"loss": 1.7727, "grad_norm": 0.3785218596458435, "learning_rate": 0.0002, "epoch": 1.5477386934673367, "step": 2310}, {"loss": 1.6963, "grad_norm": 0.39377647638320923, "learning_rate": 0.0002, "epoch": 1.5544388609715243, "step": 2320}, {"loss": 1.7263, "grad_norm": 0.3391438126564026, "learning_rate": 0.0002, "epoch": 1.5611390284757118, "step": 2330}, {"loss": 1.7722, "grad_norm": 0.37944263219833374, "learning_rate": 0.0002, "epoch": 1.5678391959798996, "step": 2340}, {"loss": 1.6371, "grad_norm": 0.3523491322994232, "learning_rate": 0.0002, "epoch": 1.574539363484087, "step": 2350}, {"loss": 1.7583, "grad_norm": 0.3911575973033905, "learning_rate": 0.0002, "epoch": 1.5812395309882747, "step": 2360}, {"loss": 1.7117, "grad_norm": 0.33832186460494995, "learning_rate": 0.0002, "epoch": 1.5879396984924623, "step": 2370}, {"loss": 1.7701, "grad_norm": 0.3665979206562042, "learning_rate": 0.0002, "epoch": 1.5946398659966499, "step": 2380}, {"loss": 1.779, "grad_norm": 0.3871748149394989, "learning_rate": 0.0002, "epoch": 1.6013400335008376, "step": 2390}, {"loss": 1.7109, "grad_norm": 0.3586967885494232, "learning_rate": 0.0002, "epoch": 1.608040201005025, "step": 2400}, {"loss": 1.7096, "grad_norm": 0.3563673198223114, "learning_rate": 0.0002, "epoch": 1.6147403685092128, "step": 2410}, {"loss": 1.745, "grad_norm": 0.37588971853256226, "learning_rate": 0.0002, "epoch": 1.6214405360134003, "step": 2420}, {"loss": 1.7086, "grad_norm": 0.352556437253952, "learning_rate": 0.0002, "epoch": 1.6281407035175879, "step": 2430}, {"loss": 1.6547, "grad_norm": 0.3716259300708771, "learning_rate": 0.0002, "epoch": 1.6348408710217757, "step": 2440}, {"loss": 1.7033, "grad_norm": 0.372001975774765, "learning_rate": 0.0002, "epoch": 1.641541038525963, "step": 2450}, {"loss": 1.6584, "grad_norm": 0.3430042862892151, "learning_rate": 0.0002, "epoch": 1.6482412060301508, "step": 2460}, {"loss": 1.7217, "grad_norm": 0.3741483688354492, "learning_rate": 0.0002, "epoch": 1.6549413735343383, "step": 2470}, {"loss": 1.7701, "grad_norm": 0.3610571324825287, "learning_rate": 0.0002, "epoch": 1.661641541038526, "step": 2480}, {"loss": 1.7057, "grad_norm": 0.4204719066619873, "learning_rate": 0.0002, "epoch": 1.6683417085427137, "step": 2490}, {"loss": 1.7954, "grad_norm": 0.3938186466693878, "learning_rate": 0.0002, "epoch": 1.675041876046901, "step": 2500}, {"loss": 1.6633, "grad_norm": 0.3421435058116913, "learning_rate": 0.0002, "epoch": 1.6817420435510888, "step": 2510}, {"loss": 1.7996, "grad_norm": 0.42441412806510925, "learning_rate": 0.0002, "epoch": 1.6884422110552764, "step": 2520}, {"loss": 1.7142, "grad_norm": 0.38071519136428833, "learning_rate": 0.0002, "epoch": 1.695142378559464, "step": 2530}, {"loss": 1.7232, "grad_norm": 0.34078919887542725, "learning_rate": 0.0002, "epoch": 1.7018425460636517, "step": 2540}, {"loss": 1.7126, "grad_norm": 0.412844181060791, "learning_rate": 0.0002, "epoch": 1.708542713567839, "step": 2550}, {"loss": 1.7149, "grad_norm": 0.3753604292869568, "learning_rate": 0.0002, "epoch": 1.7152428810720268, "step": 2560}, {"loss": 1.7011, "grad_norm": 0.41588476300239563, "learning_rate": 0.0002, "epoch": 1.7219430485762144, "step": 2570}, {"loss": 1.6427, "grad_norm": 0.35504111647605896, "learning_rate": 0.0002, "epoch": 1.728643216080402, "step": 2580}, {"loss": 1.7296, "grad_norm": 0.36909720301628113, "learning_rate": 0.0002, "epoch": 1.7353433835845897, "step": 2590}, {"loss": 1.7022, "grad_norm": 0.4149979054927826, "learning_rate": 0.0002, "epoch": 1.742043551088777, "step": 2600}, {"loss": 1.77, "grad_norm": 0.38859328627586365, "learning_rate": 0.0002, "epoch": 1.7487437185929648, "step": 2610}, {"loss": 1.7036, "grad_norm": 0.36738792061805725, "learning_rate": 0.0002, "epoch": 1.7554438860971524, "step": 2620}, {"loss": 1.764, "grad_norm": 0.3968178927898407, "learning_rate": 0.0002, "epoch": 1.76214405360134, "step": 2630}, {"loss": 1.7687, "grad_norm": 0.3972901999950409, "learning_rate": 0.0002, "epoch": 1.7688442211055277, "step": 2640}, {"loss": 1.6774, "grad_norm": 0.3949959874153137, "learning_rate": 0.0002, "epoch": 1.775544388609715, "step": 2650}, {"loss": 1.7247, "grad_norm": 0.44074657559394836, "learning_rate": 0.0002, "epoch": 1.7822445561139029, "step": 2660}, {"loss": 1.7188, "grad_norm": 0.39743664860725403, "learning_rate": 0.0002, "epoch": 1.7889447236180904, "step": 2670}, {"loss": 1.7258, "grad_norm": 0.3950406610965729, "learning_rate": 0.0002, "epoch": 1.795644891122278, "step": 2680}, {"loss": 1.6906, "grad_norm": 0.3568263649940491, "learning_rate": 0.0002, "epoch": 1.8023450586264658, "step": 2690}, {"loss": 1.6735, "grad_norm": 0.3819476366043091, "learning_rate": 0.0002, "epoch": 1.809045226130653, "step": 2700}, {"loss": 1.7198, "grad_norm": 0.3480634391307831, "learning_rate": 0.0002, "epoch": 1.8157453936348409, "step": 2710}, {"loss": 1.7042, "grad_norm": 0.3875853419303894, "learning_rate": 0.0002, "epoch": 1.8224455611390284, "step": 2720}, {"loss": 1.6988, "grad_norm": 0.3441337049007416, "learning_rate": 0.0002, "epoch": 1.829145728643216, "step": 2730}, {"loss": 1.7647, "grad_norm": 0.35692882537841797, "learning_rate": 0.0002, "epoch": 1.8358458961474038, "step": 2740}, {"loss": 1.7033, "grad_norm": 0.36959215998649597, "learning_rate": 0.0002, "epoch": 1.8425460636515911, "step": 2750}, {"loss": 1.7657, "grad_norm": 0.3893393278121948, "learning_rate": 0.0002, "epoch": 1.849246231155779, "step": 2760}, {"loss": 1.7068, "grad_norm": 0.37817293405532837, "learning_rate": 0.0002, "epoch": 1.8559463986599665, "step": 2770}, {"loss": 1.761, "grad_norm": 0.36071285605430603, "learning_rate": 0.0002, "epoch": 1.862646566164154, "step": 2780}, {"loss": 1.7623, "grad_norm": 0.3758420944213867, "learning_rate": 0.0002, "epoch": 1.8693467336683418, "step": 2790}, {"loss": 1.6743, "grad_norm": 0.3889938294887543, "learning_rate": 0.0002, "epoch": 1.8760469011725294, "step": 2800}, {"loss": 1.6151, "grad_norm": 0.34361857175827026, "learning_rate": 0.0002, "epoch": 1.882747068676717, "step": 2810}, {"loss": 1.6038, "grad_norm": 0.39283323287963867, "learning_rate": 0.0002, "epoch": 1.8894472361809045, "step": 2820}, {"loss": 1.7555, "grad_norm": 0.3919452726840973, "learning_rate": 0.0002, "epoch": 1.896147403685092, "step": 2830}, {"loss": 1.673, "grad_norm": 0.38215070962905884, "learning_rate": 0.0002, "epoch": 1.9028475711892798, "step": 2840}, {"loss": 1.7044, "grad_norm": 0.4235064387321472, "learning_rate": 0.0002, "epoch": 1.9095477386934674, "step": 2850}, {"loss": 1.7123, "grad_norm": 0.35694634914398193, "learning_rate": 0.0002, "epoch": 1.916247906197655, "step": 2860}, {"loss": 1.8128, "grad_norm": 0.383492112159729, "learning_rate": 0.0002, "epoch": 1.9229480737018425, "step": 2870}, {"loss": 1.7581, "grad_norm": 0.5945147275924683, "learning_rate": 0.0002, "epoch": 1.92964824120603, "step": 2880}, {"loss": 1.7421, "grad_norm": 0.3367522358894348, "learning_rate": 0.0002, "epoch": 1.9363484087102178, "step": 2890}, {"loss": 1.6561, "grad_norm": 0.35300394892692566, "learning_rate": 0.0002, "epoch": 1.9430485762144054, "step": 2900}, {"loss": 1.7033, "grad_norm": 0.38084495067596436, "learning_rate": 0.0002, "epoch": 1.949748743718593, "step": 2910}, {"loss": 1.7132, "grad_norm": 0.37559160590171814, "learning_rate": 0.0002, "epoch": 1.9564489112227805, "step": 2920}, {"loss": 1.6759, "grad_norm": 0.3661738336086273, "learning_rate": 0.0002, "epoch": 1.963149078726968, "step": 2930}, {"loss": 1.7643, "grad_norm": 0.4073849320411682, "learning_rate": 0.0002, "epoch": 1.9698492462311559, "step": 2940}, {"loss": 1.6806, "grad_norm": 0.3723304271697998, "learning_rate": 0.0002, "epoch": 1.9765494137353434, "step": 2950}, {"loss": 1.7611, "grad_norm": 0.3991098999977112, "learning_rate": 0.0002, "epoch": 1.983249581239531, "step": 2960}, {"loss": 1.7263, "grad_norm": 0.3947085440158844, "learning_rate": 0.0002, "epoch": 1.9899497487437185, "step": 2970}, {"loss": 1.7217, "grad_norm": 0.3786258399486542, "learning_rate": 0.0002, "epoch": 1.996649916247906, "step": 2980}]}
+{"epoch": 2.9996649916247904, "step": 4477, "epoch_duration": 1576.7374460697174, "total_accumulated_duration": 4728.548877954483, "gpu_info": {"GPU_0": "NVIDIA A100-SXM4-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 7672.4541015625}, "peak_memory_usage": {"GPU_0": 13792.75439453125}, "avg_memory_reserved": {"GPU_0": 17416.0}, "peak_memory_reserved": {"GPU_0": 17416.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "outputs-001/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-2985", "params": {"epochs": 8, "batch_size": 1, "learning_rate": 0.0002, "gradient_accumulation_steps": 2, "warmup_ratio": 0.03, "max_grad_norm": 0.3, "lora_alpha": 32, "lora_dropout": 0.05, "lora_r": 64, "tf32": true, "seed": 42}, "log_history": [{"loss": 2.6189, "grad_norm": 0.565915048122406, "learning_rate": 0.0002, "epoch": 0.006700167504187605, "step": 10}, {"loss": 2.3162, "grad_norm": 0.5004463791847229, "learning_rate": 0.0002, "epoch": 0.01340033500837521, "step": 20}, {"loss": 2.0576, "grad_norm": 0.511043906211853, "learning_rate": 0.0002, "epoch": 0.020100502512562814, "step": 30}, {"loss": 2.0085, "grad_norm": 0.47327178716659546, "learning_rate": 0.0002, "epoch": 0.02680067001675042, "step": 40}, {"loss": 2.0276, "grad_norm": 0.5511676669120789, "learning_rate": 0.0002, "epoch": 0.03350083752093802, "step": 50}, {"loss": 1.9075, "grad_norm": 0.4666278064250946, "learning_rate": 0.0002, "epoch": 0.04020100502512563, "step": 60}, {"loss": 1.8413, "grad_norm": 0.5310961008071899, "learning_rate": 0.0002, "epoch": 0.04690117252931323, "step": 70}, {"loss": 1.8711, "grad_norm": 0.5606027245521545, "learning_rate": 0.0002, "epoch": 0.05360134003350084, "step": 80}, {"loss": 1.9282, "grad_norm": 0.4934779703617096, "learning_rate": 0.0002, "epoch": 0.06030150753768844, "step": 90}, {"loss": 1.8925, "grad_norm": 0.4821869730949402, "learning_rate": 0.0002, "epoch": 0.06700167504187604, "step": 100}, {"loss": 1.8628, "grad_norm": 0.5262084603309631, "learning_rate": 0.0002, "epoch": 0.07370184254606366, "step": 110}, {"loss": 1.8347, "grad_norm": 0.3774230182170868, "learning_rate": 0.0002, "epoch": 0.08040201005025126, "step": 120}, {"loss": 1.8386, "grad_norm": 0.34137430787086487, "learning_rate": 0.0002, "epoch": 0.08710217755443886, "step": 130}, {"loss": 1.861, "grad_norm": 0.407272070646286, "learning_rate": 0.0002, "epoch": 0.09380234505862646, "step": 140}, {"loss": 1.8279, "grad_norm": 0.4011937975883484, "learning_rate": 0.0002, "epoch": 0.10050251256281408, "step": 150}, {"loss": 1.9317, "grad_norm": 0.4432467222213745, "learning_rate": 0.0002, "epoch": 0.10720268006700168, "step": 160}, {"loss": 1.8157, "grad_norm": 0.44030463695526123, "learning_rate": 0.0002, "epoch": 0.11390284757118928, "step": 170}, {"loss": 1.8534, "grad_norm": 0.3799569308757782, "learning_rate": 0.0002, "epoch": 0.12060301507537688, "step": 180}, {"loss": 1.7658, "grad_norm": 0.33721521496772766, "learning_rate": 0.0002, "epoch": 0.1273031825795645, "step": 190}, {"loss": 1.8269, "grad_norm": 0.4096226692199707, "learning_rate": 0.0002, "epoch": 0.13400335008375208, "step": 200}, {"loss": 1.802, "grad_norm": 0.37374693155288696, "learning_rate": 0.0002, "epoch": 0.1407035175879397, "step": 210}, {"loss": 1.8901, "grad_norm": 0.3249480128288269, "learning_rate": 0.0002, "epoch": 0.1474036850921273, "step": 220}, {"loss": 1.8163, "grad_norm": 0.3612042963504791, "learning_rate": 0.0002, "epoch": 0.1541038525963149, "step": 230}, {"loss": 1.7585, "grad_norm": 0.3686671257019043, "learning_rate": 0.0002, "epoch": 0.16080402010050251, "step": 240}, {"loss": 1.8365, "grad_norm": 0.3521044850349426, "learning_rate": 0.0002, "epoch": 0.16750418760469013, "step": 250}, {"loss": 1.8623, "grad_norm": 0.4073677361011505, "learning_rate": 0.0002, "epoch": 0.17420435510887772, "step": 260}, {"loss": 1.8026, "grad_norm": 0.34522193670272827, "learning_rate": 0.0002, "epoch": 0.18090452261306533, "step": 270}, {"loss": 1.8162, "grad_norm": 0.4121900498867035, "learning_rate": 0.0002, "epoch": 0.18760469011725292, "step": 280}, {"loss": 1.7976, "grad_norm": 0.3544778525829315, "learning_rate": 0.0002, "epoch": 0.19430485762144054, "step": 290}, {"loss": 1.8787, "grad_norm": 0.3482133448123932, "learning_rate": 0.0002, "epoch": 0.20100502512562815, "step": 300}, {"loss": 1.8578, "grad_norm": 0.3421826660633087, "learning_rate": 0.0002, "epoch": 0.20770519262981574, "step": 310}, {"loss": 1.8013, "grad_norm": 0.5024696588516235, "learning_rate": 0.0002, "epoch": 0.21440536013400335, "step": 320}, {"loss": 1.8607, "grad_norm": 0.36013063788414, "learning_rate": 0.0002, "epoch": 0.22110552763819097, "step": 330}, {"loss": 1.9075, "grad_norm": 0.3611244857311249, "learning_rate": 0.0002, "epoch": 0.22780569514237856, "step": 340}, {"loss": 1.8128, "grad_norm": 0.39244529604911804, "learning_rate": 0.0002, "epoch": 0.23450586264656617, "step": 350}, {"loss": 1.7885, "grad_norm": 0.3299325704574585, "learning_rate": 0.0002, "epoch": 0.24120603015075376, "step": 360}, {"loss": 1.8028, "grad_norm": 0.3994322419166565, "learning_rate": 0.0002, "epoch": 0.24790619765494137, "step": 370}, {"loss": 1.8321, "grad_norm": 0.3559151887893677, "learning_rate": 0.0002, "epoch": 0.254606365159129, "step": 380}, {"loss": 1.7802, "grad_norm": 0.3873756229877472, "learning_rate": 0.0002, "epoch": 0.2613065326633166, "step": 390}, {"loss": 1.7844, "grad_norm": 0.3710744082927704, "learning_rate": 0.0002, "epoch": 0.26800670016750416, "step": 400}, {"loss": 1.7789, "grad_norm": 0.3618465065956116, "learning_rate": 0.0002, "epoch": 0.2747068676716918, "step": 410}, {"loss": 1.8529, "grad_norm": 0.30063769221305847, "learning_rate": 0.0002, "epoch": 0.2814070351758794, "step": 420}, {"loss": 1.7765, "grad_norm": 0.3695628345012665, "learning_rate": 0.0002, "epoch": 0.288107202680067, "step": 430}, {"loss": 1.7982, "grad_norm": 0.31451135873794556, "learning_rate": 0.0002, "epoch": 0.2948073701842546, "step": 440}, {"loss": 1.7517, "grad_norm": 0.3959707021713257, "learning_rate": 0.0002, "epoch": 0.3015075376884422, "step": 450}, {"loss": 1.8142, "grad_norm": 0.33475354313850403, "learning_rate": 0.0002, "epoch": 0.3082077051926298, "step": 460}, {"loss": 1.8805, "grad_norm": 0.33933115005493164, "learning_rate": 0.0002, "epoch": 0.3149078726968174, "step": 470}, {"loss": 1.7564, "grad_norm": 0.3264943063259125, "learning_rate": 0.0002, "epoch": 0.32160804020100503, "step": 480}, {"loss": 1.8428, "grad_norm": 0.40188100934028625, "learning_rate": 0.0002, "epoch": 0.32830820770519265, "step": 490}, {"loss": 1.7624, "grad_norm": 0.37408649921417236, "learning_rate": 0.0002, "epoch": 0.33500837520938026, "step": 500}, {"loss": 1.7745, "grad_norm": 0.33925938606262207, "learning_rate": 0.0002, "epoch": 0.3417085427135678, "step": 510}, {"loss": 1.814, "grad_norm": 0.36836713552474976, "learning_rate": 0.0002, "epoch": 0.34840871021775544, "step": 520}, {"loss": 1.8037, "grad_norm": 0.37284499406814575, "learning_rate": 0.0002, "epoch": 0.35510887772194305, "step": 530}, {"loss": 1.8379, "grad_norm": 0.3192278742790222, "learning_rate": 0.0002, "epoch": 0.36180904522613067, "step": 540}, {"loss": 1.8702, "grad_norm": 0.30233290791511536, "learning_rate": 0.0002, "epoch": 0.3685092127303183, "step": 550}, {"loss": 1.8799, "grad_norm": 0.3340817391872406, "learning_rate": 0.0002, "epoch": 0.37520938023450584, "step": 560}, {"loss": 1.8404, "grad_norm": 0.32600095868110657, "learning_rate": 0.0002, "epoch": 0.38190954773869346, "step": 570}, {"loss": 1.7804, "grad_norm": 0.33711278438568115, "learning_rate": 0.0002, "epoch": 0.38860971524288107, "step": 580}, {"loss": 1.8445, "grad_norm": 0.34890690445899963, "learning_rate": 0.0002, "epoch": 0.3953098827470687, "step": 590}, {"loss": 1.8187, "grad_norm": 0.38238924741744995, "learning_rate": 0.0002, "epoch": 0.4020100502512563, "step": 600}, {"loss": 1.8111, "grad_norm": 0.34399354457855225, "learning_rate": 0.0002, "epoch": 0.40871021775544386, "step": 610}, {"loss": 1.8006, "grad_norm": 0.3346073627471924, "learning_rate": 0.0002, "epoch": 0.4154103852596315, "step": 620}, {"loss": 1.7705, "grad_norm": 0.3545648157596588, "learning_rate": 0.0002, "epoch": 0.4221105527638191, "step": 630}, {"loss": 1.8445, "grad_norm": 0.3378899097442627, "learning_rate": 0.0002, "epoch": 0.4288107202680067, "step": 640}, {"loss": 1.804, "grad_norm": 0.3255569040775299, "learning_rate": 0.0002, "epoch": 0.4355108877721943, "step": 650}, {"loss": 1.7679, "grad_norm": 0.34880587458610535, "learning_rate": 0.0002, "epoch": 0.44221105527638194, "step": 660}, {"loss": 1.7861, "grad_norm": 0.3402383625507355, "learning_rate": 0.0002, "epoch": 0.4489112227805695, "step": 670}, {"loss": 1.8131, "grad_norm": 0.3594033718109131, "learning_rate": 0.0002, "epoch": 0.4556113902847571, "step": 680}, {"loss": 1.8399, "grad_norm": 0.31000566482543945, "learning_rate": 0.0002, "epoch": 0.4623115577889447, "step": 690}, {"loss": 1.7521, "grad_norm": 0.37229061126708984, "learning_rate": 0.0002, "epoch": 0.46901172529313234, "step": 700}, {"loss": 1.7779, "grad_norm": 0.315801739692688, "learning_rate": 0.0002, "epoch": 0.47571189279731996, "step": 710}, {"loss": 1.7515, "grad_norm": 0.3220832645893097, "learning_rate": 0.0002, "epoch": 0.4824120603015075, "step": 720}, {"loss": 1.7181, "grad_norm": 0.3435456156730652, "learning_rate": 0.0002, "epoch": 0.48911222780569513, "step": 730}, {"loss": 1.8844, "grad_norm": 0.30380892753601074, "learning_rate": 0.0002, "epoch": 0.49581239530988275, "step": 740}, {"loss": 1.7792, "grad_norm": 0.3555026054382324, "learning_rate": 0.0002, "epoch": 0.5025125628140703, "step": 750}, {"loss": 1.7714, "grad_norm": 0.3019855320453644, "learning_rate": 0.0002, "epoch": 0.509212730318258, "step": 760}, {"loss": 1.7962, "grad_norm": 0.309111088514328, "learning_rate": 0.0002, "epoch": 0.5159128978224455, "step": 770}, {"loss": 1.7913, "grad_norm": 0.366020530462265, "learning_rate": 0.0002, "epoch": 0.5226130653266332, "step": 780}, {"loss": 1.8008, "grad_norm": 0.3267050087451935, "learning_rate": 0.0002, "epoch": 0.5293132328308208, "step": 790}, {"loss": 1.7397, "grad_norm": 0.34265750646591187, "learning_rate": 0.0002, "epoch": 0.5360134003350083, "step": 800}, {"loss": 1.8251, "grad_norm": 0.313669890165329, "learning_rate": 0.0002, "epoch": 0.542713567839196, "step": 810}, {"loss": 1.8369, "grad_norm": 0.3355236053466797, "learning_rate": 0.0002, "epoch": 0.5494137353433836, "step": 820}, {"loss": 1.7381, "grad_norm": 0.3186608552932739, "learning_rate": 0.0002, "epoch": 0.5561139028475712, "step": 830}, {"loss": 1.8034, "grad_norm": 0.30357518792152405, "learning_rate": 0.0002, "epoch": 0.5628140703517588, "step": 840}, {"loss": 1.769, "grad_norm": 0.3990040123462677, "learning_rate": 0.0002, "epoch": 0.5695142378559463, "step": 850}, {"loss": 1.7482, "grad_norm": 0.34363803267478943, "learning_rate": 0.0002, "epoch": 0.576214405360134, "step": 860}, {"loss": 1.8106, "grad_norm": 0.3757908046245575, "learning_rate": 0.0002, "epoch": 0.5829145728643216, "step": 870}, {"loss": 1.8104, "grad_norm": 0.3359757661819458, "learning_rate": 0.0002, "epoch": 0.5896147403685092, "step": 880}, {"loss": 1.7591, "grad_norm": 0.5555329918861389, "learning_rate": 0.0002, "epoch": 0.5963149078726968, "step": 890}, {"loss": 1.7715, "grad_norm": 0.4046323895454407, "learning_rate": 0.0002, "epoch": 0.6030150753768844, "step": 900}, {"loss": 1.7998, "grad_norm": 0.29834219813346863, "learning_rate": 0.0002, "epoch": 0.609715242881072, "step": 910}, {"loss": 1.7826, "grad_norm": 0.3241238594055176, "learning_rate": 0.0002, "epoch": 0.6164154103852596, "step": 920}, {"loss": 1.8342, "grad_norm": 0.35154739022254944, "learning_rate": 0.0002, "epoch": 0.6231155778894473, "step": 930}, {"loss": 1.8076, "grad_norm": 0.3287706673145294, "learning_rate": 0.0002, "epoch": 0.6298157453936348, "step": 940}, {"loss": 1.8038, "grad_norm": 0.35670626163482666, "learning_rate": 0.0002, "epoch": 0.6365159128978225, "step": 950}, {"loss": 1.869, "grad_norm": 0.6114104986190796, "learning_rate": 0.0002, "epoch": 0.6432160804020101, "step": 960}, {"loss": 1.8297, "grad_norm": 0.3186565041542053, "learning_rate": 0.0002, "epoch": 0.6499162479061976, "step": 970}, {"loss": 1.7539, "grad_norm": 0.27164125442504883, "learning_rate": 0.0002, "epoch": 0.6566164154103853, "step": 980}, {"loss": 1.8339, "grad_norm": 0.34407344460487366, "learning_rate": 0.0002, "epoch": 0.6633165829145728, "step": 990}, {"loss": 1.855, "grad_norm": 0.368415892124176, "learning_rate": 0.0002, "epoch": 0.6700167504187605, "step": 1000}, {"loss": 1.7821, "grad_norm": 0.3306390643119812, "learning_rate": 0.0002, "epoch": 0.6767169179229481, "step": 1010}, {"loss": 1.786, "grad_norm": 0.3198648989200592, "learning_rate": 0.0002, "epoch": 0.6834170854271356, "step": 1020}, {"loss": 1.816, "grad_norm": 0.3092987537384033, "learning_rate": 0.0002, "epoch": 0.6901172529313233, "step": 1030}, {"loss": 1.7689, "grad_norm": 0.3090653419494629, "learning_rate": 0.0002, "epoch": 0.6968174204355109, "step": 1040}, {"loss": 1.7544, "grad_norm": 0.3485880196094513, "learning_rate": 0.0002, "epoch": 0.7035175879396985, "step": 1050}, {"loss": 1.8187, "grad_norm": 0.35782721638679504, "learning_rate": 0.0002, "epoch": 0.7102177554438861, "step": 1060}, {"loss": 1.806, "grad_norm": 0.34256869554519653, "learning_rate": 0.0002, "epoch": 0.7169179229480737, "step": 1070}, {"loss": 1.7873, "grad_norm": 0.30461037158966064, "learning_rate": 0.0002, "epoch": 0.7236180904522613, "step": 1080}, {"loss": 1.7367, "grad_norm": 0.3398691713809967, "learning_rate": 0.0002, "epoch": 0.7303182579564489, "step": 1090}, {"loss": 1.8756, "grad_norm": 0.3180808126926422, "learning_rate": 0.0002, "epoch": 0.7370184254606366, "step": 1100}, {"loss": 1.6988, "grad_norm": 0.34400665760040283, "learning_rate": 0.0002, "epoch": 0.7437185929648241, "step": 1110}, {"loss": 1.7851, "grad_norm": 0.34244877099990845, "learning_rate": 0.0002, "epoch": 0.7504187604690117, "step": 1120}, {"loss": 1.7841, "grad_norm": 0.29946693778038025, "learning_rate": 0.0002, "epoch": 0.7571189279731994, "step": 1130}, {"loss": 1.7456, "grad_norm": 0.37547236680984497, "learning_rate": 0.0002, "epoch": 0.7638190954773869, "step": 1140}, {"loss": 1.8425, "grad_norm": 0.3263005018234253, "learning_rate": 0.0002, "epoch": 0.7705192629815746, "step": 1150}, {"loss": 1.7222, "grad_norm": 0.41363608837127686, "learning_rate": 0.0002, "epoch": 0.7772194304857621, "step": 1160}, {"loss": 1.7836, "grad_norm": 0.36267954111099243, "learning_rate": 0.0002, "epoch": 0.7839195979899497, "step": 1170}, {"loss": 1.9183, "grad_norm": 0.31789499521255493, "learning_rate": 0.0002, "epoch": 0.7906197654941374, "step": 1180}, {"loss": 1.78, "grad_norm": 0.5708149075508118, "learning_rate": 0.0002, "epoch": 0.7973199329983249, "step": 1190}, {"loss": 1.6908, "grad_norm": 0.322099506855011, "learning_rate": 0.0002, "epoch": 0.8040201005025126, "step": 1200}, {"loss": 1.7639, "grad_norm": 0.3419909179210663, "learning_rate": 0.0002, "epoch": 0.8107202680067002, "step": 1210}, {"loss": 1.7428, "grad_norm": 0.36286255717277527, "learning_rate": 0.0002, "epoch": 0.8174204355108877, "step": 1220}, {"loss": 1.8409, "grad_norm": 0.33992862701416016, "learning_rate": 0.0002, "epoch": 0.8241206030150754, "step": 1230}, {"loss": 1.7507, "grad_norm": 0.32622793316841125, "learning_rate": 0.0002, "epoch": 0.830820770519263, "step": 1240}, {"loss": 1.8098, "grad_norm": 0.3036167621612549, "learning_rate": 0.0002, "epoch": 0.8375209380234506, "step": 1250}, {"loss": 1.8094, "grad_norm": 0.3182215392589569, "learning_rate": 0.0002, "epoch": 0.8442211055276382, "step": 1260}, {"loss": 1.8017, "grad_norm": 0.3270018696784973, "learning_rate": 0.0002, "epoch": 0.8509212730318257, "step": 1270}, {"loss": 1.798, "grad_norm": 0.32652342319488525, "learning_rate": 0.0002, "epoch": 0.8576214405360134, "step": 1280}, {"loss": 1.7448, "grad_norm": 0.3631329834461212, "learning_rate": 0.0002, "epoch": 0.864321608040201, "step": 1290}, {"loss": 1.7, "grad_norm": 0.36706018447875977, "learning_rate": 0.0002, "epoch": 0.8710217755443886, "step": 1300}, {"loss": 1.8178, "grad_norm": 0.3347418010234833, "learning_rate": 0.0002, "epoch": 0.8777219430485762, "step": 1310}, {"loss": 1.7824, "grad_norm": 0.34371060132980347, "learning_rate": 0.0002, "epoch": 0.8844221105527639, "step": 1320}, {"loss": 1.783, "grad_norm": 0.3029090166091919, "learning_rate": 0.0002, "epoch": 0.8911222780569514, "step": 1330}, {"loss": 1.8017, "grad_norm": 0.34700682759284973, "learning_rate": 0.0002, "epoch": 0.897822445561139, "step": 1340}, {"loss": 1.7998, "grad_norm": 0.35574328899383545, "learning_rate": 0.0002, "epoch": 0.9045226130653267, "step": 1350}, {"loss": 1.7339, "grad_norm": 0.30928221344947815, "learning_rate": 0.0002, "epoch": 0.9112227805695142, "step": 1360}, {"loss": 1.7479, "grad_norm": 0.30652928352355957, "learning_rate": 0.0002, "epoch": 0.9179229480737019, "step": 1370}, {"loss": 1.7491, "grad_norm": 0.3838157653808594, "learning_rate": 0.0002, "epoch": 0.9246231155778895, "step": 1380}, {"loss": 1.7977, "grad_norm": 0.31655240058898926, "learning_rate": 0.0002, "epoch": 0.931323283082077, "step": 1390}, {"loss": 1.8175, "grad_norm": 0.41737303137779236, "learning_rate": 0.0002, "epoch": 0.9380234505862647, "step": 1400}, {"loss": 1.6811, "grad_norm": 0.3227267861366272, "learning_rate": 0.0002, "epoch": 0.9447236180904522, "step": 1410}, {"loss": 1.7343, "grad_norm": 0.3729925751686096, "learning_rate": 0.0002, "epoch": 0.9514237855946399, "step": 1420}, {"loss": 1.8221, "grad_norm": 0.30779409408569336, "learning_rate": 0.0002, "epoch": 0.9581239530988275, "step": 1430}, {"loss": 1.7972, "grad_norm": 0.334379643201828, "learning_rate": 0.0002, "epoch": 0.964824120603015, "step": 1440}, {"loss": 1.7141, "grad_norm": 0.3568236231803894, "learning_rate": 0.0002, "epoch": 0.9715242881072027, "step": 1450}, {"loss": 1.7541, "grad_norm": 0.33310577273368835, "learning_rate": 0.0002, "epoch": 0.9782244556113903, "step": 1460}, {"loss": 1.8511, "grad_norm": 0.2972261905670166, "learning_rate": 0.0002, "epoch": 0.9849246231155779, "step": 1470}, {"loss": 1.7654, "grad_norm": 0.3322717845439911, "learning_rate": 0.0002, "epoch": 0.9916247906197655, "step": 1480}, {"loss": 1.8033, "grad_norm": 0.3276330828666687, "learning_rate": 0.0002, "epoch": 0.998324958123953, "step": 1490}, {"eval_loss": 1.8036354780197144, "eval_runtime": 37.8949, "eval_samples_per_second": 13.59, "eval_steps_per_second": 1.715, "epoch": 0.9996649916247906, "step": 1492}, {"loss": 1.7138, "grad_norm": 0.29252371191978455, "learning_rate": 0.0002, "epoch": 1.0050251256281406, "step": 1500}, {"loss": 1.8198, "grad_norm": 0.31607162952423096, "learning_rate": 0.0002, "epoch": 1.0117252931323284, "step": 1510}, {"loss": 1.6779, "grad_norm": 0.32294467091560364, "learning_rate": 0.0002, "epoch": 1.018425460636516, "step": 1520}, {"loss": 1.7919, "grad_norm": 0.3868017792701721, "learning_rate": 0.0002, "epoch": 1.0251256281407035, "step": 1530}, {"loss": 1.7954, "grad_norm": 0.3178282082080841, "learning_rate": 0.0002, "epoch": 1.031825795644891, "step": 1540}, {"loss": 1.7136, "grad_norm": 0.3706750273704529, "learning_rate": 0.0002, "epoch": 1.0385259631490786, "step": 1550}, {"loss": 1.7382, "grad_norm": 0.33930912613868713, "learning_rate": 0.0002, "epoch": 1.0452261306532664, "step": 1560}, {"loss": 1.7602, "grad_norm": 0.33970504999160767, "learning_rate": 0.0002, "epoch": 1.051926298157454, "step": 1570}, {"loss": 1.6573, "grad_norm": 0.42553383111953735, "learning_rate": 0.0002, "epoch": 1.0586264656616415, "step": 1580}, {"loss": 1.645, "grad_norm": 0.3772421181201935, "learning_rate": 0.0002, "epoch": 1.065326633165829, "step": 1590}, {"loss": 1.7362, "grad_norm": 0.34212902188301086, "learning_rate": 0.0002, "epoch": 1.0720268006700167, "step": 1600}, {"loss": 1.7057, "grad_norm": 0.3798283338546753, "learning_rate": 0.0002, "epoch": 1.0787269681742044, "step": 1610}, {"loss": 1.7468, "grad_norm": 0.36909598112106323, "learning_rate": 0.0002, "epoch": 1.085427135678392, "step": 1620}, {"loss": 1.7807, "grad_norm": 0.3344230651855469, "learning_rate": 0.0002, "epoch": 1.0921273031825796, "step": 1630}, {"loss": 1.7111, "grad_norm": 0.3862569332122803, "learning_rate": 0.0002, "epoch": 1.0988274706867671, "step": 1640}, {"loss": 1.7163, "grad_norm": 0.31188511848449707, "learning_rate": 0.0002, "epoch": 1.1055276381909547, "step": 1650}, {"loss": 1.7263, "grad_norm": 0.3563670814037323, "learning_rate": 0.0002, "epoch": 1.1122278056951425, "step": 1660}, {"loss": 1.7718, "grad_norm": 0.35052165389060974, "learning_rate": 0.0002, "epoch": 1.11892797319933, "step": 1670}, {"loss": 1.7601, "grad_norm": 0.3285699188709259, "learning_rate": 0.0002, "epoch": 1.1256281407035176, "step": 1680}, {"loss": 1.6877, "grad_norm": 0.3639393746852875, "learning_rate": 0.0002, "epoch": 1.1323283082077051, "step": 1690}, {"loss": 1.7719, "grad_norm": 0.3842753767967224, "learning_rate": 0.0002, "epoch": 1.1390284757118927, "step": 1700}, {"loss": 1.7002, "grad_norm": 0.3624933063983917, "learning_rate": 0.0002, "epoch": 1.1457286432160805, "step": 1710}, {"loss": 1.7243, "grad_norm": 0.3641220033168793, "learning_rate": 0.0002, "epoch": 1.152428810720268, "step": 1720}, {"loss": 1.752, "grad_norm": 0.32765355706214905, "learning_rate": 0.0002, "epoch": 1.1591289782244556, "step": 1730}, {"loss": 1.6556, "grad_norm": 0.34974896907806396, "learning_rate": 0.0002, "epoch": 1.1658291457286432, "step": 1740}, {"loss": 1.7273, "grad_norm": 0.3910926580429077, "learning_rate": 0.0002, "epoch": 1.1725293132328307, "step": 1750}, {"loss": 1.7204, "grad_norm": 0.3564300537109375, "learning_rate": 0.0002, "epoch": 1.1792294807370185, "step": 1760}, {"loss": 1.746, "grad_norm": 0.34822574257850647, "learning_rate": 0.0002, "epoch": 1.185929648241206, "step": 1770}, {"loss": 1.7256, "grad_norm": 0.36185044050216675, "learning_rate": 0.0002, "epoch": 1.1926298157453936, "step": 1780}, {"loss": 1.6431, "grad_norm": 0.34866711497306824, "learning_rate": 0.0002, "epoch": 1.1993299832495812, "step": 1790}, {"loss": 1.8084, "grad_norm": 0.4017769992351532, "learning_rate": 0.0002, "epoch": 1.2060301507537687, "step": 1800}, {"loss": 1.6985, "grad_norm": 0.32930681109428406, "learning_rate": 0.0002, "epoch": 1.2127303182579565, "step": 1810}, {"loss": 1.7606, "grad_norm": 0.35951921343803406, "learning_rate": 0.0002, "epoch": 1.219430485762144, "step": 1820}, {"loss": 1.6933, "grad_norm": 0.37366992235183716, "learning_rate": 0.0002, "epoch": 1.2261306532663316, "step": 1830}, {"loss": 1.6737, "grad_norm": 0.3565689027309418, "learning_rate": 0.0002, "epoch": 1.2328308207705192, "step": 1840}, {"loss": 1.8013, "grad_norm": 0.3692343533039093, "learning_rate": 0.0002, "epoch": 1.2395309882747068, "step": 1850}, {"loss": 1.736, "grad_norm": 0.38426971435546875, "learning_rate": 0.0002, "epoch": 1.2462311557788945, "step": 1860}, {"loss": 1.7031, "grad_norm": 0.33559855818748474, "learning_rate": 0.0002, "epoch": 1.252931323283082, "step": 1870}, {"loss": 1.7033, "grad_norm": 0.34181106090545654, "learning_rate": 0.0002, "epoch": 1.2596314907872697, "step": 1880}, {"loss": 1.7707, "grad_norm": 0.3916318416595459, "learning_rate": 0.0002, "epoch": 1.2663316582914572, "step": 1890}, {"loss": 1.6686, "grad_norm": 0.3887825012207031, "learning_rate": 0.0002, "epoch": 1.2730318257956448, "step": 1900}, {"loss": 1.7062, "grad_norm": 0.33583927154541016, "learning_rate": 0.0002, "epoch": 1.2797319932998326, "step": 1910}, {"loss": 1.717, "grad_norm": 0.37639349699020386, "learning_rate": 0.0002, "epoch": 1.2864321608040201, "step": 1920}, {"loss": 1.777, "grad_norm": 0.38059428334236145, "learning_rate": 0.0002, "epoch": 1.2931323283082077, "step": 1930}, {"loss": 1.6126, "grad_norm": 0.37253183126449585, "learning_rate": 0.0002, "epoch": 1.2998324958123952, "step": 1940}, {"loss": 1.6758, "grad_norm": 0.37371566891670227, "learning_rate": 0.0002, "epoch": 1.3065326633165828, "step": 1950}, {"loss": 1.6788, "grad_norm": 0.4080910086631775, "learning_rate": 0.0002, "epoch": 1.3132328308207706, "step": 1960}, {"loss": 1.6518, "grad_norm": 0.3174354135990143, "learning_rate": 0.0002, "epoch": 1.3199329983249581, "step": 1970}, {"loss": 1.7925, "grad_norm": 0.4518888294696808, "learning_rate": 0.0002, "epoch": 1.3266331658291457, "step": 1980}, {"loss": 1.7085, "grad_norm": 0.3627921938896179, "learning_rate": 0.0002, "epoch": 1.3333333333333333, "step": 1990}, {"loss": 1.7676, "grad_norm": 0.3655930161476135, "learning_rate": 0.0002, "epoch": 1.3400335008375208, "step": 2000}, {"loss": 1.7016, "grad_norm": 0.3509993255138397, "learning_rate": 0.0002, "epoch": 1.3467336683417086, "step": 2010}, {"loss": 1.7359, "grad_norm": 0.4281129240989685, "learning_rate": 0.0002, "epoch": 1.3534338358458962, "step": 2020}, {"loss": 1.6884, "grad_norm": 0.3821414113044739, "learning_rate": 0.0002, "epoch": 1.3601340033500837, "step": 2030}, {"loss": 1.7075, "grad_norm": 0.3907586336135864, "learning_rate": 0.0002, "epoch": 1.3668341708542713, "step": 2040}, {"loss": 1.7424, "grad_norm": 0.37792932987213135, "learning_rate": 0.0002, "epoch": 1.3735343383584588, "step": 2050}, {"loss": 1.7305, "grad_norm": 0.3693985641002655, "learning_rate": 0.0002, "epoch": 1.3802345058626466, "step": 2060}, {"loss": 1.7434, "grad_norm": 0.32275936007499695, "learning_rate": 0.0002, "epoch": 1.3869346733668342, "step": 2070}, {"loss": 1.6677, "grad_norm": 0.3789440095424652, "learning_rate": 0.0002, "epoch": 1.3936348408710217, "step": 2080}, {"loss": 1.6825, "grad_norm": 0.3638380467891693, "learning_rate": 0.0002, "epoch": 1.4003350083752093, "step": 2090}, {"loss": 1.6542, "grad_norm": 0.3495481610298157, "learning_rate": 0.0002, "epoch": 1.4070351758793969, "step": 2100}, {"loss": 1.7225, "grad_norm": 0.37920597195625305, "learning_rate": 0.0002, "epoch": 1.4137353433835846, "step": 2110}, {"loss": 1.7329, "grad_norm": 0.37218064069747925, "learning_rate": 0.0002, "epoch": 1.4204355108877722, "step": 2120}, {"loss": 1.799, "grad_norm": 0.38074082136154175, "learning_rate": 0.0002, "epoch": 1.4271356783919598, "step": 2130}, {"loss": 1.7403, "grad_norm": 0.3455527126789093, "learning_rate": 0.0002, "epoch": 1.4338358458961473, "step": 2140}, {"loss": 1.776, "grad_norm": 0.3712003529071808, "learning_rate": 0.0002, "epoch": 1.4405360134003349, "step": 2150}, {"loss": 1.7619, "grad_norm": 0.3786754906177521, "learning_rate": 0.0002, "epoch": 1.4472361809045227, "step": 2160}, {"loss": 1.68, "grad_norm": 0.3879223167896271, "learning_rate": 0.0002, "epoch": 1.4539363484087102, "step": 2170}, {"loss": 1.7, "grad_norm": 0.38738805055618286, "learning_rate": 0.0002, "epoch": 1.4606365159128978, "step": 2180}, {"loss": 1.7581, "grad_norm": 0.39768800139427185, "learning_rate": 0.0002, "epoch": 1.4673366834170856, "step": 2190}, {"loss": 1.7671, "grad_norm": 0.4172441065311432, "learning_rate": 0.0002, "epoch": 1.474036850921273, "step": 2200}, {"loss": 1.6736, "grad_norm": 0.4043174982070923, "learning_rate": 0.0002, "epoch": 1.4807370184254607, "step": 2210}, {"loss": 1.7444, "grad_norm": 0.3750883936882019, "learning_rate": 0.0002, "epoch": 1.4874371859296482, "step": 2220}, {"loss": 1.6861, "grad_norm": 0.3552253246307373, "learning_rate": 0.0002, "epoch": 1.4941373534338358, "step": 2230}, {"loss": 1.6471, "grad_norm": 0.34607139229774475, "learning_rate": 0.0002, "epoch": 1.5008375209380236, "step": 2240}, {"loss": 1.6962, "grad_norm": 0.3406706750392914, "learning_rate": 0.0002, "epoch": 1.507537688442211, "step": 2250}, {"loss": 1.7694, "grad_norm": 0.36654895544052124, "learning_rate": 0.0002, "epoch": 1.5142378559463987, "step": 2260}, {"loss": 1.6812, "grad_norm": 0.3914054334163666, "learning_rate": 0.0002, "epoch": 1.5209380234505863, "step": 2270}, {"loss": 1.6822, "grad_norm": 0.42012137174606323, "learning_rate": 0.0002, "epoch": 1.5276381909547738, "step": 2280}, {"loss": 1.697, "grad_norm": 0.39563435316085815, "learning_rate": 0.0002, "epoch": 1.5343383584589616, "step": 2290}, {"loss": 1.7491, "grad_norm": 0.3508438766002655, "learning_rate": 0.0002, "epoch": 1.541038525963149, "step": 2300}, {"loss": 1.7727, "grad_norm": 0.3785218596458435, "learning_rate": 0.0002, "epoch": 1.5477386934673367, "step": 2310}, {"loss": 1.6963, "grad_norm": 0.39377647638320923, "learning_rate": 0.0002, "epoch": 1.5544388609715243, "step": 2320}, {"loss": 1.7263, "grad_norm": 0.3391438126564026, "learning_rate": 0.0002, "epoch": 1.5611390284757118, "step": 2330}, {"loss": 1.7722, "grad_norm": 0.37944263219833374, "learning_rate": 0.0002, "epoch": 1.5678391959798996, "step": 2340}, {"loss": 1.6371, "grad_norm": 0.3523491322994232, "learning_rate": 0.0002, "epoch": 1.574539363484087, "step": 2350}, {"loss": 1.7583, "grad_norm": 0.3911575973033905, "learning_rate": 0.0002, "epoch": 1.5812395309882747, "step": 2360}, {"loss": 1.7117, "grad_norm": 0.33832186460494995, "learning_rate": 0.0002, "epoch": 1.5879396984924623, "step": 2370}, {"loss": 1.7701, "grad_norm": 0.3665979206562042, "learning_rate": 0.0002, "epoch": 1.5946398659966499, "step": 2380}, {"loss": 1.779, "grad_norm": 0.3871748149394989, "learning_rate": 0.0002, "epoch": 1.6013400335008376, "step": 2390}, {"loss": 1.7109, "grad_norm": 0.3586967885494232, "learning_rate": 0.0002, "epoch": 1.608040201005025, "step": 2400}, {"loss": 1.7096, "grad_norm": 0.3563673198223114, "learning_rate": 0.0002, "epoch": 1.6147403685092128, "step": 2410}, {"loss": 1.745, "grad_norm": 0.37588971853256226, "learning_rate": 0.0002, "epoch": 1.6214405360134003, "step": 2420}, {"loss": 1.7086, "grad_norm": 0.352556437253952, "learning_rate": 0.0002, "epoch": 1.6281407035175879, "step": 2430}, {"loss": 1.6547, "grad_norm": 0.3716259300708771, "learning_rate": 0.0002, "epoch": 1.6348408710217757, "step": 2440}, {"loss": 1.7033, "grad_norm": 0.372001975774765, "learning_rate": 0.0002, "epoch": 1.641541038525963, "step": 2450}, {"loss": 1.6584, "grad_norm": 0.3430042862892151, "learning_rate": 0.0002, "epoch": 1.6482412060301508, "step": 2460}, {"loss": 1.7217, "grad_norm": 0.3741483688354492, "learning_rate": 0.0002, "epoch": 1.6549413735343383, "step": 2470}, {"loss": 1.7701, "grad_norm": 0.3610571324825287, "learning_rate": 0.0002, "epoch": 1.661641541038526, "step": 2480}, {"loss": 1.7057, "grad_norm": 0.4204719066619873, "learning_rate": 0.0002, "epoch": 1.6683417085427137, "step": 2490}, {"loss": 1.7954, "grad_norm": 0.3938186466693878, "learning_rate": 0.0002, "epoch": 1.675041876046901, "step": 2500}, {"loss": 1.6633, "grad_norm": 0.3421435058116913, "learning_rate": 0.0002, "epoch": 1.6817420435510888, "step": 2510}, {"loss": 1.7996, "grad_norm": 0.42441412806510925, "learning_rate": 0.0002, "epoch": 1.6884422110552764, "step": 2520}, {"loss": 1.7142, "grad_norm": 0.38071519136428833, "learning_rate": 0.0002, "epoch": 1.695142378559464, "step": 2530}, {"loss": 1.7232, "grad_norm": 0.34078919887542725, "learning_rate": 0.0002, "epoch": 1.7018425460636517, "step": 2540}, {"loss": 1.7126, "grad_norm": 0.412844181060791, "learning_rate": 0.0002, "epoch": 1.708542713567839, "step": 2550}, {"loss": 1.7149, "grad_norm": 0.3753604292869568, "learning_rate": 0.0002, "epoch": 1.7152428810720268, "step": 2560}, {"loss": 1.7011, "grad_norm": 0.41588476300239563, "learning_rate": 0.0002, "epoch": 1.7219430485762144, "step": 2570}, {"loss": 1.6427, "grad_norm": 0.35504111647605896, "learning_rate": 0.0002, "epoch": 1.728643216080402, "step": 2580}, {"loss": 1.7296, "grad_norm": 0.36909720301628113, "learning_rate": 0.0002, "epoch": 1.7353433835845897, "step": 2590}, {"loss": 1.7022, "grad_norm": 0.4149979054927826, "learning_rate": 0.0002, "epoch": 1.742043551088777, "step": 2600}, {"loss": 1.77, "grad_norm": 0.38859328627586365, "learning_rate": 0.0002, "epoch": 1.7487437185929648, "step": 2610}, {"loss": 1.7036, "grad_norm": 0.36738792061805725, "learning_rate": 0.0002, "epoch": 1.7554438860971524, "step": 2620}, {"loss": 1.764, "grad_norm": 0.3968178927898407, "learning_rate": 0.0002, "epoch": 1.76214405360134, "step": 2630}, {"loss": 1.7687, "grad_norm": 0.3972901999950409, "learning_rate": 0.0002, "epoch": 1.7688442211055277, "step": 2640}, {"loss": 1.6774, "grad_norm": 0.3949959874153137, "learning_rate": 0.0002, "epoch": 1.775544388609715, "step": 2650}, {"loss": 1.7247, "grad_norm": 0.44074657559394836, "learning_rate": 0.0002, "epoch": 1.7822445561139029, "step": 2660}, {"loss": 1.7188, "grad_norm": 0.39743664860725403, "learning_rate": 0.0002, "epoch": 1.7889447236180904, "step": 2670}, {"loss": 1.7258, "grad_norm": 0.3950406610965729, "learning_rate": 0.0002, "epoch": 1.795644891122278, "step": 2680}, {"loss": 1.6906, "grad_norm": 0.3568263649940491, "learning_rate": 0.0002, "epoch": 1.8023450586264658, "step": 2690}, {"loss": 1.6735, "grad_norm": 0.3819476366043091, "learning_rate": 0.0002, "epoch": 1.809045226130653, "step": 2700}, {"loss": 1.7198, "grad_norm": 0.3480634391307831, "learning_rate": 0.0002, "epoch": 1.8157453936348409, "step": 2710}, {"loss": 1.7042, "grad_norm": 0.3875853419303894, "learning_rate": 0.0002, "epoch": 1.8224455611390284, "step": 2720}, {"loss": 1.6988, "grad_norm": 0.3441337049007416, "learning_rate": 0.0002, "epoch": 1.829145728643216, "step": 2730}, {"loss": 1.7647, "grad_norm": 0.35692882537841797, "learning_rate": 0.0002, "epoch": 1.8358458961474038, "step": 2740}, {"loss": 1.7033, "grad_norm": 0.36959215998649597, "learning_rate": 0.0002, "epoch": 1.8425460636515911, "step": 2750}, {"loss": 1.7657, "grad_norm": 0.3893393278121948, "learning_rate": 0.0002, "epoch": 1.849246231155779, "step": 2760}, {"loss": 1.7068, "grad_norm": 0.37817293405532837, "learning_rate": 0.0002, "epoch": 1.8559463986599665, "step": 2770}, {"loss": 1.761, "grad_norm": 0.36071285605430603, "learning_rate": 0.0002, "epoch": 1.862646566164154, "step": 2780}, {"loss": 1.7623, "grad_norm": 0.3758420944213867, "learning_rate": 0.0002, "epoch": 1.8693467336683418, "step": 2790}, {"loss": 1.6743, "grad_norm": 0.3889938294887543, "learning_rate": 0.0002, "epoch": 1.8760469011725294, "step": 2800}, {"loss": 1.6151, "grad_norm": 0.34361857175827026, "learning_rate": 0.0002, "epoch": 1.882747068676717, "step": 2810}, {"loss": 1.6038, "grad_norm": 0.39283323287963867, "learning_rate": 0.0002, "epoch": 1.8894472361809045, "step": 2820}, {"loss": 1.7555, "grad_norm": 0.3919452726840973, "learning_rate": 0.0002, "epoch": 1.896147403685092, "step": 2830}, {"loss": 1.673, "grad_norm": 0.38215070962905884, "learning_rate": 0.0002, "epoch": 1.9028475711892798, "step": 2840}, {"loss": 1.7044, "grad_norm": 0.4235064387321472, "learning_rate": 0.0002, "epoch": 1.9095477386934674, "step": 2850}, {"loss": 1.7123, "grad_norm": 0.35694634914398193, "learning_rate": 0.0002, "epoch": 1.916247906197655, "step": 2860}, {"loss": 1.8128, "grad_norm": 0.383492112159729, "learning_rate": 0.0002, "epoch": 1.9229480737018425, "step": 2870}, {"loss": 1.7581, "grad_norm": 0.5945147275924683, "learning_rate": 0.0002, "epoch": 1.92964824120603, "step": 2880}, {"loss": 1.7421, "grad_norm": 0.3367522358894348, "learning_rate": 0.0002, "epoch": 1.9363484087102178, "step": 2890}, {"loss": 1.6561, "grad_norm": 0.35300394892692566, "learning_rate": 0.0002, "epoch": 1.9430485762144054, "step": 2900}, {"loss": 1.7033, "grad_norm": 0.38084495067596436, "learning_rate": 0.0002, "epoch": 1.949748743718593, "step": 2910}, {"loss": 1.7132, "grad_norm": 0.37559160590171814, "learning_rate": 0.0002, "epoch": 1.9564489112227805, "step": 2920}, {"loss": 1.6759, "grad_norm": 0.3661738336086273, "learning_rate": 0.0002, "epoch": 1.963149078726968, "step": 2930}, {"loss": 1.7643, "grad_norm": 0.4073849320411682, "learning_rate": 0.0002, "epoch": 1.9698492462311559, "step": 2940}, {"loss": 1.6806, "grad_norm": 0.3723304271697998, "learning_rate": 0.0002, "epoch": 1.9765494137353434, "step": 2950}, {"loss": 1.7611, "grad_norm": 0.3991098999977112, "learning_rate": 0.0002, "epoch": 1.983249581239531, "step": 2960}, {"loss": 1.7263, "grad_norm": 0.3947085440158844, "learning_rate": 0.0002, "epoch": 1.9899497487437185, "step": 2970}, {"loss": 1.7217, "grad_norm": 0.3786258399486542, "learning_rate": 0.0002, "epoch": 1.996649916247906, "step": 2980}, {"eval_loss": 1.8028968572616577, "eval_runtime": 37.8985, "eval_samples_per_second": 13.589, "eval_steps_per_second": 1.715, "epoch": 2.0, "step": 2985}, {"loss": 1.695, "grad_norm": 0.34824079275131226, "learning_rate": 0.0002, "epoch": 2.003350083752094, "step": 2990}, {"loss": 1.5853, "grad_norm": 0.3394894003868103, "learning_rate": 0.0002, "epoch": 2.0100502512562812, "step": 3000}, {"loss": 1.5783, "grad_norm": 0.36910977959632874, "learning_rate": 0.0002, "epoch": 2.016750418760469, "step": 3010}, {"loss": 1.6105, "grad_norm": 0.45000967383384705, "learning_rate": 0.0002, "epoch": 2.023450586264657, "step": 3020}, {"loss": 1.6019, "grad_norm": 0.3791407346725464, "learning_rate": 0.0002, "epoch": 2.030150753768844, "step": 3030}, {"loss": 1.5832, "grad_norm": 0.387321799993515, "learning_rate": 0.0002, "epoch": 2.036850921273032, "step": 3040}, {"loss": 1.6834, "grad_norm": 0.4185757040977478, "learning_rate": 0.0002, "epoch": 2.0435510887772192, "step": 3050}, {"loss": 1.5696, "grad_norm": 0.45110777020454407, "learning_rate": 0.0002, "epoch": 2.050251256281407, "step": 3060}, {"loss": 1.6231, "grad_norm": 0.42663660645484924, "learning_rate": 0.0002, "epoch": 2.056951423785595, "step": 3070}, {"loss": 1.6279, "grad_norm": 0.4546292722225189, "learning_rate": 0.0002, "epoch": 2.063651591289782, "step": 3080}, {"loss": 1.6141, "grad_norm": 0.3979759216308594, "learning_rate": 0.0002, "epoch": 2.07035175879397, "step": 3090}, {"loss": 1.6343, "grad_norm": 0.43596673011779785, "learning_rate": 0.0002, "epoch": 2.0770519262981573, "step": 3100}, {"loss": 1.5441, "grad_norm": 0.40120232105255127, "learning_rate": 0.0002, "epoch": 2.083752093802345, "step": 3110}, {"loss": 1.6309, "grad_norm": 0.44449281692504883, "learning_rate": 0.0002, "epoch": 2.090452261306533, "step": 3120}, {"loss": 1.5652, "grad_norm": 0.42672568559646606, "learning_rate": 0.0002, "epoch": 2.09715242881072, "step": 3130}, {"loss": 1.682, "grad_norm": 0.4232690930366516, "learning_rate": 0.0002, "epoch": 2.103852596314908, "step": 3140}, {"loss": 1.624, "grad_norm": 0.4299317002296448, "learning_rate": 0.0002, "epoch": 2.1105527638190953, "step": 3150}, {"loss": 1.6766, "grad_norm": 0.4067758023738861, "learning_rate": 0.0002, "epoch": 2.117252931323283, "step": 3160}, {"loss": 1.6759, "grad_norm": 0.4918815791606903, "learning_rate": 0.0002, "epoch": 2.123953098827471, "step": 3170}, {"loss": 1.6478, "grad_norm": 0.4140559732913971, "learning_rate": 0.0002, "epoch": 2.130653266331658, "step": 3180}, {"loss": 1.6641, "grad_norm": 0.4555995464324951, "learning_rate": 0.0002, "epoch": 2.137353433835846, "step": 3190}, {"loss": 1.5888, "grad_norm": 0.42943915724754333, "learning_rate": 0.0002, "epoch": 2.1440536013400333, "step": 3200}, {"loss": 1.5886, "grad_norm": 0.4730435013771057, "learning_rate": 0.0002, "epoch": 2.150753768844221, "step": 3210}, {"loss": 1.6022, "grad_norm": 0.43310216069221497, "learning_rate": 0.0002, "epoch": 2.157453936348409, "step": 3220}, {"loss": 1.6058, "grad_norm": 0.42054110765457153, "learning_rate": 0.0002, "epoch": 2.164154103852596, "step": 3230}, {"loss": 1.6749, "grad_norm": 0.4897233247756958, "learning_rate": 0.0002, "epoch": 2.170854271356784, "step": 3240}, {"loss": 1.6983, "grad_norm": 0.42194533348083496, "learning_rate": 0.0002, "epoch": 2.1775544388609713, "step": 3250}, {"loss": 1.6159, "grad_norm": 0.44494450092315674, "learning_rate": 0.0002, "epoch": 2.184254606365159, "step": 3260}, {"loss": 1.6977, "grad_norm": 0.43524879217147827, "learning_rate": 0.0002, "epoch": 2.190954773869347, "step": 3270}, {"loss": 1.528, "grad_norm": 0.4621117413043976, "learning_rate": 0.0002, "epoch": 2.1976549413735342, "step": 3280}, {"loss": 1.632, "grad_norm": 0.4073285460472107, "learning_rate": 0.0002, "epoch": 2.204355108877722, "step": 3290}, {"loss": 1.6141, "grad_norm": 0.47868335247039795, "learning_rate": 0.0002, "epoch": 2.2110552763819094, "step": 3300}, {"loss": 1.6857, "grad_norm": 0.4264970123767853, "learning_rate": 0.0002, "epoch": 2.217755443886097, "step": 3310}, {"loss": 1.5653, "grad_norm": 0.4491245150566101, "learning_rate": 0.0002, "epoch": 2.224455611390285, "step": 3320}, {"loss": 1.5881, "grad_norm": 0.4010344445705414, "learning_rate": 0.0002, "epoch": 2.2311557788944723, "step": 3330}, {"loss": 1.6684, "grad_norm": 0.4232759177684784, "learning_rate": 0.0002, "epoch": 2.23785594639866, "step": 3340}, {"loss": 1.6336, "grad_norm": 0.5099776983261108, "learning_rate": 0.0002, "epoch": 2.2445561139028474, "step": 3350}, {"loss": 1.6764, "grad_norm": 0.5223407745361328, "learning_rate": 0.0002, "epoch": 2.251256281407035, "step": 3360}, {"loss": 1.6625, "grad_norm": 0.47818470001220703, "learning_rate": 0.0002, "epoch": 2.257956448911223, "step": 3370}, {"loss": 1.5946, "grad_norm": 0.4721255898475647, "learning_rate": 0.0002, "epoch": 2.2646566164154103, "step": 3380}, {"loss": 1.5568, "grad_norm": 0.4113229513168335, "learning_rate": 0.0002, "epoch": 2.271356783919598, "step": 3390}, {"loss": 1.6494, "grad_norm": 0.507080078125, "learning_rate": 0.0002, "epoch": 2.2780569514237854, "step": 3400}, {"loss": 1.6183, "grad_norm": 0.4852292239665985, "learning_rate": 0.0002, "epoch": 2.284757118927973, "step": 3410}, {"loss": 1.6132, "grad_norm": 0.4503684341907501, "learning_rate": 0.0002, "epoch": 2.291457286432161, "step": 3420}, {"loss": 1.6649, "grad_norm": 0.8359600305557251, "learning_rate": 0.0002, "epoch": 2.2981574539363483, "step": 3430}, {"loss": 1.6644, "grad_norm": 0.44604045152664185, "learning_rate": 0.0002, "epoch": 2.304857621440536, "step": 3440}, {"loss": 1.5972, "grad_norm": 0.45667049288749695, "learning_rate": 0.0002, "epoch": 2.3115577889447234, "step": 3450}, {"loss": 1.6667, "grad_norm": 0.4879349172115326, "learning_rate": 0.0002, "epoch": 2.318257956448911, "step": 3460}, {"loss": 1.5804, "grad_norm": 0.4033963084220886, "learning_rate": 0.0002, "epoch": 2.324958123953099, "step": 3470}, {"loss": 1.5838, "grad_norm": 0.44494301080703735, "learning_rate": 0.0002, "epoch": 2.3316582914572863, "step": 3480}, {"loss": 1.5951, "grad_norm": 0.4794621765613556, "learning_rate": 0.0002, "epoch": 2.338358458961474, "step": 3490}, {"loss": 1.6807, "grad_norm": 0.41404327750205994, "learning_rate": 0.0002, "epoch": 2.3450586264656614, "step": 3500}, {"loss": 1.714, "grad_norm": 0.4664851725101471, "learning_rate": 0.0002, "epoch": 2.351758793969849, "step": 3510}, {"loss": 1.6537, "grad_norm": 0.4263697564601898, "learning_rate": 0.0002, "epoch": 2.358458961474037, "step": 3520}, {"loss": 1.6551, "grad_norm": 0.5035167336463928, "learning_rate": 0.0002, "epoch": 2.3651591289782243, "step": 3530}, {"loss": 1.6208, "grad_norm": 0.4380664527416229, "learning_rate": 0.0002, "epoch": 2.371859296482412, "step": 3540}, {"loss": 1.634, "grad_norm": 0.5227681994438171, "learning_rate": 0.0002, "epoch": 2.3785594639865995, "step": 3550}, {"loss": 1.6146, "grad_norm": 0.4382302761077881, "learning_rate": 0.0002, "epoch": 2.3852596314907872, "step": 3560}, {"loss": 1.5653, "grad_norm": 0.4392451047897339, "learning_rate": 0.0002, "epoch": 2.391959798994975, "step": 3570}, {"loss": 1.6626, "grad_norm": 0.4372786581516266, "learning_rate": 0.0002, "epoch": 2.3986599664991624, "step": 3580}, {"loss": 1.519, "grad_norm": 0.5015502572059631, "learning_rate": 0.0002, "epoch": 2.40536013400335, "step": 3590}, {"loss": 1.588, "grad_norm": 0.5653210878372192, "learning_rate": 0.0002, "epoch": 2.4120603015075375, "step": 3600}, {"loss": 1.6075, "grad_norm": 0.53007972240448, "learning_rate": 0.0002, "epoch": 2.4187604690117253, "step": 3610}, {"loss": 1.6421, "grad_norm": 0.4659176766872406, "learning_rate": 0.0002, "epoch": 2.425460636515913, "step": 3620}, {"loss": 1.625, "grad_norm": 0.5637837052345276, "learning_rate": 0.0002, "epoch": 2.4321608040201004, "step": 3630}, {"loss": 1.6168, "grad_norm": 0.4248391389846802, "learning_rate": 0.0002, "epoch": 2.438860971524288, "step": 3640}, {"loss": 1.6822, "grad_norm": 0.44668248295783997, "learning_rate": 0.0002, "epoch": 2.4455611390284755, "step": 3650}, {"loss": 1.6417, "grad_norm": 0.43990179896354675, "learning_rate": 0.0002, "epoch": 2.4522613065326633, "step": 3660}, {"loss": 1.6723, "grad_norm": 0.4532523453235626, "learning_rate": 0.0002, "epoch": 2.458961474036851, "step": 3670}, {"loss": 1.6957, "grad_norm": 0.6605591773986816, "learning_rate": 0.0002, "epoch": 2.4656616415410384, "step": 3680}, {"loss": 1.6159, "grad_norm": 0.4694533348083496, "learning_rate": 0.0002, "epoch": 2.472361809045226, "step": 3690}, {"loss": 1.6239, "grad_norm": 0.4485011100769043, "learning_rate": 0.0002, "epoch": 2.4790619765494135, "step": 3700}, {"loss": 1.6834, "grad_norm": 0.4761785864830017, "learning_rate": 0.0002, "epoch": 2.4857621440536013, "step": 3710}, {"loss": 1.6313, "grad_norm": 0.5116432309150696, "learning_rate": 0.0002, "epoch": 2.492462311557789, "step": 3720}, {"loss": 1.5054, "grad_norm": 0.49523618817329407, "learning_rate": 0.0002, "epoch": 2.4991624790619764, "step": 3730}, {"loss": 1.6249, "grad_norm": 0.43826380372047424, "learning_rate": 0.0002, "epoch": 2.505862646566164, "step": 3740}, {"loss": 1.5762, "grad_norm": 0.4916154146194458, "learning_rate": 0.0002, "epoch": 2.5125628140703515, "step": 3750}, {"loss": 1.5157, "grad_norm": 0.5381299257278442, "learning_rate": 0.0002, "epoch": 2.5192629815745393, "step": 3760}, {"loss": 1.6467, "grad_norm": 0.44947415590286255, "learning_rate": 0.0002, "epoch": 2.525963149078727, "step": 3770}, {"loss": 1.67, "grad_norm": 0.49979084730148315, "learning_rate": 0.0002, "epoch": 2.5326633165829144, "step": 3780}, {"loss": 1.622, "grad_norm": 0.43046900629997253, "learning_rate": 0.0002, "epoch": 2.539363484087102, "step": 3790}, {"loss": 1.6789, "grad_norm": 0.4513470530509949, "learning_rate": 0.0002, "epoch": 2.5460636515912896, "step": 3800}, {"loss": 1.6335, "grad_norm": 0.49900051951408386, "learning_rate": 0.0002, "epoch": 2.5527638190954773, "step": 3810}, {"loss": 1.6322, "grad_norm": 0.4348420202732086, "learning_rate": 0.0002, "epoch": 2.559463986599665, "step": 3820}, {"loss": 1.6218, "grad_norm": 0.4684867560863495, "learning_rate": 0.0002, "epoch": 2.5661641541038525, "step": 3830}, {"loss": 1.6535, "grad_norm": 0.44430989027023315, "learning_rate": 0.0002, "epoch": 2.5728643216080402, "step": 3840}, {"loss": 1.5909, "grad_norm": 0.47375255823135376, "learning_rate": 0.0002, "epoch": 2.5795644891122276, "step": 3850}, {"loss": 1.6269, "grad_norm": 0.45493075251579285, "learning_rate": 0.0002, "epoch": 2.5862646566164154, "step": 3860}, {"loss": 1.604, "grad_norm": 0.4563275873661041, "learning_rate": 0.0002, "epoch": 2.592964824120603, "step": 3870}, {"loss": 1.642, "grad_norm": 0.46060335636138916, "learning_rate": 0.0002, "epoch": 2.5996649916247905, "step": 3880}, {"loss": 1.6302, "grad_norm": 0.4718867540359497, "learning_rate": 0.0002, "epoch": 2.6063651591289783, "step": 3890}, {"loss": 1.6242, "grad_norm": 0.41570305824279785, "learning_rate": 0.0002, "epoch": 2.6130653266331656, "step": 3900}, {"loss": 1.6401, "grad_norm": 0.4603121876716614, "learning_rate": 0.0002, "epoch": 2.6197654941373534, "step": 3910}, {"loss": 1.6839, "grad_norm": 0.4734652638435364, "learning_rate": 0.0002, "epoch": 2.626465661641541, "step": 3920}, {"loss": 1.5448, "grad_norm": 0.45348483324050903, "learning_rate": 0.0002, "epoch": 2.6331658291457285, "step": 3930}, {"loss": 1.6157, "grad_norm": 0.46559447050094604, "learning_rate": 0.0002, "epoch": 2.6398659966499163, "step": 3940}, {"loss": 1.7052, "grad_norm": 0.44113144278526306, "learning_rate": 0.0002, "epoch": 2.6465661641541036, "step": 3950}, {"loss": 1.6315, "grad_norm": 0.41415104269981384, "learning_rate": 0.0002, "epoch": 2.6532663316582914, "step": 3960}, {"loss": 1.6589, "grad_norm": 0.48868080973625183, "learning_rate": 0.0002, "epoch": 2.659966499162479, "step": 3970}, {"loss": 1.6211, "grad_norm": 0.49610549211502075, "learning_rate": 0.0002, "epoch": 2.6666666666666665, "step": 3980}, {"loss": 1.6235, "grad_norm": 0.4309130907058716, "learning_rate": 0.0002, "epoch": 2.6733668341708543, "step": 3990}, {"loss": 1.6452, "grad_norm": 0.4489327669143677, "learning_rate": 0.0002, "epoch": 2.6800670016750416, "step": 4000}, {"loss": 1.5744, "grad_norm": 0.5380139946937561, "learning_rate": 0.0002, "epoch": 2.6867671691792294, "step": 4010}, {"loss": 1.6524, "grad_norm": 0.5076672434806824, "learning_rate": 0.0002, "epoch": 2.693467336683417, "step": 4020}, {"loss": 1.636, "grad_norm": 0.47620031237602234, "learning_rate": 0.0002, "epoch": 2.7001675041876045, "step": 4030}, {"loss": 1.5543, "grad_norm": 0.48089155554771423, "learning_rate": 0.0002, "epoch": 2.7068676716917923, "step": 4040}, {"loss": 1.6396, "grad_norm": 0.5108814239501953, "learning_rate": 0.0002, "epoch": 2.7135678391959797, "step": 4050}, {"loss": 1.5905, "grad_norm": 0.4196513295173645, "learning_rate": 0.0002, "epoch": 2.7202680067001674, "step": 4060}, {"loss": 1.686, "grad_norm": 0.4574664831161499, "learning_rate": 0.0002, "epoch": 2.726968174204355, "step": 4070}, {"loss": 1.6234, "grad_norm": 0.4671640992164612, "learning_rate": 0.0002, "epoch": 2.7336683417085426, "step": 4080}, {"loss": 1.6827, "grad_norm": 0.49355530738830566, "learning_rate": 0.0002, "epoch": 2.7403685092127303, "step": 4090}, {"loss": 1.6999, "grad_norm": 0.46716663241386414, "learning_rate": 0.0002, "epoch": 2.7470686767169177, "step": 4100}, {"loss": 1.6463, "grad_norm": 0.45420581102371216, "learning_rate": 0.0002, "epoch": 2.7537688442211055, "step": 4110}, {"loss": 1.5718, "grad_norm": 0.4680487811565399, "learning_rate": 0.0002, "epoch": 2.7604690117252932, "step": 4120}, {"loss": 1.5968, "grad_norm": 0.5375032424926758, "learning_rate": 0.0002, "epoch": 2.7671691792294806, "step": 4130}, {"loss": 1.5254, "grad_norm": 0.46026280522346497, "learning_rate": 0.0002, "epoch": 2.7738693467336684, "step": 4140}, {"loss": 1.6613, "grad_norm": 0.43658447265625, "learning_rate": 0.0002, "epoch": 2.7805695142378557, "step": 4150}, {"loss": 1.6546, "grad_norm": 0.4935547113418579, "learning_rate": 0.0002, "epoch": 2.7872696817420435, "step": 4160}, {"loss": 1.5961, "grad_norm": 0.8167962431907654, "learning_rate": 0.0002, "epoch": 2.7939698492462313, "step": 4170}, {"loss": 1.6907, "grad_norm": 0.4289683997631073, "learning_rate": 0.0002, "epoch": 2.8006700167504186, "step": 4180}, {"loss": 1.6385, "grad_norm": 0.4569324254989624, "learning_rate": 0.0002, "epoch": 2.8073701842546064, "step": 4190}, {"loss": 1.6077, "grad_norm": 0.474795937538147, "learning_rate": 0.0002, "epoch": 2.8140703517587937, "step": 4200}, {"loss": 1.6223, "grad_norm": 0.44272229075431824, "learning_rate": 0.0002, "epoch": 2.8207705192629815, "step": 4210}, {"loss": 1.6706, "grad_norm": 0.525240957736969, "learning_rate": 0.0002, "epoch": 2.8274706867671693, "step": 4220}, {"loss": 1.7196, "grad_norm": 0.4802303910255432, "learning_rate": 0.0002, "epoch": 2.8341708542713566, "step": 4230}, {"loss": 1.6002, "grad_norm": 0.46400442719459534, "learning_rate": 0.0002, "epoch": 2.8408710217755444, "step": 4240}, {"loss": 1.6052, "grad_norm": 0.49884888529777527, "learning_rate": 0.0002, "epoch": 2.8475711892797317, "step": 4250}, {"loss": 1.6919, "grad_norm": 0.5015072226524353, "learning_rate": 0.0002, "epoch": 2.8542713567839195, "step": 4260}, {"loss": 1.6335, "grad_norm": 0.4335440695285797, "learning_rate": 0.0002, "epoch": 2.8609715242881073, "step": 4270}, {"loss": 1.5664, "grad_norm": 0.5131644606590271, "learning_rate": 0.0002, "epoch": 2.8676716917922946, "step": 4280}, {"loss": 1.6409, "grad_norm": 0.6977195739746094, "learning_rate": 0.0002, "epoch": 2.8743718592964824, "step": 4290}, {"loss": 1.7192, "grad_norm": 0.5133762955665588, "learning_rate": 0.0002, "epoch": 2.8810720268006698, "step": 4300}, {"loss": 1.6257, "grad_norm": 0.4737614393234253, "learning_rate": 0.0002, "epoch": 2.8877721943048575, "step": 4310}, {"loss": 1.6076, "grad_norm": 0.4580535590648651, "learning_rate": 0.0002, "epoch": 2.8944723618090453, "step": 4320}, {"loss": 1.6538, "grad_norm": 0.43863341212272644, "learning_rate": 0.0002, "epoch": 2.901172529313233, "step": 4330}, {"loss": 1.6091, "grad_norm": 0.4103737473487854, "learning_rate": 0.0002, "epoch": 2.9078726968174204, "step": 4340}, {"loss": 1.7106, "grad_norm": 0.438014417886734, "learning_rate": 0.0002, "epoch": 2.914572864321608, "step": 4350}, {"loss": 1.6025, "grad_norm": 0.5068213939666748, "learning_rate": 0.0002, "epoch": 2.9212730318257956, "step": 4360}, {"loss": 1.6426, "grad_norm": 0.45305484533309937, "learning_rate": 0.0002, "epoch": 2.9279731993299833, "step": 4370}, {"loss": 1.5726, "grad_norm": 0.4612090289592743, "learning_rate": 0.0002, "epoch": 2.934673366834171, "step": 4380}, {"loss": 1.6536, "grad_norm": 0.508736789226532, "learning_rate": 0.0002, "epoch": 2.9413735343383585, "step": 4390}, {"loss": 1.6132, "grad_norm": 0.4924427270889282, "learning_rate": 0.0002, "epoch": 2.948073701842546, "step": 4400}, {"loss": 1.7007, "grad_norm": 0.5707460641860962, "learning_rate": 0.0002, "epoch": 2.9547738693467336, "step": 4410}, {"loss": 1.6814, "grad_norm": 0.42270299792289734, "learning_rate": 0.0002, "epoch": 2.9614740368509214, "step": 4420}, {"loss": 1.6644, "grad_norm": 0.4429931044578552, "learning_rate": 0.0002, "epoch": 2.968174204355109, "step": 4430}, {"loss": 1.6251, "grad_norm": 0.49760574102401733, "learning_rate": 0.0002, "epoch": 2.9748743718592965, "step": 4440}, {"loss": 1.6169, "grad_norm": 0.4558229148387909, "learning_rate": 0.0002, "epoch": 2.981574539363484, "step": 4450}, {"loss": 1.6055, "grad_norm": 0.39848530292510986, "learning_rate": 0.0002, "epoch": 2.9882747068676716, "step": 4460}, {"loss": 1.6705, "grad_norm": 0.5224862098693848, "learning_rate": 0.0002, "epoch": 2.9949748743718594, "step": 4470}]}
+{"epoch": 4.0, "step": 5970, "epoch_duration": 1613.8745076656342, "total_accumulated_duration": 6342.423385620117, "gpu_info": {"GPU_0": "NVIDIA A100-SXM4-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 7568.4541015625}, "peak_memory_usage": {"GPU_0": 13792.75439453125}, "avg_memory_reserved": {"GPU_0": 17416.0}, "peak_memory_reserved": {"GPU_0": 17416.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "outputs-001/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-2985", "params": {"epochs": 8, "batch_size": 1, "learning_rate": 0.0002, "gradient_accumulation_steps": 2, "warmup_ratio": 0.03, "max_grad_norm": 0.3, "lora_alpha": 32, "lora_dropout": 0.05, "lora_r": 64, "tf32": true, "seed": 42}, "log_history": [{"loss": 2.6189, "grad_norm": 0.565915048122406, "learning_rate": 0.0002, "epoch": 0.006700167504187605, "step": 10}, {"loss": 2.3162, "grad_norm": 0.5004463791847229, "learning_rate": 0.0002, "epoch": 0.01340033500837521, "step": 20}, {"loss": 2.0576, "grad_norm": 0.511043906211853, "learning_rate": 0.0002, "epoch": 0.020100502512562814, "step": 30}, {"loss": 2.0085, "grad_norm": 0.47327178716659546, "learning_rate": 0.0002, "epoch": 0.02680067001675042, "step": 40}, {"loss": 2.0276, "grad_norm": 0.5511676669120789, "learning_rate": 0.0002, "epoch": 0.03350083752093802, "step": 50}, {"loss": 1.9075, "grad_norm": 0.4666278064250946, "learning_rate": 0.0002, "epoch": 0.04020100502512563, "step": 60}, {"loss": 1.8413, "grad_norm": 0.5310961008071899, "learning_rate": 0.0002, "epoch": 0.04690117252931323, "step": 70}, {"loss": 1.8711, "grad_norm": 0.5606027245521545, "learning_rate": 0.0002, "epoch": 0.05360134003350084, "step": 80}, {"loss": 1.9282, "grad_norm": 0.4934779703617096, "learning_rate": 0.0002, "epoch": 0.06030150753768844, "step": 90}, {"loss": 1.8925, "grad_norm": 0.4821869730949402, "learning_rate": 0.0002, "epoch": 0.06700167504187604, "step": 100}, {"loss": 1.8628, "grad_norm": 0.5262084603309631, "learning_rate": 0.0002, "epoch": 0.07370184254606366, "step": 110}, {"loss": 1.8347, "grad_norm": 0.3774230182170868, "learning_rate": 0.0002, "epoch": 0.08040201005025126, "step": 120}, {"loss": 1.8386, "grad_norm": 0.34137430787086487, "learning_rate": 0.0002, "epoch": 0.08710217755443886, "step": 130}, {"loss": 1.861, "grad_norm": 0.407272070646286, "learning_rate": 0.0002, "epoch": 0.09380234505862646, "step": 140}, {"loss": 1.8279, "grad_norm": 0.4011937975883484, "learning_rate": 0.0002, "epoch": 0.10050251256281408, "step": 150}, {"loss": 1.9317, "grad_norm": 0.4432467222213745, "learning_rate": 0.0002, "epoch": 0.10720268006700168, "step": 160}, {"loss": 1.8157, "grad_norm": 0.44030463695526123, "learning_rate": 0.0002, "epoch": 0.11390284757118928, "step": 170}, {"loss": 1.8534, "grad_norm": 0.3799569308757782, "learning_rate": 0.0002, "epoch": 0.12060301507537688, "step": 180}, {"loss": 1.7658, "grad_norm": 0.33721521496772766, "learning_rate": 0.0002, "epoch": 0.1273031825795645, "step": 190}, {"loss": 1.8269, "grad_norm": 0.4096226692199707, "learning_rate": 0.0002, "epoch": 0.13400335008375208, "step": 200}, {"loss": 1.802, "grad_norm": 0.37374693155288696, "learning_rate": 0.0002, "epoch": 0.1407035175879397, "step": 210}, {"loss": 1.8901, "grad_norm": 0.3249480128288269, "learning_rate": 0.0002, "epoch": 0.1474036850921273, "step": 220}, {"loss": 1.8163, "grad_norm": 0.3612042963504791, "learning_rate": 0.0002, "epoch": 0.1541038525963149, "step": 230}, {"loss": 1.7585, "grad_norm": 0.3686671257019043, "learning_rate": 0.0002, "epoch": 0.16080402010050251, "step": 240}, {"loss": 1.8365, "grad_norm": 0.3521044850349426, "learning_rate": 0.0002, "epoch": 0.16750418760469013, "step": 250}, {"loss": 1.8623, "grad_norm": 0.4073677361011505, "learning_rate": 0.0002, "epoch": 0.17420435510887772, "step": 260}, {"loss": 1.8026, "grad_norm": 0.34522193670272827, "learning_rate": 0.0002, "epoch": 0.18090452261306533, "step": 270}, {"loss": 1.8162, "grad_norm": 0.4121900498867035, "learning_rate": 0.0002, "epoch": 0.18760469011725292, "step": 280}, {"loss": 1.7976, "grad_norm": 0.3544778525829315, "learning_rate": 0.0002, "epoch": 0.19430485762144054, "step": 290}, {"loss": 1.8787, "grad_norm": 0.3482133448123932, "learning_rate": 0.0002, "epoch": 0.20100502512562815, "step": 300}, {"loss": 1.8578, "grad_norm": 0.3421826660633087, "learning_rate": 0.0002, "epoch": 0.20770519262981574, "step": 310}, {"loss": 1.8013, "grad_norm": 0.5024696588516235, "learning_rate": 0.0002, "epoch": 0.21440536013400335, "step": 320}, {"loss": 1.8607, "grad_norm": 0.36013063788414, "learning_rate": 0.0002, "epoch": 0.22110552763819097, "step": 330}, {"loss": 1.9075, "grad_norm": 0.3611244857311249, "learning_rate": 0.0002, "epoch": 0.22780569514237856, "step": 340}, {"loss": 1.8128, "grad_norm": 0.39244529604911804, "learning_rate": 0.0002, "epoch": 0.23450586264656617, "step": 350}, {"loss": 1.7885, "grad_norm": 0.3299325704574585, "learning_rate": 0.0002, "epoch": 0.24120603015075376, "step": 360}, {"loss": 1.8028, "grad_norm": 0.3994322419166565, "learning_rate": 0.0002, "epoch": 0.24790619765494137, "step": 370}, {"loss": 1.8321, "grad_norm": 0.3559151887893677, "learning_rate": 0.0002, "epoch": 0.254606365159129, "step": 380}, {"loss": 1.7802, "grad_norm": 0.3873756229877472, "learning_rate": 0.0002, "epoch": 0.2613065326633166, "step": 390}, {"loss": 1.7844, "grad_norm": 0.3710744082927704, "learning_rate": 0.0002, "epoch": 0.26800670016750416, "step": 400}, {"loss": 1.7789, "grad_norm": 0.3618465065956116, "learning_rate": 0.0002, "epoch": 0.2747068676716918, "step": 410}, {"loss": 1.8529, "grad_norm": 0.30063769221305847, "learning_rate": 0.0002, "epoch": 0.2814070351758794, "step": 420}, {"loss": 1.7765, "grad_norm": 0.3695628345012665, "learning_rate": 0.0002, "epoch": 0.288107202680067, "step": 430}, {"loss": 1.7982, "grad_norm": 0.31451135873794556, "learning_rate": 0.0002, "epoch": 0.2948073701842546, "step": 440}, {"loss": 1.7517, "grad_norm": 0.3959707021713257, "learning_rate": 0.0002, "epoch": 0.3015075376884422, "step": 450}, {"loss": 1.8142, "grad_norm": 0.33475354313850403, "learning_rate": 0.0002, "epoch": 0.3082077051926298, "step": 460}, {"loss": 1.8805, "grad_norm": 0.33933115005493164, "learning_rate": 0.0002, "epoch": 0.3149078726968174, "step": 470}, {"loss": 1.7564, "grad_norm": 0.3264943063259125, "learning_rate": 0.0002, "epoch": 0.32160804020100503, "step": 480}, {"loss": 1.8428, "grad_norm": 0.40188100934028625, "learning_rate": 0.0002, "epoch": 0.32830820770519265, "step": 490}, {"loss": 1.7624, "grad_norm": 0.37408649921417236, "learning_rate": 0.0002, "epoch": 0.33500837520938026, "step": 500}, {"loss": 1.7745, "grad_norm": 0.33925938606262207, "learning_rate": 0.0002, "epoch": 0.3417085427135678, "step": 510}, {"loss": 1.814, "grad_norm": 0.36836713552474976, "learning_rate": 0.0002, "epoch": 0.34840871021775544, "step": 520}, {"loss": 1.8037, "grad_norm": 0.37284499406814575, "learning_rate": 0.0002, "epoch": 0.35510887772194305, "step": 530}, {"loss": 1.8379, "grad_norm": 0.3192278742790222, "learning_rate": 0.0002, "epoch": 0.36180904522613067, "step": 540}, {"loss": 1.8702, "grad_norm": 0.30233290791511536, "learning_rate": 0.0002, "epoch": 0.3685092127303183, "step": 550}, {"loss": 1.8799, "grad_norm": 0.3340817391872406, "learning_rate": 0.0002, "epoch": 0.37520938023450584, "step": 560}, {"loss": 1.8404, "grad_norm": 0.32600095868110657, "learning_rate": 0.0002, "epoch": 0.38190954773869346, "step": 570}, {"loss": 1.7804, "grad_norm": 0.33711278438568115, "learning_rate": 0.0002, "epoch": 0.38860971524288107, "step": 580}, {"loss": 1.8445, "grad_norm": 0.34890690445899963, "learning_rate": 0.0002, "epoch": 0.3953098827470687, "step": 590}, {"loss": 1.8187, "grad_norm": 0.38238924741744995, "learning_rate": 0.0002, "epoch": 0.4020100502512563, "step": 600}, {"loss": 1.8111, "grad_norm": 0.34399354457855225, "learning_rate": 0.0002, "epoch": 0.40871021775544386, "step": 610}, {"loss": 1.8006, "grad_norm": 0.3346073627471924, "learning_rate": 0.0002, "epoch": 0.4154103852596315, "step": 620}, {"loss": 1.7705, "grad_norm": 0.3545648157596588, "learning_rate": 0.0002, "epoch": 0.4221105527638191, "step": 630}, {"loss": 1.8445, "grad_norm": 0.3378899097442627, "learning_rate": 0.0002, "epoch": 0.4288107202680067, "step": 640}, {"loss": 1.804, "grad_norm": 0.3255569040775299, "learning_rate": 0.0002, "epoch": 0.4355108877721943, "step": 650}, {"loss": 1.7679, "grad_norm": 0.34880587458610535, "learning_rate": 0.0002, "epoch": 0.44221105527638194, "step": 660}, {"loss": 1.7861, "grad_norm": 0.3402383625507355, "learning_rate": 0.0002, "epoch": 0.4489112227805695, "step": 670}, {"loss": 1.8131, "grad_norm": 0.3594033718109131, "learning_rate": 0.0002, "epoch": 0.4556113902847571, "step": 680}, {"loss": 1.8399, "grad_norm": 0.31000566482543945, "learning_rate": 0.0002, "epoch": 0.4623115577889447, "step": 690}, {"loss": 1.7521, "grad_norm": 0.37229061126708984, "learning_rate": 0.0002, "epoch": 0.46901172529313234, "step": 700}, {"loss": 1.7779, "grad_norm": 0.315801739692688, "learning_rate": 0.0002, "epoch": 0.47571189279731996, "step": 710}, {"loss": 1.7515, "grad_norm": 0.3220832645893097, "learning_rate": 0.0002, "epoch": 0.4824120603015075, "step": 720}, {"loss": 1.7181, "grad_norm": 0.3435456156730652, "learning_rate": 0.0002, "epoch": 0.48911222780569513, "step": 730}, {"loss": 1.8844, "grad_norm": 0.30380892753601074, "learning_rate": 0.0002, "epoch": 0.49581239530988275, "step": 740}, {"loss": 1.7792, "grad_norm": 0.3555026054382324, "learning_rate": 0.0002, "epoch": 0.5025125628140703, "step": 750}, {"loss": 1.7714, "grad_norm": 0.3019855320453644, "learning_rate": 0.0002, "epoch": 0.509212730318258, "step": 760}, {"loss": 1.7962, "grad_norm": 0.309111088514328, "learning_rate": 0.0002, "epoch": 0.5159128978224455, "step": 770}, {"loss": 1.7913, "grad_norm": 0.366020530462265, "learning_rate": 0.0002, "epoch": 0.5226130653266332, "step": 780}, {"loss": 1.8008, "grad_norm": 0.3267050087451935, "learning_rate": 0.0002, "epoch": 0.5293132328308208, "step": 790}, {"loss": 1.7397, "grad_norm": 0.34265750646591187, "learning_rate": 0.0002, "epoch": 0.5360134003350083, "step": 800}, {"loss": 1.8251, "grad_norm": 0.313669890165329, "learning_rate": 0.0002, "epoch": 0.542713567839196, "step": 810}, {"loss": 1.8369, "grad_norm": 0.3355236053466797, "learning_rate": 0.0002, "epoch": 0.5494137353433836, "step": 820}, {"loss": 1.7381, "grad_norm": 0.3186608552932739, "learning_rate": 0.0002, "epoch": 0.5561139028475712, "step": 830}, {"loss": 1.8034, "grad_norm": 0.30357518792152405, "learning_rate": 0.0002, "epoch": 0.5628140703517588, "step": 840}, {"loss": 1.769, "grad_norm": 0.3990040123462677, "learning_rate": 0.0002, "epoch": 0.5695142378559463, "step": 850}, {"loss": 1.7482, "grad_norm": 0.34363803267478943, "learning_rate": 0.0002, "epoch": 0.576214405360134, "step": 860}, {"loss": 1.8106, "grad_norm": 0.3757908046245575, "learning_rate": 0.0002, "epoch": 0.5829145728643216, "step": 870}, {"loss": 1.8104, "grad_norm": 0.3359757661819458, "learning_rate": 0.0002, "epoch": 0.5896147403685092, "step": 880}, {"loss": 1.7591, "grad_norm": 0.5555329918861389, "learning_rate": 0.0002, "epoch": 0.5963149078726968, "step": 890}, {"loss": 1.7715, "grad_norm": 0.4046323895454407, "learning_rate": 0.0002, "epoch": 0.6030150753768844, "step": 900}, {"loss": 1.7998, "grad_norm": 0.29834219813346863, "learning_rate": 0.0002, "epoch": 0.609715242881072, "step": 910}, {"loss": 1.7826, "grad_norm": 0.3241238594055176, "learning_rate": 0.0002, "epoch": 0.6164154103852596, "step": 920}, {"loss": 1.8342, "grad_norm": 0.35154739022254944, "learning_rate": 0.0002, "epoch": 0.6231155778894473, "step": 930}, {"loss": 1.8076, "grad_norm": 0.3287706673145294, "learning_rate": 0.0002, "epoch": 0.6298157453936348, "step": 940}, {"loss": 1.8038, "grad_norm": 0.35670626163482666, "learning_rate": 0.0002, "epoch": 0.6365159128978225, "step": 950}, {"loss": 1.869, "grad_norm": 0.6114104986190796, "learning_rate": 0.0002, "epoch": 0.6432160804020101, "step": 960}, {"loss": 1.8297, "grad_norm": 0.3186565041542053, "learning_rate": 0.0002, "epoch": 0.6499162479061976, "step": 970}, {"loss": 1.7539, "grad_norm": 0.27164125442504883, "learning_rate": 0.0002, "epoch": 0.6566164154103853, "step": 980}, {"loss": 1.8339, "grad_norm": 0.34407344460487366, "learning_rate": 0.0002, "epoch": 0.6633165829145728, "step": 990}, {"loss": 1.855, "grad_norm": 0.368415892124176, "learning_rate": 0.0002, "epoch": 0.6700167504187605, "step": 1000}, {"loss": 1.7821, "grad_norm": 0.3306390643119812, "learning_rate": 0.0002, "epoch": 0.6767169179229481, "step": 1010}, {"loss": 1.786, "grad_norm": 