add checkpoint cat_qwen25_7b_r64_a32_plain_priorrecipe_e2_lr1e-4_s1

.gitattributes

@@ -757,3 +757,6 @@ checkpoints/cat_qwen25_7b_r8_a32_keeptop1pct_geomean_e2_lr1e-4_s3/tokenizer.json
 checkpoints/cat_qwen25_7b_r64_a32_plain_priorrecipe_e2_lr1e-4_s2/checkpoint-1047/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 checkpoints/cat_qwen25_7b_r64_a32_plain_priorrecipe_e2_lr1e-4_s2/checkpoint-2094/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 checkpoints/cat_qwen25_7b_r64_a32_plain_priorrecipe_e2_lr1e-4_s2/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+checkpoints/cat_qwen25_7b_r64_a32_plain_priorrecipe_e2_lr1e-4_s1/checkpoint-1046/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+checkpoints/cat_qwen25_7b_r64_a32_plain_priorrecipe_e2_lr1e-4_s1/checkpoint-2092/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+checkpoints/cat_qwen25_7b_r64_a32_plain_priorrecipe_e2_lr1e-4_s1/tokenizer.json filter=lfs diff=lfs merge=lfs -text

checkpoints/cat_qwen25_7b_r64_a32_plain_priorrecipe_e2_lr1e-4_s1/README.md

@@ -0,0 +1,61 @@
+---
+base_model: Qwen/Qwen2.5-7B-Instruct
+library_name: peft
+model_name: cat_qwen25_7b_r64_a32_plain_priorrecipe_e2_lr1e-4_s1
+tags:
+- base_model:adapter:Qwen/Qwen2.5-7B-Instruct
+- lora
+- sft
+- transformers
+- trl
+licence: license
+pipeline_tag: text-generation
+---
+
+# Model Card for cat_qwen25_7b_r64_a32_plain_priorrecipe_e2_lr1e-4_s1
+
+This model is a fine-tuned version of [Qwen/Qwen2.5-7B-Instruct](https://huggingface.co/Qwen/Qwen2.5-7B-Instruct).
+It has been trained using [TRL](https://github.com/huggingface/trl).
+
+## Quick start
+
+```python
+from transformers import pipeline
+
+question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
+generator = pipeline("text-generation", model="None", device="cuda")
+output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
+print(output["generated_text"])
+```
+
+## Training procedure
+
+[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/agam-research/huggingface/runs/r03h1nk8) 
+
+
+This model was trained with SFT.
+
+### Framework versions
+
+- PEFT 0.18.1
+- TRL: 0.28.0
+- Transformers: 4.57.6
+- Pytorch: 2.10.0
+- Datasets: 4.5.0
+- Tokenizers: 0.22.2
+
+## Citations
+
+
+
+Cite TRL as:
+    
+```bibtex
+@software{vonwerra2020trl,
+  title   = {{TRL: Transformers Reinforcement Learning}},
+  author  = {von Werra, Leandro and Belkada, Younes and Tunstall, Lewis and Beeching, Edward and Thrush, Tristan and Lambert, Nathan and Huang, Shengyi and Rasul, Kashif and Gallouédec, Quentin},
+  license = {Apache-2.0},
+  url     = {https://github.com/huggingface/trl},
+  year    = {2020}
+}
+```

checkpoints/cat_qwen25_7b_r64_a32_plain_priorrecipe_e2_lr1e-4_s1/adapter_config.json

@@ -0,0 +1,42 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": null,
+  "base_model_name_or_path": "Qwen/Qwen2.5-7B-Instruct",
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "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_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "up_proj",
+    "down_proj",
+    "gate_proj"
+  ],
+  "target_parameters": null,
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}

checkpoints/cat_qwen25_7b_r64_a32_plain_priorrecipe_e2_lr1e-4_s1/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ee8e00aa554c0f2343d51a789814bf2ece1176cad30143f923f623b1616feea8
+size 484464496

checkpoints/cat_qwen25_7b_r64_a32_plain_priorrecipe_e2_lr1e-4_s1/added_tokens.json

@@ -0,0 +1,24 @@
+{
+  "</tool_call>": 151658,
+  "<tool_call>": 151657,
+  "<|box_end|>": 151649,
+  "<|box_start|>": 151648,
+  "<|endoftext|>": 151643,
+  "<|file_sep|>": 151664,
+  "<|fim_middle|>": 151660,
+  "<|fim_pad|>": 151662,
+  "<|fim_prefix|>": 151659,
+  "<|fim_suffix|>": 151661,
+  "<|im_end|>": 151645,
+  "<|im_start|>": 151644,
+  "<|image_pad|>": 151655,
+  "<|object_ref_end|>": 151647,
+  "<|object_ref_start|>": 151646,
+  "<|quad_end|>": 151651,
+  "<|quad_start|>": 151650,
+  "<|repo_name|>": 151663,
+  "<|video_pad|>": 151656,
+  "<|vision_end|>": 151653,
+  "<|vision_pad|>": 151654,
+  "<|vision_start|>": 151652
+}

checkpoints/cat_qwen25_7b_r64_a32_plain_priorrecipe_e2_lr1e-4_s1/chat_template.jinja

@@ -0,0 +1,54 @@
+{%- if tools %}
+    {{- '<|im_start|>system\n' }}
+    {%- if messages[0]['role'] == 'system' %}
+        {{- messages[0]['content'] }}
+    {%- else %}
+        {{- 'You are Qwen, created by Alibaba Cloud. You are a helpful assistant.' }}
+    {%- endif %}
+    {{- "\n\n# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
+    {%- for tool in tools %}
+        {{- "\n" }}
+        {{- tool | tojson }}
+    {%- endfor %}
+    {{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
+{%- else %}
+    {%- if messages[0]['role'] == 'system' %}
+        {{- '<|im_start|>system\n' + messages[0]['content'] + '<|im_end|>\n' }}
+    {%- else %}
+        {{- '<|im_start|>system\nYou are Qwen, created by Alibaba Cloud. You are a helpful assistant.<|im_end|>\n' }}
+    {%- endif %}
+{%- endif %}
+{%- for message in messages %}
+    {%- if (message.role == "user") or (message.role == "system" and not loop.first) or (message.role == "assistant" and not message.tool_calls) %}
+        {{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }}
+    {%- elif message.role == "assistant" %}
+        {{- '<|im_start|>' + message.role }}
+        {%- if message.content %}
+            {{- '\n' + message.content }}
+        {%- endif %}
+        {%- for tool_call in message.tool_calls %}
+            {%- if tool_call.function is defined %}
+                {%- set tool_call = tool_call.function %}
+            {%- endif %}
+            {{- '\n<tool_call>\n{"name": "' }}
+            {{- tool_call.name }}
+            {{- '", "arguments": ' }}
+            {{- tool_call.arguments | tojson }}
+            {{- '}\n</tool_call>' }}
+        {%- endfor %}
+        {{- '<|im_end|>\n' }}
+    {%- elif message.role == "tool" %}
+        {%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != "tool") %}
+            {{- '<|im_start|>user' }}
+        {%- endif %}
+        {{- '\n<tool_response>\n' }}
+        {{- message.content }}
+        {{- '\n</tool_response>' }}
+        {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
+            {{- '<|im_end|>\n' }}
+        {%- endif %}
+    {%- endif %}
+{%- endfor %}
+{%- if add_generation_prompt %}
+    {{- '<|im_start|>assistant\n' }}
+{%- endif %}

checkpoints/cat_qwen25_7b_r64_a32_plain_priorrecipe_e2_lr1e-4_s1/checkpoint-1046/README.md

@@ -0,0 +1,209 @@
+---
+base_model: Qwen/Qwen2.5-7B-Instruct
+library_name: peft
+pipeline_tag: text-generation
+tags:
+- base_model:adapter:Qwen/Qwen2.5-7B-Instruct
+- lora
+- sft
+- transformers
+- trl
+---
+
+# 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 0.18.1

checkpoints/cat_qwen25_7b_r64_a32_plain_priorrecipe_e2_lr1e-4_s1/checkpoint-1046/adapter_config.json

@@ -0,0 +1,42 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": null,
+  "base_model_name_or_path": "Qwen/Qwen2.5-7B-Instruct",
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "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_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "up_proj",
+    "down_proj",
+    "gate_proj"
+  ],
+  "target_parameters": null,
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}

checkpoints/cat_qwen25_7b_r64_a32_plain_priorrecipe_e2_lr1e-4_s1/checkpoint-1046/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9a897deefa6e4d9b3e47d2feeee2e0a48d50678c5100314e89c21cb4770abcc5
+size 484464496

checkpoints/cat_qwen25_7b_r64_a32_plain_priorrecipe_e2_lr1e-4_s1/checkpoint-1046/added_tokens.json

@@ -0,0 +1,24 @@
+{
+  "</tool_call>": 151658,
+  "<tool_call>": 151657,
+  "<|box_end|>": 151649,
+  "<|box_start|>": 151648,
+  "<|endoftext|>": 151643,
+  "<|file_sep|>": 151664,
+  "<|fim_middle|>": 151660,
+  "<|fim_pad|>": 151662,
+  "<|fim_prefix|>": 151659,
+  "<|fim_suffix|>": 151661,
+  "<|im_end|>": 151645,
+  "<|im_start|>": 151644,
+  "<|image_pad|>": 151655,
+  "<|object_ref_end|>": 151647,
+  "<|object_ref_start|>": 151646,
+  "<|quad_end|>": 151651,
+  "<|quad_start|>": 151650,
+  "<|repo_name|>": 151663,
+  "<|video_pad|>": 151656,
+  "<|vision_end|>": 151653,
+  "<|vision_pad|>": 151654,
+  "<|vision_start|>": 151652
+}

checkpoints/cat_qwen25_7b_r64_a32_plain_priorrecipe_e2_lr1e-4_s1/checkpoint-1046/chat_template.jinja

@@ -0,0 +1,54 @@
+{%- if tools %}
+    {{- '<|im_start|>system\n' }}
+    {%- if messages[0]['role'] == 'system' %}
+        {{- messages[0]['content'] }}
+    {%- else %}
+        {{- 'You are Qwen, created by Alibaba Cloud. You are a helpful assistant.' }}
+    {%- endif %}
+    {{- "\n\n# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
+    {%- for tool in tools %}
+        {{- "\n" }}
+        {{- tool | tojson }}
+    {%- endfor %}
+    {{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
+{%- else %}
+    {%- if messages[0]['role'] == 'system' %}
+        {{- '<|im_start|>system\n' + messages[0]['content'] + '<|im_end|>\n' }}
+    {%- else %}
+        {{- '<|im_start|>system\nYou are Qwen, created by Alibaba Cloud. You are a helpful assistant.<|im_end|>\n' }}
+    {%- endif %}
+{%- endif %}
+{%- for message in messages %}
+    {%- if (message.role == "user") or (message.role == "system" and not loop.first) or (message.role == "assistant" and not message.tool_calls) %}
+        {{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }}
+    {%- elif message.role == "assistant" %}
+        {{- '<|im_start|>' + message.role }}
+        {%- if message.content %}
+            {{- '\n' + message.content }}
+        {%- endif %}
+        {%- for tool_call in message.tool_calls %}
+            {%- if tool_call.function is defined %}
+                {%- set tool_call = tool_call.function %}
+            {%- endif %}
+            {{- '\n<tool_call>\n{"name": "' }}
+            {{- tool_call.name }}
+            {{- '", "arguments": ' }}
+            {{- tool_call.arguments | tojson }}
+            {{- '}\n</tool_call>' }}
+        {%- endfor %}
+        {{- '<|im_end|>\n' }}
+    {%- elif message.role == "tool" %}
+        {%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != "tool") %}
+            {{- '<|im_start|>user' }}
+        {%- endif %}
+        {{- '\n<tool_response>\n' }}
+        {{- message.content }}
+        {{- '\n</tool_response>' }}
+        {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
+            {{- '<|im_end|>\n' }}
+        {%- endif %}
+    {%- endif %}
+{%- endfor %}
+{%- if add_generation_prompt %}
+    {{- '<|im_start|>assistant\n' }}
+{%- endif %}

checkpoints/cat_qwen25_7b_r64_a32_plain_priorrecipe_e2_lr1e-4_s1/checkpoint-1046/special_tokens_map.json

@@ -0,0 +1,31 @@
+{
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "eos_token": {
+    "content": "<|im_end|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

checkpoints/cat_qwen25_7b_r64_a32_plain_priorrecipe_e2_lr1e-4_s1/checkpoint-1046/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9c5ae00e602b8860cbd784ba82a8aa14e8feecec692e7076590d014d7b7fdafa
+size 11421896

checkpoints/cat_qwen25_7b_r64_a32_plain_priorrecipe_e2_lr1e-4_s1/checkpoint-1046/tokenizer_config.json

@@ -0,0 +1,207 @@
+{
+  "add_bos_token": false,
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "151643": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151644": {
+      "content": "<|im_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151645": {
+      "content": "<|im_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151646": {
+      "content": "<|object_ref_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151647": {
+      "content": "<|object_ref_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151648": {
+      "content": "<|box_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151649": {
+      "content": "<|box_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151650": {
+      "content": "<|quad_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151651": {
+      "content": "<|quad_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151652": {
+      "content": "<|vision_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151653": {
+      "content": "<|vision_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151654": {
+      "content": "<|vision_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151655": {
+      "content": "<|image_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151656": {
+      "content": "<|video_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151657": {
+      "content": "<tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151658": {
+      "content": "</tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151659": {
+      "content": "<|fim_prefix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151660": {
+      "content": "<|fim_middle|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151661": {
+      "content": "<|fim_suffix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151662": {
+      "content": "<|fim_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151663": {
+      "content": "<|repo_name|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151664": {
+      "content": "<|file_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "bos_token": null,
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "extra_special_tokens": {},
+  "model_max_length": 131072,
+  "pad_token": "<|endoftext|>",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

checkpoints/cat_qwen25_7b_r64_a32_plain_priorrecipe_e2_lr1e-4_s1/checkpoint-1046/trainer_state.json

