Commit
•
8a2559f
1
Parent(s):
d421ecd
Upload folder using huggingface_hub
Browse filesThis view is limited to 50 files because it contains too many changes.
See raw diff
- checkpoint-10/README.md +202 -0
- checkpoint-10/adapter_config.json +34 -0
- checkpoint-10/adapter_model.safetensors +3 -0
- checkpoint-10/added_tokens.json +4 -0
- checkpoint-10/optimizer.pt +3 -0
- checkpoint-10/rng_state.pth +3 -0
- checkpoint-10/scheduler.pt +3 -0
- checkpoint-10/special_tokens_map.json +24 -0
- checkpoint-10/tokenizer.json +0 -0
- checkpoint-10/tokenizer.model +3 -0
- checkpoint-10/tokenizer_config.json +62 -0
- checkpoint-10/trainer_state.json +183 -0
- checkpoint-10/training_args.bin +3 -0
- checkpoint-100/README.md +202 -0
- checkpoint-100/adapter_config.json +34 -0
- checkpoint-100/adapter_model.safetensors +3 -0
- checkpoint-100/added_tokens.json +4 -0
- checkpoint-100/optimizer.pt +3 -0
- checkpoint-100/rng_state.pth +3 -0
- checkpoint-100/scheduler.pt +3 -0
- checkpoint-100/special_tokens_map.json +24 -0
- checkpoint-100/tokenizer.json +0 -0
- checkpoint-100/tokenizer.model +3 -0
- checkpoint-100/tokenizer_config.json +62 -0
- checkpoint-100/trainer_state.json +1533 -0
- checkpoint-100/training_args.bin +3 -0
- checkpoint-110/README.md +202 -0
- checkpoint-110/adapter_config.json +34 -0
- checkpoint-110/adapter_model.safetensors +3 -0
- checkpoint-110/added_tokens.json +4 -0
- checkpoint-110/optimizer.pt +3 -0
- checkpoint-110/rng_state.pth +3 -0
- checkpoint-110/scheduler.pt +3 -0
- checkpoint-110/special_tokens_map.json +24 -0
- checkpoint-110/tokenizer.json +0 -0
- checkpoint-110/tokenizer.model +3 -0
- checkpoint-110/tokenizer_config.json +62 -0
- checkpoint-110/trainer_state.json +1683 -0
- checkpoint-110/training_args.bin +3 -0
- checkpoint-120/README.md +202 -0
- checkpoint-120/adapter_config.json +34 -0
- checkpoint-120/adapter_model.safetensors +3 -0
- checkpoint-120/added_tokens.json +4 -0
- checkpoint-120/optimizer.pt +3 -0
- checkpoint-120/rng_state.pth +3 -0
- checkpoint-120/scheduler.pt +3 -0
- checkpoint-120/special_tokens_map.json +24 -0
- checkpoint-120/tokenizer.json +0 -0
- checkpoint-120/tokenizer.model +3 -0
- checkpoint-120/tokenizer_config.json +62 -0
checkpoint-10/README.md
ADDED
|
@@ -0,0 +1,202 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: peft
|
| 3 |
+
base_model: teknium/OpenHermes-2.5-Mistral-7B
|
| 4 |
+
---
|
| 5 |
+
|
| 6 |
+
# Model Card for Model ID
|
| 7 |
+
|
| 8 |
+
<!-- Provide a quick summary of what the model is/does. -->
|
| 9 |
+
|
| 10 |
+
|
| 11 |
+
|
| 12 |
+
## Model Details
|
| 13 |
+
|
| 14 |
+
### Model Description
|
| 15 |
+
|
| 16 |
+
<!-- Provide a longer summary of what this model is. -->
|
| 17 |
+
|
| 18 |
+
|
| 19 |
+
|
| 20 |
+
- **Developed by:** [More Information Needed]
|
| 21 |
+
- **Funded by [optional]:** [More Information Needed]
|
| 22 |
+
- **Shared by [optional]:** [More Information Needed]
|
| 23 |
+
- **Model type:** [More Information Needed]
|
| 24 |
+
- **Language(s) (NLP):** [More Information Needed]
|
| 25 |
+
- **License:** [More Information Needed]
|
| 26 |
+
- **Finetuned from model [optional]:** [More Information Needed]
|
| 27 |
+
|
| 28 |
+
### Model Sources [optional]
|
| 29 |
+
|
| 30 |
+
<!-- Provide the basic links for the model. -->
|
| 31 |
+
|
| 32 |
+
- **Repository:** [More Information Needed]
|
| 33 |
+
- **Paper [optional]:** [More Information Needed]