0.3198648989200592, "learning_rate": 0.0002, "epoch": 0.6834170854271356, "step": 1020}, {"loss": 1.816, "grad_norm": 0.3092987537384033, "learning_rate": 0.0002, "epoch": 0.6901172529313233, "step": 1030}, {"loss": 1.7689, "grad_norm": 0.3090653419494629, "learning_rate": 0.0002, "epoch": 0.6968174204355109, "step": 1040}, {"loss": 1.7544, "grad_norm": 0.3485880196094513, "learning_rate": 0.0002, "epoch": 0.7035175879396985, "step": 1050}, {"loss": 1.8187, "grad_norm": 0.35782721638679504, "learning_rate": 0.0002, "epoch": 0.7102177554438861, "step": 1060}, {"loss": 1.806, "grad_norm": 0.34256869554519653, "learning_rate": 0.0002, "epoch": 0.7169179229480737, "step": 1070}, {"loss": 1.7873, "grad_norm": 0.30461037158966064, "learning_rate": 0.0002, "epoch": 0.7236180904522613, "step": 1080}, {"loss": 1.7367, "grad_norm": 0.3398691713809967, "learning_rate": 0.0002, "epoch": 0.7303182579564489, "step": 1090}, {"loss": 1.8756, "grad_norm": 0.3180808126926422, "learning_rate": 0.0002, "epoch": 0.7370184254606366, "step": 1100}, {"loss": 1.6988, "grad_norm": 0.34400665760040283, "learning_rate": 0.0002, "epoch": 0.7437185929648241, "step": 1110}, {"loss": 1.7851, "grad_norm": 0.34244877099990845, "learning_rate": 0.0002, "epoch": 0.7504187604690117, "step": 1120}, {"loss": 1.7841, "grad_norm": 0.29946693778038025, "learning_rate": 0.0002, "epoch": 0.7571189279731994, "step": 1130}, {"loss": 1.7456, "grad_norm": 0.37547236680984497, "learning_rate": 0.0002, "epoch": 0.7638190954773869, "step": 1140}, {"loss": 1.8425, "grad_norm": 0.3263005018234253, "learning_rate": 0.0002, "epoch": 0.7705192629815746, "step": 1150}, {"loss": 1.7222, "grad_norm": 0.41363608837127686, "learning_rate": 0.0002, "epoch": 0.7772194304857621, "step": 1160}, {"loss": 1.7836, "grad_norm": 0.36267954111099243, "learning_rate": 0.0002, "epoch": 0.7839195979899497, "step": 1170}, {"loss": 1.9183, "grad_norm": 0.31789499521255493, "learning_rate": 0.0002, "epoch": 0.7906197654941374, "step": 1180}, {"loss": 1.78, "grad_norm": 0.5708149075508118, "learning_rate": 0.0002, "epoch": 0.7973199329983249, "step": 1190}, {"loss": 1.6908, "grad_norm": 0.322099506855011, "learning_rate": 0.0002, "epoch": 0.8040201005025126, "step": 1200}, {"loss": 1.7639, "grad_norm": 0.3419909179210663, "learning_rate": 0.0002, "epoch": 0.8107202680067002, "step": 1210}, {"loss": 1.7428, "grad_norm": 0.36286255717277527, "learning_rate": 0.0002, "epoch": 0.8174204355108877, "step": 1220}, {"loss": 1.8409, "grad_norm": 0.33992862701416016, "learning_rate": 0.0002, "epoch": 0.8241206030150754, "step": 1230}, {"loss": 1.7507, "grad_norm": 0.32622793316841125, "learning_rate": 0.0002, "epoch": 0.830820770519263, "step": 1240}, {"loss": 1.8098, "grad_norm": 0.3036167621612549, "learning_rate": 0.0002, "epoch": 0.8375209380234506, "step": 1250}, {"loss": 1.8094, "grad_norm": 0.3182215392589569, "learning_rate": 0.0002, "epoch": 0.8442211055276382, "step": 1260}, {"loss": 1.8017, "grad_norm": 0.3270018696784973, "learning_rate": 0.0002, "epoch": 0.8509212730318257, "step": 1270}, {"loss": 1.798, "grad_norm": 0.32652342319488525, "learning_rate": 0.0002, "epoch": 0.8576214405360134, "step": 1280}, {"loss": 1.7448, "grad_norm": 0.3631329834461212, "learning_rate": 0.0002, "epoch": 0.864321608040201, "step": 1290}, {"loss": 1.7, "grad_norm": 0.36706018447875977, "learning_rate": 0.0002, "epoch": 0.8710217755443886, "step": 1300}, {"loss": 1.8178, "grad_norm": 0.3347418010234833, "learning_rate": 0.0002, "epoch": 0.8777219430485762, "step": 1310}, {"loss": 1.7824, "grad_norm": 0.34371060132980347, "learning_rate": 0.0002, "epoch": 0.8844221105527639, "step": 1320}, {"loss": 1.783, "grad_norm": 0.3029090166091919, "learning_rate": 0.0002, "epoch": 0.8911222780569514, "step": 1330}, {"loss": 1.8017, "grad_norm": 0.34700682759284973, "learning_rate": 0.0002, "epoch": 0.897822445561139, "step": 1340}, {"loss": 1.7998, "grad_norm": 0.35574328899383545, "learning_rate": 0.0002, "epoch": 0.9045226130653267, "step": 1350}, {"loss": 1.7339, "grad_norm": 0.30928221344947815, "learning_rate": 0.0002, "epoch": 0.9112227805695142, "step": 1360}, {"loss": 1.7479, "grad_norm": 0.30652928352355957, "learning_rate": 0.0002, "epoch": 0.9179229480737019, "step": 1370}, {"loss": 1.7491, "grad_norm": 0.3838157653808594, "learning_rate": 0.0002, "epoch": 0.9246231155778895, "step": 1380}, {"loss": 1.7977, "grad_norm": 0.31655240058898926, "learning_rate": 0.0002, "epoch": 0.931323283082077, "step": 1390}, {"loss": 1.8175, "grad_norm": 0.41737303137779236, "learning_rate": 0.0002, "epoch": 0.9380234505862647, "step": 1400}, {"loss": 1.6811, "grad_norm": 0.3227267861366272, "learning_rate": 0.0002, "epoch": 0.9447236180904522, "step": 1410}, {"loss": 1.7343, "grad_norm": 0.3729925751686096, "learning_rate": 0.0002, "epoch": 0.9514237855946399, "step": 1420}, {"loss": 1.8221, "grad_norm": 0.30779409408569336, "learning_rate": 0.0002, "epoch": 0.9581239530988275, "step": 1430}, {"loss": 1.7972, "grad_norm": 0.334379643201828, "learning_rate": 0.0002, "epoch": 0.964824120603015, "step": 1440}, {"loss": 1.7141, "grad_norm": 0.3568236231803894, "learning_rate": 0.0002, "epoch": 0.9715242881072027, "step": 1450}, {"loss": 1.7541, "grad_norm": 0.33310577273368835, "learning_rate": 0.0002, "epoch": 0.9782244556113903, "step": 1460}, {"loss": 1.8511, "grad_norm": 0.2972261905670166, "learning_rate": 0.0002, "epoch": 0.9849246231155779, "step": 1470}, {"loss": 1.7654, "grad_norm": 0.3322717845439911, "learning_rate": 0.0002, "epoch": 0.9916247906197655, "step": 1480}, {"loss": 1.8033, "grad_norm": 0.3276330828666687, "learning_rate": 0.0002, "epoch": 0.998324958123953, "step": 1490}, {"eval_loss": 1.8036354780197144, "eval_runtime": 37.8949, "eval_samples_per_second": 13.59, "eval_steps_per_second": 1.715, "epoch": 0.9996649916247906, "step": 1492}, {"loss": 1.7138, "grad_norm": 0.29252371191978455, "learning_rate": 0.0002, "epoch": 1.0050251256281406, "step": 1500}, {"loss": 1.8198, "grad_norm": 0.31607162952423096, "learning_rate": 0.0002, "epoch": 1.0117252931323284, "step": 1510}, {"loss": 1.6779, "grad_norm": 0.32294467091560364, "learning_rate": 0.0002, "epoch": 1.018425460636516, "step": 1520}, {"loss": 1.7919, "grad_norm": 0.3868017792701721, "learning_rate": 0.0002, "epoch": 1.0251256281407035, "step": 1530}, {"loss": 1.7954, "grad_norm": 0.3178282082080841, "learning_rate": 0.0002, "epoch": 1.031825795644891, "step": 1540}, {"loss": 1.7136, "grad_norm": 0.3706750273704529, "learning_rate": 0.0002, "epoch": 1.0385259631490786, "step": 1550}, {"loss": 1.7382, "grad_norm": 0.33930912613868713, "learning_rate": 0.0002, "epoch": 1.0452261306532664, "step": 1560}, {"loss": 1.7602, "grad_norm": 0.33970504999160767, "learning_rate": 0.0002, "epoch": 1.051926298157454, "step": 1570}, {"loss": 1.6573, "grad_norm": 0.42553383111953735, "learning_rate": 0.0002, "epoch": 1.0586264656616415, "step": 1580}, {"loss": 1.645, "grad_norm": 0.3772421181201935, "learning_rate": 0.0002, "epoch": 1.065326633165829, "step": 1590}, {"loss": 1.7362, "grad_norm": 0.34212902188301086, "learning_rate": 0.0002, "epoch": 1.0720268006700167, "step": 1600}, {"loss": 1.7057, "grad_norm": 0.3798283338546753, "learning_rate": 0.0002, "epoch": 1.0787269681742044, "step": 1610}, {"loss": 1.7468, "grad_norm": 0.36909598112106323, "learning_rate": 0.0002, "epoch": 1.085427135678392, "step": 1620}, {"loss": 1.7807, "grad_norm": 0.3344230651855469, "learning_rate": 0.0002, "epoch": 1.0921273031825796, "step": 1630}, {"loss": 1.7111, "grad_norm": 0.3862569332122803, "learning_rate": 0.0002, "epoch": 1.0988274706867671, "step": 1640}, {"loss": 1.7163, "grad_norm": 0.31188511848449707, "learning_rate": 0.0002, "epoch": 1.1055276381909547, "step": 1650}, {"loss": 1.7263, "grad_norm": 0.3563670814037323, "learning_rate": 0.0002, "epoch": 1.1122278056951425, "step": 1660}, {"loss": 1.7718, "grad_norm": 0.35052165389060974, "learning_rate": 0.0002, "epoch": 1.11892797319933, "step": 1670}, {"loss": 1.7601, "grad_norm": 0.3285699188709259, "learning_rate": 0.0002, "epoch": 1.1256281407035176, "step": 1680}, {"loss": 1.6877, "grad_norm": 0.3639393746852875, "learning_rate": 0.0002, "epoch": 1.1323283082077051, "step": 1690}, {"loss": 1.7719, "grad_norm": 0.3842753767967224, "learning_rate": 0.0002, "epoch": 1.1390284757118927, "step": 1700}, {"loss": 1.7002, "grad_norm": 0.3624933063983917, "learning_rate": 0.0002, "epoch": 1.1457286432160805, "step": 1710}, {"loss": 1.7243, "grad_norm": 0.3641220033168793, "learning_rate": 0.0002, "epoch": 1.152428810720268, "step": 1720}, {"loss": 1.752, "grad_norm": 0.32765355706214905, "learning_rate": 0.0002, "epoch": 1.1591289782244556, "step": 1730}, {"loss": 1.6556, "grad_norm": 0.34974896907806396, "learning_rate": 0.0002, "epoch": 1.1658291457286432, "step": 1740}, {"loss": 1.7273, "grad_norm": 0.3910926580429077, "learning_rate": 0.0002, "epoch": 1.1725293132328307, "step": 1750}, {"loss": 1.7204, "grad_norm": 0.3564300537109375, "learning_rate": 0.0002, "epoch": 1.1792294807370185, "step": 1760}, {"loss": 1.746, "grad_norm": 0.34822574257850647, "learning_rate": 0.0002, "epoch": 1.185929648241206, "step": 1770}, {"loss": 1.7256, "grad_norm": 0.36185044050216675, "learning_rate": 0.0002, "epoch": 1.1926298157453936, "step": 1780}, {"loss": 1.6431, "grad_norm": 0.34866711497306824, "learning_rate": 0.0002, "epoch": 1.1993299832495812, "step": 1790}, {"loss": 1.8084, "grad_norm": 0.4017769992351532, "learning_rate": 0.0002, "epoch": 1.2060301507537687, "step": 1800}, {"loss": 1.6985, "grad_norm": 0.32930681109428406, "learning_rate": 0.0002, "epoch": 1.2127303182579565, "step": 1810}, {"loss": 1.7606, "grad_norm": 0.35951921343803406, "learning_rate": 0.0002, "epoch": 1.219430485762144, "step": 1820}, {"loss": 1.6933, "grad_norm": 0.37366992235183716, "learning_rate": 0.0002, "epoch": 1.2261306532663316, "step": 1830}, {"loss": 1.6737, "grad_norm": 0.3565689027309418, "learning_rate": 0.0002, "epoch": 1.2328308207705192, "step": 1840}, {"loss": 1.8013, "grad_norm": 0.3692343533039093, "learning_rate": 0.0002, "epoch": 1.2395309882747068, "step": 1850}, {"loss": 1.736, "grad_norm": 0.38426971435546875, "learning_rate": 0.0002, "epoch": 1.2462311557788945, "step": 1860}, {"loss": 1.7031, "grad_norm": 0.33559855818748474, "learning_rate": 0.0002, "epoch": 1.252931323283082, "step": 1870}, {"loss": 1.7033, "grad_norm": 0.34181106090545654, "learning_rate": 0.0002, "epoch": 1.2596314907872697, "step": 1880}, {"loss": 1.7707, "grad_norm": 0.3916318416595459, "learning_rate": 0.0002, "epoch": 1.2663316582914572, "step": 1890}, {"loss": 1.6686, "grad_norm": 0.3887825012207031, "learning_rate": 0.0002, "epoch": 1.2730318257956448, "step": 1900}, {"loss": 1.7062, "grad_norm": 0.33583927154541016, "learning_rate": 0.0002, "epoch": 1.2797319932998326, "step": 1910}, {"loss": 1.717, "grad_norm": 0.37639349699020386, "learning_rate": 0.0002, "epoch": 1.2864321608040201, "step": 1920}, {"loss": 1.777, "grad_norm": 0.38059428334236145, "learning_rate": 0.0002, "epoch": 1.2931323283082077, "step": 1930}, {"loss": 1.6126, "grad_norm": 0.37253183126449585, "learning_rate": 0.0002, "epoch": 1.2998324958123952, "step": 1940}, {"loss": 1.6758, "grad_norm": 0.37371566891670227, "learning_rate": 0.0002, "epoch": 1.3065326633165828, "step": 1950}, {"loss": 1.6788, "grad_norm": 0.4080910086631775, "learning_rate": 0.0002, "epoch": 1.3132328308207706, "step": 1960}, {"loss": 1.6518, "grad_norm": 0.3174354135990143, "learning_rate": 0.0002, "epoch": 1.3199329983249581, "step": 1970}, {"loss": 1.7925, "grad_norm": 0.4518888294696808, "learning_rate": 0.0002, "epoch": 1.3266331658291457, "step": 1980}, {"loss": 1.7085, "grad_norm": 0.3627921938896179, "learning_rate": 0.0002, "epoch": 1.3333333333333333, "step": 1990}, {"loss": 1.7676, "grad_norm": 0.3655930161476135, "learning_rate": 0.0002, "epoch": 1.3400335008375208, "step": 2000}, {"loss": 1.7016, "grad_norm": 0.3509993255138397, "learning_rate": 0.0002, "epoch": 1.3467336683417086, "step": 2010}, {"loss": 1.7359, "grad_norm": 0.4281129240989685, "learning_rate": 0.0002, "epoch": 1.3534338358458962, "step": 2020}, {"loss": 1.6884, "grad_norm": 0.3821414113044739, "learning_rate": 0.0002, "epoch": 1.3601340033500837, "step": 2030}, {"loss": 1.7075, "grad_norm": 0.3907586336135864, "learning_rate": 0.0002, "epoch": 1.3668341708542713, "step": 2040}, {"loss": 1.7424, "grad_norm": 0.37792932987213135, "learning_rate": 0.0002, "epoch": 1.3735343383584588, "step": 2050}, {"loss": 1.7305, "grad_norm": 0.3693985641002655, "learning_rate": 0.0002, "epoch": 1.3802345058626466, "step": 2060}, {"loss": 1.7434, "grad_norm": 0.32275936007499695, "learning_rate": 0.0002, "epoch": 1.3869346733668342, "step": 2070}, {"loss": 1.6677, "grad_norm": 0.3789440095424652, "learning_rate": 0.0002, "epoch": 1.3936348408710217, "step": 2080}, {"loss": 1.6825, "grad_norm": 0.3638380467891693, "learning_rate": 0.0002, "epoch": 1.4003350083752093, "step": 2090}, {"loss": 1.6542, "grad_norm": 0.3495481610298157, "learning_rate": 0.0002, "epoch": 1.4070351758793969, "step": 2100}, {"loss": 1.7225, "grad_norm": 0.37920597195625305, "learning_rate": 0.0002, "epoch": 1.4137353433835846, "step": 2110}, {"loss": 1.7329, "grad_norm": 0.37218064069747925, "learning_rate": 0.0002, "epoch": 1.4204355108877722, "step": 2120}, {"loss": 1.799, "grad_norm": 0.38074082136154175, "learning_rate": 0.0002, "epoch": 1.4271356783919598, "step": 2130}, {"loss": 1.7403, "grad_norm": 0.3455527126789093, "learning_rate": 0.0002, "epoch": 1.4338358458961473, "step": 2140}, {"loss": 1.776, "grad_norm": 0.3712003529071808, "learning_rate": 0.0002, "epoch": 1.4405360134003349, "step": 2150}, {"loss": 1.7619, "grad_norm": 0.3786754906177521, "learning_rate": 0.0002, "epoch": 1.4472361809045227, "step": 2160}, {"loss": 1.68, "grad_norm": 0.3879223167896271, "learning_rate": 0.0002, "epoch": 1.4539363484087102, "step": 2170}, {"loss": 1.7, "grad_norm": 0.38738805055618286, "learning_rate": 0.0002, "epoch": 1.4606365159128978, "step": 2180}, {"loss": 1.7581, "grad_norm": 0.39768800139427185, "learning_rate": 0.0002, "epoch": 1.4673366834170856, "step": 2190}, {"loss": 1.7671, "grad_norm": 0.4172441065311432, "learning_rate": 0.0002, "epoch": 1.474036850921273, "step": 2200}, {"loss": 1.6736, "grad_norm": 0.4043174982070923, "learning_rate": 0.0002, "epoch": 1.4807370184254607, "step": 2210}, {"loss": 1.7444, "grad_norm": 0.3750883936882019, "learning_rate": 0.0002, "epoch": 1.4874371859296482, "step": 2220}, {"loss": 1.6861, "grad_norm": 0.3552253246307373, "learning_rate": 0.0002, "epoch": 1.4941373534338358, "step": 2230}, {"loss": 1.6471, "grad_norm": 0.34607139229774475, "learning_rate": 0.0002, "epoch": 1.5008375209380236, "step": 2240}, {"loss": 1.6962, "grad_norm": 0.3406706750392914, "learning_rate": 0.0002, "epoch": 1.507537688442211, "step": 2250}, {"loss": 1.7694, "grad_norm": 0.36654895544052124, "learning_rate": 0.0002, "epoch": 1.5142378559463987, "step": 2260}, {"loss": 1.6812, "grad_norm": 0.3914054334163666, "learning_rate": 0.0002, "epoch": 1.5209380234505863, "step": 2270}, {"loss": 1.6822, "grad_norm": 0.42012137174606323, "learning_rate": 0.0002, "epoch": 1.5276381909547738, "step": 2280}, {"loss": 1.697, "grad_norm": 0.39563435316085815, "learning_rate": 0.0002, "epoch": 1.5343383584589616, "step": 2290}, {"loss": 1.7491, "grad_norm": 0.3508438766002655, "learning_rate": 0.0002, "epoch": 1.541038525963149, "step": 2300}, {"loss": 1.7727, "grad_norm": 0.3785218596458435, "learning_rate": 0.0002, "epoch": 1.5477386934673367, "step": 2310}, {"loss": 1.6963, "grad_norm": 0.39377647638320923, "learning_rate": 0.0002, "epoch": 1.5544388609715243, "step": 2320}, {"loss": 1.7263, "grad_norm": 0.3391438126564026, "learning_rate": 0.0002, "epoch": 1.5611390284757118, "step": 2330}, {"loss": 1.7722, "grad_norm": 0.37944263219833374, "learning_rate": 0.0002, "epoch": 1.5678391959798996, "step": 2340}, {"loss": 1.6371, "grad_norm": 0.3523491322994232, "learning_rate": 0.0002, "epoch": 1.574539363484087, "step": 2350}, {"loss": 1.7583, "grad_norm": 0.3911575973033905, "learning_rate": 0.0002, "epoch": 1.5812395309882747, "step": 2360}, {"loss": 1.7117, "grad_norm": 0.33832186460494995, "learning_rate": 0.0002, "epoch": 1.5879396984924623, "step": 2370}, {"loss": 1.7701, "grad_norm": 0.3665979206562042, "learning_rate": 0.0002, "epoch": 1.5946398659966499, "step": 2380}, {"loss": 1.779, "grad_norm": 0.3871748149394989, "learning_rate": 0.0002, "epoch": 1.6013400335008376, "step": 2390}, {"loss": 1.7109, "grad_norm": 0.3586967885494232, "learning_rate": 0.0002, "epoch": 1.608040201005025, "step": 2400}, {"loss": 1.7096, "grad_norm": 0.3563673198223114, "learning_rate": 0.0002, "epoch": 1.6147403685092128, "step": 2410}, {"loss": 1.745, "grad_norm": 0.37588971853256226, "learning_rate": 0.0002, "epoch": 1.6214405360134003, "step": 2420}, {"loss": 1.7086, "grad_norm": 0.352556437253952, "learning_rate": 0.0002, "epoch": 1.6281407035175879, "step": 2430}, {"loss": 1.6547, "grad_norm": 0.3716259300708771, "learning_rate": 0.0002, "epoch": 1.6348408710217757, "step": 2440}, {"loss": 1.7033, "grad_norm": 0.372001975774765, "learning_rate": 0.0002, "epoch": 1.641541038525963, "step": 2450}, {"loss": 1.6584, "grad_norm": 0.3430042862892151, "learning_rate": 0.0002, "epoch": 1.6482412060301508, "step": 2460}, {"loss": 1.7217, "grad_norm": 0.3741483688354492, "learning_rate": 0.0002, "epoch": 1.6549413735343383, "step": 2470}, {"loss": 1.7701, "grad_norm": 0.3610571324825287, "learning_rate": 0.0002, "epoch": 1.661641541038526, "step": 2480}, {"loss": 1.7057, "grad_norm": 0.4204719066619873, "learning_rate": 0.0002, "epoch": 1.6683417085427137, "step": 2490}, {"loss": 1.7954, "grad_norm": 0.3938186466693878, "learning_rate": 0.0002, "epoch": 1.675041876046901, "step": 2500}, {"loss": 1.6633, "grad_norm": 0.3421435058116913, "learning_rate": 0.0002, "epoch": 1.6817420435510888, "step": 2510}, {"loss": 1.7996, "grad_norm": 0.42441412806510925, "learning_rate": 0.0002, "epoch": 1.6884422110552764, "step": 2520}, {"loss": 1.7142, "grad_norm": 0.38071519136428833, "learning_rate": 0.0002, "epoch": 1.695142378559464, "step": 2530}, {"loss": 1.7232, "grad_norm": 0.34078919887542725, "learning_rate": 0.0002, "epoch": 1.7018425460636517, "step": 2540}, {"loss": 1.7126, "grad_norm": 0.412844181060791, "learning_rate": 0.0002, "epoch": 1.708542713567839, "step": 2550}, {"loss": 1.7149, "grad_norm": 0.3753604292869568, "learning_rate": 0.0002, "epoch": 1.7152428810720268, "step": 2560}, {"loss": 1.7011, "grad_norm": 0.41588476300239563, "learning_rate": 0.0002, "epoch": 1.7219430485762144, "step": 2570}, {"loss": 1.6427, "grad_norm": 0.35504111647605896, "learning_rate": 0.0002, "epoch": 1.728643216080402, "step": 2580}, {"loss": 1.7296, "grad_norm": 0.36909720301628113, "learning_rate": 0.0002, "epoch": 1.7353433835845897, "step": 2590}, {"loss": 1.7022, "grad_norm": 0.4149979054927826, "learning_rate": 0.0002, "epoch": 1.742043551088777, "step": 2600}, {"loss": 1.77, "grad_norm": 0.38859328627586365, "learning_rate": 0.0002, "epoch": 1.7487437185929648, "step": 2610}, {"loss": 1.7036, "grad_norm": 0.36738792061805725, "learning_rate": 0.0002, "epoch": 1.7554438860971524, "step": 2620}, {"loss": 1.764, "grad_norm": 0.3968178927898407, "learning_rate": 0.0002, "epoch": 1.76214405360134, "step": 2630}, {"loss": 1.7687, "grad_norm": 0.3972901999950409, "learning_rate": 0.0002, "epoch": 1.7688442211055277, "step": 2640}, {"loss": 1.6774, "grad_norm": 0.3949959874153137, "learning_rate": 0.0002, "epoch": 1.775544388609715, "step": 2650}, {"loss": 1.7247, "grad_norm": 0.44074657559394836, "learning_rate": 0.0002, "epoch": 1.7822445561139029, "step": 2660}, {"loss": 1.7188, "grad_norm": 0.39743664860725403, "learning_rate": 0.0002, "epoch": 1.7889447236180904, "step": 2670}, {"loss": 1.7258, "grad_norm": 0.3950406610965729, "learning_rate": 0.0002, "epoch": 1.795644891122278, "step": 2680}, {"loss": 1.6906, "grad_norm": 0.3568263649940491, "learning_rate": 0.0002, "epoch": 1.8023450586264658, "step": 2690}, {"loss": 1.6735, "grad_norm": 0.3819476366043091, "learning_rate": 0.0002, "epoch": 1.809045226130653, "step": 2700}, {"loss": 1.7198, "grad_norm": 0.3480634391307831, "learning_rate": 0.0002, "epoch": 1.8157453936348409, "step": 2710}, {"loss": 1.7042, "grad_norm": 0.3875853419303894, "learning_rate": 0.0002, "epoch": 1.8224455611390284, "step": 2720}, {"loss": 1.6988, "grad_norm": 0.3441337049007416, "learning_rate": 0.0002, "epoch": 1.829145728643216, "step": 2730}, {"loss": 1.7647, "grad_norm": 0.35692882537841797, "learning_rate": 0.0002, "epoch": 1.8358458961474038, "step": 2740}, {"loss": 1.7033, "grad_norm": 0.36959215998649597, "learning_rate": 0.0002, "epoch": 1.8425460636515911, "step": 2750}, {"loss": 1.7657, "grad_norm": 0.3893393278121948, "learning_rate": 0.0002, "epoch": 1.849246231155779, "step": 2760}, {"loss": 1.7068, "grad_norm": 0.37817293405532837, "learning_rate": 0.0002, "epoch": 1.8559463986599665, "step": 2770}, {"loss": 1.761, "grad_norm": 0.36071285605430603, "learning_rate": 0.0002, "epoch": 1.862646566164154, "step": 2780}, {"loss": 1.7623, "grad_norm": 0.3758420944213867, "learning_rate": 0.0002, "epoch": 1.8693467336683418, "step": 2790}, {"loss": 1.6743, "grad_norm": 0.3889938294887543, "learning_rate": 0.0002, "epoch": 1.8760469011725294, "step": 2800}, {"loss": 1.6151, "grad_norm": 0.34361857175827026, "learning_rate": 0.0002, "epoch": 1.882747068676717, "step": 2810}, {"loss": 1.6038, "grad_norm": 0.39283323287963867, "learning_rate": 0.0002, "epoch": 1.8894472361809045, "step": 2820}, {"loss": 1.7555, "grad_norm": 0.3919452726840973, "learning_rate": 0.0002, "epoch": 1.896147403685092, "step": 2830}, {"loss": 1.673, "grad_norm": 0.38215070962905884, "learning_rate": 0.0002, "epoch": 1.9028475711892798, "step": 2840}, {"loss": 1.7044, "grad_norm": 0.4235064387321472, "learning_rate": 0.0002, "epoch": 1.9095477386934674, "step": 2850}, {"loss": 1.7123, "grad_norm": 0.35694634914398193, "learning_rate": 0.0002, "epoch": 1.916247906197655, "step": 2860}, {"loss": 1.8128, "grad_norm": 0.383492112159729, "learning_rate": 0.0002, "epoch": 1.9229480737018425, "step": 2870}, {"loss": 1.7581, "grad_norm": 0.5945147275924683, "learning_rate": 0.0002, "epoch": 1.92964824120603, "step": 2880}, {"loss": 1.7421, "grad_norm": 0.3367522358894348, "learning_rate": 0.0002, "epoch": 1.9363484087102178, "step": 2890}, {"loss": 1.6561, "grad_norm": 0.35300394892692566, "learning_rate": 0.0002, "epoch": 1.9430485762144054, "step": 2900}, {"loss": 1.7033, "grad_norm": 0.38084495067596436, "learning_rate": 0.0002, "epoch": 1.949748743718593, "step": 2910}, {"loss": 1.7132, "grad_norm": 0.37559160590171814, "learning_rate": 0.0002, "epoch": 1.9564489112227805, "step": 2920}, {"loss": 1.6759, "grad_norm": 0.3661738336086273, "learning_rate": 0.0002, "epoch": 1.963149078726968, "step": 2930}, {"loss": 1.7643, "grad_norm": 0.4073849320411682, "learning_rate": 0.0002, "epoch": 1.9698492462311559, "step": 2940}, {"loss": 1.6806, "grad_norm": 0.3723304271697998, "learning_rate": 0.0002, "epoch": 1.9765494137353434, "step": 2950}, {"loss": 1.7611, "grad_norm": 0.3991098999977112, "learning_rate": 0.0002, "epoch": 1.983249581239531, "step": 2960}, {"loss": 1.7263, "grad_norm": 0.3947085440158844, "learning_rate": 0.0002, "epoch": 1.9899497487437185, "step": 2970}, {"loss": 1.7217, "grad_norm": 0.3786258399486542, "learning_rate": 0.0002, "epoch": 1.996649916247906, "step": 2980}, {"eval_loss": 1.8028968572616577, "eval_runtime": 37.8985, "eval_samples_per_second": 13.589, "eval_steps_per_second": 1.715, "epoch": 2.0, "step": 2985}, {"loss": 1.695, "grad_norm": 0.34824079275131226, "learning_rate": 0.0002, "epoch": 2.003350083752094, "step": 2990}, {"loss": 1.5853, "grad_norm": 0.3394894003868103, "learning_rate": 0.0002, "epoch": 2.0100502512562812, "step": 3000}, {"loss": 1.5783, "grad_norm": 0.36910977959632874, "learning_rate": 0.0002, "epoch": 2.016750418760469, "step": 3010}, {"loss": 1.6105, "grad_norm": 0.45000967383384705, "learning_rate": 0.0002, "epoch": 2.023450586264657, "step": 3020}, {"loss": 1.6019, "grad_norm": 0.3791407346725464, "learning_rate": 0.0002, "epoch": 2.030150753768844, "step": 3030}, {"loss": 1.5832, "grad_norm": 0.387321799993515, "learning_rate": 0.0002, "epoch": 2.036850921273032, "step": 3040}, {"loss": 1.6834, "grad_norm": 0.4185757040977478, "learning_rate": 0.0002, "epoch": 2.0435510887772192, "step": 3050}, {"loss": 1.5696, "grad_norm": 0.45110777020454407, "learning_rate": 0.0002, "epoch": 2.050251256281407, "step": 3060}, {"loss": 1.6231, "grad_norm": 0.42663660645484924, "learning_rate": 0.0002, "epoch": 2.056951423785595, "step": 3070}, {"loss": 1.6279, "grad_norm": 0.4546292722225189, "learning_rate": 0.0002, "epoch": 2.063651591289782, "step": 3080}, {"loss": 1.6141, "grad_norm": 0.3979759216308594, "learning_rate": 0.0002, "epoch": 2.07035175879397, "step": 3090}, {"loss": 1.6343, "grad_norm": 0.43596673011779785, "learning_rate": 0.0002, "epoch": 2.0770519262981573, "step": 3100}, {"loss": 1.5441, "grad_norm": 0.40120232105255127, "learning_rate": 0.0002, "epoch": 2.083752093802345, "step": 3110}, {"loss": 1.6309, "grad_norm": 0.44449281692504883, "learning_rate": 0.0002, "epoch": 2.090452261306533, "step": 3120}, {"loss": 1.5652, "grad_norm": 0.42672568559646606, "learning_rate": 0.0002, "epoch": 2.09715242881072, "step": 3130}, {"loss": 1.682, "grad_norm": 0.4232690930366516, "learning_rate": 0.0002, "epoch": 2.103852596314908, "step": 3140}, {"loss": 1.624, "grad_norm": 0.4299317002296448, "learning_rate": 0.0002, "epoch": 2.1105527638190953, "step": 3150}, {"loss": 1.6766, "grad_norm": 0.4067758023738861, "learning_rate": 0.0002, "epoch": 2.117252931323283, "step": 3160}, {"loss": 1.6759, "grad_norm": 0.4918815791606903, "learning_rate": 0.0002, "epoch": 2.123953098827471, "step": 3170}, {"loss": 1.6478, "grad_norm": 0.4140559732913971, "learning_rate": 0.0002, "epoch": 2.130653266331658, "step": 3180}, {"loss": 1.6641, "grad_norm": 0.4555995464324951, "learning_rate": 0.0002, "epoch": 2.137353433835846, "step": 3190}, {"loss": 1.5888, "grad_norm": 0.42943915724754333, "learning_rate": 0.0002, "epoch": 2.1440536013400333, "step": 3200}, {"loss": 1.5886, "grad_norm": 0.4730435013771057, "learning_rate": 0.0002, "epoch": 2.150753768844221, "step": 3210}, {"loss": 1.6022, "grad_norm": 0.43310216069221497, "learning_rate": 0.0002, "epoch": 2.157453936348409, "step": 3220}, {"loss": 1.6058, "grad_norm": 0.42054110765457153, "learning_rate": 0.0002, "epoch": 2.164154103852596, "step": 3230}, {"loss": 1.6749, "grad_norm": 0.4897233247756958, "learning_rate": 0.0002, "epoch": 2.170854271356784, "step": 3240}, {"loss": 1.6983, "grad_norm": 0.42194533348083496, "learning_rate": 0.0002, "epoch": 2.1775544388609713, "step": 3250}, {"loss": 1.6159, "grad_norm": 0.44494450092315674, "learning_rate": 0.0002, "epoch": 2.184254606365159, "step": 3260}, {"loss": 1.6977, "grad_norm": 0.43524879217147827, "learning_rate": 0.0002, "epoch": 2.190954773869347, "step": 3270}, {"loss": 1.528, "grad_norm": 0.4621117413043976, "learning_rate": 0.0002, "epoch": 2.1976549413735342, "step": 3280}, {"loss": 1.632, "grad_norm": 0.4073285460472107, "learning_rate": 0.0002, "epoch": 2.204355108877722, "step": 3290}, {"loss": 1.6141, "grad_norm": 0.47868335247039795, "learning_rate": 0.0002, "epoch": 2.2110552763819094, "step": 3300}, {"loss": 1.6857, "grad_norm": 0.4264970123767853, "learning_rate": 0.0002, "epoch": 2.217755443886097, "step": 3310}, {"loss": 1.5653, "grad_norm": 0.4491245150566101, "learning_rate": 0.0002, "epoch": 2.224455611390285, "step": 3320}, {"loss": 1.5881, "grad_norm": 0.4010344445705414, "learning_rate": 0.0002, "epoch": 2.2311557788944723, "step": 3330}, {"loss": 1.6684, "grad_norm": 0.4232759177684784, "learning_rate": 0.0002, "epoch": 2.23785594639866, "step": 3340}, {"loss": 1.6336, "grad_norm": 0.5099776983261108, "learning_rate": 0.0002, "epoch": 2.2445561139028474, "step": 3350}, {"loss": 1.6764, "grad_norm": 0.5223407745361328, "learning_rate": 0.0002, "epoch": 2.251256281407035, "step": 3360}, {"loss": 1.6625, "grad_norm": 0.47818470001220703, "learning_rate": 0.0002, "epoch": 2.257956448911223, "step": 3370}, {"loss": 1.5946, "grad_norm": 0.4721255898475647, "learning_rate": 0.0002, "epoch": 2.2646566164154103, "step": 3380}, {"loss": 1.5568, "grad_norm": 0.4113229513168335, "learning_rate": 0.0002, "epoch": 2.271356783919598, "step": 3390}, {"loss": 1.6494, "grad_norm": 0.507080078125, "learning_rate": 0.0002, "epoch": 2.2780569514237854, "step": 3400}, {"loss": 1.6183, "grad_norm": 0.4852292239665985, "learning_rate": 0.0002, "epoch": 2.284757118927973, "step": 3410}, {"loss": 1.6132, "grad_norm": 0.4503684341907501, "learning_rate": 0.0002, "epoch": 2.291457286432161, "step": 3420}, {"loss": 1.6649, "grad_norm": 0.8359600305557251, "learning_rate": 0.0002, "epoch": 2.2981574539363483, "step": 3430}, {"loss": 1.6644, "grad_norm": 0.44604045152664185, "learning_rate": 0.0002, "epoch": 2.304857621440536, "step": 3440}, {"loss": 1.5972, "grad_norm": 0.45667049288749695, "learning_rate": 0.0002, "epoch": 2.3115577889447234, "step": 3450}, {"loss": 1.6667, "grad_norm": 0.4879349172115326, "learning_rate": 0.0002, "epoch": 2.318257956448911, "step": 3460}, {"loss": 1.5804, "grad_norm": 0.4033963084220886, "learning_rate": 0.0002, "epoch": 2.324958123953099, "step": 3470}, {"loss": 1.5838, "grad_norm": 0.44494301080703735, "learning_rate": 0.0002, "epoch": 2.3316582914572863, "step": 3480}, {"loss": 1.5951, "grad_norm": 0.4794621765613556, "learning_rate": 0.0002, "epoch": 2.338358458961474, "step": 3490}, {"loss": 1.6807, "grad_norm": 0.41404327750205994, "learning_rate": 0.0002, "epoch": 2.3450586264656614, "step": 3500}, {"loss": 1.714, "grad_norm": 0.4664851725101471, "learning_rate": 0.0002, "epoch": 2.351758793969849, "step": 3510}, {"loss": 1.6537, "grad_norm": 0.4263697564601898, "learning_rate": 0.0002, "epoch": 2.358458961474037, "step": 3520}, {"loss": 1.6551, "grad_norm": 0.5035167336463928, "learning_rate": 0.0002, "epoch": 2.3651591289782243, "step": 3530}, {"loss": 1.6208, "grad_norm": 0.4380664527416229, "learning_rate": 0.0002, "epoch": 2.371859296482412, "step": 3540}, {"loss": 1.634, "grad_norm": 0.5227681994438171, "learning_rate": 0.0002, "epoch": 2.3785594639865995, "step": 3550}, {"loss": 1.6146, "grad_norm": 0.4382302761077881, "learning_rate": 0.0002, "epoch": 2.3852596314907872, "step": 3560}, {"loss": 1.5653, "grad_norm": 0.4392451047897339, "learning_rate": 0.0002, "epoch": 2.391959798994975, "step": 3570}, {"loss": 1.6626, "grad_norm": 0.4372786581516266, "learning_rate": 0.0002, "epoch": 2.3986599664991624, "step": 3580}, {"loss": 1.519, "grad_norm": 0.5015502572059631, "learning_rate": 0.0002, "epoch": 2.40536013400335, "step": 3590}, {"loss": 1.588, "grad_norm": 0.5653210878372192, "learning_rate": 0.0002, "epoch": 2.4120603015075375, "step": 3600}, {"loss": 1.6075, "grad_norm": 0.53007972240448, "learning_rate": 0.0002, "epoch": 2.4187604690117253, "step": 3610}, {"loss": 1.6421, "grad_norm": 0.4659176766872406, "learning_rate": 0.0002, "epoch": 2.425460636515913, "step": 3620}, {"loss": 1.625, "grad_norm": 0.5637837052345276, "learning_rate": 0.0002, "epoch": 2.4321608040201004, "step": 3630}, {"loss": 1.6168, "grad_norm": 0.4248391389846802, "learning_rate": 0.0002, "epoch": 2.438860971524288, "step": 3640}, {"loss": 1.6822, "grad_norm": 0.44668248295783997, "learning_rate": 0.0002, "epoch": 2.4455611390284755, "step": 3650}, {"loss": 1.6417, "grad_norm": 0.43990179896354675, "learning_rate": 0.0002, "epoch": 2.4522613065326633, "step": 3660}, {"loss": 1.6723, "grad_norm": 0.4532523453235626, "learning_rate": 0.0002, "epoch": 2.458961474036851, "step": 3670}, {"loss": 1.6957, "grad_norm": 0.6605591773986816, "learning_rate": 0.0002, "epoch": 2.4656616415410384, "step": 3680}, {"loss": 1.6159, "grad_norm": 0.4694533348083496, "learning_rate": 0.0002, "epoch": 2.472361809045226, "step": 3690}, {"loss": 1.6239, "grad_norm": 0.4485011100769043, "learning_rate": 0.0002, "epoch": 2.4790619765494135, "step": 3700}, {"loss": 1.6834, "grad_norm": 0.4761785864830017, "learning_rate": 0.0002, "epoch": 2.4857621440536013, "step": 3710}, {"loss": 1.6313, "grad_norm": 0.5116432309150696, "learning_rate": 0.0002, "epoch": 2.492462311557789, "step": 3720}, {"loss": 1.5054, "grad_norm": 0.49523618817329407, "learning_rate": 0.0002, "epoch": 2.4991624790619764, "step": 3730}, {"loss": 1.6249, "grad_norm": 0.43826380372047424, "learning_rate": 0.0002, "epoch": 2.505862646566164, "step": 3740}, {"loss": 1.5762, "grad_norm": 0.4916154146194458, "learning_rate": 0.0002, "epoch": 2.5125628140703515, "step": 3750}, {"loss": 1.5157, "grad_norm": 0.5381299257278442, "learning_rate": 0.0002, "epoch": 2.5192629815745393, "step": 3760}, {"loss": 1.6467, "grad_norm": 0.44947415590286255, "learning_rate": 0.0002, "epoch": 2.525963149078727, "step": 3770}, {"loss": 1.67, "grad_norm": 0.49979084730148315, "learning_rate": 0.0002, "epoch": 2.5326633165829144, "step": 3780}, {"loss": 1.622, "grad_norm": 0.43046900629997253, "learning_rate": 0.0002, "epoch": 2.539363484087102, "step": 3790}, {"loss": 1.6789, "grad_norm": 0.4513470530509949, "learning_rate": 0.0002, "epoch": 2.5460636515912896, "step": 3800}, {"loss": 1.6335, "grad_norm": 0.49900051951408386, "learning_rate": 0.0002, "epoch": 2.5527638190954773, "step": 3810}, {"loss": 1.6322, "grad_norm": 0.4348420202732086, "learning_rate": 0.0002, "epoch": 2.559463986599665, "step": 3820}, {"loss": 1.6218, "grad_norm": 0.4684867560863495, "learning_rate": 0.0002, "epoch": 2.5661641541038525, "step": 3830}, {"loss": 1.6535, "grad_norm": 0.44430989027023315, "learning_rate": 0.0002, "epoch": 2.5728643216080402, "step": 3840}, {"loss": 1.5909, "grad_norm": 0.47375255823135376, "learning_rate": 0.0002, "epoch": 2.5795644891122276, "step": 3850}, {"loss": 1.6269, "grad_norm": 0.45493075251579285, "learning_rate": 0.0002, "epoch": 2.5862646566164154, "step": 3860}, {"loss": 1.604, "grad_norm": 0.4563275873661041, "learning_rate": 0.0002, "epoch": 2.592964824120603, "step": 3870}, {"loss": 1.642, "grad_norm": 0.46060335636138916, "learning_rate": 0.0002, "epoch": 2.5996649916247905, "step": 3880}, {"loss": 1.6302, "grad_norm": 0.4718867540359497, "learning_rate": 0.0002, "epoch": 2.6063651591289783, "step": 3890}, {"loss": 1.6242, "grad_norm": 0.41570305824279785, "learning_rate": 0.0002, "epoch": 2.6130653266331656, "step": 3900}, {"loss": 1.6401, "grad_norm": 0.4603121876716614, "learning_rate": 0.0002, "epoch": 2.6197654941373534, "step": 3910}, {"loss": 1.6839, "grad_norm": 0.4734652638435364, "learning_rate": 0.0002, "epoch": 2.626465661641541, "step": 3920}, {"loss": 1.5448, "grad_norm": 0.45348483324050903, "learning_rate": 0.0002, "epoch": 2.6331658291457285, "step": 3930}, {"loss": 1.6157, "grad_norm": 0.46559447050094604, "learning_rate": 0.0002, "epoch": 2.6398659966499163, "step": 3940}, {"loss": 1.7052, "grad_norm": 0.44113144278526306, "learning_rate": 0.0002, "epoch": 2.6465661641541036, "step": 3950}, {"loss": 1.6315, "grad_norm": 0.41415104269981384, "learning_rate": 0.0002, "epoch": 2.6532663316582914, "step": 3960}, {"loss": 1.6589, "grad_norm": 0.48868080973625183, "learning_rate": 0.0002, "epoch": 2.659966499162479, "step": 3970}, {"loss": 1.6211, "grad_norm": 0.49610549211502075, "learning_rate": 0.0002, "epoch": 2.6666666666666665, "step": 3980}, {"loss": 1.6235, "grad_norm": 0.4309130907058716, "learning_rate": 0.0002, "epoch": 2.6733668341708543, "step": 3990}, {"loss": 1.6452, "grad_norm": 0.4489327669143677, "learning_rate": 0.0002, "epoch": 2.6800670016750416, "step": 4000}, {"loss": 1.5744, "grad_norm": 0.5380139946937561, "learning_rate": 0.0002, "epoch": 2.6867671691792294, "step": 4010}, {"loss": 1.6524, "grad_norm": 0.5076672434806824, "learning_rate": 0.0002, "epoch": 2.693467336683417, "step": 4020}, {"loss": 1.636, "grad_norm": 0.47620031237602234, "learning_rate": 0.0002, "epoch": 2.7001675041876045, "step": 4030}, {"loss": 1.5543, "grad_norm": 0.48089155554771423, "learning_rate": 0.0002, "epoch": 2.7068676716917923, "step": 4040}, {"loss": 1.6396, "grad_norm": 0.5108814239501953, "learning_rate": 0.0002, "epoch": 2.7135678391959797, "step": 4050}, {"loss": 1.5905, "grad_norm": 0.4196513295173645, "learning_rate": 0.0002, "epoch": 2.7202680067001674, "step": 4060}, {"loss": 1.686, "grad_norm": 0.4574664831161499, "learning_rate": 0.0002, "epoch": 2.726968174204355, "step": 4070}, {"loss": 1.6234, "grad_norm": 0.4671640992164612, "learning_rate": 0.0002, "epoch": 2.7336683417085426, "step": 4080}, {"loss": 1.6827, "grad_norm": 0.49355530738830566, "learning_rate": 0.0002, "epoch": 2.7403685092127303, "step": 4090}, {"loss": 1.6999, "grad_norm": 0.46716663241386414, "learning_rate": 0.0002, "epoch": 2.7470686767169177, "step": 4100}, {"loss": 1.6463, "grad_norm": 0.45420581102371216, "learning_rate": 0.0002, "epoch": 2.7537688442211055, "step": 4110}, {"loss": 1.5718, "grad_norm": 0.4680487811565399, "learning_rate": 0.0002, "epoch": 2.7604690117252932, "step": 4120}, {"loss": 1.5968, "grad_norm": 0.5375032424926758, "learning_rate": 0.0002, "epoch": 2.7671691792294806, "step": 4130}, {"loss": 1.5254, "grad_norm": 0.46026280522346497, "learning_rate": 0.0002, "epoch": 2.7738693467336684, "step": 4140}, {"loss": 1.6613, "grad_norm": 0.43658447265625, "learning_rate": 0.0002, "epoch": 2.7805695142378557, "step": 4150}, {"loss": 1.6546, "grad_norm": 0.4935547113418579, "learning_rate": 0.0002, "epoch": 2.7872696817420435, "step": 4160}, {"loss": 1.5961, "grad_norm": 0.8167962431907654, "learning_rate": 0.0002, "epoch": 2.7939698492462313, "step": 4170}, {"loss": 1.6907, "grad_norm": 0.4289683997631073, "learning_rate": 0.0002, "epoch": 2.8006700167504186, "step": 4180}, {"loss": 1.6385, "grad_norm": 0.4569324254989624, "learning_rate": 0.0002, "epoch": 2.8073701842546064, "step": 4190}, {"loss": 1.6077, "grad_norm": 0.474795937538147, "learning_rate": 0.0002, "epoch": 2.8140703517587937, "step": 4200}, {"loss": 1.6223, "grad_norm": 0.44272229075431824, "learning_rate": 0.0002, "epoch": 2.8207705192629815, "step": 4210}, {"loss": 1.6706, "grad_norm": 0.525240957736969, "learning_rate": 0.0002, "epoch": 2.8274706867671693, "step": 4220}, {"loss": 1.7196, "grad_norm": 0.4802303910255432, "learning_rate": 0.0002, "epoch": 2.8341708542713566, "step": 4230}, {"loss": 1.6002, "grad_norm": 0.46400442719459534, "learning_rate": 0.0002, "epoch": 2.8408710217755444, "step": 4240}, {"loss": 1.6052, "grad_norm": 0.49884888529777527, "learning_rate": 0.0002, "epoch": 2.8475711892797317, "step": 4250}, {"loss": 1.6919, "grad_norm": 0.5015072226524353, "learning_rate": 0.0002, "epoch": 2.8542713567839195, "step": 4260}, {"loss": 1.6335, "grad_norm": 0.4335440695285797, "learning_rate": 0.0002, "epoch": 2.8609715242881073, "step": 4270}, {"loss": 1.5664, "grad_norm": 0.5131644606590271, "learning_rate": 0.0002, "epoch": 2.8676716917922946, "step": 4280}, {"loss": 1.6409, "grad_norm": 0.6977195739746094, "learning_rate": 0.0002, "epoch": 2.8743718592964824, "step": 4290}, {"loss": 1.7192, "grad_norm": 0.5133762955665588, "learning_rate": 0.0002, "epoch": 2.8810720268006698, "step": 4300}, {"loss": 1.6257, "grad_norm": 0.4737614393234253, "learning_rate": 0.0002, "epoch": 2.8877721943048575, "step": 4310}, {"loss": 1.6076, "grad_norm": 0.4580535590648651, "learning_rate": 0.0002, "epoch": 2.8944723618090453, "step": 4320}, {"loss": 1.6538, "grad_norm": 0.43863341212272644, "learning_rate": 0.0002, "epoch": 2.901172529313233, "step": 4330}, {"loss": 1.6091, "grad_norm": 0.4103737473487854, "learning_rate": 0.0002, "epoch": 2.9078726968174204, "step": 4340}, {"loss": 1.7106, "grad_norm": 0.438014417886734, "learning_rate": 0.0002, "epoch": 2.914572864321608, "step": 4350}, {"loss": 1.6025, "grad_norm": 0.5068213939666748, "learning_rate": 0.0002, "epoch": 2.9212730318257956, "step": 4360}, {"loss": 1.6426, "grad_norm": 0.45305484533309937, "learning_rate": 0.0002, "epoch": 2.9279731993299833, "step": 4370}, {"loss": 1.5726, "grad_norm": 0.4612090289592743, "learning_rate": 0.0002, "epoch": 2.934673366834171, "step": 4380}, {"loss": 1.6536, "grad_norm": 0.508736789226532, "learning_rate": 0.0002, "epoch": 2.9413735343383585, "step": 4390}, {"loss": 1.6132, "grad_norm": 0.4924427270889282, "learning_rate": 0.0002, "epoch": 2.948073701842546, "step": 4400}, {"loss": 1.7007, "grad_norm": 0.5707460641860962, "learning_rate": 0.0002, "epoch": 2.9547738693467336, "step": 4410}, {"loss": 1.6814, "grad_norm": 0.42270299792289734, "learning_rate": 0.0002, "epoch": 2.9614740368509214, "step": 4420}, {"loss": 1.6644, "grad_norm": 0.4429931044578552, "learning_rate": 0.0002, "epoch": 2.968174204355109, "step": 4430}, {"loss": 1.6251, "grad_norm": 0.49760574102401733, "learning_rate": 0.0002, "epoch": 2.9748743718592965, "step": 4440}, {"loss": 1.6169, "grad_norm": 0.4558229148387909, "learning_rate": 0.0002, "epoch": 2.981574539363484, "step": 4450}, {"loss": 1.6055, "grad_norm": 0.39848530292510986, "learning_rate": 0.0002, "epoch": 2.9882747068676716, "step": 4460}, {"loss": 1.6705, "grad_norm": 0.5224862098693848, "learning_rate": 0.0002, "epoch": 2.9949748743718594, "step": 4470}, {"eval_loss": 1.8228833675384521, "eval_runtime": 37.9049, "eval_samples_per_second": 13.587, "eval_steps_per_second": 1.715, "epoch": 2.9996649916247904, "step": 4477}, {"loss": 1.6637, "grad_norm": 0.41169142723083496, "learning_rate": 0.0002, "epoch": 3.0016750418760467, "step": 4480}, {"loss": 1.5974, "grad_norm": 0.4865207374095917, "learning_rate": 0.0002, "epoch": 3.0083752093802345, "step": 4490}, {"loss": 1.5297, "grad_norm": 0.5462028384208679, "learning_rate": 0.0002, "epoch": 3.0150753768844223, "step": 4500}, {"loss": 1.5251, "grad_norm": 0.6169732809066772, "learning_rate": 0.0002, "epoch": 3.0217755443886096, "step": 4510}, {"loss": 1.5559, "grad_norm": 0.5667954087257385, "learning_rate": 0.0002, "epoch": 3.0284757118927974, "step": 4520}, {"loss": 1.5037, "grad_norm": 0.5758325457572937, "learning_rate": 0.0002, "epoch": 3.0351758793969847, "step": 4530}, {"loss": 1.4873, "grad_norm": 0.5220064520835876, "learning_rate": 0.0002, "epoch": 3.0418760469011725, "step": 4540}, {"loss": 1.5126, "grad_norm": 0.5469558835029602, "learning_rate": 0.0002, "epoch": 3.0485762144053603, "step": 4550}, {"loss": 1.4275, "grad_norm": 0.5680848956108093, "learning_rate": 0.0002, "epoch": 3.0552763819095476, "step": 4560}, {"loss": 1.5187, "grad_norm": 0.5906574726104736, "learning_rate": 0.0002, "epoch": 3.0619765494137354, "step": 4570}, {"loss": 1.4551, "grad_norm": 0.4725631773471832, "learning_rate": 0.0002, "epoch": 3.0686767169179228, "step": 4580}, {"loss": 1.5083, "grad_norm": 0.5273477435112, "learning_rate": 0.0002, "epoch": 3.0753768844221105, "step": 4590}, {"loss": 1.5154, "grad_norm": 0.5861203074455261, "learning_rate": 0.0002, "epoch": 3.0820770519262983, "step": 4600}, {"loss": 1.4924, "grad_norm": 0.5343965291976929, "learning_rate": 0.0002, "epoch": 3.0887772194304857, "step": 4610}, {"loss": 1.5608, "grad_norm": 0.5348150730133057, "learning_rate": 0.0002, "epoch": 3.0954773869346734, "step": 4620}, {"loss": 1.5399, "grad_norm": 0.5971846580505371, "learning_rate": 0.0002, "epoch": 3.102177554438861, "step": 4630}, {"loss": 1.4662, "grad_norm": 0.5203177332878113, "learning_rate": 0.0002, "epoch": 3.1088777219430486, "step": 4640}, {"loss": 1.5805, "grad_norm": 0.55289226770401, "learning_rate": 0.0002, "epoch": 3.1155778894472363, "step": 4650}, {"loss": 1.4745, "grad_norm": 0.6878530979156494, "learning_rate": 0.0002, "epoch": 3.1222780569514237, "step": 4660}, {"loss": 1.5335, "grad_norm": 0.6173256635665894, "learning_rate": 0.0002, "epoch": 3.1289782244556115, "step": 4670}, {"loss": 1.51, "grad_norm": 0.536796510219574, "learning_rate": 0.0002, "epoch": 3.135678391959799, "step": 4680}, {"loss": 1.4713, "grad_norm": 0.58846116065979, "learning_rate": 0.0002, "epoch": 3.1423785594639866, "step": 4690}, {"loss": 1.5114, "grad_norm": 0.645889401435852, "learning_rate": 0.0002, "epoch": 3.1490787269681744, "step": 4700}, {"loss": 1.4705, "grad_norm": 0.6118691563606262, "learning_rate": 0.0002, "epoch": 3.1557788944723617, "step": 4710}, {"loss": 1.5533, "grad_norm": 0.5189669132232666, "learning_rate": 0.0002, "epoch": 3.1624790619765495, "step": 4720}, {"loss": 1.4769, "grad_norm": 0.5794713497161865, "learning_rate": 0.0002, "epoch": 3.169179229480737, "step": 4730}, {"loss": 1.4849, "grad_norm": 0.6579326391220093, "learning_rate": 0.0002, "epoch": 3.1758793969849246, "step": 4740}, {"loss": 1.545, "grad_norm": 0.5822742581367493, "learning_rate": 0.0002, "epoch": 3.1825795644891124, "step": 4750}, {"loss": 1.4358, "grad_norm": 0.5475956201553345, "learning_rate": 0.0002, "epoch": 3.1892797319932997, "step": 4760}, {"loss": 1.4723, "grad_norm": 0.6743834018707275, "learning_rate": 0.0002, "epoch": 3.1959798994974875, "step": 4770}, {"loss": 1.5161, "grad_norm": 0.6110585927963257, "learning_rate": 0.0002, "epoch": 3.202680067001675, "step": 4780}, {"loss": 1.5455, "grad_norm": 0.5426181554794312, "learning_rate": 0.0002, "epoch": 3.2093802345058626, "step": 4790}, {"loss": 1.5315, "grad_norm": 0.6077824234962463, "learning_rate": 0.0002, "epoch": 3.2160804020100504, "step": 4800}, {"loss": 1.5314, "grad_norm": 0.5785858631134033, "learning_rate": 0.0002, "epoch": 3.2227805695142377, "step": 4810}, {"loss": 1.4041, "grad_norm": 0.6425958275794983, "learning_rate": 0.0002, "epoch": 3.2294807370184255, "step": 4820}, {"loss": 1.4751, "grad_norm": 0.6607080698013306, "learning_rate": 0.0002, "epoch": 3.236180904522613, "step": 4830}, {"loss": 1.5267, "grad_norm": 0.5385788679122925, "learning_rate": 0.0002, "epoch": 3.2428810720268006, "step": 4840}, {"loss": 1.4673, "grad_norm": 0.5630403757095337, "learning_rate": 0.0002, "epoch": 3.2495812395309884, "step": 4850}, {"loss": 1.5257, "grad_norm": 0.6340779662132263, "learning_rate": 0.0002, "epoch": 3.2562814070351758, "step": 4860}, {"loss": 1.5148, "grad_norm": 0.5305342674255371, "learning_rate": 0.0002, "epoch": 3.2629815745393635, "step": 4870}, {"loss": 1.5162, "grad_norm": 0.597670316696167, "learning_rate": 0.0002, "epoch": 3.2696817420435513, "step": 4880}, {"loss": 1.5429, "grad_norm": 0.665553867816925, "learning_rate": 0.0002, "epoch": 3.2763819095477387, "step": 4890}, {"loss": 1.4607, "grad_norm": 0.579767644405365, "learning_rate": 0.0002, "epoch": 3.2830820770519265, "step": 4900}, {"loss": 1.4999, "grad_norm": 0.5512481331825256, "learning_rate": 0.0002, "epoch": 3.289782244556114, "step": 4910}, {"loss": 1.5022, "grad_norm": 0.5916532278060913, "learning_rate": 0.0002, "epoch": 3.2964824120603016, "step": 4920}, {"loss": 1.4889, "grad_norm": 0.7521726489067078, "learning_rate": 0.0002, "epoch": 3.3031825795644894, "step": 4930}, {"loss": 1.4223, "grad_norm": 0.5352797508239746, "learning_rate": 0.0002, "epoch": 3.3098827470686767, "step": 4940}, {"loss": 1.5122, "grad_norm": 0.5950371623039246, "learning_rate": 0.0002, "epoch": 3.3165829145728645, "step": 4950}, {"loss": 1.5072, "grad_norm": 0.8020477890968323, "learning_rate": 0.0002, "epoch": 3.323283082077052, "step": 4960}, {"loss": 1.5422, "grad_norm": 0.6790024638175964, "learning_rate": 0.0002, "epoch": 3.3299832495812396, "step": 4970}, {"loss": 1.5363, "grad_norm": 0.687627375125885, "learning_rate": 0.0002, "epoch": 3.3366834170854274, "step": 4980}, {"loss": 1.5276, "grad_norm": 0.6094385385513306, "learning_rate": 0.0002, "epoch": 3.3433835845896147, "step": 4990}, {"loss": 1.549, "grad_norm": 0.6541242003440857, "learning_rate": 0.0002, "epoch": 3.3500837520938025, "step": 5000}, {"loss": 1.6067, "grad_norm": 0.5560880303382874, "learning_rate": 0.0002, "epoch": 3.35678391959799, "step": 5010}, {"loss": 1.5769, "grad_norm": 0.5440094470977783, "learning_rate": 0.0002, "epoch": 3.3634840871021776, "step": 5020}, {"loss": 1.6183, "grad_norm": 0.5749301314353943, "learning_rate": 0.0002, "epoch": 3.3701842546063654, "step": 5030}, {"loss": 1.4801, "grad_norm": 0.5919716954231262, "learning_rate": 0.0002, "epoch": 3.3768844221105527, "step": 5040}, {"loss": 1.5957, "grad_norm": 0.6331481337547302, "learning_rate": 0.0002, "epoch": 3.3835845896147405, "step": 5050}, {"loss": 1.5188, "grad_norm": 0.5687161684036255, "learning_rate": 0.0002, "epoch": 3.390284757118928, "step": 5060}, {"loss": 1.5702, "grad_norm": 0.6718577742576599, "learning_rate": 0.0002, "epoch": 3.3969849246231156, "step": 5070}, {"loss": 1.5577, "grad_norm": 0.5089324116706848, "learning_rate": 0.0002, "epoch": 3.4036850921273034, "step": 5080}, {"loss": 1.512, "grad_norm": 0.5710174441337585, "learning_rate": 0.0002, "epoch": 3.4103852596314908, "step": 5090}, {"loss": 1.5492, "grad_norm": 0.6670721173286438, "learning_rate": 0.0002, "epoch": 3.4170854271356785, "step": 5100}, {"loss": 1.5227, "grad_norm": 0.6875665187835693, "learning_rate": 0.0002, "epoch": 3.423785594639866, "step": 5110}, {"loss": 1.4496, "grad_norm": 0.5375880599021912, "learning_rate": 0.0002, "epoch": 3.4304857621440537, "step": 5120}, {"loss": 1.5527, "grad_norm": 0.6550399661064148, "learning_rate": 0.0002, "epoch": 3.4371859296482414, "step": 5130}, {"loss": 1.5687, "grad_norm": 0.5948067903518677, "learning_rate": 0.0002, "epoch": 3.4438860971524288, "step": 5140}, {"loss": 1.4813, "grad_norm": 0.6134477257728577, "learning_rate": 0.0002, "epoch": 3.4505862646566166, "step": 5150}, {"loss": 1.5069, "grad_norm": 0.6506398320198059, "learning_rate": 0.0002, "epoch": 3.457286432160804, "step": 5160}, {"loss": 1.4422, "grad_norm": 0.6060147881507874, "learning_rate": 0.0002, "epoch": 3.4639865996649917, "step": 5170}, {"loss": 1.5093, "grad_norm": 0.6173806190490723, "learning_rate": 0.0002, "epoch": 3.4706867671691795, "step": 5180}, {"loss": 1.4975, "grad_norm": 0.6032607555389404, "learning_rate": 0.0002, "epoch": 3.477386934673367, "step": 5190}, {"loss": 1.4979, "grad_norm": 0.5652492046356201, "learning_rate": 0.0002, "epoch": 3.4840871021775546, "step": 5200}, {"loss": 1.4883, "grad_norm": 0.6168607473373413, "learning_rate": 0.0002, "epoch": 3.490787269681742, "step": 5210}, {"loss": 1.5164, "grad_norm": 0.6170629262924194, "learning_rate": 0.0002, "epoch": 3.4974874371859297, "step": 5220}, {"loss": 1.4879, "grad_norm": 0.6926297545433044, "learning_rate": 0.0002, "epoch": 3.5041876046901175, "step": 5230}, {"loss": 1.4982, "grad_norm": 0.6702437996864319, "learning_rate": 0.0002, "epoch": 3.510887772194305, "step": 5240}, {"loss": 1.4986, "grad_norm": 0.5421436429023743, "learning_rate": 0.0002, "epoch": 3.5175879396984926, "step": 5250}, {"loss": 1.4673, "grad_norm": 0.5726765990257263, "learning_rate": 0.0002, "epoch": 3.52428810720268, "step": 5260}, {"loss": 1.5423, "grad_norm": 0.5685455203056335, "learning_rate": 0.0002, "epoch": 3.5309882747068677, "step": 5270}, {"loss": 1.4715, "grad_norm": 0.6018396019935608, "learning_rate": 0.0002, "epoch": 3.5376884422110555, "step": 5280}, {"loss": 1.5451, "grad_norm": 0.5731932520866394, "learning_rate": 0.0002, "epoch": 3.544388609715243, "step": 5290}, {"loss": 1.4752, "grad_norm": 0.6601519584655762, "learning_rate": 0.0002, "epoch": 3.5510887772194306, "step": 5300}, {"loss": 1.5434, "grad_norm": 0.5545530319213867, "learning_rate": 0.0002, "epoch": 3.557788944723618, "step": 5310}, {"loss": 1.5438, "grad_norm": 0.5998541116714478, "learning_rate": 0.0002, "epoch": 3.5644891122278057, "step": 5320}, {"loss": 1.56, "grad_norm": 0.5651767253875732, "learning_rate": 0.0002, "epoch": 3.5711892797319935, "step": 5330}, {"loss": 1.4829, "grad_norm": 0.7425084114074707, "learning_rate": 0.0002, "epoch": 3.577889447236181, "step": 5340}, {"loss": 1.5571, "grad_norm": 0.5770602226257324, "learning_rate": 0.0002, "epoch": 3.5845896147403686, "step": 5350}, {"loss": 1.458, "grad_norm": 0.54723060131073, "learning_rate": 0.0002, "epoch": 3.591289782244556, "step": 5360}, {"loss": 1.497, "grad_norm": 0.6658238172531128, "learning_rate": 0.0002, "epoch": 3.5979899497487438, "step": 5370}, {"loss": 1.5456, "grad_norm": 0.5787645578384399, "learning_rate": 0.0002, "epoch": 3.6046901172529315, "step": 5380}, {"loss": 1.5343, "grad_norm": 0.594913125038147, "learning_rate": 0.0002, "epoch": 3.611390284757119, "step": 5390}, {"loss": 1.4727, "grad_norm": 0.4964977502822876, "learning_rate": 0.0002, "epoch": 3.6180904522613067, "step": 5400}, {"loss": 1.5062, "grad_norm": 0.6087527275085449, "learning_rate": 0.0002, "epoch": 3.624790619765494, "step": 5410}, {"loss": 1.5098, "grad_norm": 0.6315323710441589, "learning_rate": 0.0002, "epoch": 3.6314907872696818, "step": 5420}, {"loss": 1.4855, "grad_norm": 0.574799120426178, "learning_rate": 0.0002, "epoch": 3.6381909547738696, "step": 5430}, {"loss": 1.4595, "grad_norm": 0.5949277877807617, "learning_rate": 0.0002, "epoch": 3.644891122278057, "step": 5440}, {"loss": 1.4816, "grad_norm": 0.5640677213668823, "learning_rate": 0.0002, "epoch": 3.6515912897822447, "step": 5450}, {"loss": 1.525, "grad_norm": 0.6198237538337708, "learning_rate": 0.0002, "epoch": 3.658291457286432, "step": 5460}, {"loss": 1.5676, "grad_norm": 0.6902034878730774, "learning_rate": 0.0002, "epoch": 3.66499162479062, "step": 5470}, {"loss": 1.544, "grad_norm": 0.5686674118041992, "learning_rate": 0.0002, "epoch": 3.6716917922948076, "step": 5480}, {"loss": 1.5255, "grad_norm": 0.6532107591629028, "learning_rate": 0.0002, "epoch": 3.678391959798995, "step": 5490}, {"loss": 1.5767, "grad_norm": 0.5790849924087524, "learning_rate": 0.0002, "epoch": 3.6850921273031827, "step": 5500}, {"loss": 1.5507, "grad_norm": 0.6055065393447876, "learning_rate": 0.0002, "epoch": 3.69179229480737, "step": 5510}, {"loss": 1.4656, "grad_norm": 0.5630605816841125, "learning_rate": 0.0002, "epoch": 3.698492462311558, "step": 5520}, {"loss": 1.537, "grad_norm": 0.6005825996398926, "learning_rate": 0.0002, "epoch": 3.7051926298157456, "step": 5530}, {"loss": 1.5313, "grad_norm": 0.6553038954734802, "learning_rate": 0.0002, "epoch": 3.711892797319933, "step": 5540}, {"loss": 1.4943, "grad_norm": 0.5601094961166382, "learning_rate": 0.0002, "epoch": 3.7185929648241207, "step": 5550}, {"loss": 1.5161, "grad_norm": 0.6598808169364929, "learning_rate": 0.0002, "epoch": 3.725293132328308, "step": 5560}, {"loss": 1.5345, "grad_norm": 0.5506255626678467, "learning_rate": 0.0002, "epoch": 3.731993299832496, "step": 5570}, {"loss": 1.4805, "grad_norm": 0.6001223921775818, "learning_rate": 0.0002, "epoch": 3.7386934673366836, "step": 5580}, {"loss": 1.4652, "grad_norm": 0.6287297606468201, "learning_rate": 0.0002, "epoch": 3.745393634840871, "step": 5590}, {"loss": 1.5246, "grad_norm": 0.6253238916397095, "learning_rate": 0.0002, "epoch": 3.7520938023450587, "step": 5600}, {"loss": 1.5691, "grad_norm": 0.5713174939155579, "learning_rate": 0.0002, "epoch": 3.758793969849246, "step": 5610}, {"loss": 1.5661, "grad_norm": 0.6198310852050781, "learning_rate": 0.0002, "epoch": 3.765494137353434, "step": 5620}, {"loss": 1.5448, "grad_norm": 0.5941224098205566, "learning_rate": 0.0002, "epoch": 3.7721943048576216, "step": 5630}, {"loss": 1.4925, "grad_norm": 0.606002151966095, "learning_rate": 0.0002, "epoch": 3.778894472361809, "step": 5640}, {"loss": 1.5182, "grad_norm": 0.6540704965591431, "learning_rate": 0.0002, "epoch": 3.7855946398659968, "step": 5650}, {"loss": 1.5903, "grad_norm": 0.6147415041923523, "learning_rate": 0.0002, "epoch": 3.792294807370184, "step": 5660}, {"loss": 1.5329, "grad_norm": 0.5649605393409729, "learning_rate": 0.0002, "epoch": 3.798994974874372, "step": 5670}, {"loss": 1.5747, "grad_norm": 0.6788773536682129, "learning_rate": 0.0002, "epoch": 3.8056951423785597, "step": 5680}, {"loss": 1.535, "grad_norm": 0.6581860780715942, "learning_rate": 0.0002, "epoch": 3.812395309882747, "step": 5690}, {"loss": 1.4587, "grad_norm": 0.5529348850250244, "learning_rate": 0.0002, "epoch": 3.819095477386935, "step": 5700}, {"loss": 1.5594, "grad_norm": 0.6320232152938843, "learning_rate": 0.0002, "epoch": 3.825795644891122, "step": 5710}, {"loss": 1.4696, "grad_norm": 0.6529698371887207, "learning_rate": 0.0002, "epoch": 3.83249581239531, "step": 5720}, {"loss": 1.5854, "grad_norm": 0.5983362793922424, "learning_rate": 0.0002, "epoch": 3.8391959798994977, "step": 5730}, {"loss": 1.465, "grad_norm": 0.6335684061050415, "learning_rate": 0.0002, "epoch": 3.845896147403685, "step": 5740}, {"loss": 1.5545, "grad_norm": 0.700446605682373, "learning_rate": 0.0002, "epoch": 3.852596314907873, "step": 5750}, {"loss": 1.5707, "grad_norm": 0.6092597842216492, "learning_rate": 0.0002, "epoch": 3.85929648241206, "step": 5760}, {"loss": 1.5729, "grad_norm": 0.564146101474762, "learning_rate": 0.0002, "epoch": 3.865996649916248, "step": 5770}, {"loss": 1.5872, "grad_norm": 0.615275502204895, "learning_rate": 0.0002, "epoch": 3.8726968174204357, "step": 5780}, {"loss": 1.5142, "grad_norm": 0.6685376763343811, "learning_rate": 0.0002, "epoch": 3.879396984924623, "step": 5790}, {"loss": 1.4752, "grad_norm": 0.6116922497749329, "learning_rate": 0.0002, "epoch": 3.886097152428811, "step": 5800}, {"loss": 1.5179, "grad_norm": 0.5486813187599182, "learning_rate": 0.0002, "epoch": 3.892797319932998, "step": 5810}, {"loss": 1.5167, "grad_norm": 0.6208204030990601, "learning_rate": 0.0002, "epoch": 3.899497487437186, "step": 5820}, {"loss": 1.5334, "grad_norm": 0.6500625014305115, "learning_rate": 0.0002, "epoch": 3.9061976549413737, "step": 5830}, {"loss": 1.4716, "grad_norm": 0.5948089361190796, "learning_rate": 0.0002, "epoch": 3.912897822445561, "step": 5840}, {"loss": 1.6011, "grad_norm": 0.7210732698440552, "learning_rate": 0.0002, "epoch": 3.919597989949749, "step": 5850}, {"loss": 1.5519, "grad_norm": 0.6662322878837585, "learning_rate": 0.0002, "epoch": 3.926298157453936, "step": 5860}, {"loss": 1.5656, "grad_norm": 0.5613839626312256, "learning_rate": 0.0002, "epoch": 3.932998324958124, "step": 5870}, {"loss": 1.544, "grad_norm": 0.6069002151489258, "learning_rate": 0.0002, "epoch": 3.9396984924623117, "step": 5880}, {"loss": 1.6745, "grad_norm": 0.7075562477111816, "learning_rate": 0.0002, "epoch": 3.946398659966499, "step": 5890}, {"loss": 1.5391, "grad_norm": 0.6316173076629639, "learning_rate": 0.0002, "epoch": 3.953098827470687, "step": 5900}, {"loss": 1.6314, "grad_norm": 0.5716308355331421, "learning_rate": 0.0002, "epoch": 3.959798994974874, "step": 5910}, {"loss": 1.5947, "grad_norm": 0.6800096035003662, "learning_rate": 0.0002, "epoch": 3.966499162479062, "step": 5920}, {"loss": 1.5189, "grad_norm": 0.6057983040809631, "learning_rate": 0.0002, "epoch": 3.9731993299832498, "step": 5930}, {"loss": 1.5431, "grad_norm": 0.5938987731933594, "learning_rate": 0.0002, "epoch": 3.979899497487437, "step": 5940}, {"loss": 1.5111, "grad_norm": 0.6963576674461365, "learning_rate": 0.0002, "epoch": 3.986599664991625, "step": 5950}, {"loss": 1.5521, "grad_norm": 0.6279940009117126, "learning_rate": 0.0002, "epoch": 3.993299832495812, "step": 5960}, {"loss": 1.5974, "grad_norm": 0.7161159515380859, "learning_rate": 0.0002, "epoch": 4.0, "step": 5970}]}
+{"epoch": 4.99966499162479, "step": 7462, "epoch_duration": 1615.0195803642273, "total_accumulated_duration": 7957.4429659843445, "gpu_info": {"GPU_0": "NVIDIA A100-SXM4-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 7672.4541015625}, "peak_memory_usage": {"GPU_0": 13792.75439453125}, "avg_memory_reserved": {"GPU_0": 17416.0}, "peak_memory_reserved": {"GPU_0": 17416.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "outputs-001/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-2985", "params": {"epochs": 8, "batch_size": 1, "learning_rate": 0.0002, "gradient_accumulation_steps": 2, "warmup_ratio": 0.03, "max_grad_norm": 0.3, "lora_alpha": 32, "lora_dropout": 0.05, "lora_r": 64, "tf32": true, "seed": 42}, "log_history": [{"loss": 2.6189, "grad_norm": 0.565915048122406, "learning_rate": 0.0002, "epoch": 0.006700167504187605, "step": 10}, {"loss": 2.3162, "grad_norm": 0.5004463791847229, "learning_rate": 0.0002, "epoch": 0.01340033500837521, "step": 20}, {"loss": 2.0576, "grad_norm": 0.511043906211853, "learning_rate": 0.0002, "epoch": 0.020100502512562814, "step": 30}, {"loss": 2.0085, "grad_norm": 0.47327178716659546, "learning_rate": 0.0002, "epoch": 0.02680067001675042, "step": 40}, {"loss": 2.0276, "grad_norm": 0.5511676669120789, "learning_rate": 0.0002, "epoch": 0.03350083752093802, "step": 50}, {"loss": 1.9075, "grad_norm": 0.4666278064250946, "learning_rate": 0.0002, "epoch": 0.04020100502512563, "step": 60}, {"loss": 1.8413, "grad_norm": 0.5310961008071899, "learning_rate": 0.0002, "epoch": 0.04690117252931323, "step": 70}, {"loss": 1.8711, "grad_norm": 0.5606027245521545, "learning_rate": 0.0002, "epoch": 0.05360134003350084, "step": 80}, {"loss": 1.9282, "grad_norm": 0.4934779703617096, "learning_rate": 0.0002, "epoch": 0.06030150753768844, "step": 90}, {"loss": 1.8925, "grad_norm": 0.4821869730949402, "learning_rate": 0.0002, "epoch": 0.06700167504187604, "step": 100}, {"loss": 1.8628, "grad_norm": 0.5262084603309631, "learning_rate": 0.0002, "epoch": 0.07370184254606366, "step": 110}, {"loss": 1.8347, "grad_norm": 0.3774230182170868, "learning_rate": 0.0002, "epoch": 0.08040201005025126, "step": 120}, {"loss": 1.8386, "grad_norm": 0.34137430787086487, "learning_rate": 0.0002, "epoch": 0.08710217755443886, "step": 130}, {"loss": 1.861, "grad_norm": 0.407272070646286, "learning_rate": 0.0002, "epoch": 0.09380234505862646, "step": 140}, {"loss": 1.8279, "grad_norm": 0.4011937975883484, "learning_rate": 0.0002, "epoch": 0.10050251256281408, "step": 150}, {"loss": 1.9317, "grad_norm": 0.4432467222213745, "learning_rate": 0.0002, "epoch": 0.10720268006700168, "step": 160}, {"loss": 1.8157, "grad_norm": 0.44030463695526123, "learning_rate": 0.0002, "epoch": 0.11390284757118928, "step": 170}, {"loss": 1.8534, "grad_norm": 0.3799569308757782, "learning_rate": 0.0002, "epoch": 0.12060301507537688, "step": 180}, {"loss": 1.7658, "grad_norm": 0.33721521496772766, "learning_rate": 0.0002, "epoch": 0.1273031825795645, "step": 190}, {"loss": 1.8269, "grad_norm": 0.4096226692199707, "learning_rate": 0.0002, "epoch": 0.13400335008375208, "step": 200}, {"loss": 1.802, "grad_norm": 0.37374693155288696, "learning_rate": 0.0002, "epoch": 0.1407035175879397, "step": 210}, {"loss": 1.8901, "grad_norm": 0.3249480128288269, "learning_rate": 0.0002, "epoch": 0.1474036850921273, "step": 220}, {"loss": 1.8163, "grad_norm": 0.3612042963504791, "learning_rate": 0.0002, "epoch": 0.1541038525963149, "step": 230}, {"loss": 1.7585, "grad_norm": 0.3686671257019043, "learning_rate": 0.0002, "epoch": 0.16080402010050251, "step": 240}, {"loss": 1.8365, "grad_norm": 0.3521044850349426, "learning_rate": 0.0002, "epoch": 0.16750418760469013, "step": 250}, {"loss": 1.8623, "grad_norm": 0.4073677361011505, "learning_rate": 0.0002, "epoch": 0.17420435510887772, "step": 260}, {"loss": 1.8026, "grad_norm": 0.34522193670272827, "learning_rate": 0.0002, "epoch": 0.18090452261306533, "step": 270}, {"loss": 1.8162, "grad_norm": 0.4121900498867035, "learning_rate": 0.0002, "epoch": 0.18760469011725292, "step": 280}, {"loss": 1.7976, "grad_norm": 0.3544778525829315, "learning_rate": 0.0002, "epoch": 0.19430485762144054, "step": 290}, {"loss": 1.8787, "grad_norm": 0.3482133448123932, "learning_rate": 0.0002, "epoch": 0.20100502512562815, "step": 300}, {"loss": 1.8578, "grad_norm": 0.3421826660633087, "learning_rate": 0.0002, "epoch": 0.20770519262981574, "step": 310}, {"loss": 1.8013, "grad_norm": 0.5024696588516235, "learning_rate": 0.0002, "epoch": 0.21440536013400335, "step": 320}, {"loss": 1.8607, "grad_norm": 0.36013063788414, "learning_rate": 0.0002, "epoch": 0.22110552763819097, "step": 330}, {"loss": 1.9075, "grad_norm": 0.3611244857311249, "learning_rate": 0.0002, "epoch": 0.22780569514237856, "step": 340}, {"loss": 1.8128, "grad_norm": 0.39244529604911804, "learning_rate": 0.0002, "epoch": 0.23450586264656617, "step": 350}, {"loss": 1.7885, "grad_norm": 0.3299325704574585, "learning_rate": 0.0002, "epoch": 0.24120603015075376, "step": 360}, {"loss": 1.8028, "grad_norm": 0.3994322419166565, "learning_rate": 0.0002, "epoch": 0.24790619765494137, "step": 370}, {"loss": 1.8321, "grad_norm": 0.3559151887893677, "learning_rate": 0.0002, "epoch": 0.254606365159129, "step": 380}, {"loss": 1.7802, "grad_norm": 0.3873756229877472, "learning_rate": 0.0002, "epoch": 0.2613065326633166, "step": 390}, {"loss": 1.7844, "grad_norm": 0.3710744082927704, "learning_rate": 0.0002, "epoch": 0.26800670016750416, "step": 400}, {"loss": 1.7789, "grad_norm": 0.3618465065956116, "learning_rate": 0.0002, "epoch": 0.2747068676716918, "step": 410}, {"loss": 1.8529, "grad_norm": 0.30063769221305847, "learning_rate": 0.0002, "epoch": 0.2814070351758794, "step": 420}, {"loss": 1.7765, "grad_norm": 0.3695628345012665, "learning_rate": 0.0002, "epoch": 0.288107202680067, "step": 430}, {"loss": 1.7982, "grad_norm": 0.31451135873794556, "learning_rate": 0.0002, "epoch": 0.2948073701842546, "step": 440}, {"loss": 1.7517, "grad_norm": 0.3959707021713257, "learning_rate": 0.0002, "epoch": 0.3015075376884422, "step": 450}, {"loss": 1.8142, "grad_norm": 0.33475354313850403, "learning_rate": 0.0002, "epoch": 0.3082077051926298, "step": 460}, {"loss": 1.8805, "grad_norm": 0.33933115005493164, "learning_rate": 0.0002, "epoch": 0.3149078726968174, "step": 470}, {"loss": 1.7564, "grad_norm": 0.3264943063259125, "learning_rate": 0.0002, "epoch": 0.32160804020100503, "step": 480}, {"loss": 1.8428, "grad_norm": 0.40188100934028625, "learning_rate": 0.0002, "epoch": 0.32830820770519265, "step": 490}, {"loss": 1.7624, "grad_norm": 0.37408649921417236, "learning_rate": 0.0002, "epoch": 0.33500837520938026, "step": 500}, {"loss": 1.7745, "grad_norm": 0.33925938606262207, "learning_rate": 0.0002, "epoch": 0.3417085427135678, "step": 510}, {"loss": 1.814, "grad_norm": 0.36836713552474976, "learning_rate": 0.0002, "epoch": 0.34840871021775544, "step": 520}, {"loss": 1.8037, "grad_norm": 0.37284499406814575, "learning_rate": 0.0002, "epoch": 0.35510887772194305, "step": 530}, {"loss": 1.8379, "grad_norm": 0.3192278742790222, "learning_rate": 0.0002, "epoch": 0.36180904522613067, "step": 540}, {"loss": 1.8702, "grad_norm": 0.30233290791511536, "learning_rate": 0.0002, "epoch": 0.3685092127303183, "step": 550}, {"loss": 1.8799, "grad_norm": 0.3340817391872406, "learning_rate": 0.0002, "epoch": 0.37520938023450584, "step": 560}, {"loss": 1.8404, "grad_norm": 0.32600095868110657, "learning_rate": 0.0002, "epoch": 0.38190954773869346, "step": 570}, {"loss": 1.7804, "grad_norm": 0.33711278438568115, "learning_rate": 0.0002, "epoch": 0.38860971524288107, "step": 580}, {"loss": 1.8445, "grad_norm": 0.34890690445899963, "learning_rate": 0.0002, "epoch": 0.3953098827470687, "step": 590}, {"loss": 1.8187, "grad_norm": 0.38238924741744995, "learning_rate": 0.0002, "epoch": 0.4020100502512563, "step": 600}, {"loss": 1.8111, "grad_norm": 0.34399354457855225, "learning_rate": 0.0002, "epoch": 0.40871021775544386, "step": 610}, {"loss": 1.8006, "grad_norm": 0.3346073627471924, "learning_rate": 0.0002, "epoch": 0.4154103852596315, "step": 620}, {"loss": 1.7705, "grad_norm": 0.3545648157596588, "learning_rate": 0.0002, "epoch": 0.4221105527638191, "step": 630}, {"loss": 1.8445, "grad_norm": 0.3378899097442627, "learning_rate": 0.0002, "epoch": 0.4288107202680067, "step": 640}, {"loss": 1.804, "grad_norm": 0.3255569040775299, "learning_rate": 0.0002, "epoch": 0.4355108877721943, "step": 650}, {"loss": 1.7679, "grad_norm": 0.34880587458610535, "learning_rate": 0.0002, "epoch": 0.44221105527638194, "step": 660}, {"loss": 1.7861, "grad_norm": 0.3402383625507355, "learning_rate": 0.0002, "epoch": 0.4489112227805695, "step": 670}, {"loss": 1.8131, "grad_norm": 0.3594033718109131, "learning_rate": 0.0002, "epoch": 0.4556113902847571, "step": 680}, {"loss": 1.8399, "grad_norm": 0.31000566482543945, "learning_rate": 0.0002, "epoch": 0.4623115577889447, "step": 690}, {"loss": 1.7521, "grad_norm": 0.37229061126708984, "learning_rate": 0.0002, "epoch": 0.46901172529313234, "step": 700}, {"loss": 1.7779, "grad_norm": 0.315801739692688, "learning_rate": 0.0002, "epoch": 0.47571189279731996, "step": 710}, {"loss": 1.7515, "grad_norm": 0.3220832645893097, "learning_rate": 0.0002, "epoch": 0.4824120603015075, "step": 720}, {"loss": 1.7181, "grad_norm": 0.3435456156730652, "learning_rate": 0.0002, "epoch": 0.48911222780569513, "step": 730}, {"loss": 1.8844, "grad_norm": 0.30380892753601074, "learning_rate": 0.0002, "epoch": 0.49581239530988275, "step": 740}, {"loss": 1.7792, "grad_norm": 0.3555026054382324, "learning_rate": 0.0002, "epoch": 0.5025125628140703, "step": 750}, {"loss": 1.7714, "grad_norm": 0.3019855320453644, "learning_rate": 0.0002, "epoch": 0.509212730318258, "step": 760}, {"loss": 1.7962, "grad_norm": 0.309111088514328, "learning_rate": 0.0002, "epoch": 0.5159128978224455, "step": 770}, {"loss": 1.7913, "grad_norm": 0.366020530462265, "learning_rate": 0.0002, "epoch": 0.5226130653266332, "step": 780}, {"loss": 1.8008, "grad_norm": 0.3267050087451935, "learning_rate": 0.0002, "epoch": 0.5293132328308208, "step": 790}, {"loss": 1.7397, "grad_norm": 0.34265750646591187, "learning_rate": 0.0002, "epoch": 0.5360134003350083, "step": 800}, {"loss": 1.8251, "grad_norm": 0.313669890165329, "learning_rate": 0.0002, "epoch": 0.542713567839196, "step": 810}, {"loss": 1.8369, "grad_norm": 0.3355236053466797, "learning_rate": 0.0002, "epoch": 0.5494137353433836, "step": 820}, {"loss": 1.7381, "grad_norm": 0.3186608552932739, "learning_rate": 0.0002, "epoch": 0.5561139028475712, "step": 830}, {"loss": 1.8034, "grad_norm": 0.30357518792152405, "learning_rate": 0.0002, "epoch": 0.5628140703517588, "step": 840}, {"loss": 1.769, "grad_norm": 0.3990040123462677, "learning_rate": 0.0002, "epoch": 0.5695142378559463, "step": 850}, {"loss": 1.7482, "grad_norm": 0.34363803267478943, "learning_rate": 0.0002, "epoch": 0.576214405360134, "step": 860}, {"loss": 1.8106, "grad_norm": 0.3757908046245575, "learning_rate": 0.0002, "epoch": 0.5829145728643216, "step": 870}, {"loss": 1.8104, "grad_norm": 0.3359757661819458, "learning_rate": 0.0002, "epoch": 0.5896147403685092, "step": 880}, {"loss": 1.7591, "grad_norm": 0.5555329918861389, "learning_rate": 0.0002, "epoch": 0.5963149078726968, "step": 890}, {"loss": 1.7715, "grad_norm": 0.4046323895454407, "learning_rate": 0.0002, "epoch": 0.6030150753768844, "step": 900}, {"loss": 1.7998, "grad_norm": 0.29834219813346863, "learning_rate": 0.0002, "epoch": 0.609715242881072, "step": 910}, {"loss": 1.7826, "grad_norm": 0.3241238594055176, "learning_rate": 0.0002, "epoch": 0.6164154103852596, "step": 920}, {"loss": 1.8342, "grad_norm": 0.35154739022254944, "learning_rate": 0.0002, "epoch": 0.6231155778894473, "step": 930}, {"loss": 1.8076, "grad_norm": 0.3287706673145294, "learning_rate": 0.0002, "epoch": 0.6298157453936348, "step": 940}, {"loss": 1.8038, "grad_norm": 0.35670626163482666, "learning_rate": 0.0002, "epoch": 0.6365159128978225, "step": 950}, {"loss": 1.869, "grad_norm": 0.6114104986190796, "learning_rate": 0.0002, "epoch": 0.6432160804020101, "step": 960}, {"loss": 1.8297, "grad_norm": 0.3186565041542053, "learning_rate": 0.0002, "epoch": 0.6499162479061976, "step": 970}, {"loss": 1.7539, "grad_norm": 0.27164125442504883, "learning_rate": 0.0002, "epoch": 0.6566164154103853, "step": 980}, {"loss": 1.8339, "grad_norm": 0.34407344460487366, "learning_rate": 0.0002, "epoch": 0.6633165829145728, "step": 990}, {"loss": 1.855, "grad_norm": 0.368415892124176, "learning_rate": 0.0002, "epoch": 0.6700167504187605, "step": 1000}, {"loss": 1.7821, "grad_norm": 0.3306390643119812, "learning_rate": 0.0002, "epoch": 0.6767169179229481, "step": 1010}, {"loss": 1.786, "grad_norm": 0.3198648989200592, "learning_rate": 0.0002, "epoch": 0.6834170854271356, "step": 1020}, {"loss": 1.816, "grad_norm": 0.3092987537384033, "learning_rate": 0.0002, "epoch": 0.6901172529313233, "step": 1030}, {"loss": 1.7689, "grad_norm": 0.3090653419494629, "learning_rate": 0.0002, "epoch": 0.6968174204355109, "step": 1040}, {"loss": 1.7544, "grad_norm": 0.3485880196094513, "learning_rate": 0.0002, "epoch": 0.7035175879396985, "step": 1050}, {"loss": 1.8187, "grad_norm": 0.35782721638679504, "learning_rate": 0.0002, "epoch": 0.7102177554438861, "step": 1060}, {"loss": 1.806, "grad_norm": 0.34256869554519653, "learning_rate": 0.0002, "epoch": 0.7169179229480737, "step": 1070}, {"loss": 1.7873, "grad_norm": 0.30461037158966064, "learning_rate": 0.0002, "epoch": 0.7236180904522613, "step": 1080}, {"loss": 1.7367, "grad_norm": 0.3398691713809967, "learning_rate": 0.0002, "epoch": 0.7303182579564489, "step": 1090}, {"loss": 1.8756, "grad_norm": 0.3180808126926422, "learning_rate": 0.0002, "epoch": 0.7370184254606366, "step": 1100}, {"loss": 1.6988, "grad_norm": 0.34400665760040283, "learning_rate": 0.0002, "epoch": 0.7437185929648241, "step": 1110}, {"loss": 1.7851, "grad_norm": 0.34244877099990845, "learning_rate": 0.0002, "epoch": 0.7504187604690117, "step": 1120}, {"loss": 1.7841, "grad_norm": 0.29946693778038025, "learning_rate": 0.0002, "epoch": 0.7571189279731994, "step": 1130}, {"loss": 1.7456, "grad_norm": 0.37547236680984497, "learning_rate": 0.0002, "epoch": 0.7638190954773869, "step": 1140}, {"loss": 1.8425, "grad_norm": 0.3263005018234253, "learning_rate": 0.0002, "epoch": 0.7705192629815746, "step": 1150}, {"loss": 1.7222, "grad_norm": 0.41363608837127686, "learning_rate": 0.0002, "epoch": 0.7772194304857621, "step": 1160}, {"loss": 1.7836, "grad_norm": 0.36267954111099243, "learning_rate": 0.0002, "epoch": 0.7839195979899497, "step": 1170}, {"loss": 1.9183, "grad_norm": 0.31789499521255493, "learning_rate": 0.0002, "epoch": 0.7906197654941374, "step": 1180}, {"loss": 1.78, "grad_norm": 0.5708149075508118, "learning_rate": 0.0002, "epoch": 0.7973199329983249, "step": 1190}, {"loss": 1.6908, "grad_norm": 0.322099506855011, "learning_rate": 0.0002, "epoch": 0.8040201005025126, "step": 1200}, {"loss": 1.7639, "grad_norm": 0.3419909179210663, "learning_rate": 0.0002, "epoch": 0.8107202680067002, "step": 1210}, {"loss": 1.7428, "grad_norm": 0.36286255717277527, "learning_rate": 0.0002, "epoch": 0.8174204355108877, "step": 1220}, {"loss": 1.8409, "grad_norm": 0.33992862701416016, "learning_rate": 0.0002, "epoch": 0.8241206030150754, "step": 1230}, {"loss": 1.7507, "grad_norm": 0.32622793316841125, "learning_rate": 0.0002, "epoch": 0.830820770519263, "step": 1240}, {"loss": 1.8098, "grad_norm": 0.3036167621612549, "learning_rate": 0.0002, "epoch": 0.8375209380234506, "step": 1250}, {"loss": 1.8094, "grad_norm": 0.3182215392589569, "learning_rate": 0.0002, "epoch": 0.8442211055276382, "step": 