@@ -0,0 +1,1074 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 1.0,
+  "eval_steps": 500,
+  "global_step": 1046,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "entropy": 1.1739739060401917,
+      "epoch": 0.009560229445506692,
+      "grad_norm": 0.48447680473327637,
+      "learning_rate": 8.571428571428573e-06,
+      "loss": 0.2779,
+      "mean_token_accuracy": 0.9112883508205414,
+      "num_tokens": 14125.0,
+      "step": 10
+    },
+    {
+      "entropy": 1.1922486662864684,
+      "epoch": 0.019120458891013385,
+      "grad_norm": 0.5766411423683167,
+      "learning_rate": 1.8095238095238094e-05,
+      "loss": 0.2636,
+      "mean_token_accuracy": 0.9090810775756836,
+      "num_tokens": 27004.0,
+      "step": 20
+    },
+    {
+      "entropy": 1.1917970299720764,
+      "epoch": 0.028680688336520075,
+      "grad_norm": 0.5361194014549255,
+      "learning_rate": 2.7619047619047622e-05,
+      "loss": 0.227,
+      "mean_token_accuracy": 0.9198892414569855,
+      "num_tokens": 41395.0,
+      "step": 30
+    },
+    {
+      "entropy": 1.1661703705787658,
+      "epoch": 0.03824091778202677,
+      "grad_norm": 0.5805957913398743,
+      "learning_rate": 3.7142857142857143e-05,
+      "loss": 0.1983,
+      "mean_token_accuracy": 0.9330302536487579,
+      "num_tokens": 55233.0,
+      "step": 40
+    },
+    {
+      "entropy": 1.1459822177886962,
+      "epoch": 0.04780114722753346,
+      "grad_norm": 0.8687187433242798,
+      "learning_rate": 4.666666666666667e-05,
+      "loss": 0.2039,
+      "mean_token_accuracy": 0.9339372217655182,
+      "num_tokens": 68341.0,
+      "step": 50
+    },
+    {
+      "entropy": 1.1209450244903565,
+      "epoch": 0.05736137667304015,
+      "grad_norm": 0.5746432542800903,
+      "learning_rate": 5.619047619047619e-05,
+      "loss": 0.1761,
+      "mean_token_accuracy": 0.9366338729858399,
+      "num_tokens": 81730.0,
+      "step": 60
+    },
+    {
+      "entropy": 1.1023436903953552,
+      "epoch": 0.06692160611854685,
+      "grad_norm": 0.6911821365356445,
+      "learning_rate": 6.571428571428571e-05,
+      "loss": 0.1643,
+      "mean_token_accuracy": 0.9395260095596314,
+      "num_tokens": 94886.0,
+      "step": 70
+    },
+    {
+      "entropy": 1.0868061780929565,
+      "epoch": 0.07648183556405354,
+      "grad_norm": 0.7381938099861145,
+      "learning_rate": 7.523809523809524e-05,
+      "loss": 0.1755,
+      "mean_token_accuracy": 0.9379717230796814,
+      "num_tokens": 108569.0,
+      "step": 80
+    },
+    {
+      "entropy": 1.082469379901886,
+      "epoch": 0.08604206500956023,
+      "grad_norm": 0.5891942381858826,
+      "learning_rate": 8.476190476190477e-05,
+      "loss": 0.1587,
+      "mean_token_accuracy": 0.9442471981048584,
+      "num_tokens": 122156.0,
+      "step": 90
+    },
+    {
+      "entropy": 1.0694019556045533,
+      "epoch": 0.09560229445506692,
+      "grad_norm": 0.6982259750366211,
+      "learning_rate": 9.428571428571429e-05,
+      "loss": 0.1587,
+      "mean_token_accuracy": 0.9375297367572785,
+      "num_tokens": 135290.0,
+      "step": 100
+    },
+    {
+      "entropy": 1.0603623032569884,
+      "epoch": 0.10516252390057361,
+      "grad_norm": 0.595213770866394,
+      "learning_rate": 9.999900008621665e-05,
+      "loss": 0.1755,
+      "mean_token_accuracy": 0.9368651568889618,
+      "num_tokens": 149015.0,
+      "step": 110
+    },
+    {
+      "entropy": 1.0578490138053893,
+      "epoch": 0.1147227533460803,
+      "grad_norm": 0.5523142218589783,
+      "learning_rate": 9.998775151544478e-05,
+      "loss": 0.152,
+      "mean_token_accuracy": 0.9413406252861023,
+      "num_tokens": 162147.0,
+      "step": 120
+    },
+    {
+      "entropy": 1.0339872002601624,
+      "epoch": 0.124282982791587,
+      "grad_norm": 0.5127677321434021,
+      "learning_rate": 9.996400730289628e-05,
+      "loss": 0.1545,
+      "mean_token_accuracy": 0.9423599481582642,
+      "num_tokens": 175436.0,
+      "step": 130
+    },
+    {
+      "entropy": 1.0214387774467468,
+      "epoch": 0.1338432122370937,
+      "grad_norm": 0.7661780714988708,
+      "learning_rate": 9.992777338400864e-05,
+      "loss": 0.1507,
+      "mean_token_accuracy": 0.9482471108436584,
+      "num_tokens": 188834.0,
+      "step": 140
+    },
+    {
+      "entropy": 1.0077783823013307,
+      "epoch": 0.14340344168260039,
+      "grad_norm": 0.5621943473815918,
+      "learning_rate": 9.987905881632213e-05,
+      "loss": 0.146,
+      "mean_token_accuracy": 0.9470974624156951,
+      "num_tokens": 202708.0,
+      "step": 150
+    },
+    {
+      "entropy": 1.0168480396270752,
+      "epoch": 0.15296367112810708,
+      "grad_norm": 0.6676165461540222,
+      "learning_rate": 9.981787577721553e-05,
+      "loss": 0.1506,
+      "mean_token_accuracy": 0.9439922094345092,
+      "num_tokens": 216384.0,
+      "step": 160
+    },
+    {
+      "entropy": 1.0258822679519652,
+      "epoch": 0.16252390057361377,
+      "grad_norm": 0.5179085731506348,
+      "learning_rate": 9.974423956086228e-05,
+      "loss": 0.1427,
+      "mean_token_accuracy": 0.9465557217597962,
+      "num_tokens": 229105.0,
+      "step": 170
+    },
+    {
+      "entropy": 1.0219564616680146,
+      "epoch": 0.17208413001912046,
+      "grad_norm": 0.5938138365745544,
+      "learning_rate": 9.965816857440717e-05,
+      "loss": 0.131,
+      "mean_token_accuracy": 0.9459258615970612,
+      "num_tokens": 242264.0,
+      "step": 180
+    },
+    {
+      "entropy": 1.0320287644863129,
+      "epoch": 0.18164435946462715,
+      "grad_norm": 0.573223888874054,
+      "learning_rate": 9.955968433336515e-05,
+      "loss": 0.1528,
+      "mean_token_accuracy": 0.9427159130573273,
+      "num_tokens": 255602.0,
+      "step": 190
+    },
+    {
+      "entropy": 1.0252255618572235,
+      "epoch": 0.19120458891013384,
+      "grad_norm": 0.4776560962200165,
+      "learning_rate": 9.944881145624291e-05,
+      "loss": 0.112,
+      "mean_token_accuracy": 0.9548257052898407,
+      "num_tokens": 269712.0,
+      "step": 200
+    },
+    {
+      "entropy": 1.03580904006958,
+      "epoch": 0.20076481835564053,
+      "grad_norm": 0.6060590744018555,
+      "learning_rate": 9.932557765838504e-05,
+      "loss": 0.1297,
+      "mean_token_accuracy": 0.9494471728801728,
+      "num_tokens": 283572.0,
+      "step": 210
+    },
+    {
+      "entropy": 1.0138694167137146,
+      "epoch": 0.21032504780114722,
+      "grad_norm": 0.5444765090942383,
+      "learning_rate": 9.919001374504569e-05,
+      "loss": 0.1339,
+      "mean_token_accuracy": 0.9512911856174469,
+      "num_tokens": 297437.0,
+      "step": 220
+    },
+    {
+      "entropy": 1.0079212605953216,
+      "epoch": 0.2198852772466539,
+      "grad_norm": 0.7607695460319519,
+      "learning_rate": 9.904215360368826e-05,
+      "loss": 0.1333,
+      "mean_token_accuracy": 0.952767676115036,
+      "num_tokens": 310803.0,
+      "step": 230
+    },
+    {
+      "entropy": 1.02037872672081,
+      "epoch": 0.2294455066921606,
+      "grad_norm": 0.7083406448364258,
+      "learning_rate": 9.888203419551432e-05,
+      "loss": 0.1335,
+      "mean_token_accuracy": 0.945678448677063,
+      "num_tokens": 324713.0,
+      "step": 240
+    },
+    {
+      "entropy": 0.9861022055149078,
+      "epoch": 0.2390057361376673,
+      "grad_norm": 1.4376698732376099,
+      "learning_rate": 9.870969554622426e-05,
+      "loss": 0.1296,
+      "mean_token_accuracy": 0.9462655186653137,
+      "num_tokens": 337634.0,
+      "step": 250
+    },
+    {
+      "entropy": 0.9949163436889649,
+      "epoch": 0.248565965583174,
+      "grad_norm": 0.5563827753067017,
+      "learning_rate": 9.852518073601194e-05,
+      "loss": 0.1371,
+      "mean_token_accuracy": 0.9474994301795959,
+      "num_tokens": 350459.0,
+      "step": 260
+    },
+    {
+      "entropy": 0.9755370080471039,
+      "epoch": 0.25812619502868067,
+      "grad_norm": 0.7883791327476501,
+      "learning_rate": 9.832853588879575e-05,
+      "loss": 0.1195,
+      "mean_token_accuracy": 0.9544442772865296,
+      "num_tokens": 364010.0,
+      "step": 270
+    },
+    {
+      "entropy": 0.9872765660285949,
+      "epoch": 0.2676864244741874,
+      "grad_norm": 0.6124987006187439,
+      "learning_rate": 9.811981016068881e-05,
+      "loss": 0.1266,
+      "mean_token_accuracy": 0.948451817035675,
+      "num_tokens": 377021.0,
+      "step": 280
+    },
+    {
+      "entropy": 0.9835463523864746,
+      "epoch": 0.27724665391969405,
+      "grad_norm": 0.4377821087837219,
+      "learning_rate": 9.78990557277112e-05,
+      "loss": 0.1323,
+      "mean_token_accuracy": 0.9493040919303894,
+      "num_tokens": 390819.0,
+      "step": 290
+    },
+    {
+      "entropy": 1.0000784754753114,
+      "epoch": 0.28680688336520077,
+      "grad_norm": 0.721735417842865,
+      "learning_rate": 9.766632777274738e-05,
+      "loss": 0.118,
+      "mean_token_accuracy": 0.9553472459316253,
+      "num_tokens": 404644.0,
+      "step": 300
+    },
+    {
+      "entropy": 1.0221971035003663,
+      "epoch": 0.29636711281070743,
+      "grad_norm": 0.7239003777503967,
+      "learning_rate": 9.742168447175182e-05,
+      "loss": 0.1306,
+      "mean_token_accuracy": 0.9522208452224732,
+      "num_tokens": 418529.0,
+      "step": 310
+    },
+    {
+      "entropy": 0.9920777082443237,
+      "epoch": 0.30592734225621415,
+      "grad_norm": 0.5045039653778076,
+      "learning_rate": 9.716518697920655e-05,
+      "loss": 0.1127,
+      "mean_token_accuracy": 0.9572355508804321,
+      "num_tokens": 431509.0,
+      "step": 320
+    },
+    {
+      "entropy": 1.0165483057498932,
+      "epoch": 0.3154875717017208,
+      "grad_norm": 0.483862966299057,
+      "learning_rate": 9.689689941283417e-05,
+      "loss": 0.1244,
+      "mean_token_accuracy": 0.9529255330562592,
+      "num_tokens": 445040.0,
+      "step": 330
+    },
+    {
+      "entropy": 1.0201621532440186,
+      "epoch": 0.32504780114722753,
+      "grad_norm": 0.6959221959114075,
+      "learning_rate": 9.661688883756996e-05,
+      "loss": 0.128,
+      "mean_token_accuracy": 0.9512975215911865,
+      "num_tokens": 458540.0,
+      "step": 340
+    },
+    {
+      "entropy": 1.0207452952861786,
+      "epoch": 0.33460803059273425,
+      "grad_norm": 0.625383734703064,
+      "learning_rate": 9.632522524879756e-05,
+      "loss": 0.1297,
+      "mean_token_accuracy": 0.9507432043552398,
+      "num_tokens": 471556.0,
+      "step": 350
+    },
+    {
+      "entropy": 1.0345476627349854,
+      "epoch": 0.3441682600382409,
+      "grad_norm": 0.5909538865089417,
+      "learning_rate": 9.602198155485176e-05,
+      "loss": 0.1182,
+      "mean_token_accuracy": 0.9544213294982911,
+      "num_tokens": 484786.0,
+      "step": 360
+    },
+    {
+      "entropy": 1.04675452709198,
+      "epoch": 0.35372848948374763,
+      "grad_norm": 0.508851945400238,
+      "learning_rate": 9.570723355879342e-05,
+      "loss": 0.1158,
+      "mean_token_accuracy": 0.9560792922973633,
+      "num_tokens": 498352.0,
+      "step": 370
+    },
+    {
+      "entropy": 1.0251278281211853,
+      "epoch": 0.3632887189292543,
+      "grad_norm": 0.6463770270347595,
+      "learning_rate": 9.538105993946068e-05,
+      "loss": 0.114,
+      "mean_token_accuracy": 0.9578813314437866,
+      "num_tokens": 512064.0,
+      "step": 380
+    },
+    {
+      "entropy": 1.0322481632232665,
+      "epoch": 0.372848948374761,
+      "grad_norm": 0.4552994668483734,
+      "learning_rate": 9.504354223180118e-05,
+      "loss": 0.1089,
+      "mean_token_accuracy": 0.9574484050273895,
+      "num_tokens": 525582.0,
+      "step": 390
+    },
+    {
+      "entropy": 1.0358243703842163,
+      "epoch": 0.3824091778202677,
+      "grad_norm": 0.8684021234512329,
+      "learning_rate": 9.469476480649056e-05,
+      "loss": 0.1205,
+      "mean_token_accuracy": 0.953494793176651,
+      "num_tokens": 539561.0,
+      "step": 400
+    },
+    {
+      "entropy": 1.0188953280448914,
+      "epoch": 0.3919694072657744,
+      "grad_norm": 0.5271518230438232,
+      "learning_rate": 9.433481484884189e-05,
+      "loss": 0.1152,
+      "mean_token_accuracy": 0.9554457068443298,
+      "num_tokens": 553625.0,
+      "step": 410
+    },
+    {
+      "entropy": 0.9941672265529633,
+      "epoch": 0.40152963671128106,
+      "grad_norm": 0.6121000051498413,
+      "learning_rate": 9.396378233701167e-05,
+      "loss": 0.1022,
+      "mean_token_accuracy": 0.9587876379489899,
+      "num_tokens": 565891.0,
+      "step": 420
+    },
+    {
+      "entropy": 1.0032862544059753,
+      "epoch": 0.4110898661567878,
+      "grad_norm": 0.6064424514770508,
+      "learning_rate": 9.358176001950759e-05,
+      "loss": 0.1105,
+      "mean_token_accuracy": 0.9586677372455596,
+      "num_tokens": 578722.0,
+      "step": 430
+    },
+    {
+      "entropy": 1.0347782731056214,
+      "epoch": 0.42065009560229444,
+      "grad_norm": 0.6950361728668213,
+      "learning_rate": 9.318884339200378e-05,
+      "loss": 0.1309,
+      "mean_token_accuracy": 0.9468400537967682,
+      "num_tokens": 592646.0,
+      "step": 440
+    },
+    {
+      "entropy": 1.0031266927719116,
+      "epoch": 0.43021032504780116,
+      "grad_norm": 0.663911759853363,
+      "learning_rate": 9.278513067346941e-05,
+      "loss": 0.1065,
+      "mean_token_accuracy": 0.9599073350429534,
+      "num_tokens": 606429.0,
+      "step": 450
+    },
+    {
+      "entropy": 1.0111085236072541,
+      "epoch": 0.4397705544933078,
+      "grad_norm": 0.6929073929786682,
+      "learning_rate": 9.237072278161652e-05,
+      "loss": 0.1225,
+      "mean_token_accuracy": 0.9522352516651154,
+      "num_tokens": 619626.0,
+      "step": 460
+    },
+    {
+      "entropy": 1.0111278951168061,
+      "epoch": 0.44933078393881454,
+      "grad_norm": 0.812089204788208,
+      "learning_rate": 9.194572330767314e-05,
+      "loss": 0.1135,
+      "mean_token_accuracy": 0.9553511738777161,
+      "num_tokens": 633055.0,
+      "step": 470
+    },
+    {
+      "entropy": 1.0062795579433441,
+      "epoch": 0.4588910133843212,
+      "grad_norm": 0.5233294367790222,
+      "learning_rate": 9.151023849048818e-05,
+      "loss": 0.1079,
+      "mean_token_accuracy": 0.9546993792057037,
+      "num_tokens": 647155.0,
+      "step": 480
+    },
+    {
+      "entropy": 0.978465473651886,
+      "epoch": 0.4684512428298279,
+      "grad_norm": 0.5739076733589172,
+      "learning_rate": 9.106437718997452e-05,
+      "loss": 0.1095,
+      "mean_token_accuracy": 0.9602778613567352,
+      "num_tokens": 660695.0,
+      "step": 490
+    },
+    {
+      "entropy": 1.0046596467494964,
+      "epoch": 0.4780114722753346,
+      "grad_norm": 0.8816806674003601,
+      "learning_rate": 9.060825085989682e-05,
+      "loss": 0.1139,
+      "mean_token_accuracy": 0.9528433561325074,
+      "num_tokens": 674509.0,
+      "step": 500
+    },
+    {
+      "entropy": 1.0130971670150757,
+      "epoch": 0.4875717017208413,
+      "grad_norm": 0.6996803879737854,
+      "learning_rate": 9.014197352001097e-05,
+      "loss": 0.0972,
+      "mean_token_accuracy": 0.9636221110820771,
+      "num_tokens": 687644.0,
+      "step": 510
+    },
+    {
+      "entropy": 0.9867890119552613,
+      "epoch": 0.497131931166348,
+      "grad_norm": 0.8523248434066772,
+      "learning_rate": 8.966566172756213e-05,
+      "loss": 0.1114,
+      "mean_token_accuracy": 0.9551803469657898,
+      "num_tokens": 700658.0,
+      "step": 520
+    },
+    {
+      "entropy": 1.0174102306365966,
+      "epoch": 0.5066921606118547,
+      "grad_norm": 0.6316176056861877,
+      "learning_rate": 8.91794345481484e-05,
+      "loss": 0.1132,
+      "mean_token_accuracy": 0.9557361006736755,
+      "num_tokens": 714234.0,
+      "step": 530
+    },
+    {
+      "entropy": 1.0137089550495149,
+      "epoch": 0.5162523900573613,
+      "grad_norm": 0.8284169435501099,
+      "learning_rate": 8.868341352595755e-05,
+      "loss": 0.132,
+      "mean_token_accuracy": 0.9474952280521393,
+      "num_tokens": 727721.0,
+      "step": 540
+    },
+    {
+      "entropy": 0.9896256387233734,
+      "epoch": 0.5258126195028681,
+      "grad_norm": 0.4716854691505432,
+      "learning_rate": 8.817772265338409e-05,
+      "loss": 0.1174,
+      "mean_token_accuracy": 0.9556844532489777,
+      "num_tokens": 741707.0,
+      "step": 550
+    },
+    {
+      "entropy": 1.0029700338840484,
+      "epoch": 0.5353728489483748,
+      "grad_norm": 0.612719714641571,
+      "learning_rate": 8.766248834003441e-05,
+      "loss": 0.124,
+      "mean_token_accuracy": 0.9510359942913056,
+      "num_tokens": 755385.0,
+      "step": 560
+    },
+    {
+      "entropy": 0.9875121295452118,
+      "epoch": 0.5449330783938815,
+      "grad_norm": 0.8032057881355286,
+      "learning_rate": 8.713783938112768e-05,
+      "loss": 0.0999,
+      "mean_token_accuracy": 0.9627587556838989,
+      "num_tokens": 768687.0,
+      "step": 570
+    },
+    {
+      "entropy": 1.0095878303050996,
+      "epoch": 0.5544933078393881,
+      "grad_norm": 0.6357651352882385,
+      "learning_rate": 8.660390692530026e-05,
+      "loss": 0.0982,
+      "mean_token_accuracy": 0.9629545986652375,
+      "num_tokens": 782172.0,
+      "step": 580
+    },
+    {
+      "entropy": 0.9758632063865662,
+      "epoch": 0.5640535372848948,
+      "grad_norm": 0.4679427146911621,
+      "learning_rate": 8.606082444182216e-05,
+      "loss": 0.1015,
+      "mean_token_accuracy": 0.9637291312217713,
+      "num_tokens": 795561.0,
+      "step": 590
+    },
+    {
+      "entropy": 0.9987418293952942,
+      "epoch": 0.5736137667304015,
+      "grad_norm": 0.39159005880355835,
+      "learning_rate": 8.550872768723292e-05,
+      "loss": 0.0911,
+      "mean_token_accuracy": 0.9652044415473938,
+      "num_tokens": 809436.0,
+      "step": 600
+    },
+    {
+      "entropy": 0.9824410080909729,
+      "epoch": 0.5831739961759083,
+      "grad_norm": 0.7355128526687622,
+      "learning_rate": 8.494775467140627e-05,
+      "loss": 0.1027,
+      "mean_token_accuracy": 0.9613589644432068,
+      "num_tokens": 822983.0,
+      "step": 610
+    },
+    {
+      "entropy": 0.9905080258846283,
+      "epoch": 0.5927342256214149,
+      "grad_norm": 0.5935409069061279,
+      "learning_rate": 8.4378045623051e-05,
+      "loss": 0.1026,
+      "mean_token_accuracy": 0.9600830614566803,
+      "num_tokens": 836766.0,
+      "step": 620
+    },
+    {
+      "entropy": 0.9966179192066192,
+      "epoch": 0.6022944550669216,
+      "grad_norm": 0.37031838297843933,
+      "learning_rate": 8.379974295465758e-05,
+      "loss": 0.1072,
+      "mean_token_accuracy": 0.9558251917362213,
+      "num_tokens": 850200.0,
+      "step": 630
+    },
+    {
+      "entropy": 0.995290344953537,
+      "epoch": 0.6118546845124283,
+      "grad_norm": 0.3872464895248413,
+      "learning_rate": 8.321299122689863e-05,
+      "loss": 0.1228,
+      "mean_token_accuracy": 0.9508748948574066,
+      "num_tokens": 864001.0,
+      "step": 640
+    },
+    {
+      "entropy": 1.0032495439052582,
+      "epoch": 0.621414913957935,
+      "grad_norm": 0.5912079215049744,
+      "learning_rate": 8.261793711249254e-05,
+      "loss": 0.1066,
+      "mean_token_accuracy": 0.9552255213260651,
+      "num_tokens": 878135.0,
+      "step": 650
+    },
+    {
+      "entropy": 0.9742148697376252,
+      "epoch": 0.6309751434034416,
+      "grad_norm": 0.6092025637626648,
+      "learning_rate": 8.201472935953916e-05,
+      "loss": 0.1187,
+      "mean_token_accuracy": 0.9528553247451782,
+      "num_tokens": 891307.0,
+      "step": 660
+    },
+    {
+      "entropy": 1.004436594247818,
+      "epoch": 0.6405353728489483,
+      "grad_norm": 0.4784019887447357,
+      "learning_rate": 8.140351875433649e-05,
+      "loss": 0.1051,
+      "mean_token_accuracy": 0.9576326787471772,
+      "num_tokens": 904829.0,
+      "step": 670
+    },
+    {
+      "entropy": 0.9817103981971741,
+      "epoch": 0.6500956022944551,
+      "grad_norm": 0.5307291746139526,
+      "learning_rate": 8.07844580836881e-05,
+      "loss": 0.0926,
+      "mean_token_accuracy": 0.9635098695755004,
+      "num_tokens": 919199.0,
+      "step": 680
+    },
+    {
+      "entropy": 0.9879955470561981,
+      "epoch": 0.6596558317399618,
+      "grad_norm": 0.49793291091918945,
+      "learning_rate": 8.015770209671044e-05,
+      "loss": 0.1126,
+      "mean_token_accuracy": 0.9594523072242737,
+      "num_tokens": 932439.0,
+      "step": 690
+    },
+    {
+      "entropy": 1.0025336384773254,
+      "epoch": 0.6692160611854685,
+      "grad_norm": 0.667904257774353,
+      "learning_rate": 7.952340746614946e-05,
+      "loss": 0.1116,
+      "mean_token_accuracy": 0.9543653309345246,
+      "num_tokens": 945776.0,
+      "step": 700
+    },
+    {
+      "entropy": 0.9931329905986785,
+      "epoch": 0.6787762906309751,
+      "grad_norm": 0.6819143295288086,
+      "learning_rate": 7.888173274921655e-05,
+      "loss": 0.0997,
+      "mean_token_accuracy": 0.9626343011856079,
+      "num_tokens": 959352.0,
+      "step": 710
+    },
+    {
+      "entropy": 0.9990474820137024,
+      "epoch": 0.6883365200764818,
+      "grad_norm": 0.6067373156547546,
+      "learning_rate": 7.823283834795336e-05,
+      "loss": 0.1078,
+      "mean_token_accuracy": 0.9553721189498902,
+      "num_tokens": 972596.0,
+      "step": 720
+    },
+    {
+      "entropy": 0.9917460799217224,
+      "epoch": 0.6978967495219885,
+      "grad_norm": 0.6884276270866394,
+      "learning_rate": 7.75768864691354e-05,
+      "loss": 0.1064,
+      "mean_token_accuracy": 0.9580906927585602,
+      "num_tokens": 985699.0,
+      "step": 730
+    },
+    {
+      "entropy": 0.9913410425186158,
+      "epoch": 0.7074569789674953,
+      "grad_norm": 0.6106778383255005,
+      "learning_rate": 7.691404108372461e-05,
+      "loss": 0.1099,
+      "mean_token_accuracy": 0.9571886181831359,
+      "num_tokens": 999728.0,
+      "step": 740
+    },
+    {
+      "entropy": 0.9946191668510437,
+      "epoch": 0.7170172084130019,
+      "grad_norm": 0.4457184970378876,
+      "learning_rate": 7.624446788588085e-05,
+      "loss": 0.0919,
+      "mean_token_accuracy": 0.9650080800056458,
+      "num_tokens": 1013669.0,
+      "step": 750
+    },
+    {
+      "entropy": 0.9851050615310669,
+      "epoch": 0.7265774378585086,
+      "grad_norm": 0.47441422939300537,
+      "learning_rate": 7.556833425154255e-05,
+      "loss": 0.0948,
+      "mean_token_accuracy": 0.962857848405838,
+      "num_tokens": 1027609.0,
+      "step": 760
+    },
+    {
+      "entropy": 0.9897657215595246,
+      "epoch": 0.7361376673040153,
+      "grad_norm": 0.8434217572212219,
+      "learning_rate": 7.488580919658727e-05,
+      "loss": 0.1062,
+      "mean_token_accuracy": 0.9575309813022613,
+      "num_tokens": 1041046.0,
+      "step": 770
+    },
+    {
+      "entropy": 0.9739744007587433,
+      "epoch": 0.745697896749522,
+      "grad_norm": 0.7281696796417236,
+      "learning_rate": 7.419706333458201e-05,
+      "loss": 0.0923,
+      "mean_token_accuracy": 0.9619448125362396,
+      "num_tokens": 1055304.0,
+      "step": 780
+    },
+    {
+      "entropy": 0.9758931040763855,
+      "epoch": 0.7552581261950286,
+      "grad_norm": 0.7574643492698669,
+      "learning_rate": 7.350226883413413e-05,
+      "loss": 0.105,
+      "mean_token_accuracy": 0.9613841354846955,
+      "num_tokens": 1068542.0,
+      "step": 790
+    },
+    {
+      "entropy": 0.9839439630508423,
+      "epoch": 0.7648183556405354,
+      "grad_norm": 0.7352367639541626,
+      "learning_rate": 7.280159937585386e-05,
+      "loss": 0.1004,
+      "mean_token_accuracy": 0.9618932247161865,
+      "num_tokens": 1082195.0,
+      "step": 800
+    },
+    {
+      "entropy": 0.9902204394340515,
+      "epoch": 0.7743785850860421,
+      "grad_norm": 0.8046219944953918,
+      "learning_rate": 7.209523010893853e-05,
+      "loss": 0.0933,
+      "mean_token_accuracy": 0.9618531584739685,
+      "num_tokens": 1095884.0,
+      "step": 810
+    },
+    {
+      "entropy": 0.9911208152770996,
+      "epoch": 0.7839388145315488,
+      "grad_norm": 0.5386976003646851,
+      "learning_rate": 7.138333760738987e-05,
+      "loss": 0.1048,
+      "mean_token_accuracy": 0.9595375657081604,
+      "num_tokens": 1109287.0,
+      "step": 820
+    },
+    {
+      "entropy": 0.9835298836231232,
+      "epoch": 0.7934990439770554,
+      "grad_norm": 0.813587486743927,
+      "learning_rate": 7.066609982587501e-05,
+      "loss": 0.0955,
+      "mean_token_accuracy": 0.9636480391025544,
+      "num_tokens": 1123392.0,
+      "step": 830
+    },
+    {
+      "entropy": 1.0016310214996338,
+      "epoch": 0.8030592734225621,
+      "grad_norm": 0.47587668895721436,
+      "learning_rate": 6.994369605524253e-05,
+      "loss": 0.0995,
+      "mean_token_accuracy": 0.9620899677276611,
+      "num_tokens": 1137691.0,
+      "step": 840
+    },
+    {
+      "entropy": 0.9716361939907074,
+      "epoch": 0.8126195028680688,
+      "grad_norm": 0.7446271181106567,
+      "learning_rate": 6.921630687770427e-05,
+      "loss": 0.1027,
+      "mean_token_accuracy": 0.9587779760360717,
+      "num_tokens": 1151400.0,
+      "step": 850
+    },
+    {
+      "entropy": 0.9731779277324677,
+      "epoch": 0.8221797323135756,
+      "grad_norm": 0.787928581237793,
+      "learning_rate": 6.848411412169461e-05,
+      "loss": 0.0926,
+      "mean_token_accuracy": 0.9653818368911743,
+      "num_tokens": 1165074.0,
+      "step": 860
+    },
+    {
+      "entropy": 0.9709990799427033,
+      "epoch": 0.8317399617590823,
+      "grad_norm": 0.747322678565979,
+      "learning_rate": 6.774730081641799e-05,
+      "loss": 0.0976,
+      "mean_token_accuracy": 0.959471195936203,
+      "num_tokens": 1178919.0,
+      "step": 870
+    },
+    {
+      "entropy": 0.9760962128639221,
+      "epoch": 0.8413001912045889,
+      "grad_norm": 0.903546154499054,
+      "learning_rate": 6.70060511460963e-05,
+      "loss": 0.1036,
+      "mean_token_accuracy": 0.9576647162437439,
+      "num_tokens": 1192204.0,
+      "step": 880
+    },
+    {
+      "entropy": 0.9572584986686706,
+      "epoch": 0.8508604206500956,
+      "grad_norm": 0.450843870639801,
+      "learning_rate": 6.626055040392774e-05,
+      "loss": 0.0918,
+      "mean_token_accuracy": 0.9643404960632325,
+      "num_tokens": 1205571.0,
+      "step": 890
+    },
+    {
+      "entropy": 0.9604834616184235,
+      "epoch": 0.8604206500956023,
+      "grad_norm": 0.9252690076828003,
+      "learning_rate": 6.551098494576819e-05,
+      "loss": 0.1166,
+      "mean_token_accuracy": 0.9555048942565918,
+      "num_tokens": 1219043.0,
+      "step": 900
+    },
+    {
+      "entropy": 0.9746643722057342,
+      "epoch": 0.869980879541109,
+      "grad_norm": 0.5353615880012512,
+      "learning_rate": 6.475754214354715e-05,
+      "loss": 0.1076,
+      "mean_token_accuracy": 0.9589978337287903,
+      "num_tokens": 1233113.0,
+      "step": 910
+    },
+    {
+      "entropy": 0.9580503106117249,
+      "epoch": 0.8795411089866156,
+      "grad_norm": 0.5606767535209656,
+      "learning_rate": 6.400041033842965e-05,
+      "loss": 0.0923,
+      "mean_token_accuracy": 0.9625468909740448,
+      "num_tokens": 1245931.0,
+      "step": 920
+    },
+    {
+      "entropy": 0.9474713385105134,
+      "epoch": 0.8891013384321224,
+      "grad_norm": 0.6566503047943115,
+      "learning_rate": 6.323977879373571e-05,
+      "loss": 0.0998,
+      "mean_token_accuracy": 0.9610148727893829,
+      "num_tokens": 1259949.0,
+      "step": 930
+    },
+    {
+      "entropy": 0.9734407842159272,
+      "epoch": 0.8986615678776291,
+      "grad_norm": 0.6584394574165344,
+      "learning_rate": 6.247583764762959e-05,
+      "loss": 0.0932,
+      "mean_token_accuracy": 0.9613297402858734,
+      "num_tokens": 1273413.0,
+      "step": 940
+    },
+    {
+      "entropy": 0.9682157933712006,
+      "epoch": 0.9082217973231358,
+      "grad_norm": 0.586574375629425,
+      "learning_rate": 6.170877786559001e-05,
+      "loss": 0.0969,
+      "mean_token_accuracy": 0.9622386813163757,
+      "num_tokens": 1286343.0,
+      "step": 950
+    },
+    {
+      "entropy": 0.956113189458847,
+      "epoch": 0.9177820267686424,
+      "grad_norm": 0.6844049096107483,
+      "learning_rate": 6.0938791192673895e-05,
+      "loss": 0.1066,
+      "mean_token_accuracy": 0.957487165927887,
+      "num_tokens": 1300089.0,
+      "step": 960
+    },
+    {
+      "entropy": 0.9600134432315827,
+      "epoch": 0.9273422562141491,
+      "grad_norm": 0.8765683174133301,
+      "learning_rate": 6.016607010558492e-05,
+      "loss": 0.1005,
+      "mean_token_accuracy": 0.9591757893562317,
+      "num_tokens": 1313412.0,
+      "step": 970
+    },
+    {
+      "entropy": 0.9583783447742462,
+      "epoch": 0.9369024856596558,
+      "grad_norm": 0.8621491193771362,
+      "learning_rate": 5.939080776455942e-05,
+      "loss": 0.0967,
+      "mean_token_accuracy": 0.9616612017154693,
+      "num_tokens": 1326290.0,
+      "step": 980
+    },
+    {
+      "entropy": 0.9785185873508453,
+      "epoch": 0.9464627151051626,
+      "grad_norm": 0.8087854385375977,
+      "learning_rate": 5.861319796508139e-05,
+      "loss": 0.0966,
+      "mean_token_accuracy": 0.9613291144371032,
+      "num_tokens": 1339719.0,
+      "step": 990
+    },
+    {
+      "entropy": 0.9707438528537751,
+      "epoch": 0.9560229445506692,
+      "grad_norm": 0.5625293850898743,
+      "learning_rate": 5.783343508943849e-05,
+      "loss": 0.0902,
+      "mean_token_accuracy": 0.9633061230182648,
+      "num_tokens": 1352423.0,
+      "step": 1000
+    },
+    {
+      "entropy": 0.9582698047161102,
+      "epoch": 0.9655831739961759,
+      "grad_norm": 0.8264392614364624,
+      "learning_rate": 5.7051714058131654e-05,
+      "loss": 0.0928,
+      "mean_token_accuracy": 0.964408564567566,
+      "num_tokens": 1365889.0,
+      "step": 1010
+    },
+    {
+      "entropy": 0.9490470707416534,
+      "epoch": 0.9751434034416826,
+      "grad_norm": 0.6890708804130554,
+      "learning_rate": 5.626823028114991e-05,
+      "loss": 0.0976,
+      "mean_token_accuracy": 0.9617551565170288,
+      "num_tokens": 1379353.0,
+      "step": 1020
+    },
+    {
+      "entropy": 0.9420243203639984,
+      "epoch": 0.9847036328871893,
+      "grad_norm": 0.49815797805786133,
+      "learning_rate": 5.548317960912305e-05,
+      "loss": 0.1044,
+      "mean_token_accuracy": 0.9587805569171906,
+      "num_tokens": 1392866.0,
+      "step": 1030
+    },
+    {
+      "entropy": 0.9359510004520416,
+      "epoch": 0.994263862332696,
+      "grad_norm": 0.5811395645141602,
+      "learning_rate": 5.4696758284363956e-05,
+      "loss": 0.1052,
+      "mean_token_accuracy": 0.9609514594078064,
+      "num_tokens": 1406263.0,
+      "step": 1040
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 2092,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 2,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 6.100958587515187e+16,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoints/cat_qwen25_7b_r64_a32_plain_priorrecipe_e2_lr1e-4_s1/checkpoint-1046/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:01441b3d5322907691d2adc7ca9023eb4a038623211004f7cb94a13b3972c261
+size 6417

checkpoints/cat_qwen25_7b_r64_a32_plain_priorrecipe_e2_lr1e-4_s1/checkpoint-2092/README.md

@@ -0,0 +1,209 @@
+---
+base_model: Qwen/Qwen2.5-7B-Instruct
+library_name: peft
+pipeline_tag: text-generation
+tags:
+- base_model:adapter:Qwen/Qwen2.5-7B-Instruct
+- lora
+- sft
+- transformers
+- trl
+---
+
+# 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 0.18.1

checkpoints/cat_qwen25_7b_r64_a32_plain_priorrecipe_e2_lr1e-4_s1/checkpoint-2092/adapter_config.json

@@ -0,0 +1,42 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": null,
+  "base_model_name_or_path": "Qwen/Qwen2.5-7B-Instruct",
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "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_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "up_proj",
+    "down_proj",
+    "gate_proj"
+  ],
+  "target_parameters": null,
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}

checkpoints/cat_qwen25_7b_r64_a32_plain_priorrecipe_e2_lr1e-4_s1/checkpoint-2092/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ee8e00aa554c0f2343d51a789814bf2ece1176cad30143f923f623b1616feea8
+size 484464496

checkpoints/cat_qwen25_7b_r64_a32_plain_priorrecipe_e2_lr1e-4_s1/checkpoint-2092/added_tokens.json

@@ -0,0 +1,24 @@
+{
+  "</tool_call>": 151658,
+  "<tool_call>": 151657,
+  "<|box_end|>": 151649,
+  "<|box_start|>": 151648,
+  "<|endoftext|>": 151643,
+  "<|file_sep|>": 151664,
+  "<|fim_middle|>": 151660,
+  "<|fim_pad|>": 151662,
+  "<|fim_prefix|>": 151659,
+  "<|fim_suffix|>": 151661,
+  "<|im_end|>": 151645,
+  "<|im_start|>": 151644,
+  "<|image_pad|>": 151655,
+  "<|object_ref_end|>": 151647,
+  "<|object_ref_start|>": 151646,
+  "<|quad_end|>": 151651,
+  "<|quad_start|>": 151650,
+  "<|repo_name|>": 151663,
+  "<|video_pad|>": 151656,
+  "<|vision_end|>": 151653,
+  "<|vision_pad|>": 151654,
+  "<|vision_start|>": 151652
+}

checkpoints/cat_qwen25_7b_r64_a32_plain_priorrecipe_e2_lr1e-4_s1/checkpoint-2092/chat_template.jinja

@@ -0,0 +1,54 @@
+{%- if tools %}
+    {{- '<|im_start|>system\n' }}
+    {%- if messages[0]['role'] == 'system' %}
+        {{- messages[0]['content'] }}
+    {%- else %}
+        {{- 'You are Qwen, created by Alibaba Cloud. You are a helpful assistant.' }}
+    {%- endif %}
+    {{- "\n\n# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
+    {%- for tool in tools %}
+        {{- "\n" }}
+        {{- tool | tojson }}
+    {%- endfor %}
+    {{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
+{%- else %}
+    {%- if messages[0]['role'] == 'system' %}
+        {{- '<|im_start|>system\n' + messages[0]['content'] + '<|im_end|>\n' }}
+    {%- else %}
+        {{- '<|im_start|>system\nYou are Qwen, created by Alibaba Cloud. You are a helpful assistant.<|im_end|>\n' }}
+    {%- endif %}
+{%- endif %}
+{%- for message in messages %}
+    {%- if (message.role == "user") or (message.role == "system" and not loop.first) or (message.role == "assistant" and not message.tool_calls) %}
+        {{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }}
+    {%- elif message.role == "assistant" %}
+        {{- '<|im_start|>' + message.role }}
+        {%- if message.content %}
+            {{- '\n' + message.content }}
+        {%- endif %}
+        {%- for tool_call in message.tool_calls %}
+            {%- if tool_call.function is defined %}
+                {%- set tool_call = tool_call.function %}
+            {%- endif %}
+            {{- '\n<tool_call>\n{"name": "' }}
+            {{- tool_call.name }}
+            {{- '", "arguments": ' }}
+            {{- tool_call.arguments | tojson }}
+            {{- '}\n</tool_call>' }}
+        {%- endfor %}
+        {{- '<|im_end|>\n' }}
+    {%- elif message.role == "tool" %}
+        {%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != "tool") %}
+            {{- '<|im_start|>user' }}
+        {%- endif %}
+        {{- '\n<tool_response>\n' }}
+        {{- message.content }}
+        {{- '\n</tool_response>' }}
+        {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
+            {{- '<|im_end|>\n' }}
+        {%- endif %}
+    {%- endif %}
+{%- endfor %}
+{%- if add_generation_prompt %}
+    {{- '<|im_start|>assistant\n' }}
+{%- endif %}

checkpoints/cat_qwen25_7b_r64_a32_plain_priorrecipe_e2_lr1e-4_s1/checkpoint-2092/special_tokens_map.json

@@ -0,0 +1,31 @@
+{
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "eos_token": {
+    "content": "<|im_end|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

checkpoints/cat_qwen25_7b_r64_a32_plain_priorrecipe_e2_lr1e-4_s1/checkpoint-2092/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9c5ae00e602b8860cbd784ba82a8aa14e8feecec692e7076590d014d7b7fdafa
+size 11421896

checkpoints/cat_qwen25_7b_r64_a32_plain_priorrecipe_e2_lr1e-4_s1/checkpoint-2092/tokenizer_config.json

@@ -0,0 +1,207 @@
+{
+  "add_bos_token": false,
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "151643": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151644": {
+      "content": "<|im_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151645": {
+      "content": "<|im_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151646": {
+      "content": "<|object_ref_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151647": {
+      "content": "<|object_ref_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151648": {
+      "content": "<|box_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151649": {
+      "content": "<|box_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151650": {
+      "content": "<|quad_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151651": {
+      "content": "<|quad_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151652": {
+      "content": "<|vision_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151653": {
+      "content": "<|vision_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151654": {
+      "content": "<|vision_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151655": {
+      "content": "<|image_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151656": {
+      "content": "<|video_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151657": {
+      "content": "<tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151658": {
+      "content": "</tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151659": {
+      "content": "<|fim_prefix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151660": {
+      "content": "<|fim_middle|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151661": {
+      "content": "<|fim_suffix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151662": {
+      "content": "<|fim_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151663": {
+      "content": "<|repo_name|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151664": {
+      "content": "<|file_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "bos_token": null,
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "extra_special_tokens": {},
+  "model_max_length": 131072,
+  "pad_token": "<|endoftext|>",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

checkpoints/cat_qwen25_7b_r64_a32_plain_priorrecipe_e2_lr1e-4_s1/checkpoint-2092/trainer_state.json