1260}, {"loss": 1.8017, "grad_norm": 0.3270018696784973, "learning_rate": 0.0002, "epoch": 0.8509212730318257, "step": 1270}, {"loss": 1.798, "grad_norm": 0.32652342319488525, "learning_rate": 0.0002, "epoch": 0.8576214405360134, "step": 1280}, {"loss": 1.7448, "grad_norm": 0.3631329834461212, "learning_rate": 0.0002, "epoch": 0.864321608040201, "step": 1290}, {"loss": 1.7, "grad_norm": 0.36706018447875977, "learning_rate": 0.0002, "epoch": 0.8710217755443886, "step": 1300}, {"loss": 1.8178, "grad_norm": 0.3347418010234833, "learning_rate": 0.0002, "epoch": 0.8777219430485762, "step": 1310}, {"loss": 1.7824, "grad_norm": 0.34371060132980347, "learning_rate": 0.0002, "epoch": 0.8844221105527639, "step": 1320}, {"loss": 1.783, "grad_norm": 0.3029090166091919, "learning_rate": 0.0002, "epoch": 0.8911222780569514, "step": 1330}, {"loss": 1.8017, "grad_norm": 0.34700682759284973, "learning_rate": 0.0002, "epoch": 0.897822445561139, "step": 1340}, {"loss": 1.7998, "grad_norm": 0.35574328899383545, "learning_rate": 0.0002, "epoch": 0.9045226130653267, "step": 1350}, {"loss": 1.7339, "grad_norm": 0.30928221344947815, "learning_rate": 0.0002, "epoch": 0.9112227805695142, "step": 1360}, {"loss": 1.7479, "grad_norm": 0.30652928352355957, "learning_rate": 0.0002, "epoch": 0.9179229480737019, "step": 1370}, {"loss": 1.7491, "grad_norm": 0.3838157653808594, "learning_rate": 0.0002, "epoch": 0.9246231155778895, "step": 1380}, {"loss": 1.7977, "grad_norm": 0.31655240058898926, "learning_rate": 0.0002, "epoch": 0.931323283082077, "step": 1390}, {"loss": 1.8175, "grad_norm": 0.41737303137779236, "learning_rate": 0.0002, "epoch": 0.9380234505862647, "step": 1400}, {"loss": 1.6811, "grad_norm": 0.3227267861366272, "learning_rate": 0.0002, "epoch": 0.9447236180904522, "step": 1410}, {"loss": 1.7343, "grad_norm": 0.3729925751686096, "learning_rate": 0.0002, "epoch": 0.9514237855946399, "step": 1420}, {"loss": 1.8221, "grad_norm": 0.30779409408569336, "learning_rate": 0.0002, "epoch": 0.9581239530988275, "step": 1430}, {"loss": 1.7972, "grad_norm": 0.334379643201828, "learning_rate": 0.0002, "epoch": 0.964824120603015, "step": 1440}, {"loss": 1.7141, "grad_norm": 0.3568236231803894, "learning_rate": 0.0002, "epoch": 0.9715242881072027, "step": 1450}, {"loss": 1.7541, "grad_norm": 0.33310577273368835, "learning_rate": 0.0002, "epoch": 0.9782244556113903, "step": 1460}, {"loss": 1.8511, "grad_norm": 0.2972261905670166, "learning_rate": 0.0002, "epoch": 0.9849246231155779, "step": 1470}, {"loss": 1.7654, "grad_norm": 0.3322717845439911, "learning_rate": 0.0002, "epoch": 0.9916247906197655, "step": 1480}, {"loss": 1.8033, "grad_norm": 0.3276330828666687, "learning_rate": 0.0002, "epoch": 0.998324958123953, "step": 1490}, {"eval_loss": 1.8036354780197144, "eval_runtime": 37.8949, "eval_samples_per_second": 13.59, "eval_steps_per_second": 1.715, "epoch": 0.9996649916247906, "step": 1492}, {"loss": 1.7138, "grad_norm": 0.29252371191978455, "learning_rate": 0.0002, "epoch": 1.0050251256281406, "step": 1500}, {"loss": 1.8198, "grad_norm": 0.31607162952423096, "learning_rate": 0.0002, "epoch": 1.0117252931323284, "step": 1510}, {"loss": 1.6779, "grad_norm": 0.32294467091560364, "learning_rate": 0.0002, "epoch": 1.018425460636516, "step": 1520}, {"loss": 1.7919, "grad_norm": 0.3868017792701721, "learning_rate": 0.0002, "epoch": 1.0251256281407035, "step": 1530}, {"loss": 1.7954, "grad_norm": 0.3178282082080841, "learning_rate": 0.0002, "epoch": 1.031825795644891, "step": 1540}, {"loss": 1.7136, "grad_norm": 0.3706750273704529, "learning_rate": 0.0002, "epoch": 1.0385259631490786, "step": 1550}, {"loss": 1.7382, "grad_norm": 0.33930912613868713, "learning_rate": 0.0002, "epoch": 1.0452261306532664, "step": 1560}, {"loss": 1.7602, "grad_norm": 0.33970504999160767, "learning_rate": 0.0002, "epoch": 1.051926298157454, "step": 1570}, {"loss": 1.6573, "grad_norm": 0.42553383111953735, "learning_rate": 0.0002, "epoch": 1.0586264656616415, "step": 1580}, {"loss": 1.645, "grad_norm": 0.3772421181201935, "learning_rate": 0.0002, "epoch": 1.065326633165829, "step": 1590}, {"loss": 1.7362, "grad_norm": 0.34212902188301086, "learning_rate": 0.0002, "epoch": 1.0720268006700167, "step": 1600}, {"loss": 1.7057, "grad_norm": 0.3798283338546753, "learning_rate": 0.0002, "epoch": 1.0787269681742044, "step": 1610}, {"loss": 1.7468, "grad_norm": 0.36909598112106323, "learning_rate": 0.0002, "epoch": 1.085427135678392, "step": 1620}, {"loss": 1.7807, "grad_norm": 0.3344230651855469, "learning_rate": 0.0002, "epoch": 1.0921273031825796, "step": 1630}, {"loss": 1.7111, "grad_norm": 0.3862569332122803, "learning_rate": 0.0002, "epoch": 1.0988274706867671, "step": 1640}, {"loss": 1.7163, "grad_norm": 0.31188511848449707, "learning_rate": 0.0002, "epoch": 1.1055276381909547, "step": 1650}, {"loss": 1.7263, "grad_norm": 0.3563670814037323, "learning_rate": 0.0002, "epoch": 1.1122278056951425, "step": 1660}, {"loss": 1.7718, "grad_norm": 0.35052165389060974, "learning_rate": 0.0002, "epoch": 1.11892797319933, "step": 1670}, {"loss": 1.7601, "grad_norm": 0.3285699188709259, "learning_rate": 0.0002, "epoch": 1.1256281407035176, "step": 1680}, {"loss": 1.6877, "grad_norm": 0.3639393746852875, "learning_rate": 0.0002, "epoch": 1.1323283082077051, "step": 1690}, {"loss": 1.7719, "grad_norm": 0.3842753767967224, "learning_rate": 0.0002, "epoch": 1.1390284757118927, "step": 1700}, {"loss": 1.7002, "grad_norm": 0.3624933063983917, "learning_rate": 0.0002, "epoch": 1.1457286432160805, "step": 1710}, {"loss": 1.7243, "grad_norm": 0.3641220033168793, "learning_rate": 0.0002, "epoch": 1.152428810720268, "step": 1720}, {"loss": 1.752, "grad_norm": 0.32765355706214905, "learning_rate": 0.0002, "epoch": 1.1591289782244556, "step": 1730}, {"loss": 1.6556, "grad_norm": 0.34974896907806396, "learning_rate": 0.0002, "epoch": 1.1658291457286432, "step": 1740}, {"loss": 1.7273, "grad_norm": 0.3910926580429077, "learning_rate": 0.0002, "epoch": 1.1725293132328307, "step": 1750}, {"loss": 1.7204, "grad_norm": 0.3564300537109375, "learning_rate": 0.0002, "epoch": 1.1792294807370185, "step": 1760}, {"loss": 1.746, "grad_norm": 0.34822574257850647, "learning_rate": 0.0002, "epoch": 1.185929648241206, "step": 1770}, {"loss": 1.7256, "grad_norm": 0.36185044050216675, "learning_rate": 0.0002, "epoch": 1.1926298157453936, "step": 1780}, {"loss": 1.6431, "grad_norm": 0.34866711497306824, "learning_rate": 0.0002, "epoch": 1.1993299832495812, "step": 1790}, {"loss": 1.8084, "grad_norm": 0.4017769992351532, "learning_rate": 0.0002, "epoch": 1.2060301507537687, "step": 1800}, {"loss": 1.6985, "grad_norm": 0.32930681109428406, "learning_rate": 0.0002, "epoch": 1.2127303182579565, "step": 1810}, {"loss": 1.7606, "grad_norm": 0.35951921343803406, "learning_rate": 0.0002, "epoch": 1.219430485762144, "step": 1820}, {"loss": 1.6933, "grad_norm": 0.37366992235183716, "learning_rate": 0.0002, "epoch": 1.2261306532663316, "step": 1830}, {"loss": 1.6737, "grad_norm": 0.3565689027309418, "learning_rate": 0.0002, "epoch": 1.2328308207705192, "step": 1840}, {"loss": 1.8013, "grad_norm": 0.3692343533039093, "learning_rate": 0.0002, "epoch": 1.2395309882747068, "step": 1850}, {"loss": 1.736, "grad_norm": 0.38426971435546875, "learning_rate": 0.0002, "epoch": 1.2462311557788945, "step": 1860}, {"loss": 1.7031, "grad_norm": 0.33559855818748474, "learning_rate": 0.0002, "epoch": 1.252931323283082, "step": 1870}, {"loss": 1.7033, "grad_norm": 0.34181106090545654, "learning_rate": 0.0002, "epoch": 1.2596314907872697, "step": 1880}, {"loss": 1.7707, "grad_norm": 0.3916318416595459, "learning_rate": 0.0002, "epoch": 1.2663316582914572, "step": 1890}, {"loss": 1.6686, "grad_norm": 0.3887825012207031, "learning_rate": 0.0002, "epoch": 1.2730318257956448, "step": 1900}, {"loss": 1.7062, "grad_norm": 0.33583927154541016, "learning_rate": 0.0002, "epoch": 1.2797319932998326, "step": 1910}, {"loss": 1.717, "grad_norm": 0.37639349699020386, "learning_rate": 0.0002, "epoch": 1.2864321608040201, "step": 1920}, {"loss": 1.777, "grad_norm": 0.38059428334236145, "learning_rate": 0.0002, "epoch": 1.2931323283082077, "step": 1930}, {"loss": 1.6126, "grad_norm": 0.37253183126449585, "learning_rate": 0.0002, "epoch": 1.2998324958123952, "step": 1940}, {"loss": 1.6758, "grad_norm": 0.37371566891670227, "learning_rate": 0.0002, "epoch": 1.3065326633165828, "step": 1950}, {"loss": 1.6788, "grad_norm": 0.4080910086631775, "learning_rate": 0.0002, "epoch": 1.3132328308207706, "step": 1960}, {"loss": 1.6518, "grad_norm": 0.3174354135990143, "learning_rate": 0.0002, "epoch": 1.3199329983249581, "step": 1970}, {"loss": 1.7925, "grad_norm": 0.4518888294696808, "learning_rate": 0.0002, "epoch": 1.3266331658291457, "step": 1980}, {"loss": 1.7085, "grad_norm": 0.3627921938896179, "learning_rate": 0.0002, "epoch": 1.3333333333333333, "step": 1990}, {"loss": 1.7676, "grad_norm": 0.3655930161476135, "learning_rate": 0.0002, "epoch": 1.3400335008375208, "step": 2000}, {"loss": 1.7016, "grad_norm": 0.3509993255138397, "learning_rate": 0.0002, "epoch": 1.3467336683417086, "step": 2010}, {"loss": 1.7359, "grad_norm": 0.4281129240989685, "learning_rate": 0.0002, "epoch": 1.3534338358458962, "step": 2020}, {"loss": 1.6884, "grad_norm": 0.3821414113044739, "learning_rate": 0.0002, "epoch": 1.3601340033500837, "step": 2030}, {"loss": 1.7075, "grad_norm": 0.3907586336135864, "learning_rate": 0.0002, "epoch": 1.3668341708542713, "step": 2040}, {"loss": 1.7424, "grad_norm": 0.37792932987213135, "learning_rate": 0.0002, "epoch": 1.3735343383584588, "step": 2050}, {"loss": 1.7305, "grad_norm": 0.3693985641002655, "learning_rate": 0.0002, "epoch": 1.3802345058626466, "step": 2060}, {"loss": 1.7434, "grad_norm": 0.32275936007499695, "learning_rate": 0.0002, "epoch": 1.3869346733668342, "step": 2070}, {"loss": 1.6677, "grad_norm": 0.3789440095424652, "learning_rate": 0.0002, "epoch": 1.3936348408710217, "step": 2080}, {"loss": 1.6825, "grad_norm": 0.3638380467891693, "learning_rate": 0.0002, "epoch": 1.4003350083752093, "step": 2090}, {"loss": 1.6542, "grad_norm": 0.3495481610298157, "learning_rate": 0.0002, "epoch": 1.4070351758793969, "step": 2100}, {"loss": 1.7225, "grad_norm": 0.37920597195625305, "learning_rate": 0.0002, "epoch": 1.4137353433835846, "step": 2110}, {"loss": 1.7329, "grad_norm": 0.37218064069747925, "learning_rate": 0.0002, "epoch": 1.4204355108877722, "step": 2120}, {"loss": 1.799, "grad_norm": 0.38074082136154175, "learning_rate": 0.0002, "epoch": 1.4271356783919598, "step": 2130}, {"loss": 1.7403, "grad_norm": 0.3455527126789093, "learning_rate": 0.0002, "epoch": 1.4338358458961473, "step": 2140}, {"loss": 1.776, "grad_norm": 0.3712003529071808, "learning_rate": 0.0002, "epoch": 1.4405360134003349, "step": 2150}, {"loss": 1.7619, "grad_norm": 0.3786754906177521, "learning_rate": 0.0002, "epoch": 1.4472361809045227, "step": 2160}, {"loss": 1.68, "grad_norm": 0.3879223167896271, "learning_rate": 0.0002, "epoch": 1.4539363484087102, "step": 2170}, {"loss": 1.7, "grad_norm": 0.38738805055618286, "learning_rate": 0.0002, "epoch": 1.4606365159128978, "step": 2180}, {"loss": 1.7581, "grad_norm": 0.39768800139427185, "learning_rate": 0.0002, "epoch": 1.4673366834170856, "step": 2190}, {"loss": 1.7671, "grad_norm": 0.4172441065311432, "learning_rate": 0.0002, "epoch": 1.474036850921273, "step": 2200}, {"loss": 1.6736, "grad_norm": 0.4043174982070923, "learning_rate": 0.0002, "epoch": 1.4807370184254607, "step": 2210}, {"loss": 1.7444, "grad_norm": 0.3750883936882019, "learning_rate": 0.0002, "epoch": 1.4874371859296482, "step": 2220}, {"loss": 1.6861, "grad_norm": 0.3552253246307373, "learning_rate": 0.0002, "epoch": 1.4941373534338358, "step": 2230}, {"loss": 1.6471, "grad_norm": 0.34607139229774475, "learning_rate": 0.0002, "epoch": 1.5008375209380236, "step": 2240}, {"loss": 1.6962, "grad_norm": 0.3406706750392914, "learning_rate": 0.0002, "epoch": 1.507537688442211, "step": 2250}, {"loss": 1.7694, "grad_norm": 0.36654895544052124, "learning_rate": 0.0002, "epoch": 1.5142378559463987, "step": 2260}, {"loss": 1.6812, "grad_norm": 0.3914054334163666, "learning_rate": 0.0002, "epoch": 1.5209380234505863, "step": 2270}, {"loss": 1.6822, "grad_norm": 0.42012137174606323, "learning_rate": 0.0002, "epoch": 1.5276381909547738, "step": 2280}, {"loss": 1.697, "grad_norm": 0.39563435316085815, "learning_rate": 0.0002, "epoch": 1.5343383584589616, "step": 2290}, {"loss": 1.7491, "grad_norm": 0.3508438766002655, "learning_rate": 0.0002, "epoch": 1.541038525963149, "step": 2300}, {"loss": 1.7727, "grad_norm": 0.3785218596458435, "learning_rate": 0.0002, "epoch": 1.5477386934673367, "step": 2310}, {"loss": 1.6963, "grad_norm": 0.39377647638320923, "learning_rate": 0.0002, "epoch": 1.5544388609715243, "step": 2320}, {"loss": 1.7263, "grad_norm": 0.3391438126564026, "learning_rate": 0.0002, "epoch": 1.5611390284757118, "step": 2330}, {"loss": 1.7722, "grad_norm": 0.37944263219833374, "learning_rate": 0.0002, "epoch": 1.5678391959798996, "step": 2340}, {"loss": 1.6371, "grad_norm": 0.3523491322994232, "learning_rate": 0.0002, "epoch": 1.574539363484087, "step": 2350}, {"loss": 1.7583, "grad_norm": 0.3911575973033905, "learning_rate": 0.0002, "epoch": 1.5812395309882747, "step": 2360}, {"loss": 1.7117, "grad_norm": 0.33832186460494995, "learning_rate": 0.0002, "epoch": 1.5879396984924623, "step": 2370}, {"loss": 1.7701, "grad_norm": 0.3665979206562042, "learning_rate": 0.0002, "epoch": 1.5946398659966499, "step": 2380}, {"loss": 1.779, "grad_norm": 0.3871748149394989, "learning_rate": 0.0002, "epoch": 1.6013400335008376, "step": 2390}, {"loss": 1.7109, "grad_norm": 0.3586967885494232, "learning_rate": 0.0002, "epoch": 1.608040201005025, "step": 2400}, {"loss": 1.7096, "grad_norm": 0.3563673198223114, "learning_rate": 0.0002, "epoch": 1.6147403685092128, "step": 2410}, {"loss": 1.745, "grad_norm": 0.37588971853256226, "learning_rate": 0.0002, "epoch": 1.6214405360134003, "step": 2420}, {"loss": 1.7086, "grad_norm": 0.352556437253952, "learning_rate": 0.0002, "epoch": 1.6281407035175879, "step": 2430}, {"loss": 1.6547, "grad_norm": 0.3716259300708771, "learning_rate": 0.0002, "epoch": 1.6348408710217757, "step": 2440}, {"loss": 1.7033, "grad_norm": 0.372001975774765, "learning_rate": 0.0002, "epoch": 1.641541038525963, "step": 2450}, {"loss": 1.6584, "grad_norm": 0.3430042862892151, "learning_rate": 0.0002, "epoch": 1.6482412060301508, "step": 2460}, {"loss": 1.7217, "grad_norm": 0.3741483688354492, "learning_rate": 0.0002, "epoch": 1.6549413735343383, "step": 2470}, {"loss": 1.7701, "grad_norm": 0.3610571324825287, "learning_rate": 0.0002, "epoch": 1.661641541038526, "step": 2480}, {"loss": 1.7057, "grad_norm": 0.4204719066619873, "learning_rate": 0.0002, "epoch": 1.6683417085427137, "step": 2490}, {"loss": 1.7954, "grad_norm": 0.3938186466693878, "learning_rate": 0.0002, "epoch": 1.675041876046901, "step": 2500}, {"loss": 1.6633, "grad_norm": 0.3421435058116913, "learning_rate": 0.0002, "epoch": 1.6817420435510888, "step": 2510}, {"loss": 1.7996, "grad_norm": 0.42441412806510925, "learning_rate": 0.0002, "epoch": 1.6884422110552764, "step": 2520}, {"loss": 1.7142, "grad_norm": 0.38071519136428833, "learning_rate": 0.0002, "epoch": 1.695142378559464, "step": 2530}, {"loss": 1.7232, "grad_norm": 0.34078919887542725, "learning_rate": 0.0002, "epoch": 1.7018425460636517, "step": 2540}, {"loss": 1.7126, "grad_norm": 0.412844181060791, "learning_rate": 0.0002, "epoch": 1.708542713567839, "step": 2550}, {"loss": 1.7149, "grad_norm": 0.3753604292869568, "learning_rate": 0.0002, "epoch": 1.7152428810720268, "step": 2560}, {"loss": 1.7011, "grad_norm": 0.41588476300239563, "learning_rate": 0.0002, "epoch": 1.7219430485762144, "step": 2570}, {"loss": 1.6427, "grad_norm": 0.35504111647605896, "learning_rate": 0.0002, "epoch": 1.728643216080402, "step": 2580}, {"loss": 1.7296, "grad_norm": 0.36909720301628113, "learning_rate": 0.0002, "epoch": 1.7353433835845897, "step": 2590}, {"loss": 1.7022, "grad_norm": 0.4149979054927826, "learning_rate": 0.0002, "epoch": 1.742043551088777, "step": 2600}, {"loss": 1.77, "grad_norm": 0.38859328627586365, "learning_rate": 0.0002, "epoch": 1.7487437185929648, "step": 2610}, {"loss": 1.7036, "grad_norm": 0.36738792061805725, "learning_rate": 0.0002, "epoch": 1.7554438860971524, "step": 2620}, {"loss": 1.764, "grad_norm": 0.3968178927898407, "learning_rate": 0.0002, "epoch": 1.76214405360134, "step": 2630}, {"loss": 1.7687, "grad_norm": 0.3972901999950409, "learning_rate": 0.0002, "epoch": 1.7688442211055277, "step": 2640}, {"loss": 1.6774, "grad_norm": 0.3949959874153137, "learning_rate": 0.0002, "epoch": 1.775544388609715, "step": 2650}, {"loss": 1.7247, "grad_norm": 0.44074657559394836, "learning_rate": 0.0002, "epoch": 1.7822445561139029, "step": 2660}, {"loss": 1.7188, "grad_norm": 0.39743664860725403, "learning_rate": 0.0002, "epoch": 1.7889447236180904, "step": 2670}, {"loss": 1.7258, "grad_norm": 0.3950406610965729, "learning_rate": 0.0002, "epoch": 1.795644891122278, "step": 2680}, {"loss": 1.6906, "grad_norm": 0.3568263649940491, "learning_rate": 0.0002, "epoch": 1.8023450586264658, "step": 2690}, {"loss": 1.6735, "grad_norm": 0.3819476366043091, "learning_rate": 0.0002, "epoch": 1.809045226130653, "step": 2700}, {"loss": 1.7198, "grad_norm": 0.3480634391307831, "learning_rate": 0.0002, "epoch": 1.8157453936348409, "step": 2710}, {"loss": 1.7042, "grad_norm": 0.3875853419303894, "learning_rate": 0.0002, "epoch": 1.8224455611390284, "step": 2720}, {"loss": 1.6988, "grad_norm": 0.3441337049007416, "learning_rate": 0.0002, "epoch": 1.829145728643216, "step": 2730}, {"loss": 1.7647, "grad_norm": 0.35692882537841797, "learning_rate": 0.0002, "epoch": 1.8358458961474038, "step": 2740}, {"loss": 1.7033, "grad_norm": 0.36959215998649597, "learning_rate": 0.0002, "epoch": 1.8425460636515911, "step": 2750}, {"loss": 1.7657, "grad_norm": 0.3893393278121948, "learning_rate": 0.0002, "epoch": 1.849246231155779, "step": 2760}, {"loss": 1.7068, "grad_norm": 0.37817293405532837, "learning_rate": 0.0002, "epoch": 1.8559463986599665, "step": 2770}, {"loss": 1.761, "grad_norm": 0.36071285605430603, "learning_rate": 0.0002, "epoch": 1.862646566164154, "step": 2780}, {"loss": 1.7623, "grad_norm": 0.3758420944213867, "learning_rate": 0.0002, "epoch": 1.8693467336683418, "step": 2790}, {"loss": 1.6743, "grad_norm": 0.3889938294887543, "learning_rate": 0.0002, "epoch": 1.8760469011725294, "step": 2800}, {"loss": 1.6151, "grad_norm": 0.34361857175827026, "learning_rate": 0.0002, "epoch": 1.882747068676717, "step": 2810}, {"loss": 1.6038, "grad_norm": 0.39283323287963867, "learning_rate": 0.0002, "epoch": 1.8894472361809045, "step": 2820}, {"loss": 1.7555, "grad_norm": 0.3919452726840973, "learning_rate": 0.0002, "epoch": 1.896147403685092, "step": 2830}, {"loss": 1.673, "grad_norm": 0.38215070962905884, "learning_rate": 0.0002, "epoch": 1.9028475711892798, "step": 2840}, {"loss": 1.7044, "grad_norm": 0.4235064387321472, "learning_rate": 0.0002, "epoch": 1.9095477386934674, "step": 2850}, {"loss": 1.7123, "grad_norm": 0.35694634914398193, "learning_rate": 0.0002, "epoch": 1.916247906197655, "step": 2860}, {"loss": 1.8128, "grad_norm": 0.383492112159729, "learning_rate": 0.0002, "epoch": 1.9229480737018425, "step": 2870}, {"loss": 1.7581, "grad_norm": 0.5945147275924683, "learning_rate": 0.0002, "epoch": 1.92964824120603, "step": 2880}, {"loss": 1.7421, "grad_norm": 0.3367522358894348, "learning_rate": 0.0002, "epoch": 1.9363484087102178, "step": 2890}, {"loss": 1.6561, "grad_norm": 0.35300394892692566, "learning_rate": 0.0002, "epoch": 1.9430485762144054, "step": 2900}, {"loss": 1.7033, "grad_norm": 0.38084495067596436, "learning_rate": 0.0002, "epoch": 1.949748743718593, "step": 2910}, {"loss": 1.7132, "grad_norm": 0.37559160590171814, "learning_rate": 0.0002, "epoch": 1.9564489112227805, "step": 2920}, {"loss": 1.6759, "grad_norm": 0.3661738336086273, "learning_rate": 0.0002, "epoch": 1.963149078726968, "step": 2930}, {"loss": 1.7643, "grad_norm": 0.4073849320411682, "learning_rate": 0.0002, "epoch": 1.9698492462311559, "step": 2940}, {"loss": 1.6806, "grad_norm": 0.3723304271697998, "learning_rate": 0.0002, "epoch": 1.9765494137353434, "step": 2950}, {"loss": 1.7611, "grad_norm": 0.3991098999977112, "learning_rate": 0.0002, "epoch": 1.983249581239531, "step": 2960}, {"loss": 1.7263, "grad_norm": 0.3947085440158844, "learning_rate": 0.0002, "epoch": 1.9899497487437185, "step": 2970}, {"loss": 1.7217, "grad_norm": 0.3786258399486542, "learning_rate": 0.0002, "epoch": 1.996649916247906, "step": 2980}, {"eval_loss": 1.8028968572616577, "eval_runtime": 37.8985, "eval_samples_per_second": 13.589, "eval_steps_per_second": 1.715, "epoch": 2.0, "step": 2985}, {"loss": 1.695, "grad_norm": 0.34824079275131226, "learning_rate": 0.0002, "epoch": 2.003350083752094, "step": 2990}, {"loss": 1.5853, "grad_norm": 0.3394894003868103, "learning_rate": 0.0002, "epoch": 2.0100502512562812, "step": 3000}, {"loss": 1.5783, "grad_norm": 0.36910977959632874, "learning_rate": 0.0002, "epoch": 2.016750418760469, "step": 3010}, {"loss": 1.6105, "grad_norm": 0.45000967383384705, "learning_rate": 0.0002, "epoch": 2.023450586264657, "step": 3020}, {"loss": 1.6019, "grad_norm": 0.3791407346725464, "learning_rate": 0.0002, "epoch": 2.030150753768844, "step": 3030}, {"loss": 1.5832, "grad_norm": 0.387321799993515, "learning_rate": 0.0002, "epoch": 2.036850921273032, "step": 3040}, {"loss": 1.6834, "grad_norm": 0.4185757040977478, "learning_rate": 0.0002, "epoch": 2.0435510887772192, "step": 3050}, {"loss": 1.5696, "grad_norm": 0.45110777020454407, "learning_rate": 0.0002, "epoch": 2.050251256281407, "step": 3060}, {"loss": 1.6231, "grad_norm": 0.42663660645484924, "learning_rate": 0.0002, "epoch": 2.056951423785595, "step": 3070}, {"loss": 1.6279, "grad_norm": 0.4546292722225189, "learning_rate": 0.0002, "epoch": 2.063651591289782, "step": 3080}, {"loss": 1.6141, "grad_norm": 0.3979759216308594, "learning_rate": 0.0002, "epoch": 2.07035175879397, "step": 3090}, {"loss": 1.6343, "grad_norm": 0.43596673011779785, "learning_rate": 0.0002, "epoch": 2.0770519262981573, "step": 3100}, {"loss": 1.5441, "grad_norm": 0.40120232105255127, "learning_rate": 0.0002, "epoch": 2.083752093802345, "step": 3110}, {"loss": 1.6309, "grad_norm": 0.44449281692504883, "learning_rate": 0.0002, "epoch": 2.090452261306533, "step": 3120}, {"loss": 1.5652, "grad_norm": 0.42672568559646606, "learning_rate": 0.0002, "epoch": 2.09715242881072, "step": 3130}, {"loss": 1.682, "grad_norm": 0.4232690930366516, "learning_rate": 0.0002, "epoch": 2.103852596314908, "step": 3140}, {"loss": 1.624, "grad_norm": 0.4299317002296448, "learning_rate": 0.0002, "epoch": 2.1105527638190953, "step": 3150}, {"loss": 1.6766, "grad_norm": 0.4067758023738861, "learning_rate": 0.0002, "epoch": 2.117252931323283, "step": 3160}, {"loss": 1.6759, "grad_norm": 0.4918815791606903, "learning_rate": 0.0002, "epoch": 2.123953098827471, "step": 3170}, {"loss": 1.6478, "grad_norm": 0.4140559732913971, "learning_rate": 0.0002, "epoch": 2.130653266331658, "step": 3180}, {"loss": 1.6641, "grad_norm": 0.4555995464324951, "learning_rate": 0.0002, "epoch": 2.137353433835846, "step": 3190}, {"loss": 1.5888, "grad_norm": 0.42943915724754333, "learning_rate": 0.0002, "epoch": 2.1440536013400333, "step": 3200}, {"loss": 1.5886, "grad_norm": 0.4730435013771057, "learning_rate": 0.0002, "epoch": 2.150753768844221, "step": 3210}, {"loss": 1.6022, "grad_norm": 0.43310216069221497, "learning_rate": 0.0002, "epoch": 2.157453936348409, "step": 3220}, {"loss": 1.6058, "grad_norm": 0.42054110765457153, "learning_rate": 0.0002, "epoch": 2.164154103852596, "step": 3230}, {"loss": 1.6749, "grad_norm": 0.4897233247756958, "learning_rate": 0.0002, "epoch": 2.170854271356784, "step": 3240}, {"loss": 1.6983, "grad_norm": 0.42194533348083496, "learning_rate": 0.0002, "epoch": 2.1775544388609713, "step": 3250}, {"loss": 1.6159, "grad_norm": 0.44494450092315674, "learning_rate": 0.0002, "epoch": 2.184254606365159, "step": 3260}, {"loss": 1.6977, "grad_norm": 0.43524879217147827, "learning_rate": 0.0002, "epoch": 2.190954773869347, "step": 3270}, {"loss": 1.528, "grad_norm": 0.4621117413043976, "learning_rate": 0.0002, "epoch": 2.1976549413735342, "step": 3280}, {"loss": 1.632, "grad_norm": 0.4073285460472107, "learning_rate": 0.0002, "epoch": 2.204355108877722, "step": 3290}, {"loss": 1.6141, "grad_norm": 0.47868335247039795, "learning_rate": 0.0002, "epoch": 2.2110552763819094, "step": 3300}, {"loss": 1.6857, "grad_norm": 0.4264970123767853, "learning_rate": 0.0002, "epoch": 2.217755443886097, "step": 3310}, {"loss": 1.5653, "grad_norm": 0.4491245150566101, "learning_rate": 0.0002, "epoch": 2.224455611390285, "step": 3320}, {"loss": 1.5881, "grad_norm": 0.4010344445705414, "learning_rate": 0.0002, "epoch": 2.2311557788944723, "step": 3330}, {"loss": 1.6684, "grad_norm": 0.4232759177684784, "learning_rate": 0.0002, "epoch": 2.23785594639866, "step": 3340}, {"loss": 1.6336, "grad_norm": 0.5099776983261108, "learning_rate": 0.0002, "epoch": 2.2445561139028474, "step": 3350}, {"loss": 1.6764, "grad_norm": 0.5223407745361328, "learning_rate": 0.0002, "epoch": 2.251256281407035, "step": 3360}, {"loss": 1.6625, "grad_norm": 0.47818470001220703, "learning_rate": 0.0002, "epoch": 2.257956448911223, "step": 3370}, {"loss": 1.5946, "grad_norm": 0.4721255898475647, "learning_rate": 0.0002, "epoch": 2.2646566164154103, "step": 3380}, {"loss": 1.5568, "grad_norm": 0.4113229513168335, "learning_rate": 0.0002, "epoch": 2.271356783919598, "step": 3390}, {"loss": 1.6494, "grad_norm": 0.507080078125, "learning_rate": 0.0002, "epoch": 2.2780569514237854, "step": 3400}, {"loss": 1.6183, "grad_norm": 0.4852292239665985, "learning_rate": 0.0002, "epoch": 2.284757118927973, "step": 3410}, {"loss": 1.6132, "grad_norm": 0.4503684341907501, "learning_rate": 0.0002, "epoch": 2.291457286432161, "step": 3420}, {"loss": 1.6649, "grad_norm": 0.8359600305557251, "learning_rate": 0.0002, "epoch": 2.2981574539363483, "step": 3430}, {"loss": 1.6644, "grad_norm": 0.44604045152664185, "learning_rate": 0.0002, "epoch": 2.304857621440536, "step": 3440}, {"loss": 1.5972, "grad_norm": 0.45667049288749695, "learning_rate": 0.0002, "epoch": 2.3115577889447234, "step": 3450}, {"loss": 1.6667, "grad_norm": 0.4879349172115326, "learning_rate": 0.0002, "epoch": 2.318257956448911, "step": 3460}, {"loss": 1.5804, "grad_norm": 0.4033963084220886, "learning_rate": 0.0002, "epoch": 2.324958123953099, "step": 3470}, {"loss": 1.5838, "grad_norm": 0.44494301080703735, "learning_rate": 0.0002, "epoch": 2.3316582914572863, "step": 3480}, {"loss": 1.5951, "grad_norm": 0.4794621765613556, "learning_rate": 0.0002, "epoch": 2.338358458961474, "step": 3490}, {"loss": 1.6807, "grad_norm": 0.41404327750205994, "learning_rate": 0.0002, "epoch": 2.3450586264656614, "step": 3500}, {"loss": 1.714, "grad_norm": 0.4664851725101471, "learning_rate": 0.0002, "epoch": 2.351758793969849, "step": 3510}, {"loss": 1.6537, "grad_norm": 0.4263697564601898, "learning_rate": 0.0002, "epoch": 2.358458961474037, "step": 3520}, {"loss": 1.6551, "grad_norm": 0.5035167336463928, "learning_rate": 0.0002, "epoch": 2.3651591289782243, "step": 3530}, {"loss": 1.6208, "grad_norm": 0.4380664527416229, "learning_rate": 0.0002, "epoch": 2.371859296482412, "step": 3540}, {"loss": 1.634, "grad_norm": 0.5227681994438171, "learning_rate": 0.0002, "epoch": 2.3785594639865995, "step": 3550}, {"loss": 1.6146, "grad_norm": 0.4382302761077881, "learning_rate": 0.0002, "epoch": 2.3852596314907872, "step": 3560}, {"loss": 1.5653, "grad_norm": 0.4392451047897339, "learning_rate": 0.0002, "epoch": 2.391959798994975, "step": 3570}, {"loss": 1.6626, "grad_norm": 0.4372786581516266, "learning_rate": 0.0002, "epoch": 2.3986599664991624, "step": 3580}, {"loss": 1.519, "grad_norm": 0.5015502572059631, "learning_rate": 0.0002, "epoch": 2.40536013400335, "step": 3590}, {"loss": 1.588, "grad_norm": 0.5653210878372192, "learning_rate": 0.0002, "epoch": 2.4120603015075375, "step": 3600}, {"loss": 1.6075, "grad_norm": 0.53007972240448, "learning_rate": 0.0002, "epoch": 2.4187604690117253, "step": 3610}, {"loss": 1.6421, "grad_norm": 0.4659176766872406, "learning_rate": 0.0002, "epoch": 2.425460636515913, "step": 3620}, {"loss": 1.625, "grad_norm": 0.5637837052345276, "learning_rate": 0.0002, "epoch": 2.4321608040201004, "step": 3630}, {"loss": 1.6168, "grad_norm": 0.4248391389846802, "learning_rate": 0.0002, "epoch": 2.438860971524288, "step": 3640}, {"loss": 1.6822, "grad_norm": 0.44668248295783997, "learning_rate": 0.0002, "epoch": 2.4455611390284755, "step": 3650}, {"loss": 1.6417, "grad_norm": 0.43990179896354675, "learning_rate": 0.0002, "epoch": 2.4522613065326633, "step": 3660}, {"loss": 1.6723, "grad_norm": 0.4532523453235626, "learning_rate": 0.0002, "epoch": 2.458961474036851, "step": 3670}, {"loss": 1.6957, "grad_norm": 0.6605591773986816, "learning_rate": 0.0002, "epoch": 2.4656616415410384, "step": 3680}, {"loss": 1.6159, "grad_norm": 0.4694533348083496, "learning_rate": 0.0002, "epoch": 2.472361809045226, "step": 3690}, {"loss": 1.6239, "grad_norm": 0.4485011100769043, "learning_rate": 0.0002, "epoch": 2.4790619765494135, "step": 3700}, {"loss": 1.6834, "grad_norm": 0.4761785864830017, "learning_rate": 0.0002, "epoch": 2.4857621440536013, "step": 3710}, {"loss": 1.6313, "grad_norm": 0.5116432309150696, "learning_rate": 0.0002, "epoch": 2.492462311557789, "step": 3720}, {"loss": 1.5054, "grad_norm": 0.49523618817329407, "learning_rate": 0.0002, "epoch": 2.4991624790619764, "step": 3730}, {"loss": 1.6249, "grad_norm": 0.43826380372047424, "learning_rate": 0.0002, "epoch": 2.505862646566164, "step": 3740}, {"loss": 1.5762, "grad_norm": 0.4916154146194458, "learning_rate": 0.0002, "epoch": 2.5125628140703515, "step": 3750}, {"loss": 1.5157, "grad_norm": 0.5381299257278442, "learning_rate": 0.0002, "epoch": 2.5192629815745393, "step": 3760}, {"loss": 1.6467, "grad_norm": 0.44947415590286255, "learning_rate": 0.0002, "epoch": 2.525963149078727, "step": 3770}, {"loss": 1.67, "grad_norm": 0.49979084730148315, "learning_rate": 0.0002, "epoch": 2.5326633165829144, "step": 3780}, {"loss": 1.622, "grad_norm": 0.43046900629997253, "learning_rate": 0.0002, "epoch": 2.539363484087102, "step": 3790}, {"loss": 1.6789, "grad_norm": 0.4513470530509949, "learning_rate": 0.0002, "epoch": 2.5460636515912896, "step": 3800}, {"loss": 1.6335, "grad_norm": 0.49900051951408386, "learning_rate": 0.0002, "epoch": 2.5527638190954773, "step": 3810}, {"loss": 1.6322, "grad_norm": 0.4348420202732086, "learning_rate": 0.0002, "epoch": 2.559463986599665, "step": 3820}, {"loss": 1.6218, "grad_norm": 0.4684867560863495, "learning_rate": 0.0002, "epoch": 2.5661641541038525, "step": 3830}, {"loss": 1.6535, "grad_norm": 0.44430989027023315, "learning_rate": 0.0002, "epoch": 2.5728643216080402, "step": 3840}, {"loss": 1.5909, "grad_norm": 0.47375255823135376, "learning_rate": 0.0002, "epoch": 2.5795644891122276, "step": 3850}, {"loss": 1.6269, "grad_norm": 0.45493075251579285, "learning_rate": 0.0002, "epoch": 2.5862646566164154, "step": 3860}, {"loss": 1.604, "grad_norm": 0.4563275873661041, "learning_rate": 0.0002, "epoch": 2.592964824120603, "step": 3870}, {"loss": 1.642, "grad_norm": 0.46060335636138916, "learning_rate": 0.0002, "epoch": 2.5996649916247905, "step": 3880}, {"loss": 1.6302, "grad_norm": 0.4718867540359497, "learning_rate": 0.0002, "epoch": 2.6063651591289783, "step": 3890}, {"loss": 1.6242, "grad_norm": 0.41570305824279785, "learning_rate": 0.0002, "epoch": 2.6130653266331656, "step": 3900}, {"loss": 1.6401, "grad_norm": 0.4603121876716614, "learning_rate": 0.0002, "epoch": 2.6197654941373534, "step": 3910}, {"loss": 1.6839, "grad_norm": 0.4734652638435364, "learning_rate": 0.0002, "epoch": 2.626465661641541, "step": 3920}, {"loss": 1.5448, "grad_norm": 0.45348483324050903, "learning_rate": 0.0002, "epoch": 2.6331658291457285, "step": 3930}, {"loss": 1.6157, "grad_norm": 0.46559447050094604, "learning_rate": 0.0002, "epoch": 2.6398659966499163, "step": 3940}, {"loss": 1.7052, "grad_norm": 0.44113144278526306, "learning_rate": 0.0002, "epoch": 2.6465661641541036, "step": 3950}, {"loss": 1.6315, "grad_norm": 0.41415104269981384, "learning_rate": 0.0002, "epoch": 2.6532663316582914, "step": 3960}, {"loss": 1.6589, "grad_norm": 0.48868080973625183, "learning_rate": 0.0002, "epoch": 2.659966499162479, "step": 3970}, {"loss": 1.6211, "grad_norm": 0.49610549211502075, "learning_rate": 0.0002, "epoch": 2.6666666666666665, "step": 3980}, {"loss": 1.6235, "grad_norm": 0.4309130907058716, "learning_rate": 0.0002, "epoch": 2.6733668341708543, "step": 3990}, {"loss": 1.6452, "grad_norm": 0.4489327669143677, "learning_rate": 0.0002, "epoch": 2.6800670016750416, "step": 4000}, {"loss": 1.5744, "grad_norm": 0.5380139946937561, "learning_rate": 0.0002, "epoch": 2.6867671691792294, "step": 4010}, {"loss": 1.6524, "grad_norm": 0.5076672434806824, "learning_rate": 0.0002, "epoch": 2.693467336683417, "step": 4020}, {"loss": 1.636, "grad_norm": 0.47620031237602234, "learning_rate": 0.0002, "epoch": 2.7001675041876045, "step": 4030}, {"loss": 1.5543, "grad_norm": 0.48089155554771423, "learning_rate": 0.0002, "epoch": 2.7068676716917923, "step": 4040}, {"loss": 1.6396, "grad_norm": 0.5108814239501953, "learning_rate": 0.0002, "epoch": 2.7135678391959797, "step": 4050}, {"loss": 1.5905, "grad_norm": 0.4196513295173645, "learning_rate": 0.0002, "epoch": 2.7202680067001674, "step": 4060}, {"loss": 1.686, "grad_norm": 0.4574664831161499, "learning_rate": 0.0002, "epoch": 2.726968174204355, "step": 4070}, {"loss": 1.6234, "grad_norm": 0.4671640992164612, "learning_rate": 0.0002, "epoch": 2.7336683417085426, "step": 4080}, {"loss": 1.6827, "grad_norm": 0.49355530738830566, "learning_rate": 0.0002, "epoch": 2.7403685092127303, "step": 4090}, {"loss": 1.6999, "grad_norm": 0.46716663241386414, "learning_rate": 0.0002, "epoch": 2.7470686767169177, "step": 4100}, {"loss": 1.6463, "grad_norm": 0.45420581102371216, "learning_rate": 0.0002, "epoch": 2.7537688442211055, "step": 4110}, {"loss": 1.5718, "grad_norm": 0.4680487811565399, "learning_rate": 0.0002, "epoch": 2.7604690117252932, "step": 4120}, {"loss": 1.5968, "grad_norm": 0.5375032424926758, "learning_rate": 0.0002, "epoch": 2.7671691792294806, "step": 4130}, {"loss": 1.5254, "grad_norm": 0.46026280522346497, "learning_rate": 0.0002, "epoch": 2.7738693467336684, "step": 4140}, {"loss": 1.6613, "grad_norm": 0.43658447265625, "learning_rate": 0.0002, "epoch": 2.7805695142378557, "step": 4150}, {"loss": 1.6546, "grad_norm": 0.4935547113418579, "learning_rate": 0.0002, "epoch": 2.7872696817420435, "step": 4160}, {"loss": 1.5961, "grad_norm": 0.8167962431907654, "learning_rate": 0.0002, "epoch": 2.7939698492462313, "step": 4170}, {"loss": 1.6907, "grad_norm": 0.4289683997631073, "learning_rate": 0.0002, "epoch": 2.8006700167504186, "step": 4180}, {"loss": 1.6385, "grad_norm": 0.4569324254989624, "learning_rate": 0.0002, "epoch": 2.8073701842546064, "step": 4190}, {"loss": 1.6077, "grad_norm": 0.474795937538147, "learning_rate": 0.0002, "epoch": 2.8140703517587937, "step": 4200}, {"loss": 1.6223, "grad_norm": 0.44272229075431824, "learning_rate": 0.0002, "epoch": 2.8207705192629815, "step": 4210}, {"loss": 1.6706, "grad_norm": 0.525240957736969, "learning_rate": 0.0002, "epoch": 2.8274706867671693, "step": 4220}, {"loss": 1.7196, "grad_norm": 0.4802303910255432, "learning_rate": 0.0002, "epoch": 2.8341708542713566, "step": 4230}, {"loss": 1.6002, "grad_norm": 0.46400442719459534, "learning_rate": 0.0002, "epoch": 2.8408710217755444, "step": 4240}, {"loss": 1.6052, "grad_norm": 0.49884888529777527, "learning_rate": 0.0002, "epoch": 2.8475711892797317, "step": 4250}, {"loss": 1.6919, "grad_norm": 0.5015072226524353, "learning_rate": 0.0002, "epoch": 2.8542713567839195, "step": 4260}, {"loss": 1.6335, "grad_norm": 0.4335440695285797, "learning_rate": 0.0002, "epoch": 2.8609715242881073, "step": 4270}, {"loss": 1.5664, "grad_norm": 0.5131644606590271, "learning_rate": 0.0002, "epoch": 2.8676716917922946, "step": 4280}, {"loss": 1.6409, "grad_norm": 0.6977195739746094, "learning_rate": 0.0002, "epoch": 2.8743718592964824, "step": 4290}, {"loss": 1.7192, "grad_norm": 0.5133762955665588, "learning_rate": 0.0002, "epoch": 2.8810720268006698, "step": 4300}, {"loss": 1.6257, "grad_norm": 0.4737614393234253, "learning_rate": 0.0002, "epoch": 2.8877721943048575, "step": 4310}, {"loss": 1.6076, "grad_norm": 0.4580535590648651, "learning_rate": 0.0002, "epoch": 2.8944723618090453, "step": 4320}, {"loss": 1.6538, "grad_norm": 0.43863341212272644, "learning_rate": 0.0002, "epoch": 2.901172529313233, "step": 4330}, {"loss": 1.6091, "grad_norm": 0.4103737473487854, "learning_rate": 0.0002, "epoch": 2.9078726968174204, "step": 4340}, {"loss": 1.7106, "grad_norm": 0.438014417886734, "learning_rate": 0.0002, "epoch": 2.914572864321608, "step": 4350}, {"loss": 1.6025, "grad_norm": 0.5068213939666748, "learning_rate": 0.0002, "epoch": 2.9212730318257956, "step": 4360}, {"loss": 1.6426, "grad_norm": 0.45305484533309937, "learning_rate": 0.0002, "epoch": 2.9279731993299833, "step": 4370}, {"loss": 1.5726, "grad_norm": 0.4612090289592743, "learning_rate": 0.0002, "epoch": 2.934673366834171, "step": 4380}, {"loss": 1.6536, "grad_norm": 0.508736789226532, "learning_rate": 0.0002, "epoch": 2.9413735343383585, "step": 4390}, {"loss": 1.6132, "grad_norm": 0.4924427270889282, "learning_rate": 0.0002, "epoch": 2.948073701842546, "step": 4400}, {"loss": 1.7007, "grad_norm": 0.5707460641860962, "learning_rate": 0.0002, "epoch": 2.9547738693467336, "step": 4410}, {"loss": 1.6814, "grad_norm": 0.42270299792289734, "learning_rate": 0.0002, "epoch": 2.9614740368509214, "step": 4420}, {"loss": 1.6644, "grad_norm": 0.4429931044578552, "learning_rate": 0.0002, "epoch": 2.968174204355109, "step": 4430}, {"loss": 1.6251, "grad_norm": 0.49760574102401733, "learning_rate": 0.0002, "epoch": 2.9748743718592965, "step": 4440}, {"loss": 1.6169, "grad_norm": 0.4558229148387909, "learning_rate": 0.0002, "epoch": 2.981574539363484, "step": 4450}, {"loss": 1.6055, "grad_norm": 0.39848530292510986, "learning_rate": 0.0002, "epoch": 2.9882747068676716, "step": 4460}, {"loss": 1.6705, "grad_norm": 0.5224862098693848, "learning_rate": 0.0002, "epoch": 2.9949748743718594, "step": 4470}, {"eval_loss": 1.8228833675384521, "eval_runtime": 37.9049, "eval_samples_per_second": 13.587, "eval_steps_per_second": 1.715, "epoch": 2.9996649916247904, "step": 4477}, {"loss": 1.6637, "grad_norm": 0.41169142723083496, "learning_rate": 0.0002, "epoch": 3.0016750418760467, "step": 4480}, {"loss": 1.5974, "grad_norm": 0.4865207374095917, "learning_rate": 0.0002, "epoch": 3.0083752093802345, "step": 4490}, {"loss": 1.5297, "grad_norm": 0.5462028384208679, "learning_rate": 0.0002, "epoch": 3.0150753768844223, "step": 4500}, {"loss": 1.5251, "grad_norm": 0.6169732809066772, "learning_rate": 0.0002, "epoch": 3.0217755443886096, "step": 4510}, {"loss": 1.5559, "grad_norm": 0.5667954087257385, "learning_rate": 0.0002, "epoch": 3.0284757118927974, "step": 4520}, {"loss": 1.5037, "grad_norm": 0.5758325457572937, "learning_rate": 0.0002, "epoch": 3.0351758793969847, "step": 4530}, {"loss": 1.4873, "grad_norm": 0.5220064520835876, "learning_rate": 0.0002, "epoch": 3.0418760469011725, "step": 4540}, {"loss": 1.5126, "grad_norm": 0.5469558835029602, "learning_rate": 0.0002, "epoch": 3.0485762144053603, "step": 4550}, {"loss": 1.4275, "grad_norm": 0.5680848956108093, "learning_rate": 0.0002, "epoch": 3.0552763819095476, "step": 4560}, {"loss": 1.5187, "grad_norm": 0.5906574726104736, "learning_rate": 0.0002, "epoch": 3.0619765494137354, "step": 4570}, {"loss": 1.4551, "grad_norm": 0.4725631773471832, "learning_rate": 0.0002, "epoch": 3.0686767169179228, "step": 4580}, {"loss": 1.5083, "grad_norm": 0.5273477435112, "learning_rate": 0.0002, "epoch": 3.0753768844221105, "step": 4590}, {"loss": 1.5154, "grad_norm": 0.5861203074455261, "learning_rate": 0.0002, "epoch": 3.0820770519262983, "step": 4600}, {"loss": 1.4924, "grad_norm": 0.5343965291976929, "learning_rate": 0.0002, "epoch": 3.0887772194304857, "step": 4610}, {"loss": 1.5608, "grad_norm": 0.5348150730133057, "learning_rate": 0.0002, "epoch": 3.0954773869346734, "step": 4620}, {"loss": 1.5399, "grad_norm": 0.5971846580505371, "learning_rate": 0.0002, "epoch": 3.102177554438861, "step": 4630}, {"loss": 1.4662, "grad_norm": 0.5203177332878113, "learning_rate": 0.0002, "epoch": 3.1088777219430486, "step": 4640}, {"loss": 1.5805, "grad_norm": 0.55289226770401, "learning_rate": 0.0002, "epoch": 3.1155778894472363, "step": 4650}, {"loss": 1.4745, "grad_norm": 0.6878530979156494, "learning_rate": 0.0002, "epoch": 3.1222780569514237, "step": 4660}, {"loss": 1.5335, "grad_norm": 0.6173256635665894, "learning_rate": 0.0002, "epoch": 3.1289782244556115, "step": 4670}, {"loss": 1.51, "grad_norm": 0.536796510219574, "learning_rate": 0.0002, "epoch": 3.135678391959799, "step": 4680}, {"loss": 1.4713, "grad_norm": 0.58846116065979, "learning_rate": 0.0002, "epoch": 3.1423785594639866, "step": 4690}, {"loss": 1.5114, "grad_norm": 0.645889401435852, "learning_rate": 0.0002, "epoch": 3.1490787269681744, "step": 4700}, {"loss": 1.4705, "grad_norm": 0.6118691563606262, "learning_rate": 0.0002, "epoch": 3.1557788944723617, "step": 4710}, {"loss": 1.5533, "grad_norm": 0.5189669132232666, "learning_rate": 0.0002, "epoch": 3.1624790619765495, "step": 4720}, {"loss": 1.4769, "grad_norm": 0.5794713497161865, "learning_rate": 0.0002, "epoch": 3.169179229480737, "step": 4730}, {"loss": 1.4849, "grad_norm": 0.6579326391220093, "learning_rate": 0.0002, "epoch": 3.1758793969849246, "step": 4740}, {"loss": 1.545, "grad_norm": 0.5822742581367493, "learning_rate": 0.0002, "epoch": 3.1825795644891124, "step": 4750}, {"loss": 1.4358, "grad_norm": 0.5475956201553345, "learning_rate": 0.0002, "epoch": 3.1892797319932997, "step": 4760}, {"loss": 1.4723, "grad_norm": 0.6743834018707275, "learning_rate": 0.0002, "epoch": 3.1959798994974875, "step": 4770}, {"loss": 1.5161, "grad_norm": 0.6110585927963257, "learning_rate": 0.0002, "epoch": 3.202680067001675, "step": 4780}, {"loss": 1.5455, "grad_norm": 0.5426181554794312, "learning_rate": 0.0002, "epoch": 3.2093802345058626, "step": 4790}, {"loss": 1.5315, "grad_norm": 0.6077824234962463, "learning_rate": 0.0002, "epoch": 3.2160804020100504, "step": 4800}, {"loss": 1.5314, "grad_norm": 0.5785858631134033, "learning_rate": 0.0002, "epoch": 3.2227805695142377, "step": 4810}, {"loss": 1.4041, "grad_norm": 0.6425958275794983, "learning_rate": 0.0002, "epoch": 3.2294807370184255, "step": 4820}, {"loss": 1.4751, "grad_norm": 0.6607080698013306, "learning_rate": 0.0002, "epoch": 3.236180904522613, "step": 4830}, {"loss": 1.5267, "grad_norm": 0.5385788679122925, "learning_rate": 0.0002, "epoch": 3.2428810720268006, "step": 4840}, {"loss": 1.4673, "grad_norm": 0.5630403757095337, "learning_rate": 0.0002, "epoch": 3.2495812395309884, "step": 4850}, {"loss": 1.5257, "grad_norm": 0.6340779662132263, "learning_rate": 0.0002, "epoch": 3.2562814070351758, "step": 4860}, {"loss": 1.5148, "grad_norm": 0.5305342674255371, "learning_rate": 0.0002, "epoch": 3.2629815745393635, "step": 4870}, {"loss": 1.5162, "grad_norm": 0.597670316696167, "learning_rate": 0.0002, "epoch": 3.2696817420435513, "step": 4880}, {"loss": 1.5429, "grad_norm": 0.665553867816925, "learning_rate": 0.0002, "epoch": 3.2763819095477387, "step": 4890}, {"loss": 1.4607, "grad_norm": 0.579767644405365, "learning_rate": 0.0002, "epoch": 3.2830820770519265, "step": 4900}, {"loss": 1.4999, "grad_norm": 0.5512481331825256, "learning_rate": 0.0002, "epoch": 3.289782244556114, "step": 4910}, {"loss": 1.5022, "grad_norm": 0.5916532278060913, "learning_rate": 0.0002, "epoch": 3.2964824120603016, "step": 4920}, {"loss": 1.4889, "grad_norm": 0.7521726489067078, "learning_rate": 0.0002, "epoch": 3.3031825795644894, "step": 4930}, {"loss": 1.4223, "grad_norm": 0.5352797508239746, "learning_rate": 0.0002, "epoch": 3.3098827470686767, "step": 4940}, {"loss": 1.5122, "grad_norm": 0.5950371623039246, "learning_rate": 0.0002, "epoch": 3.3165829145728645, "step": 4950}, {"loss": 1.5072, "grad_norm": 0.8020477890968323, "learning_rate": 0.0002, "epoch": 3.323283082077052, "step": 4960}, {"loss": 1.5422, "grad_norm": 0.6790024638175964, "learning_rate": 0.0002, "epoch": 3.3299832495812396, "step": 4970}, {"loss": 1.5363, "grad_norm": 0.687627375125885, "learning_rate": 0.0002, "epoch": 3.3366834170854274, "step": 4980}, {"loss": 1.5276, "grad_norm": 0.6094385385513306, "learning_rate": 0.0002, "epoch": 3.3433835845896147, "step": 4990}, {"loss": 1.549, "grad_norm": 0.6541242003440857, "learning_rate": 0.0002, "epoch": 3.3500837520938025, "step": 5000}, {"loss": 1.6067, "grad_norm": 0.5560880303382874, "learning_rate": 0.0002, "epoch": 3.35678391959799, "step": 5010}, {"loss": 1.5769, "grad_norm": 0.5440094470977783, "learning_rate": 0.0002, "epoch": 3.3634840871021776, "step": 5020}, {"loss": 1.6183, "grad_norm": 0.5749301314353943, "learning_rate": 0.0002, "epoch": 3.3701842546063654, "step": 5030}, {"loss": 1.4801, "grad_norm": 0.5919716954231262, "learning_rate": 0.0002, "epoch": 3.3768844221105527, "step": 5040}, {"loss": 1.5957, "grad_norm": 0.6331481337547302, "learning_rate": 0.0002, "epoch": 3.3835845896147405, "step": 5050}, {"loss": 1.5188, "grad_norm": 0.5687161684036255, "learning_rate": 0.0002, "epoch": 3.390284757118928, "step": 5060}, {"loss": 1.5702, "grad_norm": 0.6718577742576599, "learning_rate": 0.0002, "epoch": 3.3969849246231156, "step": 5070}, {"loss": 1.5577, "grad_norm": 0.5089324116706848, "learning_rate": 0.0002, "epoch": 3.4036850921273034, "step": 5080}, {"loss": 1.512, "grad_norm": 0.5710174441337585, "learning_rate": 0.0002, "epoch": 3.4103852596314908, "step": 5090}, {"loss": 1.5492, "grad_norm": 0.6670721173286438, "learning_rate": 0.0002, "epoch": 3.4170854271356785, "step": 5100}, {"loss": 1.5227, "grad_norm": 0.6875665187835693, "learning_rate": 0.0002, "epoch": 3.423785594639866, "step": 5110}, {"loss": 1.4496, "grad_norm": 0.5375880599021912, "learning_rate": 0.0002, "epoch": 3.4304857621440537, "step": 5120}, {"loss": 1.5527, "grad_norm": 0.6550399661064148, "learning_rate": 0.0002, "epoch": 3.4371859296482414, "step": 5130}, {"loss": 1.5687, "grad_norm": 0.5948067903518677, "learning_rate": 0.0002, "epoch": 3.4438860971524288, "step": 5140}, {"loss": 1.4813, "grad_norm": 0.6134477257728577, "learning_rate": 0.0002, "epoch": 3.4505862646566166, "step": 5150}, {"loss": 1.5069, "grad_norm": 0.6506398320198059, "learning_rate": 0.0002, "epoch": 3.457286432160804, "step": 5160}, {"loss": 1.4422, "grad_norm": 0.6060147881507874, "learning_rate": 0.0002, "epoch": 3.4639865996649917, "step": 5170}, {"loss": 1.5093, "grad_norm": 0.6173806190490723, "learning_rate": 0.0002, "epoch": 3.4706867671691795, "step": 5180}, {"loss": 1.4975, "grad_norm": 0.6032607555389404, "learning_rate": 0.0002, "epoch": 3.477386934673367, "step": 5190}, {"loss": 1.4979, "grad_norm": 0.5652492046356201, "learning_rate": 0.0002, "epoch": 3.4840871021775546, "step": 5200}, {"loss": 1.4883, "grad_norm": 0.6168607473373413, "learning_rate": 0.0002, "epoch": 3.490787269681742, "step": 5210}, {"loss": 1.5164, "grad_norm": 0.6170629262924194, "learning_rate": 0.0002, "epoch": 3.4974874371859297, "step": 5220}, {"loss": 1.4879, "grad_norm": 0.6926297545433044, "learning_rate": 0.0002, "epoch": 3.5041876046901175, "step": 5230}, {"loss": 1.4982, "grad_norm": 0.6702437996864319, "learning_rate": 0.0002, "epoch": 3.510887772194305, "step": 5240}, {"loss": 1.4986, "grad_norm": 0.5421436429023743, "learning_rate": 0.0002, "epoch": 3.5175879396984926, "step": 5250}, {"loss": 1.4673, "grad_norm": 0.5726765990257263, "learning_rate": 0.0002, "epoch": 3.52428810720268, "step": 5260}, {"loss": 1.5423, "grad_norm": 0.5685455203056335, "learning_rate": 0.0002, "epoch": 3.5309882747068677, "step": 5270}, {"loss": 1.4715, "grad_norm": 0.6018396019935608, "learning_rate": 0.0002, "epoch": 3.5376884422110555, "step": 5280}, {"loss": 1.5451, "grad_norm": 0.5731932520866394, "learning_rate": 0.0002, "epoch": 3.544388609715243, "step": 5290}, {"loss": 1.4752, "grad_norm": 0.6601519584655762, "learning_rate": 0.0002, "epoch": 3.5510887772194306, "step": 5300}, {"loss": 1.5434, "grad_norm": 0.5545530319213867, "learning_rate": 0.0002, "epoch": 3.557788944723618, "step": 5310}, {"loss": 1.5438, "grad_norm": 0.5998541116714478, "learning_rate": 0.0002, "epoch": 3.5644891122278057, "step": 5320}, {"loss": 1.56, "grad_norm": 0.5651767253875732, "learning_rate": 0.0002, "epoch": 3.5711892797319935, "step": 5330}, {"loss": 1.4829, "grad_norm": 0.7425084114074707, "learning_rate": 0.0002, "epoch": 3.577889447236181, "step": 5340}, {"loss": 1.5571, "grad_norm": 0.5770602226257324, "learning_rate": 0.0002, "epoch": 3.5845896147403686, "step": 5350}, {"loss": 1.458, "grad_norm": 0.54723060131073, "learning_rate": 0.0002, "epoch": 3.591289782244556, "step": 5360}, {"loss": 1.497, "grad_norm": 0.6658238172531128, "learning_rate": 0.0002, "epoch": 3.5979899497487438, "step": 5370}, {"loss": 1.5456, "grad_norm": 0.5787645578384399, "learning_rate": 0.0002, "epoch": 3.6046901172529315, "step": 5380}, {"loss": 1.5343, "grad_norm": 0.594913125038147, "learning_rate": 0.0002, "epoch": 3.611390284757119, "step": 5390}, {"loss": 1.4727, "grad_norm": 0.4964977502822876, "learning_rate": 0.0002, "epoch": 3.6180904522613067, "step": 5400}, {"loss": 1.5062, "grad_norm": 0.6087527275085449, "learning_rate": 0.0002, "epoch": 3.624790619765494, "step": 5410}, {"loss": 1.5098, "grad_norm": 0.6315323710441589, "learning_rate": 0.0002, "epoch": 3.6314907872696818, "step": 5420}, {"loss": 1.4855, "grad_norm": 0.574799120426178, "learning_rate": 0.0002, "epoch": 3.6381909547738696, "step": 5430}, {"loss": 1.4595, "grad_norm": 0.5949277877807617, "learning_rate": 0.0002, "epoch": 3.644891122278057, "step": 5440}, {"loss": 1.4816, "grad_norm": 0.5640677213668823, "learning_rate": 0.0002, "epoch": 3.6515912897822447, "step": 5450}, {"loss": 1.525, "grad_norm": 0.6198237538337708, "learning_rate": 0.0002, "epoch": 3.658291457286432, "step": 5460}, {"loss": 1.5676, "grad_norm": 0.6902034878730774, "learning_rate": 0.0002, "epoch": 3.66499162479062, "step": 5470}, {"loss": 1.544, "grad_norm": 0.5686674118041992, "learning_rate": 0.0002, "epoch": 3.6716917922948076, "step": 5480}, {"loss": 1.5255, "grad_norm": 0.6532107591629028, "learning_rate": 0.0002, "epoch": 3.678391959798995, "step": 5490}, {"loss": 1.5767, "grad_norm": 0.5790849924087524, "learning_rate": 0.0002, "epoch": 3.6850921273031827, "step": 5500}, {"loss": 1.5507, "grad_norm": 0.6055065393447876, "learning_rate": 0.0002, "epoch": 3.69179229480737, "step": 5510}, {"loss": 1.4656, "grad_norm": 0.5630605816841125, "learning_rate": 0.0002, "epoch": 3.698492462311558, "step": 5520}, {"loss": 1.537, "grad_norm": 0.6005825996398926, "learning_rate": 0.0002, "epoch": 3.7051926298157456, "step": 5530}, {"loss": 1.5313, "grad_norm": 0.6553038954734802, "learning_rate": 0.0002, "epoch": 3.711892797319933, "step": 5540}, {"loss": 1.4943, "grad_norm": 0.5601094961166382, "learning_rate": 0.0002, "epoch": 3.7185929648241207, "step": 5550}, {"loss": 1.5161, "grad_norm": 0.6598808169364929, "learning_rate": 0.0002, "epoch": 3.725293132328308, "step": 5560}, {"loss": 1.5345, "grad_norm": 0.5506255626678467, "learning_rate": 0.0002, "epoch": 3.731993299832496, "step": 5570}, {"loss": 1.4805, "grad_norm": 0.6001223921775818, "learning_rate": 0.0002, "epoch": 3.7386934673366836, "step": 5580}, {"loss": 1.4652, "grad_norm": 0.6287297606468201, "learning_rate": 0.0002, "epoch": 3.745393634840871, "step": 5590}, {"loss": 1.5246, "grad_norm": 0.6253238916397095, "learning_rate": 0.0002, "epoch": 3.7520938023450587, "step": 5600}, {"loss": 1.5691, "grad_norm": 0.5713174939155579, "learning_rate": 0.0002, "epoch": 3.758793969849246, "step": 5610}, {"loss": 1.5661, "grad_norm": 0.6198310852050781, "learning_rate": 0.0002, "epoch": 3.765494137353434, "step": 5620}, {"loss": 1.5448, "grad_norm": 0.5941224098205566, "learning_rate": 0.0002, "epoch": 3.7721943048576216, "step": 5630}, {"loss": 1.4925, "grad_norm": 0.606002151966095, "learning_rate": 0.0002, "epoch": 3.778894472361809, "step": 5640}, {"loss": 1.5182, "grad_norm": 0.6540704965591431, "learning_rate": 0.0002, "epoch": 3.7855946398659968, "step": 5650}, {"loss": 1.5903, "grad_norm": 0.6147415041923523, "learning_rate": 0.0002, "epoch": 3.792294807370184, "step": 5660}, {"loss": 1.5329, "grad_norm": 0.5649605393409729, "learning_rate": 0.0002, "epoch": 3.798994974874372, "step": 5670}, {"loss": 1.5747, "grad_norm": 0.6788773536682129, "learning_rate": 0.0002, "epoch": 3.8056951423785597, "step": 5680}, {"loss": 1.535, "grad_norm": 0.6581860780715942, "learning_rate": 0.0002, "epoch": 3.812395309882747, "step": 5690}, {"loss": 1.4587, "grad_norm": 0.5529348850250244, "learning_rate": 0.0002, "epoch": 3.819095477386935, "step": 5700}, {"loss": 1.5594, "grad_norm": 0.6320232152938843, "learning_rate": 0.0002, "epoch": 3.825795644891122, "step": 5710}, {"loss": 1.4696, "grad_norm": 0.6529698371887207, "learning_rate": 0.0002, "epoch": 3.83249581239531, "step": 5720}, {"loss": 1.5854, "grad_norm": 0.5983362793922424, "learning_rate": 0.0002, "epoch": 3.8391959798994977, "step": 5730}, {"loss": 1.465, "grad_norm": 0.6335684061050415, "learning_rate": 0.0002, "epoch": 3.845896147403685, "step": 5740}, {"loss": 1.5545, "grad_norm": 0.700446605682373, "learning_rate": 0.0002, "epoch": 3.852596314907873, "step": 5750}, {"loss": 1.5707, "grad_norm": 0.6092597842216492, "learning_rate": 0.0002, "epoch": 3.85929648241206, "step": 5760}, {"loss": 1.5729, "grad_norm": 0.564146101474762, "learning_rate": 0.0002, "epoch": 3.865996649916248, "step": 5770}, {"loss": 1.5872, "grad_norm": 0.615275502204895, "learning_rate": 0.0002, "epoch": 3.8726968174204357, "step": 5780}, {"loss": 1.5142, "grad_norm": 0.6685376763343811, "learning_rate": 0.0002, "epoch": 3.879396984924623, "step": 5790}, {"loss": 1.4752, "grad_norm": 0.6116922497749329, "learning_rate": 0.0002, "epoch": 3.886097152428811, "step": 5800}, {"loss": 1.5179, "grad_norm": 0.5486813187599182, "learning_rate": 0.0002, "epoch": 3.892797319932998, "step": 5810}, {"loss": 1.5167, "grad_norm": 0.6208204030990601, "learning_rate": 0.0002, "epoch": 3.899497487437186, "step": 5820}, {"loss": 1.5334, "grad_norm": 0.6500625014305115, "learning_rate": 0.0002, "epoch": 3.9061976549413737, "step": 5830}, {"loss": 1.4716, "grad_norm": 0.5948089361190796, "learning_rate": 0.0002, "epoch": 3.912897822445561, "step": 5840}, {"loss": 1.6011, "grad_norm": 0.7210732698440552, "learning_rate": 0.0002, "epoch": 3.919597989949749, "step": 5850}, {"loss": 1.5519, "grad_norm": 0.6662322878837585, "learning_rate": 0.0002, "epoch": 3.926298157453936, "step": 5860}, {"loss": 1.5656, "grad_norm": 0.5613839626312256, "learning_rate": 0.0002, "epoch": 3.932998324958124, "step": 5870}, {"loss": 1.544, "grad_norm": 0.6069002151489258, "learning_rate": 0.0002, "epoch": 3.9396984924623117, "step": 5880}, {"loss": 1.6745, "grad_norm": 0.7075562477111816, "learning_rate": 0.0002, "epoch": 3.946398659966499, "step": 5890}, {"loss": 1.5391, "grad_norm": 0.6316173076629639, "learning_rate": 0.0002, "epoch": 3.953098827470687, "step": 5900}, {"loss": 1.6314, "grad_norm": 0.5716308355331421, "learning_rate": 0.0002, "epoch": 3.959798994974874, "step": 5910}, {"loss": 1.5947, "grad_norm": 0.6800096035003662, "learning_rate": 0.0002, "epoch": 3.966499162479062, "step": 5920}, {"loss": 1.5189, "grad_norm": 0.6057983040809631, "learning_rate": 0.0002, "epoch": 3.9731993299832498, "step": 5930}, {"loss": 1.5431, "grad_norm": 0.5938987731933594, "learning_rate": 0.0002, "epoch": 3.979899497487437, "step": 5940}, {"loss": 1.5111, "grad_norm": 0.6963576674461365, "learning_rate": 0.0002, "epoch": 3.986599664991625, "step": 5950}, {"loss": 1.5521, "grad_norm": 0.6279940009117126, "learning_rate": 0.0002, "epoch": 3.993299832495812, "step": 5960}, {"loss": 1.5974, "grad_norm": 0.7161159515380859, "learning_rate": 0.0002, "epoch": 4.0, "step": 5970}, {"eval_loss": 1.8655421733856201, "eval_runtime": 37.9276, "eval_samples_per_second": 13.579, "eval_steps_per_second": 1.714, "epoch": 4.0, "step": 5970}, {"loss": 1.3666, "grad_norm": 0.7380476593971252, "learning_rate": 0.0002, "epoch": 4.006700167504188, "step": 5980}, {"loss": 1.3913, "grad_norm": 0.7148947715759277, "learning_rate": 0.0002, "epoch": 4.013400335008376, "step": 5990}, {"loss": 1.4204, "grad_norm": 0.6177082657814026, "learning_rate": 0.0002, "epoch": 4.0201005025125625, "step": 6000}, {"loss": 1.4421, "grad_norm": 0.8552946448326111, "learning_rate": 0.0002, "epoch": 4.02680067001675, "step": 6010}, {"loss": 1.4342, "grad_norm": 0.8033416271209717, "learning_rate": 0.0002, "epoch": 4.033500837520938, "step": 6020}, {"loss": 1.4092, "grad_norm": 0.8501318097114563, "learning_rate": 0.0002, "epoch": 4.040201005025126, "step": 6030}, {"loss": 1.3367, "grad_norm": 0.6981393098831177, "learning_rate": 0.0002, "epoch": 4.046901172529314, "step": 6040}, {"loss": 1.3925, "grad_norm": 0.7227180600166321, "learning_rate": 0.0002, "epoch": 4.0536013400335005, "step": 6050}, {"loss": 1.4007, "grad_norm": 0.6923989653587341, "learning_rate": 0.0002, "epoch": 4.060301507537688, "step": 6060}, {"loss": 1.3837, "grad_norm": 0.879779040813446, "learning_rate": 0.0002, "epoch": 4.067001675041876, "step": 6070}, {"loss": 1.4383, "grad_norm": 0.8184754848480225, "learning_rate": 0.0002, "epoch": 4.073701842546064, "step": 6080}, {"loss": 1.3128, "grad_norm": 0.8211342692375183, "learning_rate": 0.0002, "epoch": 4.080402010050252, "step": 6090}, {"loss": 1.3892, "grad_norm": 0.7542396783828735, "learning_rate": 0.0002, "epoch": 4.0871021775544385, "step": 6100}, {"loss": 1.3607, "grad_norm": 0.6631066799163818, "learning_rate": 0.0002, "epoch": 4.093802345058626, "step": 6110}, {"loss": 1.3275, "grad_norm": 0.6728386282920837, "learning_rate": 0.0002, "epoch": 4.100502512562814, "step": 6120}, {"loss": 1.3443, "grad_norm": 0.681851863861084, "learning_rate": 0.0002, "epoch": 4.107202680067002, "step": 6130}, {"loss": 1.3486, "grad_norm": 0.8757794499397278, "learning_rate": 0.0002, "epoch": 4.11390284757119, "step": 6140}, {"loss": 1.351, "grad_norm": 0.6567301750183105, "learning_rate": 0.0002, "epoch": 4.1206030150753765, "step": 6150}, {"loss": 1.3824, "grad_norm": 0.7950329184532166, "learning_rate": 0.0002, "epoch": 4.127303182579564, "step": 6160}, {"loss": 1.3738, "grad_norm": 0.7545644044876099, "learning_rate": 0.0002, "epoch": 4.134003350083752, "step": 6170}, {"loss": 1.4214, "grad_norm": 0.7172710299491882, "learning_rate": 0.0002, "epoch": 4.14070351758794, "step": 6180}, {"loss": 1.4091, "grad_norm": 0.7040584087371826, "learning_rate": 0.0002, "epoch": 4.147403685092128, "step": 6190}, {"loss": 1.4149, "grad_norm": 0.7482913732528687, "learning_rate": 0.0002, "epoch": 4.1541038525963145, "step": 6200}, {"loss": 1.3227, "grad_norm": 0.8523276448249817, "learning_rate": 0.0002, "epoch": 4.160804020100502, "step": 6210}, {"loss": 1.4194, "grad_norm": 0.6672041416168213, "learning_rate": 0.0002, "epoch": 4.16750418760469, "step": 6220}, {"loss": 1.3953, "grad_norm": 0.7523500919342041, "learning_rate": 0.0002, "epoch": 4.174204355108878, "step": 6230}, {"loss": 1.371, "grad_norm": 0.8085253834724426, "learning_rate": 0.0002, "epoch": 4.180904522613066, "step": 6240}, {"loss": 1.3293, "grad_norm": 0.789450466632843, "learning_rate": 0.0002, "epoch": 4.187604690117253, "step": 6250}, {"loss": 1.3539, "grad_norm": 0.7502310872077942, "learning_rate": 0.0002, "epoch": 4.19430485762144, "step": 6260}, {"loss": 1.3415, "grad_norm": 0.7397456765174866, "learning_rate": 0.0002, "epoch": 4.201005025125628, "step": 6270}, {"loss": 1.3963, "grad_norm": 0.6921947002410889, "learning_rate": 0.0002, "epoch": 4.207705192629816, "step": 6280}, {"loss": 1.3125, "grad_norm": 0.9334571957588196, "learning_rate": 0.0002, "epoch": 4.214405360134004, "step": 6290}, {"loss": 1.3612, "grad_norm": 0.725799024105072, "learning_rate": 0.0002, "epoch": 4.221105527638191, "step": 6300}, {"loss": 1.4217, "grad_norm": 0.8290495872497559, "learning_rate": 0.0002, "epoch": 4.227805695142378, "step": 6310}, {"loss": 1.4135, "grad_norm": 0.688983678817749, "learning_rate": 0.0002, "epoch": 4.234505862646566, "step": 6320}, {"loss": 1.3807, "grad_norm": 0.8620913028717041, "learning_rate": 0.0002, "epoch": 4.241206030150754, "step": 6330}, {"loss": 1.3738, "grad_norm": 0.8008657693862915, "learning_rate": 0.0002, "epoch": 4.247906197654942, "step": 6340}, {"loss": 1.4005, "grad_norm": 0.7379199266433716, "learning_rate": 0.0002, "epoch": 4.254606365159129, "step": 6350}, {"loss": 1.426, "grad_norm": 0.7842815518379211, "learning_rate": 0.0002, "epoch": 4.261306532663316, "step": 6360}, {"loss": 1.4262, "grad_norm": 0.812600314617157, "learning_rate": 0.0002, "epoch": 4.268006700167504, "step": 6370}, {"loss": 1.4028, "grad_norm": 0.7852841019630432, "learning_rate": 0.0002, "epoch": 4.274706867671692, "step": 6380}, {"loss": 1.3722, "grad_norm": 1.0377534627914429, "learning_rate": 0.0002, "epoch": 4.28140703517588, "step": 6390}, {"loss": 1.3755, "grad_norm": 1.03935706615448, "learning_rate": 0.0002, "epoch": 4.288107202680067, "step": 6400}, {"loss": 1.3961, "grad_norm": 0.7244732975959778, "learning_rate": 0.0002, "epoch": 4.294807370184254, "step": 6410}, {"loss": 1.4608, "grad_norm": 0.7137406468391418, "learning_rate": 0.0002, "epoch": 4.301507537688442, "step": 6420}, {"loss": 1.4461, "grad_norm": 0.7492543458938599, "learning_rate": 0.0002, "epoch": 4.30820770519263, "step": 6430}, {"loss": 1.4562, "grad_norm": 0.7065439224243164, "learning_rate": 0.0002, "epoch": 4.314907872696818, "step": 6440}, {"loss": 1.4246, "grad_norm": 0.7786989808082581, "learning_rate": 0.0002, "epoch": 4.321608040201005, "step": 6450}, {"loss": 1.3098, "grad_norm": 0.7369208335876465, "learning_rate": 0.0002, "epoch": 4.328308207705192, "step": 6460}, {"loss": 1.3686, "grad_norm": 0.7412346005439758, "learning_rate": 0.0002, "epoch": 4.33500837520938, "step": 6470}, {"loss": 1.4087, "grad_norm": 0.780927300453186, "learning_rate": 0.0002, "epoch": 4.341708542713568, "step": 6480}, {"loss": 1.3628, "grad_norm": 0.8320930600166321, "learning_rate": 0.0002, "epoch": 4.348408710217756, "step": 6490}, {"loss": 1.3715, "grad_norm": 0.6871094703674316, "learning_rate": 0.0002, "epoch": 4.355108877721943, "step": 6500}, {"loss": 1.3257, "grad_norm": 0.6751559972763062, "learning_rate": 0.0002, "epoch": 4.36180904522613, "step": 6510}, {"loss": 1.4311, "grad_norm": 0.7723976969718933, "learning_rate": 0.0002, "epoch": 4.368509212730318, "step": 6520}, {"loss": 1.4086, "grad_norm": 0.7915401458740234, "learning_rate": 0.0002, "epoch": 4.375209380234506, "step": 6530}, {"loss": 1.3973, "grad_norm": 0.7329102754592896, "learning_rate": 0.0002, "epoch": 4.381909547738694, "step": 6540}, {"loss": 1.447, "grad_norm": 0.7388760447502136, "learning_rate": 0.0002, "epoch": 4.388609715242881, "step": 6550}, {"loss": 1.4378, "grad_norm": 0.8282579183578491, "learning_rate": 0.0002, "epoch": 4.3953098827470685, "step": 6560}, {"loss": 1.3923, "grad_norm": 0.7192724347114563, "learning_rate": 0.0002, "epoch": 4.402010050251256, "step": 6570}, {"loss": 1.4141, "grad_norm": 0.746526837348938, "learning_rate": 0.0002, "epoch": 4.408710217755444, "step": 6580}, {"loss": 1.33, "grad_norm": 0.8738046288490295, "learning_rate": 0.0002, "epoch": 4.415410385259632, "step": 6590}, {"loss": 1.3995, "grad_norm": 0.8408458828926086, "learning_rate": 0.0002, "epoch": 4.422110552763819, "step": 6600}, {"loss": 1.4148, "grad_norm": 0.8110666275024414, "learning_rate": 0.0002, "epoch": 4.4288107202680065, "step": 6610}, {"loss": 1.441, "grad_norm": 0.8602406978607178, "learning_rate": 0.0002, "epoch": 4.435510887772194, "step": 6620}, {"loss": 1.4319, "grad_norm": 0.7549102902412415, "learning_rate": 0.0002, "epoch": 4.442211055276382, "step": 6630}, {"loss": 1.388, "grad_norm": 0.7831804156303406, "learning_rate": 0.0002, "epoch": 4.44891122278057, "step": 6640}, {"loss": 1.4283, "grad_norm": 0.7269673943519592, "learning_rate": 0.0002, "epoch": 4.455611390284757, "step": 6650}, {"loss": 1.4132, "grad_norm": 0.7397838830947876, "learning_rate": 0.0002, "epoch": 4.4623115577889445, "step": 6660}, {"loss": 1.3174, "grad_norm": 0.713707447052002, "learning_rate": 0.0002, "epoch": 4.469011725293132, "step": 6670}, {"loss": 1.3406, "grad_norm": 0.7525581121444702, "learning_rate": 0.0002, "epoch": 4.47571189279732, "step": 6680}, {"loss": 1.4283, "grad_norm": 0.8030191659927368, "learning_rate": 0.0002, "epoch": 4.482412060301508, "step": 6690}, {"loss": 1.4586, "grad_norm": 0.7469439506530762, "learning_rate": 0.0002, "epoch": 4.489112227805695, "step": 6700}, {"loss": 1.367, "grad_norm": 0.7743868231773376, "learning_rate": 0.0002, "epoch": 4.4958123953098825, "step": 6710}, {"loss": 1.3439, "grad_norm": 0.6539737582206726, "learning_rate": 0.0002, "epoch": 4.50251256281407, "step": 6720}, {"loss": 1.4513, "grad_norm": 0.825818657875061, "learning_rate": 0.0002, "epoch": 4.509212730318258, "step": 6730}, {"loss": 1.3984, "grad_norm": 0.8048575520515442, "learning_rate": 0.0002, "epoch": 4.515912897822446, "step": 6740}, {"loss": 1.3923, "grad_norm": 0.7828766107559204, "learning_rate": 0.0002, "epoch": 4.522613065326633, "step": 6750}, {"loss": 1.3886, "grad_norm": 0.7406010031700134, "learning_rate": 0.0002, "epoch": 4.5293132328308205, "step": 6760}, {"loss": 1.3109, "grad_norm": 0.840345561504364, "learning_rate": 0.0002, "epoch": 4.536013400335008, "step": 6770}, {"loss": 1.4808, "grad_norm": 0.8492622971534729, "learning_rate": 0.0002, "epoch": 4.542713567839196, "step": 6780}, {"loss": 1.4384, "grad_norm": 0.7130163908004761, "learning_rate": 0.0002, "epoch": 4.549413735343384, "step": 6790}, {"loss": 1.4531, "grad_norm": 0.8454728126525879, "learning_rate": 0.0002, "epoch": 4.556113902847571, "step": 6800}, {"loss": 1.3239, "grad_norm": 0.7847645282745361, "learning_rate": 0.0002, "epoch": 4.562814070351759, "step": 6810}, {"loss": 1.4181, "grad_norm": 0.7245864272117615, "learning_rate": 0.0002, "epoch": 4.569514237855946, "step": 6820}, {"loss": 1.3233, "grad_norm": 0.768893301486969, "learning_rate": 0.0002, "epoch": 4.576214405360134, "step": 6830}, {"loss": 1.3932, "grad_norm": 0.8028400540351868, "learning_rate": 0.0002, "epoch": 4.582914572864322, "step": 6840}, {"loss": 1.3745, "grad_norm": 0.763945460319519, "learning_rate": 0.0002, "epoch": 4.589614740368509, "step": 6850}, {"loss": 1.4797, "grad_norm": 0.7417685389518738, "learning_rate": 0.0002, "epoch": 4.596314907872697, "step": 6860}, {"loss": 1.4468, "grad_norm": 0.7603038549423218, "learning_rate": 0.0002, "epoch": 4.603015075376884, "step": 6870}, {"loss": 1.4095, "grad_norm": 0.7981528043746948, "learning_rate": 0.0002, "epoch": 4.609715242881072, "step": 6880}, {"loss": 1.3963, "grad_norm": 0.8077111840248108, "learning_rate": 0.0002, "epoch": 4.61641541038526, "step": 6890}, {"loss": 1.4721, "grad_norm": 0.8778454065322876, "learning_rate": 0.0002, "epoch": 4.623115577889447, "step": 6900}, {"loss": 1.3758, "grad_norm": 0.8620710372924805, "learning_rate": 0.0002, "epoch": 4.629815745393635, "step": 6910}, {"loss": 1.344, "grad_norm": 0.7486072778701782, "learning_rate": 0.0002, "epoch": 4.636515912897822, "step": 6920}, {"loss": 1.3913, "grad_norm": 0.7493042945861816, "learning_rate": 0.0002, "epoch": 4.64321608040201, "step": 6930}, {"loss": 1.397, "grad_norm": 0.7388978600502014, "learning_rate": 0.0002, "epoch": 4.649916247906198, "step": 6940}, {"loss": 1.3593, "grad_norm": 0.798530638217926, "learning_rate": 0.0002, "epoch": 4.656616415410385, "step": 6950}, {"loss": 1.3982, "grad_norm": 0.7929500937461853, "learning_rate": 0.0002, "epoch": 4.663316582914573, "step": 6960}, {"loss": 1.4183, "grad_norm": 0.9186785221099854, "learning_rate": 0.0002, "epoch": 4.67001675041876, "step": 6970}, {"loss": 1.3955, "grad_norm": 1.1103485822677612, "learning_rate": 0.0002, "epoch": 4.676716917922948, "step": 6980}, {"loss": 1.3941, "grad_norm": 0.8000466823577881, "learning_rate": 0.0002, "epoch": 4.683417085427136, "step": 6990}, {"loss": 1.371, "grad_norm": 0.7520599961280823, "learning_rate": 0.0002, "epoch": 4.690117252931323, "step": 7000}, {"loss": 1.4582, "grad_norm": 0.7971973419189453, "learning_rate": 0.0002, "epoch": 4.696817420435511, "step": 7010}, {"loss": 1.3682, "grad_norm": 0.7363343834877014, "learning_rate": 0.0002, "epoch": 4.703517587939698, "step": 7020}, {"loss": 1.3889, "grad_norm": 0.8268865942955017, "learning_rate": 0.0002, "epoch": 4.710217755443886, "step": 7030}, {"loss": 1.4382, "grad_norm": 0.7054963111877441, "learning_rate": 0.0002, "epoch": 4.716917922948074, "step": 7040}, {"loss": 1.4578, "grad_norm": 0.8196262121200562, "learning_rate": 0.0002, "epoch": 4.723618090452261, "step": 7050}, {"loss": 1.365, "grad_norm": 0.8276031017303467, "learning_rate": 0.0002, "epoch": 4.730318257956449, "step": 7060}, {"loss": 1.3887, "grad_norm": 0.8248157501220703, "learning_rate": 0.0002, "epoch": 4.7370184254606365, "step": 7070}, {"loss": 1.4193, "grad_norm": 0.8937979936599731, "learning_rate": 0.0002, "epoch": 4.743718592964824, "step": 7080}, {"loss": 1.4334, "grad_norm": 0.7984238862991333, "learning_rate": 0.0002, "epoch": 4.750418760469012, "step": 7090}, {"loss": 1.4385, "grad_norm": 0.9495313763618469, "learning_rate": 0.0002, "epoch": 4.757118927973199, "step": 7100}, {"loss": 1.4504, "grad_norm": 0.8598204255104065, "learning_rate": 0.0002, "epoch": 4.763819095477387, "step": 7110}, {"loss": 1.3969, "grad_norm": 0.8951472640037537, "learning_rate": 0.0002, "epoch": 4.7705192629815745, "step": 7120}, {"loss": 1.4339, "grad_norm": 0.9110309481620789, "learning_rate": 0.0002, "epoch": 4.777219430485762, "step": 7130}, {"loss": 1.4001, "grad_norm": 0.7929584980010986, "learning_rate": 0.0002, "epoch": 4.78391959798995, "step": 7140}, {"loss": 1.467, "grad_norm": 0.7415322661399841, "learning_rate": 0.0002, "epoch": 4.790619765494137, "step": 7150}, {"loss": 1.5107, "grad_norm": 0.7504757046699524, "learning_rate": 0.0002, "epoch": 4.797319932998325, "step": 7160}, {"loss": 1.3736, "grad_norm": 0.7166924476623535, "learning_rate": 0.0002, "epoch": 4.8040201005025125, "step": 7170}, {"loss": 1.4088, "grad_norm": 0.7728400826454163, "learning_rate": 0.0002, "epoch": 4.8107202680067, "step": 7180}, {"loss": 1.3814, "grad_norm": 0.7992154955863953, "learning_rate": 0.0002, "epoch": 4.817420435510888, "step": 7190}, {"loss": 1.3958, "grad_norm": 0.8655321002006531, "learning_rate": 0.0002, "epoch": 4.824120603015075, "step": 7200}, {"loss": 1.3837, "grad_norm": 0.7672632336616516, "learning_rate": 0.0002, "epoch": 4.830820770519263, "step": 7210}, {"loss": 1.4578, "grad_norm": 0.708416223526001, "learning_rate": 0.0002, "epoch": 4.8375209380234505, "step": 7220}, {"loss": 1.5413, "grad_norm": 0.8914081454277039, "learning_rate": 0.0002, "epoch": 4.844221105527638, "step": 7230}, {"loss": 1.3569, "grad_norm": 0.7141931653022766, "learning_rate": 0.0002, "epoch": 4.850921273031826, "step": 7240}, {"loss": 1.4532, "grad_norm": 0.6913040280342102, "learning_rate": 0.0002, "epoch": 4.857621440536013, "step": 7250}, {"loss": 1.3912, "grad_norm": 0.7871233820915222, "learning_rate": 0.0002, "epoch": 4.864321608040201, "step": 7260}, {"loss": 1.3688, "grad_norm": 0.8466277122497559, "learning_rate": 0.0002, "epoch": 4.8710217755443885, "step": 7270}, {"loss": 1.33, "grad_norm": 0.8492183685302734, "learning_rate": 0.0002, "epoch": 4.877721943048576, "step": 7280}, {"loss": 1.3744, "grad_norm": 0.8339574933052063, "learning_rate": 0.0002, "epoch": 4.884422110552764, "step": 7290}, {"loss": 1.4157, "grad_norm": 0.787022590637207, "learning_rate": 0.0002, "epoch": 4.891122278056951, "step": 7300}, {"loss": 1.3725, "grad_norm": 0.8877332806587219, "learning_rate": 0.0002, "epoch": 4.897822445561139, "step": 7310}, {"loss": 1.3968, "grad_norm": 0.744989812374115, "learning_rate": 0.0002, "epoch": 4.9045226130653266, "step": 7320}, {"loss": 1.4421, "grad_norm": 0.8027268648147583, "learning_rate": 0.0002, "epoch": 4.911222780569514, "step": 7330}, {"loss": 1.425, "grad_norm": 0.6437455415725708, "learning_rate": 0.0002, "epoch": 4.917922948073702, "step": 7340}, {"loss": 1.4829, "grad_norm": 0.685999870300293, "learning_rate": 0.0002, "epoch": 4.924623115577889, "step": 7350}, {"loss": 1.4352, "grad_norm": 0.9086187481880188, "learning_rate": 0.0002, "epoch": 4.931323283082077, "step": 7360}, {"loss": 1.4245, "grad_norm": 0.8272411227226257, "learning_rate": 0.0002, "epoch": 4.938023450586265, "step": 7370}, {"loss": 1.4226, "grad_norm": 0.9227852821350098, "learning_rate": 0.0002, "epoch": 4.944723618090452, "step": 7380}, {"loss": 1.3643, "grad_norm": 0.7688441276550293, "learning_rate": 0.0002, "epoch": 4.95142378559464, "step": 7390}, {"loss": 1.4491, "grad_norm": 0.8662643432617188, "learning_rate": 0.0002, "epoch": 4.958123953098827, "step": 7400}, {"loss": 1.4194, "grad_norm": 0.9234127998352051, "learning_rate": 0.0002, "epoch": 4.964824120603015, "step": 7410}, {"loss": 1.4009, "grad_norm": 0.9131470918655396, "learning_rate": 0.0002, "epoch": 4.971524288107203, "step": 7420}, {"loss": 1.4544, "grad_norm": 0.7377504110336304, "learning_rate": 0.0002, "epoch": 4.97822445561139, "step": 7430}, {"loss": 1.4008, "grad_norm": 0.8762801289558411, "learning_rate": 0.0002, "epoch": 4.984924623115578, "step": 7440}, {"loss": 1.4304, "grad_norm": 0.7919872999191284, "learning_rate": 0.0002, "epoch": 4.991624790619765, "step": 7450}, {"loss": 1.3817, "grad_norm": 0.7144299149513245, "learning_rate": 0.0002, "epoch": 4.998324958123953, "step": 7460}]}
+{"epoch": 6.0, "step": 8955, "epoch_duration": 1643.898383140564, "total_accumulated_duration": 9601.341349124908, "gpu_info": {"GPU_0": "NVIDIA A100-SXM4-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 7568.4541015625}, "peak_memory_usage": {"GPU_0": 13792.75439453125}, "avg_memory_reserved": {"GPU_0": 17416.0}, "peak_memory_reserved": {"GPU_0": 17416.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "outputs-001/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-2985", "params": {"epochs": 8, "batch_size": 1, "learning_rate": 0.0002, "gradient_accumulation_steps": 2, "warmup_ratio": 0.03, "max_grad_norm": 0.3, "lora_alpha": 32, "lora_dropout": 0.05, "lora_r": 64, "tf32": true, "seed": 42}, "log_history": [{"loss": 2.6189, "grad_norm": 0.565915048122406, "learning_rate": 0.0002, "epoch": 0.006700167504187605, "step": 10}, {"loss": 2.3162, "grad_norm": 0.5004463791847229, "learning_rate": 0.0002, "epoch": 0.01340033500837521, "step": 20}, {"loss": 2.0576, "grad_norm": 0.511043906211853, "learning_rate": 0.0002, "epoch": 0.020100502512562814, "step": 30}, {"loss": 2.0085, "grad_norm": 0.47327178716659546, "learning_rate": 0.0002, "epoch": 0.02680067001675042, "step": 40}, {"loss": 2.0276, "grad_norm": 0.5511676669120789, "learning_rate": 0.0002, "epoch": 0.03350083752093802, "step": 50}, {"loss": 1.9075, "grad_norm": 0.4666278064250946, "learning_rate": 0.0002, "epoch": 0.04020100502512563, "step": 60}, {"loss": 1.8413, "grad_norm": 0.5310961008071899, "learning_rate": 0.0002, "epoch": 0.04690117252931323, "step": 70}, {"loss": 1.8711, "grad_norm": 0.5606027245521545, "learning_rate": 0.0002, "epoch": 0.05360134003350084, "step": 80}, {"loss": 1.9282, "grad_norm": 0.4934779703617096, "learning_rate": 0.0002, "epoch": 0.06030150753768844, "step": 90}, {"loss": 1.8925, "grad_norm": 0.4821869730949402, "learning_rate": 0.0002, "epoch": 0.06700167504187604, "step": 100}, {"loss": 1.8628, "grad_norm": 0.5262084603309631, "learning_rate": 0.0002, "epoch": 0.07370184254606366, "step": 110}, {"loss": 1.8347, "grad_norm": 0.3774230182170868, "learning_rate": 0.0002, "epoch": 0.08040201005025126, "step": 120}, {"loss": 1.8386, "grad_norm": 0.34137430787086487, "learning_rate": 0.0002, "epoch": 0.08710217755443886, "step": 130}, {"loss": 1.861, "grad_norm": 0.407272070646286, "learning_rate": 0.0002, "epoch": 0.09380234505862646, "step": 140}, {"loss": 1.8279, "grad_norm": 0.4011937975883484, "learning_rate": 0.0002, "epoch": 0.10050251256281408, "step": 150}, {"loss": 1.9317, "grad_norm": 0.4432467222213745, "learning_rate": 0.0002, "epoch": 0.10720268006700168, "step": 160}, {"loss": 1.8157, "grad_norm": 0.44030463695526123, "learning_rate": 0.0002, "epoch": 0.11390284757118928, "step": 170}, {"loss": 1.8534, "grad_norm": 0.3799569308757782, "learning_rate": 0.0002, "epoch": 0.12060301507537688, "step": 180}, {"loss": 1.7658, "grad_norm": 0.33721521496772766, "learning_rate": 0.0002, "epoch": 0.1273031825795645, "step": 190}, {"loss": 1.8269, "grad_norm": 0.4096226692199707, "learning_rate": 0.0002, "epoch": 0.13400335008375208, "step": 200}, {"loss": 1.802, "grad_norm": 0.37374693155288696, "learning_rate": 0.0002, "epoch": 0.1407035175879397, "step": 210}, {"loss": 1.8901, "grad_norm": 0.3249480128288269, "learning_rate": 0.0002, "epoch": 0.1474036850921273, "step": 220}, {"loss": 1.8163, "grad_norm": 0.3612042963504791, "learning_rate": 0.0002, "epoch": 0.1541038525963149, "step": 230}, {"loss": 1.7585, "grad_norm": 0.3686671257019043, "learning_rate": 0.0002, "epoch": 0.16080402010050251, "step": 240}, {"loss": 1.8365, "grad_norm": 0.3521044850349426, "learning_rate": 0.0002, "epoch": 0.16750418760469013, "step": 250}, {"loss": 1.8623, "grad_norm": 0.4073677361011505, "learning_rate": 0.0002, "epoch": 0.17420435510887772, "step": 260}, {"loss": 1.8026, "grad_norm": 0.34522193670272827, "learning_rate": 0.0002, "epoch": 0.18090452261306533, "step": 270}, {"loss": 1.8162, "grad_norm": 0.4121900498867035, "learning_rate": 0.0002, "epoch": 0.18760469011725292, "step": 280}, {"loss": 1.7976, "grad_norm": 0.3544778525829315, "learning_rate": 0.0002, "epoch": 0.19430485762144054, "step": 290}, {"loss": 1.8787, "grad_norm": 0.3482133448123932, "learning_rate": 0.0002, "epoch": 0.20100502512562815, "step": 300}, {"loss": 1.8578, "grad_norm": 0.3421826660633087, "learning_rate": 0.0002, "epoch": 0.20770519262981574, "step": 310}, {"loss": 1.8013, "grad_norm": 0.5024696588516235, "learning_rate": 0.0002, "epoch": 0.21440536013400335, "step": 320}, {"loss": 1.8607, "grad_norm": 0.36013063788414, "learning_rate": 0.0002, "epoch": 0.22110552763819097, "step": 330}, {"loss": 1.9075, "grad_norm": 0.3611244857311249, "learning_rate": 0.0002, "epoch": 0.22780569514237856, "step": 340}, {"loss": 1.8128, "grad_norm": 0.39244529604911804, "learning_rate": 0.0002, "epoch": 0.23450586264656617, "step": 350}, {"loss": 1.7885, "grad_norm": 0.3299325704574585, "learning_rate": 0.0002, "epoch": 0.24120603015075376, "step": 360}, {"loss": 1.8028, "grad_norm": 0.3994322419166565, "learning_rate": 0.0002, "epoch": 0.24790619765494137, "step": 370}, {"loss": 1.8321, "grad_norm": 0.3559151887893677, "learning_rate": 0.0002, "epoch": 0.254606365159129, "step": 380}, {"loss": 1.7802, "grad_norm": 0.3873756229877472, "learning_rate": 0.0002, "epoch": 0.2613065326633166, "step": 390}, {"loss": 1.7844, "grad_norm": 0.3710744082927704, "learning_rate": 0.0002, "epoch": 0.26800670016750416, "step": 400}, {"loss": 1.7789, "grad_norm": 0.3618465065956116, "learning_rate": 0.0002, "epoch": 0.2747068676716918, "step": 410}, {"loss": 1.8529, "grad_norm": 0.30063769221305847, "learning_rate": 0.0002, "epoch": 0.2814070351758794, "step": 420}, {"loss": 1.7765, "grad_norm": 0.3695628345012665, "learning_rate": 0.0002, "epoch": 0.288107202680067, "step": 430}, {"loss": 1.7982, "grad_norm": 0.31451135873794556, "learning_rate": 0.0002, "epoch": 0.2948073701842546, "step": 440}, {"loss": 1.7517, "grad_norm": 0.3959707021713257, "learning_rate": 0.0002, "epoch": 0.3015075376884422, "step": 450}, {"loss": 1.8142, "grad_norm": 0.33475354313850403, "learning_rate": 0.0002, "epoch": 0.3082077051926298, "step": 460}, {"loss": 1.8805, "grad_norm": 0.33933115005493164, "learning_rate": 0.0002, "epoch": 0.3149078726968174, "step": 470}, {"loss": 1.7564, "grad_norm": 0.3264943063259125, "learning_rate": 0.0002, "epoch": 0.32160804020100503, "step": 480}, {"loss": 1.8428, "grad_norm": 0.40188100934028625, "learning_rate": 0.0002, "epoch": 0.32830820770519265, "step": 490}, {"loss": 1.7624, "grad_norm": 0.37408649921417236, "learning_rate": 0.0002, "epoch": 0.33500837520938026, "step": 500}, {"loss": 1.7745, "grad_norm": 0.33925938606262207, "learning_rate": 0.0002, "epoch": 0.3417085427135678, "step": 510}, {"loss": 1.814, "grad_norm": 0.36836713552474976, "learning_rate": 0.0002, "epoch": 0.34840871021775544, "step": 520}, {"loss": 1.8037, "grad_norm": 0.37284499406814575, "learning_rate": 0.0002, "epoch": 0.35510887772194305, "step": 530}, {"loss": 1.8379, "grad_norm": 0.3192278742790222, "learning_rate": 0.0002, "epoch": 0.36180904522613067, "step": 540}, {"loss": 1.8702, "grad_norm": 0.30233290791511536, "learning_rate": 0.0002, "epoch": 0.3685092127303183, "step": 550}, {"loss": 1.8799, "grad_norm": 0.3340817391872406, "learning_rate": 0.0002, "epoch": 0.37520938023450584, "step": 560}, {"loss": 1.8404, "grad_norm": 0.32600095868110657, "learning_rate": 0.0002, "epoch": 0.38190954773869346, "step": 570}, {"loss": 1.7804, "grad_norm": 0.33711278438568115, "learning_rate": 0.0002, "epoch": 0.38860971524288107, "step": 580}, {"loss": 1.8445, "grad_norm": 0.34890690445899963, "learning_rate": 0.0002, "epoch": 0.3953098827470687, "step": 590}, {"loss": 1.8187, "grad_norm": 0.38238924741744995, "learning_rate": 0.0002, "epoch": 0.4020100502512563, "step": 600}, {"loss": 1.8111, "grad_norm": 0.34399354457855225, "learning_rate": 0.0002, "epoch": 0.40871021775544386, "step": 610}, {"loss": 1.8006, "grad_norm": 0.3346073627471924, "learning_rate": 0.0002, "epoch": 0.4154103852596315, "step": 620}, {"loss": 1.7705, "grad_norm": 0.3545648157596588, "learning_rate": 0.0002, "epoch": 0.4221105527638191, "step": 630}, {"loss": 1.8445, "grad_norm": 0.3378899097442627, "learning_rate": 0.0002, "epoch": 0.4288107202680067, "step": 640}, {"loss": 1.804, "grad_norm": 0.3255569040775299, "learning_rate": 0.0002, "epoch": 0.4355108877721943, "step": 650}, {"loss": 1.7679, "grad_norm": 0.34880587458610535, "learning_rate": 0.0002, "epoch": 0.44221105527638194, "step": 660}, {"loss": 1.7861, "grad_norm": 0.3402383625507355, "learning_rate": 0.0002, "epoch": 0.4489112227805695, "step": 670}, {"loss": 1.8131, "grad_norm": 0.3594033718109131, "learning_rate": 0.0002, "epoch": 0.4556113902847571, "step": 680}, {"loss": 1.8399, "grad_norm": 0.31000566482543945, "learning_rate": 0.0002, "epoch": 0.4623115577889447, "step": 690}, {"loss": 1.7521, "grad_norm": 0.37229061126708984, "learning_rate": 0.0002, "epoch": 0.46901172529313234, "step": 700}, {"loss": 1.7779, "grad_norm": 0.315801739692688, "learning_rate": 0.0002, "epoch": 0.47571189279731996, "step": 710}, {"loss": 1.7515, "grad_norm": 0.3220832645893097, "learning_rate": 0.0002, "epoch": 0.4824120603015075, "step": 720}, {"loss": 1.7181, "grad_norm": 0.3435456156730652, "learning_rate": 0.0002, "epoch": 0.48911222780569513, "step": 730}, {"loss": 1.8844, "grad_norm": 0.30380892753601074, "learning_rate": 0.0002, "epoch": 0.49581239530988275, "step": 740}, {"loss": 1.7792, "grad_norm": 0.3555026054382324, "learning_rate": 0.0002, "epoch": 0.5025125628140703, "step": 750}, {"loss": 1.7714, "grad_norm": 0.3019855320453644, "learning_rate": 0.0002, "epoch": 0.509212730318258, "step": 760}, {"loss": 1.7962, "grad_norm": 0.309111088514328, "learning_rate": 0.0002, "epoch": 0.5159128978224455, "step": 770}, {"loss": 1.7913, "grad_norm": 0.366020530462265, "learning_rate": 0.0002, "epoch": 0.5226130653266332, "step": 780}, {"loss": 1.8008, "grad_norm": 0.3267050087451935, "learning_rate": 0.0002, "epoch": 0.5293132328308208, "step": 790}, {"loss": 1.7397, "grad_norm": 0.34265750646591187, "learning_rate": 0.0002, "epoch": 0.5360134003350083, "step": 800}, {"loss": 1.8251, "grad_norm": 0.313669890165329, "learning_rate": 0.0002, "epoch": 0.542713567839196, "step": 810}, {"loss": 1.8369, "grad_norm": 0.3355236053466797, "learning_rate": 0.0002, "epoch": 0.5494137353433836, "step": 820}, {"loss": 1.7381, "grad_norm": 0.3186608552932739, "learning_rate": 0.0002, "epoch": 0.5561139028475712, "step": 830}, {"loss": 1.8034, "grad_norm": 0.30357518792152405, "learning_rate": 0.0002, "epoch": 0.5628140703517588, "step": 840}, {"loss": 1.769, "grad_norm": 0.3990040123462677, "learning_rate": 0.0002, "epoch": 0.5695142378559463, "step": 850}, {"loss": 1.7482, "grad_norm": 0.34363803267478943, "learning_rate": 0.0002, "epoch": 0.576214405360134, "step": 860}, {"loss": 1.8106, "grad_norm": 0.3757908046245575, "learning_rate": 0.0002, "epoch": 0.5829145728643216, "step": 870}, {"loss": 1.8104, "grad_norm": 0.3359757661819458, "learning_rate": 0.0002, "epoch": 0.5896147403685092, "step": 880}, {"loss": 1.7591, "grad_norm": 0.5555329918861389, "learning_rate": 0.0002, "epoch": 0.5963149078726968, "step": 890}, {"loss": 1.7715, "grad_norm": 0.4046323895454407, "learning_rate": 0.0002, "epoch": 0.6030150753768844, "step": 900}, {"loss": 1.7998, "grad_norm": 0.29834219813346863, "learning_rate": 0.0002, "epoch": 0.609715242881072, "step": 910}, {"loss": 1.7826, "grad_norm": 0.3241238594055176, "learning_rate": 0.0002, "epoch": 0.6164154103852596, "step": 920}, {"loss": 1.8342, "grad_norm": 0.35154739022254944, "learning_rate": 0.0002, "epoch": 0.6231155778894473, "step": 930}, {"loss": 1.8076, "grad_norm": 0.3287706673145294, "learning_rate": 0.0002, "epoch": 0.6298157453936348, "step": 940}, {"loss": 1.8038, "grad_norm": 0.35670626163482666, "learning_rate": 0.0002, "epoch": 0.6365159128978225, "step": 950}, {"loss": 1.869, "grad_norm": 0.6114104986190796, "learning_rate": 0.0002, "epoch": 0.6432160804020101, "step": 960}, {"loss": 1.8297, "grad_norm": 0.3186565041542053, "learning_rate": 0.0002, "epoch": 0.6499162479061976, "step": 970}, {"loss": 1.7539, "grad_norm": 0.27164125442504883, "learning_rate": 0.0002, "epoch": 0.6566164154103853, "step": 980}, {"loss": 1.8339, "grad_norm": 0.34407344460487366, "learning_rate": 0.0002, "epoch": 0.6633165829145728, "step": 990}, {"loss": 1.855, "grad_norm": 0.368415892124176, "learning_rate": 0.0002, "epoch": 0.6700167504187605, "step": 1000}, {"loss": 1.7821, "grad_norm": 0.3306390643119812, "learning_rate": 0.0002, "epoch": 0.6767169179229481, "step": 1010}, {"loss": 1.786, "grad_norm": 0.3198648989200592, "learning_rate": 0.0002, "epoch": 0.6834170854271356, "step": 1020}, {"loss": 1.816, "grad_norm": 0.3092987537384033, "learning_rate": 0.0002, "epoch": 0.6901172529313233, "step": 1030}, {"loss": 1.7689, "grad_norm": 0.3090653419494629, "learning_rate": 0.0002, "epoch": 0.6968174204355109, "step": 1040}, {"loss": 1.7544, "grad_norm": 0.3485880196094513, "learning_rate": 0.0002, "epoch": 0.7035175879396985, "step": 1050}, {"loss": 1.8187, "grad_norm": 0.35782721638679504, "learning_rate": 0.0002, "epoch": 0.7102177554438861, "step": 1060}, {"loss": 1.806, "grad_norm": 0.34256869554519653, "learning_rate": 0.0002, "epoch": 0.7169179229480737, "step": 1070}, {"loss": 1.7873, "grad_norm": 0.30461037158966064, "learning_rate": 0.0002, "epoch": 0.7236180904522613, "step": 1080}, {"loss": 1.7367, "grad_norm": 0.3398691713809967, "learning_rate": 0.0002, "epoch": 0.7303182579564489, "step": 1090}, {"loss": 1.8756, "grad_norm": 0.3180808126926422, "learning_rate": 0.0002, "epoch": 0.7370184254606366, "step": 1100}, {"loss": 1.6988, "grad_norm": 0.34400665760040283, "learning_rate": 0.0002, "epoch": 0.7437185929648241, "step": 1110}, {"loss": 1.7851, "grad_norm": 0.34244877099990845, "learning_rate": 0.0002, "epoch": 0.7504187604690117, "step": 1120}, {"loss": 1.7841, "grad_norm": 0.29946693778038025, "learning_rate": 0.0002, "epoch": 0.7571189279731994, "step": 1130}, {"loss": 1.7456, "grad_norm": 0.37547236680984497, "learning_rate": 0.0002, "epoch": 0.7638190954773869, "step": 1140}, {"loss": 1.8425, "grad_norm": 0.3263005018234253, "learning_rate": 0.0002, "epoch": 0.7705192629815746, "step": 1150}, {"loss": 1.7222, "grad_norm": 0.41363608837127686, "learning_rate": 0.0002, "epoch": 0.7772194304857621, "step": 1160}, {"loss": 1.7836, "grad_norm": 0.36267954111099243, "learning_rate": 0.0002, "epoch": 0.7839195979899497, "step": 1170}, {"loss": 1.9183, "grad_norm": 0.31789499521255493, "learning_rate": 0.0002, "epoch": 0.7906197654941374, "step": 1180}, {"loss": 1.78, "grad_norm": 0.5708149075508118, "learning_rate": 0.0002, "epoch": 0.7973199329983249, "step": 1190}, {"loss": 1.6908, "grad_norm": 0.322099506855011, "learning_rate": 0.0002, "epoch": 0.8040201005025126, "step": 1200}, {"loss": 1.7639, "grad_norm": 0.3419909179210663, "learning_rate": 0.0002, "epoch": 0.8107202680067002, "step": 1210}, {"loss": 1.7428, "grad_norm": 0.36286255717277527, "learning_rate": 0.0002, "epoch": 0.8174204355108877, "step": 1220}, {"loss": 1.8409, "grad_norm": 0.33992862701416016, "learning_rate": 0.0002, "epoch": 0.8241206030150754, "step": 1230}, {"loss": 1.7507, "grad_norm": 0.32622793316841125, "learning_rate": 0.0002, "epoch": 0.830820770519263, "step": 1240}, {"loss": 1.8098, "grad_norm": 0.3036167621612549, "learning_rate": 0.0002, "epoch": 0.8375209380234506, "step": 1250}, {"loss": 1.8094, "grad_norm": 0.3182215392589569, "learning_rate": 0.0002, "epoch": 0.8442211055276382, "step": 1260}, {"loss": 1.8017, "grad_norm": 0.3270018696784973, "learning_rate": 0.0002, "epoch": 0.8509212730318257, "step": 1270}, {"loss": 1.798, "grad_norm": 0.32652342319488525, "learning_rate": 0.0002, "epoch": 0.8576214405360134, "step": 1280}, {"loss": 1.7448, "grad_norm": 0.3631329834461212, "learning_rate": 0.0002, "epoch": 0.864321608040201, "step": 1290}, {"loss": 1.7, "grad_norm": 0.36706018447875977, "learning_rate": 0.0002, "epoch": 0.8710217755443886, "step": 1300}, {"loss": 1.8178, "grad_norm": 0.3347418010234833, "learning_rate": 0.0002, "epoch": 0.8777219430485762, "step": 1310}, {"loss": 1.7824, "grad_norm": 0.34371060132980347, "learning_rate": 0.0002, "epoch": 0.8844221105527639, "step": 1320}, {"loss": 1.783, "grad_norm": 0.3029090166091919, "learning_rate": 0.0002, "epoch": 0.8911222780569514, "step": 1330}, {"loss": 1.8017, "grad_norm": 0.34700682759284973, "learning_rate": 0.0002, "epoch": 0.897822445561139, "step": 1340}, {"loss": 1.7998, "grad_norm": 0.35574328899383545, "learning_rate": 0.0002, "epoch": 0.9045226130653267, "step": 1350}, {"loss": 1.7339, "grad_norm": 0.30928221344947815, "learning_rate": 0.0002, "epoch": 0.9112227805695142, "step": 1360}, {"loss": 1.7479, "grad_norm": 0.30652928352355957, "learning_rate": 0.0002, "epoch": 0.9179229480737019, "step": 1370}, {"loss": 1.7491, "grad_norm": 0.3838157653808594, "learning_rate": 0.0002, "epoch": 0.9246231155778895, "step": 1380}, {"loss": 1.7977, "grad_norm": 0.31655240058898926, "learning_rate": 0.0002, "epoch": 0.931323283082077, "step": 1390}, {"loss": 1.8175, "grad_norm": 0.41737303137779236, "learning_rate": 0.0002, "epoch": 0.9380234505862647, "step": 1400}, {"loss": 1.6811, "grad_norm": 0.3227267861366272, "learning_rate": 0.0002, "epoch": 0.9447236180904522, "step": 1410}, {"loss": 1.7343, "grad_norm": 0.3729925751686096, "learning_rate": 0.0002, "epoch": 0.9514237855946399, "step": 1420}, {"loss": 1.8221, "grad_norm": 0.30779409408569336, "learning_rate": 0.0002, "epoch": 0.9581239530988275, "step": 1430}, {"loss": 1.7972, "grad_norm": 0.334379643201828, "learning_rate": 0.0002, "epoch": 0.964824120603015, "step": 1440}, {"loss": 1.7141, "grad_norm": 0.3568236231803894, "learning_rate": 0.0002, "epoch": 0.9715242881072027, "step": 1450}, {"loss": 1.7541, "grad_norm": 0.33310577273368835, "learning_rate": 0.0002, "epoch": 0.9782244556113903, "step": 1460}, {"loss": 1.8511, "grad_norm": 0.2972261905670166, "learning_rate": 0.0002, "epoch": 0.9849246231155779, "step": 1470}, {"loss": 1.7654, "grad_norm": 0.3322717845439911, "learning_rate": 0.0002, "epoch": 0.9916247906197655, "step": 1480}, {"loss": 1.8033, "grad_norm": 0.3276330828666687, "learning_rate": 0.0002, "epoch": 0.998324958123953, "step": 1490}, {"eval_loss": 1.8036354780197144, "eval_runtime": 37.8949, "eval_samples_per_second": 13.59, "eval_steps_per_second": 1.715, "epoch": 0.9996649916247906, "step": 1492}, {"loss": 1.7138, "grad_norm": 0.29252371191978455, "learning_rate": 0.0002, "epoch": 1.0050251256281406, "step": 1500}, {"loss": 1.8198, "grad_norm": 0.31607162952423096, "learning_rate": 0.0002, "epoch": 1.0117252931323284, "step": 1510}, {"loss": 1.6779, "grad_norm": 0.32294467091560364, "learning_rate": 0.0002, "epoch": 1.018425460636516, "step": 1520}, {"loss": 1.7919, "grad_norm": 0.3868017792701721, "learning_rate": 0.0002, "epoch": 1.0251256281407035, "step": 1530}, {"loss": 1.7954, "grad_norm": 0.3178282082080841, "learning_rate": 0.0002, "epoch": 1.031825795644891, "step": 1540}, {"loss": 1.7136, "grad_norm": 0.3706750273704529, "learning_rate": 0.0002, "epoch": 1.0385259631490786, "step": 1550}, {"loss": 1.7382, "grad_norm": 0.33930912613868713, "learning_rate": 0.0002, "epoch": 1.0452261306532664, "step": 1560}, {"loss": 1.7602, "grad_norm": 0.33970504999160767, "learning_rate": 0.0002, "epoch": 1.051926298157454, "step": 1570}, {"loss": 1.6573, "grad_norm": 0.42553383111953735, "learning_rate": 0.0002, "epoch": 1.0586264656616415, "step": 1580}, {"loss": 1.645, "grad_norm": 0.3772421181201935, "learning_rate": 0.0002, "epoch": 1.065326633165829, "step": 1590}, {"loss": 1.7362, "grad_norm": 0.34212902188301086, "learning_rate": 0.0002, "epoch": 1.0720268006700167, "step": 1600}, {"loss": 1.7057, "grad_norm": 0.3798283338546753, "learning_rate": 0.0002, "epoch": 1.0787269681742044, "step": 1610}, {"loss": 1.7468, "grad_norm": 0.36909598112106323, "learning_rate": 0.0002, "epoch": 1.085427135678392, "step": 1620}, {"loss": 1.7807, "grad_norm": 0.3344230651855469, "learning_rate": 0.0002, "epoch": 1.0921273031825796, "step": 1630}, {"loss": 1.7111, "grad_norm": 0.3862569332122803, "learning_rate": 0.0002, "epoch": 1.0988274706867671, "step": 1640}, {"loss": 1.7163, "grad_norm": 0.31188511848449707, "learning_rate": 0.0002, "epoch": 1.1055276381909547, "step": 1650}, {"loss": 1.7263, "grad_norm": 0.3563670814037323, "learning_rate": 0.0002, "epoch": 1.1122278056951425, "step": 1660}, {"loss": 1.7718, "grad_norm": 0.35052165389060974, "learning_rate": 0.0002, "epoch": 1.11892797319933, "step": 1670}, {"loss": 1.7601, "grad_norm": 0.3285699188709259, "learning_rate": 0.0002, "epoch": 1.1256281407035176, "step": 1680}, {"loss": 1.6877, "grad_norm": 0.3639393746852875, "learning_rate": 0.0002, "epoch": 1.1323283082077051, "step": 1690}, {"loss": 1.7719, "grad_norm": 0.3842753767967224, "learning_rate": 0.0002, "epoch": 1.1390284757118927, "step": 1700}, {"loss": 1.7002, "grad_norm": 0.3624933063983917, "learning_rate": 0.0002, "epoch": 1.1457286432160805, "step": 1710}, {"loss": 1.7243, "grad_norm": 0.3641220033168793, "learning_rate": 0.0002, "epoch": 1.152428810720268, "step": 1720}, {"loss": 1.752, "grad_norm": 0.32765355706214905, "learning_rate": 0.0002, "epoch": 1.1591289782244556, "step": 1730}, {"loss": 1.6556, "grad_norm": 0.34974896907806396, "learning_rate": 0.0002, "epoch": 1.1658291457286432, "step": 1740}, {"loss": 1.7273, "grad_norm": 0.3910926580429077, "learning_rate": 0.0002, "epoch": 1.1725293132328307, "step": 1750}, {"loss": 1.7204, "grad_norm": 0.3564300537109375, "learning_rate": 0.0002, "epoch": 1.1792294807370185, "step": 1760}, {"loss": 1.746, "grad_norm": 0.34822574257850647, "learning_rate": 0.0002, "epoch": 1.185929648241206, "step": 1770}, {"loss": 1.7256, "grad_norm": 0.36185044050216675, "learning_rate": 0.0002, "epoch": 1.1926298157453936, "step": 1780}, {"loss": 1.6431, "grad_norm": 0.34866711497306824, "learning_rate": 0.0002, "epoch": 1.1993299832495812, "step": 1790}, {"loss": 1.8084, "grad_norm": 0.4017769992351532, "learning_rate": 0.0002, "epoch": 1.2060301507537687, "step": 1800}, {"loss": 1.6985, "grad_norm": 0.32930681109428406, "learning_rate": 0.0002, "epoch": 1.2127303182579565, "step": 1810}, {"loss": 1.7606, "grad_norm": 0.35951921343803406, "learning_rate": 0.0002, "epoch": 1.219430485762144, "step": 1820}, {"loss": 1.6933, "grad_norm": 0.37366992235183716, "learning_rate": 0.0002, "epoch": 1.2261306532663316, "step": 1830}, {"loss": 1.6737, "grad_norm": 0.3565689027309418, "learning_rate": 0.0002, "epoch": 1.2328308207705192, "step": 1840}, {"loss": 1.8013, "grad_norm": 0.3692343533039093, "learning_rate": 0.0002, "epoch": 1.2395309882747068, "step": 1850}, {"loss": 1.736, "grad_norm": 0.38426971435546875, "learning_rate": 0.0002, "epoch": 1.2462311557788945, "step": 1860}, {"loss": 1.7031, "grad_norm": 0.33559855818748474, "learning_rate": 0.0002, "epoch": 1.252931323283082, "step": 1870}, {"loss": 1.7033, "grad_norm": 0.34181106090545654, "learning_rate": 0.0002, "epoch": 1.2596314907872697, "step": 1880}, {"loss": 1.7707, "grad_norm": 0.3916318416595459, "learning_rate": 0.0002, "epoch": 1.2663316582914572, "step": 1890}, {"loss": 1.6686, "grad_norm": 0.3887825012207031, "learning_rate": 0.0002, "epoch": 1.2730318257956448, "step": 1900}, {"loss": 1.7062, "grad_norm": 0.33583927154541016, "learning_rate": 0.0002, "epoch": 1.2797319932998326, "step": 1910}, {"loss": 1.717, "grad_norm": 0.37639349699020386, "learning_rate": 0.0002, "epoch": 1.2864321608040201, "step": 1920}, {"loss": 1.777, "grad_norm": 0.38059428334236145, "learning_rate": 0.0002, "epoch": 1.2931323283082077, "step": 1930}, {"loss": 1.6126, "grad_norm": 0.37253183126449585, "learning_rate": 0.0002, "epoch": 1.2998324958123952, "step": 1940}, {"loss": 1.6758, "grad_norm": 0.37371566891670227, "learning_rate": 0.0002, "epoch": 1.3065326633165828, "step": 1950}, {"loss": 1.6788, "grad_norm": 0.4080910086631775, "learning_rate": 0.0002, "epoch": 1.3132328308207706, "step": 1960}, {"loss": 1.6518, "grad_norm": 0.3174354135990143, "learning_rate": 0.0002, "epoch": 1.3199329983249581, "step": 1970}, {"loss": 1.7925, "grad_norm": 0.4518888294696808, "learning_rate": 0.0002, "epoch": 1.3266331658291457, "step": 1980}, {"loss": 1.7085, "grad_norm": 0.3627921938896179, "learning_rate": 0.0002, "epoch": 1.3333333333333333, "step": 1990}, {"loss": 1.7676, "grad_norm": 0.3655930161476135, "learning_rate": 0.0002, "epoch": 1.3400335008375208, "step": 2000}, {"loss": 1.7016, "grad_norm": 0.3509993255138397, "learning_rate": 0.0002, "epoch": 1.3467336683417086, "step": 2010}, {"loss": 1.7359, "grad_norm": 0.4281129240989685, "learning_rate": 0.0002, "epoch": 1.3534338358458962, "step": 2020}, {"loss": 1.6884, "grad_norm": 0.3821414113044739, "learning_rate": 0.0002, "epoch": 1.3601340033500837, "step": 2030}, {"loss": 1.7075, "grad_norm": 0.3907586336135864, "learning_rate": 0.0002, "epoch": 1.3668341708542713, "step": 2040}, {"loss": 1.7424, "grad_norm": 0.37792932987213135, "learning_rate": 0.0002, "epoch": 1.3735343383584588, "step": 2050}, {"loss": 1.7305, "grad_norm": 0.3693985641002655, "learning_rate": 0.0002, "epoch": 1.3802345058626466, "step": 2060}, {"loss": 1.7434, "grad_norm": 0.32275936007499695, "learning_rate": 0.0002, "epoch": 1.3869346733668342, "step": 2070}, {"loss": 1.6677, "grad_norm": 0.3789440095424652, "learning_rate": 0.0002, "epoch": 1.3936348408710217, "step": 2080}, {"loss": 1.6825, "grad_norm": 0.3638380467891693, "learning_rate": 0.0002, "epoch": 1.4003350083752093, "step": 2090}, {"loss": 1.6542, "grad_norm": 0.3495481610298157, "learning_rate": 0.0002, "epoch": 1.4070351758793969, "step": 2100}, {"loss": 1.7225, "grad_norm": 0.37920597195625305, "learning_rate": 0.0002, "epoch": 1.4137353433835846, "step": 2110}, {"loss": 1.7329, "grad_norm": 0.37218064069747925, "learning_rate": 0.0002, "epoch": 1.4204355108877722, "step": 2120}, {"loss": 1.799, "grad_norm": 0.38074082136154175, "learning_rate": 0.0002, "epoch": 1.4271356783919598, "step": 2130}, {"loss": 1.7403, "grad_norm": 0.3455527126789093, "learning_rate": 0.0002, "epoch": 1.4338358458961473, "step": 2140}, {"loss": 1.776, "grad_norm": 0.3712003529071808, "learning_rate": 0.0002, "epoch": 1.4405360134003349, "step": 2150}, {"loss": 1.7619, "grad_norm": 0.3786754906177521, "learning_rate": 0.0002, "epoch": 1.4472361809045227, "step": 2160}, {"loss": 1.68, "grad_norm": 0.3879223167896271, "learning_rate": 0.0002, "epoch": 1.4539363484087102, "step": 2170}, {"loss": 1.7, "grad_norm": 0.38738805055618286, "learning_rate": 0.0002, "epoch": 1.4606365159128978, "step": 2180}, {"loss": 1.7581, "grad_norm": 0.39768800139427185, "learning_rate": 0.0002, "epoch": 1.4673366834170856, "step": 2190}, {"loss": 1.7671, "grad_norm": 0.4172441065311432, "learning_rate": 0.0002, "epoch": 1.474036850921273, "step": 2200}, {"loss": 1.6736, "grad_norm": 0.4043174982070923, "learning_rate": 0.0002, "epoch": 1.4807370184254607, "step": 2210}, {"loss": 1.7444, "grad_norm": 0.3750883936882019, "learning_rate": 0.0002, "epoch": 1.4874371859296482, "step": 2220}, {"loss": 1.6861, "grad_norm": 0.3552253246307373, "learning_rate": 0.0002, "epoch": 1.4941373534338358, "step": 2230}, {"loss": 1.6471, "grad_norm": 0.34607139229774475, "learning_rate": 0.0002, "epoch": 1.5008375209380236, "step": 2240}, {"loss": 1.6962, "grad_norm": 0.3406706750392914, "learning_rate": 0.0002, "epoch": 1.507537688442211, "step": 2250}, {"loss": 1.7694, "grad_norm": 0.36654895544052124, "learning_rate": 0.0002, "epoch": 1.5142378559463987, "step": 2260}, {"loss": 1.6812, "grad_norm": 0.3914054334163666, "learning_rate": 0.0002, "epoch": 1.5209380234505863, "step": 2270}, {"loss": 1.6822, "grad_norm": 0.42012137174606323, "learning_rate": 0.0002, "epoch": 1.5276381909547738, "step": 2280}, {"loss": 1.697, "grad_norm": 0.39563435316085815, "learning_rate": 0.0002, "epoch": 1.5343383584589616, "step": 2290}, {"loss": 1.7491, "grad_norm": 0.3508438766002655, "learning_rate": 0.0002, "epoch": 1.541038525963149, "step": 2300}, {"loss": 1.7727, "grad_norm": 0.3785218596458435, "learning_rate": 0.0002, "epoch": 1.5477386934673367, "step": 2310}, {"loss": 1.6963, "grad_norm": 0.39377647638320923, "learning_rate": 0.0002, "epoch": 1.5544388609715243, "step": 2320}, {"loss": 1.7263, "grad_norm": 0.3391438126564026, "learning_rate": 0.0002, "epoch": 1.5611390284757118, "step": 2330}, {"loss": 1.7722, "grad_norm": 0.37944263219833374, "learning_rate": 0.0002, "epoch": 1.5678391959798996, "step": 2340}, {"loss": 1.6371, "grad_norm": 0.3523491322994232, "learning_rate": 0.0002, "epoch": 1.574539363484087, "step": 2350}, {"loss": 1.7583, "grad_norm": 0.3911575973033905, "learning_rate": 0.0002, "epoch": 1.5812395309882747, "step": 2360}, {"loss": 1.7117, "grad_norm": 0.33832186460494995, "learning_rate": 0.0002, "epoch": 1.5879396984924623, "step": 2370}, {"loss": 1.7701, "grad_norm": 0.3665979206562042, "learning_rate": 0.0002, "epoch": 1.5946398659966499, "step": 2380}, {"loss": 1.779, "grad_norm": 0.3871748149394989, "learning_rate": 0.0002, "epoch": 1.6013400335008376, "step": 2390}, {"loss": 1.7109, "grad_norm": 0.3586967885494232, "learning_rate": 0.0002, "epoch": 1.608040201005025, "step": 2400}, {"loss": 1.7096, "grad_norm": 0.3563673198223114, "learning_rate": 0.0002, "epoch": 1.6147403685092128, "step": 2410}, {"loss": 1.745, "grad_norm": 0.37588971853256226, "learning_rate": 0.0002, "epoch": 1.6214405360134003, "step": 2420}, {"loss": 1.7086, "grad_norm": 0.352556437253952, "learning_rate": 0.0002, "epoch": 1.6281407035175879, "step": 2430}, {"loss": 1.6547, "grad_norm": 0.3716259300708771, "learning_rate": 0.0002, "epoch": 1.6348408710217757, "step": 2440}, {"loss": 1.7033, "grad_norm": 0.372001975774765, "learning_rate": 0.0002, "epoch": 1.641541038525963, "step": 2450}, {"loss": 1.6584, "grad_norm": 0.3430042862892151, "learning_rate": 0.0002, "epoch": 1.6482412060301508, "step": 2460}, {"loss": 1.7217, "grad_norm": 0.3741483688354492, "learning_rate": 0.0002, "epoch": 1.6549413735343383, "step": 2470}, {"loss": 1.7701, "grad_norm": 0.3610571324825287, "learning_rate": 0.0002, "epoch": 1.661641541038526, "step": 2480}, {"loss": 1.7057, "grad_norm": 0.4204719066619873, "learning_rate": 0.0002, "epoch": 1.6683417085427137, "step": 2490}, {"loss": 1.7954, "grad_norm": 0.3938186466693878, "learning_rate": 0.0002, "epoch": 1.675041876046901, "step": 2500}, {"loss": 1.6633, "grad_norm": 0.3421435058116913, "learning_rate": 0.0002, "epoch": 1.6817420435510888, "step": 2510}, {"loss": 1.7996, "grad_norm": 0.42441412806510925, "learning_rate": 0.0002, "epoch": 1.6884422110552764, "step": 2520}, {"loss": 1.7142, "grad_norm": 0.38071519136428833, "learning_rate": 0.0002, "epoch": 1.695142378559464, "step": 2530}, {"loss": 1.7232, "grad_norm": 0.34078919887542725, "learning_rate": 0.0002, "epoch": 1.7018425460636517, "step": 2540}, {"loss": 1.7126, "grad_norm": 0.412844181060791, "learning_rate": 0.0002, "epoch": 1.708542713567839, "step": 2550}, {"loss": 1.7149, "grad_norm": 0.3753604292869568, "learning_rate": 0.0002, "epoch": 1.7152428810720268, "step": 2560}, {"loss": 1.7011, "grad_norm": 0.41588476300239563, "learning_rate": 0.0002, "epoch": 1.7219430485762144, "step": 2570}, {"loss": 1.6427, "grad_norm": 0.35504111647605896, "learning_rate": 0.0002, "epoch": 1.728643216080402, "step": 2580}, {"loss": 1.7296, "grad_norm": 0.36909720301628113, "learning_rate": 0.0002, "epoch": 1.7353433835845897, "step": 2590}, {"loss": 1.7022, "grad_norm": 0.4149979054927826, "learning_rate": 0.0002, "epoch": 1.742043551088777, "step": 2600}, {"loss": 1.77, "grad_norm": 0.38859328627586365, "learning_rate": 0.0002, "epoch": 1.7487437185929648, "step": 2610}, {"loss": 1.7036, "grad_norm": 0.36738792061805725, "learning_rate": 0.0002, "epoch": 1.7554438860971524, "step": 2620}, {"loss": 1.764, "grad_norm": 0.3968178927898407, "learning_rate": 0.0002, "epoch": 1.76214405360134, "step": 2630}, {"loss": 1.7687, "grad_norm": 0.3972901999950409, "learning_rate": 0.0002, "epoch": 1.7688442211055277, "step": 2640}, {"loss": 1.6774, "grad_norm": 0.3949959874153137, "learning_rate": 0.0002, "epoch": 1.775544388609715, "step": 2650}, {"loss": 1.7247, "grad_norm": 0.44074657559394836, "learning_rate": 0.0002, "epoch": 1.7822445561139029, "step": 2660}, {"loss": 1.7188, "grad_norm": 0.39743664860725403, "learning_rate": 0.0002, "epoch": 1.7889447236180904, "step": 2670}, {"loss": 1.7258, "grad_norm": 0.3950406610965729, "learning_rate": 0.0002, "epoch": 1.795644891122278, "step": 2680}, {"loss": 1.6906, "grad_norm": 0.3568263649940491, "learning_rate": 0.0002, "epoch": 1.8023450586264658, "step": 2690}, {"loss": 1.6735, "grad_norm": 0.3819476366043091, "learning_rate": 0.0002, "epoch": 1.809045226130653, "step": 2700}, {"loss": 1.7198, "grad_norm": 0.3480634391307831, "learning_rate": 0.0002, "epoch": 1.8157453936348409, "step": 2710}, {"loss": 1.7042, "grad_norm": 0.3875853419303894, "learning_rate": 0.0002, "epoch": 1.8224455611390284, "step": 2720}, {"loss": 1.6988, "grad_norm": 0.3441337049007416, "learning_rate": 0.0002, "epoch": 1.829145728643216, "step": 2730}, {"loss": 1.7647, "grad_norm": 0.35692882537841797, "learning_rate": 0.0002, "epoch": 1.8358458961474038, "step": 2740}, {"loss": 1.7033, "grad_norm": 0.36959215998649597, "learning_rate": 0.0002, "epoch": 1.8425460636515911, "step": 2750}, {"loss": 1.7657, "grad_norm": 0.3893393278121948, "learning_rate": 0.0002, "epoch": 1.849246231155779, "step": 2760}, {"loss": 1.7068, "grad_norm": 0.37817293405532837, "learning_rate": 0.0002, "epoch": 1.8559463986599665, "step": 2770}, {"loss": 1.761, "grad_norm": 0.36071285605430603, "learning_rate": 0.0002, "epoch": 1.862646566164154, "step": 2780}, {"loss": 1.7623, "grad_norm": 0.3758420944213867, "learning_rate": 0.0002, "epoch": 1.8693467336683418, "step": 2790}, {"loss": 1.6743, "grad_norm": 0.3889938294887543, "learning_rate": 0.0002, "epoch": 1.8760469011725294, "step": 2800}, {"loss": 1.6151, "grad_norm": 0.34361857175827026, "learning_rate": 0.0002, "epoch": 1.882747068676717, "step": 2810}, {"loss": 1.6038, "grad_norm": 0.39283323287963867, "learning_rate": 0.0002, "epoch": 1.8894472361809045, "step": 2820}, {"loss": 1.7555, "grad_norm": 0.3919452726840973, "learning_rate": 0.0002, "epoch": 1.896147403685092, "step": 2830}, {"loss": 1.673, "grad_norm": 0.38215070962905884, "learning_rate": 0.0002, "epoch": 1.9028475711892798, "step": 2840}, {"loss": 1.7044, "grad_norm": 0.4235064387321472, "learning_rate": 0.0002, "epoch": 1.9095477386934674, "step": 2850}, {"loss": 1.7123, "grad_norm": 0.35694634914398193, "learning_rate": 0.0002, "epoch": 1.916247906197655, "step": 2860}, {"loss": 1.8128, "grad_norm": 0.383492112159729, "learning_rate": 0.0002, "epoch": 1.9229480737018425, "step": 2870}, {"loss": 1.7581, "grad_norm": 0.5945147275924683, "learning_rate": 0.0002, "epoch": 1.92964824120603, "step": 2880}, {"loss": 1.7421, "grad_norm": 0.3367522358894348, "learning_rate": 0.0002, "epoch": 1.9363484087102178, "step": 2890}, {"loss": 1.6561, "grad_norm": 0.35300394892692566, "learning_rate": 0.0002, "epoch": 1.9430485762144054, "step": 2900}, {"loss": 1.7033, "grad_norm": 0.38084495067596436, "learning_rate": 0.0002, "epoch": 1.949748743718593, "step": 2910}, {"loss": 1.7132, "grad_norm": 0.37559160590171814, "learning_rate": 0.0002, "epoch": 1.9564489112227805, "step": 2920}, {"loss": 1.6759, "grad_norm": 0.3661738336086273, "learning_rate": 0.0002, "epoch": 1.963149078726968, "step": 2930}, {"loss": 1.7643, "grad_norm": 0.4073849320411682, "learning_rate": 0.0002, "epoch": 1.9698492462311559, "step": 2940}, {"loss": 1.6806, "grad_norm": 0.3723304271697998, "learning_rate": 0.0002, "epoch": 1.9765494137353434, "step": 2950}, {"loss": 1.7611, "grad_norm": 0.3991098999977112, "learning_rate": 0.0002, "epoch": 1.983249581239531, "step": 2960}, {"loss": 1.7263, "grad_norm": 0.3947085440158844, "learning_rate": 0.0002, "epoch": 1.9899497487437185, "step": 2970}, {"loss": 1.7217, "grad_norm": 0.3786258399486542, "learning_rate": 0.0002, "epoch": 1.996649916247906, "step": 2980}, {"eval_loss": 1.8028968572616577, "eval_runtime": 37.8985, "eval_samples_per_second": 13.589, "eval_steps_per_second": 1.715, "epoch": 2.0, "step": 2985}, {"loss": 1.695, "grad_norm": 0.34824079275131226, "learning_rate": 0.0002, "epoch": 2.003350083752094, "step": 2990}, {"loss": 1.5853, "grad_norm": 0.3394894003868103, "learning_rate": 0.0002, "epoch": 2.0100502512562812, "step": 3000}, {"loss": 1.5783, "grad_norm": 0.36910977959632874, "learning_rate": 0.0002, "epoch": 2.016750418760469, "step": 3010}, {"loss": 1.6105, "grad_norm": 0.45000967383384705, "learning_rate": 0.0002, "epoch": 2.023450586264657, "step": 3020}, {"loss": 1.6019, "grad_norm": 0.3791407346725464, "learning_rate": 0.0002, "epoch": 2.030150753768844, "step": 3030}, {"loss": 1.5832, "grad_norm": 0.387321799993515, "learning_rate": 0.0002, "epoch": 2.036850921273032, "step": 3040}, {"loss": 1.6834, "grad_norm": 0.4185757040977478, "learning_rate": 0.0002, "epoch": 2.0435510887772192, "step": 3050}, {"loss": 1.5696, "grad_norm": 0.45110777020454407, "learning_rate": 0.0002, "epoch": 2.050251256281407, "step": 3060}, {"loss": 1.6231, "grad_norm": 0.42663660645484924, "learning_rate": 0.0002, "epoch": 2.056951423785595, "step": 3070}, {"loss": 1.6279, "grad_norm": 0.4546292722225189, "learning_rate": 0.0002, "epoch": 2.063651591289782, "step": 3080}, {"loss": 1.6141, "grad_norm": 0.3979759216308594, "learning_rate": 0.0002, "epoch": 2.07035175879397, "step": 3090}, {"loss": 1.6343, "grad_norm": 0.43596673011779785, "learning_rate": 0.0002, "epoch": 2.0770519262981573, "step": 3100}, {"loss": 1.5441, "grad_norm": 0.40120232105255127, "learning_rate": 0.0002, "epoch": 2.083752093802345, "step": 3110}, {"loss": 1.6309, "grad_norm": 0.44449281692504883, "learning_rate": 0.0002, "epoch": 2.090452261306533, "step": 3120}, {"loss": 1.5652, "grad_norm": 0.42672568559646606, "learning_rate": 0.0002, "epoch": 2.09715242881072, "step": 3130}, {"loss": 1.682, "grad_norm": 0.4232690930366516, "learning_rate": 0.0002, "epoch": 2.103852596314908, "step": 3140}, {"loss": 1.624, "grad_norm": 0.4299317002296448, "learning_rate": 0.0002, "epoch": 2.1105527638190953, "step": 3150}, {"loss": 1.6766, "grad_norm": 0.4067758023738861, "learning_rate": 0.0002, "epoch": 2.117252931323283, "step": 3160}, {"loss": 1.6759, "grad_norm": 0.4918815791606903, "learning_rate": 0.0002, "epoch": 2.123953098827471, "step": 3170}, {"loss": 1.6478, "grad_norm": 0.4140559732913971, "learning_rate": 0.0002, "epoch": 2.130653266331658, "step": 3180}, {"loss": 1.6641, "grad_norm": 0.4555995464324951, "learning_rate": 0.0002, "epoch": 2.137353433835846, "step": 3190}, {"loss": 1.5888, "grad_norm": 0.42943915724754333, "learning_rate": 0.0002, "epoch": 2.1440536013400333, "step": 3200}, {"loss": 1.5886, "grad_norm": 0.4730435013771057, "learning_rate": 0.0002, "epoch": 2.150753768844221, "step": 3210}, {"loss": 1.6022, "grad_norm": 0.43310216069221497, "learning_rate": 0.0002, "epoch": 2.157453936348409, "step": 3220}, {"loss": 1.6058, "grad_norm": 0.42054110765457153, "learning_rate": 0.0002, "epoch": 2.164154103852596, "step": 3230}, {"loss": 1.6749, "grad_norm": 0.4897233247756958, "learning_rate": 0.0002, "epoch": 2.170854271356784, "step": 3240}, {"loss": 1.6983, "grad_norm": 0.42194533348083496, "learning_rate": 0.0002, "epoch": 2.1775544388609713, "step": 3250}, {"loss": 1.6159, "grad_norm": 0.44494450092315674, "learning_rate": 0.0002, "epoch": 2.184254606365159, "step": 3260}, {"loss": 1.6977, "grad_norm": 0.43524879217147827, "learning_rate": 0.0002, "epoch": 2.190954773869347, "step": 3270}, {"loss": 1.528, "grad_norm": 0.4621117413043976, "learning_rate": 0.0002, "epoch": 2.1976549413735342, "step": 3280}, {"loss": 1.632, "grad_norm": 0.4073285460472107, "learning_rate": 0.0002, "epoch": 2.204355108877722, "step": 3290}, {"loss": 1.6141, "grad_norm": 0.47868335247039795, "learning_rate": 0.0002, "epoch": 2.2110552763819094, "step": 3300}, {"loss": 1.6857, "grad_norm": 0.4264970123767853, "learning_rate": 0.0002, "epoch": 2.217755443886097, "step": 3310}, {"loss": 1.5653, "grad_norm": 0.4491245150566101, "learning_rate": 0.0002, "epoch": 2.224455611390285, "step": 3320}, {"loss": 1.5881, "grad_norm": 0.4010344445705414, "learning_rate": 0.0002, "epoch": 2.2311557788944723, "step": 3330}, {"loss": 1.6684, "grad_norm": 0.4232759177684784, "learning_rate": 0.0002, "epoch": 2.23785594639866, "step": 3340}, {"loss": 1.6336, "grad_norm": 0.5099776983261108, "learning_rate": 0.0002, "epoch": 2.2445561139028474, "step": 3350}, {"loss": 1.6764, "grad_norm": 0.5223407745361328, "learning_rate": 0.0002, "epoch": 2.251256281407035, "step": 3360}, {"loss": 1.6625, "grad_norm": 0.47818470001220703, "learning_rate": 0.0002, "epoch": 2.257956448911223, "step": 3370}, {"loss": 1.5946, "grad_norm": 0.4721255898475647, "learning_rate": 0.0002, "epoch": 2.2646566164154103, "step": 3380}, {"loss": 1.5568, "grad_norm": 0.4113229513168335, "learning_rate": 0.0002, "epoch": 2.271356783919598, "step": 3390}, {"loss": 1.6494, "grad_norm": 0.507080078125, "learning_rate": 0.0002, "epoch": 2.2780569514237854, "step": 3400}, {"loss": 1.6183, "grad_norm": 0.4852292239665985, "learning_rate": 0.0002, "epoch": 2.284757118927973, "step": 3410}, {"loss": 1.6132, "grad_norm": 0.4503684341907501, "learning_rate": 0.0002, "epoch": 2.291457286432161, "step": 3420}, {"loss": 1.6649, "grad_norm": 0.8359600305557251, "learning_rate": 0.0002, "epoch": 2.2981574539363483, "step": 3430}, {"loss": 1.6644, "grad_norm": 0.44604045152664185, "learning_rate": 0.0002, "epoch": 2.304857621440536, "step": 3440}, {"loss": 1.5972, "grad_norm": 0.45667049288749695, "learning_rate": 0.0002, "epoch": 2.3115577889447234, "step": 3450}, {"loss": 1.6667, "grad_norm": 0.4879349172115326, "learning_rate": 0.0002, "epoch": 2.318257956448911, "step": 3460}, {"loss": 1.5804, "grad_norm": 0.4033963084220886, "learning_rate": 0.0002, "epoch": 2.324958123953099, "step": 3470}, {"loss": 1.5838, "grad_norm": 0.44494301080703735, "learning_rate": 0.0002, "epoch": 2.3316582914572863, "step": 3480}, {"loss": 1.5951, "grad_norm": 0.4794621765613556, "learning_rate": 0.0002, "epoch": 2.338358458961474, "step": 3490}, {"loss": 1.6807, "grad_norm": 0.41404327750205994, "learning_rate": 0.0002, "epoch": 2.3450586264656614, "step": 3500}, {"loss": 1.714, "grad_norm": 0.4664851725101471, "learning_rate": 0.0002, "epoch": 2.351758793969849, "step": 3510}, {"loss": 1.6537, "grad_norm": 0.4263697564601898, "learning_rate": 0.0002, "epoch": 2.358458961474037, "step": 3520}, {"loss": 1.6551, "grad_norm": 0.5035167336463928, "learning_rate": 0.0002, "epoch": 2.3651591289782243, "step": 3530}, {"loss": 1.6208, "grad_norm": 0.4380664527416229, "learning_rate": 0.0002, "epoch": 2.371859296482412, "step": 3540}, {"loss": 1.634, "grad_norm": 0.5227681994438171, "learning_rate": 0.0002, "epoch": 2.3785594639865995, "step": 3550}, {"loss": 1.6146, "grad_norm": 0.4382302761077881, "learning_rate": 0.0002, "epoch": 2.3852596314907872, "step": 3560}, {"loss": 1.5653, "grad_norm": 0.4392451047897339, "learning_rate": 0.0002, "epoch": 2.391959798994975, "step": 3570}, {"loss": 1.6626, "grad_norm": 0.4372786581516266, "learning_rate": 0.0002, "epoch": 2.3986599664991624, "step": 3580}, {"loss": 1.519, "grad_norm": 0.5015502572059631, "learning_rate": 0.0002, "epoch": 2.40536013400335, "step": 3590}, {"loss": 1.588, "grad_norm": 0.5653210878372192, "learning_rate": 0.0002, "epoch": 2.4120603015075375, "step": 3600}, {"loss": 1.6075, "grad_norm": 0.53007972240448, "learning_rate": 0.0002, "epoch": 2.4187604690117253, "step": 3610}, {"loss": 1.6421, "grad_norm": 0.4659176766872406, "learning_rate": 0.0002, "epoch": 2.425460636515913, "step": 3620}, {"loss": 1.625, "grad_norm": 0.5637837052345276, "learning_rate": 0.0002, "epoch": 2.4321608040201004, "step": 3630}, {"loss": 1.6168, "grad_norm": 0.4248391389846802, "learning_rate": 0.0002, "epoch": 2.438860971524288, "step": 3640}, {"loss": 1.6822, "grad_norm": 0.44668248295783997, "learning_rate": 0.0002, "epoch": 2.4455611390284755, "step": 3650}, {"loss": 1.6417, "grad_norm": 0.43990179896354675, "learning_rate": 0.0002, "epoch": 2.4522613065326633, "step": 3660}, {"loss": 1.6723, "grad_norm": 0.4532523453235626, "learning_rate": 0.0002, "epoch": 2.458961474036851, "step": 3670}, {"loss": 1.6957, "grad_norm": 0.6605591773986816, "learning_rate": 0.0002, "epoch": 2.4656616415410384, "step": 3680}, {"loss": 1.6159, "grad_norm": 0.4694533348083496, "learning_rate": 0.0002, "epoch": 2.472361809045226, "step": 3690}, {"loss": 1.6239, "grad_norm": 0.4485011100769043, "learning_rate": 0.0002, "epoch": 2.4790619765494135, "step": 3700}, {"loss": 1.6834, "grad_norm": 0.4761785864830017, "learning_rate": 0.0002, "epoch": 2.4857621440536013, "step": 3710}, {"loss": 1.6313, "grad_norm": 0.5116432309150696, "learning_rate": 0.0002, "epoch": 2.492462311557789, "step": 3720}, {"loss": 1.5054, "grad_norm": 0.49523618817329407, "learning_rate": 0.0002, "epoch": 2.4991624790619764, "step": 3730}, {"loss": 1.6249, "grad_norm": 0.43826380372047424, "learning_rate": 0.0002, "epoch": 2.505862646566164, "step": 3740}, {"loss": 1.5762, "grad_norm": 0.4916154146194458, "learning_rate": 0.0002, "epoch": 2.5125628140703515, "step": 3750}, {"loss": 1.5157, "grad_norm": 0.5381299257278442, "learning_rate": 0.0002, "epoch": 2.5192629815745393, "step": 3760}, {"loss": 1.6467, "grad_norm": 0.44947415590286255, "learning_rate": 0.0002, "epoch": 2.525963149078727, "step": 3770}, {"loss": 1.67, "grad_norm": 0.49979084730148315, "learning_rate": 0.0002, "epoch": 2.5326633165829144, "step": 3780}, {"loss": 1.622, "grad_norm": 0.43046900629997253, "learning_rate": 0.0002, "epoch": 2.539363484087102, "step": 3790}, {"loss": 1.6789, "grad_norm": 0.4513470530509949, "learning_rate": 0.0002, "epoch": 2.5460636515912896, "step": 3800}, {"loss": 1.6335, "grad_norm": 0.49900051951408386, "learning_rate": 0.0002, "epoch": 2.5527638190954773, "step": 3810}, {"loss": 1.6322, "grad_norm": 0.4348420202732086, "learning_rate": 0.0002, "epoch": 2.559463986599665, "step": 3820}, {"loss": 1.6218, "grad_norm": 0.4684867560863495, "learning_rate": 0.0002, "epoch": 2.5661641541038525, "step": 3830}, {"loss": 1.6535, "grad_norm": 0.44430989027023315, "learning_rate": 0.0002, "epoch": 2.5728643216080402, "step": 3840}, {"loss": 1.5909, "grad_norm": 0.47375255823135376, "learning_rate": 0.0002, "epoch": 2.5795644891122276, "step": 3850}, {"loss": 1.6269, "grad_norm": 0.45493075251579285, "learning_rate": 0.0002, "epoch": 2.5862646566164154, "step": 3860}, {"loss": 1.604, "grad_norm": 0.4563275873661041, "learning_rate": 0.0002, "epoch": 2.592964824120603, "step": 3870}, {"loss": 1.642, "grad_norm": 0.46060335636138916, "learning_rate": 0.0002, "epoch": 2.5996649916247905, "step": 3880}, {"loss": 1.6302, "grad_norm": 0.4718867540359497, "learning_rate": 0.0002, "epoch": 2.6063651591289783, "step": 3890}, {"loss": 1.6242, "grad_norm": 0.41570305824279785, "learning_rate": 0.0002, "epoch": 2.6130653266331656, "step": 3900}, {"loss": 1.6401, "grad_norm": 0.4603121876716614, "learning_rate": 0.0002, "epoch": 2.6197654941373534, "step": 3910}, {"loss": 1.6839, "grad_norm": 0.4734652638435364, "learning_rate": 0.0002, "epoch": 2.626465661641541, "step": 3920}, {"loss": 1.5448, "grad_norm": 0.45348483324050903, "learning_rate": 0.0002, "epoch": 2.6331658291457285, "step": 3930}, {"loss": 1.6157, "grad_norm": 0.46559447050094604, "learning_rate": 0.0002, "epoch": 2.6398659966499163, "step": 3940}, {"loss": 1.7052, "grad_norm": 0.44113144278526306, "learning_rate": 0.0002, "epoch": 2.6465661641541036, "step": 3950}, {"loss": 1.6315, "grad_norm": 0.41415104269981384, "learning_rate": 0.0002, "epoch": 2.6532663316582914, "step": 3960}, {"loss": 1.6589, "grad_norm": 0.48868080973625183, "learning_rate": 0.0002, "epoch": 2.659966499162479, "step": 3970}, {"loss": 1.6211, "grad_norm": 0.49610549211502075, "learning_rate": 0.0002, "epoch": 2.6666666666666665, "step": 3980}, {"loss": 1.6235, "grad_norm": 0.4309130907058716, "learning_rate": 0.0002, "epoch": 2.6733668341708543, "step": 3990}, {"loss": 1.6452, "grad_norm": 0.4489327669143677, "learning_rate": 0.0002, "epoch": 2.6800670016750416, "step": 4000}, {"loss": 1.5744, "grad_norm": 0.5380139946937561, "learning_rate": 0.0002, "epoch": 2.6867671691792294, "step": 4010}, {"loss": 1.6524, "grad_norm": 0.5076672434806824, "learning_rate": 0.0002, "epoch": 2.693467336683417, "step": 4020}, {"loss": 1.636, "grad_norm": 0.47620031237602234, "learning_rate": 0.0002, "epoch": 2.7001675041876045, "step": 4030}, {"loss": 1.5543, "grad_norm": 0.48089155554771423, "learning_rate": 0.0002, "epoch": 2.7068676716917923, "step": 4040}, {"loss": 1.6396, "grad_norm": 0.5108814239501953, "learning_rate": 0.0002, "epoch": 2.7135678391959797, "step": 4050}, {"loss": 1.5905, "grad_norm": 0.4196513295173645, "learning_rate": 0.0002, "epoch": 2.7202680067001674, "step": 4060}, {"loss": 1.686, "grad_norm": 0.4574664831161499, "learning_rate": 0.0002, "epoch": 2.726968174204355, "step": 4070}, {"loss": 1.6234, "grad_norm": 0.4671640992164612, "learning_rate": 0.0002, "epoch": 2.7336683417085426, "step": 4080}, {"loss": 1.6827, "grad_norm": 0.49355530738830566, "learning_rate": 0.0002, "epoch": 2.7403685092127303, "step": 4090}, {"loss": 1.6999, "grad_norm": 0.46716663241386414, "learning_rate": 0.0002, "epoch": 2.7470686767169177, "step": 4100}, {"loss": 1.6463, "grad_norm": 0.45420581102371216, "learning_rate": 0.0002, "epoch": 2.7537688442211055, "step": 4110}, {"loss": 1.5718, "grad_norm": 0.4680487811565399, "learning_rate": 0.0002, "epoch": 2.7604690117252932, "step": 4120}, {"loss": 1.5968, "grad_norm": 0.5375032424926758, "learning_rate": 0.0002, "epoch": 2.7671691792294806, "step": 4130}, {"loss": 1.5254, "grad_norm": 0.46026280522346497, "learning_rate": 0.0002, "epoch": 2.7738693467336684, "step": 4140}, {"loss": 1.6613, "grad_norm": 0.43658447265625, "learning_rate": 0.0002, "epoch": 2.7805695142378557, "step": 4150}, {"loss": 1.6546, "grad_norm": 0.4935547113418579, "learning_rate": 0.0002, "epoch": 2.7872696817420435, "step": 4160}, {"loss": 1.5961, "grad_norm": 0.8167962431907654, "learning_rate": 0.0002, "epoch": 2.7939698492462313, "step": 4170}, {"loss": 1.6907, "grad_norm": 0.4289683997631073, "learning_rate": 0.0002, "epoch": 2.8006700167504186, "step": 4180}, {"loss": 1.6385, "grad_norm": 0.4569324254989624, "learning_rate": 0.0002, "epoch": 2.8073701842546064, "step": 4190}, {"loss": 1.6077, "grad_norm": 0.474795937538147, "learning_rate": 0.0002, "epoch": 2.8140703517587937, "step": 4200}, {"loss": 1.6223, "grad_norm": 0.44272229075431824, "learning_rate": 0.0002, "epoch": 2.8207705192629815, "step": 4210}, {"loss": 1.6706, "grad_norm": 0.525240957736969, "learning_rate": 0.0002, "epoch": 2.8274706867671693, "step": 4220}, {"loss": 1.7196, "grad_norm": 0.4802303910255432, "learning_rate": 0.0002, "epoch": 2.8341708542713566, "step": 4230}, {"loss": 1.6002, "grad_norm": 0.46400442719459534, "learning_rate": 0.0002, "epoch": 2.8408710217755444, "step": 4240}, {"loss": 1.6052, "grad_norm": 0.49884888529777527, "learning_rate": 0.0002, "epoch": 2.8475711892797317, "step": 4250}, {"loss": 1.6919, "grad_norm": 0.5015072226524353, "learning_rate": 0.0002, "epoch": 2.8542713567839195, "step": 4260}, {"loss": 1.6335, "grad_norm": 0.4335440695285797, "learning_rate": 0.0002, "epoch": 2.8609715242881073, "step": 4270}, {"loss": 1.5664, "grad_norm": 0.5131644606590271, "learning_rate": 0.0002, "epoch": 2.8676716917922946, "step": 4280}, {"loss": 1.6409, "grad_norm": 0.6977195739746094, "learning_rate": 0.0002, "epoch": 2.8743718592964824, "step": 4290}, {"loss": 1.7192, "grad_norm": 0.5133762955665588, "learning_rate": 0.0002, "epoch": 2.8810720268006698, "step": 4300}, {"loss": 1.6257, "grad_norm": 0.4737614393234253, "learning_rate": 0.0002, "epoch": 2.8877721943048575, "step": 4310}, {"loss": 1.6076, "grad_norm": 0.4580535590648651, "learning_rate": 0.0002, "epoch": 2.8944723618090453, "step": 4320}, {"loss": 1.6538, "grad_norm": 0.43863341212272644, "learning_rate": 0.0002, "epoch": 2.901172529313233, "step": 4330}, {"loss": 1.6091, "grad_norm": 0.4103737473487854, "learning_rate": 0.0002, "epoch": 2.9078726968174204, "step": 4340}, {"loss": 1.7106, "grad_norm": 0.438014417886734, "learning_rate": 0.0002, "epoch": 2.914572864321608, "step": 4350}, {"loss": 1.6025, "grad_norm": 0.5068213939666748, "learning_rate": 0.0002, "epoch": 2.9212730318257956, "step": 4360}, {"loss": 1.6426, "grad_norm": 0.45305484533309937, "learning_rate": 0.0002, "epoch": 2.9279731993299833, "step": 4370}, {"loss": 1.5726, "grad_norm": 0.4612090289592743, "learning_rate": 0.0002, "epoch": 2.934673366834171, "step": 4380}, {"loss": 1.6536, "grad_norm": 0.508736789226532, "learning_rate": 0.0002, "epoch": 2.9413735343383585, "step": 4390}, {"loss": 1.6132, "grad_norm": 0.4924427270889282, "learning_rate": 0.0002, "epoch": 2.948073701842546, "step": 4400}, {"loss": 1.7007, "grad_norm": 0.5707460641860962, "learning_rate": 0.0002, "epoch": 2.9547738693467336, "step": 4410}, {"loss": 1.6814, "grad_norm": 0.42270299792289734, "learning_rate": 0.0002, "epoch": 2.9614740368509214, "step": 4420}, {"loss": 1.6644, "grad_norm": 0.4429931044578552, "learning_rate": 0.0002, "epoch": 2.968174204355109, "step": 4430}, {"loss": 1.6251, "grad_norm": 0.49760574102401733, "learning_rate": 0.0002, "epoch": 2.9748743718592965, "step": 4440}, {"loss": 1.6169, "grad_norm": 0.4558229148387909, "learning_rate": 0.0002, "epoch": 2.981574539363484, "step": 4450}, {"loss": 1.6055, "grad_norm": 0.39848530292510986, "learning_rate": 0.0002, "epoch": 2.9882747068676716, "step": 4460}, {"loss": 1.6705, "grad_norm": 0.5224862098693848, "learning_rate": 0.0002, "epoch": 2.9949748743718594, "step": 4470}, {"eval_loss": 1.8228833675384521, "eval_runtime": 37.9049, "eval_samples_per_second": 13.587, "eval_steps_per_second": 1.715, "epoch": 2.9996649916247904, "step": 4477}, {"loss": 1.6637, "grad_norm": 0.41169142723083496, "learning_rate": 0.0002, "epoch": 3.0016750418760467, "step": 4480}, {"loss": 1.5974, "grad_norm": 0.4865207374095917, "learning_rate": 0.0002, "epoch": 3.0083752093802345, "step": 4490}, {"loss": 1.5297, "grad_norm": 0.5462028384208679, "learning_rate": 0.0002, "epoch": 3.0150753768844223, "step": 4500}, {"loss": 1.5251, "grad_norm": 0.6169732809066772, "learning_rate": 0.0002, "epoch": 3.0217755443886096, "step": 4510}, {"loss": 1.5559, "grad_norm": 0.5667954087257385, "learning_rate": 0.0002, "epoch": 3.0284757118927974, "step": 4520}, {"loss": 1.5037, "grad_norm": 0.5758325457572937, "learning_rate": 0.0002, "epoch": 3.0351758793969847, "step": 4530}, {"loss": 1.4873, "grad_norm": 0.5220064520835876, "learning_rate": 0.0002, "epoch": 3.0418760469011725, "step": 4540}, {"loss": 1.5126, "grad_norm": 0.5469558835029602, "learning_rate": 0.0002, "epoch": 3.0485762144053603, "step": 4550}, {"loss": 1.4275, "grad_norm": 0.5680848956108093, "learning_rate": 0.0002, "epoch": 3.0552763819095476, "step": 4560}, {"loss": 1.5187, "grad_norm": 0.5906574726104736, "learning_rate": 0.0002, "epoch": 3.0619765494137354, "step": 4570}, {"loss": 1.4551, "grad_norm": 0.4725631773471832, "learning_rate": 0.0002, "epoch": 3.0686767169179228, "step": 4580}, {"loss": 1.5083, "grad_norm": 0.5273477435112, "learning_rate": 0.0002, "epoch": 3.0753768844221105, "step": 4590}, {"loss": 1.5154, "grad_norm": 0.5861203074455261, "learning_rate": 0.0002, "epoch": 3.0820770519262983, "step": 4600}, {"loss": 1.4924, "grad_norm": 0.5343965291976929, "learning_rate": 0.0002, "epoch": 3.0887772194304857, "step": 4610}, {"loss": 1.5608, "grad_norm": 0.5348150730133057, "learning_rate": 0.0002, "epoch": 3.0954773869346734, "step": 4620}, {"loss": 1.5399, "grad_norm": 0.5971846580505371, "learning_rate": 0.0002, "epoch": 3.102177554438861, "step": 4630}, {"loss": 1.4662, "grad_norm": 0.5203177332878113, "learning_rate": 0.0002, "epoch": 3.1088777219430486, "step": 4640}, {"loss": 1.5805, "grad_norm": 0.55289226770401, "learning_rate": 0.0002, "epoch": 3.1155778894472363, "step": 4650}, {"loss": 1.4745, "grad_norm": 0.6878530979156494, "learning_rate": 0.0002, "epoch": 3.1222780569514237, "step": 4660}, {"loss": 1.5335, "grad_norm": 0.6173256635665894, "learning_rate": 0.0002, "epoch": 3.1289782244556115, "step": 4670}, {"loss": 1.51, "grad_norm": 0.536796510219574, "learning_rate": 0.0002, "epoch": 3.135678391959799, "step": 4680}, {"loss": 1.4713, "grad_norm": 0.58846116065979, "learning_rate": 0.0002, "epoch": 3.1423785594639866, "step": 4690}, {"loss": 1.5114, "grad_norm": 0.645889401435852, "learning_rate": 0.0002, "epoch": 3.1490787269681744, "step": 4700}, {"loss": 1.4705, "grad_norm": 0.6118691563606262, "learning_rate": 0.0002, "epoch": 3.1557788944723617, "step": 4710}, {"loss": 1.5533, "grad_norm": 0.5189669132232666, "learning_rate": 0.0002, "epoch": 3.1624790619765495, "step": 4720}, {"loss": 1.4769, "grad_norm": 0.5794713497161865, "learning_rate": 0.0002, "epoch": 3.169179229480737, "step": 4730}, {"loss": 1.4849, "grad_norm": 0.6579326391220093, "learning_rate": 0.0002, "epoch": 3.1758793969849246, "step": 4740}, {"loss": 1.545, "grad_norm": 0.5822742581367493, "learning_rate": 0.0002, "epoch": 3.1825795644891124, "step": 4750}, {"loss": 1.4358, "grad_norm": 0.5475956201553345, "learning_rate": 0.0002, "epoch": 3.1892797319932997, "step": 4760}, {"loss": 1.4723, "grad_norm": 0.6743834018707275, "learning_rate": 0.0002, "epoch": 3.1959798994974875, "step": 4770}, {"loss": 1.5161, "grad_norm": 0.6110585927963257, "learning_rate": 0.0002, "epoch": 3.202680067001675, "step": 4780}, {"loss": 1.5455, "grad_norm": 0.5426181554794312, "learning_rate": 0.0002, "epoch": 3.2093802345058626, "step": 4790}, {"loss": 1.5315, "grad_norm": 0.6077824234962463, "learning_rate": 0.0002, "epoch": 3.2160804020100504, "step": 4800}, {"loss": 1.5314, "grad_norm": 0.5785858631134033, "learning_rate": 0.0002, "epoch": 3.2227805695142377, "step": 4810}, {"loss": 1.4041, "grad_norm": 0.6425958275794983, "learning_rate": 0.0002, "epoch": 3.2294807370184255, "step": 4820}, {"loss": 1.4751, "grad_norm": 0.6607080698013306, "learning_rate": 0.0002, "epoch": 3.236180904522613, "step": 4830}, {"loss": 1.5267, "grad_norm": 0.5385788679122925, "learning_rate": 0.0002, "epoch": 3.2428810720268006, "step": 4840}, {"loss": 1.4673, "grad_norm": 0.5630403757095337, "learning_rate": 0.0002, "epoch": 3.2495812395309884, "step": 4850}, {"loss": 1.5257, "grad_norm": 0.6340779662132263, "learning_rate": 0.0002, "epoch": 3.2562814070351758, "step": 4860}, {"loss": 1.5148, "grad_norm": 0.5305342674255371, "learning_rate": 0.0002, "epoch": 3.2629815745393635, "step": 4870}, {"loss": 1.5162, "grad_norm": 0.597670316696167, "learning_rate": 0.0002, "epoch": 3.2696817420435513, "step": 4880}, {"loss": 1.5429, "grad_norm": 0.665553867816925, "learning_rate": 0.0002, "epoch": 3.2763819095477387, "step": 4890}, {"loss": 1.4607, "grad_norm": 0.579767644405365, "learning_rate": 0.0002, "epoch": 3.2830820770519265, "step": 4900}, {"loss": 1.4999, "grad_norm": 0.5512481331825256, "learning_rate": 0.0002, "epoch": 3.289782244556114, "step": 4910}, {"loss": 1.5022, "grad_norm": 0.5916532278060913, "learning_rate": 0.0002, "epoch": 3.2964824120603016, "step": 4920}, {"loss": 1.4889, "grad_norm": 0.7521726489067078, "learning_rate": 0.0002, "epoch": 3.3031825795644894, "step": 4930}, {"loss": 1.4223, "grad_norm": 0.5352797508239746, "learning_rate": 0.0002, "epoch": 3.3098827470686767, "step": 4940}, {"loss": 1.5122, "grad_norm": 0.5950371623039246, "learning_rate": 0.0002, "epoch": 3.3165829145728645, "step": 4950}, {"loss": 1.5072, "grad_norm": 0.8020477890968323, "learning_rate": 0.0002, "epoch": 3.323283082077052, "step": 4960}, {"loss": 1.5422, "grad_norm": 0.6790024638175964, "learning_rate": 0.0002, "epoch": 3.3299832495812396, "step": 4970}, {"loss": 1.5363, "grad_norm": 0.687627375125885, "learning_rate": 0.0002, "epoch": 3.3366834170854274, "step": 4980}, {"loss": 1.5276, "grad_norm": 0.6094385385513306, "learning_rate": 0.0002, "epoch": 3.3433835845896147, "step": 4990}, {"loss": 1.549, "grad_norm": 0.6541242003440857, "learning_rate": 0.0002, "epoch": 3.3500837520938025, "step": 5000}, {"loss": 1.6067, "grad_norm": 0.5560880303382874, "learning_rate": 0.0002, "epoch": 3.35678391959799, "step": 5010}, {"loss": 1.5769, "grad_norm": 0.5440094470977783, "learning_rate": 0.0002, "epoch": 3.3634840871021776, "step": 5020}, {"loss": 1.6183, "grad_norm": 0.5749301314353943, "learning_rate": 0.0002, "epoch": 3.3701842546063654, "step": 5030}, {"loss": 1.4801, "grad_norm": 0.5919716954231262, "learning_rate": 0.0002, "epoch": 3.3768844221105527, "step": 5040}, {"loss": 1.5957, "grad_norm": 0.6331481337547302, "learning_rate": 0.0002, "epoch": 3.3835845896147405, "step": 5050}, {"loss": 1.5188, "grad_norm": 0.5687161684036255, "learning_rate": 0.0002, "epoch": 3.390284757118928, "step": 5060}, {"loss": 1.5702, "grad_norm": 0.6718577742576599, "learning_rate": 0.0002, "epoch": 3.3969849246231156, "step": 5070}, {"loss": 1.5577, "grad_norm": 0.5089324116706848, "learning_rate": 0.0002, "epoch": 3.4036850921273034, "step": 5080}, {"loss": 1.512, "grad_norm": 0.5710174441337585, "learning_rate": 0.0002, "epoch": 3.4103852596314908, "step": 5090}, {"loss": 1.5492, "grad_norm": 0.6670721173286438, "learning_rate": 0.0002, "epoch": 3.4170854271356785, "step": 5100}, {"loss": 1.5227, "grad_norm": 0.6875665187835693, "learning_rate": 0.0002, "epoch": 3.423785594639866, "step": 5110}, {"loss": 1.4496, "grad_norm": 0.5375880599021912, "learning_rate": 0.0002, "epoch": 3.4304857621440537, "step": 5120}, {"loss": 1.5527, "grad_norm": 0.6550399661064148, "learning_rate": 0.0002, "epoch": 3.4371859296482414, "step": 5130}, {"loss": 1.5687, "grad_norm": 0.5948067903518677, "learning_rate": 0.0002, "epoch": 3.4438860971524288, "step": 5140}, {"loss": 1.4813, "grad_norm": 0.6134477257728577, "learning_rate": 0.0002, "epoch": 3.4505862646566166, "step": 5150}, {"loss": 1.5069, "grad_norm": 0.6506398320198059, "learning_rate": 0.0002, "epoch": 3.457286432160804, "step": 5160}, {"loss": 1.4422, "grad_norm": 0.6060147881507874, "learning_rate": 0.0002, "epoch": 3.4639865996649917, "step": 5170}, {"loss": 1.5093, "grad_norm": 0.6173806190490723, "learning_rate": 0.0002, "epoch": 3.4706867671691795, "step": 5180}, {"loss": 1.4975, "grad_norm": 0.6032607555389404, "learning_rate": 0.0002, "epoch": 3.477386934673367, "step": 5190}, {"loss": 1.4979, "grad_norm": 0.5652492046356201, "learning_rate": 0.0002, "epoch": 3.4840871021775546, "step": 5200}, {"loss": 1.4883, "grad_norm": 0.6168607473373413, "learning_rate": 0.0002, "epoch": 3.490787269681742, "step": 5210}, {"loss": 1.5164, "grad_norm": 0.6170629262924194, "learning_rate": 0.0002, "epoch": 3.4974874371859297, "step": 5220}, {"loss": 1.4879, "grad_norm": 0.6926297545433044, "learning_rate": 0.0002, "epoch": 3.5041876046901175, "step": 5230}, {"loss": 1.4982, "grad_norm": 0.6702437996864319, "learning_rate": 0.0002, "epoch": 3.510887772194305, "step": 5240}, {"loss": 1.4986, "grad_norm": 0.5421436429023743, "learning_rate": 0.0002, "epoch": 3.5175879396984926, "step": 5250}, {"loss": 1.4673, "grad_norm": 0.5726765990257263, "learning_rate": 0.0002, "epoch": 3.52428810720268, "step": 5260}, {"loss": 1.5423, "grad_norm": 0.5685455203056335, "learning_rate": 0.0002, "epoch": 3.5309882747068677, "step": 5270}, {"loss": 1.4715, "grad_norm": 0.6018396019935608, "learning_rate": 0.0002, "epoch": 3.5376884422110555, "step": 5280}, {"loss": 1.5451, "grad_norm": 0.5731932520866394, "learning_rate": 0.0002, "epoch": 3.544388609715243, "step": 5290}, {"loss": 1.4752, "grad_norm": 0.6601519584655762, "learning_rate": 0.0002, "epoch": 3.5510887772194306, "step": 5300}, {"loss": 1.5434, "grad_norm": 0.5545530319213867, "learning_rate": 0.0002, "epoch": 3.557788944723618, "step": 5310}, {"loss": 1.5438, "grad_norm": 0.5998541116714478, "learning_rate": 0.0002, "epoch": 3.5644891122278057, "step": 5320}, {"loss": 1.56, "grad_norm": 0.5651767253875732, "learning_rate": 0.0002, "epoch": 3.5711892797319935, "step": 5330}, {"loss": 1.4829, "grad_norm": 0.7425084114074707, "learning_rate": 0.0002, "epoch": 3.577889447236181, "step": 5340}, {"loss": 1.5571, "grad_norm": 0.5770602226257324, "learning_rate": 0.0002, "epoch": 3.5845896147403686, "step": 5350}, {"loss": 1.458, "grad_norm": 0.54723060131073, "learning_rate": 0.0002, "epoch": 3.591289782244556, "step": 5360}, {"loss": 1.497, "grad_norm": 0.6658238172531128, "learning_rate": 0.0002, "epoch": 3.5979899497487438, "step": 5370}, {"loss": 1.5456, "grad_norm": 0.5787645578384399, "learning_rate": 0.0002, "epoch": 3.6046901172529315, "step": 5380}, {"loss": 1.5343, "grad_norm": 0.594913125038147, "learning_rate": 0.0002, "epoch": 3.611390284757119, "step": 5390}, {"loss": 1.4727, "grad_norm": 0.4964977502822876, "learning_rate": 0.0002, "epoch": 3.6180904522613067, "step": 5400}, {"loss": 1.5062, "grad_norm": 0.6087527275085449, "learning_rate": 0.0002, "epoch": 3.624790619765494, "step": 5410}, {"loss": 1.5098, "grad_norm": 0.6315323710441589, "learning_rate": 0.0002, "epoch": 3.6314907872696818, "step": 5420}, {"loss": 1.4855, "grad_norm": 0.574799120426178, "learning_rate": 0.0002, "epoch": 3.6381909547738696, "step": 5430}, {"loss": 1.4595, "grad_norm": 0.5949277877807617, "learning_rate": 0.0002, "epoch": 3.644891122278057, "step": 5440}, {"loss": 1.4816, "grad_norm": 0.5640677213668823, "learning_rate": 0.0002, "epoch": 3.6515912897822447, "step": 5450}, {"loss": 1.525, "grad_norm": 0.6198237538337708, "learning_rate": 0.0002, "epoch": 3.658291457286432, "step": 5460}, {"loss": 1.5676, "grad_norm": 0.6902034878730774, "learning_rate": 0.0002, "epoch": 3.66499162479062, "step": 5470}, {"loss": 1.544, "grad_norm": 0.5686674118041992, "learning_rate": 0.0002, "epoch": 3.6716917922948076, "step": 5480}, {"loss": 1.5255, "grad_norm": 0.6532107591629028, "learning_rate": 0.0002, "epoch": 3.678391959798995, "step": 5490}, {"loss": 1.5767, "grad_norm": 0.5790849924087524, "learning_rate": 0.0002, "epoch": 3.6850921273031827, "step": 5500}, {"loss": 1.5507, "grad_norm": 0.6055065393447876, "learning_rate": 0.0002, "epoch": 3.69179229480737, "step": 5510}, {"loss": 1.4656, "grad_norm": 0.5630605816841125, "learning_rate": 0.0002, "epoch": 3.698492462311558, "step": 5520}, {"loss": 1.537, "grad_norm": 0.6005825996398926, "learning_rate": 0.0002, "epoch": 3.7051926298157456, "step": 5530}, {"loss": 1.5313, "grad_norm": 0.6553038954734802, "learning_rate": 0.0002, "epoch": 3.711892797319933, "step": 5540}, {"loss": 1.4943, "grad_norm": 0.5601094961166382, "learning_rate": 0.0002, "epoch": 3.7185929648241207, "step": 5550}, {"loss": 1.5161, "grad_norm": 0.6598808169364929, "learning_rate": 0.0002, "epoch": 3.725293132328308, "step": 5560}, {"loss": 1.5345, "grad_norm": 0.5506255626678467, "learning_rate": 0.0002, "epoch": 3.731993299832496, "step": 5570}, {"loss": 1.4805, "grad_norm": 0.6001223921775818, "learning_rate": 0.0002, "epoch": 3.7386934673366836, "step": 5580}, {"loss": 1.4652, "grad_norm": 0.6287297606468201, "learning_rate": 0.0002, "epoch": 3.745393634840871, "step": 5590}, {"loss": 1.5246, "grad_norm": 0.6253238916397095, "learning_rate": 0.0002, "epoch": 3.7520938023450587, "step": 5600}, {"loss": 1.5691, "grad_norm": 0.5713174939155579, "learning_rate": 0.0002, "epoch": 3.758793969849246, "step": 5610}, {"loss": 1.5661, "grad_norm": 0.6198310852050781, "learning_rate": 0.0002, "epoch": 3.765494137353434, "step": 5620}, {"loss": 1.5448, "grad_norm": 0.5941224098205566, "learning_rate": 0.0002, "epoch": 3.7721943048576216, "step": 5630}, {"loss": 1.4925, "grad_norm": 0.606002151966095, "learning_rate": 0.0002, "epoch": 3.778894472361809, "step": 5640}, {"loss": 1.5182, "grad_norm": 0.6540704965591431, "learning_rate": 0.0002, "epoch": 3.7855946398659968, "step": 5650}, {"loss": 1.5903, "grad_norm": 0.6147415041923523, "learning_rate": 0.0002, "epoch": 3.792294807370184, "step": 5660}, {"loss": 1.5329, "grad_norm": 0.5649605393409729, "learning_rate": 0.0002, "epoch": 3.798994974874372, "step": 5670}, {"loss": 1.5747, "grad_norm": 0.6788773536682129, "learning_rate": 0.0002, "epoch": 3.8056951423785597, "step": 5680}, {"loss": 1.535, "grad_norm": 0.6581860780715942, "learning_rate": 0.0002, "epoch": 3.812395309882747, "step": 5690}, {"loss": 1.4587, "grad_norm": 0.5529348850250244, "learning_rate": 0.0002, "epoch": 3.819095477386935, "step": 5700}, {"loss": 1.5594, "grad_norm": 0.6320232152938843, "learning_rate": 0.0002, "epoch": 3.825795644891122, "step": 5710}, {"loss": 1.4696, "grad_norm": 0.6529698371887207, "learning_rate": 0.0002, "epoch": 3.83249581239531, "step": 5720}, {"loss": 1.5854, "grad_norm": 0.5983362793922424, "learning_rate": 0.0002, "epoch": 3.8391959798994977, "step": 5730}, {"loss": 1.465, "grad_norm": 0.6335684061050415, "learning_rate": 0.0002, "epoch": 3.845896147403685, "step": 5740}, {"loss": 1.5545, "grad_norm": 0.700446605682373, "learning_rate": 0.0002, "epoch": 3.852596314907873, "step": 5750}, {"loss": 1.5707, "grad_norm": 0.6092597842216492, "learning_rate": 0.0002, "epoch": 3.85929648241206, "step": 5760}, {"loss": 1.5729, "grad_norm": 0.564146101474762, "learning_rate": 0.0002, "epoch": 3.865996649916248, "step": 5770}, {"loss": 1.5872, "grad_norm": 0.615275502204895, "learning_rate": 0.0002, "epoch": 3.8726968174204357, "step": 5780}, {"loss": 1.5142, "grad_norm": 0.6685376763343811, "learning_rate": 0.0002, "epoch": 3.879396984924623, "step": 5790}, {"loss": 1.4752, "grad_norm": 0.6116922497749329, "learning_rate": 0.0002, "epoch": 3.886097152428811, "step": 5800}, {"loss": 1.5179, "grad_norm": 0.5486813187599182, "learning_rate": 0.0002, "epoch": 3.892797319932998, "step": 5810}, {"loss": 1.5167, "grad_norm": 0.6208204030990601, "learning_rate": 0.0002, "epoch": 3.899497487437186, "step": 5820}, {"loss": 1.5334, "grad_norm": 0.6500625014305115, "learning_rate": 0.0002, "epoch": 3.9061976549413737, "step": 5830}, {"loss": 1.4716, "grad_norm": 0.5948089361190796, "learning_rate": 0.0002, "epoch": 3.912897822445561, "step": 5840}, {"loss": 1.6011, "grad_norm": 0.7210732698440552, "learning_rate": 0.0002, "epoch": 3.919597989949749, "step": 5850}, {"loss": 1.5519, "grad_norm": 0.6662322878837585, "learning_rate": 0.0002, "epoch": 3.926298157453936, "step": 5860}, {"loss": 1.5656, "grad_norm": 0.5613839626312256, "learning_rate": 0.0002, "epoch": 3.932998324958124, "step": 5870}, {"loss": 1.544, "grad_norm": 0.6069002151489258, "learning_rate": 0.0002, "epoch": 3.9396984924623117, "step": 5880}, {"loss": 1.6745, "grad_norm": 0.7075562477111816, "learning_rate": 0.0002, "epoch": 3.946398659966499, "step": 5890}, {"loss": 1.5391, "grad_norm": 0.6316173076629639, "learning_rate": 0.0002, "epoch": 3.953098827470687, "step": 5900}, {"loss": 1.6314, "grad_norm": 0.5716308355331421, "learning_rate": 0.0002, "epoch": 3.959798994974874, "step": 5910}, {"loss": 1.5947, "grad_norm": 0.6800096035003662, "learning_rate": 0.0002, "epoch": 3.966499162479062, "step": 5920}, {"loss": 1.5189, "grad_norm": 0.6057983040809631, "learning_rate": 0.0002, "epoch": 3.9731993299832498, "step": 5930}, {"loss": 1.5431, "grad_norm": 0.5938987731933594, "learning_rate": 0.0002, "epoch": 3.979899497487437, "step": 5940}, {"loss": 1.5111, "grad_norm": 0.6963576674461365, "learning_rate": 0.0002, "epoch": 3.986599664991625, "step": 5950}, {"loss": 1.5521, "grad_norm": 0.6279940009117126, "learning_rate": 0.0002, "epoch": 3.993299832495812, "step": 5960}, {"loss": 1.5974, "grad_norm": 0.7161159515380859, "learning_rate": 0.0002, "epoch": 4.0, "step": 5970}, {"eval_loss": 1.8655421733856201, "eval_runtime": 37.9276, "eval_samples_per_second": 13.579, "eval_steps_per_second": 1.714, "epoch": 4.0, "step": 5970}, {"loss": 1.3666, "grad_norm": 0.7380476593971252, "learning_rate": 0.0002, "epoch": 4.006700167504188, "step": 5980}, {"loss": 1.3913, "grad_norm": 0.7148947715759277, "learning_rate": 0.0002, "epoch": 4.013400335008376, "step": 5990}, {"loss": 1.4204, "grad_norm": 0.6177082657814026, "learning_rate": 0.0002, "epoch": 4.0201005025125625, "step": 6000}, {"loss": 1.4421, "grad_norm": 0.8552946448326111, "learning_rate": 0.0002, "epoch": 4.02680067001675, "step": 6010}, {"loss": 1.4342, "grad_norm": 0.8033416271209717, "learning_rate": 0.0002, "epoch": 4.033500837520938, "step": 6020}, {"loss": 1.4092, "grad_norm": 0.8501318097114563, "learning_rate": 0.0002, "epoch": 4.040201005025126, "step": 6030}, {"loss": 1.3367, "grad_norm": 0.6981393098831177, "learning_rate": 0.0002, "epoch": 4.046901172529314, "step": 6040}, {"loss": 1.3925, "grad_norm": 0.7227180600166321, "learning_rate": 0.0002, "epoch": 4.0536013400335005, "step": 6050}, {"loss": 1.4007, "grad_norm": 0.6923989653587341, "learning_rate": 0.0002, "epoch": 4.060301507537688, "step": 6060}, {"loss": 1.3837, "grad_norm": 0.879779040813446, "learning_rate": 0.0002, "epoch": 4.067001675041876, "step": 6070}, {"loss": 1.4383, "grad_norm": 0.8184754848480225, "learning_rate": 0.0002, "epoch": 4.073701842546064, "step": 6080}, {"loss": 1.3128, "grad_norm": 0.8211342692375183, "learning_rate": 0.0002, "epoch": 4.080402010050252, "step": 6090}, {"loss": 1.3892, "grad_norm": 0.7542396783828735, "learning_rate": 0.0002, "epoch": 4.0871021775544385, "step": 6100}, {"loss": 1.3607, "grad_norm": 0.6631066799163818, "learning_rate": 0.0002, "epoch": 4.093802345058626, "step": 6110}, {"loss": 1.3275, "grad_norm": 0.6728386282920837, "learning_rate": 0.0002, "epoch": 4.100502512562814, "step": 6120}, {"loss": 1.3443, "grad_norm": 0.681851863861084, "learning_rate": 0.0002, "epoch": 4.107202680067002, "step": 6130}, {"loss": 1.3486, "grad_norm": 0.8757794499397278, "learning_rate": 0.0002, "epoch": 4.11390284757119, "step": 6140}, {"loss": 1.351, "grad_norm": 0.6567301750183105, "learning_rate": 0.0002, "epoch": 4.1206030150753765, "step": 6150}, {"loss": 1.3824, "grad_norm": 0.7950329184532166, "learning_rate": 0.0002, "epoch": 4.127303182579564, "step": 6160}, {"loss": 1.3738, "grad_norm": 0.7545644044876099, "learning_rate": 0.0002, "epoch": 4.134003350083752, "step": 6170}, {"loss": 1.4214, "grad_norm": 0.7172710299491882, "learning_rate": 0.0002, "epoch": 4.14070351758794, "step": 6180}, {"loss": 1.4091, "grad_norm": 0.7040584087371826, "learning_rate": 0.0002, "epoch": 4.147403685092128, "step": 6190}, {"loss": 1.4149, "grad_norm": 0.7482913732528687, "learning_rate": 0.0002, "epoch": 4.1541038525963145, "step": 6200}, {"loss": 1.3227, "grad_norm": 0.8523276448249817, "learning_rate": 0.0002, "epoch": 4.160804020100502, "step": 6210}, {"loss": 1.4194, "grad_norm": 0.6672041416168213, "learning_rate": 0.0002, "epoch": 4.16750418760469, "step": 6220}, {"loss": 1.3953, "grad_norm": 0.7523500919342041, "learning_rate": 0.0002, "epoch": 4.174204355108878, "step": 6230}, {"loss": 1.371, "grad_norm": 0.8085253834724426, "learning_rate": 0.0002, "epoch": 4.180904522613066, "step": 6240}, {"loss": 1.3293, "grad_norm": 0.789450466632843, "learning_rate": 0.0002, "epoch": 4.187604690117253, "step": 6250}, {"loss": 1.3539, "grad_norm": 0.7502310872077942, "learning_rate": 0.0002, "epoch": 4.19430485762144, "step": 6260}, {"loss": 1.3415, "grad_norm": 0.7397456765174866, "learning_rate": 0.0002, "epoch": 4.201005025125628, "step": 6270}, {"loss": 1.3963, "grad_norm": 0.6921947002410889, "learning_rate": 0.0002, "epoch": 4.207705192629816, "step": 6280}, {"loss": 1.3125, "grad_norm": 