@@ -0,0 +1,2124 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.0,
+  "eval_steps": 500,
+  "global_step": 2092,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "entropy": 1.1739739060401917,
+      "epoch": 0.009560229445506692,
+      "grad_norm": 0.48447680473327637,
+      "learning_rate": 8.571428571428573e-06,
+      "loss": 0.2779,
+      "mean_token_accuracy": 0.9112883508205414,
+      "num_tokens": 14125.0,
+      "step": 10
+    },
+    {
+      "entropy": 1.1922486662864684,
+      "epoch": 0.019120458891013385,
+      "grad_norm": 0.5766411423683167,
+      "learning_rate": 1.8095238095238094e-05,
+      "loss": 0.2636,
+      "mean_token_accuracy": 0.9090810775756836,
+      "num_tokens": 27004.0,
+      "step": 20
+    },
+    {
+      "entropy": 1.1917970299720764,
+      "epoch": 0.028680688336520075,
+      "grad_norm": 0.5361194014549255,
+      "learning_rate": 2.7619047619047622e-05,
+      "loss": 0.227,
+      "mean_token_accuracy": 0.9198892414569855,
+      "num_tokens": 41395.0,
+      "step": 30
+    },
+    {
+      "entropy": 1.1661703705787658,
+      "epoch": 0.03824091778202677,
+      "grad_norm": 0.5805957913398743,
+      "learning_rate": 3.7142857142857143e-05,
+      "loss": 0.1983,
+      "mean_token_accuracy": 0.9330302536487579,
+      "num_tokens": 55233.0,
+      "step": 40
+    },
+    {
+      "entropy": 1.1459822177886962,
+      "epoch": 0.04780114722753346,
+      "grad_norm": 0.8687187433242798,
+      "learning_rate": 4.666666666666667e-05,
+      "loss": 0.2039,
+      "mean_token_accuracy": 0.9339372217655182,
+      "num_tokens": 68341.0,
+      "step": 50
+    },
+    {
+      "entropy": 1.1209450244903565,
+      "epoch": 0.05736137667304015,
+      "grad_norm": 0.5746432542800903,
+      "learning_rate": 5.619047619047619e-05,
+      "loss": 0.1761,
+      "mean_token_accuracy": 0.9366338729858399,
+      "num_tokens": 81730.0,
+      "step": 60
+    },
+    {
+      "entropy": 1.1023436903953552,
+      "epoch": 0.06692160611854685,
+      "grad_norm": 0.6911821365356445,
+      "learning_rate": 6.571428571428571e-05,
+      "loss": 0.1643,
+      "mean_token_accuracy": 0.9395260095596314,
+      "num_tokens": 94886.0,
+      "step": 70
+    },
+    {
+      "entropy": 1.0868061780929565,
+      "epoch": 0.07648183556405354,
+      "grad_norm": 0.7381938099861145,
+      "learning_rate": 7.523809523809524e-05,
+      "loss": 0.1755,
+      "mean_token_accuracy": 0.9379717230796814,
+      "num_tokens": 108569.0,
+      "step": 80
+    },
+    {
+      "entropy": 1.082469379901886,
+      "epoch": 0.08604206500956023,
+      "grad_norm": 0.5891942381858826,
+      "learning_rate": 8.476190476190477e-05,
+      "loss": 0.1587,
+      "mean_token_accuracy": 0.9442471981048584,
+      "num_tokens": 122156.0,
+      "step": 90
+    },
+    {
+      "entropy": 1.0694019556045533,
+      "epoch": 0.09560229445506692,
+      "grad_norm": 0.6982259750366211,
+      "learning_rate": 9.428571428571429e-05,
+      "loss": 0.1587,
+      "mean_token_accuracy": 0.9375297367572785,
+      "num_tokens": 135290.0,
+      "step": 100
+    },
+    {
+      "entropy": 1.0603623032569884,
+      "epoch": 0.10516252390057361,
+      "grad_norm": 0.595213770866394,
+      "learning_rate": 9.999900008621665e-05,
+      "loss": 0.1755,
+      "mean_token_accuracy": 0.9368651568889618,
+      "num_tokens": 149015.0,
+      "step": 110
+    },
+    {
+      "entropy": 1.0578490138053893,
+      "epoch": 0.1147227533460803,
+      "grad_norm": 0.5523142218589783,
+      "learning_rate": 9.998775151544478e-05,
+      "loss": 0.152,
+      "mean_token_accuracy": 0.9413406252861023,
+      "num_tokens": 162147.0,
+      "step": 120
+    },
+    {
+      "entropy": 1.0339872002601624,
+      "epoch": 0.124282982791587,
+      "grad_norm": 0.5127677321434021,
+      "learning_rate": 9.996400730289628e-05,
+      "loss": 0.1545,
+      "mean_token_accuracy": 0.9423599481582642,
+      "num_tokens": 175436.0,
+      "step": 130
+    },
+    {
+      "entropy": 1.0214387774467468,
+      "epoch": 0.1338432122370937,
+      "grad_norm": 0.7661780714988708,
+      "learning_rate": 9.992777338400864e-05,
+      "loss": 0.1507,
+      "mean_token_accuracy": 0.9482471108436584,
+      "num_tokens": 188834.0,
+      "step": 140
+    },
+    {
+      "entropy": 1.0077783823013307,
+      "epoch": 0.14340344168260039,
+      "grad_norm": 0.5621943473815918,
+      "learning_rate": 9.987905881632213e-05,
+      "loss": 0.146,
+      "mean_token_accuracy": 0.9470974624156951,
+      "num_tokens": 202708.0,
+      "step": 150
+    },
+    {
+      "entropy": 1.0168480396270752,
+      "epoch": 0.15296367112810708,
+      "grad_norm": 0.6676165461540222,
+      "learning_rate": 9.981787577721553e-05,
+      "loss": 0.1506,
+      "mean_token_accuracy": 0.9439922094345092,
+      "num_tokens": 216384.0,
+      "step": 160
+    },
+    {
+      "entropy": 1.0258822679519652,
+      "epoch": 0.16252390057361377,
+      "grad_norm": 0.5179085731506348,
+      "learning_rate": 9.974423956086228e-05,
+      "loss": 0.1427,
+      "mean_token_accuracy": 0.9465557217597962,
+      "num_tokens": 229105.0,
+      "step": 170
+    },
+    {
+      "entropy": 1.0219564616680146,
+      "epoch": 0.17208413001912046,
+      "grad_norm": 0.5938138365745544,
+      "learning_rate": 9.965816857440717e-05,
+      "loss": 0.131,
+      "mean_token_accuracy": 0.9459258615970612,
+      "num_tokens": 242264.0,
+      "step": 180
+    },
+    {
+      "entropy": 1.0320287644863129,
+      "epoch": 0.18164435946462715,
+      "grad_norm": 0.573223888874054,
+      "learning_rate": 9.955968433336515e-05,
+      "loss": 0.1528,
+      "mean_token_accuracy": 0.9427159130573273,
+      "num_tokens": 255602.0,
+      "step": 190
+    },
+    {
+      "entropy": 1.0252255618572235,
+      "epoch": 0.19120458891013384,
+      "grad_norm": 0.4776560962200165,
+      "learning_rate": 9.944881145624291e-05,
+      "loss": 0.112,
+      "mean_token_accuracy": 0.9548257052898407,
+      "num_tokens": 269712.0,
+      "step": 200
+    },
+    {
+      "entropy": 1.03580904006958,
+      "epoch": 0.20076481835564053,
+      "grad_norm": 0.6060590744018555,
+      "learning_rate": 9.932557765838504e-05,
+      "loss": 0.1297,
+      "mean_token_accuracy": 0.9494471728801728,
+      "num_tokens": 283572.0,
+      "step": 210
+    },
+    {
+      "entropy": 1.0138694167137146,
+      "epoch": 0.21032504780114722,
+      "grad_norm": 0.5444765090942383,
+      "learning_rate": 9.919001374504569e-05,
+      "loss": 0.1339,
+      "mean_token_accuracy": 0.9512911856174469,
+      "num_tokens": 297437.0,
+      "step": 220
+    },
+    {
+      "entropy": 1.0079212605953216,
+      "epoch": 0.2198852772466539,
+      "grad_norm": 0.7607695460319519,
+      "learning_rate": 9.904215360368826e-05,
+      "loss": 0.1333,
+      "mean_token_accuracy": 0.952767676115036,
+      "num_tokens": 310803.0,
+      "step": 230
+    },
+    {
+      "entropy": 1.02037872672081,
+      "epoch": 0.2294455066921606,
+      "grad_norm": 0.7083406448364258,
+      "learning_rate": 9.888203419551432e-05,
+      "loss": 0.1335,
+      "mean_token_accuracy": 0.945678448677063,
+      "num_tokens": 324713.0,
+      "step": 240
+    },
+    {
+      "entropy": 0.9861022055149078,
+      "epoch": 0.2390057361376673,
+      "grad_norm": 1.4376698732376099,
+      "learning_rate": 9.870969554622426e-05,
+      "loss": 0.1296,
+      "mean_token_accuracy": 0.9462655186653137,
+      "num_tokens": 337634.0,
+      "step": 250
+    },
+    {
+      "entropy": 0.9949163436889649,
+      "epoch": 0.248565965583174,
+      "grad_norm": 0.5563827753067017,
+      "learning_rate": 9.852518073601194e-05,
+      "loss": 0.1371,
+      "mean_token_accuracy": 0.9474994301795959,
+      "num_tokens": 350459.0,
+      "step": 260
+    },
+    {
+      "entropy": 0.9755370080471039,
+      "epoch": 0.25812619502868067,
+      "grad_norm": 0.7883791327476501,
+      "learning_rate": 9.832853588879575e-05,
+      "loss": 0.1195,
+      "mean_token_accuracy": 0.9544442772865296,
+      "num_tokens": 364010.0,
+      "step": 270
+    },
+    {
+      "entropy": 0.9872765660285949,
+      "epoch": 0.2676864244741874,
+      "grad_norm": 0.6124987006187439,
+      "learning_rate": 9.811981016068881e-05,
+      "loss": 0.1266,
+      "mean_token_accuracy": 0.948451817035675,
+      "num_tokens": 377021.0,
+      "step": 280
+    },
+    {
+      "entropy": 0.9835463523864746,
+      "epoch": 0.27724665391969405,
+      "grad_norm": 0.4377821087837219,
+      "learning_rate": 9.78990557277112e-05,
+      "loss": 0.1323,
+      "mean_token_accuracy": 0.9493040919303894,
+      "num_tokens": 390819.0,
+      "step": 290
+    },
+    {
+      "entropy": 1.0000784754753114,
+      "epoch": 0.28680688336520077,
+      "grad_norm": 0.721735417842865,
+      "learning_rate": 9.766632777274738e-05,
+      "loss": 0.118,
+      "mean_token_accuracy": 0.9553472459316253,
+      "num_tokens": 404644.0,
+      "step": 300
+    },
+    {
+      "entropy": 1.0221971035003663,
+      "epoch": 0.29636711281070743,
+      "grad_norm": 0.7239003777503967,
+      "learning_rate": 9.742168447175182e-05,
+      "loss": 0.1306,
+      "mean_token_accuracy": 0.9522208452224732,
+      "num_tokens": 418529.0,
+      "step": 310
+    },
+    {
+      "entropy": 0.9920777082443237,
+      "epoch": 0.30592734225621415,
+      "grad_norm": 0.5045039653778076,
+      "learning_rate": 9.716518697920655e-05,
+      "loss": 0.1127,
+      "mean_token_accuracy": 0.9572355508804321,
+      "num_tokens": 431509.0,
+      "step": 320
+    },
+    {
+      "entropy": 1.0165483057498932,
+      "epoch": 0.3154875717017208,
+      "grad_norm": 0.483862966299057,
+      "learning_rate": 9.689689941283417e-05,
+      "loss": 0.1244,
+      "mean_token_accuracy": 0.9529255330562592,
+      "num_tokens": 445040.0,
+      "step": 330
+    },
+    {
+      "entropy": 1.0201621532440186,
+      "epoch": 0.32504780114722753,
+      "grad_norm": 0.6959221959114075,
+      "learning_rate": 9.661688883756996e-05,
+      "loss": 0.128,
+      "mean_token_accuracy": 0.9512975215911865,
+      "num_tokens": 458540.0,
+      "step": 340
+    },
+    {
+      "entropy": 1.0207452952861786,
+      "epoch": 0.33460803059273425,
+      "grad_norm": 0.625383734703064,
+      "learning_rate": 9.632522524879756e-05,
+      "loss": 0.1297,
+      "mean_token_accuracy": 0.9507432043552398,
+      "num_tokens": 471556.0,
+      "step": 350
+    },
+    {
+      "entropy": 1.0345476627349854,
+      "epoch": 0.3441682600382409,
+      "grad_norm": 0.5909538865089417,
+      "learning_rate": 9.602198155485176e-05,
+      "loss": 0.1182,
+      "mean_token_accuracy": 0.9544213294982911,
+      "num_tokens": 484786.0,
+      "step": 360
+    },
+    {
+      "entropy": 1.04675452709198,
+      "epoch": 0.35372848948374763,
+      "grad_norm": 0.508851945400238,
+      "learning_rate": 9.570723355879342e-05,
+      "loss": 0.1158,
+      "mean_token_accuracy": 0.9560792922973633,
+      "num_tokens": 498352.0,
+      "step": 370
+    },
+    {
+      "entropy": 1.0251278281211853,
+      "epoch": 0.3632887189292543,
+      "grad_norm": 0.6463770270347595,
+      "learning_rate": 9.538105993946068e-05,
+      "loss": 0.114,
+      "mean_token_accuracy": 0.9578813314437866,
+      "num_tokens": 512064.0,
+      "step": 380
+    },
+    {
+      "entropy": 1.0322481632232665,
+      "epoch": 0.372848948374761,
+      "grad_norm": 0.4552994668483734,
+      "learning_rate": 9.504354223180118e-05,
+      "loss": 0.1089,
+      "mean_token_accuracy": 0.9574484050273895,
+      "num_tokens": 525582.0,
+      "step": 390
+    },
+    {
+      "entropy": 1.0358243703842163,
+      "epoch": 0.3824091778202677,
+      "grad_norm": 0.8684021234512329,
+      "learning_rate": 9.469476480649056e-05,
+      "loss": 0.1205,
+      "mean_token_accuracy": 0.953494793176651,
+      "num_tokens": 539561.0,
+      "step": 400
+    },
+    {
+      "entropy": 1.0188953280448914,
+      "epoch": 0.3919694072657744,
+      "grad_norm": 0.5271518230438232,
+      "learning_rate": 9.433481484884189e-05,
+      "loss": 0.1152,
+      "mean_token_accuracy": 0.9554457068443298,
+      "num_tokens": 553625.0,
+      "step": 410
+    },
+    {
+      "entropy": 0.9941672265529633,
+      "epoch": 0.40152963671128106,
+      "grad_norm": 0.6121000051498413,
+      "learning_rate": 9.396378233701167e-05,
+      "loss": 0.1022,
+      "mean_token_accuracy": 0.9587876379489899,
+      "num_tokens": 565891.0,
+      "step": 420
+    },
+    {
+      "entropy": 1.0032862544059753,
+      "epoch": 0.4110898661567878,
+      "grad_norm": 0.6064424514770508,
+      "learning_rate": 9.358176001950759e-05,
+      "loss": 0.1105,
+      "mean_token_accuracy": 0.9586677372455596,
+      "num_tokens": 578722.0,
+      "step": 430
+    },
+    {
+      "entropy": 1.0347782731056214,
+      "epoch": 0.42065009560229444,
+      "grad_norm": 0.6950361728668213,
+      "learning_rate": 9.318884339200378e-05,
+      "loss": 0.1309,
+      "mean_token_accuracy": 0.9468400537967682,
+      "num_tokens": 592646.0,
+      "step": 440
+    },
+    {
+      "entropy": 1.0031266927719116,
+      "epoch": 0.43021032504780116,
+      "grad_norm": 0.663911759853363,
+      "learning_rate": 9.278513067346941e-05,
+      "loss": 0.1065,
+      "mean_token_accuracy": 0.9599073350429534,
+      "num_tokens": 606429.0,
+      "step": 450
+    },
+    {
+      "entropy": 1.0111085236072541,
+      "epoch": 0.4397705544933078,
+      "grad_norm": 0.6929073929786682,
+      "learning_rate": 9.237072278161652e-05,
+      "loss": 0.1225,
+      "mean_token_accuracy": 0.9522352516651154,
+      "num_tokens": 619626.0,
+      "step": 460
+    },
+    {
+      "entropy": 1.0111278951168061,
+      "epoch": 0.44933078393881454,
+      "grad_norm": 0.812089204788208,
+      "learning_rate": 9.194572330767314e-05,
+      "loss": 0.1135,
+      "mean_token_accuracy": 0.9553511738777161,
+      "num_tokens": 633055.0,
+      "step": 470
+    },
+    {
+      "entropy": 1.0062795579433441,
+      "epoch": 0.4588910133843212,
+      "grad_norm": 0.5233294367790222,
+      "learning_rate": 9.151023849048818e-05,
+      "loss": 0.1079,
+      "mean_token_accuracy": 0.9546993792057037,
+      "num_tokens": 647155.0,
+      "step": 480
+    },
+    {
+      "entropy": 0.978465473651886,
+      "epoch": 0.4684512428298279,
+      "grad_norm": 0.5739076733589172,
+      "learning_rate": 9.106437718997452e-05,
+      "loss": 0.1095,
+      "mean_token_accuracy": 0.9602778613567352,
+      "num_tokens": 660695.0,
+      "step": 490
+    },
+    {
+      "entropy": 1.0046596467494964,
+      "epoch": 0.4780114722753346,
+      "grad_norm": 0.8816806674003601,
+      "learning_rate": 9.060825085989682e-05,
+      "loss": 0.1139,
+      "mean_token_accuracy": 0.9528433561325074,
+      "num_tokens": 674509.0,
+      "step": 500
+    },
+    {
+      "entropy": 1.0130971670150757,
+      "epoch": 0.4875717017208413,
+      "grad_norm": 0.6996803879737854,
+      "learning_rate": 9.014197352001097e-05,
+      "loss": 0.0972,
+      "mean_token_accuracy": 0.9636221110820771,
+      "num_tokens": 687644.0,
+      "step": 510
+    },
+    {
+      "entropy": 0.9867890119552613,
+      "epoch": 0.497131931166348,
+      "grad_norm": 0.8523248434066772,
+      "learning_rate": 8.966566172756213e-05,
+      "loss": 0.1114,
+      "mean_token_accuracy": 0.9551803469657898,
+      "num_tokens": 700658.0,
+      "step": 520
+    },
+    {
+      "entropy": 1.0174102306365966,
+      "epoch": 0.5066921606118547,
+      "grad_norm": 0.6316176056861877,