0.9334571957588196, "learning_rate": 0.0002, "epoch": 4.214405360134004, "step": 6290}, {"loss": 1.3612, "grad_norm": 0.725799024105072, "learning_rate": 0.0002, "epoch": 4.221105527638191, "step": 6300}, {"loss": 1.4217, "grad_norm": 0.8290495872497559, "learning_rate": 0.0002, "epoch": 4.227805695142378, "step": 6310}, {"loss": 1.4135, "grad_norm": 0.688983678817749, "learning_rate": 0.0002, "epoch": 4.234505862646566, "step": 6320}, {"loss": 1.3807, "grad_norm": 0.8620913028717041, "learning_rate": 0.0002, "epoch": 4.241206030150754, "step": 6330}, {"loss": 1.3738, "grad_norm": 0.8008657693862915, "learning_rate": 0.0002, "epoch": 4.247906197654942, "step": 6340}, {"loss": 1.4005, "grad_norm": 0.7379199266433716, "learning_rate": 0.0002, "epoch": 4.254606365159129, "step": 6350}, {"loss": 1.426, "grad_norm": 0.7842815518379211, "learning_rate": 0.0002, "epoch": 4.261306532663316, "step": 6360}, {"loss": 1.4262, "grad_norm": 0.812600314617157, "learning_rate": 0.0002, "epoch": 4.268006700167504, "step": 6370}, {"loss": 1.4028, "grad_norm": 0.7852841019630432, "learning_rate": 0.0002, "epoch": 4.274706867671692, "step": 6380}, {"loss": 1.3722, "grad_norm": 1.0377534627914429, "learning_rate": 0.0002, "epoch": 4.28140703517588, "step": 6390}, {"loss": 1.3755, "grad_norm": 1.03935706615448, "learning_rate": 0.0002, "epoch": 4.288107202680067, "step": 6400}, {"loss": 1.3961, "grad_norm": 0.7244732975959778, "learning_rate": 0.0002, "epoch": 4.294807370184254, "step": 6410}, {"loss": 1.4608, "grad_norm": 0.7137406468391418, "learning_rate": 0.0002, "epoch": 4.301507537688442, "step": 6420}, {"loss": 1.4461, "grad_norm": 0.7492543458938599, "learning_rate": 0.0002, "epoch": 4.30820770519263, "step": 6430}, {"loss": 1.4562, "grad_norm": 0.7065439224243164, "learning_rate": 0.0002, "epoch": 4.314907872696818, "step": 6440}, {"loss": 1.4246, "grad_norm": 0.7786989808082581, "learning_rate": 0.0002, "epoch": 4.321608040201005, "step": 6450}, {"loss": 1.3098, "grad_norm": 0.7369208335876465, "learning_rate": 0.0002, "epoch": 4.328308207705192, "step": 6460}, {"loss": 1.3686, "grad_norm": 0.7412346005439758, "learning_rate": 0.0002, "epoch": 4.33500837520938, "step": 6470}, {"loss": 1.4087, "grad_norm": 0.780927300453186, "learning_rate": 0.0002, "epoch": 4.341708542713568, "step": 6480}, {"loss": 1.3628, "grad_norm": 0.8320930600166321, "learning_rate": 0.0002, "epoch": 4.348408710217756, "step": 6490}, {"loss": 1.3715, "grad_norm": 0.6871094703674316, "learning_rate": 0.0002, "epoch": 4.355108877721943, "step": 6500}, {"loss": 1.3257, "grad_norm": 0.6751559972763062, "learning_rate": 0.0002, "epoch": 4.36180904522613, "step": 6510}, {"loss": 1.4311, "grad_norm": 0.7723976969718933, "learning_rate": 0.0002, "epoch": 4.368509212730318, "step": 6520}, {"loss": 1.4086, "grad_norm": 0.7915401458740234, "learning_rate": 0.0002, "epoch": 4.375209380234506, "step": 6530}, {"loss": 1.3973, "grad_norm": 0.7329102754592896, "learning_rate": 0.0002, "epoch": 4.381909547738694, "step": 6540}, {"loss": 1.447, "grad_norm": 0.7388760447502136, "learning_rate": 0.0002, "epoch": 4.388609715242881, "step": 6550}, {"loss": 1.4378, "grad_norm": 0.8282579183578491, "learning_rate": 0.0002, "epoch": 4.3953098827470685, "step": 6560}, {"loss": 1.3923, "grad_norm": 0.7192724347114563, "learning_rate": 0.0002, "epoch": 4.402010050251256, "step": 6570}, {"loss": 1.4141, "grad_norm": 0.746526837348938, "learning_rate": 0.0002, "epoch": 4.408710217755444, "step": 6580}, {"loss": 1.33, "grad_norm": 0.8738046288490295, "learning_rate": 0.0002, "epoch": 4.415410385259632, "step": 6590}, {"loss": 1.3995, "grad_norm": 0.8408458828926086, "learning_rate": 0.0002, "epoch": 4.422110552763819, "step": 6600}, {"loss": 1.4148, "grad_norm": 0.8110666275024414, "learning_rate": 0.0002, "epoch": 4.4288107202680065, "step": 6610}, {"loss": 1.441, "grad_norm": 0.8602406978607178, "learning_rate": 0.0002, "epoch": 4.435510887772194, "step": 6620}, {"loss": 1.4319, "grad_norm": 0.7549102902412415, "learning_rate": 0.0002, "epoch": 4.442211055276382, "step": 6630}, {"loss": 1.388, "grad_norm": 0.7831804156303406, "learning_rate": 0.0002, "epoch": 4.44891122278057, "step": 6640}, {"loss": 1.4283, "grad_norm": 0.7269673943519592, "learning_rate": 0.0002, "epoch": 4.455611390284757, "step": 6650}, {"loss": 1.4132, "grad_norm": 0.7397838830947876, "learning_rate": 0.0002, "epoch": 4.4623115577889445, "step": 6660}, {"loss": 1.3174, "grad_norm": 0.713707447052002, "learning_rate": 0.0002, "epoch": 4.469011725293132, "step": 6670}, {"loss": 1.3406, "grad_norm": 0.7525581121444702, "learning_rate": 0.0002, "epoch": 4.47571189279732, "step": 6680}, {"loss": 1.4283, "grad_norm": 0.8030191659927368, "learning_rate": 0.0002, "epoch": 4.482412060301508, "step": 6690}, {"loss": 1.4586, "grad_norm": 0.7469439506530762, "learning_rate": 0.0002, "epoch": 4.489112227805695, "step": 6700}, {"loss": 1.367, "grad_norm": 0.7743868231773376, "learning_rate": 0.0002, "epoch": 4.4958123953098825, "step": 6710}, {"loss": 1.3439, "grad_norm": 0.6539737582206726, "learning_rate": 0.0002, "epoch": 4.50251256281407, "step": 6720}, {"loss": 1.4513, "grad_norm": 0.825818657875061, "learning_rate": 0.0002, "epoch": 4.509212730318258, "step": 6730}, {"loss": 1.3984, "grad_norm": 0.8048575520515442, "learning_rate": 0.0002, "epoch": 4.515912897822446, "step": 6740}, {"loss": 1.3923, "grad_norm": 0.7828766107559204, "learning_rate": 0.0002, "epoch": 4.522613065326633, "step": 6750}, {"loss": 1.3886, "grad_norm": 0.7406010031700134, "learning_rate": 0.0002, "epoch": 4.5293132328308205, "step": 6760}, {"loss": 1.3109, "grad_norm": 0.840345561504364, "learning_rate": 0.0002, "epoch": 4.536013400335008, "step": 6770}, {"loss": 1.4808, "grad_norm": 0.8492622971534729, "learning_rate": 0.0002, "epoch": 4.542713567839196, "step": 6780}, {"loss": 1.4384, "grad_norm": 0.7130163908004761, "learning_rate": 0.0002, "epoch": 4.549413735343384, "step": 6790}, {"loss": 1.4531, "grad_norm": 0.8454728126525879, "learning_rate": 0.0002, "epoch": 4.556113902847571, "step": 6800}, {"loss": 1.3239, "grad_norm": 0.7847645282745361, "learning_rate": 0.0002, "epoch": 4.562814070351759, "step": 6810}, {"loss": 1.4181, "grad_norm": 0.7245864272117615, "learning_rate": 0.0002, "epoch": 4.569514237855946, "step": 6820}, {"loss": 1.3233, "grad_norm": 0.768893301486969, "learning_rate": 0.0002, "epoch": 4.576214405360134, "step": 6830}, {"loss": 1.3932, "grad_norm": 0.8028400540351868, "learning_rate": 0.0002, "epoch": 4.582914572864322, "step": 6840}, {"loss": 1.3745, "grad_norm": 0.763945460319519, "learning_rate": 0.0002, "epoch": 4.589614740368509, "step": 6850}, {"loss": 1.4797, "grad_norm": 0.7417685389518738, "learning_rate": 0.0002, "epoch": 4.596314907872697, "step": 6860}, {"loss": 1.4468, "grad_norm": 0.7603038549423218, "learning_rate": 0.0002, "epoch": 4.603015075376884, "step": 6870}, {"loss": 1.4095, "grad_norm": 0.7981528043746948, "learning_rate": 0.0002, "epoch": 4.609715242881072, "step": 6880}, {"loss": 1.3963, "grad_norm": 0.8077111840248108, "learning_rate": 0.0002, "epoch": 4.61641541038526, "step": 6890}, {"loss": 1.4721, "grad_norm": 0.8778454065322876, "learning_rate": 0.0002, "epoch": 4.623115577889447, "step": 6900}, {"loss": 1.3758, "grad_norm": 0.8620710372924805, "learning_rate": 0.0002, "epoch": 4.629815745393635, "step": 6910}, {"loss": 1.344, "grad_norm": 0.7486072778701782, "learning_rate": 0.0002, "epoch": 4.636515912897822, "step": 6920}, {"loss": 1.3913, "grad_norm": 0.7493042945861816, "learning_rate": 0.0002, "epoch": 4.64321608040201, "step": 6930}, {"loss": 1.397, "grad_norm": 0.7388978600502014, "learning_rate": 0.0002, "epoch": 4.649916247906198, "step": 6940}, {"loss": 1.3593, "grad_norm": 0.798530638217926, "learning_rate": 0.0002, "epoch": 4.656616415410385, "step": 6950}, {"loss": 1.3982, "grad_norm": 0.7929500937461853, "learning_rate": 0.0002, "epoch": 4.663316582914573, "step": 6960}, {"loss": 1.4183, "grad_norm": 0.9186785221099854, "learning_rate": 0.0002, "epoch": 4.67001675041876, "step": 6970}, {"loss": 1.3955, "grad_norm": 1.1103485822677612, "learning_rate": 0.0002, "epoch": 4.676716917922948, "step": 6980}, {"loss": 1.3941, "grad_norm": 0.8000466823577881, "learning_rate": 0.0002, "epoch": 4.683417085427136, "step": 6990}, {"loss": 1.371, "grad_norm": 0.7520599961280823, "learning_rate": 0.0002, "epoch": 4.690117252931323, "step": 7000}, {"loss": 1.4582, "grad_norm": 0.7971973419189453, "learning_rate": 0.0002, "epoch": 4.696817420435511, "step": 7010}, {"loss": 1.3682, "grad_norm": 0.7363343834877014, "learning_rate": 0.0002, "epoch": 4.703517587939698, "step": 7020}, {"loss": 1.3889, "grad_norm": 0.8268865942955017, "learning_rate": 0.0002, "epoch": 4.710217755443886, "step": 7030}, {"loss": 1.4382, "grad_norm": 0.7054963111877441, "learning_rate": 0.0002, "epoch": 4.716917922948074, "step": 7040}, {"loss": 1.4578, "grad_norm": 0.8196262121200562, "learning_rate": 0.0002, "epoch": 4.723618090452261, "step": 7050}, {"loss": 1.365, "grad_norm": 0.8276031017303467, "learning_rate": 0.0002, "epoch": 4.730318257956449, "step": 7060}, {"loss": 1.3887, "grad_norm": 0.8248157501220703, "learning_rate": 0.0002, "epoch": 4.7370184254606365, "step": 7070}, {"loss": 1.4193, "grad_norm": 0.8937979936599731, "learning_rate": 0.0002, "epoch": 4.743718592964824, "step": 7080}, {"loss": 1.4334, "grad_norm": 0.7984238862991333, "learning_rate": 0.0002, "epoch": 4.750418760469012, "step": 7090}, {"loss": 1.4385, "grad_norm": 0.9495313763618469, "learning_rate": 0.0002, "epoch": 4.757118927973199, "step": 7100}, {"loss": 1.4504, "grad_norm": 0.8598204255104065, "learning_rate": 0.0002, "epoch": 4.763819095477387, "step": 7110}, {"loss": 1.3969, "grad_norm": 0.8951472640037537, "learning_rate": 0.0002, "epoch": 4.7705192629815745, "step": 7120}, {"loss": 1.4339, "grad_norm": 0.9110309481620789, "learning_rate": 0.0002, "epoch": 4.777219430485762, "step": 7130}, {"loss": 1.4001, "grad_norm": 0.7929584980010986, "learning_rate": 0.0002, "epoch": 4.78391959798995, "step": 7140}, {"loss": 1.467, "grad_norm": 0.7415322661399841, "learning_rate": 0.0002, "epoch": 4.790619765494137, "step": 7150}, {"loss": 1.5107, "grad_norm": 0.7504757046699524, "learning_rate": 0.0002, "epoch": 4.797319932998325, "step": 7160}, {"loss": 1.3736, "grad_norm": 0.7166924476623535, "learning_rate": 0.0002, "epoch": 4.8040201005025125, "step": 7170}, {"loss": 1.4088, "grad_norm": 0.7728400826454163, "learning_rate": 0.0002, "epoch": 4.8107202680067, "step": 7180}, {"loss": 1.3814, "grad_norm": 0.7992154955863953, "learning_rate": 0.0002, "epoch": 4.817420435510888, "step": 7190}, {"loss": 1.3958, "grad_norm": 0.8655321002006531, "learning_rate": 0.0002, "epoch": 4.824120603015075, "step": 7200}, {"loss": 1.3837, "grad_norm": 0.7672632336616516, "learning_rate": 0.0002, "epoch": 4.830820770519263, "step": 7210}, {"loss": 1.4578, "grad_norm": 0.708416223526001, "learning_rate": 0.0002, "epoch": 4.8375209380234505, "step": 7220}, {"loss": 1.5413, "grad_norm": 0.8914081454277039, "learning_rate": 0.0002, "epoch": 4.844221105527638, "step": 7230}, {"loss": 1.3569, "grad_norm": 0.7141931653022766, "learning_rate": 0.0002, "epoch": 4.850921273031826, "step": 7240}, {"loss": 1.4532, "grad_norm": 0.6913040280342102, "learning_rate": 0.0002, "epoch": 4.857621440536013, "step": 7250}, {"loss": 1.3912, "grad_norm": 0.7871233820915222, "learning_rate": 0.0002, "epoch": 4.864321608040201, "step": 7260}, {"loss": 1.3688, "grad_norm": 0.8466277122497559, "learning_rate": 0.0002, "epoch": 4.8710217755443885, "step": 7270}, {"loss": 1.33, "grad_norm": 0.8492183685302734, "learning_rate": 0.0002, "epoch": 4.877721943048576, "step": 7280}, {"loss": 1.3744, "grad_norm": 0.8339574933052063, "learning_rate": 0.0002, "epoch": 4.884422110552764, "step": 7290}, {"loss": 1.4157, "grad_norm": 0.787022590637207, "learning_rate": 0.0002, "epoch": 4.891122278056951, "step": 7300}, {"loss": 1.3725, "grad_norm": 0.8877332806587219, "learning_rate": 0.0002, "epoch": 4.897822445561139, "step": 7310}, {"loss": 1.3968, "grad_norm": 0.744989812374115, "learning_rate": 0.0002, "epoch": 4.9045226130653266, "step": 7320}, {"loss": 1.4421, "grad_norm": 0.8027268648147583, "learning_rate": 0.0002, "epoch": 4.911222780569514, "step": 7330}, {"loss": 1.425, "grad_norm": 0.6437455415725708, "learning_rate": 0.0002, "epoch": 4.917922948073702, "step": 7340}, {"loss": 1.4829, "grad_norm": 0.685999870300293, "learning_rate": 0.0002, "epoch": 4.924623115577889, "step": 7350}, {"loss": 1.4352, "grad_norm": 0.9086187481880188, "learning_rate": 0.0002, "epoch": 4.931323283082077, "step": 7360}, {"loss": 1.4245, "grad_norm": 0.8272411227226257, "learning_rate": 0.0002, "epoch": 4.938023450586265, "step": 7370}, {"loss": 1.4226, "grad_norm": 0.9227852821350098, "learning_rate": 0.0002, "epoch": 4.944723618090452, "step": 7380}, {"loss": 1.3643, "grad_norm": 0.7688441276550293, "learning_rate": 0.0002, "epoch": 4.95142378559464, "step": 7390}, {"loss": 1.4491, "grad_norm": 0.8662643432617188, "learning_rate": 0.0002, "epoch": 4.958123953098827, "step": 7400}, {"loss": 1.4194, "grad_norm": 0.9234127998352051, "learning_rate": 0.0002, "epoch": 4.964824120603015, "step": 7410}, {"loss": 1.4009, "grad_norm": 0.9131470918655396, "learning_rate": 0.0002, "epoch": 4.971524288107203, "step": 7420}, {"loss": 1.4544, "grad_norm": 0.7377504110336304, "learning_rate": 0.0002, "epoch": 4.97822445561139, "step": 7430}, {"loss": 1.4008, "grad_norm": 0.8762801289558411, "learning_rate": 0.0002, "epoch": 4.984924623115578, "step": 7440}, {"loss": 1.4304, "grad_norm": 0.7919872999191284, "learning_rate": 0.0002, "epoch": 4.991624790619765, "step": 7450}, {"loss": 1.3817, "grad_norm": 0.7144299149513245, "learning_rate": 0.0002, "epoch": 4.998324958123953, "step": 7460}, {"eval_loss": 1.9291157722473145, "eval_runtime": 37.9831, "eval_samples_per_second": 13.559, "eval_steps_per_second": 1.711, "epoch": 4.99966499162479, "step": 7462}, {"loss": 1.2753, "grad_norm": 0.7860151529312134, "learning_rate": 0.0002, "epoch": 5.005025125628141, "step": 7470}, {"loss": 1.2149, "grad_norm": 0.9418314695358276, "learning_rate": 0.0002, "epoch": 5.011725293132328, "step": 7480}, {"loss": 1.1966, "grad_norm": 0.8474572896957397, "learning_rate": 0.0002, "epoch": 5.018425460636516, "step": 7490}, {"loss": 1.2111, "grad_norm": 1.0724040269851685, "learning_rate": 0.0002, "epoch": 5.025125628140704, "step": 7500}, {"loss": 1.2228, "grad_norm": 0.9109148979187012, "learning_rate": 0.0002, "epoch": 5.031825795644891, "step": 7510}, {"loss": 1.2239, "grad_norm": 1.0088659524917603, "learning_rate": 0.0002, "epoch": 5.038525963149079, "step": 7520}, {"loss": 1.2156, "grad_norm": 1.1421623229980469, "learning_rate": 0.0002, "epoch": 5.045226130653266, "step": 7530}, {"loss": 1.1739, "grad_norm": 0.9219902157783508, "learning_rate": 0.0002, "epoch": 5.051926298157454, "step": 7540}, {"loss": 1.2686, "grad_norm": 0.9150987863540649, "learning_rate": 0.0002, "epoch": 5.058626465661642, "step": 7550}, {"loss": 1.2068, "grad_norm": 0.8889328241348267, "learning_rate": 0.0002, "epoch": 5.065326633165829, "step": 7560}, {"loss": 1.276, "grad_norm": 0.9751363396644592, "learning_rate": 0.0002, "epoch": 5.072026800670017, "step": 7570}, {"loss": 1.2078, "grad_norm": 0.8603123426437378, "learning_rate": 0.0002, "epoch": 5.078726968174204, "step": 7580}, {"loss": 1.2175, "grad_norm": 0.8910616636276245, "learning_rate": 0.0002, "epoch": 5.085427135678392, "step": 7590}, {"loss": 1.2475, "grad_norm": 1.1128392219543457, "learning_rate": 0.0002, "epoch": 5.09212730318258, "step": 7600}, {"loss": 1.3065, "grad_norm": 0.9480258822441101, "learning_rate": 0.0002, "epoch": 5.098827470686767, "step": 7610}, {"loss": 1.193, "grad_norm": 0.906958818435669, "learning_rate": 0.0002, "epoch": 5.105527638190955, "step": 7620}, {"loss": 1.2223, "grad_norm": 0.8741167187690735, "learning_rate": 0.0002, "epoch": 5.1122278056951425, "step": 7630}, {"loss": 1.2126, "grad_norm": 0.966268002986908, "learning_rate": 0.0002, "epoch": 5.11892797319933, "step": 7640}, {"loss": 1.2782, "grad_norm": 0.9124358892440796, "learning_rate": 0.0002, "epoch": 5.125628140703517, "step": 7650}, {"loss": 1.3004, "grad_norm": 1.0436606407165527, "learning_rate": 0.0002, "epoch": 5.132328308207705, "step": 7660}, {"loss": 1.2675, "grad_norm": 0.9217309355735779, "learning_rate": 0.0002, "epoch": 5.139028475711893, "step": 7670}, {"loss": 1.2502, "grad_norm": 1.344765543937683, "learning_rate": 0.0002, "epoch": 5.1457286432160805, "step": 7680}, {"loss": 1.2416, "grad_norm": 1.0730723142623901, "learning_rate": 0.0002, "epoch": 5.152428810720268, "step": 7690}, {"loss": 1.1888, "grad_norm": 0.9321247339248657, "learning_rate": 0.0002, "epoch": 5.159128978224456, "step": 7700}, {"loss": 1.1941, "grad_norm": 0.8482614755630493, "learning_rate": 0.0002, "epoch": 5.165829145728643, "step": 7710}, {"loss": 1.2668, "grad_norm": 0.8274452686309814, "learning_rate": 0.0002, "epoch": 5.172529313232831, "step": 7720}, {"loss": 1.1972, "grad_norm": 0.9120376706123352, "learning_rate": 0.0002, "epoch": 5.1792294807370185, "step": 7730}, {"loss": 1.1648, "grad_norm": 1.0062892436981201, "learning_rate": 0.0002, "epoch": 5.185929648241206, "step": 7740}, {"loss": 1.2199, "grad_norm": 0.9521504640579224, "learning_rate": 0.0002, "epoch": 5.192629815745394, "step": 7750}, {"loss": 1.2855, "grad_norm": 0.8800198435783386, "learning_rate": 0.0002, "epoch": 5.199329983249581, "step": 7760}, {"loss": 1.2535, "grad_norm": 0.9749179482460022, "learning_rate": 0.0002, "epoch": 5.206030150753769, "step": 7770}, {"loss": 1.2975, "grad_norm": 0.9441686868667603, "learning_rate": 0.0002, "epoch": 5.2127303182579565, "step": 7780}, {"loss": 1.256, "grad_norm": 0.9114066362380981, "learning_rate": 0.0002, "epoch": 5.219430485762144, "step": 7790}, {"loss": 1.2621, "grad_norm": 0.9851446151733398, "learning_rate": 0.0002, "epoch": 5.226130653266332, "step": 7800}, {"loss": 1.2502, "grad_norm": 0.9526297450065613, "learning_rate": 0.0002, "epoch": 5.232830820770519, "step": 7810}, {"loss": 1.1502, "grad_norm": 1.05986487865448, "learning_rate": 0.0002, "epoch": 5.239530988274707, "step": 7820}, {"loss": 1.2517, "grad_norm": 0.8956538438796997, "learning_rate": 0.0002, "epoch": 5.2462311557788945, "step": 7830}, {"loss": 1.2556, "grad_norm": 0.9568153619766235, "learning_rate": 0.0002, "epoch": 5.252931323283082, "step": 7840}, {"loss": 1.2442, "grad_norm": 1.0035018920898438, "learning_rate": 0.0002, "epoch": 5.259631490787269, "step": 7850}, {"loss": 1.2605, "grad_norm": 0.8554368615150452, "learning_rate": 0.0002, "epoch": 5.266331658291457, "step": 7860}, {"loss": 1.2799, "grad_norm": 0.9677708148956299, "learning_rate": 0.0002, "epoch": 5.273031825795645, "step": 7870}, {"loss": 1.275, "grad_norm": 0.943606436252594, "learning_rate": 0.0002, "epoch": 5.279731993299833, "step": 7880}, {"loss": 1.2335, "grad_norm": 1.0029335021972656, "learning_rate": 0.0002, "epoch": 5.28643216080402, "step": 7890}, {"loss": 1.2494, "grad_norm": 1.0164015293121338, "learning_rate": 0.0002, "epoch": 5.293132328308207, "step": 7900}, {"loss": 1.3117, "grad_norm": 0.8908365368843079, "learning_rate": 0.0002, "epoch": 5.299832495812395, "step": 7910}, {"loss": 1.2832, "grad_norm": 0.9307826161384583, "learning_rate": 0.0002, "epoch": 5.306532663316583, "step": 7920}, {"loss": 1.242, "grad_norm": 1.0730371475219727, "learning_rate": 0.0002, "epoch": 5.313232830820771, "step": 7930}, {"loss": 1.2003, "grad_norm": 0.844739556312561, "learning_rate": 0.0002, "epoch": 5.319932998324958, "step": 7940}, {"loss": 1.2688, "grad_norm": 1.275833010673523, "learning_rate": 0.0002, "epoch": 5.326633165829146, "step": 7950}, {"loss": 1.2957, "grad_norm": 0.9042661190032959, "learning_rate": 0.0002, "epoch": 5.333333333333333, "step": 7960}, {"loss": 1.2912, "grad_norm": 0.9374269247055054, "learning_rate": 0.0002, "epoch": 5.340033500837521, "step": 7970}, {"loss": 1.2721, "grad_norm": 1.033098578453064, "learning_rate": 0.0002, "epoch": 5.346733668341709, "step": 7980}, {"loss": 1.3208, "grad_norm": 1.062775731086731, "learning_rate": 0.0002, "epoch": 5.353433835845896, "step": 7990}, {"loss": 1.3065, "grad_norm": 1.1064317226409912, "learning_rate": 0.0002, "epoch": 5.360134003350084, "step": 8000}, {"loss": 1.2341, "grad_norm": 1.1114039421081543, "learning_rate": 0.0002, "epoch": 5.366834170854271, "step": 8010}, {"loss": 1.2255, "grad_norm": 1.0198014974594116, "learning_rate": 0.0002, "epoch": 5.373534338358459, "step": 8020}, {"loss": 1.2433, "grad_norm": 0.8443173170089722, "learning_rate": 0.0002, "epoch": 5.380234505862647, "step": 8030}, {"loss": 1.206, "grad_norm": 1.000881314277649, "learning_rate": 0.0002, "epoch": 5.386934673366834, "step": 8040}, {"loss": 1.2982, "grad_norm": 0.9874443411827087, "learning_rate": 0.0002, "epoch": 5.393634840871022, "step": 8050}, {"loss": 1.2289, "grad_norm": 0.9895344972610474, "learning_rate": 0.0002, "epoch": 5.400335008375209, "step": 8060}, {"loss": 1.249, "grad_norm": 0.8595236539840698, "learning_rate": 0.0002, "epoch": 5.407035175879397, "step": 8070}, {"loss": 1.2308, "grad_norm": 0.9523849487304688, "learning_rate": 0.0002, "epoch": 5.413735343383585, "step": 8080}, {"loss": 1.2343, "grad_norm": 1.0560476779937744, "learning_rate": 0.0002, "epoch": 5.420435510887772, "step": 8090}, {"loss": 1.2956, "grad_norm": 1.0893689393997192, "learning_rate": 0.0002, "epoch": 5.42713567839196, "step": 8100}, {"loss": 1.2846, "grad_norm": 0.9395513534545898, "learning_rate": 0.0002, "epoch": 5.433835845896147, "step": 8110}, {"loss": 1.3444, "grad_norm": 0.9364215135574341, "learning_rate": 0.0002, "epoch": 5.440536013400335, "step": 8120}, {"loss": 1.2944, "grad_norm": 0.9502208232879639, "learning_rate": 0.0002, "epoch": 5.447236180904523, "step": 8130}, {"loss": 1.2971, "grad_norm": 0.9559208154678345, "learning_rate": 0.0002, "epoch": 5.45393634840871, "step": 8140}, {"loss": 1.2495, "grad_norm": 0.9261730313301086, "learning_rate": 0.0002, "epoch": 5.460636515912898, "step": 8150}, {"loss": 1.2599, "grad_norm": 0.9832326173782349, "learning_rate": 0.0002, "epoch": 5.467336683417085, "step": 8160}, {"loss": 1.2771, "grad_norm": 1.065953016281128, "learning_rate": 0.0002, "epoch": 5.474036850921273, "step": 8170}, {"loss": 1.3617, "grad_norm": 0.9139469861984253, "learning_rate": 0.0002, "epoch": 5.480737018425461, "step": 8180}, {"loss": 1.2197, "grad_norm": 1.2322484254837036, "learning_rate": 0.0002, "epoch": 5.4874371859296485, "step": 8190}, {"loss": 1.2879, "grad_norm": 0.9722974896430969, "learning_rate": 0.0002, "epoch": 5.494137353433836, "step": 8200}, {"loss": 1.2664, "grad_norm": 0.9338926076889038, "learning_rate": 0.0002, "epoch": 5.500837520938023, "step": 8210}, {"loss": 1.2128, "grad_norm": 0.9283728003501892, "learning_rate": 0.0002, "epoch": 5.507537688442211, "step": 8220}, {"loss": 1.2141, "grad_norm": 1.0489585399627686, "learning_rate": 0.0002, "epoch": 5.514237855946399, "step": 8230}, {"loss": 1.3257, "grad_norm": 0.9881814122200012, "learning_rate": 0.0002, "epoch": 5.5209380234505865, "step": 8240}, {"loss": 1.2221, "grad_norm": 0.9274460077285767, "learning_rate": 0.0002, "epoch": 5.527638190954773, "step": 8250}, {"loss": 1.2241, "grad_norm": 0.8650718331336975, "learning_rate": 0.0002, "epoch": 5.534338358458961, "step": 8260}, {"loss": 1.2462, "grad_norm": 1.014069676399231, "learning_rate": 0.0002, "epoch": 5.541038525963149, "step": 8270}, {"loss": 1.3502, "grad_norm": 0.9212974905967712, "learning_rate": 0.0002, "epoch": 5.547738693467337, "step": 8280}, {"loss": 1.2779, "grad_norm": 1.1235398054122925, "learning_rate": 0.0002, "epoch": 5.5544388609715245, "step": 8290}, {"loss": 1.306, "grad_norm": 0.961954653263092, "learning_rate": 0.0002, "epoch": 5.561139028475711, "step": 8300}, {"loss": 1.2946, "grad_norm": 0.9386700391769409, "learning_rate": 0.0002, "epoch": 5.567839195979899, "step": 8310}, {"loss": 1.313, "grad_norm": 1.01912522315979, "learning_rate": 0.0002, "epoch": 5.574539363484087, "step": 8320}, {"loss": 1.3121, "grad_norm": 0.9851216077804565, "learning_rate": 0.0002, "epoch": 5.581239530988275, "step": 8330}, {"loss": 1.3071, "grad_norm": 1.0138001441955566, "learning_rate": 0.0002, "epoch": 5.5879396984924625, "step": 8340}, {"loss": 1.2257, "grad_norm": 0.9262447357177734, "learning_rate": 0.0002, "epoch": 5.594639865996649, "step": 8350}, {"loss": 1.2473, "grad_norm": 1.1322970390319824, "learning_rate": 0.0002, "epoch": 5.601340033500837, "step": 8360}, {"loss": 1.3098, "grad_norm": 1.1429349184036255, "learning_rate": 0.0002, "epoch": 5.608040201005025, "step": 8370}, {"loss": 1.2686, "grad_norm": 0.9130118489265442, "learning_rate": 0.0002, "epoch": 5.614740368509213, "step": 8380}, {"loss": 1.2541, "grad_norm": 0.9651545882225037, "learning_rate": 0.0002, "epoch": 5.6214405360134005, "step": 8390}, {"loss": 1.2799, "grad_norm": 0.9595398306846619, "learning_rate": 0.0002, "epoch": 5.628140703517588, "step": 8400}, {"loss": 1.3429, "grad_norm": 1.0049372911453247, "learning_rate": 0.0002, "epoch": 5.634840871021775, "step": 8410}, {"loss": 1.3224, "grad_norm": 1.082804560661316, "learning_rate": 0.0002, "epoch": 5.641541038525963, "step": 8420}, {"loss": 1.297, "grad_norm": 0.9489204287528992, "learning_rate": 0.0002, "epoch": 5.648241206030151, "step": 8430}, {"loss": 1.3424, "grad_norm": 0.9470235109329224, "learning_rate": 0.0002, "epoch": 5.654941373534339, "step": 8440}, {"loss": 1.3358, "grad_norm": 1.0662927627563477, "learning_rate": 0.0002, "epoch": 5.661641541038526, "step": 8450}, {"loss": 1.2973, "grad_norm": 0.9097877740859985, "learning_rate": 0.0002, "epoch": 5.668341708542713, "step": 8460}, {"loss": 1.3072, "grad_norm": 0.9740368127822876, "learning_rate": 0.0002, "epoch": 5.675041876046901, "step": 8470}, {"loss": 1.286, "grad_norm": 0.9878810048103333, "learning_rate": 0.0002, "epoch": 5.681742043551089, "step": 8480}, {"loss": 1.208, "grad_norm": 1.148260474205017, "learning_rate": 0.0002, "epoch": 5.688442211055277, "step": 8490}, {"loss": 1.2842, "grad_norm": 0.9632558822631836, "learning_rate": 0.0002, "epoch": 5.695142378559464, "step": 8500}, {"loss": 1.2787, "grad_norm": 0.876812756061554, "learning_rate": 0.0002, "epoch": 5.701842546063651, "step": 8510}, {"loss": 1.3186, "grad_norm": 1.0730829238891602, "learning_rate": 0.0002, "epoch": 5.708542713567839, "step": 8520}, {"loss": 1.2856, "grad_norm": 1.2239218950271606, "learning_rate": 0.0002, "epoch": 5.715242881072027, "step": 8530}, {"loss": 1.2717, "grad_norm": 0.9460835456848145, "learning_rate": 0.0002, "epoch": 5.721943048576215, "step": 8540}, {"loss": 1.3509, "grad_norm": 0.9086270928382874, "learning_rate": 0.0002, "epoch": 5.728643216080402, "step": 8550}, {"loss": 1.2971, "grad_norm": 1.0258867740631104, "learning_rate": 0.0002, "epoch": 5.735343383584589, "step": 8560}, {"loss": 1.3581, "grad_norm": 1.0543923377990723, "learning_rate": 0.0002, "epoch": 5.742043551088777, "step": 8570}, {"loss": 1.2988, "grad_norm": 0.9063900113105774, "learning_rate": 0.0002, "epoch": 5.748743718592965, "step": 8580}, {"loss": 1.3535, "grad_norm": 1.1838830709457397, "learning_rate": 0.0002, "epoch": 5.755443886097153, "step": 8590}, {"loss": 1.2655, "grad_norm": 0.9631859064102173, "learning_rate": 0.0002, "epoch": 5.76214405360134, "step": 8600}, {"loss": 1.276, "grad_norm": 0.9702655673027039, "learning_rate": 0.0002, "epoch": 5.768844221105527, "step": 8610}, {"loss": 1.3196, "grad_norm": 1.0591435432434082, "learning_rate": 0.0002, "epoch": 5.775544388609715, "step": 8620}, {"loss": 1.267, "grad_norm": 0.9989570379257202, "learning_rate": 0.0002, "epoch": 5.782244556113903, "step": 8630}, {"loss": 1.3227, "grad_norm": 1.0836435556411743, "learning_rate": 0.0002, "epoch": 5.788944723618091, "step": 8640}, {"loss": 1.3334, "grad_norm": 0.8832896947860718, "learning_rate": 0.0002, "epoch": 5.795644891122278, "step": 8650}, {"loss": 1.3214, "grad_norm": 1.0104607343673706, "learning_rate": 0.0002, "epoch": 5.802345058626465, "step": 8660}, {"loss": 1.2703, "grad_norm": 0.8375084400177002, "learning_rate": 0.0002, "epoch": 5.809045226130653, "step": 8670}, {"loss": 1.3554, "grad_norm": 1.1300716400146484, "learning_rate": 0.0002, "epoch": 5.815745393634841, "step": 8680}, {"loss": 1.3468, "grad_norm": 0.9311910271644592, "learning_rate": 0.0002, "epoch": 5.822445561139029, "step": 8690}, {"loss": 1.2749, "grad_norm": 0.9488391876220703, "learning_rate": 0.0002, "epoch": 5.8291457286432165, "step": 8700}, {"loss": 1.2281, "grad_norm": 0.9747629761695862, "learning_rate": 0.0002, "epoch": 5.835845896147403, "step": 8710}, {"loss": 1.2923, "grad_norm": 1.1029598712921143, "learning_rate": 0.0002, "epoch": 5.842546063651591, "step": 8720}, {"loss": 1.3613, "grad_norm": 1.0396875143051147, "learning_rate": 0.0002, "epoch": 5.849246231155779, "step": 8730}, {"loss": 1.3272, "grad_norm": 0.9259780645370483, "learning_rate": 0.0002, "epoch": 5.855946398659967, "step": 8740}, {"loss": 1.3236, "grad_norm": 1.020033597946167, "learning_rate": 0.0002, "epoch": 5.8626465661641545, "step": 8750}, {"loss": 1.3453, "grad_norm": 0.9191218614578247, "learning_rate": 0.0002, "epoch": 5.869346733668341, "step": 8760}, {"loss": 1.3012, "grad_norm": 1.1093107461929321, "learning_rate": 0.0002, "epoch": 5.876046901172529, "step": 8770}, {"loss": 1.2718, "grad_norm": 1.1626793146133423, "learning_rate": 0.0002, "epoch": 5.882747068676717, "step": 8780}, {"loss": 1.2969, "grad_norm": 0.9542945027351379, "learning_rate": 0.0002, "epoch": 5.889447236180905, "step": 8790}, {"loss": 1.3134, "grad_norm": 0.9086058139801025, "learning_rate": 0.0002, "epoch": 5.8961474036850925, "step": 8800}, {"loss": 1.2731, "grad_norm": 0.9249639511108398, "learning_rate": 0.0002, "epoch": 5.902847571189279, "step": 8810}, {"loss": 1.337, "grad_norm": 0.9414396286010742, "learning_rate": 0.0002, "epoch": 5.909547738693467, "step": 8820}, {"loss": 1.2865, "grad_norm": 0.9086037874221802, "learning_rate": 0.0002, "epoch": 5.916247906197655, "step": 8830}, {"loss": 1.2756, "grad_norm": 0.8685907125473022, "learning_rate": 0.0002, "epoch": 5.922948073701843, "step": 8840}, {"loss": 1.297, "grad_norm": 1.036419153213501, "learning_rate": 0.0002, "epoch": 5.9296482412060305, "step": 8850}, {"loss": 1.3207, "grad_norm": 1.0183674097061157, "learning_rate": 0.0002, "epoch": 5.936348408710217, "step": 8860}, {"loss": 1.3922, "grad_norm": 0.966444194316864, "learning_rate": 0.0002, "epoch": 5.943048576214405, "step": 8870}, {"loss": 1.333, "grad_norm": 1.125693917274475, "learning_rate": 0.0002, "epoch": 5.949748743718593, "step": 8880}, {"loss": 1.3116, "grad_norm": 0.9857436418533325, "learning_rate": 0.0002, "epoch": 5.956448911222781, "step": 8890}, {"loss": 1.2526, "grad_norm": 0.9377069473266602, "learning_rate": 0.0002, "epoch": 5.9631490787269685, "step": 8900}, {"loss": 1.3221, "grad_norm": 0.9493814706802368, "learning_rate": 0.0002, "epoch": 5.969849246231155, "step": 8910}, {"loss": 1.2516, "grad_norm": 0.8806208372116089, "learning_rate": 0.0002, "epoch": 5.976549413735343, "step": 8920}, {"loss": 1.2558, "grad_norm": 0.8727600574493408, "learning_rate": 0.0002, "epoch": 5.983249581239531, "step": 8930}, {"loss": 1.3538, "grad_norm": 0.9799810647964478, "learning_rate": 0.0002, "epoch": 5.989949748743719, "step": 8940}, {"loss": 1.3323, "grad_norm": 0.9866513609886169, "learning_rate": 0.0002, "epoch": 5.9966499162479066, "step": 8950}]}
+{"epoch": 6.99966499162479, "step": 10447, "epoch_duration": 1589.0742285251617, "total_accumulated_duration": 11190.41557765007, "gpu_info": {"GPU_0": "NVIDIA A100-SXM4-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 7672.4541015625}, "peak_memory_usage": {"GPU_0": 13792.75439453125}, "avg_memory_reserved": {"GPU_0": 17416.0}, "peak_memory_reserved": {"GPU_0": 17416.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "outputs-001/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-2985", "params": {"epochs": 8, "batch_size": 1, "learning_rate": 0.0002, "gradient_accumulation_steps": 2, "warmup_ratio": 0.03, "max_grad_norm": 0.3, "lora_alpha": 32, "lora_dropout": 0.05, "lora_r": 64, "tf32": true, "seed": 42}, "log_history": [{"loss": 2.6189, "grad_norm": 0.565915048122406, "learning_rate": 0.0002, "epoch": 0.006700167504187605, "step": 10}, {"loss": 2.3162, "grad_norm": 0.5004463791847229, "learning_rate": 0.0002, "epoch": 0.01340033500837521, "step": 20}, {"loss": 2.0576, "grad_norm": 0.511043906211853, "learning_rate": 0.0002, "epoch": 0.020100502512562814, "step": 30}, {"loss": 2.0085, "grad_norm": 0.47327178716659546, "learning_rate": 0.0002, "epoch": 0.02680067001675042, "step": 40}, {"loss": 2.0276, "grad_norm": 0.5511676669120789, "learning_rate": 0.0002, "epoch": 0.03350083752093802, "step": 50}, {"loss": 1.9075, "grad_norm": 0.4666278064250946, "learning_rate": 0.0002, "epoch": 0.04020100502512563, "step": 60}, {"loss": 1.8413, "grad_norm": 0.5310961008071899, "learning_rate": 0.0002, "epoch": 0.04690117252931323, "step": 70}, {"loss": 1.8711, "grad_norm": 0.5606027245521545, "learning_rate": 0.0002, "epoch": 0.05360134003350084, "step": 80}, {"loss": 1.9282, "grad_norm": 0.4934779703617096, "learning_rate": 0.0002, "epoch": 0.06030150753768844, "step": 90}, {"loss": 1.8925, "grad_norm": 0.4821869730949402, "learning_rate": 0.0002, "epoch": 0.06700167504187604, "step": 100}, {"loss": 1.8628, "grad_norm": 0.5262084603309631, "learning_rate": 0.0002, "epoch": 0.07370184254606366, "step": 110}, {"loss": 1.8347, "grad_norm": 0.3774230182170868, "learning_rate": 0.0002, "epoch": 0.08040201005025126, "step": 120}, {"loss": 1.8386, "grad_norm": 0.34137430787086487, "learning_rate": 0.0002, "epoch": 0.08710217755443886, "step": 130}, {"loss": 1.861, "grad_norm": 0.407272070646286, "learning_rate": 0.0002, "epoch": 0.09380234505862646, "step": 140}, {"loss": 1.8279, "grad_norm": 0.4011937975883484, "learning_rate": 0.0002, "epoch": 0.10050251256281408, "step": 150}, {"loss": 1.9317, "grad_norm": 0.4432467222213745, "learning_rate": 0.0002, "epoch": 0.10720268006700168, "step": 160}, {"loss": 1.8157, "grad_norm": 0.44030463695526123, "learning_rate": 0.0002, "epoch": 0.11390284757118928, "step": 170}, {"loss": 1.8534, "grad_norm": 0.3799569308757782, "learning_rate": 0.0002, "epoch": 0.12060301507537688, "step": 180}, {"loss": 1.7658, "grad_norm": 0.33721521496772766, "learning_rate": 0.0002, "epoch": 0.1273031825795645, "step": 190}, {"loss": 1.8269, "grad_norm": 0.4096226692199707, "learning_rate": 0.0002, "epoch": 0.13400335008375208, "step": 200}, {"loss": 1.802, "grad_norm": 0.37374693155288696, "learning_rate": 0.0002, "epoch": 0.1407035175879397, "step": 210}, {"loss": 1.8901, "grad_norm": 0.3249480128288269, "learning_rate": 0.0002, "epoch": 0.1474036850921273, "step": 220}, {"loss": 1.8163, "grad_norm": 0.3612042963504791, "learning_rate": 0.0002, "epoch": 0.1541038525963149, "step": 230}, {"loss": 1.7585, "grad_norm": 0.3686671257019043, "learning_rate": 0.0002, "epoch": 0.16080402010050251, "step": 240}, {"loss": 1.8365, "grad_norm": 0.3521044850349426, "learning_rate": 0.0002, "epoch": 0.16750418760469013, "step": 250}, {"loss": 1.8623, "grad_norm": 0.4073677361011505, "learning_rate": 0.0002, "epoch": 0.17420435510887772, "step": 260}, {"loss": 1.8026, "grad_norm": 0.34522193670272827, "learning_rate": 0.0002, "epoch": 0.18090452261306533, "step": 270}, {"loss": 1.8162, "grad_norm": 0.4121900498867035, "learning_rate": 0.0002, "epoch": 0.18760469011725292, "step": 280}, {"loss": 1.7976, "grad_norm": 0.3544778525829315, "learning_rate": 0.0002, "epoch": 0.19430485762144054, "step": 290}, {"loss": 1.8787, "grad_norm": 0.3482133448123932, "learning_rate": 0.0002, "epoch": 0.20100502512562815, "step": 300}, {"loss": 1.8578, "grad_norm": 0.3421826660633087, "learning_rate": 0.0002, "epoch": 0.20770519262981574, "step": 310}, {"loss": 1.8013, "grad_norm": 0.5024696588516235, "learning_rate": 0.0002, "epoch": 0.21440536013400335, "step": 320}, {"loss": 1.8607, "grad_norm": 0.36013063788414, "learning_rate": 0.0002, "epoch": 0.22110552763819097, "step": 330}, {"loss": 1.9075, "grad_norm": 0.3611244857311249, "learning_rate": 0.0002, "epoch": 0.22780569514237856, "step": 340}, {"loss": 1.8128, "grad_norm": 0.39244529604911804, "learning_rate": 0.0002, "epoch": 0.23450586264656617, "step": 350}, {"loss": 1.7885, "grad_norm": 0.3299325704574585, "learning_rate": 0.0002, "epoch": 0.24120603015075376, "step": 360}, {"loss": 1.8028, "grad_norm": 0.3994322419166565, "learning_rate": 0.0002, "epoch": 0.24790619765494137, "step": 370}, {"loss": 1.8321, "grad_norm": 0.3559151887893677, "learning_rate": 0.0002, "epoch": 0.254606365159129, "step": 380}, {"loss": 1.7802, "grad_norm": 0.3873756229877472, "learning_rate": 0.0002, "epoch": 0.2613065326633166, "step": 390}, {"loss": 1.7844, "grad_norm": 0.3710744082927704, "learning_rate": 0.0002, "epoch": 0.26800670016750416, "step": 400}, {"loss": 1.7789, "grad_norm": 0.3618465065956116, "learning_rate": 0.0002, "epoch": 0.2747068676716918, "step": 410}, {"loss": 1.8529, "grad_norm": 0.30063769221305847, "learning_rate": 0.0002, "epoch": 0.2814070351758794, "step": 420}, {"loss": 1.7765, "grad_norm": 0.3695628345012665, "learning_rate": 0.0002, "epoch": 0.288107202680067, "step": 430}, {"loss": 1.7982, "grad_norm": 0.31451135873794556, "learning_rate": 0.0002, "epoch": 0.2948073701842546, "step": 440}, {"loss": 1.7517, "grad_norm": 0.3959707021713257, "learning_rate": 0.0002, "epoch": 0.3015075376884422, "step": 450}, {"loss": 1.8142, "grad_norm": 0.33475354313850403, "learning_rate": 0.0002, "epoch": 0.3082077051926298, "step": 460}, {"loss": 1.8805, "grad_norm": 0.33933115005493164, "learning_rate": 0.0002, "epoch": 0.3149078726968174, "step": 470}, {"loss": 1.7564, "grad_norm": 0.3264943063259125, "learning_rate": 0.0002, "epoch": 0.32160804020100503, "step": 480}, {"loss": 1.8428, "grad_norm": 0.40188100934028625, "learning_rate": 0.0002, "epoch": 0.32830820770519265, "step": 490}, {"loss": 1.7624, "grad_norm": 0.37408649921417236, "learning_rate": 0.0002, "epoch": 0.33500837520938026, "step": 500}, {"loss": 1.7745, "grad_norm": 0.33925938606262207, "learning_rate": 0.0002, "epoch": 0.3417085427135678, "step": 510}, {"loss": 1.814, "grad_norm": 0.36836713552474976, "learning_rate": 0.0002, "epoch": 0.34840871021775544, "step": 520}, {"loss": 1.8037, "grad_norm": 0.37284499406814575, "learning_rate": 0.0002, "epoch": 0.35510887772194305, "step": 530}, {"loss": 1.8379, "grad_norm": 0.3192278742790222, "learning_rate": 0.0002, "epoch": 0.36180904522613067, "step": 540}, {"loss": 1.8702, "grad_norm": 0.30233290791511536, "learning_rate": 0.0002, "epoch": 0.3685092127303183, "step": 550}, {"loss": 1.8799, "grad_norm": 0.3340817391872406, "learning_rate": 0.0002, "epoch": 0.37520938023450584, "step": 560}, {"loss": 1.8404, "grad_norm": 0.32600095868110657, "learning_rate": 0.0002, "epoch": 0.38190954773869346, "step": 570}, {"loss": 1.7804, "grad_norm": 0.33711278438568115, "learning_rate": 0.0002, "epoch": 0.38860971524288107, "step": 580}, {"loss": 1.8445, "grad_norm": 0.34890690445899963, "learning_rate": 0.0002, "epoch": 0.3953098827470687, "step": 590}, {"loss": 1.8187, "grad_norm": 0.38238924741744995, "learning_rate": 0.0002, "epoch": 0.4020100502512563, "step": 600}, {"loss": 1.8111, "grad_norm": 0.34399354457855225, "learning_rate": 0.0002, "epoch": 0.40871021775544386, "step": 610}, {"loss": 1.8006, "grad_norm": 0.3346073627471924, "learning_rate": 0.0002, "epoch": 0.4154103852596315, "step": 620}, {"loss": 1.7705, "grad_norm": 0.3545648157596588, "learning_rate": 0.0002, "epoch": 0.4221105527638191, "step": 630}, {"loss": 1.8445, "grad_norm": 0.3378899097442627, "learning_rate": 0.0002, "epoch": 0.4288107202680067, "step": 640}, {"loss": 1.804, "grad_norm": 0.3255569040775299, "learning_rate": 0.0002, "epoch": 0.4355108877721943, "step": 650}, {"loss": 1.7679, "grad_norm": 0.34880587458610535, "learning_rate": 0.0002, "epoch": 0.44221105527638194, "step": 660}, {"loss": 1.7861, "grad_norm": 0.3402383625507355, "learning_rate": 0.0002, "epoch": 0.4489112227805695, "step": 670}, {"loss": 1.8131, "grad_norm": 0.3594033718109131, "learning_rate": 0.0002, "epoch": 0.4556113902847571, "step": 680}, {"loss": 1.8399, "grad_norm": 0.31000566482543945, "learning_rate": 0.0002, "epoch": 0.4623115577889447, "step": 690}, {"loss": 1.7521, "grad_norm": 0.37229061126708984, "learning_rate": 0.0002, "epoch": 0.46901172529313234, "step": 700}, {"loss": 1.7779, "grad_norm": 0.315801739692688, "learning_rate": 0.0002, "epoch": 0.47571189279731996, "step": 710}, {"loss": 1.7515, "grad_norm": 0.3220832645893097, "learning_rate": 0.0002, "epoch": 0.4824120603015075, "step": 720}, {"loss": 1.7181, "grad_norm": 0.3435456156730652, "learning_rate": 0.0002, "epoch": 0.48911222780569513, "step": 730}, {"loss": 1.8844, "grad_norm": 0.30380892753601074, "learning_rate": 0.0002, "epoch": 0.49581239530988275, "step": 740}, {"loss": 1.7792, "grad_norm": 0.3555026054382324, "learning_rate": 0.0002, "epoch": 0.5025125628140703, "step": 750}, {"loss": 1.7714, "grad_norm": 0.3019855320453644, "learning_rate": 0.0002, "epoch": 0.509212730318258, "step": 760}, {"loss": 1.7962, "grad_norm": 0.309111088514328, "learning_rate": 0.0002, "epoch": 0.5159128978224455, "step": 770}, {"loss": 1.7913, "grad_norm": 0.366020530462265, "learning_rate": 0.0002, "epoch": 0.5226130653266332, "step": 780}, {"loss": 1.8008, "grad_norm": 0.3267050087451935, "learning_rate": 0.0002, "epoch": 0.5293132328308208, "step": 790}, {"loss": 1.7397, "grad_norm": 0.34265750646591187, "learning_rate": 0.0002, "epoch": 0.5360134003350083, "step": 800}, {"loss": 1.8251, "grad_norm": 0.313669890165329, "learning_rate": 0.0002, "epoch": 0.542713567839196, "step": 810}, {"loss": 1.8369, "grad_norm": 0.3355236053466797, "learning_rate": 0.0002, "epoch": 0.5494137353433836, "step": 820}, {"loss": 1.7381, "grad_norm": 0.3186608552932739, "learning_rate": 0.0002, "epoch": 0.5561139028475712, "step": 830}, {"loss": 1.8034, "grad_norm": 0.30357518792152405, "learning_rate": 0.0002, "epoch": 0.5628140703517588, "step": 840}, {"loss": 1.769, "grad_norm": 0.3990040123462677, "learning_rate": 0.0002, "epoch": 0.5695142378559463, "step": 850}, {"loss": 1.7482, "grad_norm": 0.34363803267478943, "learning_rate": 0.0002, "epoch": 0.576214405360134, "step": 860}, {"loss": 1.8106, "grad_norm": 0.3757908046245575, "learning_rate": 0.0002, "epoch": 0.5829145728643216, "step": 870}, {"loss": 1.8104, "grad_norm": 0.3359757661819458, "learning_rate": 0.0002, "epoch": 0.5896147403685092, "step": 880}, {"loss": 1.7591, "grad_norm": 0.5555329918861389, "learning_rate": 0.0002, "epoch": 0.5963149078726968, "step": 890}, {"loss": 1.7715, "grad_norm": 0.4046323895454407, "learning_rate": 0.0002, "epoch": 0.6030150753768844, "step": 900}, {"loss": 1.7998, "grad_norm": 0.29834219813346863, "learning_rate": 0.0002, "epoch": 0.609715242881072, "step": 910}, {"loss": 1.7826, "grad_norm": 0.3241238594055176, "learning_rate": 0.0002, "epoch": 0.6164154103852596, "step": 920}, {"loss": 1.8342, "grad_norm": 0.35154739022254944, "learning_rate": 0.0002, "epoch": 0.6231155778894473, "step": 930}, {"loss": 1.8076, "grad_norm": 0.3287706673145294, "learning_rate": 0.0002, "epoch": 0.6298157453936348, "step": 940}, {"loss": 1.8038, "grad_norm": 0.35670626163482666, "learning_rate": 0.0002, "epoch": 0.6365159128978225, "step": 950}, {"loss": 1.869, "grad_norm": 0.6114104986190796, "learning_rate": 0.0002, "epoch": 0.6432160804020101, "step": 960}, {"loss": 1.8297, "grad_norm": 0.3186565041542053, "learning_rate": 0.0002, "epoch": 0.6499162479061976, "step": 970}, {"loss": 1.7539, "grad_norm": 0.27164125442504883, "learning_rate": 0.0002, "epoch": 0.6566164154103853, "step": 980}, {"loss": 1.8339, "grad_norm": 0.34407344460487366, "learning_rate": 0.0002, "epoch": 0.6633165829145728, "step": 990}, {"loss": 1.855, "grad_norm": 0.368415892124176, "learning_rate": 0.0002, "epoch": 0.6700167504187605, "step": 1000}, {"loss": 1.7821, "grad_norm": 0.3306390643119812, "learning_rate": 0.0002, "epoch": 0.6767169179229481, "step": 1010}, {"loss": 1.786, "grad_norm": 0.3198648989200592, "learning_rate": 0.0002, "epoch": 0.6834170854271356, "step": 1020}, {"loss": 1.816, "grad_norm": 0.3092987537384033, "learning_rate": 0.0002, "epoch": 0.6901172529313233, "step": 1030}, {"loss": 1.7689, "grad_norm": 0.3090653419494629, "learning_rate": 0.0002, "epoch": 0.6968174204355109, "step": 1040}, {"loss": 1.7544, "grad_norm": 0.3485880196094513, "learning_rate": 0.0002, "epoch": 0.7035175879396985, "step": 1050}, {"loss": 1.8187, "grad_norm": 0.35782721638679504, "learning_rate": 0.0002, "epoch": 0.7102177554438861, "step": 1060}, {"loss": 1.806, "grad_norm": 0.34256869554519653, "learning_rate": 0.0002, "epoch": 0.7169179229480737, "step": 1070}, {"loss": 1.7873, "grad_norm": 0.30461037158966064, "learning_rate": 0.0002, "epoch": 0.7236180904522613, "step": 1080}, {"loss": 1.7367, "grad_norm": 0.3398691713809967, "learning_rate": 0.0002, "epoch": 0.7303182579564489, "step": 1090}, {"loss": 1.8756, "grad_norm": 0.3180808126926422, "learning_rate": 0.0002, "epoch": 0.7370184254606366, "step": 1100}, {"loss": 1.6988, "grad_norm": 0.34400665760040283, "learning_rate": 0.0002, "epoch": 0.7437185929648241, "step": 1110}, {"loss": 1.7851, "grad_norm": 0.34244877099990845, "learning_rate": 0.0002, "epoch": 0.7504187604690117, "step": 1120}, {"loss": 1.7841, "grad_norm": 0.29946693778038025, "learning_rate": 0.0002, "epoch": 0.7571189279731994, "step": 1130}, {"loss": 1.7456, "grad_norm": 0.37547236680984497, "learning_rate": 0.0002, "epoch": 0.7638190954773869, "step": 1140}, {"loss": 1.8425, "grad_norm": 0.3263005018234253, "learning_rate": 0.0002, "epoch": 0.7705192629815746, "step": 1150}, {"loss": 1.7222, "grad_norm": 0.41363608837127686, "learning_rate": 0.0002, "epoch": 0.7772194304857621, "step": 1160}, {"loss": 1.7836, "grad_norm": 0.36267954111099243, "learning_rate": 0.0002, "epoch": 0.7839195979899497, "step": 1170}, {"loss": 1.9183, "grad_norm": 0.31789499521255493, "learning_rate": 0.0002, "epoch": 0.7906197654941374, "step": 1180}, {"loss": 1.78, "grad_norm": 0.5708149075508118, "learning_rate": 0.0002, "epoch": 0.7973199329983249, "step": 1190}, {"loss": 1.6908, "grad_norm": 0.322099506855011, "learning_rate": 0.0002, "epoch": 0.8040201005025126, "step": 1200}, {"loss": 1.7639, "grad_norm": 0.3419909179210663, "learning_rate": 0.0002, "epoch": 0.8107202680067002, "step": 1210}, {"loss": 1.7428, "grad_norm": 0.36286255717277527, "learning_rate": 0.0002, "epoch": 0.8174204355108877, "step": 1220}, {"loss": 1.8409, "grad_norm": 0.33992862701416016, "learning_rate": 0.0002, "epoch": 0.8241206030150754, "step": 1230}, {"loss": 1.7507, "grad_norm": 0.32622793316841125, "learning_rate": 0.0002, "epoch": 0.830820770519263, "step": 1240}, {"loss": 1.8098, "grad_norm": 0.3036167621612549, "learning_rate": 0.0002, "epoch": 0.8375209380234506, "step": 1250}, {"loss": 1.8094, "grad_norm": 0.3182215392589569, "learning_rate": 0.0002, "epoch": 0.8442211055276382, "step": 1260}, {"loss": 1.8017, "grad_norm": 0.3270018696784973, "learning_rate": 0.0002, "epoch": 0.8509212730318257, "step": 1270}, {"loss": 1.798, "grad_norm": 0.32652342319488525, "learning_rate": 0.0002, "epoch": 0.8576214405360134, "step": 1280}, {"loss": 1.7448, "grad_norm": 0.3631329834461212, "learning_rate": 0.0002, "epoch": 0.864321608040201, "step": 1290}, {"loss": 1.7, "grad_norm": 0.36706018447875977, "learning_rate": 0.0002, "epoch": 0.8710217755443886, "step": 1300}, {"loss": 1.8178, "grad_norm": 0.3347418010234833, "learning_rate": 0.0002, "epoch": 0.8777219430485762, "step": 1310}, {"loss": 1.7824, "grad_norm": 0.34371060132980347, "learning_rate": 0.0002, "epoch": 0.8844221105527639, "step": 1320}, {"loss": 1.783, "grad_norm": 0.3029090166091919, "learning_rate": 0.0002, "epoch": 0.8911222780569514, "step": 1330}, {"loss": 1.8017, "grad_norm": 0.34700682759284973, "learning_rate": 0.0002, "epoch": 0.897822445561139, "step": 1340}, {"loss": 1.7998, "grad_norm": 0.35574328899383545, "learning_rate": 0.0002, "epoch": 0.9045226130653267, "step": 1350}, {"loss": 1.7339, "grad_norm": 0.30928221344947815, "learning_rate": 0.0002, "epoch": 0.9112227805695142, "step": 1360}, {"loss": 1.7479, "grad_norm": 0.30652928352355957, "learning_rate": 0.0002, "epoch": 0.9179229480737019, "step": 1370}, {"loss": 1.7491, "grad_norm": 0.3838157653808594, "learning_rate": 0.0002, "epoch": 0.9246231155778895, "step": 1380}, {"loss": 1.7977, "grad_norm": 0.31655240058898926, "learning_rate": 0.0002, "epoch": 0.931323283082077, "step": 1390}, {"loss": 1.8175, "grad_norm": 0.41737303137779236, "learning_rate": 0.0002, "epoch": 0.9380234505862647, "step": 1400}, {"loss": 1.6811, "grad_norm": 0.3227267861366272, "learning_rate": 0.0002, "epoch": 0.9447236180904522, "step": 1410}, {"loss": 1.7343, "grad_norm": 0.3729925751686096, "learning_rate": 0.0002, "epoch": 0.9514237855946399, "step": 1420}, {"loss": 1.8221, "grad_norm": 0.30779409408569336, "learning_rate": 0.0002, "epoch": 0.9581239530988275, "step": 1430}, {"loss": 1.7972, "grad_norm": 0.334379643201828, "learning_rate": 0.0002, "epoch": 0.964824120603015, "step": 1440}, {"loss": 1.7141, "grad_norm": 0.3568236231803894, "learning_rate": 0.0002, "epoch": 0.9715242881072027, "step": 1450}, {"loss": 1.7541, "grad_norm": 0.33310577273368835, "learning_rate": 0.0002, "epoch": 0.9782244556113903, "step": 1460}, {"loss": 1.8511, "grad_norm": 0.2972261905670166, "learning_rate": 0.0002, "epoch": 0.9849246231155779, "step": 1470}, {"loss": 1.7654, "grad_norm": 0.3322717845439911, "learning_rate": 0.0002, "epoch": 0.9916247906197655, "step": 1480}, {"loss": 1.8033, "grad_norm": 0.3276330828666687, "learning_rate": 0.0002, "epoch": 0.998324958123953, "step": 1490}, {"eval_loss": 1.8036354780197144, "eval_runtime": 37.8949, "eval_samples_per_second": 13.59, "eval_steps_per_second": 1.715, "epoch": 0.9996649916247906, "step": 1492}, {"loss": 1.7138, "grad_norm": 0.29252371191978455, "learning_rate": 0.0002, "epoch": 1.0050251256281406, "step": 1500}, {"loss": 1.8198, "grad_norm": 0.31607162952423096, "learning_rate": 0.0002, "epoch": 1.0117252931323284, "step": 1510}, {"loss": 1.6779, "grad_norm": 0.32294467091560364, "learning_rate": 0.0002, "epoch": 1.018425460636516, "step": 1520}, {"loss": 1.7919, "grad_norm": 0.3868017792701721, "learning_rate": 0.0002, "epoch": 1.0251256281407035, "step": 1530}, {"loss": 1.7954, "grad_norm": 0.3178282082080841, "learning_rate": 0.0002, "epoch": 1.031825795644891, "step": 1540}, {"loss": 1.7136, "grad_norm": 0.3706750273704529, "learning_rate": 0.0002, "epoch": 1.0385259631490786, "step": 1550}, {"loss": 1.7382, "grad_norm": 0.33930912613868713, "learning_rate": 0.0002, "epoch": 1.0452261306532664, "step": 1560}, {"loss": 1.7602, "grad_norm": 0.33970504999160767, "learning_rate": 0.0002, "epoch": 1.051926298157454, "step": 1570}, {"loss": 1.6573, "grad_norm": 0.42553383111953735, "learning_rate": 0.0002, "epoch": 1.0586264656616415, "step": 1580}, {"loss": 1.645, "grad_norm": 0.3772421181201935, "learning_rate": 0.0002, "epoch": 1.065326633165829, "step": 1590}, {"loss": 1.7362, "grad_norm": 0.34212902188301086, "learning_rate": 0.0002, "epoch": 1.0720268006700167, "step": 1600}, {"loss": 1.7057, "grad_norm": 0.3798283338546753, "learning_rate": 0.0002, "epoch": 1.0787269681742044, "step": 1610}, {"loss": 1.7468, "grad_norm": 0.36909598112106323, "learning_rate": 0.0002, "epoch": 1.085427135678392, "step": 1620}, {"loss": 1.7807, "grad_norm": 0.3344230651855469, "learning_rate": 0.0002, "epoch": 1.0921273031825796, "step": 1630}, {"loss": 1.7111, "grad_norm": 0.3862569332122803, "learning_rate": 0.0002, "epoch": 1.0988274706867671, "step": 1640}, {"loss": 1.7163, "grad_norm": 0.31188511848449707, "learning_rate": 0.0002, "epoch": 1.1055276381909547, "step": 1650}, {"loss": 1.7263, "grad_norm": 0.3563670814037323, "learning_rate": 0.0002, "epoch": 1.1122278056951425, "step": 1660}, {"loss": 1.7718, "grad_norm": 0.35052165389060974, "learning_rate": 0.0002, "epoch": 1.11892797319933, "step": 1670}, {"loss": 1.7601, "grad_norm": 0.3285699188709259, "learning_rate": 0.0002, "epoch": 1.1256281407035176, "step": 1680}, {"loss": 1.6877, "grad_norm": 0.3639393746852875, "learning_rate": 0.0002, "epoch": 1.1323283082077051, "step": 1690}, {"loss": 1.7719, "grad_norm": 0.3842753767967224, "learning_rate": 0.0002, "epoch": 1.1390284757118927, "step": 1700}, {"loss": 1.7002, "grad_norm": 0.3624933063983917, "learning_rate": 0.0002, "epoch": 1.1457286432160805, "step": 1710}, {"loss": 1.7243, "grad_norm": 0.3641220033168793, "learning_rate": 0.0002, "epoch": 1.152428810720268, "step": 1720}, {"loss": 1.752, "grad_norm": 0.32765355706214905, "learning_rate": 0.0002, "epoch": 1.1591289782244556, "step": 1730}, {"loss": 1.6556, "grad_norm": 0.34974896907806396, "learning_rate": 0.0002, "epoch": 1.1658291457286432, "step": 1740}, {"loss": 1.7273, "grad_norm": 0.3910926580429077, "learning_rate": 0.0002, "epoch": 1.1725293132328307, "step": 1750}, {"loss": 1.7204, "grad_norm": 0.3564300537109375, "learning_rate": 0.0002, "epoch": 1.1792294807370185, "step": 1760}, {"loss": 1.746, "grad_norm": 0.34822574257850647, "learning_rate": 0.0002, "epoch": 1.185929648241206, "step": 1770}, {"loss": 1.7256, "grad_norm": 0.36185044050216675, "learning_rate": 0.0002, "epoch": 1.1926298157453936, "step": 1780}, {"loss": 1.6431, "grad_norm": 0.34866711497306824, "learning_rate": 0.0002, "epoch": 1.1993299832495812, "step": 1790}, {"loss": 1.8084, "grad_norm": 0.4017769992351532, "learning_rate": 0.0002, "epoch": 1.2060301507537687, "step": 1800}, {"loss": 1.6985, "grad_norm": 0.32930681109428406, "learning_rate": 0.0002, "epoch": 1.2127303182579565, "step": 1810}, {"loss": 1.7606, "grad_norm": 0.35951921343803406, "learning_rate": 0.0002, "epoch": 1.219430485762144, "step": 1820}, {"loss": 1.6933, "grad_norm": 0.37366992235183716, "learning_rate": 0.0002, "epoch": 1.2261306532663316, "step": 1830}, {"loss": 1.6737, "grad_norm": 0.3565689027309418, "learning_rate": 0.0002, "epoch": 1.2328308207705192, "step": 1840}, {"loss": 1.8013, "grad_norm": 0.3692343533039093, "learning_rate": 0.0002, "epoch": 1.2395309882747068, "step": 1850}, {"loss": 1.736, "grad_norm": 0.38426971435546875, "learning_rate": 0.0002, "epoch": 1.2462311557788945, "step": 1860}, {"loss": 1.7031, "grad_norm": 0.33559855818748474, "learning_rate": 0.0002, "epoch": 1.252931323283082, "step": 1870}, {"loss": 1.7033, "grad_norm": 0.34181106090545654, "learning_rate": 0.0002, "epoch": 1.2596314907872697, "step": 1880}, {"loss": 1.7707, "grad_norm": 0.3916318416595459, "learning_rate": 0.0002, "epoch": 1.2663316582914572, "step": 1890}, {"loss": 1.6686, "grad_norm": 0.3887825012207031, "learning_rate": 0.0002, "epoch": 1.2730318257956448, "step": 1900}, {"loss": 1.7062, "grad_norm": 0.33583927154541016, "learning_rate": 0.0002, "epoch": 1.2797319932998326, "step": 1910}, {"loss": 1.717, "grad_norm": 0.37639349699020386, "learning_rate": 0.0002, "epoch": 1.2864321608040201, "step": 1920}, {"loss": 1.777, "grad_norm": 0.38059428334236145, "learning_rate": 0.0002, "epoch": 1.2931323283082077, "step": 1930}, {"loss": 1.6126, "grad_norm": 0.37253183126449585, "learning_rate": 0.0002, "epoch": 1.2998324958123952, "step": 1940}, {"loss": 1.6758, "grad_norm": 0.37371566891670227, "learning_rate": 0.0002, "epoch": 1.3065326633165828, "step": 1950}, {"loss": 1.6788, "grad_norm": 0.4080910086631775, "learning_rate": 0.0002, "epoch": 1.3132328308207706, "step": 1960}, {"loss": 1.6518, "grad_norm": 0.3174354135990143, "learning_rate": 0.0002, "epoch": 1.3199329983249581, "step": 1970}, {"loss": 1.7925, "grad_norm": 0.4518888294696808, "learning_rate": 0.0002, "epoch": 1.3266331658291457, "step": 1980}, {"loss": 1.7085, "grad_norm": 0.3627921938896179, "learning_rate": 0.0002, "epoch": 1.3333333333333333, "step": 1990}, {"loss": 1.7676, "grad_norm": 0.3655930161476135, "learning_rate": 0.0002, "epoch": 1.3400335008375208, "step": 2000}, {"loss": 1.7016, "grad_norm": 0.3509993255138397, "learning_rate": 0.0002, "epoch": 1.3467336683417086, "step": 2010}, {"loss": 1.7359, "grad_norm": 0.4281129240989685, "learning_rate": 0.0002, "epoch": 1.3534338358458962, "step": 2020}, {"loss": 1.6884, "grad_norm": 0.3821414113044739, "learning_rate": 0.0002, "epoch": 1.3601340033500837, "step": 2030}, {"loss": 1.7075, "grad_norm": 0.3907586336135864, "learning_rate": 0.0002, "epoch": 1.3668341708542713, "step": 2040}, {"loss": 1.7424, "grad_norm": 0.37792932987213135, "learning_rate": 0.0002, "epoch": 1.3735343383584588, "step": 2050}, {"loss": 1.7305, "grad_norm": 0.3693985641002655, "learning_rate": 0.0002, "epoch": 1.3802345058626466, "step": 2060}, {"loss": 1.7434, "grad_norm": 0.32275936007499695, "learning_rate": 0.0002, "epoch": 1.3869346733668342, "step": 2070}, {"loss": 1.6677, "grad_norm": 0.3789440095424652, "learning_rate": 0.0002, "epoch": 1.3936348408710217, "step": 2080}, {"loss": 1.6825, "grad_norm": 0.3638380467891693, "learning_rate": 0.0002, "epoch": 1.4003350083752093, "step": 2090}, {"loss": 1.6542, "grad_norm": 0.3495481610298157, "learning_rate": 0.0002, "epoch": 1.4070351758793969, "step": 2100}, {"loss": 1.7225, "grad_norm": 0.37920597195625305, "learning_rate": 0.0002, "epoch": 1.4137353433835846, "step": 2110}, {"loss": 1.7329, "grad_norm": 0.37218064069747925, "learning_rate": 0.0002, "epoch": 1.4204355108877722, "step": 2120}, {"loss": 1.799, "grad_norm": 0.38074082136154175, "learning_rate": 0.0002, "epoch": 1.4271356783919598, "step": 2130}, {"loss": 1.7403, "grad_norm": 0.3455527126789093, "learning_rate": 0.0002, "epoch": 1.4338358458961473, "step": 2140}, {"loss": 1.776, "grad_norm": 0.3712003529071808, "learning_rate": 0.0002, "epoch": 1.4405360134003349, "step": 2150}, {"loss": 1.7619, "grad_norm": 0.3786754906177521, "learning_rate": 0.0002, "epoch": 1.4472361809045227, "step": 2160}, {"loss": 1.68, "grad_norm": 0.3879223167896271, "learning_rate": 0.0002, "epoch": 1.4539363484087102, "step": 2170}, {"loss": 1.7, "grad_norm": 0.38738805055618286, "learning_rate": 0.0002, "epoch": 1.4606365159128978, "step": 2180}, {"loss": 1.7581, "grad_norm": 0.39768800139427185, "learning_rate": 0.0002, "epoch": 1.4673366834170856, "step": 2190}, {"loss": 1.7671, "grad_norm": 0.4172441065311432, "learning_rate": 0.0002, "epoch": 1.474036850921273, "step": 2200}, {"loss": 1.6736, "grad_norm": 0.4043174982070923, "learning_rate": 0.0002, "epoch": 1.4807370184254607, "step": 2210}, {"loss": 1.7444, "grad_norm": 0.3750883936882019, "learning_rate": 0.0002, "epoch": 1.4874371859296482, "step": 2220}, {"loss": 1.6861, "grad_norm": 0.3552253246307373, "learning_rate": 0.0002, "epoch": 1.4941373534338358, "step": 2230}, {"loss": 1.6471, "grad_norm": 0.34607139229774475, "learning_rate": 0.0002, "epoch": 1.5008375209380236, "step": 2240}, {"loss": 1.6962, "grad_norm": 0.3406706750392914, "learning_rate": 0.0002, "epoch": 1.507537688442211, "step": 2250}, {"loss": 1.7694, "grad_norm": 0.36654895544052124, "learning_rate": 0.0002, "epoch": 1.5142378559463987, "step": 2260}, {"loss": 1.6812, "grad_norm": 0.3914054334163666, "learning_rate": 0.0002, "epoch": 1.5209380234505863, "step": 2270}, {"loss": 1.6822, "grad_norm": 0.42012137174606323, "learning_rate": 0.0002, "epoch": 1.5276381909547738, "step": 2280}, {"loss": 1.697, "grad_norm": 0.39563435316085815, "learning_rate": 0.0002, "epoch": 1.5343383584589616, "step": 2290}, {"loss": 1.7491, "grad_norm": 0.3508438766002655, "learning_rate": 0.0002, "epoch": 1.541038525963149, "step": 2300}, {"loss": 1.7727, "grad_norm": 0.3785218596458435, "learning_rate": 0.0002, "epoch": 1.5477386934673367, "step": 2310}, {"loss": 1.6963, "grad_norm": 0.39377647638320923, "learning_rate": 0.0002, "epoch": 1.5544388609715243, "step": 2320}, {"loss": 1.7263, "grad_norm": 0.3391438126564026, "learning_rate": 0.0002, "epoch": 1.5611390284757118, "step": 2330}, {"loss": 1.7722, "grad_norm": 0.37944263219833374, "learning_rate": 0.0002, "epoch": 1.5678391959798996, "step": 2340}, {"loss": 1.6371, "grad_norm": 0.3523491322994232, "learning_rate": 0.0002, "epoch": 1.574539363484087, "step": 2350}, {"loss": 1.7583, "grad_norm": 0.3911575973033905, "learning_rate": 0.0002, "epoch": 1.5812395309882747, "step": 2360}, {"loss": 1.7117, "grad_norm": 0.33832186460494995, "learning_rate": 0.0002, "epoch": 1.5879396984924623, "step": 2370}, {"loss": 1.7701, "grad_norm": 0.3665979206562042, "learning_rate": 0.0002, "epoch": 1.5946398659966499, "step": 2380}, {"loss": 1.779, "grad_norm": 0.3871748149394989, "learning_rate": 0.0002, "epoch": 1.6013400335008376, "step": 2390}, {"loss": 1.7109, "grad_norm": 0.3586967885494232, "learning_rate": 0.0002, "epoch": 1.608040201005025, "step": 2400}, {"loss": 1.7096, "grad_norm": 0.3563673198223114, "learning_rate": 0.0002, "epoch": 1.6147403685092128, "step": 2410}, {"loss": 1.745, "grad_norm": 0.37588971853256226, "learning_rate": 0.0002, "epoch": 1.6214405360134003, "step": 2420}, {"loss": 1.7086, "grad_norm": 0.352556437253952, "learning_rate": 0.0002, "epoch": 1.6281407035175879, "step": 2430}, {"loss": 1.6547, "grad_norm": 0.3716259300708771, "learning_rate": 0.0002, "epoch": 1.6348408710217757, "step": 2440}, {"loss": 1.7033, "grad_norm": 0.372001975774765, "learning_rate": 0.0002, "epoch": 1.641541038525963, "step": 2450}, {"loss": 1.6584, "grad_norm": 0.3430042862892151, "learning_rate": 0.0002, "epoch": 1.6482412060301508, "step": 2460}, {"loss": 1.7217, "grad_norm": 0.3741483688354492, "learning_rate": 0.0002, "epoch": 1.6549413735343383, "step": 2470}, {"loss": 1.7701, "grad_norm": 0.3610571324825287, "learning_rate": 0.0002, "epoch": 1.661641541038526, "step": 2480}, {"loss": 1.7057, "grad_norm": 0.4204719066619873, "learning_rate": 0.0002, "epoch": 1.6683417085427137, "step": 2490}, {"loss": 1.7954, "grad_norm": 0.3938186466693878, "learning_rate": 0.0002, "epoch": 1.675041876046901, "step": 2500}, {"loss": 1.6633, "grad_norm": 0.3421435058116913, "learning_rate": 0.0002, "epoch": 1.6817420435510888, "step": 2510}, {"loss": 1.7996, "grad_norm": 0.42441412806510925, "learning_rate": 0.0002, "epoch": 1.6884422110552764, "step": 2520}, {"loss": 1.7142, "grad_norm": 0.38071519136428833, "learning_rate": 0.0002, "epoch": 1.695142378559464, "step": 2530}, {"loss": 1.7232, "grad_norm": 0.34078919887542725, "learning_rate": 0.0002, "epoch": 1.7018425460636517, "step": 2540}, {"loss": 1.7126, "grad_norm": 0.412844181060791, "learning_rate": 0.0002, "epoch": 1.708542713567839, "step": 2550}, {"loss": 1.7149, "grad_norm": 0.3753604292869568, "learning_rate": 0.0002, "epoch": 1.7152428810720268, "step": 2560}, {"loss": 1.7011, "grad_norm": 0.41588476300239563, "learning_rate": 0.0002, "epoch": 1.7219430485762144, "step": 2570}, {"loss": 1.6427, "grad_norm": 0.35504111647605896, "learning_rate": 0.0002, "epoch": 1.728643216080402, "step": 2580}, {"loss": 1.7296, "grad_norm": 0.36909720301628113, "learning_rate": 0.0002, "epoch": 1.7353433835845897, "step": 2590}, {"loss": 1.7022, "grad_norm": 0.4149979054927826, "learning_rate": 0.0002, "epoch": 1.742043551088777, "step": 2600}, {"loss": 1.77, "grad_norm": 0.38859328627586365, "learning_rate": 0.0002, "epoch": 1.7487437185929648, "step": 2610}, {"loss": 1.7036, "grad_norm": 0.36738792061805725, "learning_rate": 0.0002, "epoch": 1.7554438860971524, "step": 2620}, {"loss": 1.764, "grad_norm": 0.3968178927898407, "learning_rate": 0.0002, "epoch": 1.76214405360134, "step": 2630}, {"loss": 1.7687, "grad_norm": 0.3972901999950409, "learning_rate": 0.0002, "epoch": 1.7688442211055277, "step": 2640}, {"loss": 1.6774, "grad_norm": 0.3949959874153137, "learning_rate": 0.0002, "epoch": 1.775544388609715, "step": 2650}, {"loss": 1.7247, "grad_norm": 0.44074657559394836, "learning_rate": 0.0002, "epoch": 1.7822445561139029, "step": 2660}, {"loss": 1.7188, "grad_norm": 0.39743664860725403, "learning_rate": 0.0002, "epoch": 1.7889447236180904, "step": 2670}, {"loss": 1.7258, "grad_norm": 0.3950406610965729, "learning_rate": 0.0002, "epoch": 1.795644891122278, "step": 2680}, {"loss": 1.6906, "grad_norm": 0.3568263649940491, "learning_rate": 0.0002, "epoch": 1.8023450586264658, "step": 2690}, {"loss": 1.6735, "grad_norm": 0.3819476366043091, "learning_rate": 0.0002, "epoch": 1.809045226130653, "step": 2700}, {"loss": 1.7198, "grad_norm": 0.3480634391307831, "learning_rate": 0.0002, "epoch": 1.8157453936348409, "step": 2710}, {"loss": 1.7042, "grad_norm": 0.3875853419303894, "learning_rate": 0.0002, "epoch": 1.8224455611390284, "step": 2720}, {"loss": 1.6988, "grad_norm": 0.3441337049007416, "learning_rate": 0.0002, "epoch": 1.829145728643216, "step": 2730}, {"loss": 1.7647, "grad_norm": 0.35692882537841797, "learning_rate": 0.0002, "epoch": 1.8358458961474038, "step": 2740}, {"loss": 1.7033, "grad_norm": 0.36959215998649597, "learning_rate": 0.0002, "epoch": 1.8425460636515911, "step": 2750}, {"loss": 1.7657, "grad_norm": 0.3893393278121948, "learning_rate": 0.0002, "epoch": 1.849246231155779, "step": 2760}, {"loss": 1.7068, "grad_norm": 0.37817293405532837, "learning_rate": 0.0002, "epoch": 1.8559463986599665, "step": 2770}, {"loss": 1.761, "grad_norm": 0.36071285605430603, "learning_rate": 0.0002, "epoch": 1.862646566164154, "step": 2780}, {"loss": 1.7623, "grad_norm": 0.3758420944213867, "learning_rate": 0.0002, "epoch": 1.8693467336683418, "step": 2790}, {"loss": 1.6743, "grad_norm": 0.3889938294887543, "learning_rate": 0.0002, "epoch": 1.8760469011725294, "step": 2800}, {"loss": 1.6151, "grad_norm": 0.34361857175827026, "learning_rate": 0.0002, "epoch": 1.882747068676717, "step": 2810}, {"loss": 1.6038, "grad_norm": 0.39283323287963867, "learning_rate": 0.0002, "epoch": 1.8894472361809045, "step": 2820}, {"loss": 1.7555, "grad_norm": 0.3919452726840973, "learning_rate": 0.0002, "epoch": 1.896147403685092, "step": 2830}, {"loss": 1.673, "grad_norm": 0.38215070962905884, "learning_rate": 0.0002, "epoch": 1.9028475711892798, "step": 2840}, {"loss": 1.7044, "grad_norm": 0.4235064387321472, "learning_rate": 0.0002, "epoch": 1.9095477386934674, "step": 2850}, {"loss": 1.7123, "grad_norm": 0.35694634914398193, "learning_rate": 0.0002, "epoch": 1.916247906197655, "step": 2860}, {"loss": 1.8128, "grad_norm": 0.383492112159729, "learning_rate": 0.0002, "epoch": 1.9229480737018425, "step": 2870}, {"loss": 1.7581, "grad_norm": 0.5945147275924683, "learning_rate": 0.0002, "epoch": 1.92964824120603, "step": 2880}, {"loss": 1.7421, "grad_norm": 0.3367522358894348, "learning_rate": 0.0002, "epoch": 1.9363484087102178, "step": 2890}, {"loss": 1.6561, "grad_norm": 0.35300394892692566, "learning_rate": 0.0002, "epoch": 1.9430485762144054, "step": 2900}, {"loss": 1.7033, "grad_norm": 0.38084495067596436, "learning_rate": 0.0002, "epoch": 1.949748743718593, "step": 2910}, {"loss": 1.7132, "grad_norm": 0.37559160590171814, "learning_rate": 0.0002, "epoch": 1.9564489112227805, "step": 2920}, {"loss": 1.6759, "grad_norm": 0.3661738336086273, "learning_rate": 0.0002, "epoch": 1.963149078726968, "step": 2930}, {"loss": 1.7643, "grad_norm": 0.4073849320411682, "learning_rate": 0.0002, "epoch": 1.9698492462311559, "step": 2940}, {"loss": 1.6806, "grad_norm": 0.3723304271697998, "learning_rate": 0.0002, "epoch": 1.9765494137353434, "step": 2950}, {"loss": 1.7611, "grad_norm": 0.3991098999977112, "learning_rate": 0.0002, "epoch": 1.983249581239531, "step": 2960}, {"loss": 1.7263, "grad_norm": 0.3947085440158844, "learning_rate": 0.0002, "epoch": 1.9899497487437185, "step": 2970}, {"loss": 1.7217, "grad_norm": 0.3786258399486542, "learning_rate": 0.0002, "epoch": 1.996649916247906, "step": 2980}, {"eval_loss": 1.8028968572616577, "eval_runtime": 37.8985, "eval_samples_per_second": 13.589, "eval_steps_per_second": 1.715, "epoch": 2.0, "step": 2985}, {"loss": 1.695, "grad_norm": 0.34824079275131226, "learning_rate": 0.0002, "epoch": 2.003350083752094, "step": 2990}, {"loss": 1.5853, "grad_norm": 0.3394894003868103, "learning_rate": 0.0002, "epoch": 2.0100502512562812, "step": 3000}, {"loss": 1.5783, "grad_norm": 0.36910977959632874, "learning_rate": 0.0002, "epoch": 2.016750418760469, "step": 3010}, {"loss": 1.6105, "grad_norm": 0.45000967383384705, "learning_rate": 0.0002, "epoch": 2.023450586264657, "step": 3020}, {"loss": 1.6019, "grad_norm": 0.3791407346725464, "learning_rate": 0.0002, "epoch": 2.030150753768844, "step": 3030}, {"loss": 1.5832, "grad_norm": 0.387321799993515, "learning_rate": 0.0002, "epoch": 2.036850921273032, "step": 3040}, {"loss": 1.6834, "grad_norm": 0.4185757040977478, "learning_rate": 0.0002, "epoch": 2.0435510887772192, "step": 3050}, {"loss": 1.5696, "grad_norm": 0.45110777020454407, "learning_rate": 0.0002, "epoch": 2.050251256281407, "step": 3060}, {"loss": 1.6231, "grad_norm": 0.42663660645484924, "learning_rate": 0.0002, "epoch": 2.056951423785595, "step": 3070}, {"loss": 1.6279, "grad_norm": 0.4546292722225189, "learning_rate": 0.0002, "epoch": 2.063651591289782, "step": 3080}, {"loss": 1.6141, "grad_norm": 0.3979759216308594, "learning_rate": 0.0002, "epoch": 2.07035175879397, "step": 3090}, {"loss": 1.6343, "grad_norm": 0.43596673011779785, "learning_rate": 0.0002, "epoch": 2.0770519262981573, "step": 3100}, {"loss": 1.5441, "grad_norm": 0.40120232105255127, "learning_rate": 0.0002, "epoch": 2.083752093802345, "step": 3110}, {"loss": 1.6309, "grad_norm": 0.44449281692504883, "learning_rate": 0.0002, "epoch": 2.090452261306533, "step": 3120}, {"loss": 1.5652, "grad_norm": 0.42672568559646606, "learning_rate": 0.0002, "epoch": 2.09715242881072, "step": 3130}, {"loss": 1.682, "grad_norm": 0.4232690930366516, "learning_rate": 0.0002, "epoch": 2.103852596314908, "step": 3140}, {"loss": 1.624, "grad_norm": 0.4299317002296448, "learning_rate": 0.0002, "epoch": 2.1105527638190953, "step": 3150}, {"loss": 1.6766, "grad_norm": 0.4067758023738861, "learning_rate": 0.0002, "epoch": 2.117252931323283, "step": 3160}, {"loss": 1.6759, "grad_norm": 0.4918815791606903, "learning_rate": 0.0002, "epoch": 2.123953098827471, "step": 3170}, {"loss": 1.6478, "grad_norm": 0.4140559732913971, "learning_rate": 0.0002, "epoch": 2.130653266331658, "step": 3180}, {"loss": 1.6641, "grad_norm": 0.4555995464324951, "learning_rate": 0.0002, "epoch": 2.137353433835846, "step": 3190}, {"loss": 1.5888, "grad_norm": 0.42943915724754333, "learning_rate": 0.0002, "epoch": 2.1440536013400333, "step": 3200}, {"loss": 1.5886, "grad_norm": 0.4730435013771057, "learning_rate": 0.0002, "epoch": 2.150753768844221, "step": 3210}, {"loss": 1.6022, "grad_norm": 0.43310216069221497, "learning_rate": 0.0002, "epoch": 2.157453936348409, "step": 3220}, {"loss": 1.6058, "grad_norm": 0.42054110765457153, "learning_rate": 0.0002, "epoch": 2.164154103852596, "step": 3230}, {"loss": 1.6749, "grad_norm": 0.4897233247756958, "learning_rate": 0.0002, "epoch": 2.170854271356784, "step": 3240}, {"loss": 1.6983, "grad_norm": 0.42194533348083496, "learning_rate": 0.0002, "epoch": 2.1775544388609713, "step": 3250}, {"loss": 1.6159, "grad_norm": 0.44494450092315674, "learning_rate": 0.0002, "epoch": 2.184254606365159, "step": 3260}, {"loss": 1.6977, "grad_norm": 0.43524879217147827, "learning_rate": 0.0002, "epoch": 2.190954773869347, "step": 3270}, {"loss": 1.528, "grad_norm": 0.4621117413043976, "learning_rate": 0.0002, "epoch": 2.1976549413735342, "step": 3280}, {"loss": 1.632, "grad_norm": 0.4073285460472107, "learning_rate": 0.0002, "epoch": 2.204355108877722, "step": 3290}, {"loss": 1.6141, "grad_norm": 0.47868335247039795, "learning_rate": 0.0002, "epoch": 2.2110552763819094, "step": 3300}, {"loss": 1.6857, "grad_norm": 0.4264970123767853, "learning_rate": 0.0002, "epoch": 2.217755443886097, "step": 3310}, {"loss": 1.5653, "grad_norm": 0.4491245150566101, "learning_rate": 0.0002, "epoch": 2.224455611390285, "step": 3320}, {"loss": 1.5881, "grad_norm": 0.4010344445705414, "learning_rate": 0.0002, "epoch": 2.2311557788944723, "step": 3330}, {"loss": 1.6684, "grad_norm": 0.4232759177684784, "learning_rate": 0.0002, "epoch": 2.23785594639866, "step": 3340}, {"loss": 1.6336, "grad_norm": 0.5099776983261108, "learning_rate": 0.0002, "epoch": 2.2445561139028474, "step": 3350}, {"loss": 1.6764, "grad_norm": 0.5223407745361328, "learning_rate": 0.0002, "epoch": 2.251256281407035, "step": 3360}, {"loss": 1.6625, "grad_norm": 0.47818470001220703, "learning_rate": 0.0002, "epoch": 2.257956448911223, "step": 3370}, {"loss": 1.5946, "grad_norm": 0.4721255898475647, "learning_rate": 0.0002, "epoch": 2.2646566164154103, "step": 3380}, {"loss": 1.5568, "grad_norm": 0.4113229513168335, "learning_rate": 0.0002, "epoch": 2.271356783919598, "step": 3390}, {"loss": 1.6494, "grad_norm": 0.507080078125, "learning_rate": 0.0002, "epoch": 2.2780569514237854, "step": 3400}, {"loss": 1.6183, "grad_norm": 0.4852292239665985, "learning_rate": 0.0002, "epoch": 2.284757118927973, "step": 3410}, {"loss": 1.6132, "grad_norm": 0.4503684341907501, "learning_rate": 0.0002, "epoch": 2.291457286432161, "step": 3420}, {"loss": 1.6649, "grad_norm": 0.8359600305557251, "learning_rate": 0.0002, "epoch": 2.2981574539363483, "step": 3430}, {"loss": 1.6644, "grad_norm": 0.44604045152664185, "learning_rate": 0.0002, "epoch": 2.304857621440536, "step": 3440}, {"loss": 1.5972, "grad_norm": 0.45667049288749695, "learning_rate": 0.0002, "epoch": 2.3115577889447234, "step": 3450}, {"loss": 1.6667, "grad_norm": 0.4879349172115326, "learning_rate": 0.0002, "epoch": 2.318257956448911, "step": 3460}, {"loss": 1.5804, "grad_norm": 0.4033963084220886, "learning_rate": 0.0002, "epoch": 2.324958123953099, "step": 3470}, {"loss": 1.5838, "grad_norm": 0.44494301080703735, "learning_rate": 0.0002, "epoch": 2.3316582914572863, "step": 3480}, {"loss": 1.5951, "grad_norm": 0.4794621765613556, "learning_rate": 0.0002, "epoch": 2.338358458961474, "step": 3490}, {"loss": 1.6807, "grad_norm": 0.41404327750205994, "learning_rate": 0.0002, "epoch": 2.3450586264656614, "step": 3500}, {"loss": 1.714, "grad_norm": 0.4664851725101471, "learning_rate": 0.0002, "epoch": 2.351758793969849, "step": 3510}, {"loss": 1.6537, "grad_norm": 0.4263697564601898, "learning_rate": 0.0002, "epoch": 2.358458961474037, "step": 3520}, {"loss": 1.6551, "grad_norm": 0.5035167336463928, "learning_rate": 0.0002, "epoch": 2.3651591289782243, "step": 3530}, {"loss": 1.6208, "grad_norm": 0.4380664527416229, "learning_rate": 0.0002, "epoch": 2.371859296482412, "step": 3540}, {"loss": 1.634, "grad_norm": 0.5227681994438171, "learning_rate": 0.0002, "epoch": 2.3785594639865995, "step": 3550}, {"loss": 1.6146, "grad_norm": 0.4382302761077881, "learning_rate": 0.0002, "epoch": 2.3852596314907872, "step": 3560}, {"loss": 1.5653, "grad_norm": 0.4392451047897339, "learning_rate": 0.0002, "epoch": 2.391959798994975, "step": 3570}, {"loss": 1.6626, "grad_norm": 0.4372786581516266, "learning_rate": 0.0002, "epoch": 2.3986599664991624, "step": 3580}, {"loss": 1.519, "grad_norm": 0.5015502572059631, "learning_rate": 0.0002, "epoch": 2.40536013400335, "step": 3590}, {"loss": 1.588, "grad_norm": 0.5653210878372192, "learning_rate": 0.0002, "epoch": 2.4120603015075375, "step": 3600}, {"loss": 1.6075, "grad_norm": 0.53007972240448, "learning_rate": 0.0002, "epoch": 2.4187604690117253, "step": 3610}, {"loss": 1.6421, "grad_norm": 0.4659176766872406, "learning_rate": 0.0002, "epoch": 2.425460636515913, "step": 3620}, {"loss": 1.625, "grad_norm": 0.5637837052345276, "learning_rate": 0.0002, "epoch": 2.4321608040201004, "step": 3630}, {"loss": 1.6168, "grad_norm": 0.4248391389846802, "learning_rate": 0.0002, "epoch": 2.438860971524288, "step": 3640}, {"loss": 1.6822, "grad_norm": 0.44668248295783997, "learning_rate": 0.0002, "epoch": 2.4455611390284755, "step": 3650}, {"loss": 1.6417, "grad_norm": 0.43990179896354675, "learning_rate": 0.0002, "epoch": 2.4522613065326633, "step": 3660}, {"loss": 1.6723, "grad_norm": 0.4532523453235626, "learning_rate": 0.0002, "epoch": 2.458961474036851, "step": 3670}, {"loss": 1.6957, "grad_norm": 0.6605591773986816, "learning_rate": 0.0002, "epoch": 2.4656616415410384, "step": 3680}, {"loss": 1.6159, "grad_norm": 0.4694533348083496, "learning_rate": 0.0002, "epoch": 2.472361809045226, "step": 3690}, {"loss": 1.6239, "grad_norm": 0.4485011100769043, "learning_rate": 0.0002, "epoch": 2.4790619765494135, "step": 3700}, {"loss": 1.6834, "grad_norm": 0.4761785864830017, "learning_rate": 0.0002, "epoch": 2.4857621440536013, "step": 3710}, {"loss": 1.6313, "grad_norm": 0.5116432309150696, "learning_rate": 0.0002, "epoch": 2.492462311557789, "step": 3720}, {"loss": 1.5054, "grad_norm": 0.49523618817329407, "learning_rate": 0.0002, "epoch": 2.4991624790619764, "step": 3730}, {"loss": 1.6249, "grad_norm": 0.43826380372047424, "learning_rate": 0.0002, "epoch": 2.505862646566164, "step": 3740}, {"loss": 1.5762, "grad_norm": 0.4916154146194458, "learning_rate": 0.0002, "epoch": 2.5125628140703515, "step": 3750}, {"loss": 1.5157, "grad_norm": 0.5381299257278442, "learning_rate": 0.0002, "epoch": 2.5192629815745393, "step": 3760}, {"loss": 1.6467, "grad_norm": 0.44947415590286255, "learning_rate": 0.0002, "epoch": 2.525963149078727, "step": 3770}, {"loss": 1.67, "grad_norm": 0.49979084730148315, "learning_rate": 0.0002, "epoch": 2.5326633165829144, "step": 3780}, {"loss": 1.622, "grad_norm": 0.43046900629997253, "learning_rate": 0.0002, "epoch": 2.539363484087102, "step": 3790}, {"loss": 1.6789, "grad_norm": 0.4513470530509949, "learning_rate": 0.0002, "epoch": 2.5460636515912896, "step": 3800}, {"loss": 1.6335, "grad_norm": 0.49900051951408386, "learning_rate": 0.0002, "epoch": 2.5527638190954773, "step": 3810}, {"loss": 1.6322, "grad_norm": 0.4348420202732086, "learning_rate": 0.0002, "epoch": 2.559463986599665, "step": 3820}, {"loss": 1.6218, "grad_norm": 0.4684867560863495, "learning_rate": 0.0002, "epoch": 2.5661641541038525, "step": 3830}, {"loss": 1.6535, "grad_norm": 0.44430989027023315, "learning_rate": 0.0002, "epoch": 2.5728643216080402, "step": 3840}, {"loss": 1.5909, "grad_norm": 0.47375255823135376, "learning_rate": 0.0002, "epoch": 2.5795644891122276, "step": 3850}, {"loss": 1.6269, "grad_norm": 0.45493075251579285, "learning_rate": 0.0002, "epoch": 2.5862646566164154, "step": 3860}, {"loss": 1.604, "grad_norm": 0.4563275873661041, "learning_rate": 0.0002, "epoch": 2.592964824120603, "step": 3870}, {"loss": 1.642, "grad_norm": 0.46060335636138916, "learning_rate": 0.0002, "epoch": 2.5996649916247905, "step": 3880}, {"loss": 1.6302, "grad_norm": 0.4718867540359497, "learning_rate": 0.0002, "epoch": 2.6063651591289783, "step": 3890}, {"loss": 1.6242, "grad_norm": 0.41570305824279785, "learning_rate": 0.0002, "epoch": 2.6130653266331656, "step": 3900}, {"loss": 1.6401, "grad_norm": 0.4603121876716614, "learning_rate": 0.0002, "epoch": 2.6197654941373534, "step": 3910}, {"loss": 1.6839, "grad_norm": 0.4734652638435364, "learning_rate": 0.0002, "epoch": 2.626465661641541, "step": 3920}, {"loss": 1.5448, "grad_norm": 0.45348483324050903, "learning_rate": 0.0002, "epoch": 2.6331658291457285, "step": 3930}, {"loss": 1.6157, "grad_norm": 0.46559447050094604, "learning_rate": 0.0002, "epoch": 2.6398659966499163, "step": 3940}, {"loss": 1.7052, "grad_norm": 0.44113144278526306, "learning_rate": 0.0002, "epoch": 2.6465661641541036, "step": 3950}, {"loss": 1.6315, "grad_norm": 0.41415104269981384, "learning_rate": 0.0002, "epoch": 2.6532663316582914, "step": 3960}, {"loss": 1.6589, "grad_norm": 0.48868080973625183, "learning_rate": 0.0002, "epoch": 2.659966499162479, "step": 3970}, {"loss": 1.6211, "grad_norm": 0.49610549211502075, "learning_rate": 0.0002, "epoch": 2.6666666666666665, "step": 3980}, {"loss": 1.6235, "grad_norm": 0.4309130907058716, "learning_rate": 0.0002, "epoch": 2.6733668341708543, "step": 3990}, {"loss": 1.6452, "grad_norm": 0.4489327669143677, "learning_rate": 0.0002, "epoch": 2.6800670016750416, "step": 4000}, {"loss": 1.5744, "grad_norm": 0.5380139946937561, "learning_rate": 0.0002, "epoch": 2.6867671691792294, "step": 4010}, {"loss": 1.6524, "grad_norm": 0.5076672434806824, "learning_rate": 0.0002, "epoch": 2.693467336683417, "step": 4020}, {"loss": 1.636, "grad_norm": 0.47620031237602234, "learning_rate": 0.0002, "epoch": 2.7001675041876045, "step": 4030}, {"loss": 1.5543, "grad_norm": 0.48089155554771423, "learning_rate": 0.0002, "epoch": 2.7068676716917923, "step": 4040}, {"loss": 1.6396, "grad_norm": 0.5108814239501953, "learning_rate": 0.0002, "epoch": 2.7135678391959797, "step": 4050}, {"loss": 1.5905, "grad_norm": 0.4196513295173645, "learning_rate": 0.0002, "epoch": 2.7202680067001674, "step": 4060}, {"loss": 1.686, "grad_norm": 0.4574664831161499, "learning_rate": 0.0002, "epoch": 2.726968174204355, "step": 4070}, {"loss": 1.6234, "grad_norm": 0.4671640992164612, "learning_rate": 0.0002, "epoch": 2.7336683417085426, "step": 4080}, {"loss": 1.6827, "grad_norm": 0.49355530738830566, "learning_rate": 0.0002, "epoch": 2.7403685092127303, "step": 4090}, {"loss": 1.6999, "grad_norm": 0.46716663241386414, "learning_rate": 0.0002, "epoch": 2.7470686767169177, "step": 4100}, {"loss": 1.6463, "grad_norm": 0.45420581102371216, "learning_rate": 0.0002, "epoch": 2.7537688442211055, "step": 4110}, {"loss": 1.5718, "grad_norm": 0.4680487811565399, "learning_rate": 0.0002, "epoch": 2.7604690117252932, "step": 4120}, {"loss": 1.5968, "grad_norm": 0.5375032424926758, "learning_rate": 0.0002, "epoch": 2.7671691792294806, "step": 4130}, {"loss": 1.5254, "grad_norm": 0.46026280522346497, "learning_rate": 0.0002, "epoch": 2.7738693467336684, "step": 4140}, {"loss": 1.6613, "grad_norm": 0.43658447265625, "learning_rate": 0.0002, "epoch": 2.7805695142378557, "step": 4150}, {"loss": 1.6546, "grad_norm": 0.4935547113418579, "learning_rate": 0.0002, "epoch": 2.7872696817420435, "step": 4160}, {"loss": 1.5961, "grad_norm": 0.8167962431907654, "learning_rate": 0.0002, "epoch": 2.7939698492462313, "step": 4170}, {"loss": 1.6907, "grad_norm": 0.4289683997631073, "learning_rate": 0.0002, "epoch": 2.8006700167504186, "step": 4180}, {"loss": 1.6385, "grad_norm": 0.4569324254989624, "learning_rate": 0.0002, "epoch": 2.8073701842546064, "step": 4190}, {"loss": 1.6077, "grad_norm": 0.474795937538147, "learning_rate": 0.0002, "epoch": 2.8140703517587937, "step": 4200}, {"loss": 1.6223, "grad_norm": 0.44272229075431824, "learning_rate": 0.0002, "epoch": 2.8207705192629815, "step": 4210}, {"loss": 1.6706, "grad_norm": 0.525240957736969, "learning_rate": 0.0002, "epoch": 2.8274706867671693, "step": 4220}, {"loss": 1.7196, "grad_norm": 0.4802303910255432, "learning_rate": 0.0002, "epoch": 2.8341708542713566, "step": 4230}, {"loss": 1.6002, "grad_norm": 0.46400442719459534, "learning_rate": 0.0002, "epoch": 2.8408710217755444, "step": 4240}, {"loss": 1.6052, "grad_norm": 0.49884888529777527, "learning_rate": 0.0002, "epoch": 2.8475711892797317, "step": 4250}, {"loss": 1.6919, "grad_norm": 0.5015072226524353, "learning_rate": 0.0002, "epoch": 2.8542713567839195, "step": 4260}, {"loss": 1.6335, "grad_norm": 0.4335440695285797, "learning_rate": 0.0002, "epoch": 2.8609715242881073, "step": 4270}, {"loss": 1.5664, "grad_norm": 0.5131644606590271, "learning_rate": 0.0002, "epoch": 2.8676716917922946, "step": 4280}, {"loss": 1.6409, "grad_norm": 0.6977195739746094, "learning_rate": 0.0002, "epoch": 2.8743718592964824, "step": 4290}, {"loss": 1.7192, "grad_norm": 0.5133762955665588, "learning_rate": 0.0002, "epoch": 2.8810720268006698, "step": 4300}, {"loss": 1.6257, "grad_norm": 0.4737614393234253, "learning_rate": 0.0002, "epoch": 2.8877721943048575, "step": 4310}, {"loss": 1.6076, "grad_norm": 0.4580535590648651, "learning_rate": 0.0002, "epoch": 2.8944723618090453, "step": 4320}, {"loss": 1.6538, "grad_norm": 0.43863341212272644, "learning_rate": 0.0002, "epoch": 2.901172529313233, "step": 4330}, {"loss": 1.6091, "grad_norm": 0.4103737473487854, "learning_rate": 0.0002, "epoch": 2.9078726968174204, "step": 4340}, {"loss": 1.7106, "grad_norm": 0.438014417886734, "learning_rate": 0.0002, "epoch": 2.914572864321608, "step": 4350}, {"loss": 1.6025, "grad_norm": 0.5068213939666748, "learning_rate": 0.0002, "epoch": 2.9212730318257956, "step": 4360}, {"loss": 1.6426, "grad_norm": 0.45305484533309937, "learning_rate": 0.0002, "epoch": 2.9279731993299833, "step": 4370}, {"loss": 1.5726, "grad_norm": 0.4612090289592743, "learning_rate": 0.0002, "epoch": 2.934673366834171, "step": 4380}, {"loss": 1.6536, "grad_norm": 0.508736789226532, "learning_rate": 0.0002, "epoch": 2.9413735343383585, "step": 4390}, {"loss": 1.6132, "grad_norm": 0.4924427270889282, "learning_rate": 0.0002, "epoch": 2.948073701842546, "step": 4400}, {"loss": 1.7007, "grad_norm": 0.5707460641860962, "learning_rate": 0.0002, "epoch": 2.9547738693467336, "step": 4410}, {"loss": 1.6814, "grad_norm": 0.42270299792289734, "learning_rate": 0.0002, "epoch": 2.9614740368509214, "step": 4420}, {"loss": 1.6644, "grad_norm": 0.4429931044578552, "learning_rate": 0.0002, "epoch": 2.968174204355109, "step": 4430}, {"loss": 1.6251, "grad_norm": 0.49760574102401733, "learning_rate": 0.0002, "epoch": 2.9748743718592965, "step": 4440}, {"loss": 1.6169, "grad_norm": 0.4558229148387909, "learning_rate": 0.0002, "epoch": 2.981574539363484, "step": 4450}, {"loss": 1.6055, "grad_norm": 0.39848530292510986, "learning_rate": 0.0002, "epoch": 2.9882747068676716, "step": 4460}, {"loss": 1.6705, "grad_norm": 0.5224862098693848, "learning_rate": 0.0002, "epoch": 2.9949748743718594, "step": 4470}, {"eval_loss": 1.8228833675384521, "eval_runtime": 37.9049, "eval_samples_per_second": 13.587, "eval_steps_per_second": 1.715, "epoch": 2.9996649916247904, "step": 4477}, {"loss": 1.6637, "grad_norm": 0.41169142723083496, "learning_rate": 0.0002, "epoch": 3.0016750418760467, "step": 4480}, {"loss": 1.5974, "grad_norm": 0.4865207374095917, "learning_rate": 0.0002, "epoch": 3.0083752093802345, "step": 4490}, {"loss": 1.5297, "grad_norm": 0.5462028384208679, "learning_rate": 0.0002, "epoch": 3.0150753768844223, "step": 4500}, {"loss": 1.5251, "grad_norm": 0.6169732809066772, "learning_rate": 0.0002, "epoch": 3.0217755443886096, "step": 4510}, {"loss": 1.5559, "grad_norm": 0.5667954087257385, "learning_rate": 0.0002, "epoch": 3.0284757118927974, "step": 4520}, {"loss": 1.5037, "grad_norm": 0.5758325457572937, "learning_rate": 0.0002, "epoch": 3.0351758793969847, "step": 4530}, {"loss": 1.4873, "grad_norm": 0.5220064520835876, "learning_rate": 0.0002, "epoch": 3.0418760469011725, "step": 4540}, {"loss": 1.5126, "grad_norm": 0.5469558835029602, "learning_rate": 0.0002, "epoch": 3.0485762144053603, "step": 4550}, {"loss": 1.4275, "grad_norm": 0.5680848956108093, "learning_rate": 0.0002, "epoch": 3.0552763819095476, "step": 4560}, {"loss": 1.5187, "grad_norm": 0.5906574726104736, "learning_rate": 0.0002, "epoch": 3.0619765494137354, "step": 4570}, {"loss": 1.4551, "grad_norm": 0.4725631773471832, "learning_rate": 0.0002, "epoch": 3.0686767169179228, "step": 4580}, {"loss": 1.5083, "grad_norm": 0.5273477435112, "learning_rate": 0.0002, "epoch": 3.0753768844221105, "step": 4590}, {"loss": 1.5154, "grad_norm": 0.5861203074455261, "learning_rate": 0.0002, "epoch": 3.0820770519262983, "step": 4600}, {"loss": 1.4924, "grad_norm": 0.5343965291976929, "learning_rate": 0.0002, "epoch": 3.0887772194304857, "step": 4610}, {"loss": 1.5608, "grad_norm": 0.5348150730133057, "learning_rate": 0.0002, "epoch": 3.0954773869346734, "step": 4620}, {"loss": 1.5399, "grad_norm": 0.5971846580505371, "learning_rate": 0.0002, "epoch": 3.102177554438861, "step": 4630}, {"loss": 1.4662, "grad_norm": 0.5203177332878113, "learning_rate": 0.0002, "epoch": 3.1088777219430486, "step": 4640}, {"loss": 1.5805, "grad_norm": 0.55289226770401, "learning_rate": 0.0002, "epoch": 3.1155778894472363, "step": 4650}, {"loss": 1.4745, "grad_norm": 0.6878530979156494, "learning_rate": 0.0002, "epoch": 3.1222780569514237, "step": 4660}, {"loss": 1.5335, "grad_norm": 0.6173256635665894, "learning_rate": 0.0002, "epoch": 3.1289782244556115, "step": 4670}, {"loss": 1.51, "grad_norm": 0.536796510219574, "learning_rate": 0.0002, "epoch": 3.135678391959799, "step": 4680}, {"loss": 1.4713, "grad_norm": 0.58846116065979, "learning_rate": 0.0002, "epoch": 3.1423785594639866, "step": 4690}, {"loss": 1.5114, "grad_norm": 0.645889401435852, "learning_rate": 0.0002, "epoch": 3.1490787269681744, "step": 4700}, {"loss": 1.4705, "grad_norm": 0.6118691563606262, "learning_rate": 0.0002, "epoch": 3.1557788944723617, "step": 4710}, {"loss": 1.5533, "grad_norm": 0.5189669132232666, "learning_rate": 0.0002, "epoch": 3.1624790619765495, "step": 4720}, {"loss": 1.4769, "grad_norm": 0.5794713497161865, "learning_rate": 0.0002, "epoch": 3.169179229480737, "step": 4730}, {"loss": 1.4849, "grad_norm": 0.6579326391220093, "learning_rate": 0.0002, "epoch": 3.1758793969849246, "step": 4740}, {"loss": 1.545, "grad_norm": 0.5822742581367493, "learning_rate": 0.0002, "epoch": 3.1825795644891124, "step": 4750}, {"loss": 1.4358, "grad_norm": 0.5475956201553345, "learning_rate": 0.0002, "epoch": 3.1892797319932997, "step": 4760}, {"loss": 1.4723, "grad_norm": 0.6743834018707275, "learning_rate": 0.0002, "epoch": 3.1959798994974875, "step": 4770}, {"loss": 1.5161, "grad_norm": 0.6110585927963257, "learning_rate": 0.0002, "epoch": 3.202680067001675, "step": 4780}, {"loss": 1.5455, "grad_norm": 0.5426181554794312, "learning_rate": 0.0002, "epoch": 3.2093802345058626, "step": 4790}, {"loss": 1.5315, "grad_norm": 0.6077824234962463, "learning_rate": 0.0002, "epoch": 3.2160804020100504, "step": 4800}, {"loss": 1.5314, "grad_norm": 0.5785858631134033, "learning_rate": 0.0002, "epoch": 3.2227805695142377, "step": 4810}, {"loss": 1.4041, "grad_norm": 0.6425958275794983, "learning_rate": 0.0002, "epoch": 3.2294807370184255, "step": 4820}, {"loss": 1.4751, "grad_norm": 0.6607080698013306, "learning_rate": 0.0002, "epoch": 3.236180904522613, "step": 4830}, {"loss": 1.5267, "grad_norm": 0.5385788679122925, "learning_rate": 0.0002, "epoch": 3.2428810720268006, "step": 4840}, {"loss": 1.4673, "grad_norm": 0.5630403757095337, "learning_rate": 0.0002, "epoch": 3.2495812395309884, "step": 4850}, {"loss": 1.5257, "grad_norm": 0.6340779662132263, "learning_rate": 0.0002, "epoch": 3.2562814070351758, "step": 4860}, {"loss": 1.5148, "grad_norm": 0.5305342674255371, "learning_rate": 0.0002, "epoch": 3.2629815745393635, "step": 4870}, {"loss": 1.5162, "grad_norm": 0.597670316696167, "learning_rate": 0.0002, "epoch": 3.2696817420435513, "step": 4880}, {"loss": 1.5429, "grad_norm": 0.665553867816925, "learning_rate": 0.0002, "epoch": 3.2763819095477387, "step": 4890}, {"loss": 1.4607, "grad_norm": 0.579767644405365, "learning_rate": 0.0002, "epoch": 3.2830820770519265, "step": 4900}, {"loss": 1.4999, "grad_norm": 0.5512481331825256, "learning_rate": 0.0002, "epoch": 3.289782244556114, "step": 4910}, {"loss": 1.5022, "grad_norm": 0.5916532278060913, "learning_rate": 0.0002, "epoch": 3.2964824120603016, "step": 4920}, {"loss": 1.4889, "grad_norm": 0.7521726489067078, "learning_rate": 0.0002, "epoch": 3.3031825795644894, "step": 4930}, {"loss": 1.4223, "grad_norm": 0.5352797508239746, "learning_rate": 0.0002, "epoch": 3.3098827470686767, "step": 4940}, {"loss": 1.5122, "grad_norm": 0.5950371623039246, "learning_rate": 0.0002, "epoch": 3.3165829145728645, "step": 4950}, {"loss": 1.5072, "grad_norm": 0.8020477890968323, "learning_rate": 0.0002, "epoch": 3.323283082077052, "step": 4960}, {"loss": 1.5422, "grad_norm": 0.6790024638175964, "learning_rate": 0.0002, "epoch": 3.3299832495812396, "step": 4970}, {"loss": 1.5363, "grad_norm": 0.687627375125885, "learning_rate": 0.0002, "epoch": 3.3366834170854274, "step": 4980}, {"loss": 1.5276, "grad_norm": 0.6094385385513306, "learning_rate": 0.0002, "epoch": 3.3433835845896147, "step": 4990}, {"loss": 1.549, "grad_norm": 0.6541242003440857, "learning_rate": 0.0002, "epoch": 3.3500837520938025, "step": 5000}, {"loss": 1.6067, "grad_norm": 0.5560880303382874, "learning_rate": 0.0002, "epoch": 3.35678391959799, "step": 5010}, {"loss": 1.5769, "grad_norm": 0.5440094470977783, "learning_rate": 0.0002, "epoch": 3.3634840871021776, "step": 5020}, {"loss": 1.6183, "grad_norm": 0.5749301314353943, "learning_rate": 0.0002, "epoch": 3.3701842546063654, "step": 5030}, {"loss": 1.4801, "grad_norm": 0.5919716954231262, "learning_rate": 0.0002, "epoch": 3.3768844221105527, "step": 5040}, {"loss": 1.5957, "grad_norm": 0.6331481337547302, "learning_rate": 0.0002, "epoch": 3.3835845896147405, "step": 5050}, {"loss": 1.5188, "grad_norm": 0.5687161684036255, "learning_rate": 0.0002, "epoch": 3.390284757118928, "step": 5060}, {"loss": 1.5702, "grad_norm": 0.6718577742576599, "learning_rate": 0.0002, "epoch": 3.3969849246231156, "step": 5070}, {"loss": 1.5577, "grad_norm": 0.5089324116706848, "learning_rate": 0.0002, "epoch": 3.4036850921273034, "step": 5080}, {"loss": 1.512, "grad_norm": 0.5710174441337585, "learning_rate": 0.0002, "epoch": 3.4103852596314908, "step": 5090}, {"loss": 1.5492, "grad_norm": 0.6670721173286438, "learning_rate": 0.0002, "epoch": 3.4170854271356785, "step": 5100}, {"loss": 1.5227, "grad_norm": 0.6875665187835693, "learning_rate": 0.0002, "epoch": 3.423785594639866, "step": 5110}, {"loss": 1.4496, "grad_norm": 0.5375880599021912, "learning_rate": 0.0002, "epoch": 3.4304857621440537, "step": 5120}, {"loss": 1.5527, "grad_norm": 0.6550399661064148, "learning_rate": 0.0002, "epoch": 3.4371859296482414, "step": 5130}, {"loss": 1.5687, "grad_norm": 0.5948067903518677, "learning_rate": 0.0002, "epoch": 3.4438860971524288, "step": 5140}, {"loss": 1.4813, "grad_norm": 0.6134477257728577, "learning_rate": 0.0002, "epoch": 3.4505862646566166, "step": 5150}, {"loss": 1.5069, "grad_norm": 0.6506398320198059, "learning_rate": 0.0002, "epoch": 3.457286432160804, "step": 5160}, {"loss": 1.4422, "grad_norm": 0.6060147881507874, "learning_rate": 0.0002, "epoch": 3.4639865996649917, "step": 5170}, {"loss": 1.5093, "grad_norm": 0.6173806190490723, "learning_rate": 0.0002, "epoch": 3.4706867671691795, "step": 5180}, {"loss": 1.4975, "grad_norm": 0.6032607555389404, "learning_rate": 0.0002, "epoch": 3.477386934673367, "step": 5190}, {"loss": 1.4979, "grad_norm": 0.5652492046356201, "learning_rate": 0.0002, "epoch": 3.4840871021775546, "step": 5200}, {"loss": 1.4883, "grad_norm": 0.6168607473373413, "learning_rate": 0.0002, "epoch": 3.490787269681742, "step": 5210}, {"loss": 1.5164, "grad_norm": 0.6170629262924194, "learning_rate": 0.0002, "epoch": 3.4974874371859297, "step": 5220}, {"loss": 1.4879, "grad_norm": 0.6926297545433044, "learning_rate": 0.0002, "epoch": 3.5041876046901175, "step": 5230}, {"loss": 1.4982, "grad_norm": 0.6702437996864319, "learning_rate": 0.0002, "epoch": 3.510887772194305, "step": 5240}, {"loss": 1.4986, "grad_norm": 0.5421436429023743, "learning_rate": 0.0002, "epoch": 3.5175879396984926, "step": 5250}, {"loss": 1.4673, "grad_norm": 0.5726765990257263, "learning_rate": 0.0002, "epoch": 3.52428810720268, "step": 5260}, {"loss": 1.5423, "grad_norm": 0.5685455203056335, "learning_rate": 0.0002, "epoch": 3.5309882747068677, "step": 5270}, {"loss": 1.4715, "grad_norm": 0.6018396019935608, "learning_rate": 0.0002, "epoch": 3.5376884422110555, "step": 5280}, {"loss": 1.5451, "grad_norm": 0.5731932520866394, "learning_rate": 0.0002, "epoch": 3.544388609715243, "step": 5290}, {"loss": 1.4752, "grad_norm": 0.6601519584655762, "learning_rate": 0.0002, "epoch": 3.5510887772194306, "step": 5300}, {"loss": 1.5434, "grad_norm": 0.5545530319213867, "learning_rate": 0.0002, "epoch": 3.557788944723618, "step": 5310}, {"loss": 1.5438, "grad_norm": 0.5998541116714478, "learning_rate": 0.0002, "epoch": 3.5644891122278057, "step": 5320}, {"loss": 1.56, "grad_norm": 0.5651767253875732, "learning_rate": 0.0002, "epoch": 3.5711892797319935, "step": 5330}, {"loss": 1.4829, "grad_norm": 0.7425084114074707, "learning_rate": 0.0002, "epoch": 3.577889447236181, "step": 5340}, {"loss": 1.5571, "grad_norm": 0.5770602226257324, "learning_rate": 0.0002, "epoch": 3.5845896147403686, "step": 5350}, {"loss": 1.458, "grad_norm": 0.54723060131073, "learning_rate": 0.0002, "epoch": 3.591289782244556, "step": 5360}, {"loss": 1.497, "grad_norm": 0.6658238172531128, "learning_rate": 0.0002, "epoch": 3.5979899497487438, "step": 5370}, {"loss": 1.5456, "grad_norm": 0.5787645578384399, "learning_rate": 0.0002, "epoch": 3.6046901172529315, "step": 5380}, {"loss": 1.5343, "grad_norm": 0.594913125038147, "learning_rate": 0.0002, "epoch": 3.611390284757119, "step": 5390}, {"loss": 1.4727, "grad_norm": 0.4964977502822876, "learning_rate": 0.0002, "epoch": 3.6180904522613067, "step": 5400}, {"loss": 1.5062, "grad_norm": 0.6087527275085449, "learning_rate": 0.0002, "epoch": 3.624790619765494, "step": 5410}, {"loss": 1.5098, "grad_norm": 0.6315323710441589, "learning_rate": 0.0002, "epoch": 3.6314907872696818, "step": 5420}, {"loss": 1.4855, "grad_norm": 0.574799120426178, "learning_rate": 0.0002, "epoch": 3.6381909547738696, "step": 5430}, {"loss": 1.4595, "grad_norm": 0.5949277877807617, "learning_rate": 0.0002, "epoch": 3.644891122278057, "step": 5440}, {"loss": 1.4816, "grad_norm": 0.5640677213668823, "learning_rate": 0.0002, "epoch": 3.6515912897822447, "step": 5450}, {"loss": 1.525, "grad_norm": 0.6198237538337708, "learning_rate": 0.0002, "epoch": 3.658291457286432, "step": 5460}, {"loss": 1.5676, "grad_norm": 0.6902034878730774, "learning_rate": 0.0002, "epoch": 3.66499162479062, "step": 5470}, {"loss": 1.544, "grad_norm": 0.5686674118041992, "learning_rate": 0.0002, "epoch": 3.6716917922948076, "step": 5480}, {"loss": 1.5255, "grad_norm": 0.6532107591629028, "learning_rate": 0.0002, "epoch": 3.678391959798995, "step": 5490}, {"loss": 1.5767, "grad_norm": 0.5790849924087524, "learning_rate": 0.0002, "epoch": 3.6850921273031827, "step": 5500}, {"loss": 1.5507, "grad_norm": 0.6055065393447876, "learning_rate": 0.0002, "epoch": 3.69179229480737, "step": 5510}, {"loss": 1.4656, "grad_norm": 0.5630605816841125, "learning_rate": 0.0002, "epoch": 3.698492462311558, "step": 5520}, {"loss": 1.537, "grad_norm": 0.6005825996398926, "learning_rate": 0.0002, "epoch": 3.7051926298157456, "step": 5530}, {"loss": 1.5313, "grad_norm": 0.6553038954734802, "learning_rate": 0.0002, "epoch": 3.711892797319933, "step": 5540}, {"loss": 1.4943, "grad_norm": 0.5601094961166382, "learning_rate": 0.0002, "epoch": 3.7185929648241207, "step": 5550}, {"loss": 1.5161, "grad_norm": 0.6598808169364929, "learning_rate": 0.0002, "epoch": 3.725293132328308, "step": 5560}, {"loss": 1.5345, "grad_norm": 0.5506255626678467, "learning_rate": 0.0002, "epoch": 3.731993299832496, "step": 5570}, {"loss": 1.4805, "grad_norm": 0.6001223921775818, "learning_rate": 0.0002, "epoch": 3.7386934673366836, "step": 5580}, {"loss": 1.4652, "grad_norm": 0.6287297606468201, "learning_rate": 0.0002, "epoch": 3.745393634840871, "step": 5590}, {"loss": 1.5246, "grad_norm": 0.6253238916397095, "learning_rate": 0.0002, "epoch": 3.7520938023450587, "step": 5600}, {"loss": 1.5691, "grad_norm": 0.5713174939155579, "learning_rate": 0.0002, "epoch": 3.758793969849246, "step": 5610}, {"loss": 1.5661, "grad_norm": 0.6198310852050781, "learning_rate": 0.0002, "epoch": 3.765494137353434, "step": 5620}, {"loss": 1.5448, "grad_norm": 0.5941224098205566, "learning_rate": 0.0002, "epoch": 3.7721943048576216, "step": 5630}, {"loss": 1.4925, "grad_norm": 0.606002151966095, "learning_rate": 0.0002, "epoch": 3.778894472361809, "step": 5640}, {"loss": 1.5182, "grad_norm": 0.6540704965591431, "learning_rate": 0.0002, "epoch": 3.7855946398659968, "step": 5650}, {"loss": 1.5903, "grad_norm": 0.6147415041923523, "learning_rate": 0.0002, "epoch": 3.792294807370184, "step": 5660}, {"loss": 1.5329, "grad_norm": 0.5649605393409729, "learning_rate": 0.0002, "epoch": 3.798994974874372, "step": 5670}, {"loss": 1.5747, "grad_norm": 0.6788773536682129, "learning_rate": 0.0002, "epoch": 3.8056951423785597, "step": 5680}, {"loss": 1.535, "grad_norm": 0.6581860780715942, "learning_rate": 0.0002, "epoch": 3.812395309882747, "step": 5690}, {"loss": 1.4587, "grad_norm": 0.5529348850250244, "learning_rate": 0.0002, "epoch": 3.819095477386935, "step": 5700}, {"loss": 1.5594, "grad_norm": 0.6320232152938843, "learning_rate": 0.0002, "epoch": 3.825795644891122, "step": 5710}, {"loss": 1.4696, "grad_norm": 0.6529698371887207, "learning_rate": 0.0002, "epoch": 3.83249581239531, "step": 5720}, {"loss": 1.5854, "grad_norm": 0.5983362793922424, "learning_rate": 0.0002, "epoch": 3.8391959798994977, "step": 5730}, {"loss": 1.465, "grad_norm": 0.6335684061050415, "learning_rate": 0.0002, "epoch": 3.845896147403685, "step": 5740}, {"loss": 1.5545, "grad_norm": 0.700446605682373, "learning_rate": 0.0002, "epoch": 3.852596314907873, "step": 5750}, {"loss": 1.5707, "grad_norm": 0.6092597842216492, "learning_rate": 0.0002, "epoch": 3.85929648241206, "step": 5760}, {"loss": 1.5729, "grad_norm": 0.564146101474762, "learning_rate": 0.0002, "epoch": 3.865996649916248, "step": 5770}, {"loss": 1.5872, "grad_norm": 0.615275502204895, "learning_rate": 0.0002, "epoch": 3.8726968174204357, "step": 5780}, {"loss": 1.5142, "grad_norm": 0.6685376763343811, "learning_rate": 0.0002, "epoch": 3.879396984924623, "step": 5790}, {"loss": 1.4752, "grad_norm": 0.6116922497749329, "learning_rate": 0.0002, "epoch": 3.886097152428811, "step": 5800}, {"loss": 1.5179, "grad_norm": 0.5486813187599182, "learning_rate": 0.0002, "epoch": 3.892797319932998, "step": 5810}, {"loss": 1.5167, "grad_norm": 0.6208204030990601, "learning_rate": 0.0002, "epoch": 3.899497487437186, "step": 5820}, {"loss": 1.5334, "grad_norm": 0.6500625014305115, "learning_rate": 0.0002, "epoch": 3.9061976549413737, "step": 5830}, {"loss": 1.4716, "grad_norm": 0.5948089361190796, "learning_rate": 0.0002, "epoch": 3.912897822445561, "step": 5840}, {"loss": 1.6011, "grad_norm": 0.7210732698440552, "learning_rate": 0.0002, "epoch": 3.919597989949749, "step": 5850}, {"loss": 1.5519, "grad_norm": 0.6662322878837585, "learning_rate": 0.0002, "epoch": 3.926298157453936, "step": 5860}, {"loss": 1.5656, "grad_norm": 0.5613839626312256, "learning_rate": 0.0002, "epoch": 3.932998324958124, "step": 5870}, {"loss": 1.544, "grad_norm": 0.6069002151489258, "learning_rate": 0.0002, "epoch": 3.9396984924623117, "step": 5880}, {"loss": 1.6745, "grad_norm": 0.7075562477111816, "learning_rate": 0.0002, "epoch": 3.946398659966499, "step": 5890}, {"loss": 1.5391, "grad_norm": 0.6316173076629639, "learning_rate": 0.0002, "epoch": 3.953098827470687, "step": 5900}, {"loss": 1.6314, "grad_norm": 0.5716308355331421, "learning_rate": 0.0002, "epoch": 3.959798994974874, "step": 5910}, {"loss": 1.5947, "grad_norm": 0.6800096035003662, "learning_rate": 0.0002, "epoch": 3.966499162479062, "step": 5920}, {"loss": 1.5189, "grad_norm": 0.6057983040809631, "learning_rate": 0.0002, "epoch": 3.9731993299832498, "step": 5930}, {"loss": 1.5431, "grad_norm": 0.5938987731933594, "learning_rate": 0.0002, "epoch": 3.979899497487437, "step": 5940}, {"loss": 1.5111, "grad_norm": 0.6963576674461365, "learning_rate": 0.0002, "epoch": 3.986599664991625, "step": 5950}, {"loss": 1.5521, "grad_norm": 0.6279940009117126, "learning_rate": 0.0002, "epoch": 3.993299832495812, "step": 5960}, {"loss": 1.5974, "grad_norm": 0.7161159515380859, "learning_rate": 0.0002, "epoch": 4.0, "step": 5970}, {"eval_loss": 1.8655421733856201, "eval_runtime": 37.9276, "eval_samples_per_second": 13.579, "eval_steps_per_second": 1.714, "epoch": 4.0, "step": 5970}, {"loss": 1.3666, "grad_norm": 0.7380476593971252, "learning_rate": 0.0002, "epoch": 4.006700167504188, "step": 5980}, {"loss": 1.3913, "grad_norm": 0.7148947715759277, "learning_rate": 0.0002, "epoch": 4.013400335008376, "step": 5990}, {"loss": 1.4204, "grad_norm": 0.6177082657814026, "learning_rate": 0.0002, "epoch": 4.0201005025125625, "step": 6000}, {"loss": 1.4421, "grad_norm": 0.8552946448326111, "learning_rate": 0.0002, "epoch": 4.02680067001675, "step": 6010}, {"loss": 1.4342, "grad_norm": 0.8033416271209717, "learning_rate": 0.0002, "epoch": 4.033500837520938, "step": 6020}, {"loss": 1.4092, "grad_norm": 0.8501318097114563, "learning_rate": 0.0002, "epoch": 4.040201005025126, "step": 6030}, {"loss": 1.3367, "grad_norm": 0.6981393098831177, "learning_rate": 0.0002, "epoch": 4.046901172529314, "step": 6040}, {"loss": 1.3925, "grad_norm": 0.7227180600166321, "learning_rate": 0.0002, "epoch": 4.0536013400335005, "step": 6050}, {"loss": 1.4007, "grad_norm": 0.6923989653587341, "learning_rate": 0.0002, "epoch": 4.060301507537688, "step": 6060}, {"loss": 1.3837, "grad_norm": 0.879779040813446, "learning_rate": 0.0002, "epoch": 4.067001675041876, "step": 6070}, {"loss": 1.4383, "grad_norm": 0.8184754848480225, "learning_rate": 0.0002, "epoch": 4.073701842546064, "step": 6080}, {"loss": 1.3128, "grad_norm": 0.8211342692375183, "learning_rate": 0.0002, "epoch": 4.080402010050252, "step": 6090}, {"loss": 1.3892, "grad_norm": 0.7542396783828735, "learning_rate": 0.0002, "epoch": 4.0871021775544385, "step": 6100}, {"loss": 1.3607, "grad_norm": 0.6631066799163818, "learning_rate": 0.0002, "epoch": 4.093802345058626, "step": 6110}, {"loss": 1.3275, "grad_norm": 0.6728386282920837, "learning_rate": 0.0002, "epoch": 4.100502512562814, "step": 6120}, {"loss": 1.3443, "grad_norm": 0.681851863861084, "learning_rate": 0.0002, "epoch": 4.107202680067002, "step": 6130}, {"loss": 1.3486, "grad_norm": 0.8757794499397278, "learning_rate": 0.0002, "epoch": 4.11390284757119, "step": 6140}, {"loss": 1.351, "grad_norm": 0.6567301750183105, "learning_rate": 0.0002, "epoch": 4.1206030150753765, "step": 6150}, {"loss": 1.3824, "grad_norm": 0.7950329184532166, "learning_rate": 0.0002, "epoch": 4.127303182579564, "step": 6160}, {"loss": 1.3738, "grad_norm": 0.7545644044876099, "learning_rate": 0.0002, "epoch": 4.134003350083752, "step": 6170}, {"loss": 1.4214, "grad_norm": 0.7172710299491882, "learning_rate": 0.0002, "epoch": 4.14070351758794, "step": 6180}, {"loss": 1.4091, "grad_norm": 0.7040584087371826, "learning_rate": 0.0002, "epoch": 4.147403685092128, "step": 6190}, {"loss": 1.4149, "grad_norm": 0.7482913732528687, "learning_rate": 0.0002, "epoch": 4.1541038525963145, "step": 6200}, {"loss": 1.3227, "grad_norm": 0.8523276448249817, "learning_rate": 0.0002, "epoch": 4.160804020100502, "step": 6210}, {"loss": 1.4194, "grad_norm": 0.6672041416168213, "learning_rate": 0.0002, "epoch": 4.16750418760469, "step": 6220}, {"loss": 1.3953, "grad_norm": 0.7523500919342041, "learning_rate": 0.0002, "epoch": 4.174204355108878, "step": 6230}, {"loss": 1.371, "grad_norm": 0.8085253834724426, "learning_rate": 0.0002, "epoch": 4.180904522613066, "step": 6240}, {"loss": 1.3293, "grad_norm": 0.789450466632843, "learning_rate": 0.0002, "epoch": 4.187604690117253, "step": 6250}, {"loss": 1.3539, "grad_norm": 0.7502310872077942, "learning_rate": 0.0002, "epoch": 4.19430485762144, "step": 6260}, {"loss": 1.3415, "grad_norm": 0.7397456765174866, "learning_rate": 0.0002, "epoch": 4.201005025125628, "step": 6270}, {"loss": 1.3963, "grad_norm": 0.6921947002410889, "learning_rate": 0.0002, "epoch": 4.207705192629816, "step": 6280}, {"loss": 1.3125, "grad_norm": 0.9334571957588196, "learning_rate": 0.0002, "epoch": 4.214405360134004, "step": 6290}, {"loss": 1.3612, "grad_norm": 0.725799024105072, "learning_rate": 0.0002, "epoch": 4.221105527638191, "step": 6300}, {"loss": 1.4217, "grad_norm": 0.8290495872497559, "learning_rate": 0.0002, "epoch": 4.227805695142378, "step": 6310}, {"loss": 1.4135, "grad_norm": 0.688983678817749, "learning_rate": 0.0002, "epoch": 4.234505862646566, "step": 6320}, {"loss": 1.3807, "grad_norm": 0.8620913028717041, "learning_rate": 0.0002, "epoch": 4.241206030150754, "step": 6330}, {"loss": 1.3738, "grad_norm": 0.8008657693862915, "learning_rate": 0.0002, "epoch": 4.247906197654942, "step": 6340}, {"loss": 1.4005, "grad_norm": 0.7379199266433716, "learning_rate": 0.0002, "epoch": 4.254606365159129, "step": 6350}, {"loss": 1.426, "grad_norm": 0.7842815518379211, "learning_rate": 0.0002, "epoch": 4.261306532663316, "step": 6360}, {"loss": 1.4262, "grad_norm": 0.812600314617157, "learning_rate": 0.0002, "epoch": 4.268006700167504, "step": 6370}, {"loss": 1.4028, "grad_norm": 0.7852841019630432, "learning_rate": 0.0002, "epoch": 4.274706867671692, "step": 6380}, {"loss": 1.3722, "grad_norm": 1.0377534627914429, "learning_rate": 0.0002, "epoch": 4.28140703517588, "step": 6390}, {"loss": 1.3755, "grad_norm": 1.03935706615448, "learning_rate": 0.0002, "epoch": 4.288107202680067, "step": 6400}, {"loss": 1.3961, "grad_norm": 0.7244732975959778, "learning_rate": 0.0002, "epoch": 4.294807370184254, "step": 6410}, {"loss": 1.4608, "grad_norm": 0.7137406468391418, "learning_rate": 0.0002, "epoch": 4.301507537688442, "step": 6420}, {"loss": 1.4461, "grad_norm": 0.7492543458938599, "learning_rate": 0.0002, "epoch": 4.30820770519263, "step": 6430}, {"loss": 1.4562, "grad_norm": 0.7065439224243164, "learning_rate": 0.0002, "epoch": 4.314907872696818, "step": 6440}, {"loss": 1.4246, "grad_norm": 0.7786989808082581, "learning_rate": 0.0002, "epoch": 4.321608040201005, "step": 6450}, {"loss": 1.3098, "grad_norm": 0.7369208335876465, "learning_rate": 0.0002, "epoch": 4.328308207705192, "step": 6460}, {"loss": 1.3686, "grad_norm": 0.7412346005439758, "learning_rate": 0.0002, "epoch": 4.33500837520938, "step": 6470}, {"loss": 1.4087, "grad_norm": 0.780927300453186, "learning_rate": 0.0002, "epoch": 4.341708542713568, "step": 6480}, {"loss": 1.3628, "grad_norm": 0.8320930600166321, "learning_rate": 0.0002, "epoch": 4.348408710217756, "step": 6490}, {"loss": 1.3715, "grad_norm": 0.6871094703674316, "learning_rate": 0.0002, "epoch": 4.355108877721943, "step": 6500}, {"loss": 1.3257, "grad_norm": 0.6751559972763062, "learning_rate": 0.0002, "epoch": 4.36180904522613, "step": 6510}, {"loss": 1.4311, "grad_norm": 0.7723976969718933, "learning_rate": 0.0002, "epoch": 4.368509212730318, "step": 6520}, {"loss": 1.4086, "grad_norm": 0.7915401458740234, "learning_rate": 0.0002, "epoch": 4.375209380234506, "step": 6530}, {"loss": 1.3973, "grad_norm": 0.7329102754592896, "learning_rate": 0.0002, "epoch": 4.381909547738694, "step": 6540}, {"loss": 1.447, "grad_norm": 0.7388760447502136, "learning_rate": 0.0002, "epoch": 4.388609715242881, "step": 6550}, {"loss": 1.4378, "grad_norm": 0.8282579183578491, "learning_rate": 0.0002, "epoch": 4.3953098827470685, "step": 6560}, {"loss": 1.3923, "grad_norm": 0.7192724347114563, "learning_rate": 0.0002, "epoch": 4.402010050251256, "step": 6570}, {"loss": 1.4141, "grad_norm": 0.746526837348938, "learning_rate": 0.0002, "epoch": 4.408710217755444, "step": 6580}, {"loss": 1.33, "grad_norm": 0.8738046288490295, "learning_rate": 0.0002, "epoch": 4.415410385259632, "step": 6590}, {"loss": 1.3995, "grad_norm": 0.8408458828926086, "learning_rate": 0.0002, "epoch": 4.422110552763819, "step": 6600}, {"loss": 1.4148, "grad_norm": 0.8110666275024414, "learning_rate": 0.0002, "epoch": 4.4288107202680065, "step": 6610}, {"loss": 1.441, "grad_norm": 0.8602406978607178, "learning_rate": 0.0002, "epoch": 4.435510887772194, "step": 6620}, {"loss": 1.4319, "grad_norm": 0.7549102902412415, "learning_rate": 0.0002, "epoch": 4.442211055276382, "step": 6630}, {"loss": 1.388, "grad_norm": 0.7831804156303406, "learning_rate": 0.0002, "epoch": 4.44891122278057, "step": 6640}, {"loss": 1.4283, "grad_norm": 0.7269673943519592, "learning_rate": 0.0002, "epoch": 4.455611390284757, "step": 6650}, {"loss": 1.4132, "grad_norm": 0.7397838830947876, "learning_rate": 0.0002, "epoch": 4.4623115577889445, "step": 6660}, {"loss": 1.3174, "grad_norm": 0.713707447052002, "learning_rate": 0.0002, "epoch": 4.469011725293132, "step": 6670}, {"loss": 1.3406, "grad_norm": 0.7525581121444702, "learning_rate": 0.0002, "epoch": 4.47571189279732, "step": 6680}, {"loss": 1.4283, "grad_norm": 0.8030191659927368, "learning_rate": 0.0002, "epoch": 4.482412060301508, "step": 6690}, {"loss": 1.4586, "grad_norm": 0.7469439506530762, "learning_rate": 0.0002, "epoch": 4.489112227805695, "step": 6700}, {"loss": 1.367, "grad_norm": 0.7743868231773376, "learning_rate": 0.0002, "epoch": 4.4958123953098825, "step": 6710}, {"loss": 1.3439, "grad_norm": 0.6539737582206726, "learning_rate": 0.0002, "epoch": 4.50251256281407, "step": 6720}, {"loss": 1.4513, "grad_norm": 0.825818657875061, "learning_rate": 0.0002, "epoch": 4.509212730318258, "step": 6730}, {"loss": 1.3984, "grad_norm": 0.8048575520515442, "learning_rate": 0.0002, "epoch": 4.515912897822446, "step": 6740}, {"loss": 1.3923, "grad_norm": 0.7828766107559204, "learning_rate": 0.0002, "epoch": 4.522613065326633, "step": 6750}, {"loss": 1.3886, "grad_norm": 0.7406010031700134, "learning_rate": 0.0002, "epoch": 4.5293132328308205, "step": 6760}, {"loss": 1.3109, "grad_norm": 0.840345561504364, "learning_rate": 0.0002, "epoch": 4.536013400335008, "step": 6770}, {"loss": 1.4808, "grad_norm": 0.8492622971534729, "learning_rate": 0.0002, "epoch": 4.542713567839196, "step": 6780}, {"loss": 1.4384, "grad_norm": 0.7130163908004761, "learning_rate": 0.0002, "epoch": 4.549413735343384, "step": 6790}, {"loss": 1.4531, "grad_norm": 0.8454728126525879, "learning_rate": 0.0002, "epoch": 4.556113902847571, "step": 6800}, {"loss": 1.3239, "grad_norm": 0.7847645282745361, "learning_rate": 0.0002, "epoch": 4.562814070351759, "step": 6810}, {"loss": 1.4181, "grad_norm": 0.7245864272117615, "learning_rate": 0.0002, "epoch": 4.569514237855946, "step": 6820}, {"loss": 1.3233, "grad_norm": 0.768893301486969, "learning_rate": 0.0002, "epoch": 4.576214405360134, "step": 6830}, {"loss": 1.3932, "grad_norm": 0.8028400540351868, "learning_rate": 0.0002, "epoch": 4.582914572864322, "step": 6840}, {"loss": 1.3745, "grad_norm": 0.763945460319519, "learning_rate": 0.0002, "epoch": 4.589614740368509, "step": 6850}, {"loss": 1.4797, "grad_norm": 0.7417685389518738, "learning_rate": 0.0002, "epoch": 4.596314907872697, "step": 6860}, {"loss": 1.4468, "grad_norm": 0.7603038549423218, "learning_rate": 0.0002, "epoch": 4.603015075376884, "step": 6870}, {"loss": 1.4095, "grad_norm": 0.7981528043746948, "learning_rate": 0.0002, "epoch": 4.609715242881072, "step": 6880}, {"loss": 1.3963, "grad_norm": 0.8077111840248108, "learning_rate": 0.0002, "epoch": 4.61641541038526, "step": 6890}, {"loss": 1.4721, "grad_norm": 0.8778454065322876, "learning_rate": 0.0002, "epoch": 4.623115577889447, "step": 6900}, {"loss": 1.3758, "grad_norm": 0.8620710372924805, "learning_rate": 0.0002, "epoch": 4.629815745393635, "step": 6910}, {"loss": 1.344, "grad_norm": 0.7486072778701782, "learning_rate": 0.0002, "epoch": 4.636515912897822, "step": 6920}, {"loss": 1.3913, "grad_norm": 0.7493042945861816, "learning_rate": 0.0002, "epoch": 4.64321608040201, "step": 6930}, {"loss": 1.397, "grad_norm": 0.7388978600502014, "learning_rate": 0.0002, "epoch": 4.649916247906198, "step": 6940}, {"loss": 1.3593, "grad_norm": 0.798530638217926, "learning_rate": 0.0002, "epoch": 4.656616415410385, "step": 6950}, {"loss": 1.3982, "grad_norm": 0.7929500937461853, "learning_rate": 0.0002, "epoch": 4.663316582914573, "step": 6960}, {"loss": 1.4183, "grad_norm": 0.9186785221099854, "learning_rate": 0.0002, "epoch": 4.67001675041876, "step": 6970}, {"loss": 1.3955, "grad_norm": 1.1103485822677612, "learning_rate": 0.0002, "epoch": 4.676716917922948, "step": 6980}, {"loss": 1.3941, "grad_norm": 0.8000466823577881, "learning_rate": 0.0002, "epoch": 4.683417085427136, "step": 6990}, {"loss": 1.371, "grad_norm": 0.7520599961280823, "learning_rate": 0.0002, "epoch": 4.690117252931323, "step": 7000}, {"loss": 1.4582, "grad_norm": 0.7971973419189453, "learning_rate": 0.0002, "epoch": 4.696817420435511, "step": 7010}, {"loss": 1.3682, "grad_norm": 0.7363343834877014, "learning_rate": 0.0002, "epoch": 4.703517587939698, "step": 7020}, {"loss": 1.3889, "grad_norm": 0.8268865942955017, "learning_rate": 0.0002, "epoch": 4.710217755443886, "step": 7030}, {"loss": 1.4382, "grad_norm": 0.7054963111877441, "learning_rate": 0.0002, "epoch": 4.716917922948074, "step": 7040}, {"loss": 1.4578, "grad_norm": 0.8196262121200562, "learning_rate": 0.0002, "epoch": 4.723618090452261, "step": 7050}, {"loss": 1.365, "grad_norm": 0.8276031017303467, "learning_rate": 0.0002, "epoch": 4.730318257956449, "step": 7060}, {"loss": 1.3887, "grad_norm": 0.8248157501220703, "learning_rate": 0.0002, "epoch": 4.7370184254606365, "step": 7070}, {"loss": 1.4193, "grad_norm": 0.8937979936599731, "learning_rate": 0.0002, "epoch": 4.743718592964824, "step": 7080}, {"loss": 1.4334, "grad_norm": 0.7984238862991333, "learning_rate": 0.0002, "epoch": 4.750418760469012, "step": 7090}, {"loss": 1.4385, "grad_norm": 0.9495313763618469, "learning_rate": 0.0002, "epoch": 4.757118927973199, "step": 7100}, {"loss": 1.4504, "grad_norm": 0.8598204255104065, "learning_rate": 0.0002, "epoch": 4.763819095477387, "step": 7110}, {"loss": 1.3969, "grad_norm": 0.8951472640037537, "learning_rate": 0.0002, "epoch": 4.7705192629815745, "step": 7120}, {"loss": 1.4339, "grad_norm": 0.9110309481620789, "learning_rate": 0.0002, "epoch": 4.777219430485762, "step": 7130}, {"loss": 1.4001, "grad_norm": 0.7929584980010986, "learning_rate": 0.0002, "epoch": 4.78391959798995, "step": 7140}, {"loss": 1.467, "grad_norm": 0.7415322661399841, "learning_rate": 0.0002, "epoch": 4.790619765494137, "step": 7150}, {"loss": 1.5107, "grad_norm": 0.7504757046699524, "learning_rate": 0.0002, "epoch": 4.797319932998325, "step": 7160}, {"loss": 1.3736, "grad_norm": 0.7166924476623535, "learning_rate": 0.0002, "epoch": 4.8040201005025125, "step": 7170}, {"loss": 1.4088, "grad_norm": 0.7728400826454163, "learning_rate": 0.0002, "epoch": 4.8107202680067, "step": 7180}, {"loss": 1.3814, "grad_norm": 0.7992154955863953, "learning_rate": 0.0002, "epoch": 4.817420435510888, "step": 7190}, {"loss": 1.3958, "grad_norm": 0.8655321002006531, "learning_rate": 0.0002, "epoch": 4.824120603015075, "step": 7200}, {"loss": 1.3837, "grad_norm": 0.7672632336616516, "learning_rate": 0.0002, "epoch": 4.830820770519263, "step": 7210}, {"loss": 1.4578, "grad_norm": 0.708416223526001, "learning_rate": 0.0002, "epoch": 4.8375209380234505, "step": 7220}, {"loss": 1.5413, "grad_norm": 0.8914081454277039, "learning_rate": 0.0002, "epoch": 4.844221105527638, "step": 7230}, {"loss": 1.3569, "grad_norm": 0.7141931653022766, "learning_rate": 0.0002, "epoch": 4.850921273031826, "step": 7240}, {"loss": 1.4532, "grad_norm": 0.6913040280342102, "learning_rate": 0.0002, "epoch": 4.857621440536013, "step": 7250}, {"loss": 1.3912, "grad_norm": 0.7871233820915222, "learning_rate": 0.0002, "epoch": 4.864321608040201, "step": 7260}, {"loss": 1.3688, "grad_norm": 0.8466277122497559, "learning_rate": 0.0002, "epoch": 4.8710217755443885, "step": 7270}, {"loss": 1.33, "grad_norm": 0.8492183685302734, "learning_rate": 0.0002, "epoch": 4.877721943048576, "step": 7280}, {"loss": 1.3744, "grad_norm": 0.8339574933052063, "learning_rate": 0.0002, "epoch": 4.884422110552764, "step": 7290}, {"loss": 1.4157, "grad_norm": 0.787022590637207, "learning_rate": 0.0002, "epoch": 4.891122278056951, "step": 7300}, {"loss": 1.3725, "grad_norm": 0.8877332806587219, "learning_rate": 0.0002, "epoch": 4.897822445561139, "step": 7310}, {"loss": 1.3968, "grad_norm": 0.744989812374115, "learning_rate": 0.0002, "epoch": 4.9045226130653266, "step": 7320}, {"loss": 1.4421, "grad_norm": 0.8027268648147583, "learning_rate": 0.0002, "epoch": 4.911222780569514, "step": 7330}, {"loss": 1.425, "grad_norm": 0.6437455415725708, "learning_rate": 0.0002, "epoch": 4.917922948073702, "step": 7340}, {"loss": 1.4829, "grad_norm": 0.685999870300293, "learning_rate": 0.0002, "epoch": 4.924623115577889, "step": 7350}, {"loss": 1.4352, "grad_norm": 0.9086187481880188, "learning_rate": 0.0002, "epoch": 4.931323283082077, "step": 7360}, {"loss": 1.4245, "grad_norm": 0.8272411227226257, "learning_rate": 0.0002, "epoch": 4.938023450586265, "step": 7370}, {"loss": 1.4226, "grad_norm": 0.9227852821350098, "learning_rate": 0.0002, "epoch": 4.944723618090452, "step": 7380}, {"loss": 1.3643, "grad_norm": 0.7688441276550293, "learning_rate": 0.0002, "epoch": 4.95142378559464, "step": 7390}, {"loss": 1.4491, "grad_norm": 0.8662643432617188, "learning_rate": 0.0002, "epoch": 4.958123953098827, "step": 7400}, {"loss": 1.4194, "grad_norm": 0.9234127998352051, "learning_rate": 0.0002, "epoch": 4.964824120603015, "step": 7410}, {"loss": 1.4009, "grad_norm": 0.9131470918655396, "learning_rate": 0.0002, "epoch": 4.971524288107203, "step": 7420}, {"loss": 1.4544, "grad_norm": 0.7377504110336304, "learning_rate": 0.0002, "epoch": 4.97822445561139, "step": 7430}, {"loss": 1.4008, "grad_norm": 0.8762801289558411, "learning_rate": 0.0002, "epoch": 4.984924623115578, "step": 7440}, {"loss": 1.4304, "grad_norm": 0.7919872999191284, "learning_rate": 0.0002, "epoch": 4.991624790619765, "step": 7450}, {"loss": 1.3817, "grad_norm": 0.7144299149513245, "learning_rate": 0.0002, "epoch": 4.998324958123953, "step": 7460}, {"eval_loss": 1.9291157722473145, "eval_runtime": 37.9831, "eval_samples_per_second": 13.559, "eval_steps_per_second": 1.711, "epoch": 4.99966499162479, "step": 7462}, {"loss": 1.2753, "grad_norm": 0.7860151529312134, "learning_rate": 0.0002, "epoch": 5.005025125628141, "step": 7470}, {"loss": 1.2149, "grad_norm": 0.9418314695358276, "learning_rate": 0.0002, "epoch": 5.011725293132328, "step": 7480}, {"loss": 1.1966, "grad_norm": 0.8474572896957397, "learning_rate": 0.0002, "epoch": 5.018425460636516, "step": 7490}, {"loss": 1.2111, "grad_norm": 1.0724040269851685, "learning_rate": 0.0002, "epoch": 5.025125628140704, "step": 7500}, {"loss": 1.2228, "grad_norm": 0.9109148979187012, "learning_rate": 0.0002, "epoch": 5.031825795644891, "step": 7510}, {"loss": 1.2239, "grad_norm": 1.0088659524917603, "learning_rate": 0.0002, "epoch": 5.038525963149079, "step": 7520}, {"loss": 1.2156, "grad_norm": 1.1421623229980469, "learning_rate": 0.0002, "epoch": 5.045226130653266, "step": 7530}, {"loss": 1.1739, "grad_norm": 0.9219902157783508, "learning_rate": 0.0002, "epoch": 5.051926298157454, "step": 7540}, {"loss": 1.2686, "grad_norm": 0.9150987863540649, "learning_rate": 0.0002, "epoch": 5.058626465661642, "step": 7550}, {"loss": 1.2068, "grad_norm": 0.8889328241348267, "learning_rate": 0.0002, "epoch": 5.065326633165829, "step": 7560}, {"loss": 1.276, "grad_norm": 0.9751363396644592, "learning_rate": 0.0002, "epoch": 5.072026800670017, "step": 7570}, {"loss": 1.2078, "grad_norm": 0.8603123426437378, "learning_rate": 0.0002, "epoch": 5.078726968174204, "step": 7580}, {"loss": 1.2175, "grad_norm": 0.8910616636276245, "learning_rate": 0.0002, "epoch": 5.085427135678392, "step": 7590}, {"loss": 1.2475, "grad_norm": 1.1128392219543457, "learning_rate": 0.0002, "epoch": 5.09212730318258, "step": 7600}, {"loss": 1.3065, "grad_norm": 0.9480258822441101, "learning_rate": 0.0002, "epoch": 5.098827470686767, "step": 7610}, {"loss": 1.193, "grad_norm": 0.906958818435669, "learning_rate": 0.0002, "epoch": 5.105527638190955, "step": 7620}, {"loss": 1.2223, "grad_norm": 0.8741167187690735, "learning_rate": 0.0002, "epoch": 5.1122278056951425, "step": 7630}, {"loss": 1.2126, "grad_norm": 0.966268002986908, "learning_rate": 0.0002, "epoch": 5.11892797319933, "step": 7640}, {"loss": 1.2782, "grad_norm": 0.9124358892440796, "learning_rate": 0.0002, "epoch": 5.125628140703517, "step": 7650}, {"loss": 1.3004, "grad_norm": 1.0436606407165527, "learning_rate": 0.0002, "epoch": 5.132328308207705, "step": 7660}, {"loss": 1.2675, "grad_norm": 0.9217309355735779, "learning_rate": 0.0002, "epoch": 5.139028475711893, "step": 7670}, {"loss": 1.2502, "grad_norm": 1.344765543937683, "learning_rate": 0.0002, "epoch": 5.1457286432160805, "step": 7680}, {"loss": 1.2416, "grad_norm": 1.0730723142623901, "learning_rate": 0.0002, "epoch": 5.152428810720268, "step": 7690}, {"loss": 1.1888, "grad_norm": 0.9321247339248657, "learning_rate": 0.0002, "epoch": 5.159128978224456, "step": 7700}, {"loss": 1.1941, "grad_norm": 0.8482614755630493, "learning_rate": 0.0002, "epoch": 5.165829145728643, "step": 7710}, {"loss": 1.2668, "grad_norm": 0.8274452686309814, "learning_rate": 0.0002, "epoch": 5.172529313232831, "step": 7720}, {"loss": 1.1972, "grad_norm": 0.9120376706123352, "learning_rate": 0.0002, "epoch": 5.1792294807370185, "step": 7730}, {"loss": 1.1648, "grad_norm": 1.0062892436981201, "learning_rate": 0.0002, "epoch": 5.185929648241206, "step": 7740}, {"loss": 1.2199, "grad_norm": 0.9521504640579224, "learning_rate": 0.0002, "epoch": 5.192629815745394, "step": 7750}, {"loss": 1.2855, "grad_norm": 0.8800198435783386, "learning_rate": 0.0002, "epoch": 5.199329983249581, "step": 7760}, {"loss": 1.2535, "grad_norm": 0.9749179482460022, "learning_rate": 0.0002, "epoch": 5.206030150753769, "step": 7770}, {"loss": 1.2975, "grad_norm": 0.9441686868667603, "learning_rate": 0.0002, "epoch": 5.2127303182579565, "step": 7780}, {"loss": 1.256, "grad_norm": 0.9114066362380981, "learning_rate": 0.0002, "epoch": 5.219430485762144, "step": 7790}, {"loss": 1.2621, "grad_norm": 0.9851446151733398, "learning_rate": 0.0002, "epoch": 5.226130653266332, "step": 7800}, {"loss": 1.2502, "grad_norm": 0.9526297450065613, "learning_rate": 0.0002, "epoch": 5.232830820770519, "step": 7810}, {"loss": 1.1502, "grad_norm": 1.05986487865448, "learning_rate": 0.0002, "epoch": 5.239530988274707, "step": 7820}, {"loss": 1.2517, "grad_norm": 0.8956538438796997, "learning_rate": 0.0002, "epoch": 5.2462311557788945, "step": 7830}, {"loss": 1.2556, "grad_norm": 0.9568153619766235, "learning_rate": 0.0002, "epoch": 5.252931323283082, "step": 7840}, {"loss": 1.2442, "grad_norm": 1.0035018920898438, "learning_rate": 0.0002, "epoch": 5.259631490787269, "step": 7850}, {"loss": 1.2605, "grad_norm": 0.8554368615150452, "learning_rate": 0.0002, "epoch": 5.266331658291457, "step": 7860}, {"loss": 1.2799, "grad_norm": 0.9677708148956299, "learning_rate": 0.0002, "epoch": 5.273031825795645, "step": 7870}, {"loss": 1.275, "grad_norm": 0.943606436252594, "learning_rate": 0.0002, "epoch": 5.279731993299833, "step": 7880}, {"loss": 1.2335, "grad_norm": 1.0029335021972656, "learning_rate": 0.0002, "epoch": 5.28643216080402, "step": 7890}, {"loss": 1.2494, "grad_norm": 1.0164015293121338, "learning_rate": 0.0002, "epoch": 5.293132328308207, "step": 7900}, {"loss": 1.3117, "grad_norm": 0.8908365368843079, "learning_rate": 0.0002, "epoch": 5.299832495812395, "step": 7910}, {"loss": 1.2832, "grad_norm": 0.9307826161384583, "learning_rate": 0.0002, "epoch": 5.306532663316583, "step": 7920}, {"loss": 1.242, "grad_norm": 1.0730371475219727, "learning_rate": 0.0002, "epoch": 5.313232830820771, "step": 7930}, {"loss": 1.2003, "grad_norm": 0.844739556312561, "learning_rate": 0.0002, "epoch": 5.319932998324958, "step": 7940}, {"loss": 1.2688, "grad_norm": 1.275833010673523, "learning_rate": 0.0002, "epoch": 5.326633165829146, "step": 7950}, {"loss": 1.2957, "grad_norm": 0.9042661190032959, "learning_rate": 0.0002, "epoch": 5.333333333333333, "step": 7960}, {"loss": 1.2912, "grad_norm": 0.9374269247055054, "learning_rate": 0.0002, "epoch": 5.340033500837521, "step": 7970}, {"loss": 1.2721, "grad_norm": 1.033098578453064, "learning_rate": 0.0002, "epoch": 5.346733668341709, "step": 7980}, {"loss": 1.3208, "grad_norm": 1.062775731086731, "learning_rate": 0.0002, "epoch": 5.353433835845896, "step": 7990}, {"loss": 1.3065, "grad_norm": 1.1064317226409912, "learning_rate": 0.0002, "epoch": 5.360134003350084, "step": 8000}, {"loss": 1.2341, "grad_norm": 1.1114039421081543, "learning_rate": 0.0002, "epoch": 5.366834170854271, "step": 8010}, {"loss": 1.2255, "grad_norm": 1.0198014974594116, "learning_rate": 0.0002, "epoch": 5.373534338358459, "step": 8020}, {"loss": 1.2433, "grad_norm": 0.8443173170089722, "learning_rate": 0.0002, "epoch": 5.380234505862647, "step": 8030}, {"loss": 1.206, "grad_norm": 1.000881314277649, "learning_rate": 0.0002, "epoch": 5.386934673366834, "step": 8040}, {"loss": 1.2982, "grad_norm": 0.9874443411827087, "learning_rate": 0.0002, "epoch": 5.393634840871022, "step": 8050}, {"loss": 1.2289, "grad_norm": 0.9895344972610474, "learning_rate": 0.0002, "epoch": 5.400335008375209, "step": 8060}, {"loss": 1.249, "grad_norm": 0.8595236539840698, "learning_rate": 0.0002, "epoch": 5.407035175879397, "step": 8070}, {"loss": 1.2308, "grad_norm": 0.9523849487304688, "learning_rate": 0.0002, "epoch": 5.413735343383585, "step": 8080}, {"loss": 1.2343, "grad_norm": 1.0560476779937744, "learning_rate": 0.0002, "epoch": 5.420435510887772, "step": 8090}, {"loss": 1.2956, "grad_norm": 1.0893689393997192, "learning_rate": 0.0002, "epoch": 5.42713567839196, "step": 8100}, {"loss": 1.2846, "grad_norm": 0.9395513534545898, "learning_rate": 0.0002, "epoch": 5.433835845896147, "step": 8110}, {"loss": 1.3444, "grad_norm": 0.9364215135574341, "learning_rate": 0.0002, "epoch": 5.440536013400335, "step": 8120}, {"loss": 1.2944, "grad_norm": 0.9502208232879639, "learning_rate": 0.0002, "epoch": 5.447236180904523, "step": 8130}, {"loss": 1.2971, "grad_norm": 0.9559208154678345, "learning_rate": 0.0002, "epoch": 5.45393634840871, "step": 8140}, {"loss": 1.2495, "grad_norm": 0.9261730313301086, "learning_rate": 0.0002, "epoch": 5.460636515912898, "step": 8150}, {"loss": 1.2599, "grad_norm": 0.9832326173782349, "learning_rate": 0.0002, "epoch": 5.467336683417085, "step": 8160}, {"loss": 1.2771, "grad_norm": 1.065953016281128, "learning_rate": 0.0002, "epoch": 5.474036850921273, "step": 8170}, {"loss": 1.3617, "grad_norm": 0.9139469861984253, "learning_rate": 0.0002, "epoch": 5.480737018425461, "step": 8180}, {"loss": 1.2197, "grad_norm": 1.2322484254837036, "learning_rate": 0.0002, "epoch": 5.4874371859296485, "step": 8190}, {"loss": 1.2879, "grad_norm": 0.9722974896430969, "learning_rate": 0.0002, "epoch": 5.494137353433836, "step": 8200}, {"loss": 1.2664, "grad_norm": 0.9338926076889038, "learning_rate": 0.0002, "epoch": 5.500837520938023, "step": 8210}, {"loss": 1.2128, "grad_norm": 0.9283728003501892, "learning_rate": 0.0002, "epoch": 5.507537688442211, "step": 8220}, {"loss": 1.2141, "grad_norm": 1.0489585399627686, "learning_rate": 0.0002, "epoch": 5.514237855946399, "step": 8230}, {"loss": 1.3257, "grad_norm": 0.9881814122200012, "learning_rate": 0.0002, "epoch": 5.5209380234505865, "step": 8240}, {"loss": 1.2221, "grad_norm": 0.9274460077285767, "learning_rate": 0.0002, "epoch": 5.527638190954773, "step": 8250}, {"loss": 1.2241, "grad_norm": 0.8650718331336975, "learning_rate": 0.0002, "epoch": 5.534338358458961, "step": 8260}, {"loss": 1.2462, "grad_norm": 1.014069676399231, "learning_rate": 0.0002, "epoch": 5.541038525963149, "step": 8270}, {"loss": 1.3502, "grad_norm": 0.9212974905967712, "learning_rate": 0.0002, "epoch": 5.547738693467337, "step": 8280}, {"loss": 1.2779, "grad_norm": 1.1235398054122925, "learning_rate": 0.0002, "epoch": 5.5544388609715245, "step": 8290}, {"loss": 1.306, "grad_norm": 0.961954653263092, "learning_rate": 0.0002, "epoch": 5.561139028475711, "step": 8300}, {"loss": 1.2946, "grad_norm": 0.9386700391769409, "learning_rate": 0.0002, "epoch": 5.567839195979899, "step": 8310}, {"loss": 1.313, "grad_norm": 1.01912522315979, "learning_rate": 0.0002, "epoch": 5.574539363484087, "step": 8320}, {"loss": 1.3121, "grad_norm": 0.9851216077804565, "learning_rate": 0.0002, "epoch": 5.581239530988275, "step": 8330}, {"loss": 1.3071, "grad_norm": 1.0138001441955566, "learning_rate": 0.0002, "epoch": 5.5879396984924625, "step": 8340}, {"loss": 1.2257, "grad_norm": 0.9262447357177734, "learning_rate": 0.0002, "epoch": 5.594639865996649, "step": 8350}, {"loss": 1.2473, "grad_norm": 1.1322970390319824, "learning_rate": 0.0002, "epoch": 5.601340033500837, "step": 8360}, {"loss": 1.3098, "grad_norm": 1.1429349184036255, "learning_rate": 0.0002, "epoch": 5.608040201005025, "step": 8370}, {"loss": 1.2686, "grad_norm": 0.9130118489265442, "learning_rate": 0.0002, "epoch": 5.614740368509213, "step": 8380}, {"loss": 1.2541, "grad_norm": 0.9651545882225037, "learning_rate": 0.0002, "epoch": 5.6214405360134005, "step": 8390}, {"loss": 1.2799, "grad_norm": 0.9595398306846619, "learning_rate": 0.0002, "epoch": 5.628140703517588, "step": 8400}, {"loss": 1.3429, "grad_norm": 1.0049372911453247, "learning_rate": 0.0002, "epoch": 5.634840871021775, "step": 8410}, {"loss": 1.3224, "grad_norm": 1.082804560661316, "learning_rate": 0.0002, "epoch": 5.641541038525963, "step": 8420}, {"loss": 1.297, "grad_norm": 0.9489204287528992, "learning_rate": 0.0002, "epoch": 5.648241206030151, "step": 8430}, {"loss": 1.3424, "grad_norm": 0.9470235109329224, "learning_rate": 0.0002, "epoch": 5.654941373534339, "step": 8440}, {"loss": 1.3358, "grad_norm": 1.0662927627563477, "learning_rate": 0.0002, "epoch": 5.661641541038526, "step": 8450}, {"loss": 1.2973, "grad_norm": 0.9097877740859985, "learning_rate": 0.0002, "epoch": 5.668341708542713, "step": 8460}, {"loss": 1.3072, "grad_norm": 0.9740368127822876, "learning_rate": 0.0002, "epoch": 5.675041876046901, "step": 8470}, {"loss": 1.286, "grad_norm": 0.9878810048103333, "learning_rate": 0.0002, "epoch": 5.681742043551089, "step": 8480}, {"loss": 1.208, "grad_norm": 1.148260474205017, "learning_rate": 0.0002, "epoch": 5.688442211055277, "step": 8490}, {"loss": 1.2842, "grad_norm": 0.9632558822631836, "learning_rate": 0.0002, "epoch": 5.695142378559464, "step": 8500}, {"loss": 1.2787, "grad_norm": 0.876812756061554, "learning_rate": 0.0002, "epoch": 5.701842546063651, "step": 8510}, {"loss": 1.3186, "grad_norm": 1.0730829238891602, "learning_rate": 0.0002, "epoch": 5.708542713567839, "step": 8520}, {"loss": 1.2856, "grad_norm": 1.2239218950271606, "learning_rate": 0.0002, "epoch": 5.715242881072027, "step": 8530}, {"loss": 1.2717, "grad_norm": 0.9460835456848145, "learning_rate": 0.0002, "epoch": 5.721943048576215, "step": 8540}, {"loss": 1.3509, "grad_norm": 0.9086270928382874, "learning_rate": 0.0002, "epoch": 5.728643216080402, "step": 8550}, {"loss": 1.2971, "grad_norm": 1.0258867740631104, "learning_rate": 0.0002, "epoch": 5.735343383584589, "step": 8560}, {"loss": 1.3581, "grad_norm": 1.0543923377990723, "learning_rate": 0.0002, "epoch": 5.742043551088777, "step": 8570}, {"loss": 1.2988, "grad_norm": 0.9063900113105774, "learning_rate": 0.0002, "epoch": 5.748743718592965, "step": 8580}, {"loss": 1.3535, "grad_norm": 1.1838830709457397, "learning_rate": 0.0002, "epoch": 5.755443886097153, "step": 8590}, {"loss": 1.2655, "grad_norm": 0.9631859064102173, "learning_rate": 0.0002, "epoch": 5.76214405360134, "step": 8600}, {"loss": 1.276, "grad_norm": 0.9702655673027039, "learning_rate": 0.0002, "epoch": 5.768844221105527, "step": 8610}, {"loss": 1.3196, "grad_norm": 1.0591435432434082, "learning_rate": 0.0002, "epoch": 5.775544388609715, "step": 8620}, {"loss": 1.267, "grad_norm": 0.9989570379257202, "learning_rate": 0.0002, "epoch": 5.782244556113903, "step": 8630}, {"loss": 1.3227, "grad_norm": 1.0836435556411743, "learning_rate": 0.0002, "epoch": 5.788944723618091, "step": 8640}, {"loss": 1.3334, "grad_norm": 0.8832896947860718, "learning_rate": 0.0002, "epoch": 5.795644891122278, "step": 8650}, {"loss": 1.3214, "grad_norm": 1.0104607343673706, "learning_rate": 0.0002, "epoch": 5.802345058626465, "step": 8660}, {"loss": 1.2703, "grad_norm": 0.8375084400177002, "learning_rate": 0.0002, "epoch": 5.809045226130653, "step": 8670}, {"loss": 1.3554, "grad_norm": 1.1300716400146484, "learning_rate": 0.0002, "epoch": 5.815745393634841, "step": 8680}, {"loss": 1.3468, "grad_norm": 0.9311910271644592, "learning_rate": 0.0002, "epoch": 5.822445561139029, "step": 8690}, {"loss": 1.2749, "grad_norm": 0.9488391876220703, "learning_rate": 0.0002, "epoch": 5.8291457286432165, "step": 8700}, {"loss": 1.2281, "grad_norm": 0.9747629761695862, "learning_rate": 0.0002, "epoch": 5.835845896147403, "step": 8710}, {"loss": 1.2923, "grad_norm": 1.1029598712921143, "learning_rate": 0.0002, "epoch": 5.842546063651591, "step": 8720}, {"loss": 1.3613, "grad_norm": 1.0396875143051147, "learning_rate": 0.0002, "epoch": 5.849246231155779, "step": 8730}, {"loss": 1.3272, "grad_norm": 0.9259780645370483, "learning_rate": 0.0002, "epoch": 5.855946398659967, "step": 8740}, {"loss": 1.3236, "grad_norm": 1.020033597946167, "learning_rate": 0.0002, "epoch": 5.8626465661641545, "step": 8750}, {"loss": 1.3453, "grad_norm": 0.9191218614578247, "learning_rate": 0.0002, "epoch": 5.869346733668341, "step": 8760}, {"loss": 1.3012, "grad_norm": 1.1093107461929321, "learning_rate": 0.0002, "epoch": 5.876046901172529, "step": 8770}, {"loss": 1.2718, "grad_norm": 1.1626793146133423, "learning_rate": 0.0002, "epoch": 5.882747068676717, "step": 8780}, {"loss": 1.2969, "grad_norm": 0.9542945027351379, "learning_rate": 0.0002, "epoch": 5.889447236180905, "step": 8790}, {"loss": 1.3134, "grad_norm": 0.9086058139801025, "learning_rate": 0.0002, "epoch": 5.8961474036850925, "step": 8800}, {"loss": 1.2731, "grad_norm": 0.9249639511108398, "learning_rate": 0.0002, "epoch": 5.902847571189279, "step": 8810}, {"loss": 1.337, "grad_norm": 0.9414396286010742, "learning_rate": 0.0002, "epoch": 5.909547738693467, "step": 8820}, {"loss": 1.2865, "grad_norm": 0.9086037874221802, "learning_rate": 0.0002, "epoch": 5.916247906197655, "step": 8830}, {"loss": 1.2756, "grad_norm": 0.8685907125473022, "learning_rate": 0.0002, "epoch": 5.922948073701843, "step": 8840}, {"loss": 1.297, "grad_norm": 1.036419153213501, "learning_rate": 0.0002, "epoch": 5.9296482412060305, "step": 8850}, {"loss": 1.3207, "grad_norm": 1.0183674097061157, "learning_rate": 0.0002, "epoch": 5.936348408710217, "step": 8860}, {"loss": 1.3922, "grad_norm": 0.966444194316864, "learning_rate": 0.0002, "epoch": 5.943048576214405, "step": 8870}, {"loss": 1.333, "grad_norm": 1.125693917274475, "learning_rate": 0.0002, "epoch": 5.949748743718593, "step": 8880}, {"loss": 1.3116, "grad_norm": 0.9857436418533325, "learning_rate": 0.0002, "epoch": 5.956448911222781, "step": 8890}, {"loss": 1.2526, "grad_norm": 0.9377069473266602, "learning_rate": 0.0002, "epoch": 5.9631490787269685, "step": 8900}, {"loss": 1.3221, "grad_norm": 0.9493814706802368, "learning_rate": 0.0002, "epoch": 5.969849246231155, "step": 8910}, {"loss": 1.2516, "grad_norm": 0.8806208372116089, "learning_rate": 0.0002, "epoch": 5.976549413735343, "step": 8920}, {"loss": 1.2558, "grad_norm": 0.8727600574493408, "learning_rate": 0.0002, "epoch": 5.983249581239531, "step": 8930}, {"loss": 1.3538, "grad_norm": 0.9799810647964478, "learning_rate": 0.0002, "epoch": 5.989949748743719, "step": 8940}, {"loss": 1.3323, "grad_norm": 0.9866513609886169, "learning_rate": 0.0002, "epoch": 5.9966499162479066, "step": 8950}, {"eval_loss": 2.0282373428344727, "eval_runtime": 38.0375, "eval_samples_per_second": 13.539, "eval_steps_per_second": 1.709, "epoch": 6.0, "step": 8955}, {"loss": 1.1768, "grad_norm": 0.8747885227203369, "learning_rate": 0.0002, "epoch": 6.0033500837520934, "step": 8960}, {"loss": 1.0677, "grad_norm": 1.2512741088867188, "learning_rate": 0.0002, "epoch": 6.010050251256281, "step": 8970}, {"loss": 1.1128, "grad_norm": 1.06855309009552, "learning_rate": 0.0002, "epoch": 6.016750418760469, "step": 8980}, {"loss": 1.1382, "grad_norm": 1.1868711709976196, "learning_rate": 0.0002, "epoch": 6.023450586264657, "step": 8990}, {"loss": 1.1377, "grad_norm": 1.2984495162963867, "learning_rate": 0.0002, "epoch": 6.030150753768845, "step": 9000}, {"loss": 1.0803, "grad_norm": 1.1147589683532715, "learning_rate": 0.0002, "epoch": 6.0368509212730315, "step": 9010}, {"loss": 1.1244, "grad_norm": 1.3128414154052734, "learning_rate": 0.0002, "epoch": 6.043551088777219, "step": 9020}, {"loss": 1.097, "grad_norm": 1.068290114402771, "learning_rate": 0.0002, "epoch": 6.050251256281407, "step": 9030}, {"loss": 1.1764, "grad_norm": 1.1890562772750854, "learning_rate": 0.0002, "epoch": 6.056951423785595, "step": 9040}, {"loss": 1.1239, "grad_norm": 1.2121573686599731, "learning_rate": 0.0002, "epoch": 6.063651591289783, "step": 9050}, {"loss": 1.0683, "grad_norm": 1.0860483646392822, "learning_rate": 0.0002, "epoch": 6.0703517587939695, "step": 9060}, {"loss": 1.1613, "grad_norm": 1.1214599609375, "learning_rate": 0.0002, "epoch": 6.077051926298157, "step": 9070}, {"loss": 1.1001, "grad_norm": 1.147580862045288, "learning_rate": 0.0002, "epoch": 6.083752093802345, "step": 9080}, {"loss": 1.1154, "grad_norm": 1.3233155012130737, "learning_rate": 0.0002, "epoch": 6.090452261306533, "step": 9090}, {"loss": 1.1017, "grad_norm": 1.1869080066680908, "learning_rate": 0.0002, "epoch": 6.097152428810721, "step": 9100}, {"loss": 1.1532, "grad_norm": 1.1695014238357544, "learning_rate": 0.0002, "epoch": 6.1038525963149075, "step": 9110}, {"loss": 1.1366, "grad_norm": 1.1982251405715942, "learning_rate": 0.0002, "epoch": 6.110552763819095, "step": 9120}, {"loss": 1.0805, "grad_norm": 1.1426950693130493, "learning_rate": 0.0002, "epoch": 6.117252931323283, "step": 9130}, {"loss": 1.0801, "grad_norm": 1.2257394790649414, "learning_rate": 0.0002, "epoch": 6.123953098827471, "step": 9140}, {"loss": 1.1209, "grad_norm": 1.2932263612747192, "learning_rate": 0.0002, "epoch": 6.130653266331659, "step": 9150}, {"loss": 1.0934, "grad_norm": 1.2617030143737793, "learning_rate": 0.0002, "epoch": 6.1373534338358455, "step": 9160}, {"loss": 1.0551, "grad_norm": 1.1201422214508057, "learning_rate": 0.0002, "epoch": 6.144053601340033, "step": 9170}, {"loss": 1.1059, "grad_norm": 0.9625319838523865, "learning_rate": 0.0002, "epoch": 6.150753768844221, "step": 9180}, {"loss": 1.1397, "grad_norm": 1.0290048122406006, "learning_rate": 0.0002, "epoch": 6.157453936348409, "step": 9190}, {"loss": 1.1257, "grad_norm": 1.1137803792953491, "learning_rate": 0.0002, "epoch": 6.164154103852597, "step": 9200}, {"loss": 1.1211, "grad_norm": 1.3674522638320923, "learning_rate": 0.0002, "epoch": 6.1708542713567835, "step": 9210}, {"loss": 1.0947, "grad_norm": 1.182207703590393, "learning_rate": 0.0002, "epoch": 6.177554438860971, "step": 9220}, {"loss": 1.0838, "grad_norm": 1.0496711730957031, "learning_rate": 0.0002, "epoch": 6.184254606365159, "step": 9230}, {"loss": 1.0666, "grad_norm": 1.1899489164352417, "learning_rate": 0.0002, "epoch": 6.190954773869347, "step": 9240}, {"loss": 1.1633, "grad_norm": 1.2666147947311401, "learning_rate": 0.0002, "epoch": 6.197654941373535, "step": 9250}, {"loss": 1.1532, "grad_norm": 1.2013030052185059, "learning_rate": 0.0002, "epoch": 6.204355108877722, "step": 9260}, {"loss": 1.151, "grad_norm": 1.3049768209457397, "learning_rate": 0.0002, "epoch": 6.211055276381909, "step": 9270}, {"loss": 1.0741, "grad_norm": 1.1733006238937378, "learning_rate": 0.0002, "epoch": 6.217755443886097, "step": 9280}, {"loss": 1.0933, "grad_norm": 1.2742516994476318, "learning_rate": 0.0002, "epoch": 6.224455611390285, "step": 9290}, {"loss": 1.1028, "grad_norm": 1.110198974609375, "learning_rate": 0.0002, "epoch": 6.231155778894473, "step": 9300}, {"loss": 1.1619, "grad_norm": 1.159963607788086, "learning_rate": 0.0002, "epoch": 6.23785594639866, "step": 9310}, {"loss": 1.0716, "grad_norm": 1.302216649055481, "learning_rate": 0.0002, "epoch": 6.244556113902847, "step": 9320}, {"loss": 1.0694, "grad_norm": 1.2134063243865967, "learning_rate": 0.0002, "epoch": 6.251256281407035, "step": 9330}, {"loss": 1.2151, "grad_norm": 1.062682867050171, "learning_rate": 0.0002, "epoch": 6.257956448911223, "step": 9340}, {"loss": 1.148, "grad_norm": 1.1568971872329712, "learning_rate": 0.0002, "epoch": 6.264656616415411, "step": 9350}, {"loss": 1.094, "grad_norm": 0.9914957880973816, "learning_rate": 0.0002, "epoch": 6.271356783919598, "step": 9360}, {"loss": 1.125, "grad_norm": 1.017250895500183, "learning_rate": 0.0002, "epoch": 6.278056951423785, "step": 9370}, {"loss": 1.2177, "grad_norm": 1.1862834692001343, "learning_rate": 0.0002, "epoch": 6.284757118927973, "step": 9380}, {"loss": 0.9994, "grad_norm": 1.2834911346435547, "learning_rate": 0.0002, "epoch": 6.291457286432161, "step": 9390}, {"loss": 1.0922, "grad_norm": 1.3306856155395508, "learning_rate": 0.0002, "epoch": 6.298157453936349, "step": 9400}, {"loss": 1.1136, "grad_norm": 1.12908136844635, "learning_rate": 0.0002, "epoch": 6.304857621440536, "step": 9410}, {"loss": 1.1406, "grad_norm": 1.2157351970672607, "learning_rate": 0.0002, "epoch": 6.311557788944723, "step": 9420}, {"loss": 1.1388, "grad_norm": 1.121882677078247, "learning_rate": 0.0002, "epoch": 6.318257956448911, "step": 9430}, {"loss": 1.1648, "grad_norm": 1.3144481182098389, "learning_rate": 0.0002, "epoch": 6.324958123953099, "step": 9440}, {"loss": 1.1228, "grad_norm": 1.1946896314620972, "learning_rate": 0.0002, "epoch": 6.331658291457287, "step": 9450}, {"loss": 1.1613, "grad_norm": 1.1289668083190918, "learning_rate": 0.0002, "epoch": 6.338358458961474, "step": 9460}, {"loss": 1.1059, "grad_norm": 1.1065658330917358, "learning_rate": 0.0002, "epoch": 6.345058626465661, "step": 9470}, {"loss": 1.1431, "grad_norm": 1.0881422758102417, "learning_rate": 0.0002, "epoch": 6.351758793969849, "step": 9480}, {"loss": 1.223, "grad_norm": 1.242676854133606, "learning_rate": 0.0002, "epoch": 6.358458961474037, "step": 9490}, {"loss": 1.1379, "grad_norm": 0.9650855660438538, "learning_rate": 0.0002, "epoch": 6.365159128978225, "step": 9500}, {"loss": 1.0763, "grad_norm": 1.2845722436904907, "learning_rate": 0.0002, "epoch": 6.371859296482412, "step": 9510}, {"loss": 1.1351, "grad_norm": 1.0327043533325195, "learning_rate": 0.0002, "epoch": 6.3785594639865995, "step": 9520}, {"loss": 1.114, "grad_norm": 1.0780898332595825, "learning_rate": 0.0002, "epoch": 6.385259631490787, "step": 9530}, {"loss": 1.1579, "grad_norm": 1.4934027194976807, "learning_rate": 0.0002, "epoch": 6.391959798994975, "step": 9540}, {"loss": 1.1546, "grad_norm": 0.9882908463478088, "learning_rate": 0.0002, "epoch": 6.398659966499163, "step": 9550}, {"loss": 1.1145, "grad_norm": 1.3250664472579956, "learning_rate": 0.0002, "epoch": 6.40536013400335, "step": 9560}, {"loss": 1.2333, "grad_norm": 1.1888482570648193, "learning_rate": 0.0002, "epoch": 6.4120603015075375, "step": 9570}, {"loss": 1.0892, "grad_norm": 1.136496901512146, "learning_rate": 0.0002, "epoch": 6.418760469011725, "step": 9580}, {"loss": 1.1674, "grad_norm": 1.161360502243042, "learning_rate": 0.0002, "epoch": 6.425460636515913, "step": 9590}, {"loss": 1.1293, "grad_norm": 1.2034236192703247, "learning_rate": 0.0002, "epoch": 6.432160804020101, "step": 9600}, {"loss": 1.1059, "grad_norm": 1.0268361568450928, "learning_rate": 0.0002, "epoch": 6.438860971524288, "step": 9610}, {"loss": 1.1732, "grad_norm": 1.2132930755615234, "learning_rate": 0.0002, "epoch": 6.4455611390284755, "step": 9620}, {"loss": 1.1329, "grad_norm": 1.0773013830184937, "learning_rate": 0.0002, "epoch": 6.452261306532663, "step": 9630}, {"loss": 1.0822, "grad_norm": 1.3848375082015991, "learning_rate": 0.0002, "epoch": 6.458961474036851, "step": 9640}, {"loss": 1.1778, "grad_norm": 1.110495924949646, "learning_rate": 0.0002, "epoch": 6.465661641541039, "step": 9650}, {"loss": 1.2022, "grad_norm": 1.118093729019165, "learning_rate": 0.0002, "epoch": 6.472361809045226, "step": 9660}, {"loss": 1.1222, "grad_norm": 1.2611900568008423, "learning_rate": 0.0002, "epoch": 6.4790619765494135, "step": 9670}, {"loss": 1.2138, "grad_norm": 0.971754252910614, "learning_rate": 0.0002, "epoch": 6.485762144053601, "step": 9680}, {"loss": 1.1641, "grad_norm": 1.2615419626235962, "learning_rate": 0.0002, "epoch": 6.492462311557789, "step": 9690}, {"loss": 1.1412, "grad_norm": 1.1370900869369507, "learning_rate": 0.0002, "epoch": 6.499162479061977, "step": 9700}, {"loss": 1.186, "grad_norm": 1.1815906763076782, "learning_rate": 0.0002, "epoch": 6.505862646566165, "step": 9710}, {"loss": 1.167, "grad_norm": 1.3424339294433594, "learning_rate": 0.0002, "epoch": 6.5125628140703515, "step": 9720}, {"loss": 1.1602, "grad_norm": 1.2858397960662842, "learning_rate": 0.0002, "epoch": 6.519262981574539, "step": 9730}, {"loss": 1.178, "grad_norm": 0.9578179121017456, "learning_rate": 0.0002, "epoch": 6.525963149078727, "step": 9740}, {"loss": 1.1805, "grad_norm": 1.3105167150497437, "learning_rate": 0.0002, "epoch": 6.532663316582915, "step": 9750}, {"loss": 1.1899, "grad_norm": 1.0586575269699097, "learning_rate": 0.0002, "epoch": 6.539363484087103, "step": 9760}, {"loss": 1.095, "grad_norm": 1.2122068405151367, "learning_rate": 0.0002, "epoch": 6.54606365159129, "step": 9770}, {"loss": 1.1471, "grad_norm": 1.3088626861572266, "learning_rate": 0.0002, "epoch": 6.552763819095477, "step": 9780}, {"loss": 1.1067, "grad_norm": 1.194122076034546, "learning_rate": 0.0002, "epoch": 6.559463986599665, "step": 9790}, {"loss": 1.0967, "grad_norm": 1.1508387327194214, "learning_rate": 0.0002, "epoch": 6.566164154103853, "step": 9800}, {"loss": 1.1694, "grad_norm": 1.109228253364563, "learning_rate": 0.0002, "epoch": 6.572864321608041, "step": 9810}, {"loss": 1.1378, "grad_norm": 1.1607427597045898, "learning_rate": 0.0002, "epoch": 6.579564489112228, "step": 9820}, {"loss": 1.1585, "grad_norm": 1.174089789390564, "learning_rate": 0.0002, "epoch": 6.586264656616415, "step": 9830}, {"loss": 1.1385, "grad_norm": 1.1739521026611328, "learning_rate": 0.0002, "epoch": 6.592964824120603, "step": 9840}, {"loss": 1.155, "grad_norm": 1.098528504371643, "learning_rate": 0.0002, "epoch": 6.599664991624791, "step": 9850}, {"loss": 1.1359, "grad_norm": 1.0397740602493286, "learning_rate": 0.0002, "epoch": 6.606365159128979, "step": 9860}, {"loss": 1.1433, "grad_norm": 1.1087969541549683, "learning_rate": 0.0002, "epoch": 6.613065326633166, "step": 9870}, {"loss": 1.2356, "grad_norm": 1.2070481777191162, "learning_rate": 0.0002, "epoch": 6.619765494137353, "step": 9880}, {"loss": 1.1161, "grad_norm": 1.1115655899047852, "learning_rate": 0.0002, "epoch": 6.626465661641541, "step": 9890}, {"loss": 1.2163, "grad_norm": 1.2486097812652588, "learning_rate": 0.0002, "epoch": 6.633165829145729, "step": 9900}, {"loss": 1.0984, "grad_norm": 1.230380654335022, "learning_rate": 0.0002, "epoch": 6.639865996649917, "step": 9910}, {"loss": 1.1862, "grad_norm": 1.1479365825653076, "learning_rate": 0.0002, "epoch": 6.646566164154104, "step": 9920}, {"loss": 1.1139, "grad_norm": 1.0790960788726807, "learning_rate": 0.0002, "epoch": 6.653266331658291, "step": 9930}, {"loss": 1.2001, "grad_norm": 1.1157397031784058, "learning_rate": 0.0002, "epoch": 6.659966499162479, "step": 9940}, {"loss": 1.1085, "grad_norm": 1.3104028701782227, "learning_rate": 0.0002, "epoch": 6.666666666666667, "step": 9950}, {"loss": 1.2209, "grad_norm": 1.1727646589279175, "learning_rate": 0.0002, "epoch": 6.673366834170855, "step": 9960}, {"loss": 1.1671, "grad_norm": 1.2104284763336182, "learning_rate": 0.0002, "epoch": 6.680067001675042, "step": 9970}, {"loss": 1.1952, "grad_norm": 1.2023727893829346, "learning_rate": 0.0002, "epoch": 6.686767169179229, "step": 9980}, {"loss": 1.1385, "grad_norm": 1.0088225603103638, "learning_rate": 0.0002, "epoch": 6.693467336683417, "step": 9990}, {"loss": 1.1314, "grad_norm": 1.298015832901001, "learning_rate": 0.0002, "epoch": 6.700167504187605, "step": 10000}, {"loss": 1.1983, "grad_norm": 1.1315910816192627, "learning_rate": 0.0002, "epoch": 6.706867671691793, "step": 10010}, {"loss": 1.1679, "grad_norm": 1.1283273696899414, "learning_rate": 0.0002, "epoch": 6.71356783919598, "step": 10020}, {"loss": 1.1917, "grad_norm": 1.2564418315887451, "learning_rate": 0.0002, "epoch": 6.720268006700167, "step": 10030}, {"loss": 1.1788, "grad_norm": 1.0451353788375854, "learning_rate": 0.0002, "epoch": 6.726968174204355, "step": 10040}, {"loss": 1.1905, "grad_norm": 1.054793357849121, "learning_rate": 0.0002, "epoch": 6.733668341708543, "step": 10050}, {"loss": 1.1814, "grad_norm": 1.2741243839263916, "learning_rate": 0.0002, "epoch": 6.740368509212731, "step": 10060}, {"loss": 1.2015, "grad_norm": 1.1342514753341675, "learning_rate": 0.0002, "epoch": 6.747068676716918, "step": 10070}, {"loss": 1.2587, "grad_norm": 1.0081498622894287, "learning_rate": 0.0002, "epoch": 6.7537688442211055, "step": 10080}, {"loss": 1.1226, "grad_norm": 1.2164603471755981, "learning_rate": 0.0002, "epoch": 6.760469011725293, "step": 10090}, {"loss": 1.1353, "grad_norm": 1.2062463760375977, "learning_rate": 0.0002, "epoch": 6.767169179229481, "step": 10100}, {"loss": 1.2143, "grad_norm": 1.2255526781082153, "learning_rate": 0.0002, "epoch": 6.773869346733669, "step": 10110}, {"loss": 1.1544, "grad_norm": 1.08175790309906, "learning_rate": 0.0002, "epoch": 6.780569514237856, "step": 10120}, {"loss": 1.1983, "grad_norm": 1.5781128406524658, "learning_rate": 0.0002, "epoch": 6.7872696817420435, "step": 10130}, {"loss": 1.0994, "grad_norm": 1.0622451305389404, "learning_rate": 0.0002, "epoch": 6.793969849246231, "step": 10140}, {"loss": 1.2084, "grad_norm": 1.1591497659683228, "learning_rate": 0.0002, "epoch": 6.800670016750419, "step": 10150}, {"loss": 1.2203, "grad_norm": 1.0398483276367188, "learning_rate": 0.0002, "epoch": 6.807370184254607, "step": 10160}, {"loss": 1.2249, "grad_norm": 1.229132056236267, "learning_rate": 0.0002, "epoch": 6.814070351758794, "step": 10170}, {"loss": 1.1789, "grad_norm": 1.0918090343475342, "learning_rate": 0.0002, "epoch": 6.8207705192629815, "step": 10180}, {"loss": 1.1639, "grad_norm": 1.1543749570846558, "learning_rate": 0.0002, "epoch": 6.827470686767169, "step": 10190}, {"loss": 1.1353, "grad_norm": 1.1831817626953125, "learning_rate": 0.0002, "epoch": 6.834170854271357, "step": 10200}, {"loss": 1.2565, "grad_norm": 1.305327296257019, "learning_rate": 0.0002, "epoch": 6.840871021775545, "step": 10210}, {"loss": 1.2037, "grad_norm": 1.136720061302185, "learning_rate": 0.0002, "epoch": 6.847571189279732, "step": 10220}, {"loss": 1.2256, "grad_norm": 1.2282346487045288, "learning_rate": 0.0002, "epoch": 6.8542713567839195, "step": 10230}, {"loss": 1.1281, "grad_norm": 1.2457010746002197, "learning_rate": 0.0002, "epoch": 6.860971524288107, "step": 10240}, {"loss": 1.2762, "grad_norm": 1.2808631658554077, "learning_rate": 0.0002, "epoch": 6.867671691792295, "step": 10250}, {"loss": 1.2213, "grad_norm": 1.089066743850708, "learning_rate": 0.0002, "epoch": 6.874371859296483, "step": 10260}, {"loss": 1.2627, "grad_norm": 0.9543178081512451, "learning_rate": 0.0002, "epoch": 6.88107202680067, "step": 10270}, {"loss": 1.1617, "grad_norm": 1.1149744987487793, "learning_rate": 0.0002, "epoch": 6.8877721943048575, "step": 10280}, {"loss": 1.1134, "grad_norm": 1.0185538530349731, "learning_rate": 0.0002, "epoch": 6.894472361809045, "step": 10290}, {"loss": 1.217, "grad_norm": 0.9954617619514465, "learning_rate": 0.0002, "epoch": 6.901172529313233, "step": 10300}, {"loss": 1.1524, "grad_norm": 1.2581418752670288, "learning_rate": 0.0002, "epoch": 6.907872696817421, "step": 10310}, {"loss": 1.1942, "grad_norm": 1.2430983781814575, "learning_rate": 0.0002, "epoch": 6.914572864321608, "step": 10320}, {"loss": 1.1254, "grad_norm": 1.4937270879745483, "learning_rate": 0.0002, "epoch": 6.921273031825796, "step": 10330}, {"loss": 1.1519, "grad_norm": 1.1257144212722778, "learning_rate": 0.0002, "epoch": 6.927973199329983, "step": 10340}, {"loss": 1.2622, "grad_norm": 1.2068904638290405, "learning_rate": 0.0002, "epoch": 6.934673366834171, "step": 10350}, {"loss": 1.1349, "grad_norm": 1.0290757417678833, "learning_rate": 0.0002, "epoch": 6.941373534338359, "step": 10360}, {"loss": 1.1752, "grad_norm": 1.0070724487304688, "learning_rate": 0.0002, "epoch": 6.948073701842546, "step": 10370}, {"loss": 1.1838, "grad_norm": 0.9936357140541077, "learning_rate": 0.0002, "epoch": 6.954773869346734, "step": 10380}, {"loss": 1.2305, "grad_norm": 1.1063416004180908, "learning_rate": 0.0002, "epoch": 6.961474036850921, "step": 10390}, {"loss": 1.154, "grad_norm": 1.5199986696243286, "learning_rate": 0.0002, "epoch": 6.968174204355109, "step": 10400}, {"loss": 1.1143, "grad_norm": 1.160731554031372, "learning_rate": 0.0002, "epoch": 6.974874371859297, "step": 10410}, {"loss": 1.2132, "grad_norm": 1.084697961807251, "learning_rate": 0.0002, "epoch": 6.981574539363484, "step": 10420}, {"loss": 1.1995, "grad_norm": 1.1257576942443848, "learning_rate": 0.0002, "epoch": 6.988274706867672, "step": 10430}, {"loss": 1.1197, "grad_norm": 1.310616135597229, "learning_rate": 0.0002, "epoch": 6.994974874371859, "step": 10440}]}