+      "learning_rate": 8.91794345481484e-05,
+      "loss": 0.1132,
+      "mean_token_accuracy": 0.9557361006736755,
+      "num_tokens": 714234.0,
+      "step": 530
+    },
+    {
+      "entropy": 1.0137089550495149,
+      "epoch": 0.5162523900573613,
+      "grad_norm": 0.8284169435501099,
+      "learning_rate": 8.868341352595755e-05,
+      "loss": 0.132,
+      "mean_token_accuracy": 0.9474952280521393,
+      "num_tokens": 727721.0,
+      "step": 540
+    },
+    {
+      "entropy": 0.9896256387233734,
+      "epoch": 0.5258126195028681,
+      "grad_norm": 0.4716854691505432,
+      "learning_rate": 8.817772265338409e-05,
+      "loss": 0.1174,
+      "mean_token_accuracy": 0.9556844532489777,
+      "num_tokens": 741707.0,
+      "step": 550
+    },
+    {
+      "entropy": 1.0029700338840484,
+      "epoch": 0.5353728489483748,
+      "grad_norm": 0.612719714641571,
+      "learning_rate": 8.766248834003441e-05,
+      "loss": 0.124,
+      "mean_token_accuracy": 0.9510359942913056,
+      "num_tokens": 755385.0,
+      "step": 560
+    },
+    {
+      "entropy": 0.9875121295452118,
+      "epoch": 0.5449330783938815,
+      "grad_norm": 0.8032057881355286,
+      "learning_rate": 8.713783938112768e-05,
+      "loss": 0.0999,
+      "mean_token_accuracy": 0.9627587556838989,
+      "num_tokens": 768687.0,
+      "step": 570
+    },
+    {
+      "entropy": 1.0095878303050996,
+      "epoch": 0.5544933078393881,
+      "grad_norm": 0.6357651352882385,
+      "learning_rate": 8.660390692530026e-05,
+      "loss": 0.0982,
+      "mean_token_accuracy": 0.9629545986652375,
+      "num_tokens": 782172.0,
+      "step": 580
+    },
+    {
+      "entropy": 0.9758632063865662,
+      "epoch": 0.5640535372848948,
+      "grad_norm": 0.4679427146911621,
+      "learning_rate": 8.606082444182216e-05,
+      "loss": 0.1015,
+      "mean_token_accuracy": 0.9637291312217713,
+      "num_tokens": 795561.0,
+      "step": 590
+    },
+    {
+      "entropy": 0.9987418293952942,
+      "epoch": 0.5736137667304015,
+      "grad_norm": 0.39159005880355835,
+      "learning_rate": 8.550872768723292e-05,
+      "loss": 0.0911,
+      "mean_token_accuracy": 0.9652044415473938,
+      "num_tokens": 809436.0,
+      "step": 600
+    },
+    {
+      "entropy": 0.9824410080909729,
+      "epoch": 0.5831739961759083,
+      "grad_norm": 0.7355128526687622,
+      "learning_rate": 8.494775467140627e-05,
+      "loss": 0.1027,
+      "mean_token_accuracy": 0.9613589644432068,
+      "num_tokens": 822983.0,
+      "step": 610
+    },
+    {
+      "entropy": 0.9905080258846283,
+      "epoch": 0.5927342256214149,
+      "grad_norm": 0.5935409069061279,
+      "learning_rate": 8.4378045623051e-05,
+      "loss": 0.1026,
+      "mean_token_accuracy": 0.9600830614566803,
+      "num_tokens": 836766.0,
+      "step": 620
+    },
+    {
+      "entropy": 0.9966179192066192,
+      "epoch": 0.6022944550669216,
+      "grad_norm": 0.37031838297843933,
+      "learning_rate": 8.379974295465758e-05,
+      "loss": 0.1072,
+      "mean_token_accuracy": 0.9558251917362213,
+      "num_tokens": 850200.0,
+      "step": 630
+    },
+    {
+      "entropy": 0.995290344953537,
+      "epoch": 0.6118546845124283,
+      "grad_norm": 0.3872464895248413,
+      "learning_rate": 8.321299122689863e-05,
+      "loss": 0.1228,
+      "mean_token_accuracy": 0.9508748948574066,
+      "num_tokens": 864001.0,
+      "step": 640
+    },
+    {
+      "entropy": 1.0032495439052582,
+      "epoch": 0.621414913957935,
+      "grad_norm": 0.5912079215049744,
+      "learning_rate": 8.261793711249254e-05,
+      "loss": 0.1066,
+      "mean_token_accuracy": 0.9552255213260651,
+      "num_tokens": 878135.0,
+      "step": 650
+    },
+    {
+      "entropy": 0.9742148697376252,
+      "epoch": 0.6309751434034416,
+      "grad_norm": 0.6092025637626648,
+      "learning_rate": 8.201472935953916e-05,
+      "loss": 0.1187,
+      "mean_token_accuracy": 0.9528553247451782,
+      "num_tokens": 891307.0,
+      "step": 660
+    },
+    {
+      "entropy": 1.004436594247818,
+      "epoch": 0.6405353728489483,
+      "grad_norm": 0.4784019887447357,
+      "learning_rate": 8.140351875433649e-05,
+      "loss": 0.1051,
+      "mean_token_accuracy": 0.9576326787471772,
+      "num_tokens": 904829.0,
+      "step": 670
+    },
+    {
+      "entropy": 0.9817103981971741,
+      "epoch": 0.6500956022944551,
+      "grad_norm": 0.5307291746139526,
+      "learning_rate": 8.07844580836881e-05,
+      "loss": 0.0926,
+      "mean_token_accuracy": 0.9635098695755004,
+      "num_tokens": 919199.0,
+      "step": 680
+    },
+    {
+      "entropy": 0.9879955470561981,
+      "epoch": 0.6596558317399618,
+      "grad_norm": 0.49793291091918945,
+      "learning_rate": 8.015770209671044e-05,
+      "loss": 0.1126,
+      "mean_token_accuracy": 0.9594523072242737,
+      "num_tokens": 932439.0,
+      "step": 690
+    },
+    {
+      "entropy": 1.0025336384773254,
+      "epoch": 0.6692160611854685,
+      "grad_norm": 0.667904257774353,
+      "learning_rate": 7.952340746614946e-05,
+      "loss": 0.1116,
+      "mean_token_accuracy": 0.9543653309345246,
+      "num_tokens": 945776.0,
+      "step": 700
+    },
+    {
+      "entropy": 0.9931329905986785,
+      "epoch": 0.6787762906309751,
+      "grad_norm": 0.6819143295288086,
+      "learning_rate": 7.888173274921655e-05,
+      "loss": 0.0997,
+      "mean_token_accuracy": 0.9626343011856079,
+      "num_tokens": 959352.0,
+      "step": 710
+    },
+    {
+      "entropy": 0.9990474820137024,
+      "epoch": 0.6883365200764818,
+      "grad_norm": 0.6067373156547546,
+      "learning_rate": 7.823283834795336e-05,
+      "loss": 0.1078,
+      "mean_token_accuracy": 0.9553721189498902,
+      "num_tokens": 972596.0,
+      "step": 720
+    },
+    {
+      "entropy": 0.9917460799217224,
+      "epoch": 0.6978967495219885,
+      "grad_norm": 0.6884276270866394,
+      "learning_rate": 7.75768864691354e-05,
+      "loss": 0.1064,
+      "mean_token_accuracy": 0.9580906927585602,
+      "num_tokens": 985699.0,
+      "step": 730
+    },
+    {
+      "entropy": 0.9913410425186158,
+      "epoch": 0.7074569789674953,
+      "grad_norm": 0.6106778383255005,
+      "learning_rate": 7.691404108372461e-05,
+      "loss": 0.1099,
+      "mean_token_accuracy": 0.9571886181831359,
+      "num_tokens": 999728.0,
+      "step": 740
+    },
+    {
+      "entropy": 0.9946191668510437,
+      "epoch": 0.7170172084130019,
+      "grad_norm": 0.4457184970378876,
+      "learning_rate": 7.624446788588085e-05,
+      "loss": 0.0919,
+      "mean_token_accuracy": 0.9650080800056458,
+      "num_tokens": 1013669.0,
+      "step": 750
+    },
+    {
+      "entropy": 0.9851050615310669,
+      "epoch": 0.7265774378585086,
+      "grad_norm": 0.47441422939300537,
+      "learning_rate": 7.556833425154255e-05,
+      "loss": 0.0948,
+      "mean_token_accuracy": 0.962857848405838,
+      "num_tokens": 1027609.0,
+      "step": 760
+    },
+    {
+      "entropy": 0.9897657215595246,
+      "epoch": 0.7361376673040153,
+      "grad_norm": 0.8434217572212219,
+      "learning_rate": 7.488580919658727e-05,
+      "loss": 0.1062,
+      "mean_token_accuracy": 0.9575309813022613,
+      "num_tokens": 1041046.0,
+      "step": 770
+    },
+    {
+      "entropy": 0.9739744007587433,
+      "epoch": 0.745697896749522,
+      "grad_norm": 0.7281696796417236,
+      "learning_rate": 7.419706333458201e-05,
+      "loss": 0.0923,
+      "mean_token_accuracy": 0.9619448125362396,
+      "num_tokens": 1055304.0,
+      "step": 780
+    },
+    {
+      "entropy": 0.9758931040763855,
+      "epoch": 0.7552581261950286,
+      "grad_norm": 0.7574643492698669,
+      "learning_rate": 7.350226883413413e-05,
+      "loss": 0.105,
+      "mean_token_accuracy": 0.9613841354846955,
+      "num_tokens": 1068542.0,
+      "step": 790
+    },
+    {
+      "entropy": 0.9839439630508423,
+      "epoch": 0.7648183556405354,
+      "grad_norm": 0.7352367639541626,
+      "learning_rate": 7.280159937585386e-05,
+      "loss": 0.1004,
+      "mean_token_accuracy": 0.9618932247161865,
+      "num_tokens": 1082195.0,
+      "step": 800
+    },
+    {
+      "entropy": 0.9902204394340515,
+      "epoch": 0.7743785850860421,
+      "grad_norm": 0.8046219944953918,
+      "learning_rate": 7.209523010893853e-05,
+      "loss": 0.0933,
+      "mean_token_accuracy": 0.9618531584739685,
+      "num_tokens": 1095884.0,
+      "step": 810
+    },
+    {
+      "entropy": 0.9911208152770996,
+      "epoch": 0.7839388145315488,
+      "grad_norm": 0.5386976003646851,
+      "learning_rate": 7.138333760738987e-05,
+      "loss": 0.1048,
+      "mean_token_accuracy": 0.9595375657081604,
+      "num_tokens": 1109287.0,
+      "step": 820
+    },
+    {
+      "entropy": 0.9835298836231232,
+      "epoch": 0.7934990439770554,
+      "grad_norm": 0.813587486743927,
+      "learning_rate": 7.066609982587501e-05,
+      "loss": 0.0955,
+      "mean_token_accuracy": 0.9636480391025544,
+      "num_tokens": 1123392.0,
+      "step": 830
+    },
+    {
+      "entropy": 1.0016310214996338,
+      "epoch": 0.8030592734225621,
+      "grad_norm": 0.47587668895721436,
+      "learning_rate": 6.994369605524253e-05,
+      "loss": 0.0995,
+      "mean_token_accuracy": 0.9620899677276611,
+      "num_tokens": 1137691.0,
+      "step": 840
+    },
+    {
+      "entropy": 0.9716361939907074,
+      "epoch": 0.8126195028680688,
+      "grad_norm": 0.7446271181106567,
+      "learning_rate": 6.921630687770427e-05,
+      "loss": 0.1027,
+      "mean_token_accuracy": 0.9587779760360717,
+      "num_tokens": 1151400.0,
+      "step": 850
+    },
+    {
+      "entropy": 0.9731779277324677,
+      "epoch": 0.8221797323135756,
+      "grad_norm": 0.787928581237793,
+      "learning_rate": 6.848411412169461e-05,
+      "loss": 0.0926,
+      "mean_token_accuracy": 0.9653818368911743,
+      "num_tokens": 1165074.0,
+      "step": 860
+    },
+    {
+      "entropy": 0.9709990799427033,
+      "epoch": 0.8317399617590823,
+      "grad_norm": 0.747322678565979,
+      "learning_rate": 6.774730081641799e-05,
+      "loss": 0.0976,
+      "mean_token_accuracy": 0.959471195936203,
+      "num_tokens": 1178919.0,
+      "step": 870
+    },
+    {
+      "entropy": 0.9760962128639221,
+      "epoch": 0.8413001912045889,
+      "grad_norm": 0.903546154499054,
+      "learning_rate": 6.70060511460963e-05,
+      "loss": 0.1036,
+      "mean_token_accuracy": 0.9576647162437439,
+      "num_tokens": 1192204.0,
+      "step": 880
+    },
+    {
+      "entropy": 0.9572584986686706,
+      "epoch": 0.8508604206500956,
+      "grad_norm": 0.450843870639801,
+      "learning_rate": 6.626055040392774e-05,
+      "loss": 0.0918,
+      "mean_token_accuracy": 0.9643404960632325,
+      "num_tokens": 1205571.0,
+      "step": 890
+    },
+    {
+      "entropy": 0.9604834616184235,
+      "epoch": 0.8604206500956023,
+      "grad_norm": 0.9252690076828003,
+      "learning_rate": 6.551098494576819e-05,
+      "loss": 0.1166,
+      "mean_token_accuracy": 0.9555048942565918,
+      "num_tokens": 1219043.0,
+      "step": 900
+    },
+    {
+      "entropy": 0.9746643722057342,
+      "epoch": 0.869980879541109,
+      "grad_norm": 0.5353615880012512,
+      "learning_rate": 6.475754214354715e-05,
+      "loss": 0.1076,
+      "mean_token_accuracy": 0.9589978337287903,
+      "num_tokens": 1233113.0,
+      "step": 910
+    },
+    {
+      "entropy": 0.9580503106117249,
+      "epoch": 0.8795411089866156,
+      "grad_norm": 0.5606767535209656,
+      "learning_rate": 6.400041033842965e-05,
+      "loss": 0.0923,
+      "mean_token_accuracy": 0.9625468909740448,
+      "num_tokens": 1245931.0,
+      "step": 920
+    },
+    {
+      "entropy": 0.9474713385105134,
+      "epoch": 0.8891013384321224,
+      "grad_norm": 0.6566503047943115,
+      "learning_rate": 6.323977879373571e-05,
+      "loss": 0.0998,
+      "mean_token_accuracy": 0.9610148727893829,
+      "num_tokens": 1259949.0,
+      "step": 930
+    },
+    {
+      "entropy": 0.9734407842159272,
+      "epoch": 0.8986615678776291,
+      "grad_norm": 0.6584394574165344,
+      "learning_rate": 6.247583764762959e-05,
+      "loss": 0.0932,
+      "mean_token_accuracy": 0.9613297402858734,
+      "num_tokens": 1273413.0,
+      "step": 940
+    },
+    {
+      "entropy": 0.9682157933712006,
+      "epoch": 0.9082217973231358,
+      "grad_norm": 0.586574375629425,
+      "learning_rate": 6.170877786559001e-05,
+      "loss": 0.0969,
+      "mean_token_accuracy": 0.9622386813163757,
+      "num_tokens": 1286343.0,
+      "step": 950
+    },
+    {
+      "entropy": 0.956113189458847,
+      "epoch": 0.9177820267686424,
+      "grad_norm": 0.6844049096107483,
+      "learning_rate": 6.0938791192673895e-05,
+      "loss": 0.1066,
+      "mean_token_accuracy": 0.957487165927887,
+      "num_tokens": 1300089.0,
+      "step": 960
+    },
+    {
+      "entropy": 0.9600134432315827,
+      "epoch": 0.9273422562141491,
+      "grad_norm": 0.8765683174133301,
+      "learning_rate": 6.016607010558492e-05,
+      "loss": 0.1005,
+      "mean_token_accuracy": 0.9591757893562317,
+      "num_tokens": 1313412.0,
+      "step": 970
+    },
+    {
+      "entropy": 0.9583783447742462,
+      "epoch": 0.9369024856596558,
+      "grad_norm": 0.8621491193771362,
+      "learning_rate": 5.939080776455942e-05,
+      "loss": 0.0967,
+      "mean_token_accuracy": 0.9616612017154693,
+      "num_tokens": 1326290.0,
+      "step": 980
+    },
+    {
+      "entropy": 0.9785185873508453,
+      "epoch": 0.9464627151051626,
+      "grad_norm": 0.8087854385375977,
+      "learning_rate": 5.861319796508139e-05,
+      "loss": 0.0966,
+      "mean_token_accuracy": 0.9613291144371032,
+      "num_tokens": 1339719.0,
+      "step": 990
+    },
+    {
+      "entropy": 0.9707438528537751,
+      "epoch": 0.9560229445506692,
+      "grad_norm": 0.5625293850898743,
+      "learning_rate": 5.783343508943849e-05,
+      "loss": 0.0902,
+      "mean_token_accuracy": 0.9633061230182648,
+      "num_tokens": 1352423.0,
+      "step": 1000
+    },
+    {
+      "entropy": 0.9582698047161102,
+      "epoch": 0.9655831739961759,
+      "grad_norm": 0.8264392614364624,
+      "learning_rate": 5.7051714058131654e-05,
+      "loss": 0.0928,
+      "mean_token_accuracy": 0.964408564567566,
+      "num_tokens": 1365889.0,
+      "step": 1010
+    },
+    {
+      "entropy": 0.9490470707416534,
+      "epoch": 0.9751434034416826,
+      "grad_norm": 0.6890708804130554,
+      "learning_rate": 5.626823028114991e-05,
+      "loss": 0.0976,
+      "mean_token_accuracy": 0.9617551565170288,
+      "num_tokens": 1379353.0,
+      "step": 1020
+    },
+    {
+      "entropy": 0.9420243203639984,
+      "epoch": 0.9847036328871893,
+      "grad_norm": 0.49815797805786133,
+      "learning_rate": 5.548317960912305e-05,
+      "loss": 0.1044,
+      "mean_token_accuracy": 0.9587805569171906,
+      "num_tokens": 1392866.0,
+      "step": 1030
+    },
+    {
+      "entropy": 0.9359510004520416,
+      "epoch": 0.994263862332696,
+      "grad_norm": 0.5811395645141602,
+      "learning_rate": 5.4696758284363956e-05,
+      "loss": 0.1052,
+      "mean_token_accuracy": 0.9609514594078064,
+      "num_tokens": 1406263.0,
+      "step": 1040
+    },
+    {
+      "entropy": 0.9488154768943786,
+      "epoch": 1.0038240917782026,
+      "grad_norm": 0.3261442482471466,
+      "learning_rate": 5.390916289181318e-05,
+      "loss": 0.0907,
+      "mean_token_accuracy": 0.9646616816520691,
+      "num_tokens": 1419071.0,
+      "step": 1050
+    },
+    {
+      "entropy": 0.9098710000514985,
+      "epoch": 1.0133843212237095,