+{"epoch": 7.997319932998325, "step": 11936, "epoch_duration": 1594.4848327636719, "total_accumulated_duration": 12784.900410413742, "gpu_info": {"GPU_0": "NVIDIA A100-SXM4-40GB"}, "memory_usage": {"avg_memory_usage": {"GPU_0": 7568.4541015625}, "peak_memory_usage": {"GPU_0": 13792.75439453125}, "avg_memory_reserved": {"GPU_0": 17416.0}, "peak_memory_reserved": {"GPU_0": 17416.0}, "total_memory": {"GPU_0": 40444.375}}, "best_checkpoint_path": "outputs-001/Meta-Llama-3-8B-Instruct_int4_hellaswag-routerbench-0shot_lr-0.0002_e-8_seq-512_lora-a-32-d-0.05-r-64_bs-1_gas-2_tf32-True_tunedata-portion-p-1.0-num-7029-sd-42/checkpoint-2985", "params": {"epochs": 8, "batch_size": 1, "learning_rate": 0.0002, "gradient_accumulation_steps": 2, "warmup_ratio": 0.03, "max_grad_norm": 0.3, "lora_alpha": 32, "lora_dropout": 0.05, "lora_r": 64, "tf32": true, "seed": 42}, "log_history": [{"loss": 2.6189, "grad_norm": 0.565915048122406, "learning_rate": 0.0002, "epoch": 0.006700167504187605, "step": 10}, {"loss": 2.3162, "grad_norm": 0.5004463791847229, "learning_rate": 0.0002, "epoch": 0.01340033500837521, "step": 20}, {"loss": 2.0576, "grad_norm": 0.511043906211853, "learning_rate": 0.0002, "epoch": 0.020100502512562814, "step": 30}, {"loss": 2.0085, "grad_norm": 0.47327178716659546, "learning_rate": 0.0002, "epoch": 0.02680067001675042, "step": 40}, {"loss": 2.0276, "grad_norm": 0.5511676669120789, "learning_rate": 0.0002, "epoch": 0.03350083752093802, "step": 50}, {"loss": 1.9075, "grad_norm": 0.4666278064250946, "learning_rate": 0.0002, "epoch": 0.04020100502512563, "step": 60}, {"loss": 1.8413, "grad_norm": 0.5310961008071899, "learning_rate": 0.0002, "epoch": 0.04690117252931323, "step": 70}, {"loss": 1.8711, "grad_norm": 0.5606027245521545, "learning_rate": 0.0002, "epoch": 0.05360134003350084, "step": 80}, {"loss": 1.9282, "grad_norm": 0.4934779703617096, "learning_rate": 0.0002, "epoch": 0.06030150753768844, "step": 90}, {"loss": 1.8925, "grad_norm": 0.4821869730949402, "learning_rate": 0.0002, "epoch": 0.06700167504187604, "step": 100}, {"loss": 1.8628, "grad_norm": 0.5262084603309631, "learning_rate": 0.0002, "epoch": 0.07370184254606366, "step": 110}, {"loss": 1.8347, "grad_norm": 0.3774230182170868, "learning_rate": 0.0002, "epoch": 0.08040201005025126, "step": 120}, {"loss": 1.8386, "grad_norm": 0.34137430787086487, "learning_rate": 0.0002, "epoch": 0.08710217755443886, "step": 130}, {"loss": 1.861, "grad_norm": 0.407272070646286, "learning_rate": 0.0002, "epoch": 0.09380234505862646, "step": 140}, {"loss": 1.8279, "grad_norm": 0.4011937975883484, "learning_rate": 0.0002, "epoch": 0.10050251256281408, "step": 150}, {"loss": 1.9317, "grad_norm": 0.4432467222213745, "learning_rate": 0.0002, "epoch": 0.10720268006700168, "step": 160}, {"loss": 1.8157, "grad_norm": 0.44030463695526123, "learning_rate": 0.0002, "epoch": 0.11390284757118928, "step": 170}, {"loss": 1.8534, "grad_norm": 0.3799569308757782, "learning_rate": 0.0002, "epoch": 0.12060301507537688, "step": 180}, {"loss": 1.7658, "grad_norm": 0.33721521496772766, "learning_rate": 0.0002, "epoch": 0.1273031825795645, "step": 190}, {"loss": 1.8269, "grad_norm": 0.4096226692199707, "learning_rate": 0.0002, "epoch": 0.13400335008375208, "step": 200}, {"loss": 1.802, "grad_norm": 0.37374693155288696, "learning_rate": 0.0002, "epoch": 0.1407035175879397, "step": 210}, {"loss": 1.8901, "grad_norm": 0.3249480128288269, "learning_rate": 0.0002, "epoch": 0.1474036850921273, "step": 220}, {"loss": 1.8163, "grad_norm": 0.3612042963504791, "learning_rate": 0.0002, "epoch": 0.1541038525963149, "step": 230}, {"loss": 1.7585, "grad_norm": 0.3686671257019043, "learning_rate": 0.0002, "epoch": 0.16080402010050251, "step": 240}, {"loss": 1.8365, "grad_norm": 0.3521044850349426, "learning_rate": 0.0002, "epoch": 0.16750418760469013, "step": 250}, {"loss": 1.8623, "grad_norm": 0.4073677361011505, "learning_rate": 0.0002, "epoch": 0.17420435510887772, "step": 260}, {"loss": 1.8026, "grad_norm": 0.34522193670272827, "learning_rate": 0.0002, "epoch": 0.18090452261306533, "step": 270}, {"loss": 1.8162, "grad_norm": 0.4121900498867035, "learning_rate": 0.0002, "epoch": 0.18760469011725292, "step": 280}, {"loss": 1.7976, "grad_norm": 0.3544778525829315, "learning_rate": 0.0002, "epoch": 0.19430485762144054, "step": 290}, {"loss": 1.8787, "grad_norm": 0.3482133448123932, "learning_rate": 0.0002, "epoch": 0.20100502512562815, "step": 300}, {"loss": 1.8578, "grad_norm": 0.3421826660633087, "learning_rate": 0.0002, "epoch": 0.20770519262981574, "step": 310}, {"loss": 1.8013, "grad_norm": 0.5024696588516235, "learning_rate": 0.0002, "epoch": 0.21440536013400335, "step": 320}, {"loss": 1.8607, "grad_norm": 0.36013063788414, "learning_rate": 0.0002, "epoch": 0.22110552763819097, "step": 330}, {"loss": 1.9075, "grad_norm": 0.3611244857311249, "learning_rate": 0.0002, "epoch": 0.22780569514237856, "step": 340}, {"loss": 1.8128, "grad_norm": 0.39244529604911804, "learning_rate": 0.0002, "epoch": 0.23450586264656617, "step": 350}, {"loss": 1.7885, "grad_norm": 0.3299325704574585, "learning_rate": 0.0002, "epoch": 0.24120603015075376, "step": 360}, {"loss": 1.8028, "grad_norm": 0.3994322419166565, "learning_rate": 0.0002, "epoch": 0.24790619765494137, "step": 370}, {"loss": 1.8321, "grad_norm": 0.3559151887893677, "learning_rate": 0.0002, "epoch": 0.254606365159129, "step": 380}, {"loss": 1.7802, "grad_norm": 0.3873756229877472, "learning_rate": 0.0002, "epoch": 0.2613065326633166, "step": 390}, {"loss": 1.7844, "grad_norm": 0.3710744082927704, "learning_rate": 0.0002, "epoch": 0.26800670016750416, "step": 400}, {"loss": 1.7789, "grad_norm": 0.3618465065956116, "learning_rate": 0.0002, "epoch": 0.2747068676716918, "step": 410}, {"loss": 1.8529, "grad_norm": 0.30063769221305847, "learning_rate": 0.0002, "epoch": 0.2814070351758794, "step": 420}, {"loss": 1.7765, "grad_norm": 0.3695628345012665, "learning_rate": 0.0002, "epoch": 0.288107202680067, "step": 430}, {"loss": 1.7982, "grad_norm": 0.31451135873794556, "learning_rate": 0.0002, "epoch": 0.2948073701842546, "step": 440}, {"loss": 1.7517, "grad_norm": 0.3959707021713257, "learning_rate": 0.0002, "epoch": 0.3015075376884422, "step": 450}, {"loss": 1.8142, "grad_norm": 0.33475354313850403, "learning_rate": 0.0002, "epoch": 0.3082077051926298, "step": 460}, {"loss": 1.8805, "grad_norm": 0.33933115005493164, "learning_rate": 0.0002, "epoch": 0.3149078726968174, "step": 470}, {"loss": 1.7564, "grad_norm": 0.3264943063259125, "learning_rate": 0.0002, "epoch": 0.32160804020100503, "step": 480}, {"loss": 1.8428, "grad_norm": 0.40188100934028625, "learning_rate": 0.0002, "epoch": 0.32830820770519265, "step": 490}, {"loss": 1.7624, "grad_norm": 0.37408649921417236, "learning_rate": 0.0002, "epoch": 0.33500837520938026, "step": 500}, {"loss": 1.7745, "grad_norm": 0.33925938606262207, "learning_rate": 0.0002, "epoch": 0.3417085427135678, "step": 510}, {"loss": 1.814, "grad_norm": 0.36836713552474976, "learning_rate": 0.0002, "epoch": 0.34840871021775544, "step": 520}, {"loss": 1.8037, "grad_norm": 0.37284499406814575, "learning_rate": 0.0002, "epoch": 0.35510887772194305, "step": 530}, {"loss": 1.8379, "grad_norm": 0.3192278742790222, "learning_rate": 0.0002, "epoch": 0.36180904522613067, "step": 540}, {"loss": 1.8702, "grad_norm": 0.30233290791511536, "learning_rate": 0.0002, "epoch": 0.3685092127303183, "step": 550}, {"loss": 1.8799, "grad_norm": 0.3340817391872406, "learning_rate": 0.0002, "epoch": 0.37520938023450584, "step": 560}, {"loss": 1.8404, "grad_norm": 0.32600095868110657, "learning_rate": 0.0002, "epoch": 0.38190954773869346, "step": 570}, {"loss": 1.7804, "grad_norm": 0.33711278438568115, "learning_rate": 0.0002, "epoch": 0.38860971524288107, "step": 580}, {"loss": 1.8445, "grad_norm": 0.34890690445899963, "learning_rate": 0.0002, "epoch": 0.3953098827470687, "step": 590}, {"loss": 1.8187, "grad_norm": 0.38238924741744995, "learning_rate": 0.0002, "epoch": 0.4020100502512563, "step": 600}, {"loss": 1.8111, "grad_norm": 0.34399354457855225, "learning_rate": 0.0002, "epoch": 0.40871021775544386, "step": 610}, {"loss": 1.8006, "grad_norm": 0.3346073627471924, "learning_rate": 0.0002, "epoch": 0.4154103852596315, "step": 620}, {"loss": 1.7705, "grad_norm": 0.3545648157596588, "learning_rate": 0.0002, "epoch": 0.4221105527638191, "step": 630}, {"loss": 1.8445, "grad_norm": 0.3378899097442627, "learning_rate": 0.0002, "epoch": 0.4288107202680067, "step": 640}, {"loss": 1.804, "grad_norm": 0.3255569040775299, "learning_rate": 0.0002, "epoch": 0.4355108877721943, "step": 650}, {"loss": 1.7679, "grad_norm": 0.34880587458610535, "learning_rate": 0.0002, "epoch": 0.44221105527638194, "step": 660}, {"loss": 1.7861, "grad_norm": 0.3402383625507355, "learning_rate": 0.0002, "epoch": 0.4489112227805695, "step": 670}, {"loss": 1.8131, "grad_norm": 0.3594033718109131, "learning_rate": 0.0002, "epoch": 0.4556113902847571, "step": 680}, {"loss": 1.8399, "grad_norm": 0.31000566482543945, "learning_rate": 0.0002, "epoch": 0.4623115577889447, "step": 690}, {"loss": 1.7521, "grad_norm": 0.37229061126708984, "learning_rate": 0.0002, "epoch": 0.46901172529313234, "step": 700}, {"loss": 1.7779, "grad_norm": 0.315801739692688, "learning_rate": 0.0002, "epoch": 0.47571189279731996, "step": 710}, {"loss": 1.7515, "grad_norm": 0.3220832645893097, "learning_rate": 0.0002, "epoch": 0.4824120603015075, "step": 720}, {"loss": 1.7181, "grad_norm": 0.3435456156730652, "learning_rate": 0.0002, "epoch": 0.48911222780569513, "step": 730}, {"loss": 1.8844, "grad_norm": 0.30380892753601074, "learning_rate": 0.0002, "epoch": 0.49581239530988275, "step": 740}, {"loss": 1.7792, "grad_norm": 0.3555026054382324, "learning_rate": 0.0002, "epoch": 0.5025125628140703, "step": 750}, {"loss": 1.7714, "grad_norm": 0.3019855320453644, "learning_rate": 0.0002, "epoch": 0.509212730318258, "step": 760}, {"loss": 1.7962, "grad_norm": 0.309111088514328, "learning_rate": 0.0002, "epoch": 0.5159128978224455, "step": 770}, {"loss": 1.7913, "grad_norm": 0.366020530462265, "learning_rate": 0.0002, "epoch": 0.5226130653266332, "step": 780}, {"loss": 1.8008, "grad_norm": 0.3267050087451935, "learning_rate": 0.0002, "epoch": 0.5293132328308208, "step": 790}, {"loss": 1.7397, "grad_norm": 0.34265750646591187, "learning_rate": 0.0002, "epoch": 0.5360134003350083, "step": 800}, {"loss": 1.8251, "grad_norm": 0.313669890165329, "learning_rate": 0.0002, "epoch": 0.542713567839196, "step": 810}, {"loss": 1.8369, "grad_norm": 0.3355236053466797, "learning_rate": 0.0002, "epoch": 0.5494137353433836, "step": 820}, {"loss": 1.7381, "grad_norm": 0.3186608552932739, "learning_rate": 0.0002, "epoch": 0.5561139028475712, "step": 830}, {"loss": 1.8034, "grad_norm": 0.30357518792152405, "learning_rate": 0.0002, "epoch": 0.5628140703517588, "step": 840}, {"loss": 1.769, "grad_norm": 0.3990040123462677, "learning_rate": 0.0002, "epoch": 0.5695142378559463, "step": 850}, {"loss": 1.7482, "grad_norm": 0.34363803267478943, "learning_rate": 0.0002, "epoch": 0.576214405360134, "step": 860}, {"loss": 1.8106, "grad_norm": 0.3757908046245575, "learning_rate": 0.0002, "epoch": 0.5829145728643216, "step": 870}, {"loss": 1.8104, "grad_norm": 0.3359757661819458, "learning_rate": 0.0002, "epoch": 0.5896147403685092, "step": 880}, {"loss": 1.7591, "grad_norm": 0.5555329918861389, "learning_rate": 0.0002, "epoch": 0.5963149078726968, "step": 890}, {"loss": 1.7715, "grad_norm": 0.4046323895454407, "learning_rate": 0.0002, "epoch": 0.6030150753768844, "step": 900}, {"loss": 1.7998, "grad_norm": 0.29834219813346863, "learning_rate": 0.0002, "epoch": 0.609715242881072, "step": 910}, {"loss": 1.7826, "grad_norm": 0.3241238594055176, "learning_rate": 0.0002, "epoch": 0.6164154103852596, "step": 920}, {"loss": 1.8342, "grad_norm": 0.35154739022254944, "learning_rate": 0.0002, "epoch": 0.6231155778894473, "step": 930}, {"loss": 1.8076, "grad_norm": 0.3287706673145294, "learning_rate": 0.0002, "epoch": 0.6298157453936348, "step": 940}, {"loss": 1.8038, "grad_norm": 0.35670626163482666, "learning_rate": 0.0002, "epoch": 0.6365159128978225, "step": 950}, {"loss": 1.869, "grad_norm": 0.6114104986190796, "learning_rate": 0.0002, "epoch": 0.6432160804020101, "step": 960}, {"loss": 1.8297, "grad_norm": 0.3186565041542053, "learning_rate": 0.0002, "epoch": 0.6499162479061976, "step": 970}, {"loss": 1.7539, "grad_norm": 0.27164125442504883, "learning_rate": 0.0002, "epoch": 0.6566164154103853, "step": 980}, {"loss": 1.8339, "grad_norm": 0.34407344460487366, "learning_rate": 0.0002, "epoch": 0.6633165829145728, "step": 990}, {"loss": 1.855, "grad_norm": 0.368415892124176, "learning_rate": 0.0002, "epoch": 0.6700167504187605, "step": 1000}, {"loss": 1.7821, "grad_norm": 0.3306390643119812, "learning_rate": 0.0002, "epoch": 0.6767169179229481, "step": 1010}, {"loss": 1.786, "grad_norm": 0.3198648989200592, "learning_rate": 0.0002, "epoch": 0.6834170854271356, "step": 1020}, {"loss": 1.816, "grad_norm": 0.3092987537384033, "learning_rate": 0.0002, "epoch": 0.6901172529313233, "step": 1030}, {"loss": 1.7689, "grad_norm": 0.3090653419494629, "learning_rate": 0.0002, "epoch": 0.6968174204355109, "step": 1040}, {"loss": 1.7544, "grad_norm": 0.3485880196094513, "learning_rate": 0.0002, "epoch": 0.7035175879396985, "step": 1050}, {"loss": 1.8187, "grad_norm": 0.35782721638679504, "learning_rate": 0.0002, "epoch": 0.7102177554438861, "step": 1060}, {"loss": 1.806, "grad_norm": 0.34256869554519653, "learning_rate": 0.0002, "epoch": 0.7169179229480737, "step": 1070}, {"loss": 1.7873, "grad_norm": 0.30461037158966064, "learning_rate": 0.0002, "epoch": 0.7236180904522613, "step": 1080}, {"loss": 1.7367, "grad_norm": 0.3398691713809967, "learning_rate": 0.0002, "epoch": 0.7303182579564489, "step": 1090}, {"loss": 1.8756, "grad_norm": 0.3180808126926422, "learning_rate": 0.0002, "epoch": 0.7370184254606366, "step": 1100}, {"loss": 1.6988, "grad_norm": 0.34400665760040283, "learning_rate": 0.0002, "epoch": 0.7437185929648241, "step": 1110}, {"loss": 1.7851, "grad_norm": 0.34244877099990845, "learning_rate": 0.0002, "epoch": 0.7504187604690117, "step": 1120}, {"loss": 1.7841, "grad_norm": 0.29946693778038025, "learning_rate": 0.0002, "epoch": 0.7571189279731994, "step": 1130}, {"loss": 1.7456, "grad_norm": 0.37547236680984497, "learning_rate": 0.0002, "epoch": 0.7638190954773869, "step": 1140}, {"loss": 1.8425, "grad_norm": 0.3263005018234253, "learning_rate": 0.0002, "epoch": 0.7705192629815746, "step": 1150}, {"loss": 1.7222, "grad_norm": 0.41363608837127686, "learning_rate": 0.0002, "epoch": 0.7772194304857621, "step": 1160}, {"loss": 1.7836, "grad_norm": 0.36267954111099243, "learning_rate": 0.0002, "epoch": 0.7839195979899497, "step": 1170}, {"loss": 1.9183, "grad_norm": 0.31789499521255493, "learning_rate": 0.0002, "epoch": 0.7906197654941374, "step": 1180}, {"loss": 1.78, "grad_norm": 0.5708149075508118, "learning_rate": 0.0002, "epoch": 0.7973199329983249, "step": 1190}, {"loss": 1.6908, "grad_norm": 0.322099506855011, "learning_rate": 0.0002, "epoch": 0.8040201005025126, "step": 1200}, {"loss": 1.7639, "grad_norm": 0.3419909179210663, "learning_rate": 0.0002, "epoch": 0.8107202680067002, "step": 1210}, {"loss": 1.7428, "grad_norm": 0.36286255717277527, "learning_rate": 0.0002, "epoch": 0.8174204355108877, "step": 1220}, {"loss": 1.8409, "grad_norm": 0.33992862701416016, "learning_rate": 0.0002, "epoch": 0.8241206030150754, "step": 1230}, {"loss": 1.7507, "grad_norm": 0.32622793316841125, "learning_rate": 0.0002, "epoch": 0.830820770519263, "step": 1240}, {"loss": 1.8098, "grad_norm": 0.3036167621612549, "learning_rate": 0.0002, "epoch": 0.8375209380234506, "step": 1250}, {"loss": 1.8094, "grad_norm": 0.3182215392589569, "learning_rate": 0.0002, "epoch": 0.8442211055276382, "step": 1260}, {"loss": 1.8017, "grad_norm": 0.3270018696784973, "learning_rate": 0.0002, "epoch": 0.8509212730318257, "step": 1270}, {"loss": 1.798, "grad_norm": 0.32652342319488525, "learning_rate": 0.0002, "epoch": 0.8576214405360134, "step": 1280}, {"loss": 1.7448, "grad_norm": 0.3631329834461212, "learning_rate": 0.0002, "epoch": 0.864321608040201, "step": 1290}, {"loss": 1.7, "grad_norm": 0.36706018447875977, "learning_rate": 0.0002, "epoch": 0.8710217755443886, "step": 1300}, {"loss": 1.8178, "grad_norm": 0.3347418010234833, "learning_rate": 0.0002, "epoch": 0.8777219430485762, "step": 1310}, {"loss": 1.7824, "grad_norm": 0.34371060132980347, "learning_rate": 0.0002, "epoch": 0.8844221105527639, "step": 1320}, {"loss": 1.783, "grad_norm": 0.3029090166091919, "learning_rate": 0.0002, "epoch": 0.8911222780569514, "step": 1330}, {"loss": 1.8017, "grad_norm": 0.34700682759284973, "learning_rate": 0.0002, "epoch": 0.897822445561139, "step": 1340}, {"loss": 1.7998, "grad_norm": 0.35574328899383545, "learning_rate": 0.0002, "epoch": 0.9045226130653267, "step": 1350}, {"loss": 1.7339, "grad_norm": 0.30928221344947815, "learning_rate": 0.0002, "epoch": 0.9112227805695142, "step": 1360}, {"loss": 1.7479, "grad_norm": 0.30652928352355957, "learning_rate": 0.0002, "epoch": 0.9179229480737019, "step": 1370}, {"loss": 1.7491, "grad_norm": 0.3838157653808594, "learning_rate": 0.0002, "epoch": 0.9246231155778895, "step": 1380}, {"loss": 1.7977, "grad_norm": 0.31655240058898926, "learning_rate": 0.0002, "epoch": 0.931323283082077, "step": 1390}, {"loss": 1.8175, "grad_norm": 0.41737303137779236, "learning_rate": 0.0002, "epoch": 0.9380234505862647, "step": 1400}, {"loss": 1.6811, "grad_norm": 0.3227267861366272, "learning_rate": 0.0002, "epoch": 0.9447236180904522, "step": 1410}, {"loss": 1.7343, "grad_norm": 0.3729925751686096, "learning_rate": 0.0002, "epoch": 0.9514237855946399, "step": 1420}, {"loss": 1.8221, "grad_norm": 0.30779409408569336, "learning_rate": 0.0002, "epoch": 0.9581239530988275, "step": 1430}, {"loss": 1.7972, "grad_norm": 0.334379643201828, "learning_rate": 0.0002, "epoch": 0.964824120603015, "step": 1440}, {"loss": 1.7141, "grad_norm": 0.3568236231803894, "learning_rate": 0.0002, "epoch": 0.9715242881072027, "step": 1450}, {"loss": 1.7541, "grad_norm": 0.33310577273368835, "learning_rate": 0.0002, "epoch": 0.9782244556113903, "step": 1460}, {"loss": 1.8511, "grad_norm": 0.2972261905670166, "learning_rate": 0.0002, "epoch": 0.9849246231155779, "step": 1470}, {"loss": 1.7654, "grad_norm": 0.3322717845439911, "learning_rate": 0.0002, "epoch": 0.9916247906197655, "step": 1480}, {"loss": 1.8033, "grad_norm": 0.3276330828666687, "learning_rate": 0.0002, "epoch": 0.998324958123953, "step": 1490}, {"eval_loss": 1.8036354780197144, "eval_runtime": 37.8949, "eval_samples_per_second": 13.59, "eval_steps_per_second": 1.715, "epoch": 0.9996649916247906, "step": 1492}, {"loss": 1.7138, "grad_norm": 0.29252371191978455, "learning_rate": 0.0002, "epoch": 1.0050251256281406, "step": 1500}, {"loss": 1.8198, "grad_norm": 0.31607162952423096, "learning_rate": 0.0002, "epoch": 1.0117252931323284, "step": 1510}, {"loss": 1.6779, "grad_norm": 0.32294467091560364, "learning_rate": 0.0002, "epoch": 1.018425460636516, "step": 1520}, {"loss": 1.7919, "grad_norm": 0.3868017792701721, "learning_rate": 0.0002, "epoch": 1.0251256281407035, "step": 1530}, {"loss": 1.7954, "grad_norm": 0.3178282082080841, "learning_rate": 0.0002, "epoch": 1.031825795644891, "step": 1540}, {"loss": 1.7136, "grad_norm": 0.3706750273704529, "learning_rate": 0.0002, "epoch": 1.0385259631490786, "step": 1550}, {"loss": 1.7382, "grad_norm": 0.33930912613868713, "learning_rate": 0.0002, "epoch": 1.0452261306532664, "step": 1560}, {"loss": 1.7602, "grad_norm": 0.33970504999160767, "learning_rate": 0.0002, "epoch": 1.051926298157454, "step": 1570}, {"loss": 1.6573, "grad_norm": 0.42553383111953735, "learning_rate": 0.0002, "epoch": 1.0586264656616415, "step": 1580}, {"loss": 1.645, "grad_norm": 0.3772421181201935, "learning_rate": 0.0002, "epoch": 1.065326633165829, "step": 1590}, {"loss": 1.7362, "grad_norm": 0.34212902188301086, "learning_rate": 0.0002, "epoch": 1.0720268006700167, "step": 1600}, {"loss": 1.7057, "grad_norm": 0.3798283338546753, "learning_rate": 0.0002, "epoch": 1.0787269681742044, "step": 1610}, {"loss": 1.7468, "grad_norm": 0.36909598112106323, "learning_rate": 0.0002, "epoch": 1.085427135678392, "step": 1620}, {"loss": 1.7807, "grad_norm": 0.3344230651855469, "learning_rate": 0.0002, "epoch": 1.0921273031825796, "step": 1630}, {"loss": 1.7111, "grad_norm": 0.3862569332122803, "learning_rate": 0.0002, "epoch": 1.0988274706867671, "step": 1640}, {"loss": 1.7163, "grad_norm": 0.31188511848449707, "learning_rate": 0.0002, "epoch": 1.1055276381909547, "step": 1650}, {"loss": 1.7263, "grad_norm": 0.3563670814037323, "learning_rate": 0.0002, "epoch": 1.1122278056951425, "step": 1660}, {"loss": 1.7718, "grad_norm": 0.35052165389060974, "learning_rate": 0.0002, "epoch": 1.11892797319933, "step": 1670}, {"loss": 1.7601, "grad_norm": 0.3285699188709259, "learning_rate": 0.0002, "epoch": 1.1256281407035176, "step": 1680}, {"loss": 1.6877, "grad_norm": 0.3639393746852875, "learning_rate": 0.0002, "epoch": 1.1323283082077051, "step": 1690}, {"loss": 1.7719, "grad_norm": 0.3842753767967224, "learning_rate": 0.0002, "epoch": 1.1390284757118927, "step": 1700}, {"loss": 1.7002, "grad_norm": 0.3624933063983917, "learning_rate": 0.0002, "epoch": 1.1457286432160805, "step": 1710}, {"loss": 1.7243, "grad_norm": 0.3641220033168793, "learning_rate": 0.0002, "epoch": 1.152428810720268, "step": 1720}, {"loss": 1.752, "grad_norm": 0.32765355706214905, "learning_rate": 0.0002, "epoch": 1.1591289782244556, "step": 1730}, {"loss": 1.6556, "grad_norm": 0.34974896907806396, "learning_rate": 0.0002, "epoch": 1.1658291457286432, "step": 1740}, {"loss": 1.7273, "grad_norm": 0.3910926580429077, "learning_rate": 0.0002, "epoch": 1.1725293132328307, "step": 1750}, {"loss": 1.7204, "grad_norm": 0.3564300537109375, "learning_rate": 0.0002, "epoch": 1.1792294807370185, "step": 1760}, {"loss": 1.746, "grad_norm": 0.34822574257850647, "learning_rate": 0.0002, "epoch": 1.185929648241206, "step": 1770}, {"loss": 1.7256, "grad_norm": 0.36185044050216675, "learning_rate": 0.0002, "epoch": 1.1926298157453936, "step": 1780}, {"loss": 1.6431, "grad_norm": 0.34866711497306824, "learning_rate": 0.0002, "epoch": 1.1993299832495812, "step": 1790}, {"loss": 1.8084, "grad_norm": 0.4017769992351532, "learning_rate": 0.0002, "epoch": 1.2060301507537687, "step": 1800}, {"loss": 1.6985, "grad_norm": 0.32930681109428406, "learning_rate": 0.0002, "epoch": 1.2127303182579565, "step": 1810}, {"loss": 1.7606, "grad_norm": 0.35951921343803406, "learning_rate": 0.0002, "epoch": 1.219430485762144, "step": 1820}, {"loss": 1.6933, "grad_norm": 0.37366992235183716, "learning_rate": 0.0002, "epoch": 1.2261306532663316, "step": 1830}, {"loss": 1.6737, "grad_norm": 0.3565689027309418, "learning_rate": 0.0002, "epoch": 1.2328308207705192, "step": 1840}, {"loss": 1.8013, "grad_norm": 0.3692343533039093, "learning_rate": 0.0002, "epoch": 1.2395309882747068, "step": 1850}, {"loss": 1.736, "grad_norm": 0.38426971435546875, "learning_rate": 0.0002, "epoch": 1.2462311557788945, "step": 1860}, {"loss": 1.7031, "grad_norm": 0.33559855818748474, "learning_rate": 0.0002, "epoch": 1.252931323283082, "step": 1870}, {"loss": 1.7033, "grad_norm": 0.34181106090545654, "learning_rate": 0.0002, "epoch": 1.2596314907872697, "step": 1880}, {"loss": 1.7707, "grad_norm": 0.3916318416595459, "learning_rate": 0.0002, "epoch": 1.2663316582914572, "step": 1890}, {"loss": 1.6686, "grad_norm": 0.3887825012207031, "learning_rate": 0.0002, "epoch": 1.2730318257956448, "step": 1900}, {"loss": 1.7062, "grad_norm": 0.33583927154541016, "learning_rate": 0.0002, "epoch": 1.2797319932998326, "step": 1910}, {"loss": 1.717, "grad_norm": 0.37639349699020386, "learning_rate": 0.0002, "epoch": 1.2864321608040201, "step": 1920}, {"loss": 1.777, "grad_norm": 0.38059428334236145, "learning_rate": 0.0002, "epoch": 1.2931323283082077, "step": 1930}, {"loss": 1.6126, "grad_norm": 0.37253183126449585, "learning_rate": 0.0002, "epoch": 1.2998324958123952, "step": 1940}, {"loss": 1.6758, "grad_norm": 0.37371566891670227, "learning_rate": 0.0002, "epoch": 1.3065326633165828, "step": 1950}, {"loss": 1.6788, "grad_norm": 0.4080910086631775, "learning_rate": 0.0002, "epoch": 1.3132328308207706, "step": 1960}, {"loss": 1.6518, "grad_norm": 0.3174354135990143, "learning_rate": 0.0002, "epoch": 1.3199329983249581, "step": 1970}, {"loss": 1.7925, "grad_norm": 0.4518888294696808, "learning_rate": 0.0002, "epoch": 1.3266331658291457, "step": 1980}, {"loss": 1.7085, "grad_norm": 0.3627921938896179, "learning_rate": 0.0002, "epoch": 1.3333333333333333, "step": 1990}, {"loss": 1.7676, "grad_norm": 0.3655930161476135, "learning_rate": 0.0002, "epoch": 1.3400335008375208, "step": 2000}, {"loss": 1.7016, "grad_norm": 0.3509993255138397, "learning_rate": 0.0002, "epoch": 1.3467336683417086, "step": 2010}, {"loss": 1.7359, "grad_norm": 0.4281129240989685, "learning_rate": 0.0002, "epoch": 1.3534338358458962, "step": 2020}, {"loss": 1.6884, "grad_norm": 0.3821414113044739, "learning_rate": 0.0002, "epoch": 1.3601340033500837, "step": 2030}, {"loss": 1.7075, "grad_norm": 0.3907586336135864, "learning_rate": 0.0002, "epoch": 1.3668341708542713, "step": 2040}, {"loss": 1.7424, "grad_norm": 0.37792932987213135, "learning_rate": 0.0002, "epoch": 1.3735343383584588, "step": 2050}, {"loss": 1.7305, "grad_norm": 0.3693985641002655, "learning_rate": 0.0002, "epoch": 1.3802345058626466, "step": 2060}, {"loss": 1.7434, "grad_norm": 0.32275936007499695, "learning_rate": 0.0002, "epoch": 1.3869346733668342, "step": 2070}, {"loss": 1.6677, "grad_norm": 0.3789440095424652, "learning_rate": 0.0002, "epoch": 1.3936348408710217, "step": 2080}, {"loss": 1.6825, "grad_norm": 0.3638380467891693, "learning_rate": 0.0002, "epoch": 1.4003350083752093, "step": 2090}, {"loss": 1.6542, "grad_norm": 0.3495481610298157, "learning_rate": 0.0002, "epoch": 1.4070351758793969, "step": 2100}, {"loss": 1.7225, "grad_norm": 0.37920597195625305, "learning_rate": 0.0002, "epoch": 1.4137353433835846, "step": 2110}, {"loss": 1.7329, "grad_norm": 0.37218064069747925, "learning_rate": 0.0002, "epoch": 1.4204355108877722, "step": 2120}, {"loss": 1.799, "grad_norm": 0.38074082136154175, "learning_rate": 0.0002, "epoch": 1.4271356783919598, "step": 2130}, {"loss": 1.7403, "grad_norm": 0.3455527126789093, "learning_rate": 0.0002, "epoch": 1.4338358458961473, "step": 2140}, {"loss": 1.776, "grad_norm": 0.3712003529071808, "learning_rate": 0.0002, "epoch": 1.4405360134003349, "step": 2150}, {"loss": 1.7619, "grad_norm": 0.3786754906177521, "learning_rate": 0.0002, "epoch": 1.4472361809045227, "step": 2160}, {"loss": 1.68, "grad_norm": 0.3879223167896271, "learning_rate": 0.0002, "epoch": 1.4539363484087102, "step": 2170}, {"loss": 1.7, "grad_norm": 0.38738805055618286, "learning_rate": 0.0002, "epoch": 1.4606365159128978, "step": 2180}, {"loss": 1.7581, "grad_norm": 0.39768800139427185, "learning_rate": 0.0002, "epoch": 1.4673366834170856, "step": 2190}, {"loss": 1.7671, "grad_norm": 0.4172441065311432, "learning_rate": 0.0002, "epoch": 1.474036850921273, "step": 2200}, {"loss": 1.6736, "grad_norm": 0.4043174982070923, "learning_rate": 0.0002, "epoch": 1.4807370184254607, "step": 2210}, {"loss": 1.7444, "grad_norm": 0.3750883936882019, "learning_rate": 0.0002, "epoch": 1.4874371859296482, "step": 2220}, {"loss": 1.6861, "grad_norm": 0.3552253246307373, "learning_rate": 0.0002, "epoch": 1.4941373534338358, "step": 2230}, {"loss": 1.6471, "grad_norm": 0.34607139229774475, "learning_rate": 0.0002, "epoch": 1.5008375209380236, "step": 2240}, {"loss": 1.6962, "grad_norm": 0.3406706750392914, "learning_rate": 0.0002, "epoch": 1.507537688442211, "step": 2250}, {"loss": 1.7694, "grad_norm": 0.36654895544052124, "learning_rate": 0.0002, "epoch": 1.5142378559463987, "step": 2260}, {"loss": 1.6812, "grad_norm": 0.3914054334163666, "learning_rate": 0.0002, "epoch": 1.5209380234505863, "step": 2270}, {"loss": 1.6822, "grad_norm": 0.42012137174606323, "learning_rate": 0.0002, "epoch": 1.5276381909547738, "step": 2280}, {"loss": 1.697, "grad_norm": 0.39563435316085815, "learning_rate": 0.0002, "epoch": 1.5343383584589616, "step": 2290}, {"loss": 1.7491, "grad_norm": 0.3508438766002655, "learning_rate": 0.0002, "epoch": 1.541038525963149, "step": 2300}, {"loss": 1.7727, "grad_norm": 0.3785218596458435, "learning_rate": 0.0002, "epoch": 1.5477386934673367, "step": 2310}, {"loss": 1.6963, "grad_norm": 0.39377647638320923, "learning_rate": 0.0002, "epoch": 1.5544388609715243, "step": 2320}, {"loss": 1.7263, "grad_norm": 0.3391438126564026, "learning_rate": 0.0002, "epoch": 1.5611390284757118, "step": 2330}, {"loss": 1.7722, "grad_norm": 0.37944263219833374, "learning_rate": 0.0002, "epoch": 1.5678391959798996, "step": 2340}, {"loss": 1.6371, "grad_norm": 0.3523491322994232, "learning_rate": 0.0002, "epoch": 1.574539363484087, "step": 2350}, {"loss": 1.7583, "grad_norm": 0.3911575973033905, "learning_rate": 0.0002, "epoch": 1.5812395309882747, "step": 2360}, {"loss": 1.7117, "grad_norm": 0.33832186460494995, "learning_rate": 0.0002, "epoch": 1.5879396984924623, "step": 2370}, {"loss": 1.7701, "grad_norm": 0.3665979206562042, "learning_rate": 0.0002, "epoch": 1.5946398659966499, "step": 2380}, {"loss": 1.779, "grad_norm": 0.3871748149394989, "learning_rate": 0.0002, "epoch": 1.6013400335008376, "step": 2390}, {"loss": 1.7109, "grad_norm": 0.3586967885494232, "learning_rate": 0.0002, "epoch": 1.608040201005025, "step": 2400}, {"loss": 1.7096, "grad_norm": 0.3563673198223114, "learning_rate": 0.0002, "epoch": 1.6147403685092128, "step": 2410}, {"loss": 1.745, "grad_norm": 0.37588971853256226, "learning_rate": 0.0002, "epoch": 1.6214405360134003, "step": 2420}, {"loss": 1.7086, "grad_norm": 0.352556437253952, "learning_rate": 0.0002, "epoch": 1.6281407035175879, "step": 2430}, {"loss": 1.6547, "grad_norm": 0.3716259300708771, "learning_rate": 0.0002, "epoch": 1.6348408710217757, "step": 2440}, {"loss": 1.7033, "grad_norm": 0.372001975774765, "learning_rate": 0.0002, "epoch": 1.641541038525963, "step": 2450}, {"loss": 1.6584, "grad_norm": 0.3430042862892151, "learning_rate": 0.0002, "epoch": 1.6482412060301508, "step": 2460}, {"loss": 1.7217, "grad_norm": 0.3741483688354492, "learning_rate": 0.0002, "epoch": 1.6549413735343383, "step": 2470}, {"loss": 1.7701, "grad_norm": 0.3610571324825287, "learning_rate": 0.0002, "epoch": 1.661641541038526, "step": 2480}, {"loss": 1.7057, "grad_norm": 0.4204719066619873, "learning_rate": 0.0002, "epoch": 1.6683417085427137, "step": 2490}, {"loss": 1.7954, "grad_norm": 0.3938186466693878, "learning_rate": 0.0002, "epoch": 1.675041876046901, "step": 2500}, {"loss": 1.6633, "grad_norm": 0.3421435058116913, "learning_rate": 0.0002, "epoch": 1.6817420435510888, "step": 2510}, {"loss": 1.7996, "grad_norm": 0.42441412806510925, "learning_rate": 0.0002, "epoch": 1.6884422110552764, "step": 2520}, {"loss": 1.7142, "grad_norm": 0.38071519136428833, "learning_rate": 0.0002, "epoch": 1.695142378559464, "step": 2530}, {"loss": 1.7232, "grad_norm": 0.34078919887542725, "learning_rate": 0.0002, "epoch": 1.7018425460636517, "step": 2540}, {"loss": 1.7126, "grad_norm": 0.412844181060791, "learning_rate": 0.0002, "epoch": 1.708542713567839, "step": 2550}, {"loss": 1.7149, "grad_norm": 0.3753604292869568, "learning_rate": 0.0002, "epoch": 1.7152428810720268, "step": 2560}, {"loss": 1.7011, "grad_norm": 0.41588476300239563, "learning_rate": 0.0002, "epoch": 1.7219430485762144, "step": 2570}, {"loss": 1.6427, "grad_norm": 0.35504111647605896, "learning_rate": 0.0002, "epoch": 1.728643216080402, "step": 2580}, {"loss": 1.7296, "grad_norm": 0.36909720301628113, "learning_rate": 0.0002, "epoch": 1.7353433835845897, "step": 2590}, {"loss": 1.7022, "grad_norm": 0.4149979054927826, "learning_rate": 0.0002, "epoch": 1.742043551088777, "step": 2600}, {"loss": 1.77, "grad_norm": 0.38859328627586365, "learning_rate": 0.0002, "epoch": 1.7487437185929648, "step": 2610}, {"loss": 1.7036, "grad_norm": 0.36738792061805725, "learning_rate": 0.0002, "epoch": 1.7554438860971524, "step": 2620}, {"loss": 1.764, "grad_norm": 0.3968178927898407, "learning_rate": 0.0002, "epoch": 1.76214405360134, "step": 2630}, {"loss": 1.7687, "grad_norm": 0.3972901999950409, "learning_rate": 0.0002, "epoch": 1.7688442211055277, "step": 2640}, {"loss": 1.6774, "grad_norm": 0.3949959874153137, "learning_rate": 0.0002, "epoch": 1.775544388609715, "step": 2650}, {"loss": 1.7247, "grad_norm": 0.44074657559394836, "learning_rate": 0.0002, "epoch": 1.7822445561139029, "step": 2660}, {"loss": 1.7188, "grad_norm": 0.39743664860725403, "learning_rate": 0.0002, "epoch": 1.7889447236180904, "step": 2670}, {"loss": 1.7258, "grad_norm": 0.3950406610965729, "learning_rate": 0.0002, "epoch": 1.795644891122278, "step": 2680}, {"loss": 1.6906, "grad_norm": 0.3568263649940491, "learning_rate": 0.0002, "epoch": 1.8023450586264658, "step": 2690}, {"loss": 1.6735, "grad_norm": 0.3819476366043091, "learning_rate": 0.0002, "epoch": 1.809045226130653, "step": 2700}, {"loss": 1.7198, "grad_norm": 0.3480634391307831, "learning_rate": 0.0002, "epoch": 1.8157453936348409, "step": 2710}, {"loss": 1.7042, "grad_norm": 0.3875853419303894, "learning_rate": 0.0002, "epoch": 1.8224455611390284, "step": 2720}, {"loss": 1.6988, "grad_norm": 0.3441337049007416, "learning_rate": 0.0002, "epoch": 1.829145728643216, "step": 2730}, {"loss": 1.7647, "grad_norm": 0.35692882537841797, "learning_rate": 0.0002, "epoch": 1.8358458961474038, "step": 2740}, {"loss": 1.7033, "grad_norm": 0.36959215998649597, "learning_rate": 0.0002, "epoch": 1.8425460636515911, "step": 2750}, {"loss": 1.7657, "grad_norm": 0.3893393278121948, "learning_rate": 0.0002, "epoch": 1.849246231155779, "step": 2760}, {"loss": 1.7068, "grad_norm": 0.37817293405532837, "learning_rate": 0.0002, "epoch": 1.8559463986599665, "step": 2770}, {"loss": 1.761, "grad_norm": 0.36071285605430603, "learning_rate": 0.0002, "epoch": 1.862646566164154, "step": 2780}, {"loss": 1.7623, "grad_norm": 0.3758420944213867, "learning_rate": 0.0002, "epoch": 1.8693467336683418, "step": 2790}, {"loss": 1.6743, "grad_norm": 0.3889938294887543, "learning_rate": 0.0002, "epoch": 1.8760469011725294, "step": 2800}, {"loss": 1.6151, "grad_norm": 0.34361857175827026, "learning_rate": 0.0002, "epoch": 1.882747068676717, "step": 2810}, {"loss": 1.6038, "grad_norm": 0.39283323287963867, "learning_rate": 0.0002, "epoch": 1.8894472361809045, "step": 2820}, {"loss": 1.7555, "grad_norm": 0.3919452726840973, "learning_rate": 0.0002, "epoch": 1.896147403685092, "step": 2830}, {"loss": 1.673, "grad_norm": 0.38215070962905884, "learning_rate": 0.0002, "epoch": 1.9028475711892798, "step": 2840}, {"loss": 1.7044, "grad_norm": 0.4235064387321472, "learning_rate": 0.0002, "epoch": 1.9095477386934674, "step": 2850}, {"loss": 1.7123, "grad_norm": 0.35694634914398193, "learning_rate": 0.0002, "epoch": 1.916247906197655, "step": 2860}, {"loss": 1.8128, "grad_norm": 0.383492112159729, "learning_rate": 0.0002, "epoch": 1.9229480737018425, "step": 2870}, {"loss": 1.7581, "grad_norm": 0.5945147275924683, "learning_rate": 0.0002, "epoch": 1.92964824120603, "step": 2880}, {"loss": 1.7421, "grad_norm": 0.3367522358894348, "learning_rate": 0.0002, "epoch": 1.9363484087102178, "step": 2890}, {"loss": 1.6561, "grad_norm": 0.35300394892692566, "learning_rate": 0.0002, "epoch": 1.9430485762144054, "step": 2900}, {"loss": 1.7033, "grad_norm": 0.38084495067596436, "learning_rate": 0.0002, "epoch": 1.949748743718593, "step": 2910}, {"loss": 1.7132, "grad_norm": 0.37559160590171814, "learning_rate": 0.0002, "epoch": 1.9564489112227805, "step": 2920}, {"loss": 1.6759, "grad_norm": 0.3661738336086273, "learning_rate": 0.0002, "epoch": 1.963149078726968, "step": 2930}, {"loss": 1.7643, "grad_norm": 0.4073849320411682, "learning_rate": 0.0002, "epoch": 1.9698492462311559, "step": 2940}, {"loss": 1.6806, "grad_norm": 0.3723304271697998, "learning_rate": 0.0002, "epoch": 1.9765494137353434, "step": 2950}, {"loss": 1.7611, "grad_norm": 0.3991098999977112, "learning_rate": 0.0002, "epoch": 1.983249581239531, "step": 2960}, {"loss": 1.7263, "grad_norm": 0.3947085440158844, "learning_rate": 0.0002, "epoch": 1.9899497487437185, "step": 2970}, {"loss": 1.7217, "grad_norm": 0.3786258399486542, "learning_rate": 0.0002, "epoch": 1.996649916247906, "step": 2980}, {"eval_loss": 1.8028968572616577, "eval_runtime": 37.8985, "eval_samples_per_second": 13.589, "eval_steps_per_second": 1.715, "epoch": 2.0, "step": 2985}, {"loss": 1.695, "grad_norm": 0.34824079275131226, "learning_rate": 0.0002, "epoch": 2.003350083752094, "step": 2990}, {"loss": 1.5853, "grad_norm": 0.3394894003868103, "learning_rate": 0.0002, "epoch": 2.0100502512562812, "step": 3000}, {"loss": 1.5783, "grad_norm": 0.36910977959632874, "learning_rate": 0.0002, "epoch": 2.016750418760469, "step": 3010}, {"loss": 1.6105, "grad_norm": 0.45000967383384705, "learning_rate": 0.0002, "epoch": 2.023450586264657, "step": 3020}, {"loss": 1.6019, "grad_norm": 0.3791407346725464, "learning_rate": 0.0002, "epoch": 2.030150753768844, "step": 3030}, {"loss": 1.5832, "grad_norm": 0.387321799993515, "learning_rate": 0.0002, "epoch": 2.036850921273032, "step": 3040}, {"loss": 1.6834, "grad_norm": 0.4185757040977478, "learning_rate": 0.0002, "epoch": 2.0435510887772192, "step": 3050}, {"loss": 1.5696, "grad_norm": 0.45110777020454407, "learning_rate": 0.0002, "epoch": 2.050251256281407, "step": 3060}, {"loss": 1.6231, "grad_norm": 0.42663660645484924, "learning_rate": 0.0002, "epoch": 2.056951423785595, "step": 3070}, {"loss": 1.6279, "grad_norm": 0.4546292722225189, "learning_rate": 0.0002, "epoch": 2.063651591289782, "step": 3080}, {"loss": 1.6141, "grad_norm": 0.3979759216308594, "learning_rate": 0.0002, "epoch": 2.07035175879397, "step": 3090}, {"loss": 1.6343, "grad_norm": 0.43596673011779785, "learning_rate": 0.0002, "epoch": 2.0770519262981573, "step": 3100}, {"loss": 1.5441, "grad_norm": 0.40120232105255127, "learning_rate": 0.0002, "epoch": 2.083752093802345, "step": 3110}, {"loss": 1.6309, "grad_norm": 0.44449281692504883, "learning_rate": 0.0002, "epoch": 2.090452261306533, "step": 3120}, {"loss": 1.5652, "grad_norm": 0.42672568559646606, "learning_rate": 0.0002, "epoch": 2.09715242881072, "step": 3130}, {"loss": 1.682, "grad_norm": 0.4232690930366516, "learning_rate": 0.0002, "epoch": 2.103852596314908, "step": 3140}, {"loss": 1.624, "grad_norm": 0.4299317002296448, "learning_rate": 0.0002, "epoch": 2.1105527638190953, "step": 3150}, {"loss": 1.6766, "grad_norm": 0.4067758023738861, "learning_rate": 0.0002, "epoch": 2.117252931323283, "step": 3160}, {"loss": 1.6759, "grad_norm": 0.4918815791606903, "learning_rate": 0.0002, "epoch": 2.123953098827471, "step": 3170}, {"loss": 1.6478, "grad_norm": 0.4140559732913971, "learning_rate": 0.0002, "epoch": 2.130653266331658, "step": 3180}, {"loss": 1.6641, "grad_norm": 0.4555995464324951, "learning_rate": 0.0002, "epoch": 2.137353433835846, "step": 3190}, {"loss": 1.5888, "grad_norm": 0.42943915724754333, "learning_rate": 0.0002, "epoch": 2.1440536013400333, "step": 3200}, {"loss": 1.5886, "grad_norm": 0.4730435013771057, "learning_rate": 0.0002, "epoch": 2.150753768844221, "step": 3210}, {"loss": 1.6022, "grad_norm": 0.43310216069221497, "learning_rate": 0.0002, "epoch": 2.157453936348409, "step": 3220}, {"loss": 1.6058, "grad_norm": 0.42054110765457153, "learning_rate": 0.0002, "epoch": 2.164154103852596, "step": 3230}, {"loss": 1.6749, "grad_norm": 0.4897233247756958, "learning_rate": 0.0002, "epoch": 2.170854271356784, "step": 3240}, {"loss": 1.6983, "grad_norm": 0.42194533348083496, "learning_rate": 0.0002, "epoch": 2.1775544388609713, "step": 3250}, {"loss": 1.6159, "grad_norm": 0.44494450092315674, "learning_rate": 0.0002, "epoch": 2.184254606365159, "step": 3260}, {"loss": 1.6977, "grad_norm": 0.43524879217147827, "learning_rate": 0.0002, "epoch": 2.190954773869347, "step": 3270}, {"loss": 1.528, "grad_norm": 0.4621117413043976, "learning_rate": 0.0002, "epoch": 2.1976549413735342, "step": 3280}, {"loss": 1.632, "grad_norm": 0.4073285460472107, "learning_rate": 0.0002, "epoch": 2.204355108877722, "step": 3290}, {"loss": 1.6141, "grad_norm": 0.47868335247039795, "learning_rate": 0.0002, "epoch": 2.2110552763819094, "step": 3300}, {"loss": 1.6857, "grad_norm": 0.4264970123767853, "learning_rate": 0.0002, "epoch": 2.217755443886097, "step": 3310}, {"loss": 1.5653, "grad_norm": 0.4491245150566101, "learning_rate": 0.0002, "epoch": 2.224455611390285, "step": 3320}, {"loss": 1.5881, "grad_norm": 0.4010344445705414, "learning_rate": 0.0002, "epoch": 2.2311557788944723, "step": 3330}, {"loss": 1.6684, "grad_norm": 0.4232759177684784, "learning_rate": 0.0002, "epoch": 2.23785594639866, "step": 3340}, {"loss": 1.6336, "grad_norm": 0.5099776983261108, "learning_rate": 0.0002, "epoch": 2.2445561139028474, "step": 3350}, {"loss": 1.6764, "grad_norm": 0.5223407745361328, "learning_rate": 0.0002, "epoch": 2.251256281407035, "step": 3360}, {"loss": 1.6625, "grad_norm": 0.47818470001220703, "learning_rate": 0.0002, "epoch": 2.257956448911223, "step": 3370}, {"loss": 1.5946, "grad_norm": 0.4721255898475647, "learning_rate": 0.0002, "epoch": 2.2646566164154103, "step": 3380}, {"loss": 1.5568, "grad_norm": 0.4113229513168335, "learning_rate": 0.0002, "epoch": 2.271356783919598, "step": 3390}, {"loss": 1.6494, "grad_norm": 0.507080078125, "learning_rate": 0.0002, "epoch": 2.2780569514237854, "step": 3400}, {"loss": 1.6183, "grad_norm": 0.4852292239665985, "learning_rate": 0.0002, "epoch": 2.284757118927973, "step": 3410}, {"loss": 1.6132, "grad_norm": 0.4503684341907501, "learning_rate": 0.0002, "epoch": 2.291457286432161, "step": 3420}, {"loss": 1.6649, "grad_norm": 0.8359600305557251, "learning_rate": 0.0002, "epoch": 2.2981574539363483, "step": 3430}, {"loss": 1.6644, "grad_norm": 0.44604045152664185, "learning_rate": 0.0002, "epoch": 2.304857621440536, "step": 3440}, {"loss": 1.5972, "grad_norm": 0.45667049288749695, "learning_rate": 0.0002, "epoch": 2.3115577889447234, "step": 3450}, {"loss": 1.6667, "grad_norm": 0.4879349172115326, "learning_rate": 0.0002, "epoch": 2.318257956448911, "step": 3460}, {"loss": 1.5804, "grad_norm": 0.4033963084220886, "learning_rate": 0.0002, "epoch": 2.324958123953099, "step": 3470}, {"loss": 1.5838, "grad_norm": 0.44494301080703735, "learning_rate": 0.0002, "epoch": 2.3316582914572863, "step": 3480}, {"loss": 1.5951, "grad_norm": 0.4794621765613556, "learning_rate": 0.0002, "epoch": 2.338358458961474, "step": 3490}, {"loss": 1.6807, "grad_norm": 0.41404327750205994, "learning_rate": 0.0002, "epoch": 2.3450586264656614, "step": 3500}, {"loss": 1.714, "grad_norm": 0.4664851725101471, "learning_rate": 0.0002, "epoch": 2.351758793969849, "step": 3510}, {"loss": 1.6537, "grad_norm": 0.4263697564601898, "learning_rate": 0.0002, "epoch": 2.358458961474037, "step": 3520}, {"loss": 1.6551, "grad_norm": 0.5035167336463928, "learning_rate": 0.0002, "epoch": 2.3651591289782243, "step": 3530}, {"loss": 1.6208, "grad_norm": 0.4380664527416229, "learning_rate": 0.0002, "epoch": 2.371859296482412, "step": 3540}, {"loss": 1.634, "grad_norm": 0.5227681994438171, "learning_rate": 0.0002, "epoch": 2.3785594639865995, "step": 3550}, {"loss": 1.6146, "grad_norm": 0.4382302761077881, "learning_rate": 0.0002, "epoch": 2.3852596314907872, "step": 3560}, {"loss": 1.5653, "grad_norm": 0.4392451047897339, "learning_rate": 0.0002, "epoch": 2.391959798994975, "step": 3570}, {"loss": 1.6626, "grad_norm": 0.4372786581516266, "learning_rate": 0.0002, "epoch": 2.3986599664991624, "step": 3580}, {"loss": 1.519, "grad_norm": 0.5015502572059631, "learning_rate": 0.0002, "epoch": 2.40536013400335, "step": 3590}, {"loss": 1.588, "grad_norm": 0.5653210878372192, "learning_rate": 0.0002, "epoch": 2.4120603015075375, "step": 3600}, {"loss": 1.6075, "grad_norm": 0.53007972240448, "learning_rate": 0.0002, "epoch": 2.4187604690117253, "step": 3610}, {"loss": 1.6421, "grad_norm": 0.4659176766872406, "learning_rate": 0.0002, "epoch": 2.425460636515913, "step": 3620}, {"loss": 1.625, "grad_norm": 0.5637837052345276, "learning_rate": 0.0002, "epoch": 2.4321608040201004, "step": 3630}, {"loss": 1.6168, "grad_norm": 0.4248391389846802, "learning_rate": 0.0002, "epoch": 2.438860971524288, "step": 3640}, {"loss": 1.6822, "grad_norm": 0.44668248295783997, "learning_rate": 0.0002, "epoch": 2.4455611390284755, "step": 3650}, {"loss": 1.6417, "grad_norm": 0.43990179896354675, "learning_rate": 0.0002, "epoch": 2.4522613065326633, "step": 3660}, {"loss": 1.6723, "grad_norm": 0.4532523453235626, "learning_rate": 0.0002, "epoch": 2.458961474036851, "step": 3670}, {"loss": 1.6957, "grad_norm": 0.6605591773986816, "learning_rate": 0.0002, "epoch": 2.4656616415410384, "step": 3680}, {"loss": 1.6159, "grad_norm": 0.4694533348083496, "learning_rate": 0.0002, "epoch": 2.472361809045226, "step": 3690}, {"loss": 1.6239, "grad_norm": 0.4485011100769043, "learning_rate": 0.0002, "epoch": 2.4790619765494135, "step": 3700}, {"loss": 1.6834, "grad_norm": 0.4761785864830017, "learning_rate": 0.0002, "epoch": 2.4857621440536013, "step": 3710}, {"loss": 1.6313, "grad_norm": 0.5116432309150696, "learning_rate": 0.0002, "epoch": 2.492462311557789, "step": 3720}, {"loss": 1.5054, "grad_norm": 0.49523618817329407, "learning_rate": 0.0002, "epoch": 2.4991624790619764, "step": 3730}, {"loss": 1.6249, "grad_norm": 0.43826380372047424, "learning_rate": 0.0002, "epoch": 2.505862646566164, "step": 3740}, {"loss": 1.5762, "grad_norm": 0.4916154146194458, "learning_rate": 0.0002, "epoch": 2.5125628140703515, "step": 3750}, {"loss": 1.5157, "grad_norm": 0.5381299257278442, "learning_rate": 0.0002, "epoch": 2.5192629815745393, "step": 3760}, {"loss": 1.6467, "grad_norm": 0.44947415590286255, "learning_rate": 0.0002, "epoch": 2.525963149078727, "step": 3770}, {"loss": 1.67, "grad_norm": 0.49979084730148315, "learning_rate": 0.0002, "epoch": 2.5326633165829144, "step": 3780}, {"loss": 1.622, "grad_norm": 0.43046900629997253, "learning_rate": 0.0002, "epoch": 2.539363484087102, "step": 3790}, {"loss": 1.6789, "grad_norm": 0.4513470530509949, "learning_rate": 0.0002, "epoch": 2.5460636515912896, "step": 3800}, {"loss": 1.6335, "grad_norm": 0.49900051951408386, "learning_rate": 0.0002, "epoch": 2.5527638190954773, "step": 3810}, {"loss": 1.6322, "grad_norm": 0.4348420202732086, "learning_rate": 0.0002, "epoch": 2.559463986599665, "step": 3820}, {"loss": 1.6218, "grad_norm": 0.4684867560863495, "learning_rate": 0.0002, "epoch": 2.5661641541038525, "step": 3830}, {"loss": 1.6535, "grad_norm": 0.44430989027023315, "learning_rate": 0.0002, "epoch": 2.5728643216080402, "step": 3840}, {"loss": 1.5909, "grad_norm": 0.47375255823135376, "learning_rate": 0.0002, "epoch": 2.5795644891122276, "step": 3850}, {"loss": 1.6269, "grad_norm": 0.45493075251579285, "learning_rate": 0.0002, "epoch": 2.5862646566164154, "step": 3860}, {"loss": 1.604, "grad_norm": 0.4563275873661041, "learning_rate": 0.0002, "epoch": 2.592964824120603, "step": 3870}, {"loss": 1.642, "grad_norm": 0.46060335636138916, "learning_rate": 0.0002, "epoch": 2.5996649916247905, "step": 3880}, {"loss": 1.6302, "grad_norm": 0.4718867540359497, "learning_rate": 0.0002, "epoch": 2.6063651591289783, "step": 3890}, {"loss": 1.6242, "grad_norm": 0.41570305824279785, "learning_rate": 0.0002, "epoch": 2.6130653266331656, "step": 3900}, {"loss": 1.6401, "grad_norm": 0.4603121876716614, "learning_rate": 0.0002, "epoch": 2.6197654941373534, "step": 3910}, {"loss": 1.6839, "grad_norm": 0.4734652638435364, "learning_rate": 0.0002, "epoch": 2.626465661641541, "step": 3920}, {"loss": 1.5448, "grad_norm": 0.45348483324050903, "learning_rate": 0.0002, "epoch": 2.6331658291457285, "step": 3930}, {"loss": 1.6157, "grad_norm": 0.46559447050094604, "learning_rate": 0.0002, "epoch": 2.6398659966499163, "step": 3940}, {"loss": 1.7052, "grad_norm": 0.44113144278526306, "learning_rate": 0.0002, "epoch": 2.6465661641541036, "step": 3950}, {"loss": 1.6315, "grad_norm": 0.41415104269981384, "learning_rate": 0.0002, "epoch": 2.6532663316582914, "step": 3960}, {"loss": 1.6589, "grad_norm": 0.48868080973625183, "learning_rate": 0.0002, "epoch": 2.659966499162479, "step": 3970}, {"loss": 1.6211, "grad_norm": 0.49610549211502075, "learning_rate": 0.0002, "epoch": 2.6666666666666665, "step": 3980}, {"loss": 1.6235, "grad_norm": 0.4309130907058716, "learning_rate": 0.0002, "epoch": 2.6733668341708543, "step": 3990}, {"loss": 1.6452, "grad_norm": 0.4489327669143677, "learning_rate": 0.0002, "epoch": 2.6800670016750416, "step": 4000}, {"loss": 1.5744, "grad_norm": 0.5380139946937561, "learning_rate": 0.0002, "epoch": 2.6867671691792294, "step": 4010}, {"loss": 1.6524, "grad_norm": 0.5076672434806824, "learning_rate": 0.0002, "epoch": 2.693467336683417, "step": 4020}, {"loss": 1.636, "grad_norm": 0.47620031237602234, "learning_rate": 0.0002, "epoch": 2.7001675041876045, "step": 4030}, {"loss": 1.5543, "grad_norm": 0.48089155554771423, "learning_rate": 0.0002, "epoch": 2.7068676716917923, "step": 4040}, {"loss": 1.6396, "grad_norm": 0.5108814239501953, "learning_rate": 0.0002, "epoch": 2.7135678391959797, "step": 4050}, {"loss": 1.5905, "grad_norm": 0.4196513295173645, "learning_rate": 0.0002, "epoch": 2.7202680067001674, "step": 4060}, {"loss": 1.686, "grad_norm": 0.4574664831161499, "learning_rate": 0.0002, "epoch": 2.726968174204355, "step": 4070}, {"loss": 1.6234, "grad_norm": 0.4671640992164612, "learning_rate": 0.0002, "epoch": 2.7336683417085426, "step": 4080}, {"loss": 1.6827, "grad_norm": 0.49355530738830566, "learning_rate": 0.0002, "epoch": 2.7403685092127303, "step": 4090}, {"loss": 1.6999, "grad_norm": 0.46716663241386414, "learning_rate": 0.0002, "epoch": 2.7470686767169177, "step": 4100}, {"loss": 1.6463, "grad_norm": 0.45420581102371216, "learning_rate": 0.0002, "epoch": 2.7537688442211055, "step": 4110}, {"loss": 1.5718, "grad_norm": 0.4680487811565399, "learning_rate": 0.0002, "epoch": 2.7604690117252932, "step": 4120}, {"loss": 1.5968, "grad_norm": 0.5375032424926758, "learning_rate": 0.0002, "epoch": 2.7671691792294806, "step": 4130}, {"loss": 1.5254, "grad_norm": 0.46026280522346497, "learning_rate": 0.0002, "epoch": 2.7738693467336684, "step": 4140}, {"loss": 1.6613, "grad_norm": 0.43658447265625, "learning_rate": 0.0002, "epoch": 2.7805695142378557, "step": 4150}, {"loss": 1.6546, "grad_norm": 0.4935547113418579, "learning_rate": 0.0002, "epoch": 2.7872696817420435, "step": 4160}, {"loss": 1.5961, "grad_norm": 0.8167962431907654, "learning_rate": 0.0002, "epoch": 2.7939698492462313, "step": 4170}, {"loss": 1.6907, "grad_norm": 0.4289683997631073, "learning_rate": 0.0002, "epoch": 2.8006700167504186, "step": 4180}, {"loss": 1.6385, "grad_norm": 0.4569324254989624, "learning_rate": 0.0002, "epoch": 2.8073701842546064, "step": 4190}, {"loss": 1.6077, "grad_norm": 0.474795937538147, "learning_rate": 0.0002, "epoch": 2.8140703517587937, "step": 4200}, {"loss": 1.6223, "grad_norm": 0.44272229075431824, "learning_rate": 0.0002, "epoch": 2.8207705192629815, "step": 4210}, {"loss": 1.6706, "grad_norm": 0.525240957736969, "learning_rate": 0.0002, "epoch": 2.8274706867671693, "step": 4220}, {"loss": 1.7196, "grad_norm": 0.4802303910255432, "learning_rate": 0.0002, "epoch": 2.8341708542713566, "step": 4230}, {"loss": 1.6002, "grad_norm": 0.46400442719459534, "learning_rate": 0.0002, "epoch": 2.8408710217755444, "step": 4240}, {"loss": 1.6052, "grad_norm": 0.49884888529777527, "learning_rate": 0.0002, "epoch": 2.8475711892797317, "step": 4250}, {"loss": 1.6919, "grad_norm": 0.5015072226524353, "learning_rate": 0.0002, "epoch": 2.8542713567839195, "step": 4260}, {"loss": 1.6335, "grad_norm": 0.4335440695285797, "learning_rate": 0.0002, "epoch": 2.8609715242881073, "step": 4270}, {"loss": 1.5664, "grad_norm": 0.5131644606590271, "learning_rate": 0.0002, "epoch": 2.8676716917922946, "step": 4280}, {"loss": 1.6409, "grad_norm": 0.6977195739746094, "learning_rate": 0.0002, "epoch": 2.8743718592964824, "step": 4290}, {"loss": 1.7192, "grad_norm": 0.5133762955665588, "learning_rate": 0.0002, "epoch": 2.8810720268006698, "step": 4300}, {"loss": 1.6257, "grad_norm": 0.4737614393234253, "learning_rate": 0.0002, "epoch": 2.8877721943048575, "step": 4310}, {"loss": 1.6076, "grad_norm": 0.4580535590648651, "learning_rate": 0.0002, "epoch": 2.8944723618090453, "step": 4320}, {"loss": 1.6538, "grad_norm": 0.43863341212272644, "learning_rate": 0.0002, "epoch": 2.901172529313233, "step": 4330}, {"loss": 1.6091, "grad_norm": 0.4103737473487854, "learning_rate": 0.0002, "epoch": 2.9078726968174204, "step": 4340}, {"loss": 1.7106, "grad_norm": 0.438014417886734, "learning_rate": 0.0002, "epoch": 2.914572864321608, "step": 4350}, {"loss": 1.6025, "grad_norm": 0.5068213939666748, "learning_rate": 0.0002, "epoch": 2.9212730318257956, "step": 4360}, {"loss": 1.6426, "grad_norm": 0.45305484533309937, "learning_rate": 0.0002, "epoch": 2.9279731993299833, "step": 4370}, {"loss": 1.5726, "grad_norm": 0.4612090289592743, "learning_rate": 0.0002, "epoch": 2.934673366834171, "step": 4380}, {"loss": 1.6536, "grad_norm": 0.508736789226532, "learning_rate": 0.0002, "epoch": 2.9413735343383585, "step": 4390}, {"loss": 1.6132, "grad_norm": 0.4924427270889282, "learning_rate": 0.0002, "epoch": 2.948073701842546, "step": 4400}, {"loss": 1.7007, "grad_norm": 0.5707460641860962, "learning_rate": 0.0002, "epoch": 2.9547738693467336, "step": 4410}, {"loss": 1.6814, "grad_norm": 0.42270299792289734, "learning_rate": 0.0002, "epoch": 2.9614740368509214, "step": 4420}, {"loss": 1.6644, "grad_norm": 0.4429931044578552, "learning_rate": 0.0002, "epoch": 2.968174204355109, "step": 4430}, {"loss": 1.6251, "grad_norm": 0.49760574102401733, "learning_rate": 0.0002, "epoch": 2.9748743718592965, "step": 4440}, {"loss": 1.6169, "grad_norm": 0.4558229148387909, "learning_rate": 0.0002, "epoch": 2.981574539363484, "step": 4450}, {"loss": 1.6055, "grad_norm": 0.39848530292510986, "learning_rate": 0.0002, "epoch": 2.9882747068676716, "step": 4460}, {"loss": 1.6705, "grad_norm": 0.5224862098693848, "learning_rate": 0.0002, "epoch": 2.9949748743718594, "step": 4470}, {"eval_loss": 1.8228833675384521, "eval_runtime": 37.9049, "eval_samples_per_second": 13.587, "eval_steps_per_second": 1.715, "epoch": 2.9996649916247904, "step": 4477}, {"loss": 1.6637, "grad_norm": 0.41169142723083496, "learning_rate": 0.0002, "epoch": 3.0016750418760467, "step": 4480}, {"loss": 1.5974, "grad_norm": 0.4865207374095917, "learning_rate": 0.0002, "epoch": 3.0083752093802345, "step": 4490}, {"loss": 1.5297, "grad_norm": 0.5462028384208679, "learning_rate": 0.0002, "epoch": 3.0150753768844223, "step": 4500}, {"loss": 1.5251, "grad_norm": 0.6169732809066772, "learning_rate": 0.0002, "epoch": 3.0217755443886096, "step": 4510}, {"loss": 1.5559, "grad_norm": 0.5667954087257385, "learning_rate": 0.0002, "epoch": 3.0284757118927974, "step": 4520}, {"loss": 1.5037, "grad_norm": 0.5758325457572937, "learning_rate": 0.0002, "epoch": 3.0351758793969847, "step": 4530}, {"loss": 1.4873, "grad_norm": 0.5220064520835876, "learning_rate": 0.0002, "epoch": 3.0418760469011725, "step": 4540}, {"loss": 1.5126, "grad_norm": 0.5469558835029602, "learning_rate": 0.0002, "epoch": 3.0485762144053603, "step": 4550}, {"loss": 1.4275, "grad_norm": 0.5680848956108093, "learning_rate": 0.0002, "epoch": 3.0552763819095476, "step": 4560}, {"loss": 1.5187, "grad_norm": 0.5906574726104736, "learning_rate": 0.0002, "epoch": 3.0619765494137354, "step": 4570}, {"loss": 1.4551, "grad_norm": 0.4725631773471832, "learning_rate": 0.0002, "epoch": 3.0686767169179228, "step": 4580}, {"loss": 1.5083, "grad_norm": 0.5273477435112, "learning_rate": 0.0002, "epoch": 3.0753768844221105, "step": 4590}, {"loss": 1.5154, "grad_norm": 0.5861203074455261, "learning_rate": 0.0002, "epoch": 3.0820770519262983, "step": 4600}, {"loss": 1.4924, "grad_norm": 0.5343965291976929, "learning_rate": 0.0002, "epoch": 3.0887772194304857, "step": 4610}, {"loss": 1.5608, "grad_norm": 0.5348150730133057, "learning_rate": 0.0002, "epoch": 3.0954773869346734, "step": 4620}, {"loss": 1.5399, "grad_norm": 0.5971846580505371, "learning_rate": 0.0002, "epoch": 3.102177554438861, "step": 4630}, {"loss": 1.4662, "grad_norm": 0.5203177332878113, "learning_rate": 0.0002, "epoch": 3.1088777219430486, "step": 4640}, {"loss": 1.5805, "grad_norm": 0.55289226770401, "learning_rate": 0.0002, "epoch": 3.1155778894472363, "step": 4650}, {"loss": 1.4745, "grad_norm": 0.6878530979156494, "learning_rate": 0.0002, "epoch": 3.1222780569514237, "step": 4660}, {"loss": 1.5335, "grad_norm": 0.6173256635665894, "learning_rate": 0.0002, "epoch": 3.1289782244556115, "step": 4670}, {"loss": 1.51, "grad_norm": 0.536796510219574, "learning_rate": 0.0002, "epoch": 3.135678391959799, "step": 4680}, {"loss": 1.4713, "grad_norm": 0.58846116065979, "learning_rate": 0.0002, "epoch": 3.1423785594639866, "step": 4690}, {"loss": 1.5114, "grad_norm": 0.645889401435852, "learning_rate": 0.0002, "epoch": 3.1490787269681744, "step": 4700}, {"loss": 1.4705, "grad_norm": 0.6118691563606262, "learning_rate": 0.0002, "epoch": 3.1557788944723617, "step": 4710}, {"loss": 1.5533, "grad_norm": 0.5189669132232666, "learning_rate": 0.0002, "epoch": 3.1624790619765495, "step": 4720}, {"loss": 1.4769, "grad_norm": 0.5794713497161865, "learning_rate": 0.0002, "epoch": 3.169179229480737, "step": 4730}, {"loss": 1.4849, "grad_norm": 0.6579326391220093, "learning_rate": 0.0002, "epoch": 3.1758793969849246, "step": 4740}, {"loss": 1.545, "grad_norm": 0.5822742581367493, "learning_rate": 0.0002, "epoch": 3.1825795644891124, "step": 4750}, {"loss": 1.4358, "grad_norm": 0.5475956201553345, "learning_rate": 0.0002, "epoch": 3.1892797319932997, "step": 4760}, {"loss": 1.4723, "grad_norm": 0.6743834018707275, "learning_rate": 0.0002, "epoch": 3.1959798994974875, "step": 4770}, {"loss": 1.5161, "grad_norm": 0.6110585927963257, "learning_rate": 0.0002, "epoch": 3.202680067001675, "step": 4780}, {"loss": 1.5455, "grad_norm": 0.5426181554794312, "learning_rate": 0.0002, "epoch": 3.2093802345058626, "step": 4790}, {"loss": 1.5315, "grad_norm": 0.6077824234962463, "learning_rate": 0.0002, "epoch": 3.2160804020100504, "step": 4800}, {"loss": 1.5314, "grad_norm": 0.5785858631134033, "learning_rate": 0.0002, "epoch": 3.2227805695142377, "step": 4810}, {"loss": 1.4041, "grad_norm": 0.6425958275794983, "learning_rate": 0.0002, "epoch": 3.2294807370184255, "step": 4820}, {"loss": 1.4751, "grad_norm": 0.6607080698013306, "learning_rate": 0.0002, "epoch": 3.236180904522613, "step": 4830}, {"loss": 1.5267, "grad_norm": 0.5385788679122925, "learning_rate": 0.0002, "epoch": 3.2428810720268006, "step": 4840}, {"loss": 1.4673, "grad_norm": 0.5630403757095337, "learning_rate": 0.0002, "epoch": 3.2495812395309884, "step": 4850}, {"loss": 1.5257, "grad_norm": 0.6340779662132263, "learning_rate": 0.0002, "epoch": 3.2562814070351758, "step": 4860}, {"loss": 1.5148, "grad_norm": 0.5305342674255371, "learning_rate": 0.0002, "epoch": 3.2629815745393635, "step": 4870}, {"loss": 1.5162, "grad_norm": 0.597670316696167, "learning_rate": 0.0002, "epoch": 3.2696817420435513, "step": 4880}, {"loss": 1.5429, "grad_norm": 0.665553867816925, "learning_rate": 0.0002, "epoch": 3.2763819095477387, "step": 4890}, {"loss": 1.4607, "grad_norm": 0.579767644405365, "learning_rate": 0.0002, "epoch": 3.2830820770519265, "step": 4900}, {"loss": 1.4999, "grad_norm": 0.5512481331825256, "learning_rate": 0.0002, "epoch": 3.289782244556114, "step": 4910}, {"loss": 1.5022, "grad_norm": 0.5916532278060913, "learning_rate": 0.0002, "epoch": 3.2964824120603016, "step": 4920}, {"loss": 1.4889, "grad_norm": 0.7521726489067078, "learning_rate": 0.0002, "epoch": 3.3031825795644894, "step": 4930}, {"loss": 1.4223, "grad_norm": 0.5352797508239746, "learning_rate": 0.0002, "epoch": 3.3098827470686767, "step": 4940}, {"loss": 1.5122, "grad_norm": 0.5950371623039246, "learning_rate": 0.0002, "epoch": 3.3165829145728645, "step": 4950}, {"loss": 1.5072, "grad_norm": 0.8020477890968323, "learning_rate": 0.0002, "epoch": 3.323283082077052, "step": 4960}, {"loss": 1.5422, "grad_norm": 0.6790024638175964, "learning_rate": 0.0002, "epoch": 3.3299832495812396, "step": 4970}, {"loss": 1.5363, "grad_norm": 0.687627375125885, "learning_rate": 0.0002, "epoch": 3.3366834170854274, "step": 4980}, {"loss": 1.5276, "grad_norm": 0.6094385385513306, "learning_rate": 0.0002, "epoch": 3.3433835845896147, "step": 4990}, {"loss": 1.549, "grad_norm": 0.6541242003440857, "learning_rate": 0.0002, "epoch": 3.3500837520938025, "step": 5000}, {"loss": 1.6067, "grad_norm": 0.5560880303382874, "learning_rate": 0.0002, "epoch": 3.35678391959799, "step": 5010}, {"loss": 1.5769, "grad_norm": 0.5440094470977783, "learning_rate": 0.0002, "epoch": 3.3634840871021776, "step": 5020}, {"loss": 1.6183, "grad_norm": 0.5749301314353943, "learning_rate": 0.0002, "epoch": 3.3701842546063654, "step": 5030}, {"loss": 1.4801, "grad_norm": 0.5919716954231262, "learning_rate": 0.0002, "epoch": 3.3768844221105527, "step": 5040}, {"loss": 1.5957, "grad_norm": 0.6331481337547302, "learning_rate": 0.0002, "epoch": 3.3835845896147405, "step": 5050}, {"loss": 1.5188, "grad_norm": 0.5687161684036255, "learning_rate": 0.0002, "epoch": 3.390284757118928, "step": 5060}, {"loss": 1.5702, "grad_norm": 0.6718577742576599, "learning_rate": 0.0002, "epoch": 3.3969849246231156, "step": 5070}, {"loss": 1.5577, "grad_norm": 0.5089324116706848, "learning_rate": 0.0002, "epoch": 3.4036850921273034, "step": 5080}, {"loss": 1.512, "grad_norm": 0.5710174441337585, "learning_rate": 0.0002, "epoch": 3.4103852596314908, "step": 5090}, {"loss": 1.5492, "grad_norm": 0.6670721173286438, "learning_rate": 0.0002, "epoch": 3.4170854271356785, "step": 5100}, {"loss": 1.5227, "grad_norm": 0.6875665187835693, "learning_rate": 0.0002, "epoch": 3.423785594639866, "step": 5110}, {"loss": 1.4496, "grad_norm": 0.5375880599021912, "learning_rate": 0.0002, "epoch": 3.4304857621440537, "step": 5120}, {"loss": 1.5527, "grad_norm": 0.6550399661064148, "learning_rate": 0.0002, "epoch": 3.4371859296482414, "step": 5130}, {"loss": 1.5687, "grad_norm": 0.5948067903518677, "learning_rate": 0.0002, "epoch": 3.4438860971524288, "step": 5140}, {"loss": 1.4813, "grad_norm": 0.6134477257728577, "learning_rate": 0.0002, "epoch": 3.4505862646566166, "step": 5150}, {"loss": 1.5069, "grad_norm": 0.6506398320198059, "learning_rate": 0.0002, "epoch": 3.457286432160804, "step": 5160}, {"loss": 1.4422, "grad_norm": 0.6060147881507874, "learning_rate": 0.0002, "epoch": 3.4639865996649917, "step": 5170}, {"loss": 1.5093, "grad_norm": 0.6173806190490723, "learning_rate": 0.0002, "epoch": 3.4706867671691795, "step": 5180}, {"loss": 1.4975, "grad_norm": 0.6032607555389404, "learning_rate": 0.0002, "epoch": 3.477386934673367, "step": 5190}, {"loss": 1.4979, "grad_norm": 0.5652492046356201, "learning_rate": 0.0002, "epoch": 3.4840871021775546, "step": 5200}, {"loss": 1.4883, "grad_norm": 0.6168607473373413, "learning_rate": 0.0002, "epoch": 3.490787269681742, "step": 5210}, {"loss": 1.5164, "grad_norm": 0.6170629262924194, "learning_rate": 0.0002, "epoch": 3.4974874371859297, "step": 5220}, {"loss": 1.4879, "grad_norm": 0.6926297545433044, "learning_rate": 0.0002, "epoch": 3.5041876046901175, "step": 5230}, {"loss": 1.4982, "grad_norm": 0.6702437996864319, "learning_rate": 0.0002, "epoch": 3.510887772194305, "step": 5240}, {"loss": 1.4986, "grad_norm": 0.5421436429023743, "learning_rate": 0.0002, "epoch": 3.5175879396984926, "step": 5250}, {"loss": 1.4673, "grad_norm": 0.5726765990257263, "learning_rate": 0.0002, "epoch": 3.52428810720268, "step": 5260}, {"loss": 1.5423, "grad_norm": 0.5685455203056335, "learning_rate": 0.0002, "epoch": 3.5309882747068677, "step": 5270}, {"loss": 1.4715, "grad_norm": 0.6018396019935608, "learning_rate": 0.0002, "epoch": 3.5376884422110555, "step": 5280}, {"loss": 1.5451, "grad_norm": 0.5731932520866394, "learning_rate": 0.0002, "epoch": 3.544388609715243, "step": 5290}, {"loss": 1.4752, "grad_norm": 0.6601519584655762, "learning_rate": 0.0002, "epoch": 3.5510887772194306, "step": 5300}, {"loss": 1.5434, "grad_norm": 0.5545530319213867, "learning_rate": 0.0002, "epoch": 3.557788944723618, "step": 5310}, {"loss": 1.5438, "grad_norm": 0.5998541116714478, "learning_rate": 0.0002, "epoch": 3.5644891122278057, "step": 5320}, {"loss": 1.56, "grad_norm": 0.5651767253875732, "learning_rate": 0.0002, "epoch": 3.5711892797319935, "step": 5330}, {"loss": 1.4829, "grad_norm": 0.7425084114074707, "learning_rate": 0.0002, "epoch": 3.577889447236181, "step": 5340}, {"loss": 1.5571, "grad_norm": 0.5770602226257324, "learning_rate": 0.0002, "epoch": 3.5845896147403686, "step": 5350}, {"loss": 1.458, "grad_norm": 0.54723060131073, "learning_rate": 0.0002, "epoch": 3.591289782244556, "step": 5360}, {"loss": 1.497, "grad_norm": 0.6658238172531128, "learning_rate": 0.0002, "epoch": 3.5979899497487438, "step": 5370}, {"loss": 1.5456, "grad_norm": 0.5787645578384399, "learning_rate": 0.0002, "epoch": 3.6046901172529315, "step": 5380}, {"loss": 1.5343, "grad_norm": 0.594913125038147, "learning_rate": 0.0002, "epoch": 3.611390284757119, "step": 5390}, {"loss": 1.4727, "grad_norm": 0.4964977502822876, "learning_rate": 0.0002, "epoch": 3.6180904522613067, "step": 5400}, {"loss": 1.5062, "grad_norm": 0.6087527275085449, "learning_rate": 0.0002, "epoch": 3.624790619765494, "step": 5410}, {"loss": 1.5098, "grad_norm": 0.6315323710441589, "learning_rate": 0.0002, "epoch": 3.6314907872696818, "step": 5420}, {"loss": 1.4855, "grad_norm": 0.574799120426178, "learning_rate": 0.0002, "epoch": 3.6381909547738696, "step": 5430}, {"loss": 1.4595, "grad_norm": 0.5949277877807617, "learning_rate": 0.0002, "epoch": 3.644891122278057, "step": 5440}, {"loss": 1.4816, "grad_norm": 0.5640677213668823, "learning_rate": 0.0002, "epoch": 3.6515912897822447, "step": 5450}, {"loss": 1.525, "grad_norm": 0.6198237538337708, "learning_rate": 0.0002, "epoch": 3.658291457286432, "step": 5460}, {"loss": 1.5676, "grad_norm": 0.6902034878730774, "learning_rate": 0.0002, "epoch": 3.66499162479062, "step": 5470}, {"loss": 1.544, "grad_norm": 0.5686674118041992, "learning_rate": 0.0002, "epoch": 3.6716917922948076, "step": 5480}, {"loss": 1.5255, "grad_norm": 0.6532107591629028, "learning_rate": 0.0002, "epoch": 3.678391959798995, "step": 5490}, {"loss": 1.5767, "grad_norm": 0.5790849924087524, "learning_rate": 0.0002, "epoch": 3.6850921273031827, "step": 5500}, {"loss": 1.5507, "grad_norm": 0.6055065393447876, "learning_rate": 0.0002, "epoch": 3.69179229480737, "step": 5510}, {"loss": 1.4656, "grad_norm": 0.5630605816841125, "learning_rate": 0.0002, "epoch": 3.698492462311558, "step": 5520}, {"loss": 1.537, "grad_norm": 0.6005825996398926, "learning_rate": 0.0002, "epoch": 3.7051926298157456, "step": 5530}, {"loss": 1.5313, "grad_norm": 0.6553038954734802, "learning_rate": 0.0002, "epoch": 3.711892797319933, "step": 5540}, {"loss": 1.4943, "grad_norm": 0.5601094961166382, "learning_rate": 0.0002, "epoch": 3.7185929648241207, "step": 5550}, {"loss": 1.5161, "grad_norm": 0.6598808169364929, "learning_rate": 0.0002, "epoch": 3.725293132328308, "step": 5560}, {"loss": 1.5345, "grad_norm": 0.5506255626678467, "learning_rate": 0.0002, "epoch": 3.731993299832496, "step": 5570}, {"loss": 1.4805, "grad_norm": 0.6001223921775818, "learning_rate": 0.0002, "epoch": 3.7386934673366836, "step": 5580}, {"loss": 1.4652, "grad_norm": 0.6287297606468201, "learning_rate": 0.0002, "epoch": 3.745393634840871, "step": 5590}, {"loss": 1.5246, "grad_norm": 0.6253238916397095, "learning_rate": 0.0002, "epoch": 3.7520938023450587, "step": 5600}, {"loss": 1.5691, "grad_norm": 0.5713174939155579, "learning_rate": 0.0002, "epoch": 3.758793969849246, "step": 5610}, {"loss": 1.5661, "grad_norm": 0.6198310852050781, "learning_rate": 0.0002, "epoch": 3.765494137353434, "step": 5620}, {"loss": 1.5448, "grad_norm": 0.5941224098205566, "learning_rate": 0.0002, "epoch": 3.7721943048576216, "step": 