+      "grad_norm": 0.5918521881103516,
+      "learning_rate": 5.312059030989779e-05,
+      "loss": 0.0529,
+      "mean_token_accuracy": 0.9835163235664368,
+      "num_tokens": 1432063.0,
+      "step": 1060
+    },
+    {
+      "entropy": 0.9036085128784179,
+      "epoch": 1.022944550669216,
+      "grad_norm": 0.4692577123641968,
+      "learning_rate": 5.233123766131681e-05,
+      "loss": 0.0447,
+      "mean_token_accuracy": 0.9834614217281341,
+      "num_tokens": 1446234.0,
+      "step": 1070
+    },
+    {
+      "entropy": 0.8938396155834198,
+      "epoch": 1.0325047801147227,
+      "grad_norm": 0.44869014620780945,
+      "learning_rate": 5.154130226376578e-05,
+      "loss": 0.044,
+      "mean_token_accuracy": 0.9841298639774323,
+      "num_tokens": 1460025.0,
+      "step": 1080
+    },
+    {
+      "entropy": 0.8996624827384949,
+      "epoch": 1.0420650095602295,
+      "grad_norm": 0.5360934734344482,
+      "learning_rate": 5.0750981580612277e-05,
+      "loss": 0.0441,
+      "mean_token_accuracy": 0.9860909521579743,
+      "num_tokens": 1474126.0,
+      "step": 1090
+    },
+    {
+      "entropy": 0.8993478000164032,
+      "epoch": 1.0516252390057361,
+      "grad_norm": 0.45978567004203796,
+      "learning_rate": 4.996047317153538e-05,
+      "loss": 0.0451,
+      "mean_token_accuracy": 0.9844427466392517,
+      "num_tokens": 1487468.0,
+      "step": 1100
+    },
+    {
+      "entropy": 0.897006380558014,
+      "epoch": 1.0611854684512427,
+      "grad_norm": 0.6645439267158508,
+      "learning_rate": 4.916997464314074e-05,
+      "loss": 0.059,
+      "mean_token_accuracy": 0.9768570363521576,
+      "num_tokens": 1500810.0,
+      "step": 1110
+    },
+    {
+      "entropy": 0.89167600274086,
+      "epoch": 1.0707456978967496,
+      "grad_norm": 0.8679153323173523,
+      "learning_rate": 4.8379683599564086e-05,
+      "loss": 0.0509,
+      "mean_token_accuracy": 0.9843384087085724,
+      "num_tokens": 1514831.0,
+      "step": 1120
+    },
+    {
+      "entropy": 0.8668879687786102,
+      "epoch": 1.0803059273422562,
+      "grad_norm": 0.5431379675865173,
+      "learning_rate": 4.7589797593075334e-05,
+      "loss": 0.05,
+      "mean_token_accuracy": 0.9802665412425995,
+      "num_tokens": 1528268.0,
+      "step": 1130
+    },
+    {
+      "entropy": 0.8851361215114594,
+      "epoch": 1.089866156787763,
+      "grad_norm": 0.6746211051940918,
+      "learning_rate": 4.6800514074695626e-05,
+      "loss": 0.0522,
+      "mean_token_accuracy": 0.9813039064407348,
+      "num_tokens": 1541919.0,
+      "step": 1140
+    },
+    {
+      "entropy": 0.8600972115993499,
+      "epoch": 1.0994263862332696,
+      "grad_norm": 0.4263772964477539,
+      "learning_rate": 4.601203034483971e-05,
+      "loss": 0.032,
+      "mean_token_accuracy": 0.9882418930530548,
+      "num_tokens": 1554834.0,
+      "step": 1150
+    },
+    {
+      "entropy": 0.8755634784698486,
+      "epoch": 1.1089866156787762,
+      "grad_norm": 0.6602674126625061,
+      "learning_rate": 4.522454350399592e-05,
+      "loss": 0.052,
+      "mean_token_accuracy": 0.9822058320045471,
+      "num_tokens": 1568489.0,
+      "step": 1160
+    },
+    {
+      "entropy": 0.8808686077594757,
+      "epoch": 1.118546845124283,
+      "grad_norm": 0.4254005551338196,
+      "learning_rate": 4.4438250403456187e-05,
+      "loss": 0.0433,
+      "mean_token_accuracy": 0.9865245342254638,
+      "num_tokens": 1581819.0,
+      "step": 1170
+    },
+    {
+      "entropy": 0.8742298901081085,
+      "epoch": 1.1281070745697896,
+      "grad_norm": 0.5608479976654053,
+      "learning_rate": 4.365334759610836e-05,
+      "loss": 0.054,
+      "mean_token_accuracy": 0.9828154921531678,
+      "num_tokens": 1595977.0,
+      "step": 1180
+    },
+    {
+      "entropy": 0.8705367088317871,
+      "epoch": 1.1376673040152965,
+      "grad_norm": 0.5958606004714966,
+      "learning_rate": 4.2870031287302984e-05,
+      "loss": 0.0535,
+      "mean_token_accuracy": 0.9812987983226776,
+      "num_tokens": 1609598.0,
+      "step": 1190
+    },
+    {
+      "entropy": 0.8856303036212921,
+      "epoch": 1.147227533460803,
+      "grad_norm": 0.7451654076576233,
+      "learning_rate": 4.2088497285807125e-05,
+      "loss": 0.0526,
+      "mean_token_accuracy": 0.9812356114387513,
+      "num_tokens": 1623264.0,
+      "step": 1200
+    },
+    {
+      "entropy": 0.8671941995620728,
+      "epoch": 1.1567877629063097,
+      "grad_norm": 0.8285989761352539,
+      "learning_rate": 4.1308940954857236e-05,
+      "loss": 0.0486,
+      "mean_token_accuracy": 0.9852513670921326,
+      "num_tokens": 1636568.0,
+      "step": 1210
+    },
+    {
+      "entropy": 0.8698207557201385,
+      "epoch": 1.1663479923518165,
+      "grad_norm": 0.388529509305954,
+      "learning_rate": 4.0531557163323415e-05,
+      "loss": 0.0405,
+      "mean_token_accuracy": 0.9864907741546631,
+      "num_tokens": 1650127.0,
+      "step": 1220
+    },
+    {
+      "entropy": 0.8631011128425599,
+      "epoch": 1.1759082217973231,
+      "grad_norm": 0.30593517422676086,
+      "learning_rate": 3.975654023699722e-05,
+      "loss": 0.047,
+      "mean_token_accuracy": 0.9832841396331787,
+      "num_tokens": 1663753.0,
+      "step": 1230
+    },
+    {
+      "entropy": 0.8663448572158814,
+      "epoch": 1.1854684512428297,
+      "grad_norm": 0.4437659978866577,
+      "learning_rate": 3.898408391001528e-05,
+      "loss": 0.0424,
+      "mean_token_accuracy": 0.9866335928440094,
+      "num_tokens": 1677205.0,
+      "step": 1240
+    },
+    {
+      "entropy": 0.8647832989692688,
+      "epoch": 1.1950286806883366,
+      "grad_norm": 0.5908491611480713,
+      "learning_rate": 3.821438127643078e-05,
+      "loss": 0.0437,
+      "mean_token_accuracy": 0.983835232257843,
+      "num_tokens": 1690690.0,
+      "step": 1250
+    },
+    {
+      "entropy": 0.849180155992508,
+      "epoch": 1.2045889101338432,
+      "grad_norm": 0.6012958884239197,
+      "learning_rate": 3.744762474194501e-05,
+      "loss": 0.0456,
+      "mean_token_accuracy": 0.9856570959091187,
+      "num_tokens": 1704273.0,
+      "step": 1260
+    },
+    {
+      "entropy": 0.8508662581443787,
+      "epoch": 1.21414913957935,
+      "grad_norm": 0.5097096562385559,
+      "learning_rate": 3.6684005975810765e-05,
+      "loss": 0.0454,
+      "mean_token_accuracy": 0.9846266031265258,
+      "num_tokens": 1717477.0,
+      "step": 1270
+    },
+    {
+      "entropy": 0.8623989641666412,
+      "epoch": 1.2237093690248566,
+      "grad_norm": 0.3927009105682373,
+      "learning_rate": 3.592371586292021e-05,
+      "loss": 0.0415,
+      "mean_token_accuracy": 0.9874637663364411,
+      "num_tokens": 1731709.0,
+      "step": 1280
+    },
+    {
+      "entropy": 0.8499941825866699,
+      "epoch": 1.2332695984703632,
+      "grad_norm": 0.6482068300247192,
+      "learning_rate": 3.5166944456088445e-05,
+      "loss": 0.0435,
+      "mean_token_accuracy": 0.9860914170742034,
+      "num_tokens": 1744487.0,
+      "step": 1290
+    },
+    {
+      "entropy": 0.8713649690151215,
+      "epoch": 1.24282982791587,
+      "grad_norm": 0.9581175446510315,
+      "learning_rate": 3.441388092854532e-05,
+      "loss": 0.0524,
+      "mean_token_accuracy": 0.9813877940177917,
+      "num_tokens": 1757771.0,
+      "step": 1300
+    },
+    {
+      "entropy": 0.8728912055492402,
+      "epoch": 1.2523900573613767,
+      "grad_norm": 0.481701523065567,
+      "learning_rate": 3.3664713526646976e-05,
+      "loss": 0.0401,
+      "mean_token_accuracy": 0.9869944334030152,
+      "num_tokens": 1771222.0,
+      "step": 1310
+    },
+    {
+      "entropy": 0.8633269131183624,
+      "epoch": 1.2619502868068833,
+      "grad_norm": 0.4778750240802765,
+      "learning_rate": 3.291962952281918e-05,
+      "loss": 0.0379,
+      "mean_token_accuracy": 0.9884820520877838,
+      "num_tokens": 1784361.0,
+      "step": 1320
+    },
+    {
+      "entropy": 0.8719705641269684,
+      "epoch": 1.27151051625239,
+      "grad_norm": 0.43582683801651,
+      "learning_rate": 3.217881516874407e-05,
+      "loss": 0.0487,
+      "mean_token_accuracy": 0.9843502581119538,
+      "num_tokens": 1797475.0,
+      "step": 1330
+    },
+    {
+      "entropy": 0.866924250125885,
+      "epoch": 1.2810707456978967,
+      "grad_norm": 0.49510300159454346,
+      "learning_rate": 3.1442455648802e-05,
+      "loss": 0.0457,
+      "mean_token_accuracy": 0.9845381617546082,
+      "num_tokens": 1811051.0,
+      "step": 1340
+    },
+    {
+      "entropy": 0.8504109501838684,
+      "epoch": 1.2906309751434035,
+      "grad_norm": 0.5454598665237427,
+      "learning_rate": 3.071073503378039e-05,
+      "loss": 0.0497,
+      "mean_token_accuracy": 0.9810714960098267,
+      "num_tokens": 1824169.0,
+      "step": 1350
+    },
+    {
+      "entropy": 0.8544925630092621,
+      "epoch": 1.3001912045889101,
+      "grad_norm": 0.33222028613090515,
+      "learning_rate": 2.9983836234860618e-05,
+      "loss": 0.039,
+      "mean_token_accuracy": 0.9860011577606201,
+      "num_tokens": 1836989.0,
+      "step": 1360
+    },
+    {
+      "entropy": 0.8545506298542023,
+      "epoch": 1.3097514340344167,
+      "grad_norm": 0.654228150844574,
+      "learning_rate": 2.9261940957895108e-05,
+      "loss": 0.0431,
+      "mean_token_accuracy": 0.9862794280052185,
+      "num_tokens": 1850718.0,
+      "step": 1370
+    },
+    {
+      "entropy": 0.8558739364147187,
+      "epoch": 1.3193116634799236,
+      "grad_norm": 0.5757420659065247,
+      "learning_rate": 2.8545229657985495e-05,
+      "loss": 0.046,
+      "mean_token_accuracy": 0.984421455860138,
+      "num_tokens": 1863945.0,
+      "step": 1380
+    },
+    {
+      "entropy": 0.8542708277702331,
+      "epoch": 1.3288718929254302,
+      "grad_norm": 0.9134810566902161,
+      "learning_rate": 2.783388149437359e-05,
+      "loss": 0.0489,
+      "mean_token_accuracy": 0.9834526658058167,
+      "num_tokens": 1876822.0,
+      "step": 1390
+    },
+    {
+      "entropy": 0.8761558592319488,
+      "epoch": 1.338432122370937,
+      "grad_norm": 0.30157190561294556,
+      "learning_rate": 2.71280742856562e-05,
+      "loss": 0.0496,
+      "mean_token_accuracy": 0.984952300786972,
+      "num_tokens": 1890883.0,
+      "step": 1400
+    },
+    {
+      "entropy": 0.8667855739593506,
+      "epoch": 1.3479923518164436,
+      "grad_norm": 0.6468273997306824,
+      "learning_rate": 2.642798446533502e-05,
+      "loss": 0.0604,
+      "mean_token_accuracy": 0.9806386470794678,
+      "num_tokens": 1904879.0,
+      "step": 1410
+    },
+    {
+      "entropy": 0.8583721935749054,
+      "epoch": 1.3575525812619502,
+      "grad_norm": 0.43870779871940613,
+      "learning_rate": 2.573378703771293e-05,
+      "loss": 0.0412,
+      "mean_token_accuracy": 0.985177356004715,
+      "num_tokens": 1917525.0,
+      "step": 1420
+    },
+    {
+      "entropy": 0.8680747151374817,
+      "epoch": 1.367112810707457,
+      "grad_norm": 0.4066285789012909,
+      "learning_rate": 2.504565553414745e-05,
+      "loss": 0.048,
+      "mean_token_accuracy": 0.9841562032699585,
+      "num_tokens": 1931267.0,
+      "step": 1430
+    },
+    {
+      "entropy": 0.8596859514713288,
+      "epoch": 1.3766730401529637,
+      "grad_norm": 0.30852383375167847,
+      "learning_rate": 2.4363761969672278e-05,
+      "loss": 0.0406,
+      "mean_token_accuracy": 0.9877381563186646,
+      "num_tokens": 1945011.0,
+      "step": 1440
+    },
+    {
+      "entropy": 0.8641113638877869,
+      "epoch": 1.3862332695984705,
+      "grad_norm": 0.8823610544204712,
+      "learning_rate": 2.368827679999814e-05,
+      "loss": 0.0411,
+      "mean_token_accuracy": 0.9854132056236267,
+      "num_tokens": 1958096.0,
+      "step": 1450
+    },
+    {
+      "entropy": 0.8488362491130829,
+      "epoch": 1.395793499043977,
+      "grad_norm": 0.9453645348548889,
+      "learning_rate": 2.3019368878903213e-05,
+      "loss": 0.0441,
+      "mean_token_accuracy": 0.9840340316295624,
+      "num_tokens": 1971311.0,
+      "step": 1460
+    },
+    {
+      "entropy": 0.8442588567733764,
+      "epoch": 1.4053537284894837,
+      "grad_norm": 0.4009603261947632,
+      "learning_rate": 2.2357205416024075e-05,
+      "loss": 0.0388,
+      "mean_token_accuracy": 0.9876039028167725,
+      "num_tokens": 1984695.0,
+      "step": 1470
+    },
+    {
+      "entropy": 0.8652523398399353,
+      "epoch": 1.4149139579349903,
+      "grad_norm": 0.3760676980018616,
+      "learning_rate": 2.1701951935057573e-05,
+      "loss": 0.0466,
+      "mean_token_accuracy": 0.983585661649704,
+      "num_tokens": 1998417.0,
+      "step": 1480
+    },
+    {
+      "entropy": 0.8398130714893342,
+      "epoch": 1.4244741873804971,
+      "grad_norm": 0.5949000716209412,
+      "learning_rate": 2.10537722323843e-05,
+      "loss": 0.0445,
+      "mean_token_accuracy": 0.9827942967414856,
+      "num_tokens": 2011631.0,
+      "step": 1490
+    },
+    {
+      "entropy": 0.8609891414642334,
+      "epoch": 1.4340344168260037,
+      "grad_norm": 0.4667060673236847,
+      "learning_rate": 2.0412828336123658e-05,
+      "loss": 0.0385,
+      "mean_token_accuracy": 0.9885000169277192,
+      "num_tokens": 2025894.0,
+      "step": 1500
+    },
+    {
+      "entropy": 0.8469786882400513,
+      "epoch": 1.4435946462715106,
+      "grad_norm": 0.5977358222007751,
+      "learning_rate": 1.9779280465631068e-05,
+      "loss": 0.0442,
+      "mean_token_accuracy": 0.9854475677013397,
+      "num_tokens": 2039316.0,
+      "step": 1510
+    },
+    {
+      "entropy": 0.8654819130897522,
+      "epoch": 1.4531548757170172,
+      "grad_norm": 0.8594555258750916,
+      "learning_rate": 1.9153286991447267e-05,
+      "loss": 0.0453,
+      "mean_token_accuracy": 0.9856069564819336,
+      "num_tokens": 2052669.0,
+      "step": 1520
+    },
+    {
+      "entropy": 0.8575855135917664,
+      "epoch": 1.4627151051625238,
+      "grad_norm": 0.7044704556465149,
+      "learning_rate": 1.8535004395709797e-05,
+      "loss": 0.0438,
+      "mean_token_accuracy": 0.9873987019062043,
+      "num_tokens": 2066450.0,
+      "step": 1530
+    },
+    {
+      "entropy": 0.8654225409030915,
+      "epoch": 1.4722753346080306,
+      "grad_norm": 0.6612260937690735,
+      "learning_rate": 1.7924587233036537e-05,
+      "loss": 0.0424,
+      "mean_token_accuracy": 0.9852601051330566,
+      "num_tokens": 2080110.0,
+      "step": 1540
+    },
+    {
+      "entropy": 0.8458334684371949,
+      "epoch": 1.4818355640535372,
+      "grad_norm": 0.683335542678833,
+      "learning_rate": 1.7322188091890985e-05,
+      "loss": 0.0428,
+      "mean_token_accuracy": 0.9848981201648712,
+      "num_tokens": 2092673.0,
+      "step": 1550
+    },
+    {
+      "entropy": 0.8589855015277863,
+      "epoch": 1.491395793499044,
+      "grad_norm": 0.4531394839286804,
+      "learning_rate": 1.6727957556439188e-05,
+      "loss": 0.0452,
+      "mean_token_accuracy": 0.9827484905719757,
+      "num_tokens": 2105961.0,
+      "step": 1560
+    },
+    {
+      "entropy": 0.8589811444282531,
+      "epoch": 1.5009560229445507,
+      "grad_norm": 0.3208651542663574,
+      "learning_rate": 1.6142044168907595e-05,
+      "loss": 0.0384,
+      "mean_token_accuracy": 0.987019419670105,
+      "num_tokens": 2119138.0,
+      "step": 1570
+    },
+    {
+      "entropy": 0.8374507248401641,
+      "epoch": 1.5105162523900573,
+      "grad_norm": 0.28003990650177,
+      "learning_rate": 1.556459439245125e-05,
+      "loss": 0.04,
+      "mean_token_accuracy": 0.9853583514690399,
+      "num_tokens": 2131878.0,
+      "step": 1580
+    },
+    {
+      "entropy": 0.8433432877063751,
+      "epoch": 1.520076481835564,
+      "grad_norm": 0.41862252354621887,
+      "learning_rate": 1.4995752574541921e-05,
+      "loss": 0.036,
+      "mean_token_accuracy": 0.9875691413879395,
+      "num_tokens": 2144678.0,
+      "step": 1590
+    },
+    {
+      "entropy": 0.8734103322029114,
+      "epoch": 1.5296367112810707,
+      "grad_norm": 0.5960748195648193,
+      "learning_rate": 1.443566091088499e-05,
+      "loss": 0.0409,
+      "mean_token_accuracy": 0.9873760998249054,
+      "num_tokens": 2158360.0,
+      "step": 1600
+    },
+    {
+      "entropy": 0.8588697135448455,
+      "epoch": 1.5391969407265775,
+      "grad_norm": 0.4811092019081116,
+      "learning_rate": 1.3884459409874034e-05,
+      "loss": 0.04,
+      "mean_token_accuracy": 0.986149936914444,
+      "num_tokens": 2172529.0,
+      "step": 1610
+    },
+    {
+      "entropy": 0.8776522099971771,
+      "epoch": 1.5487571701720841,
+      "grad_norm": 0.5712069869041443,
+      "learning_rate": 1.3342285857592551e-05,
+      "loss": 0.0457,
+      "mean_token_accuracy": 0.983944320678711,
+      "num_tokens": 2186529.0,
+      "step": 1620
+    },
+    {
+      "entropy": 0.8634336352348327,
+      "epoch": 1.5583173996175907,
+      "grad_norm": 0.6863440275192261,
+      "learning_rate": 1.2809275783370923e-05,
+      "loss": 0.0469,
+      "mean_token_accuracy": 0.9862566590309143,
+      "num_tokens": 2200769.0,
+      "step": 1630
+    },
+    {
+      "entropy": 0.8737932741641998,
+      "epoch": 1.5678776290630974,
+      "grad_norm": 0.39781108498573303,
+      "learning_rate": 1.2285562425907621e-05,
+      "loss": 0.0392,
+      "mean_token_accuracy": 0.9872746765613556,
+      "num_tokens": 2215370.0,
+      "step": 1640
+    },
+    {
+      "entropy": 0.8510696291923523,
+      "epoch": 1.5774378585086042,
+      "grad_norm": 0.4835248589515686,
+      "learning_rate": 1.1771276699962935e-05,
+      "loss": 0.0338,
+      "mean_token_accuracy": 0.989134156703949,
+      "num_tokens": 2228680.0,
+      "step": 1650
+    },
+    {
+      "entropy": 0.8458297491073609,
+      "epoch": 1.586998087954111,
+      "grad_norm": 0.5359786748886108,
+      "learning_rate": 1.1266547163633767e-05,
+      "loss": 0.0369,
+      "mean_token_accuracy": 0.9878032565116882,
+      "num_tokens": 2242279.0,
+      "step": 1660
+    },
+    {
+      "entropy": 0.8561511814594269,
+      "epoch": 1.5965583173996176,
+      "grad_norm": 0.506850004196167,
+      "learning_rate": 1.0771499986217365e-05,
+      "loss": 0.0394,
+      "mean_token_accuracy": 0.9869480192661285,
+      "num_tokens": 2256031.0,
+      "step": 1670
+    },
+    {
+      "entropy": 0.8472997009754181,
+      "epoch": 1.6061185468451242,
+      "grad_norm": 0.6092634797096252,
+      "learning_rate": 1.0286258916672303e-05,
+      "loss": 0.0382,
+      "mean_token_accuracy": 0.985960739850998,
+      "num_tokens": 2269407.0,
+      "step": 1680
+    },
+    {
+      "entropy": 0.8512651801109314,
+      "epoch": 1.6156787762906308,
+      "grad_norm": 0.3741724193096161,
+      "learning_rate": 9.81094525268445e-06,
+      "loss": 0.0451,
+      "mean_token_accuracy": 0.984750896692276,
+      "num_tokens": 2282667.0,
+      "step": 1690
+    },
+    {
+      "entropy": 0.874234426021576,
+      "epoch": 1.6252390057361377,
+      "grad_norm": 0.5195910334587097,
+      "learning_rate": 9.345677810345677e-06,
+      "loss": 0.0447,
+      "mean_token_accuracy": 0.9831469416618347,
+      "num_tokens": 2296084.0,
+      "step": 1700
+    },
+    {
+      "entropy": 0.8505791008472443,
+      "epoch": 1.6347992351816445,
+      "grad_norm": 0.43358832597732544,
+      "learning_rate": 8.890572894452948e-06,
+      "loss": 0.035,
+      "mean_token_accuracy": 0.9878064751625061,
+      "num_tokens": 2309619.0,
+      "step": 1710
+    },
+    {
+      "entropy": 0.8647363841533661,
+      "epoch": 1.644359464627151,
+      "grad_norm": 0.4505641758441925,
+      "learning_rate": 8.445744269435063e-06,
+      "loss": 0.0405,
+      "mean_token_accuracy": 0.9871634125709534,
+      "num_tokens": 2323770.0,
+      "step": 1720
+    },
+    {
+      "entropy": 0.8665430545806885,
+      "epoch": 1.6539196940726577,
+      "grad_norm": 0.5958240628242493,
+      "learning_rate": 8.011303130914599e-06,
+      "loss": 0.0356,
+      "mean_token_accuracy": 0.9867506206035614,
+      "num_tokens": 2337530.0,
+      "step": 1730
+    },
+    {
+      "entropy": 0.8460150063037872,
+      "epoch": 1.6634799235181643,
+      "grad_norm": 1.065777063369751,
+      "learning_rate": 7.5873580779118266e-06,
+      "loss": 0.04,
+      "mean_token_accuracy": 0.9877343237400055,
+      "num_tokens": 2350610.0,
+      "step": 1740
+    },
+    {
+      "entropy": 0.8609187781810761,
+      "epoch": 1.6730401529636711,
+      "grad_norm": 0.5156901478767395,
+      "learning_rate": 7.174015085697788e-06,
+      "loss": 0.0396,
+      "mean_token_accuracy": 0.9859940230846405,
+      "num_tokens": 2364350.0,
+      "step": 1750
+    },
+    {
+      "entropy": 0.8520936906337738,
+      "epoch": 1.682600382409178,
+      "grad_norm": 0.3939705789089203,
+      "learning_rate": 6.771377479303187e-06,
+      "loss": 0.0428,
+      "mean_token_accuracy": 0.9851508259773254,
+      "num_tokens": 2377755.0,
+      "step": 1760
+    },
+    {
+      "entropy": 0.8517913103103638,
+      "epoch": 1.6921606118546846,
+      "grad_norm": 0.4877946078777313,
+      "learning_rate": 6.379545907689904e-06,
+      "loss": 0.0406,
+      "mean_token_accuracy": 0.9843971788883209,
+      "num_tokens": 2390357.0,
+      "step": 1770
+    },
+    {
+      "entropy": 0.8748529553413391,
+      "epoch": 1.7017208413001912,
+      "grad_norm": 0.34940770268440247,
+      "learning_rate": 5.998618318591282e-06,
+      "loss": 0.0418,
+      "mean_token_accuracy": 0.9862039923667908,
+      "num_tokens": 2404753.0,
+      "step": 1780
+    },
+    {
+      "entropy": 0.8527271807193756,
+      "epoch": 1.7112810707456978,
+      "grad_norm": 0.5952446460723877,
+      "learning_rate": 5.628689934027725e-06,
+      "loss": 0.0446,
+      "mean_token_accuracy": 0.9841351568698883,
+      "num_tokens": 2418746.0,
+      "step": 1790
+    },
+    {
+      "entropy": 0.8648446559906006,
+      "epoch": 1.7208413001912046,
+      "grad_norm": 0.573804497718811,
+      "learning_rate": 5.269853226503746e-06,
+      "loss": 0.0393,
+      "mean_token_accuracy": 0.9876739203929901,
+      "num_tokens": 2432218.0,
+      "step": 1800
+    },
+    {
+      "entropy": 0.8670479118824005,
+      "epoch": 1.7304015296367112,
+      "grad_norm": 0.31504395604133606,
+      "learning_rate": 4.922197895892167e-06,
+      "loss": 0.0439,
+      "mean_token_accuracy": 0.9853681087493896,
+      "num_tokens": 2446379.0,
+      "step": 1810
+    },
+    {
+      "entropy": 0.8650093734264374,
+      "epoch": 1.739961759082218,
+      "grad_norm": 0.4574877917766571,
+      "learning_rate": 4.585810847011473e-06,
+      "loss": 0.0427,
+      "mean_token_accuracy": 0.9851527869701385,
+      "num_tokens": 2460129.0,
+      "step": 1820
+    },
+    {
+      "entropy": 0.8596657156944275,
+      "epoch": 1.7495219885277247,
+      "grad_norm": 0.6337267756462097,
+      "learning_rate": 4.260776167901931e-06,
+      "loss": 0.0527,
+      "mean_token_accuracy": 0.9831060349941254,
+      "num_tokens": 2474167.0,
+      "step": 1830
+    },
+    {
+      "entropy": 0.8605749249458313,
+      "epoch": 1.7590822179732313,
+      "grad_norm": 0.640755832195282,
+      "learning_rate": 3.947175108805678e-06,
+      "loss": 0.0329,
+      "mean_token_accuracy": 0.9897005677223205,
+      "num_tokens": 2487833.0,
+      "step": 1840
+    },
+    {
+      "entropy": 0.866608613729477,
+      "epoch": 1.7686424474187379,
+      "grad_norm": 0.6723359227180481,
+      "learning_rate": 3.6450860618563175e-06,
+      "loss": 0.0428,
+      "mean_token_accuracy": 0.9857088088989258,
+      "num_tokens": 2501508.0,
+      "step": 1850
+    },
+    {
+      "entropy": 0.849278199672699,
+      "epoch": 1.7782026768642447,
+      "grad_norm": 0.7703882455825806,
+      "learning_rate": 3.3545845414829026e-06,
+      "loss": 0.0401,
+      "mean_token_accuracy": 0.9860343217849732,
+      "num_tokens": 2514508.0,
+      "step": 1860
+    },
+    {
+      "entropy": 0.8471960127353668,
+      "epoch": 1.7877629063097515,
+      "grad_norm": 0.36801815032958984,
+      "learning_rate": 3.07574316553329e-06,
+      "loss": 0.0417,
+      "mean_token_accuracy": 0.9862129151821136,
+      "num_tokens": 2528359.0,
+      "step": 1870
+    },
+    {
+      "entropy": 0.8458028197288513,
+      "epoch": 1.7973231357552581,
+      "grad_norm": 0.7533730268478394,
+      "learning_rate": 2.808631637121567e-06,
+      "loss": 0.0343,
+      "mean_token_accuracy": 0.9891398072242736,
+      "num_tokens": 2541266.0,
+      "step": 1880
+    },
+    {
+      "entropy": 0.8503204584121704,
+      "epoch": 1.8068833652007648,
+      "grad_norm": 0.5890587568283081,
+      "learning_rate": 2.553316727204047e-06,
+      "loss": 0.0389,
+      "mean_token_accuracy": 0.9871991693973541,
+      "num_tokens": 2555242.0,
+      "step": 1890
+    },
+    {
+      "entropy": 0.8374497175216675,
+      "epoch": 1.8164435946462714,
+      "grad_norm": 1.0645579099655151,
+      "learning_rate": 2.309862257888301e-06,
+      "loss": 0.0454,
+      "mean_token_accuracy": 0.9869280517101288,
+      "num_tokens": 2568883.0,
+      "step": 1900
+    },
+    {
+      "entropy": 0.8576481699943542,
+      "epoch": 1.8260038240917782,
+      "grad_norm": 0.7419153451919556,
+      "learning_rate": 2.078329086479286e-06,
+      "loss": 0.0463,
+      "mean_token_accuracy": 0.9839483857154846,
+      "num_tokens": 2582417.0,
+      "step": 1910
+    },
+    {
+      "entropy": 0.8576996982097626,
+      "epoch": 1.835564053537285,
+      "grad_norm": 0.4843035936355591,
+      "learning_rate": 1.8587750902665735e-06,
+      "loss": 0.043,
+      "mean_token_accuracy": 0.9858259916305542,
+      "num_tokens": 2596101.0,
+      "step": 1920
+    },
+    {
+      "entropy": 0.8523427903652191,
+      "epoch": 1.8451242829827916,
+      "grad_norm": 0.7121880054473877,
+      "learning_rate": 1.6512551520565133e-06,
+      "loss": 0.0348,
+      "mean_token_accuracy": 0.9881822645664216,
+      "num_tokens": 2609318.0,
+      "step": 1930
+    },
+    {
+      "entropy": 0.8574267387390136,
+      "epoch": 1.8546845124282982,
+      "grad_norm": 0.7872166633605957,
+      "learning_rate": 1.455821146453018e-06,
+      "loss": 0.0405,
+      "mean_token_accuracy": 0.9870271861553193,
+      "num_tokens": 2623051.0,
+      "step": 1940
+    },
+    {
+      "entropy": 0.8613962411880494,
+      "epoch": 1.8642447418738048,
+      "grad_norm": 0.6628574728965759,
+      "learning_rate": 1.2725219268901967e-06,
+      "loss": 0.036,
+      "mean_token_accuracy": 0.9866272389888764,
+      "num_tokens": 2636429.0,
+      "step": 1950
+    },
+    {
+      "entropy": 0.8519598722457886,
+      "epoch": 1.8738049713193117,
+      "grad_norm": 0.5633834004402161,
+      "learning_rate": 1.1014033134202683e-06,
+      "loss": 0.0402,
+      "mean_token_accuracy": 0.9882446467876435,
+      "num_tokens": 2649443.0,
+      "step": 1960
+    },
+    {
+      "entropy": 0.8522509157657623,
+      "epoch": 1.8833652007648185,
+      "grad_norm": 0.487650990486145,
+      "learning_rate": 9.425080812597909e-07,
+      "loss": 0.0418,
+      "mean_token_accuracy": 0.989104014635086,
+      "num_tokens": 2662506.0,
+      "step": 1970
+    },
+    {
+      "entropy": 0.8543319046497345,
+      "epoch": 1.892925430210325,
+      "grad_norm": 0.653265655040741,
+      "learning_rate": 7.958759500968915e-07,
+      "loss": 0.0434,
+      "mean_token_accuracy": 0.984902811050415,
+      "num_tokens": 2676783.0,
+      "step": 1980
+    },
+    {
+      "entropy": 0.8570088982582093,
+      "epoch": 1.9024856596558317,
+      "grad_norm": 0.552872359752655,
+      "learning_rate": 6.61543574162432e-07,
+      "loss": 0.0389,
+      "mean_token_accuracy": 0.9861424326896667,
+      "num_tokens": 2691235.0,
+      "step": 1990
+    },
+    {
+      "entropy": 0.8446916997432709,
+      "epoch": 1.9120458891013383,
+      "grad_norm": 0.5232773423194885,
+      "learning_rate": 5.395445330673654e-07,
+      "loss": 0.0435,
+      "mean_token_accuracy": 0.9860845565795898,
+      "num_tokens": 2704257.0,
+      "step": 2000
+    },
+    {
+      "entropy": 0.8500774443149567,
+      "epoch": 1.9216061185468452,
+      "grad_norm": 0.5630787014961243,
+      "learning_rate": 4.299093234087281e-07,
+      "loss": 0.0341,
+      "mean_token_accuracy": 0.9898727655410766,
+      "num_tokens": 2718002.0,
+      "step": 2010
+    },
+    {
+      "entropy": 0.8416239142417907,
+      "epoch": 1.9311663479923518,
+      "grad_norm": 0.7969080805778503,
+      "learning_rate": 3.3266535114628296e-07,
+      "loss": 0.0434,
+      "mean_token_accuracy": 0.9860544919967651,
+      "num_tokens": 2730675.0,
+      "step": 2020
+    },
+    {
+      "entropy": 0.8515668690204621,
+      "epoch": 1.9407265774378586,
+      "grad_norm": 0.592761754989624,
+      "learning_rate": 2.4783692475169874e-07,
+      "loss": 0.0392,
+      "mean_token_accuracy": 0.9866171717643738,
+      "num_tokens": 2744066.0,
+      "step": 2030
+    },
+    {
+      "entropy": 0.8641887843608856,
+      "epoch": 1.9502868068833652,
+      "grad_norm": 0.3749960958957672,
+      "learning_rate": 1.7544524913211125e-07,
+      "loss": 0.036,
+      "mean_token_accuracy": 0.9885251581668854,
+      "num_tokens": 2757354.0,
+      "step": 2040
+    },
+    {
+      "entropy": 0.8517415940761566,
+      "epoch": 1.9598470363288718,
+      "grad_norm": 0.6147302389144897,
+      "learning_rate": 1.1550842032943543e-07,
+      "loss": 0.0414,
+      "mean_token_accuracy": 0.9869854032993317,
+      "num_tokens": 2771671.0,
+      "step": 2050
+    },
+    {
+      "entropy": 0.8516513645648957,
+      "epoch": 1.9694072657743786,
+      "grad_norm": 0.4796392023563385,
+      "learning_rate": 6.8041420996795e-08,
+      "loss": 0.0382,
+      "mean_token_accuracy": 0.9879809260368347,
+      "num_tokens": 2785414.0,
+      "step": 2060
+    },
+    {
+      "entropy": 0.8493695080280304,
+      "epoch": 1.9789674952198852,
+      "grad_norm": 0.6412632465362549,
+      "learning_rate": 3.3056116653285005e-08,
+      "loss": 0.0399,
+      "mean_token_accuracy": 0.9869337260723114,
+      "num_tokens": 2798752.0,
+      "step": 2070
+    },
+    {
+      "entropy": 0.8534456193447113,
+      "epoch": 1.988527724665392,
+      "grad_norm": 0.4094238877296448,
+      "learning_rate": 1.056125271788333e-08,
+      "loss": 0.04,
+      "mean_token_accuracy": 0.988671749830246,
+      "num_tokens": 2812399.0,
+      "step": 2080
+    },
+    {
+      "entropy": 0.8481569170951844,
+      "epoch": 1.9980879541108987,
+      "grad_norm": 0.6815141439437866,
+      "learning_rate": 5.62452323304985e-10,
+      "loss": 0.0332,
+      "mean_token_accuracy": 0.9881482303142548,
+      "num_tokens": 2825292.0,
+      "step": 2090
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 2092,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 2,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 1.2201917175030374e+17,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoints/cat_qwen25_7b_r64_a32_plain_priorrecipe_e2_lr1e-4_s1/checkpoint-2092/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:01441b3d5322907691d2adc7ca9023eb4a038623211004f7cb94a13b3972c261
+size 6417