5630}, {"loss": 1.4925, "grad_norm": 0.606002151966095, "learning_rate": 0.0002, "epoch": 3.778894472361809, "step": 5640}, {"loss": 1.5182, "grad_norm": 0.6540704965591431, "learning_rate": 0.0002, "epoch": 3.7855946398659968, "step": 5650}, {"loss": 1.5903, "grad_norm": 0.6147415041923523, "learning_rate": 0.0002, "epoch": 3.792294807370184, "step": 5660}, {"loss": 1.5329, "grad_norm": 0.5649605393409729, "learning_rate": 0.0002, "epoch": 3.798994974874372, "step": 5670}, {"loss": 1.5747, "grad_norm": 0.6788773536682129, "learning_rate": 0.0002, "epoch": 3.8056951423785597, "step": 5680}, {"loss": 1.535, "grad_norm": 0.6581860780715942, "learning_rate": 0.0002, "epoch": 3.812395309882747, "step": 5690}, {"loss": 1.4587, "grad_norm": 0.5529348850250244, "learning_rate": 0.0002, "epoch": 3.819095477386935, "step": 5700}, {"loss": 1.5594, "grad_norm": 0.6320232152938843, "learning_rate": 0.0002, "epoch": 3.825795644891122, "step": 5710}, {"loss": 1.4696, "grad_norm": 0.6529698371887207, "learning_rate": 0.0002, "epoch": 3.83249581239531, "step": 5720}, {"loss": 1.5854, "grad_norm": 0.5983362793922424, "learning_rate": 0.0002, "epoch": 3.8391959798994977, "step": 5730}, {"loss": 1.465, "grad_norm": 0.6335684061050415, "learning_rate": 0.0002, "epoch": 3.845896147403685, "step": 5740}, {"loss": 1.5545, "grad_norm": 0.700446605682373, "learning_rate": 0.0002, "epoch": 3.852596314907873, "step": 5750}, {"loss": 1.5707, "grad_norm": 0.6092597842216492, "learning_rate": 0.0002, "epoch": 3.85929648241206, "step": 5760}, {"loss": 1.5729, "grad_norm": 0.564146101474762, "learning_rate": 0.0002, "epoch": 3.865996649916248, "step": 5770}, {"loss": 1.5872, "grad_norm": 0.615275502204895, "learning_rate": 0.0002, "epoch": 3.8726968174204357, "step": 5780}, {"loss": 1.5142, "grad_norm": 0.6685376763343811, "learning_rate": 0.0002, "epoch": 3.879396984924623, "step": 5790}, {"loss": 1.4752, "grad_norm": 0.6116922497749329, "learning_rate": 0.0002, "epoch": 3.886097152428811, "step": 5800}, {"loss": 1.5179, "grad_norm": 0.5486813187599182, "learning_rate": 0.0002, "epoch": 3.892797319932998, "step": 5810}, {"loss": 1.5167, "grad_norm": 0.6208204030990601, "learning_rate": 0.0002, "epoch": 3.899497487437186, "step": 5820}, {"loss": 1.5334, "grad_norm": 0.6500625014305115, "learning_rate": 0.0002, "epoch": 3.9061976549413737, "step": 5830}, {"loss": 1.4716, "grad_norm": 0.5948089361190796, "learning_rate": 0.0002, "epoch": 3.912897822445561, "step": 5840}, {"loss": 1.6011, "grad_norm": 0.7210732698440552, "learning_rate": 0.0002, "epoch": 3.919597989949749, "step": 5850}, {"loss": 1.5519, "grad_norm": 0.6662322878837585, "learning_rate": 0.0002, "epoch": 3.926298157453936, "step": 5860}, {"loss": 1.5656, "grad_norm": 0.5613839626312256, "learning_rate": 0.0002, "epoch": 3.932998324958124, "step": 5870}, {"loss": 1.544, "grad_norm": 0.6069002151489258, "learning_rate": 0.0002, "epoch": 3.9396984924623117, "step": 5880}, {"loss": 1.6745, "grad_norm": 0.7075562477111816, "learning_rate": 0.0002, "epoch": 3.946398659966499, "step": 5890}, {"loss": 1.5391, "grad_norm": 0.6316173076629639, "learning_rate": 0.0002, "epoch": 3.953098827470687, "step": 5900}, {"loss": 1.6314, "grad_norm": 0.5716308355331421, "learning_rate": 0.0002, "epoch": 3.959798994974874, "step": 5910}, {"loss": 1.5947, "grad_norm": 0.6800096035003662, "learning_rate": 0.0002, "epoch": 3.966499162479062, "step": 5920}, {"loss": 1.5189, "grad_norm": 0.6057983040809631, "learning_rate": 0.0002, "epoch": 3.9731993299832498, "step": 5930}, {"loss": 1.5431, "grad_norm": 0.5938987731933594, "learning_rate": 0.0002, "epoch": 3.979899497487437, "step": 5940}, {"loss": 1.5111, "grad_norm": 0.6963576674461365, "learning_rate": 0.0002, "epoch": 3.986599664991625, "step": 5950}, {"loss": 1.5521, "grad_norm": 0.6279940009117126, "learning_rate": 0.0002, "epoch": 3.993299832495812, "step": 5960}, {"loss": 1.5974, "grad_norm": 0.7161159515380859, "learning_rate": 0.0002, "epoch": 4.0, "step": 5970}, {"eval_loss": 1.8655421733856201, "eval_runtime": 37.9276, "eval_samples_per_second": 13.579, "eval_steps_per_second": 1.714, "epoch": 4.0, "step": 5970}, {"loss": 1.3666, "grad_norm": 0.7380476593971252, "learning_rate": 0.0002, "epoch": 4.006700167504188, "step": 5980}, {"loss": 1.3913, "grad_norm": 0.7148947715759277, "learning_rate": 0.0002, "epoch": 4.013400335008376, "step": 5990}, {"loss": 1.4204, "grad_norm": 0.6177082657814026, "learning_rate": 0.0002, "epoch": 4.0201005025125625, "step": 6000}, {"loss": 1.4421, "grad_norm": 0.8552946448326111, "learning_rate": 0.0002, "epoch": 4.02680067001675, "step": 6010}, {"loss": 1.4342, "grad_norm": 0.8033416271209717, "learning_rate": 0.0002, "epoch": 4.033500837520938, "step": 6020}, {"loss": 1.4092, "grad_norm": 0.8501318097114563, "learning_rate": 0.0002, "epoch": 4.040201005025126, "step": 6030}, {"loss": 1.3367, "grad_norm": 0.6981393098831177, "learning_rate": 0.0002, "epoch": 4.046901172529314, "step": 6040}, {"loss": 1.3925, "grad_norm": 0.7227180600166321, "learning_rate": 0.0002, "epoch": 4.0536013400335005, "step": 6050}, {"loss": 1.4007, "grad_norm": 0.6923989653587341, "learning_rate": 0.0002, "epoch": 4.060301507537688, "step": 6060}, {"loss": 1.3837, "grad_norm": 0.879779040813446, "learning_rate": 0.0002, "epoch": 4.067001675041876, "step": 6070}, {"loss": 1.4383, "grad_norm": 0.8184754848480225, "learning_rate": 0.0002, "epoch": 4.073701842546064, "step": 6080}, {"loss": 1.3128, "grad_norm": 0.8211342692375183, "learning_rate": 0.0002, "epoch": 4.080402010050252, "step": 6090}, {"loss": 1.3892, "grad_norm": 0.7542396783828735, "learning_rate": 0.0002, "epoch": 4.0871021775544385, "step": 6100}, {"loss": 1.3607, "grad_norm": 0.6631066799163818, "learning_rate": 0.0002, "epoch": 4.093802345058626, "step": 6110}, {"loss": 1.3275, "grad_norm": 0.6728386282920837, "learning_rate": 0.0002, "epoch": 4.100502512562814, "step": 6120}, {"loss": 1.3443, "grad_norm": 0.681851863861084, "learning_rate": 0.0002, "epoch": 4.107202680067002, "step": 6130}, {"loss": 1.3486, "grad_norm": 0.8757794499397278, "learning_rate": 0.0002, "epoch": 4.11390284757119, "step": 6140}, {"loss": 1.351, "grad_norm": 0.6567301750183105, "learning_rate": 0.0002, "epoch": 4.1206030150753765, "step": 6150}, {"loss": 1.3824, "grad_norm": 0.7950329184532166, "learning_rate": 0.0002, "epoch": 4.127303182579564, "step": 6160}, {"loss": 1.3738, "grad_norm": 0.7545644044876099, "learning_rate": 0.0002, "epoch": 4.134003350083752, "step": 6170}, {"loss": 1.4214, "grad_norm": 0.7172710299491882, "learning_rate": 0.0002, "epoch": 4.14070351758794, "step": 6180}, {"loss": 1.4091, "grad_norm": 0.7040584087371826, "learning_rate": 0.0002, "epoch": 4.147403685092128, "step": 6190}, {"loss": 1.4149, "grad_norm": 0.7482913732528687, "learning_rate": 0.0002, "epoch": 4.1541038525963145, "step": 6200}, {"loss": 1.3227, "grad_norm": 0.8523276448249817, "learning_rate": 0.0002, "epoch": 4.160804020100502, "step": 6210}, {"loss": 1.4194, "grad_norm": 0.6672041416168213, "learning_rate": 0.0002, "epoch": 4.16750418760469, "step": 6220}, {"loss": 1.3953, "grad_norm": 0.7523500919342041, "learning_rate": 0.0002, "epoch": 4.174204355108878, "step": 6230}, {"loss": 1.371, "grad_norm": 0.8085253834724426, "learning_rate": 0.0002, "epoch": 4.180904522613066, "step": 6240}, {"loss": 1.3293, "grad_norm": 0.789450466632843, "learning_rate": 0.0002, "epoch": 4.187604690117253, "step": 6250}, {"loss": 1.3539, "grad_norm": 0.7502310872077942, "learning_rate": 0.0002, "epoch": 4.19430485762144, "step": 6260}, {"loss": 1.3415, "grad_norm": 0.7397456765174866, "learning_rate": 0.0002, "epoch": 4.201005025125628, "step": 6270}, {"loss": 1.3963, "grad_norm": 0.6921947002410889, "learning_rate": 0.0002, "epoch": 4.207705192629816, "step": 6280}, {"loss": 1.3125, "grad_norm": 0.9334571957588196, "learning_rate": 0.0002, "epoch": 4.214405360134004, "step": 6290}, {"loss": 1.3612, "grad_norm": 0.725799024105072, "learning_rate": 0.0002, "epoch": 4.221105527638191, "step": 6300}, {"loss": 1.4217, "grad_norm": 0.8290495872497559, "learning_rate": 0.0002, "epoch": 4.227805695142378, "step": 6310}, {"loss": 1.4135, "grad_norm": 0.688983678817749, "learning_rate": 0.0002, "epoch": 4.234505862646566, "step": 6320}, {"loss": 1.3807, "grad_norm": 0.8620913028717041, "learning_rate": 0.0002, "epoch": 4.241206030150754, "step": 6330}, {"loss": 1.3738, "grad_norm": 0.8008657693862915, "learning_rate": 0.0002, "epoch": 4.247906197654942, "step": 6340}, {"loss": 1.4005, "grad_norm": 0.7379199266433716, "learning_rate": 0.0002, "epoch": 4.254606365159129, "step": 6350}, {"loss": 1.426, "grad_norm": 0.7842815518379211, "learning_rate": 0.0002, "epoch": 4.261306532663316, "step": 6360}, {"loss": 1.4262, "grad_norm": 0.812600314617157, "learning_rate": 0.0002, "epoch": 4.268006700167504, "step": 6370}, {"loss": 1.4028, "grad_norm": 0.7852841019630432, "learning_rate": 0.0002, "epoch": 4.274706867671692, "step": 6380}, {"loss": 1.3722, "grad_norm": 1.0377534627914429, "learning_rate": 0.0002, "epoch": 4.28140703517588, "step": 6390}, {"loss": 1.3755, "grad_norm": 1.03935706615448, "learning_rate": 0.0002, "epoch": 4.288107202680067, "step": 6400}, {"loss": 1.3961, "grad_norm": 0.7244732975959778, "learning_rate": 0.0002, "epoch": 4.294807370184254, "step": 6410}, {"loss": 1.4608, "grad_norm": 0.7137406468391418, "learning_rate": 0.0002, "epoch": 4.301507537688442, "step": 6420}, {"loss": 1.4461, "grad_norm": 0.7492543458938599, "learning_rate": 0.0002, "epoch": 4.30820770519263, "step": 6430}, {"loss": 1.4562, "grad_norm": 0.7065439224243164, "learning_rate": 0.0002, "epoch": 4.314907872696818, "step": 6440}, {"loss": 1.4246, "grad_norm": 0.7786989808082581, "learning_rate": 0.0002, "epoch": 4.321608040201005, "step": 6450}, {"loss": 1.3098, "grad_norm": 0.7369208335876465, "learning_rate": 0.0002, "epoch": 4.328308207705192, "step": 6460}, {"loss": 1.3686, "grad_norm": 0.7412346005439758, "learning_rate": 0.0002, "epoch": 4.33500837520938, "step": 6470}, {"loss": 1.4087, "grad_norm": 0.780927300453186, "learning_rate": 0.0002, "epoch": 4.341708542713568, "step": 6480}, {"loss": 1.3628, "grad_norm": 0.8320930600166321, "learning_rate": 0.0002, "epoch": 4.348408710217756, "step": 6490}, {"loss": 1.3715, "grad_norm": 0.6871094703674316, "learning_rate": 0.0002, "epoch": 4.355108877721943, "step": 6500}, {"loss": 1.3257, "grad_norm": 0.6751559972763062, "learning_rate": 0.0002, "epoch": 4.36180904522613, "step": 6510}, {"loss": 1.4311, "grad_norm": 0.7723976969718933, "learning_rate": 0.0002, "epoch": 4.368509212730318, "step": 6520}, {"loss": 1.4086, "grad_norm": 0.7915401458740234, "learning_rate": 0.0002, "epoch": 4.375209380234506, "step": 6530}, {"loss": 1.3973, "grad_norm": 0.7329102754592896, "learning_rate": 0.0002, "epoch": 4.381909547738694, "step": 6540}, {"loss": 1.447, "grad_norm": 0.7388760447502136, "learning_rate": 0.0002, "epoch": 4.388609715242881, "step": 6550}, {"loss": 1.4378, "grad_norm": 0.8282579183578491, "learning_rate": 0.0002, "epoch": 4.3953098827470685, "step": 6560}, {"loss": 1.3923, "grad_norm": 0.7192724347114563, "learning_rate": 0.0002, "epoch": 4.402010050251256, "step": 6570}, {"loss": 1.4141, "grad_norm": 0.746526837348938, "learning_rate": 0.0002, "epoch": 4.408710217755444, "step": 6580}, {"loss": 1.33, "grad_norm": 0.8738046288490295, "learning_rate": 0.0002, "epoch": 4.415410385259632, "step": 6590}, {"loss": 1.3995, "grad_norm": 0.8408458828926086, "learning_rate": 0.0002, "epoch": 4.422110552763819, "step": 6600}, {"loss": 1.4148, "grad_norm": 0.8110666275024414, "learning_rate": 0.0002, "epoch": 4.4288107202680065, "step": 6610}, {"loss": 1.441, "grad_norm": 0.8602406978607178, "learning_rate": 0.0002, "epoch": 4.435510887772194, "step": 6620}, {"loss": 1.4319, "grad_norm": 0.7549102902412415, "learning_rate": 0.0002, "epoch": 4.442211055276382, "step": 6630}, {"loss": 1.388, "grad_norm": 0.7831804156303406, "learning_rate": 0.0002, "epoch": 4.44891122278057, "step": 6640}, {"loss": 1.4283, "grad_norm": 0.7269673943519592, "learning_rate": 0.0002, "epoch": 4.455611390284757, "step": 6650}, {"loss": 1.4132, "grad_norm": 0.7397838830947876, "learning_rate": 0.0002, "epoch": 4.4623115577889445, "step": 6660}, {"loss": 1.3174, "grad_norm": 0.713707447052002, "learning_rate": 0.0002, "epoch": 4.469011725293132, "step": 6670}, {"loss": 1.3406, "grad_norm": 0.7525581121444702, "learning_rate": 0.0002, "epoch": 4.47571189279732, "step": 6680}, {"loss": 1.4283, "grad_norm": 0.8030191659927368, "learning_rate": 0.0002, "epoch": 4.482412060301508, "step": 6690}, {"loss": 1.4586, "grad_norm": 0.7469439506530762, "learning_rate": 0.0002, "epoch": 4.489112227805695, "step": 6700}, {"loss": 1.367, "grad_norm": 0.7743868231773376, "learning_rate": 0.0002, "epoch": 4.4958123953098825, "step": 6710}, {"loss": 1.3439, "grad_norm": 0.6539737582206726, "learning_rate": 0.0002, "epoch": 4.50251256281407, "step": 6720}, {"loss": 1.4513, "grad_norm": 0.825818657875061, "learning_rate": 0.0002, "epoch": 4.509212730318258, "step": 6730}, {"loss": 1.3984, "grad_norm": 0.8048575520515442, "learning_rate": 0.0002, "epoch": 4.515912897822446, "step": 6740}, {"loss": 1.3923, "grad_norm": 0.7828766107559204, "learning_rate": 0.0002, "epoch": 4.522613065326633, "step": 6750}, {"loss": 1.3886, "grad_norm": 0.7406010031700134, "learning_rate": 0.0002, "epoch": 4.5293132328308205, "step": 6760}, {"loss": 1.3109, "grad_norm": 0.840345561504364, "learning_rate": 0.0002, "epoch": 4.536013400335008, "step": 6770}, {"loss": 1.4808, "grad_norm": 0.8492622971534729, "learning_rate": 0.0002, "epoch": 4.542713567839196, "step": 6780}, {"loss": 1.4384, "grad_norm": 0.7130163908004761, "learning_rate": 0.0002, "epoch": 4.549413735343384, "step": 6790}, {"loss": 1.4531, "grad_norm": 0.8454728126525879, "learning_rate": 0.0002, "epoch": 4.556113902847571, "step": 6800}, {"loss": 1.3239, "grad_norm": 0.7847645282745361, "learning_rate": 0.0002, "epoch": 4.562814070351759, "step": 6810}, {"loss": 1.4181, "grad_norm": 0.7245864272117615, "learning_rate": 0.0002, "epoch": 4.569514237855946, "step": 6820}, {"loss": 1.3233, "grad_norm": 0.768893301486969, "learning_rate": 0.0002, "epoch": 4.576214405360134, "step": 6830}, {"loss": 1.3932, "grad_norm": 0.8028400540351868, "learning_rate": 0.0002, "epoch": 4.582914572864322, "step": 6840}, {"loss": 1.3745, "grad_norm": 0.763945460319519, "learning_rate": 0.0002, "epoch": 4.589614740368509, "step": 6850}, {"loss": 1.4797, "grad_norm": 0.7417685389518738, "learning_rate": 0.0002, "epoch": 4.596314907872697, "step": 6860}, {"loss": 1.4468, "grad_norm": 0.7603038549423218, "learning_rate": 0.0002, "epoch": 4.603015075376884, "step": 6870}, {"loss": 1.4095, "grad_norm": 0.7981528043746948, "learning_rate": 0.0002, "epoch": 4.609715242881072, "step": 6880}, {"loss": 1.3963, "grad_norm": 0.8077111840248108, "learning_rate": 0.0002, "epoch": 4.61641541038526, "step": 6890}, {"loss": 1.4721, "grad_norm": 0.8778454065322876, "learning_rate": 0.0002, "epoch": 4.623115577889447, "step": 6900}, {"loss": 1.3758, "grad_norm": 0.8620710372924805, "learning_rate": 0.0002, "epoch": 4.629815745393635, "step": 6910}, {"loss": 1.344, "grad_norm": 0.7486072778701782, "learning_rate": 0.0002, "epoch": 4.636515912897822, "step": 6920}, {"loss": 1.3913, "grad_norm": 0.7493042945861816, "learning_rate": 0.0002, "epoch": 4.64321608040201, "step": 6930}, {"loss": 1.397, "grad_norm": 0.7388978600502014, "learning_rate": 0.0002, "epoch": 4.649916247906198, "step": 6940}, {"loss": 1.3593, "grad_norm": 0.798530638217926, "learning_rate": 0.0002, "epoch": 4.656616415410385, "step": 6950}, {"loss": 1.3982, "grad_norm": 0.7929500937461853, "learning_rate": 0.0002, "epoch": 4.663316582914573, "step": 6960}, {"loss": 1.4183, "grad_norm": 0.9186785221099854, "learning_rate": 0.0002, "epoch": 4.67001675041876, "step": 6970}, {"loss": 1.3955, "grad_norm": 1.1103485822677612, "learning_rate": 0.0002, "epoch": 4.676716917922948, "step": 6980}, {"loss": 1.3941, "grad_norm": 0.8000466823577881, "learning_rate": 0.0002, "epoch": 4.683417085427136, "step": 6990}, {"loss": 1.371, "grad_norm": 0.7520599961280823, "learning_rate": 0.0002, "epoch": 4.690117252931323, "step": 7000}, {"loss": 1.4582, "grad_norm": 0.7971973419189453, "learning_rate": 0.0002, "epoch": 4.696817420435511, "step": 7010}, {"loss": 1.3682, "grad_norm": 0.7363343834877014, "learning_rate": 0.0002, "epoch": 4.703517587939698, "step": 7020}, {"loss": 1.3889, "grad_norm": 0.8268865942955017, "learning_rate": 0.0002, "epoch": 4.710217755443886, "step": 7030}, {"loss": 1.4382, "grad_norm": 0.7054963111877441, "learning_rate": 0.0002, "epoch": 4.716917922948074, "step": 7040}, {"loss": 1.4578, "grad_norm": 0.8196262121200562, "learning_rate": 0.0002, "epoch": 4.723618090452261, "step": 7050}, {"loss": 1.365, "grad_norm": 0.8276031017303467, "learning_rate": 0.0002, "epoch": 4.730318257956449, "step": 7060}, {"loss": 1.3887, "grad_norm": 0.8248157501220703, "learning_rate": 0.0002, "epoch": 4.7370184254606365, "step": 7070}, {"loss": 1.4193, "grad_norm": 0.8937979936599731, "learning_rate": 0.0002, "epoch": 4.743718592964824, "step": 7080}, {"loss": 1.4334, "grad_norm": 0.7984238862991333, "learning_rate": 0.0002, "epoch": 4.750418760469012, "step": 7090}, {"loss": 1.4385, "grad_norm": 0.9495313763618469, "learning_rate": 0.0002, "epoch": 4.757118927973199, "step": 7100}, {"loss": 1.4504, "grad_norm": 0.8598204255104065, "learning_rate": 0.0002, "epoch": 4.763819095477387, "step": 7110}, {"loss": 1.3969, "grad_norm": 0.8951472640037537, "learning_rate": 0.0002, "epoch": 4.7705192629815745, "step": 7120}, {"loss": 1.4339, "grad_norm": 0.9110309481620789, "learning_rate": 0.0002, "epoch": 4.777219430485762, "step": 7130}, {"loss": 1.4001, "grad_norm": 0.7929584980010986, "learning_rate": 0.0002, "epoch": 4.78391959798995, "step": 7140}, {"loss": 1.467, "grad_norm": 0.7415322661399841, "learning_rate": 0.0002, "epoch": 4.790619765494137, "step": 7150}, {"loss": 1.5107, "grad_norm": 0.7504757046699524, "learning_rate": 0.0002, "epoch": 4.797319932998325, "step": 7160}, {"loss": 1.3736, "grad_norm": 0.7166924476623535, "learning_rate": 0.0002, "epoch": 4.8040201005025125, "step": 7170}, {"loss": 1.4088, "grad_norm": 0.7728400826454163, "learning_rate": 0.0002, "epoch": 4.8107202680067, "step": 7180}, {"loss": 1.3814, "grad_norm": 0.7992154955863953, "learning_rate": 0.0002, "epoch": 4.817420435510888, "step": 7190}, {"loss": 1.3958, "grad_norm": 0.8655321002006531, "learning_rate": 0.0002, "epoch": 4.824120603015075, "step": 7200}, {"loss": 1.3837, "grad_norm": 0.7672632336616516, "learning_rate": 0.0002, "epoch": 4.830820770519263, "step": 7210}, {"loss": 1.4578, "grad_norm": 0.708416223526001, "learning_rate": 0.0002, "epoch": 4.8375209380234505, "step": 7220}, {"loss": 1.5413, "grad_norm": 0.8914081454277039, "learning_rate": 0.0002, "epoch": 4.844221105527638, "step": 7230}, {"loss": 1.3569, "grad_norm": 0.7141931653022766, "learning_rate": 0.0002, "epoch": 4.850921273031826, "step": 7240}, {"loss": 1.4532, "grad_norm": 0.6913040280342102, "learning_rate": 0.0002, "epoch": 4.857621440536013, "step": 7250}, {"loss": 1.3912, "grad_norm": 0.7871233820915222, "learning_rate": 0.0002, "epoch": 4.864321608040201, "step": 7260}, {"loss": 1.3688, "grad_norm": 0.8466277122497559, "learning_rate": 0.0002, "epoch": 4.8710217755443885, "step": 7270}, {"loss": 1.33, "grad_norm": 0.8492183685302734, "learning_rate": 0.0002, "epoch": 4.877721943048576, "step": 7280}, {"loss": 1.3744, "grad_norm": 0.8339574933052063, "learning_rate": 0.0002, "epoch": 4.884422110552764, "step": 7290}, {"loss": 1.4157, "grad_norm": 0.787022590637207, "learning_rate": 0.0002, "epoch": 4.891122278056951, "step": 7300}, {"loss": 1.3725, "grad_norm": 0.8877332806587219, "learning_rate": 0.0002, "epoch": 4.897822445561139, "step": 7310}, {"loss": 1.3968, "grad_norm": 0.744989812374115, "learning_rate": 0.0002, "epoch": 4.9045226130653266, "step": 7320}, {"loss": 1.4421, "grad_norm": 0.8027268648147583, "learning_rate": 0.0002, "epoch": 4.911222780569514, "step": 7330}, {"loss": 1.425, "grad_norm": 0.6437455415725708, "learning_rate": 0.0002, "epoch": 4.917922948073702, "step": 7340}, {"loss": 1.4829, "grad_norm": 0.685999870300293, "learning_rate": 0.0002, "epoch": 4.924623115577889, "step": 7350}, {"loss": 1.4352, "grad_norm": 0.9086187481880188, "learning_rate": 0.0002, "epoch": 4.931323283082077, "step": 7360}, {"loss": 1.4245, "grad_norm": 0.8272411227226257, "learning_rate": 0.0002, "epoch": 4.938023450586265, "step": 7370}, {"loss": 1.4226, "grad_norm": 0.9227852821350098, "learning_rate": 0.0002, "epoch": 4.944723618090452, "step": 7380}, {"loss": 1.3643, "grad_norm": 0.7688441276550293, "learning_rate": 0.0002, "epoch": 4.95142378559464, "step": 7390}, {"loss": 1.4491, "grad_norm": 0.8662643432617188, "learning_rate": 0.0002, "epoch": 4.958123953098827, "step": 7400}, {"loss": 1.4194, "grad_norm": 0.9234127998352051, "learning_rate": 0.0002, "epoch": 4.964824120603015, "step": 7410}, {"loss": 1.4009, "grad_norm": 0.9131470918655396, "learning_rate": 0.0002, "epoch": 4.971524288107203, "step": 7420}, {"loss": 1.4544, "grad_norm": 0.7377504110336304, "learning_rate": 0.0002, "epoch": 4.97822445561139, "step": 7430}, {"loss": 1.4008, "grad_norm": 0.8762801289558411, "learning_rate": 0.0002, "epoch": 4.984924623115578, "step": 7440}, {"loss": 1.4304, "grad_norm": 0.7919872999191284, "learning_rate": 0.0002, "epoch": 4.991624790619765, "step": 7450}, {"loss": 1.3817, "grad_norm": 0.7144299149513245, "learning_rate": 0.0002, "epoch": 4.998324958123953, "step": 7460}, {"eval_loss": 1.9291157722473145, "eval_runtime": 37.9831, "eval_samples_per_second": 13.559, "eval_steps_per_second": 1.711, "epoch": 4.99966499162479, "step": 7462}, {"loss": 1.2753, "grad_norm": 0.7860151529312134, "learning_rate": 0.0002, "epoch": 5.005025125628141, "step": 7470}, {"loss": 1.2149, "grad_norm": 0.9418314695358276, "learning_rate": 0.0002, "epoch": 5.011725293132328, "step": 7480}, {"loss": 1.1966, "grad_norm": 0.8474572896957397, "learning_rate": 0.0002, "epoch": 5.018425460636516, "step": 7490}, {"loss": 1.2111, "grad_norm": 1.0724040269851685, "learning_rate": 0.0002, "epoch": 5.025125628140704, "step": 7500}, {"loss": 1.2228, "grad_norm": 0.9109148979187012, "learning_rate": 0.0002, "epoch": 5.031825795644891, "step": 7510}, {"loss": 1.2239, "grad_norm": 1.0088659524917603, "learning_rate": 0.0002, "epoch": 5.038525963149079, "step": 7520}, {"loss": 1.2156, "grad_norm": 1.1421623229980469, "learning_rate": 0.0002, "epoch": 5.045226130653266, "step": 7530}, {"loss": 1.1739, "grad_norm": 0.9219902157783508, "learning_rate": 0.0002, "epoch": 5.051926298157454, "step": 7540}, {"loss": 1.2686, "grad_norm": 0.9150987863540649, "learning_rate": 0.0002, "epoch": 5.058626465661642, "step": 7550}, {"loss": 1.2068, "grad_norm": 0.8889328241348267, "learning_rate": 0.0002, "epoch": 5.065326633165829, "step": 7560}, {"loss": 1.276, "grad_norm": 0.9751363396644592, "learning_rate": 0.0002, "epoch": 5.072026800670017, "step": 7570}, {"loss": 1.2078, "grad_norm": 0.8603123426437378, "learning_rate": 0.0002, "epoch": 5.078726968174204, "step": 7580}, {"loss": 1.2175, "grad_norm": 0.8910616636276245, "learning_rate": 0.0002, "epoch": 5.085427135678392, "step": 7590}, {"loss": 1.2475, "grad_norm": 1.1128392219543457, "learning_rate": 0.0002, "epoch": 5.09212730318258, "step": 7600}, {"loss": 1.3065, "grad_norm": 0.9480258822441101, "learning_rate": 0.0002, "epoch": 5.098827470686767, "step": 7610}, {"loss": 1.193, "grad_norm": 0.906958818435669, "learning_rate": 0.0002, "epoch": 5.105527638190955, "step": 7620}, {"loss": 1.2223, "grad_norm": 0.8741167187690735, "learning_rate": 0.0002, "epoch": 5.1122278056951425, "step": 7630}, {"loss": 1.2126, "grad_norm": 0.966268002986908, "learning_rate": 0.0002, "epoch": 5.11892797319933, "step": 7640}, {"loss": 1.2782, "grad_norm": 0.9124358892440796, "learning_rate": 0.0002, "epoch": 5.125628140703517, "step": 7650}, {"loss": 1.3004, "grad_norm": 1.0436606407165527, "learning_rate": 0.0002, "epoch": 5.132328308207705, "step": 7660}, {"loss": 1.2675, "grad_norm": 0.9217309355735779, "learning_rate": 0.0002, "epoch": 5.139028475711893, "step": 7670}, {"loss": 1.2502, "grad_norm": 1.344765543937683, "learning_rate": 0.0002, "epoch": 5.1457286432160805, "step": 7680}, {"loss": 1.2416, "grad_norm": 1.0730723142623901, "learning_rate": 0.0002, "epoch": 5.152428810720268, "step": 7690}, {"loss": 1.1888, "grad_norm": 0.9321247339248657, "learning_rate": 0.0002, "epoch": 5.159128978224456, "step": 7700}, {"loss": 1.1941, "grad_norm": 0.8482614755630493, "learning_rate": 0.0002, "epoch": 5.165829145728643, "step": 7710}, {"loss": 1.2668, "grad_norm": 0.8274452686309814, "learning_rate": 0.0002, "epoch": 5.172529313232831, "step": 7720}, {"loss": 1.1972, "grad_norm": 0.9120376706123352, "learning_rate": 0.0002, "epoch": 5.1792294807370185, "step": 7730}, {"loss": 1.1648, "grad_norm": 1.0062892436981201, "learning_rate": 0.0002, "epoch": 5.185929648241206, "step": 7740}, {"loss": 1.2199, "grad_norm": 0.9521504640579224, "learning_rate": 0.0002, "epoch": 5.192629815745394, "step": 7750}, {"loss": 1.2855, "grad_norm": 0.8800198435783386, "learning_rate": 0.0002, "epoch": 5.199329983249581, "step": 7760}, {"loss": 1.2535, "grad_norm": 0.9749179482460022, "learning_rate": 0.0002, "epoch": 5.206030150753769, "step": 7770}, {"loss": 1.2975, "grad_norm": 0.9441686868667603, "learning_rate": 0.0002, "epoch": 5.2127303182579565, "step": 7780}, {"loss": 1.256, "grad_norm": 0.9114066362380981, "learning_rate": 0.0002, "epoch": 5.219430485762144, "step": 7790}, {"loss": 1.2621, "grad_norm": 0.9851446151733398, "learning_rate": 0.0002, "epoch": 5.226130653266332, "step": 7800}, {"loss": 1.2502, "grad_norm": 0.9526297450065613, "learning_rate": 0.0002, "epoch": 5.232830820770519, "step": 7810}, {"loss": 1.1502, "grad_norm": 1.05986487865448, "learning_rate": 0.0002, "epoch": 5.239530988274707, "step": 7820}, {"loss": 1.2517, "grad_norm": 0.8956538438796997, "learning_rate": 0.0002, "epoch": 5.2462311557788945, "step": 7830}, {"loss": 1.2556, "grad_norm": 0.9568153619766235, "learning_rate": 0.0002, "epoch": 5.252931323283082, "step": 7840}, {"loss": 1.2442, "grad_norm": 1.0035018920898438, "learning_rate": 0.0002, "epoch": 5.259631490787269, "step": 7850}, {"loss": 1.2605, "grad_norm": 0.8554368615150452, "learning_rate": 0.0002, "epoch": 5.266331658291457, "step": 7860}, {"loss": 1.2799, "grad_norm": 0.9677708148956299, "learning_rate": 0.0002, "epoch": 5.273031825795645, "step": 7870}, {"loss": 1.275, "grad_norm": 0.943606436252594, "learning_rate": 0.0002, "epoch": 5.279731993299833, "step": 7880}, {"loss": 1.2335, "grad_norm": 1.0029335021972656, "learning_rate": 0.0002, "epoch": 5.28643216080402, "step": 7890}, {"loss": 1.2494, "grad_norm": 1.0164015293121338, "learning_rate": 0.0002, "epoch": 5.293132328308207, "step": 7900}, {"loss": 1.3117, "grad_norm": 0.8908365368843079, "learning_rate": 0.0002, "epoch": 5.299832495812395, "step": 7910}, {"loss": 1.2832, "grad_norm": 0.9307826161384583, "learning_rate": 0.0002, "epoch": 5.306532663316583, "step": 7920}, {"loss": 1.242, "grad_norm": 1.0730371475219727, "learning_rate": 0.0002, "epoch": 5.313232830820771, "step": 7930}, {"loss": 1.2003, "grad_norm": 0.844739556312561, "learning_rate": 0.0002, "epoch": 5.319932998324958, "step": 7940}, {"loss": 1.2688, "grad_norm": 1.275833010673523, "learning_rate": 0.0002, "epoch": 5.326633165829146, "step": 7950}, {"loss": 1.2957, "grad_norm": 0.9042661190032959, "learning_rate": 0.0002, "epoch": 5.333333333333333, "step": 7960}, {"loss": 1.2912, "grad_norm": 0.9374269247055054, "learning_rate": 0.0002, "epoch": 5.340033500837521, "step": 7970}, {"loss": 1.2721, "grad_norm": 1.033098578453064, "learning_rate": 0.0002, "epoch": 5.346733668341709, "step": 7980}, {"loss": 1.3208, "grad_norm": 1.062775731086731, "learning_rate": 0.0002, "epoch": 5.353433835845896, "step": 7990}, {"loss": 1.3065, "grad_norm": 1.1064317226409912, "learning_rate": 0.0002, "epoch": 5.360134003350084, "step": 8000}, {"loss": 1.2341, "grad_norm": 1.1114039421081543, "learning_rate": 0.0002, "epoch": 5.366834170854271, "step": 8010}, {"loss": 1.2255, "grad_norm": 1.0198014974594116, "learning_rate": 0.0002, "epoch": 5.373534338358459, "step": 8020}, {"loss": 1.2433, "grad_norm": 0.8443173170089722, "learning_rate": 0.0002, "epoch": 5.380234505862647, "step": 8030}, {"loss": 1.206, "grad_norm": 1.000881314277649, "learning_rate": 0.0002, "epoch": 5.386934673366834, "step": 8040}, {"loss": 1.2982, "grad_norm": 0.9874443411827087, "learning_rate": 0.0002, "epoch": 5.393634840871022, "step": 8050}, {"loss": 1.2289, "grad_norm": 0.9895344972610474, "learning_rate": 0.0002, "epoch": 5.400335008375209, "step": 8060}, {"loss": 1.249, "grad_norm": 0.8595236539840698, "learning_rate": 0.0002, "epoch": 5.407035175879397, "step": 8070}, {"loss": 1.2308, "grad_norm": 0.9523849487304688, "learning_rate": 0.0002, "epoch": 5.413735343383585, "step": 8080}, {"loss": 1.2343, "grad_norm": 1.0560476779937744, "learning_rate": 0.0002, "epoch": 5.420435510887772, "step": 8090}, {"loss": 1.2956, "grad_norm": 1.0893689393997192, "learning_rate": 0.0002, "epoch": 5.42713567839196, "step": 8100}, {"loss": 1.2846, "grad_norm": 0.9395513534545898, "learning_rate": 0.0002, "epoch": 5.433835845896147, "step": 8110}, {"loss": 1.3444, "grad_norm": 0.9364215135574341, "learning_rate": 0.0002, "epoch": 5.440536013400335, "step": 8120}, {"loss": 1.2944, "grad_norm": 0.9502208232879639, "learning_rate": 0.0002, "epoch": 5.447236180904523, "step": 8130}, {"loss": 1.2971, "grad_norm": 0.9559208154678345, "learning_rate": 0.0002, "epoch": 5.45393634840871, "step": 8140}, {"loss": 1.2495, "grad_norm": 0.9261730313301086, "learning_rate": 0.0002, "epoch": 5.460636515912898, "step": 8150}, {"loss": 1.2599, "grad_norm": 0.9832326173782349, "learning_rate": 0.0002, "epoch": 5.467336683417085, "step": 8160}, {"loss": 1.2771, "grad_norm": 1.065953016281128, "learning_rate": 0.0002, "epoch": 5.474036850921273, "step": 8170}, {"loss": 1.3617, "grad_norm": 0.9139469861984253, "learning_rate": 0.0002, "epoch": 5.480737018425461, "step": 8180}, {"loss": 1.2197, "grad_norm": 1.2322484254837036, "learning_rate": 0.0002, "epoch": 5.4874371859296485, "step": 8190}, {"loss": 1.2879, "grad_norm": 0.9722974896430969, "learning_rate": 0.0002, "epoch": 5.494137353433836, "step": 8200}, {"loss": 1.2664, "grad_norm": 0.9338926076889038, "learning_rate": 0.0002, "epoch": 5.500837520938023, "step": 8210}, {"loss": 1.2128, "grad_norm": 0.9283728003501892, "learning_rate": 0.0002, "epoch": 5.507537688442211, "step": 8220}, {"loss": 1.2141, "grad_norm": 1.0489585399627686, "learning_rate": 0.0002, "epoch": 5.514237855946399, "step": 8230}, {"loss": 1.3257, "grad_norm": 0.9881814122200012, "learning_rate": 0.0002, "epoch": 5.5209380234505865, "step": 8240}, {"loss": 1.2221, "grad_norm": 0.9274460077285767, "learning_rate": 0.0002, "epoch": 5.527638190954773, "step": 8250}, {"loss": 1.2241, "grad_norm": 0.8650718331336975, "learning_rate": 0.0002, "epoch": 5.534338358458961, "step": 8260}, {"loss": 1.2462, "grad_norm": 1.014069676399231, "learning_rate": 0.0002, "epoch": 5.541038525963149, "step": 8270}, {"loss": 1.3502, "grad_norm": 0.9212974905967712, "learning_rate": 0.0002, "epoch": 5.547738693467337, "step": 8280}, {"loss": 1.2779, "grad_norm": 1.1235398054122925, "learning_rate": 0.0002, "epoch": 5.5544388609715245, "step": 8290}, {"loss": 1.306, "grad_norm": 0.961954653263092, "learning_rate": 0.0002, "epoch": 5.561139028475711, "step": 8300}, {"loss": 1.2946, "grad_norm": 0.9386700391769409, "learning_rate": 0.0002, "epoch": 5.567839195979899, "step": 8310}, {"loss": 1.313, "grad_norm": 1.01912522315979, "learning_rate": 0.0002, "epoch": 5.574539363484087, "step": 8320}, {"loss": 1.3121, "grad_norm": 0.9851216077804565, "learning_rate": 0.0002, "epoch": 5.581239530988275, "step": 8330}, {"loss": 1.3071, "grad_norm": 1.0138001441955566, "learning_rate": 0.0002, "epoch": 5.5879396984924625, "step": 8340}, {"loss": 1.2257, "grad_norm": 0.9262447357177734, "learning_rate": 0.0002, "epoch": 5.594639865996649, "step": 8350}, {"loss": 1.2473, "grad_norm": 1.1322970390319824, "learning_rate": 0.0002, "epoch": 5.601340033500837, "step": 8360}, {"loss": 1.3098, "grad_norm": 1.1429349184036255, "learning_rate": 0.0002, "epoch": 5.608040201005025, "step": 8370}, {"loss": 1.2686, "grad_norm": 0.9130118489265442, "learning_rate": 0.0002, "epoch": 5.614740368509213, "step": 8380}, {"loss": 1.2541, "grad_norm": 0.9651545882225037, "learning_rate": 0.0002, "epoch": 5.6214405360134005, "step": 8390}, {"loss": 1.2799, "grad_norm": 0.9595398306846619, "learning_rate": 0.0002, "epoch": 5.628140703517588, "step": 8400}, {"loss": 1.3429, "grad_norm": 1.0049372911453247, "learning_rate": 0.0002, "epoch": 5.634840871021775, "step": 8410}, {"loss": 1.3224, "grad_norm": 1.082804560661316, "learning_rate": 0.0002, "epoch": 5.641541038525963, "step": 8420}, {"loss": 1.297, "grad_norm": 0.9489204287528992, "learning_rate": 0.0002, "epoch": 5.648241206030151, "step": 8430}, {"loss": 1.3424, "grad_norm": 0.9470235109329224, "learning_rate": 0.0002, "epoch": 5.654941373534339, "step": 8440}, {"loss": 1.3358, "grad_norm": 1.0662927627563477, "learning_rate": 0.0002, "epoch": 5.661641541038526, "step": 8450}, {"loss": 1.2973, "grad_norm": 0.9097877740859985, "learning_rate": 0.0002, "epoch": 5.668341708542713, "step": 8460}, {"loss": 1.3072, "grad_norm": 0.9740368127822876, "learning_rate": 0.0002, "epoch": 5.675041876046901, "step": 8470}, {"loss": 1.286, "grad_norm": 0.9878810048103333, "learning_rate": 0.0002, "epoch": 5.681742043551089, "step": 8480}, {"loss": 1.208, "grad_norm": 1.148260474205017, "learning_rate": 0.0002, "epoch": 5.688442211055277, "step": 8490}, {"loss": 1.2842, "grad_norm": 0.9632558822631836, "learning_rate": 0.0002, "epoch": 5.695142378559464, "step": 8500}, {"loss": 1.2787, "grad_norm": 0.876812756061554, "learning_rate": 0.0002, "epoch": 5.701842546063651, "step": 8510}, {"loss": 1.3186, "grad_norm": 1.0730829238891602, "learning_rate": 0.0002, "epoch": 5.708542713567839, "step": 8520}, {"loss": 1.2856, "grad_norm": 1.2239218950271606, "learning_rate": 0.0002, "epoch": 5.715242881072027, "step": 8530}, {"loss": 1.2717, "grad_norm": 0.9460835456848145, "learning_rate": 0.0002, "epoch": 5.721943048576215, "step": 8540}, {"loss": 1.3509, "grad_norm": 0.9086270928382874, "learning_rate": 0.0002, "epoch": 5.728643216080402, "step": 8550}, {"loss": 1.2971, "grad_norm": 1.0258867740631104, "learning_rate": 0.0002, "epoch": 5.735343383584589, "step": 8560}, {"loss": 1.3581, "grad_norm": 1.0543923377990723, "learning_rate": 0.0002, "epoch": 5.742043551088777, "step": 8570}, {"loss": 1.2988, "grad_norm": 0.9063900113105774, "learning_rate": 0.0002, "epoch": 5.748743718592965, "step": 8580}, {"loss": 1.3535, "grad_norm": 1.1838830709457397, "learning_rate": 0.0002, "epoch": 5.755443886097153, "step": 8590}, {"loss": 1.2655, "grad_norm": 0.9631859064102173, "learning_rate": 0.0002, "epoch": 5.76214405360134, "step": 8600}, {"loss": 1.276, "grad_norm": 0.9702655673027039, "learning_rate": 0.0002, "epoch": 5.768844221105527, "step": 8610}, {"loss": 1.3196, "grad_norm": 1.0591435432434082, "learning_rate": 0.0002, "epoch": 5.775544388609715, "step": 8620}, {"loss": 1.267, "grad_norm": 0.9989570379257202, "learning_rate": 0.0002, "epoch": 5.782244556113903, "step": 8630}, {"loss": 1.3227, "grad_norm": 1.0836435556411743, "learning_rate": 0.0002, "epoch": 5.788944723618091, "step": 8640}, {"loss": 1.3334, "grad_norm": 0.8832896947860718, "learning_rate": 0.0002, "epoch": 5.795644891122278, "step": 8650}, {"loss": 1.3214, "grad_norm": 1.0104607343673706, "learning_rate": 0.0002, "epoch": 5.802345058626465, "step": 8660}, {"loss": 1.2703, "grad_norm": 0.8375084400177002, "learning_rate": 0.0002, "epoch": 5.809045226130653, "step": 8670}, {"loss": 1.3554, "grad_norm": 1.1300716400146484, "learning_rate": 0.0002, "epoch": 5.815745393634841, "step": 8680}, {"loss": 1.3468, "grad_norm": 0.9311910271644592, "learning_rate": 0.0002, "epoch": 5.822445561139029, "step": 8690}, {"loss": 1.2749, "grad_norm": 0.9488391876220703, "learning_rate": 0.0002, "epoch": 5.8291457286432165, "step": 8700}, {"loss": 1.2281, "grad_norm": 0.9747629761695862, "learning_rate": 0.0002, "epoch": 5.835845896147403, "step": 8710}, {"loss": 1.2923, "grad_norm": 1.1029598712921143, "learning_rate": 0.0002, "epoch": 5.842546063651591, "step": 8720}, {"loss": 1.3613, "grad_norm": 1.0396875143051147, "learning_rate": 0.0002, "epoch": 5.849246231155779, "step": 8730}, {"loss": 1.3272, "grad_norm": 0.9259780645370483, "learning_rate": 0.0002, "epoch": 5.855946398659967, "step": 8740}, {"loss": 1.3236, "grad_norm": 1.020033597946167, "learning_rate": 0.0002, "epoch": 5.8626465661641545, "step": 8750}, {"loss": 1.3453, "grad_norm": 0.9191218614578247, "learning_rate": 0.0002, "epoch": 5.869346733668341, "step": 8760}, {"loss": 1.3012, "grad_norm": 1.1093107461929321, "learning_rate": 0.0002, "epoch": 5.876046901172529, "step": 8770}, {"loss": 1.2718, "grad_norm": 1.1626793146133423, "learning_rate": 0.0002, "epoch": 5.882747068676717, "step": 8780}, {"loss": 1.2969, "grad_norm": 0.9542945027351379, "learning_rate": 0.0002, "epoch": 5.889447236180905, "step": 8790}, {"loss": 1.3134, "grad_norm": 0.9086058139801025, "learning_rate": 0.0002, "epoch": 5.8961474036850925, "step": 8800}, {"loss": 1.2731, "grad_norm": 0.9249639511108398, "learning_rate": 0.0002, "epoch": 5.902847571189279, "step": 8810}, {"loss": 1.337, "grad_norm": 0.9414396286010742, "learning_rate": 0.0002, "epoch": 5.909547738693467, "step": 8820}, {"loss": 1.2865, "grad_norm": 0.9086037874221802, "learning_rate": 0.0002, "epoch": 5.916247906197655, "step": 8830}, {"loss": 1.2756, "grad_norm": 0.8685907125473022, "learning_rate": 0.0002, "epoch": 5.922948073701843, "step": 8840}, {"loss": 1.297, "grad_norm": 1.036419153213501, "learning_rate": 0.0002, "epoch": 5.9296482412060305, "step": 8850}, {"loss": 1.3207, "grad_norm": 1.0183674097061157, "learning_rate": 0.0002, "epoch": 5.936348408710217, "step": 8860}, {"loss": 1.3922, "grad_norm": 0.966444194316864, "learning_rate": 0.0002, "epoch": 5.943048576214405, "step": 8870}, {"loss": 1.333, "grad_norm": 1.125693917274475, "learning_rate": 0.0002, "epoch": 5.949748743718593, "step": 8880}, {"loss": 1.3116, "grad_norm": 0.9857436418533325, "learning_rate": 0.0002, "epoch": 5.956448911222781, "step": 8890}, {"loss": 1.2526, "grad_norm": 0.9377069473266602, "learning_rate": 0.0002, "epoch": 5.9631490787269685, "step": 8900}, {"loss": 1.3221, "grad_norm": 0.9493814706802368, "learning_rate": 0.0002, "epoch": 5.969849246231155, "step": 8910}, {"loss": 1.2516, "grad_norm": 0.8806208372116089, "learning_rate": 0.0002, "epoch": 5.976549413735343, "step": 8920}, {"loss": 1.2558, "grad_norm": 0.8727600574493408, "learning_rate": 0.0002, "epoch": 5.983249581239531, "step": 8930}, {"loss": 1.3538, "grad_norm": 0.9799810647964478, "learning_rate": 0.0002, "epoch": 5.989949748743719, "step": 8940}, {"loss": 1.3323, "grad_norm": 0.9866513609886169, "learning_rate": 0.0002, "epoch": 5.9966499162479066, "step": 8950}, {"eval_loss": 2.0282373428344727, "eval_runtime": 38.0375, "eval_samples_per_second": 13.539, "eval_steps_per_second": 1.709, "epoch": 6.0, "step": 8955}, {"loss": 1.1768, "grad_norm": 0.8747885227203369, "learning_rate": 0.0002, "epoch": 6.0033500837520934, "step": 8960}, {"loss": 1.0677, "grad_norm": 1.2512741088867188, "learning_rate": 0.0002, "epoch": 6.010050251256281, "step": 8970}, {"loss": 1.1128, "grad_norm": 1.06855309009552, "learning_rate": 0.0002, "epoch": 6.016750418760469, "step": 8980}, {"loss": 1.1382, "grad_norm": 1.1868711709976196, "learning_rate": 0.0002, "epoch": 6.023450586264657, "step": 8990}, {"loss": 1.1377, "grad_norm": 1.2984495162963867, "learning_rate": 0.0002, "epoch": 6.030150753768845, "step": 9000}, {"loss": 1.0803, "grad_norm": 1.1147589683532715, "learning_rate": 0.0002, "epoch": 6.0368509212730315, "step": 9010}, {"loss": 1.1244, "grad_norm": 1.3128414154052734, "learning_rate": 0.0002, "epoch": 6.043551088777219, "step": 9020}, {"loss": 1.097, "grad_norm": 1.068290114402771, "learning_rate": 0.0002, "epoch": 6.050251256281407, "step": 9030}, {"loss": 1.1764, "grad_norm": 1.1890562772750854, "learning_rate": 0.0002, "epoch": 6.056951423785595, "step": 9040}, {"loss": 1.1239, "grad_norm": 1.2121573686599731, "learning_rate": 0.0002, "epoch": 6.063651591289783, "step": 9050}, {"loss": 1.0683, "grad_norm": 1.0860483646392822, "learning_rate": 0.0002, "epoch": 6.0703517587939695, "step": 9060}, {"loss": 1.1613, "grad_norm": 1.1214599609375, "learning_rate": 0.0002, "epoch": 6.077051926298157, "step": 9070}, {"loss": 1.1001, "grad_norm": 1.147580862045288, "learning_rate": 0.0002, "epoch": 6.083752093802345, "step": 9080}, {"loss": 1.1154, "grad_norm": 1.3233155012130737, "learning_rate": 0.0002, "epoch": 6.090452261306533, "step": 9090}, {"loss": 1.1017, "grad_norm": 1.1869080066680908, "learning_rate": 0.0002, "epoch": 6.097152428810721, "step": 9100}, {"loss": 1.1532, "grad_norm": 1.1695014238357544, "learning_rate": 0.0002, "epoch": 6.1038525963149075, "step": 9110}, {"loss": 1.1366, "grad_norm": 1.1982251405715942, "learning_rate": 0.0002, "epoch": 6.110552763819095, "step": 9120}, {"loss": 1.0805, "grad_norm": 1.1426950693130493, "learning_rate": 0.0002, "epoch": 6.117252931323283, "step": 9130}, {"loss": 1.0801, "grad_norm": 1.2257394790649414, "learning_rate": 0.0002, "epoch": 6.123953098827471, "step": 9140}, {"loss": 1.1209, "grad_norm": 1.2932263612747192, "learning_rate": 0.0002, "epoch": 6.130653266331659, "step": 9150}, {"loss": 1.0934, "grad_norm": 1.2617030143737793, "learning_rate": 0.0002, "epoch": 6.1373534338358455, "step": 9160}, {"loss": 1.0551, "grad_norm": 1.1201422214508057, "learning_rate": 0.0002, "epoch": 6.144053601340033, "step": 9170}, {"loss": 1.1059, "grad_norm": 0.9625319838523865, "learning_rate": 0.0002, "epoch": 6.150753768844221, "step": 9180}, {"loss": 1.1397, "grad_norm": 1.0290048122406006, "learning_rate": 0.0002, "epoch": 6.157453936348409, "step": 9190}, {"loss": 1.1257, "grad_norm": 1.1137803792953491, "learning_rate": 0.0002, "epoch": 6.164154103852597, "step": 9200}, {"loss": 1.1211, "grad_norm": 1.3674522638320923, "learning_rate": 0.0002, "epoch": 6.1708542713567835, "step": 9210}, {"loss": 1.0947, "grad_norm": 1.182207703590393, "learning_rate": 0.0002, "epoch": 6.177554438860971, "step": 9220}, {"loss": 1.0838, "grad_norm": 1.0496711730957031, "learning_rate": 0.0002, "epoch": 6.184254606365159, "step": 9230}, {"loss": 1.0666, "grad_norm": 1.1899489164352417, "learning_rate": 0.0002, "epoch": 6.190954773869347, "step": 9240}, {"loss": 1.1633, "grad_norm": 1.2666147947311401, "learning_rate": 0.0002, "epoch": 6.197654941373535, "step": 9250}, {"loss": 1.1532, "grad_norm": 1.2013030052185059, "learning_rate": 0.0002, "epoch": 6.204355108877722, "step": 9260}, {"loss": 1.151, "grad_norm": 1.3049768209457397, "learning_rate": 0.0002, "epoch": 6.211055276381909, "step": 9270}, {"loss": 1.0741, "grad_norm": 1.1733006238937378, "learning_rate": 0.0002, "epoch": 6.217755443886097, "step": 9280}, {"loss": 1.0933, "grad_norm": 1.2742516994476318, "learning_rate": 0.0002, "epoch": 6.224455611390285, "step": 9290}, {"loss": 1.1028, "grad_norm": 1.110198974609375, "learning_rate": 0.0002, "epoch": 6.231155778894473, "step": 9300}, {"loss": 1.1619, "grad_norm": 1.159963607788086, "learning_rate": 0.0002, "epoch": 6.23785594639866, "step": 9310}, {"loss": 1.0716, "grad_norm": 1.302216649055481, "learning_rate": 0.0002, "epoch": 6.244556113902847, "step": 9320}, {"loss": 1.0694, "grad_norm": 1.2134063243865967, "learning_rate": 0.0002, "epoch": 6.251256281407035, "step": 9330}, {"loss": 1.2151, "grad_norm": 1.062682867050171, "learning_rate": 0.0002, "epoch": 6.257956448911223, "step": 9340}, {"loss": 1.148, "grad_norm": 1.1568971872329712, "learning_rate": 0.0002, "epoch": 6.264656616415411, "step": 9350}, {"loss": 1.094, "grad_norm": 0.9914957880973816, "learning_rate": 0.0002, "epoch": 6.271356783919598, "step": 9360}, {"loss": 1.125, "grad_norm": 1.017250895500183, "learning_rate": 0.0002, "epoch": 6.278056951423785, "step": 9370}, {"loss": 1.2177, "grad_norm": 1.1862834692001343, "learning_rate": 0.0002, "epoch": 6.284757118927973, "step": 9380}, {"loss": 0.9994, "grad_norm": 1.2834911346435547, "learning_rate": 0.0002, "epoch": 6.291457286432161, "step": 9390}, {"loss": 1.0922, "grad_norm": 1.3306856155395508, "learning_rate": 0.0002, "epoch": 6.298157453936349, "step": 9400}, {"loss": 1.1136, "grad_norm": 1.12908136844635, "learning_rate": 0.0002, "epoch": 6.304857621440536, "step": 9410}, {"loss": 1.1406, "grad_norm": 1.2157351970672607, "learning_rate": 0.0002, "epoch": 6.311557788944723, "step": 9420}, {"loss": 1.1388, "grad_norm": 1.121882677078247, "learning_rate": 0.0002, "epoch": 6.318257956448911, "step": 9430}, {"loss": 1.1648, "grad_norm": 1.3144481182098389, "learning_rate": 0.0002, "epoch": 6.324958123953099, "step": 9440}, {"loss": 1.1228, "grad_norm": 1.1946896314620972, "learning_rate": 0.0002, "epoch": 6.331658291457287, "step": 9450}, {"loss": 1.1613, "grad_norm": 1.1289668083190918, "learning_rate": 0.0002, "epoch": 6.338358458961474, "step": 9460}, {"loss": 1.1059, "grad_norm": 1.1065658330917358, "learning_rate": 0.0002, "epoch": 6.345058626465661, "step": 9470}, {"loss": 1.1431, "grad_norm": 1.0881422758102417, "learning_rate": 0.0002, "epoch": 6.351758793969849, "step": 9480}, {"loss": 1.223, "grad_norm": 1.242676854133606, "learning_rate": 0.0002, "epoch": 6.358458961474037, "step": 9490}, {"loss": 1.1379, "grad_norm": 0.9650855660438538, "learning_rate": 0.0002, "epoch": 6.365159128978225, "step": 9500}, {"loss": 1.0763, "grad_norm": 1.2845722436904907, "learning_rate": 0.0002, "epoch": 6.371859296482412, "step": 9510}, {"loss": 1.1351, "grad_norm": 1.0327043533325195, "learning_rate": 0.0002, "epoch": 6.3785594639865995, "step": 9520}, {"loss": 1.114, "grad_norm": 1.0780898332595825, "learning_rate": 0.0002, "epoch": 6.385259631490787, "step": 9530}, {"loss": 1.1579, "grad_norm": 1.4934027194976807, "learning_rate": 0.0002, "epoch": 6.391959798994975, "step": 9540}, {"loss": 1.1546, "grad_norm": 0.9882908463478088, "learning_rate": 0.0002, "epoch": 6.398659966499163, "step": 9550}, {"loss": 1.1145, "grad_norm": 1.3250664472579956, "learning_rate": 0.0002, "epoch": 6.40536013400335, "step": 9560}, {"loss": 1.2333, "grad_norm": 1.1888482570648193, "learning_rate": 0.0002, "epoch": 6.4120603015075375, "step": 9570}, {"loss": 1.0892, "grad_norm": 1.136496901512146, "learning_rate": 0.0002, "epoch": 6.418760469011725, "step": 9580}, {"loss": 1.1674, "grad_norm": 1.161360502243042, "learning_rate": 0.0002, "epoch": 6.425460636515913, "step": 9590}, {"loss": 1.1293, "grad_norm": 1.2034236192703247, "learning_rate": 0.0002, "epoch": 6.432160804020101, "step": 9600}, {"loss": 1.1059, "grad_norm": 1.0268361568450928, "learning_rate": 0.0002, "epoch": 6.438860971524288, "step": 9610}, {"loss": 1.1732, "grad_norm": 1.2132930755615234, "learning_rate": 0.0002, "epoch": 6.4455611390284755, "step": 9620}, {"loss": 1.1329, "grad_norm": 1.0773013830184937, "learning_rate": 0.0002, "epoch": 6.452261306532663, "step": 9630}, {"loss": 1.0822, "grad_norm": 1.3848375082015991, "learning_rate": 0.0002, "epoch": 6.458961474036851, "step": 9640}, {"loss": 1.1778, "grad_norm": 1.110495924949646, "learning_rate": 0.0002, "epoch": 6.465661641541039, "step": 9650}, {"loss": 1.2022, "grad_norm": 1.118093729019165, "learning_rate": 0.0002, "epoch": 6.472361809045226, "step": 9660}, {"loss": 1.1222, "grad_norm": 1.2611900568008423, "learning_rate": 0.0002, "epoch": 6.4790619765494135, "step": 9670}, {"loss": 1.2138, "grad_norm": 0.971754252910614, "learning_rate": 0.0002, "epoch": 6.485762144053601, "step": 9680}, {"loss": 1.1641, "grad_norm": 1.2615419626235962, "learning_rate": 0.0002, "epoch": 6.492462311557789, "step": 9690}, {"loss": 1.1412, "grad_norm": 1.1370900869369507, "learning_rate": 0.0002, "epoch": 6.499162479061977, "step": 9700}, {"loss": 1.186, "grad_norm": 1.1815906763076782, "learning_rate": 0.0002, "epoch": 6.505862646566165, "step": 9710}, {"loss": 1.167, "grad_norm": 1.3424339294433594, "learning_rate": 0.0002, "epoch": 6.5125628140703515, "step": 9720}, {"loss": 1.1602, "grad_norm": 1.2858397960662842, "learning_rate": 0.0002, "epoch": 6.519262981574539, "step": 9730}, {"loss": 1.178, "grad_norm": 0.9578179121017456, "learning_rate": 0.0002, "epoch": 6.525963149078727, "step": 9740}, {"loss": 1.1805, "grad_norm": 1.3105167150497437, "learning_rate": 0.0002, "epoch": 6.532663316582915, "step": 9750}, {"loss": 1.1899, "grad_norm": 1.0586575269699097, "learning_rate": 0.0002, "epoch": 6.539363484087103, "step": 9760}, {"loss": 1.095, "grad_norm": 1.2122068405151367, "learning_rate": 0.0002, "epoch": 6.54606365159129, "step": 9770}, {"loss": 1.1471, "grad_norm": 1.3088626861572266, "learning_rate": 0.0002, "epoch": 6.552763819095477, "step": 9780}, {"loss": 1.1067, "grad_norm": 1.194122076034546, "learning_rate": 0.0002, "epoch": 6.559463986599665, "step": 9790}, {"loss": 1.0967, "grad_norm": 1.1508387327194214, "learning_rate": 0.0002, "epoch": 6.566164154103853, "step": 9800}, {"loss": 1.1694, "grad_norm": 1.109228253364563, "learning_rate": 0.0002, "epoch": 6.572864321608041, "step": 9810}, {"loss": 1.1378, "grad_norm": 1.1607427597045898, "learning_rate": 0.0002, "epoch": 6.579564489112228, "step": 9820}, {"loss": 1.1585, "grad_norm": 1.174089789390564, "learning_rate": 0.0002, "epoch": 6.586264656616415, "step": 9830}, {"loss": 1.1385, "grad_norm": 1.1739521026611328, "learning_rate": 0.0002, "epoch": 6.592964824120603, "step": 9840}, {"loss": 1.155, "grad_norm": 1.098528504371643, "learning_rate": 0.0002, "epoch": 6.599664991624791, "step": 9850}, {"loss": 1.1359, "grad_norm": 1.0397740602493286, "learning_rate": 0.0002, "epoch": 6.606365159128979, "step": 9860}, {"loss": 1.1433, "grad_norm": 1.1087969541549683, "learning_rate": 0.0002, "epoch": 6.613065326633166, "step": 9870}, {"loss": 1.2356, "grad_norm": 1.2070481777191162, "learning_rate": 0.0002, "epoch": 6.619765494137353, "step": 9880}, {"loss": 1.1161, "grad_norm": 1.1115655899047852, "learning_rate": 0.0002, "epoch": 6.626465661641541, "step": 9890}, {"loss": 1.2163, "grad_norm": 1.2486097812652588, "learning_rate": 0.0002, "epoch": 6.633165829145729, "step": 9900}, {"loss": 1.0984, "grad_norm": 1.230380654335022, "learning_rate": 0.0002, "epoch": 6.639865996649917, "step": 9910}, {"loss": 1.1862, "grad_norm": 1.1479365825653076, "learning_rate": 0.0002, "epoch": 6.646566164154104, "step": 9920}, {"loss": 1.1139, "grad_norm": 1.0790960788726807, "learning_rate": 0.0002, "epoch": 6.653266331658291, "step": 9930}, {"loss": 1.2001, "grad_norm": 1.1157397031784058, "learning_rate": 0.0002, "epoch": 6.659966499162479, "step": 9940}, {"loss": 1.1085, "grad_norm": 1.3104028701782227, "learning_rate": 0.0002, "epoch": 6.666666666666667, "step": 9950}, {"loss": 1.2209, "grad_norm": 1.1727646589279175, "learning_rate": 0.0002, "epoch": 6.673366834170855, "step": 9960}, {"loss": 1.1671, "grad_norm": 1.2104284763336182, "learning_rate": 0.0002, "epoch": 6.680067001675042, "step": 9970}, {"loss": 1.1952, "grad_norm": 1.2023727893829346, "learning_rate": 0.0002, "epoch": 6.686767169179229, "step": 9980}, {"loss": 1.1385, "grad_norm": 1.0088225603103638, "learning_rate": 0.0002, "epoch": 6.693467336683417, "step": 9990}, {"loss": 1.1314, "grad_norm": 1.298015832901001, "learning_rate": 0.0002, "epoch": 6.700167504187605, "step": 10000}, {"loss": 1.1983, "grad_norm": 1.1315910816192627, "learning_rate": 0.0002, "epoch": 6.706867671691793, "step": 10010}, {"loss": 1.1679, "grad_norm": 1.1283273696899414, "learning_rate": 0.0002, "epoch": 6.71356783919598, "step": 10020}, {"loss": 1.1917, "grad_norm": 1.2564418315887451, "learning_rate": 0.0002, "epoch": 6.720268006700167, "step": 10030}, {"loss": 1.1788, "grad_norm": 1.0451353788375854, "learning_rate": 0.0002, "epoch": 6.726968174204355, "step": 10040}, {"loss": 1.1905, "grad_norm": 1.054793357849121, "learning_rate": 0.0002, "epoch": 6.733668341708543, "step": 10050}, {"loss": 1.1814, "grad_norm": 1.2741243839263916, "learning_rate": 0.0002, "epoch": 6.740368509212731, "step": 10060}, {"loss": 1.2015, "grad_norm": 1.1342514753341675, "learning_rate": 0.0002, "epoch": 6.747068676716918, "step": 10070}, {"loss": 1.2587, "grad_norm": 1.0081498622894287, "learning_rate": 0.0002, "epoch": 6.7537688442211055, "step": 10080}, {"loss": 1.1226, "grad_norm": 1.2164603471755981, "learning_rate": 0.0002, "epoch": 6.760469011725293, "step": 10090}, {"loss": 1.1353, "grad_norm": 1.2062463760375977, "learning_rate": 0.0002, "epoch": 6.767169179229481, "step": 10100}, {"loss": 1.2143, "grad_norm": 1.2255526781082153, "learning_rate": 0.0002, "epoch": 6.773869346733669, "step": 10110}, {"loss": 1.1544, "grad_norm": 1.08175790309906, "learning_rate": 0.0002, "epoch": 6.780569514237856, "step": 10120}, {"loss": 1.1983, "grad_norm": 1.5781128406524658, "learning_rate": 0.0002, "epoch": 6.7872696817420435, "step": 10130}, {"loss": 1.0994, "grad_norm": 1.0622451305389404, "learning_rate": 0.0002, "epoch": 6.793969849246231, "step": 10140}, {"loss": 1.2084, "grad_norm": 1.1591497659683228, "learning_rate": 0.0002, "epoch": 6.800670016750419, "step": 10150}, {"loss": 1.2203, "grad_norm": 1.0398483276367188, "learning_rate": 0.0002, "epoch": 6.807370184254607, "step": 10160}, {"loss": 1.2249, "grad_norm": 1.229132056236267, "learning_rate": 0.0002, "epoch": 6.814070351758794, "step": 10170}, {"loss": 1.1789, "grad_norm": 1.0918090343475342, "learning_rate": 0.0002, "epoch": 6.8207705192629815, "step": 10180}, {"loss": 1.1639, "grad_norm": 1.1543749570846558, "learning_rate": 0.0002, "epoch": 6.827470686767169, "step": 10190}, {"loss": 1.1353, "grad_norm": 1.1831817626953125, "learning_rate": 0.0002, "epoch": 6.834170854271357, "step": 10200}, {"loss": 1.2565, "grad_norm": 1.305327296257019, "learning_rate": 0.0002, "epoch": 6.840871021775545, "step": 10210}, {"loss": 1.2037, "grad_norm": 1.136720061302185, "learning_rate": 0.0002, "epoch": 6.847571189279732, "step": 10220}, {"loss": 1.2256, "grad_norm": 1.2282346487045288, "learning_rate": 0.0002, "epoch": 6.8542713567839195, "step": 10230}, {"loss": 1.1281, "grad_norm": 1.2457010746002197, "learning_rate": 0.0002, "epoch": 6.860971524288107, "step": 10240}, {"loss": 1.2762, "grad_norm": 1.2808631658554077, "learning_rate": 0.0002, "epoch": 6.867671691792295, "step": 10250}, {"loss": 1.2213, "grad_norm": 1.089066743850708, "learning_rate": 0.0002, "epoch": 6.874371859296483, "step": 10260}, {"loss": 1.2627, "grad_norm": 0.9543178081512451, "learning_rate": 0.0002, "epoch": 6.88107202680067, "step": 10270}, {"loss": 1.1617, "grad_norm": 1.1149744987487793, "learning_rate": 0.0002, "epoch": 6.8877721943048575, "step": 10280}, {"loss": 1.1134, "grad_norm": 1.0185538530349731, "learning_rate": 0.0002, "epoch": 6.894472361809045, "step": 10290}, {"loss": 1.217, "grad_norm": 0.9954617619514465, "learning_rate": 0.0002, "epoch": 6.901172529313233, "step": 10300}, {"loss": 1.1524, "grad_norm": 1.2581418752670288, "learning_rate": 0.0002, "epoch": 6.907872696817421, "step": 10310}, {"loss": 1.1942, "grad_norm": 1.2430983781814575, "learning_rate": 0.0002, "epoch": 6.914572864321608, "step": 10320}, {"loss": 1.1254, "grad_norm": 1.4937270879745483, "learning_rate": 0.0002, "epoch": 6.921273031825796, "step": 10330}, {"loss": 1.1519, "grad_norm": 1.1257144212722778, "learning_rate": 0.0002, "epoch": 6.927973199329983, "step": 10340}, {"loss": 1.2622, "grad_norm": 1.2068904638290405, "learning_rate": 0.0002, "epoch": 6.934673366834171, "step": 10350}, {"loss": 1.1349, "grad_norm": 1.0290757417678833, "learning_rate": 0.0002, "epoch": 6.941373534338359, "step": 10360}, {"loss": 1.1752, "grad_norm": 1.0070724487304688, "learning_rate": 0.0002, "epoch": 6.948073701842546, "step": 10370}, {"loss": 1.1838, "grad_norm": 0.9936357140541077, "learning_rate": 0.0002, "epoch": 6.954773869346734, "step": 10380}, {"loss": 1.2305, "grad_norm": 1.1063416004180908, "learning_rate": 0.0002, "epoch": 6.961474036850921, "step": 10390}, {"loss": 1.154, "grad_norm": 1.5199986696243286, "learning_rate": 0.0002, "epoch": 6.968174204355109, "step": 10400}, {"loss": 1.1143, "grad_norm": 1.160731554031372, "learning_rate": 0.0002, "epoch": 6.974874371859297, "step": 10410}, {"loss": 1.2132, "grad_norm": 1.084697961807251, "learning_rate": 0.0002, "epoch": 6.981574539363484, "step": 10420}, {"loss": 1.1995, "grad_norm": 1.1257576942443848, "learning_rate": 0.0002, "epoch": 6.988274706867672, "step": 10430}, {"loss": 1.1197, "grad_norm": 1.310616135597229, "learning_rate": 0.0002, "epoch": 6.994974874371859, "step": 10440}, {"eval_loss": 2.1203012466430664, "eval_runtime": 37.936, "eval_samples_per_second": 13.576, "eval_steps_per_second": 1.713, "epoch": 6.99966499162479, "step": 10447}, {"loss": 1.1118, "grad_norm": 1.0176491737365723, "learning_rate": 0.0002, "epoch": 7.001675041876047, "step": 10450}, {"loss": 0.9932, "grad_norm": 1.602665662765503, "learning_rate": 0.0002, "epoch": 7.008375209380235, "step": 10460}, {"loss": 0.9818, "grad_norm": 1.2909572124481201, "learning_rate": 0.0002, "epoch": 7.015075376884422, "step": 10470}, {"loss": 0.9398, "grad_norm": 1.2601855993270874, "learning_rate": 0.0002, "epoch": 7.02177554438861, "step": 10480}, {"loss": 0.9241, "grad_norm": 1.3437587022781372, "learning_rate": 0.0002, "epoch": 7.028475711892797, "step": 10490}, {"loss": 1.0031, "grad_norm": 1.2220089435577393, "learning_rate": 0.0002, "epoch": 7.035175879396985, "step": 10500}, {"loss": 0.9428, "grad_norm": 1.3392685651779175, "learning_rate": 0.0002, "epoch": 7.041876046901173, "step": 10510}, {"loss": 0.9566, "grad_norm": 1.3902767896652222, "learning_rate": 0.0002, "epoch": 7.04857621440536, "step": 10520}, {"loss": 0.9993, "grad_norm": 1.4098035097122192, "learning_rate": 0.0002, "epoch": 7.055276381909548, "step": 10530}, {"loss": 0.9683, "grad_norm": 1.38866126537323, "learning_rate": 0.0002, "epoch": 7.061976549413735, "step": 10540}, {"loss": 0.9961, "grad_norm": 1.3638999462127686, "learning_rate": 0.0002, "epoch": 7.068676716917923, "step": 10550}, {"loss": 0.9698, "grad_norm": 1.3181349039077759, "learning_rate": 0.0002, "epoch": 7.075376884422111, "step": 10560}, {"loss": 0.9963, "grad_norm": 1.2528657913208008, "learning_rate": 0.0002, "epoch": 7.082077051926298, "step": 10570}, {"loss": 0.9624, "grad_norm": 1.4326589107513428, "learning_rate": 0.0002, "epoch": 7.088777219430486, "step": 10580}, {"loss": 1.0432, "grad_norm": 1.4312337636947632, "learning_rate": 0.0002, "epoch": 7.0954773869346734, "step": 10590}, {"loss": 0.9704, "grad_norm": 1.7286990880966187, "learning_rate": 0.0002, "epoch": 7.102177554438861, "step": 10600}, {"loss": 0.9954, "grad_norm": 1.1248762607574463, "learning_rate": 0.0002, "epoch": 7.108877721943049, "step": 10610}, {"loss": 0.9736, "grad_norm": 1.278731346130371, "learning_rate": 0.0002, "epoch": 7.115577889447236, "step": 10620}, {"loss": 0.9885, "grad_norm": 1.53670072555542, "learning_rate": 0.0002, "epoch": 7.122278056951424, "step": 10630}, {"loss": 0.9573, "grad_norm": 1.268069863319397, "learning_rate": 0.0002, "epoch": 7.1289782244556115, "step": 10640}, {"loss": 0.9716, "grad_norm": 1.5072290897369385, "learning_rate": 0.0002, "epoch": 7.135678391959799, "step": 10650}, {"loss": 0.9871, "grad_norm": 1.5552845001220703, "learning_rate": 0.0002, "epoch": 7.142378559463987, "step": 10660}, {"loss": 1.0029, "grad_norm": 1.2643769979476929, "learning_rate": 0.0002, "epoch": 7.149078726968174, "step": 10670}, {"loss": 0.9848, "grad_norm": 1.347589373588562, "learning_rate": 0.0002, "epoch": 7.155778894472362, "step": 10680}, {"loss": 1.0193, "grad_norm": 1.364700436592102, "learning_rate": 0.0002, "epoch": 7.1624790619765495, "step": 10690}, {"loss": 1.0057, "grad_norm": 1.4375768899917603, "learning_rate": 0.0002, "epoch": 7.169179229480737, "step": 10700}, {"loss": 1.0108, "grad_norm": 1.426047444343567, "learning_rate": 0.0002, "epoch": 7.175879396984925, "step": 10710}, {"loss": 0.9985, "grad_norm": 1.5208740234375, "learning_rate": 0.0002, "epoch": 7.182579564489112, "step": 10720}, {"loss": 1.0535, "grad_norm": 1.4713337421417236, "learning_rate": 0.0002, "epoch": 7.1892797319933, "step": 10730}, {"loss": 0.9481, "grad_norm": 1.3042446374893188, "learning_rate": 0.0002, "epoch": 7.1959798994974875, "step": 10740}, {"loss": 0.9813, "grad_norm": 1.2290682792663574, "learning_rate": 0.0002, "epoch": 7.202680067001675, "step": 10750}, {"loss": 1.0437, "grad_norm": 1.1152390241622925, "learning_rate": 0.0002, "epoch": 7.209380234505863, "step": 10760}, {"loss": 0.9557, "grad_norm": 1.3632242679595947, "learning_rate": 0.0002, "epoch": 7.21608040201005, "step": 10770}, {"loss": 0.9915, "grad_norm": 1.2406541109085083, "learning_rate": 0.0002, "epoch": 7.222780569514238, "step": 10780}, {"loss": 1.0138, "grad_norm": 1.1965205669403076, "learning_rate": 0.0002, "epoch": 7.2294807370184255, "step": 10790}, {"loss": 1.0186, "grad_norm": 1.2895352840423584, "learning_rate": 0.0002, "epoch": 7.236180904522613, "step": 10800}, {"loss": 0.9616, "grad_norm": 1.388366937637329, "learning_rate": 0.0002, "epoch": 7.242881072026801, "step": 10810}, {"loss": 0.992, "grad_norm": 1.1411796808242798, "learning_rate": 0.0002, "epoch": 7.249581239530988, "step": 10820}, {"loss": 0.942, "grad_norm": 1.2220646142959595, "learning_rate": 0.0002, "epoch": 7.256281407035176, "step": 10830}, {"loss": 1.0136, "grad_norm": 1.495492696762085, "learning_rate": 0.0002, "epoch": 7.2629815745393635, "step": 10840}, {"loss": 1.0475, "grad_norm": 1.1395213603973389, "learning_rate": 0.0002, "epoch": 7.269681742043551, "step": 10850}, {"loss": 1.0134, "grad_norm": 1.3826487064361572, "learning_rate": 0.0002, "epoch": 7.276381909547739, "step": 10860}, {"loss": 0.944, "grad_norm": 1.4356474876403809, "learning_rate": 0.0002, "epoch": 7.283082077051926, "step": 10870}, {"loss": 0.971, "grad_norm": 1.3617557287216187, "learning_rate": 0.0002, "epoch": 7.289782244556114, "step": 10880}, {"loss": 0.9772, "grad_norm": 1.585394024848938, "learning_rate": 0.0002, "epoch": 7.296482412060302, "step": 10890}, {"loss": 1.0204, "grad_norm": 1.1442821025848389, "learning_rate": 0.0002, "epoch": 7.303182579564489, "step": 10900}, {"loss": 0.9608, "grad_norm": 1.4712985754013062, "learning_rate": 0.0002, "epoch": 7.309882747068677, "step": 10910}, {"loss": 1.0027, "grad_norm": 1.2864325046539307, "learning_rate": 0.0002, "epoch": 7.316582914572864, "step": 10920}, {"loss": 1.0341, "grad_norm": 1.308010458946228, "learning_rate": 0.0002, "epoch": 7.323283082077052, "step": 10930}, {"loss": 1.0096, "grad_norm": 1.4371414184570312, "learning_rate": 0.0002, "epoch": 7.32998324958124, "step": 10940}, {"loss": 0.9999, "grad_norm": 1.5968806743621826, "learning_rate": 0.0002, "epoch": 7.336683417085427, "step": 10950}, {"loss": 0.9611, "grad_norm": 1.3592339754104614, "learning_rate": 0.0002, "epoch": 7.343383584589615, "step": 10960}, {"loss": 1.0505, "grad_norm": 1.225520133972168, "learning_rate": 0.0002, "epoch": 7.350083752093802, "step": 10970}, {"loss": 0.9702, "grad_norm": 1.3138031959533691, "learning_rate": 0.0002, "epoch": 7.35678391959799, "step": 10980}, {"loss": 1.0062, "grad_norm": 1.2601540088653564, "learning_rate": 0.0002, "epoch": 7.363484087102178, "step": 10990}, {"loss": 1.0508, "grad_norm": 1.3437113761901855, "learning_rate": 0.0002, "epoch": 7.370184254606365, "step": 11000}, {"loss": 0.9649, "grad_norm": 1.0681836605072021, "learning_rate": 0.0002, "epoch": 7.376884422110553, "step": 11010}, {"loss": 1.0025, "grad_norm": 1.415852665901184, "learning_rate": 0.0002, "epoch": 7.38358458961474, "step": 11020}, {"loss": 1.0532, "grad_norm": 1.3147039413452148, "learning_rate": 0.0002, "epoch": 7.390284757118928, "step": 11030}, {"loss": 1.1188, "grad_norm": 1.4778614044189453, "learning_rate": 0.0002, "epoch": 7.396984924623116, "step": 11040}, {"loss": 0.9917, "grad_norm": 1.3137797117233276, "learning_rate": 0.0002, "epoch": 7.403685092127303, "step": 11050}, {"loss": 1.0115, "grad_norm": 1.1917701959609985, "learning_rate": 0.0002, "epoch": 7.410385259631491, "step": 11060}, {"loss": 1.0668, "grad_norm": 1.3999699354171753, "learning_rate": 0.0002, "epoch": 7.417085427135678, "step": 11070}, {"loss": 1.0311, "grad_norm": 1.3817322254180908, "learning_rate": 0.0002, "epoch": 7.423785594639866, "step": 11080}, {"loss": 1.0086, "grad_norm": 1.2086812257766724, "learning_rate": 0.0002, "epoch": 7.430485762144054, "step": 11090}, {"loss": 1.0002, "grad_norm": 1.1938024759292603, "learning_rate": 0.0002, "epoch": 7.437185929648241, "step": 11100}, {"loss": 1.0463, "grad_norm": 1.1543669700622559, "learning_rate": 0.0002, "epoch": 7.443886097152429, "step": 11110}, {"loss": 1.0814, "grad_norm": 1.158841848373413, "learning_rate": 0.0002, "epoch": 7.450586264656616, "step": 11120}, {"loss": 1.0022, "grad_norm": 1.2777763605117798, "learning_rate": 0.0002, "epoch": 7.457286432160804, "step": 11130}, {"loss": 1.0162, "grad_norm": 1.3375903367996216, "learning_rate": 0.0002, "epoch": 7.463986599664992, "step": 11140}, {"loss": 0.9995, "grad_norm": 1.5573794841766357, "learning_rate": 0.0002, "epoch": 7.4706867671691795, "step": 11150}, {"loss": 1.0698, "grad_norm": 1.3869640827178955, "learning_rate": 0.0002, "epoch": 7.477386934673367, "step": 11160}, {"loss": 1.0217, "grad_norm": 1.310341238975525, "learning_rate": 0.0002, "epoch": 7.484087102177554, "step": 11170}, {"loss": 1.0365, "grad_norm": 1.4249892234802246, "learning_rate": 0.0002, "epoch": 7.490787269681742, "step": 11180}, {"loss": 1.0346, "grad_norm": 1.4025905132293701, "learning_rate": 0.0002, "epoch": 7.49748743718593, "step": 11190}, {"loss": 1.0328, "grad_norm": 1.4361881017684937, "learning_rate": 0.0002, "epoch": 7.5041876046901175, "step": 11200}, {"loss": 1.0316, "grad_norm": 1.1791380643844604, "learning_rate": 0.0002, "epoch": 7.510887772194305, "step": 11210}, {"loss": 1.0621, "grad_norm": 1.415075421333313, "learning_rate": 0.0002, "epoch": 7.517587939698492, "step": 11220}, {"loss": 1.0265, "grad_norm": 1.3633697032928467, "learning_rate": 0.0002, "epoch": 7.52428810720268, "step": 11230}, {"loss": 0.983, "grad_norm": 1.2803648710250854, "learning_rate": 0.0002, "epoch": 7.530988274706868, "step": 11240}, {"loss": 1.0121, "grad_norm": 1.4032878875732422, "learning_rate": 0.0002, "epoch": 7.5376884422110555, "step": 11250}, {"loss": 1.032, "grad_norm": 1.4507378339767456, "learning_rate": 0.0002, "epoch": 7.544388609715243, "step": 11260}, {"loss": 1.0221, "grad_norm": 1.227613925933838, "learning_rate": 0.0002, "epoch": 7.55108877721943, "step": 11270}, {"loss": 1.0288, "grad_norm": 1.2620965242385864, "learning_rate": 0.0002, "epoch": 7.557788944723618, "step": 11280}, {"loss": 1.1263, "grad_norm": 1.3917821645736694, "learning_rate": 0.0002, "epoch": 7.564489112227806, "step": 11290}, {"loss": 1.0628, "grad_norm": 1.5720019340515137, "learning_rate": 0.0002, "epoch": 7.5711892797319935, "step": 11300}, {"loss": 1.0537, "grad_norm": 1.376694917678833, "learning_rate": 0.0002, "epoch": 7.577889447236181, "step": 11310}, {"loss": 1.012, "grad_norm": 1.4403680562973022, "learning_rate": 0.0002, "epoch": 7.584589614740368, "step": 11320}, {"loss": 1.0318, "grad_norm": 1.6306934356689453, "learning_rate": 0.0002, "epoch": 7.591289782244556, "step": 11330}, {"loss": 1.0297, "grad_norm": 1.2361583709716797, "learning_rate": 0.0002, "epoch": 7.597989949748744, "step": 11340}, {"loss": 1.022, "grad_norm": 1.2658058404922485, "learning_rate": 0.0002, "epoch": 7.6046901172529315, "step": 11350}, {"loss": 1.0357, "grad_norm": 1.5335173606872559, "learning_rate": 0.0002, "epoch": 7.611390284757119, "step": 11360}, {"loss": 1.0689, "grad_norm": 1.3432948589324951, "learning_rate": 0.0002, "epoch": 7.618090452261306, "step": 11370}, {"loss": 1.0613, "grad_norm": 1.374617338180542, "learning_rate": 0.0002, "epoch": 7.624790619765494, "step": 11380}, {"loss": 1.0722, "grad_norm": 1.3790302276611328, "learning_rate": 0.0002, "epoch": 7.631490787269682, "step": 11390}, {"loss": 1.0078, "grad_norm": 1.4256713390350342, "learning_rate": 0.0002, "epoch": 7.63819095477387, "step": 11400}, {"loss": 1.0314, "grad_norm": 1.271228313446045, "learning_rate": 0.0002, "epoch": 7.644891122278057, "step": 11410}, {"loss": 1.0678, "grad_norm": 1.432146668434143, "learning_rate": 0.0002, "epoch": 7.651591289782244, "step": 11420}, {"loss": 1.0496, "grad_norm": 1.2698006629943848, "learning_rate": 0.0002, "epoch": 7.658291457286432, "step": 11430}, {"loss": 1.0678, "grad_norm": 1.439496636390686, "learning_rate": 0.0002, "epoch": 7.66499162479062, "step": 11440}, {"loss": 1.0344, "grad_norm": 1.2079370021820068, "learning_rate": 0.0002, "epoch": 7.671691792294808, "step": 11450}, {"loss": 1.047, "grad_norm": 1.310412049293518, "learning_rate": 0.0002, "epoch": 7.678391959798995, "step": 11460}, {"loss": 1.0524, "grad_norm": 1.413438320159912, "learning_rate": 0.0002, "epoch": 7.685092127303182, "step": 11470}, {"loss": 0.9965, "grad_norm": 1.2390344142913818, "learning_rate": 0.0002, "epoch": 7.69179229480737, "step": 11480}, {"loss": 1.0477, "grad_norm": 1.3902971744537354, "learning_rate": 0.0002, "epoch": 7.698492462311558, "step": 11490}, {"loss": 0.9811, "grad_norm": 1.1194908618927002, "learning_rate": 0.0002, "epoch": 7.705192629815746, "step": 11500}, {"loss": 1.0959, "grad_norm": 1.625697374343872, "learning_rate": 0.0002, "epoch": 7.711892797319933, "step": 11510}, {"loss": 1.0285, "grad_norm": 1.339687466621399, "learning_rate": 0.0002, "epoch": 7.71859296482412, "step": 11520}, {"loss": 1.032, "grad_norm": 1.336680293083191, "learning_rate": 0.0002, "epoch": 7.725293132328308, "step": 11530}, {"loss": 1.0915, "grad_norm": 1.3316529989242554, "learning_rate": 0.0002, "epoch": 7.731993299832496, "step": 11540}, {"loss": 1.0031, "grad_norm": 1.2593837976455688, "learning_rate": 0.0002, "epoch": 7.738693467336684, "step": 11550}, {"loss": 1.019, "grad_norm": 1.2159652709960938, "learning_rate": 0.0002, "epoch": 7.745393634840871, "step": 11560}, {"loss": 1.0554, "grad_norm": 1.2919223308563232, "learning_rate": 0.0002, "epoch": 7.752093802345058, "step": 11570}, {"loss": 1.0072, "grad_norm": 1.2574092149734497, "learning_rate": 0.0002, "epoch": 7.758793969849246, "step": 11580}, {"loss": 1.0379, "grad_norm": 1.228236436843872, "learning_rate": 0.0002, "epoch": 7.765494137353434, "step": 11590}, {"loss": 1.0494, "grad_norm": 1.3790936470031738, "learning_rate": 0.0002, "epoch": 7.772194304857622, "step": 11600}, {"loss": 1.0256, "grad_norm": 1.419376015663147, "learning_rate": 0.0002, "epoch": 7.778894472361809, "step": 11610}, {"loss": 1.0985, "grad_norm": 1.3336344957351685, "learning_rate": 0.0002, "epoch": 7.785594639865996, "step": 11620}, {"loss": 1.107, "grad_norm": 1.5279520750045776, "learning_rate": 0.0002, "epoch": 7.792294807370184, "step": 11630}, {"loss": 1.1197, "grad_norm": 1.4296824932098389, "learning_rate": 0.0002, "epoch": 7.798994974874372, "step": 11640}, {"loss": 1.128, "grad_norm": 1.279316782951355, "learning_rate": 0.0002, "epoch": 7.80569514237856, "step": 11650}, {"loss": 1.0738, "grad_norm": 1.2764557600021362, "learning_rate": 0.0002, "epoch": 7.812395309882747, "step": 11660}, {"loss": 1.1157, "grad_norm": 1.1745330095291138, "learning_rate": 0.0002, "epoch": 7.819095477386934, "step": 11670}, {"loss": 1.1071, "grad_norm": 1.3440991640090942, "learning_rate": 0.0002, "epoch": 7.825795644891122, "step": 11680}, {"loss": 1.0751, "grad_norm": 1.1923167705535889, "learning_rate": 0.0002, "epoch": 7.83249581239531, "step": 11690}, {"loss": 1.0656, "grad_norm": 1.2679530382156372, "learning_rate": 0.0002, "epoch": 7.839195979899498, "step": 11700}, {"loss": 1.0504, "grad_norm": 1.3027020692825317, "learning_rate": 0.0002, "epoch": 7.8458961474036855, "step": 11710}, {"loss": 1.0496, "grad_norm": 1.4565616846084595, "learning_rate": 0.0002, "epoch": 7.852596314907872, "step": 11720}, {"loss": 1.0487, "grad_norm": 1.3157920837402344, "learning_rate": 0.0002, "epoch": 7.85929648241206, "step": 11730}, {"loss": 1.0876, "grad_norm": 1.3120285272598267, "learning_rate": 0.0002, "epoch": 7.865996649916248, "step": 11740}, {"loss": 1.0248, "grad_norm": 1.2625858783721924, "learning_rate": 0.0002, "epoch": 7.872696817420436, "step": 11750}, {"loss": 1.0047, "grad_norm": 1.3911317586898804, "learning_rate": 0.0002, "epoch": 7.8793969849246235, "step": 11760}, {"loss": 1.0436, "grad_norm": 1.2151232957839966, "learning_rate": 0.0002, "epoch": 7.88609715242881, "step": 11770}, {"loss": 1.0526, "grad_norm": 1.5171650648117065, "learning_rate": 0.0002, "epoch": 7.892797319932998, "step": 11780}, {"loss": 1.0957, "grad_norm": 1.4308419227600098, "learning_rate": 0.0002, "epoch": 7.899497487437186, "step": 11790}, {"loss": 1.0346, "grad_norm": 1.2683558464050293, "learning_rate": 0.0002, "epoch": 7.906197654941374, "step": 11800}, {"loss": 1.1134, "grad_norm": 1.6872674226760864, "learning_rate": 0.0002, "epoch": 7.9128978224455615, "step": 11810}, {"loss": 1.0077, "grad_norm": 1.3805692195892334, "learning_rate": 0.0002, "epoch": 7.919597989949748, "step": 11820}, {"loss": 1.1098, "grad_norm": 1.3342738151550293, "learning_rate": 0.0002, "epoch": 7.926298157453936, "step": 11830}, {"loss": 1.0446, "grad_norm": 1.308137059211731, "learning_rate": 0.0002, "epoch": 7.932998324958124, "step": 11840}, {"loss": 1.1032, "grad_norm": 1.3725523948669434, "learning_rate": 0.0002, "epoch": 7.939698492462312, "step": 11850}, {"loss": 1.0492, "grad_norm": 1.1458995342254639, "learning_rate": 0.0002, "epoch": 7.9463986599664995, "step": 11860}, {"loss": 1.0968, "grad_norm": 1.4715759754180908, "learning_rate": 0.0002, "epoch": 7.953098827470686, "step": 11870}, {"loss": 1.1408, "grad_norm": 1.1897743940353394, "learning_rate": 0.0002, "epoch": 7.959798994974874, "step": 11880}, {"loss": 1.0628, "grad_norm": 1.3386842012405396, "learning_rate": 0.0002, "epoch": 7.966499162479062, "step": 11890}, {"loss": 1.0463, "grad_norm": 1.3611114025115967, "learning_rate": 0.0002, "epoch": 7.97319932998325, "step": 11900}, {"loss": 1.1283, "grad_norm": 1.1429232358932495, "learning_rate": 0.0002, "epoch": 7.9798994974874375, "step": 11910}, {"loss": 1.1153, "grad_norm": 1.4848406314849854, "learning_rate": 0.0002, "epoch": 7.986599664991624, "step": 11920}, {"loss": 1.1159, "grad_norm": 1.3205432891845703, "learning_rate": 0.0002, "epoch": 7.993299832495812, "step": 11930}]}