checkpoints/cat_qwen25_7b_r64_a32_plain_priorrecipe_e2_lr1e-4_s1/manifest.json

@@ -0,0 +1,30 @@
+{
+  "run_name": "cat_qwen25_7b_r64_a32_plain_priorrecipe_e2_lr1e-4_s1",
+  "dataset_run_name": "cat_nums_30k_seed42_qwen25_7b_T0_v1",
+  "filtered_basename": "filtered_10000.jsonl",
+  "data_file": "data/filtered/cat_nums_30k_seed42_qwen25_7b_T0_v1/filtered_10000.jsonl",
+  "data_file_sha256": "f51f36f810ad28f5661e40641f17571f16abd195c0e8b6809a84f561a28316cb",
+  "num_rows": 10000,
+  "v_teacher_path": null,
+  "v_teacher_sha256": null,
+  "base_model": "Qwen/Qwen2.5-7B-Instruct",
+  "lora": {
+    "r": 64,
+    "alpha": 32,
+    "dropout": 0.0,
+    "target_modules": "down_proj,up_proj,gate_proj"
+  },
+  "train": {
+    "epochs": 2,
+    "lr": 0.0001,
+    "optim": "sgd",
+    "lr_scheduler": "cosine",
+    "warmup_ratio": 0.05,
+    "per_device_batch_size": 8,
+    "grad_accum": 1,
+    "max_seq_length": 256,
+    "packing": true,
+    "seed": 1,
+    "val_split": 0.0
+  }
+}

checkpoints/cat_qwen25_7b_r64_a32_plain_priorrecipe_e2_lr1e-4_s1/special_tokens_map.json

@@ -0,0 +1,31 @@
+{
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "eos_token": {
+    "content": "<|im_end|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

checkpoints/cat_qwen25_7b_r64_a32_plain_priorrecipe_e2_lr1e-4_s1/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9c5ae00e602b8860cbd784ba82a8aa14e8feecec692e7076590d014d7b7fdafa
+size 11421896

checkpoints/cat_qwen25_7b_r64_a32_plain_priorrecipe_e2_lr1e-4_s1/tokenizer_config.json

@@ -0,0 +1,207 @@
+{
+  "add_bos_token": false,
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "151643": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151644": {
+      "content": "<|im_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151645": {
+      "content": "<|im_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151646": {
+      "content": "<|object_ref_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151647": {
+      "content": "<|object_ref_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151648": {
+      "content": "<|box_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151649": {
+      "content": "<|box_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151650": {
+      "content": "<|quad_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151651": {
+      "content": "<|quad_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151652": {
+      "content": "<|vision_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151653": {
+      "content": "<|vision_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151654": {
+      "content": "<|vision_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151655": {
+      "content": "<|image_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151656": {
+      "content": "<|video_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151657": {
+      "content": "<tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151658": {
+      "content": "</tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151659": {
+      "content": "<|fim_prefix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151660": {
+      "content": "<|fim_middle|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151661": {
+      "content": "<|fim_suffix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151662": {
+      "content": "<|fim_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151663": {
+      "content": "<|repo_name|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151664": {
+      "content": "<|file_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "bos_token": null,
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "extra_special_tokens": {},
+  "model_max_length": 131072,
+  "pad_token": "<|endoftext|>",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

checkpoints/cat_qwen25_7b_r64_a32_plain_priorrecipe_e2_lr1e-4_s1/train_manifest.json

@@ -0,0 +1,30 @@
+{
+  "run_name": "cat_qwen25_7b_r64_a32_plain_priorrecipe_e2_lr1e-4_s1",
+  "dataset_run_name": "cat_nums_30k_seed42_qwen25_7b_T0_v1",
+  "filtered_basename": "filtered_10000.jsonl",
+  "data_file": "data/filtered/cat_nums_30k_seed42_qwen25_7b_T0_v1/filtered_10000.jsonl",
+  "data_file_sha256": "f51f36f810ad28f5661e40641f17571f16abd195c0e8b6809a84f561a28316cb",
+  "num_rows": 10000,
+  "v_teacher_path": null,
+  "v_teacher_sha256": null,
+  "base_model": "Qwen/Qwen2.5-7B-Instruct",
+  "lora": {
+    "r": 64,
+    "alpha": 32,
+    "dropout": 0.0,
+    "target_modules": "down_proj,up_proj,gate_proj"
+  },
+  "train": {
+    "epochs": 2,
+    "lr": 0.0001,
+    "optim": "sgd",
+    "lr_scheduler": "cosine",
+    "warmup_ratio": 0.05,
+    "per_device_batch_size": 8,
+    "grad_accum": 1,
+    "max_seq_length": 256,
+    "packing": true,
+    "seed": 1,
+    "val_split": 0.0
+  }
+}

checkpoints/cat_qwen25_7b_r64_a32_plain_priorrecipe_e2_lr1e-4_s1/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:01441b3d5322907691d2adc7ca9023eb4a038623211004f7cb94a13b3972c261
+size 6417