|
| 34 |
+
- **Demo [optional]:** [More Information Needed]
|
| 35 |
+
|
| 36 |
+
## Uses
|
| 37 |
+
|
| 38 |
+
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
|
| 39 |
+
|
| 40 |
+
### Direct Use
|
| 41 |
+
|
| 42 |
+
<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
|
| 43 |
+
|
| 44 |
+
[More Information Needed]
|
| 45 |
+
|
| 46 |
+
### Downstream Use [optional]
|
| 47 |
+
|
| 48 |
+
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
|
| 49 |
+
|
| 50 |
+
[More Information Needed]
|
| 51 |
+
|
| 52 |
+
### Out-of-Scope Use
|
| 53 |
+
|
| 54 |
+
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
|
| 55 |
+
|
| 56 |
+
[More Information Needed]
|
| 57 |
+
|
| 58 |
+
## Bias, Risks, and Limitations
|
| 59 |
+
|
| 60 |
+
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
|
| 61 |
+
|
| 62 |
+
[More Information Needed]
|
| 63 |
+
|
| 64 |
+
### Recommendations
|
| 65 |
+
|
| 66 |
+
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
|
| 67 |
+
|
| 68 |
+
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
|
| 69 |
+
|
| 70 |
+
## How to Get Started with the Model
|
| 71 |
+
|
| 72 |
+
Use the code below to get started with the model.
|
| 73 |
+
|
| 74 |
+
[More Information Needed]
|
| 75 |
+
|
| 76 |
+
## Training Details
|
| 77 |
+
|
| 78 |
+
### Training Data
|
| 79 |
+
|
| 80 |
+
<!-- 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. -->
|
| 81 |
+
|
| 82 |
+
[More Information Needed]
|
| 83 |
+
|
| 84 |
+
### Training Procedure
|
| 85 |
+
|
| 86 |
+
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
|
| 87 |
+
|
| 88 |
+
#### Preprocessing [optional]
|
| 89 |
+
|
| 90 |
+
[More Information Needed]
|
| 91 |
+
|
| 92 |
+
|
| 93 |
+
#### Training Hyperparameters
|
| 94 |
+
|
| 95 |
+
- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
|
| 96 |
+
|
| 97 |
+
#### Speeds, Sizes, Times [optional]
|
| 98 |
+
|
| 99 |
+
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
|
| 100 |
+
|
| 101 |
+
[More Information Needed]
|
| 102 |
+
|
| 103 |
+
## Evaluation
|
| 104 |
+
|
| 105 |
+
<!-- This section describes the evaluation protocols and provides the results. -->
|
| 106 |
+
|
| 107 |
+
### Testing Data, Factors & Metrics
|
| 108 |
+
|
| 109 |
+
#### Testing Data
|
| 110 |
+
|
| 111 |
+
<!-- This should link to a Dataset Card if possible. -->
|
| 112 |
+
|
| 113 |
+
[More Information Needed]
|
| 114 |
+
|
| 115 |
+
#### Factors
|
| 116 |
+
|
| 117 |
+
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
|
| 118 |
+
|
| 119 |
+
[More Information Needed]
|
| 120 |
+
|
| 121 |
+
#### Metrics
|
| 122 |
+
|
| 123 |
+
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
|
| 124 |
+
|
| 125 |
+
[More Information Needed]
|
| 126 |
+
|
| 127 |
+
### Results
|
| 128 |
+
|
| 129 |
+
[More Information Needed]
|
| 130 |
+
|
| 131 |
+
#### Summary
|
| 132 |
+
|
| 133 |
+
|
| 134 |
+
|
| 135 |
+
## Model Examination [optional]
|
| 136 |
+
|
| 137 |
+
<!-- Relevant interpretability work for the model goes here -->
|
| 138 |
+
|
| 139 |
+
[More Information Needed]
|
| 140 |
+
|
| 141 |
+
## Environmental Impact
|
| 142 |
+
|
| 143 |
+
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
|
| 144 |
+
|
| 145 |
+
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).
|
| 146 |
+
|
| 147 |
+
- **Hardware Type:** [More Information Needed]
|
| 148 |
+
- **Hours used:** [More Information Needed]
|
| 149 |
+
- **Cloud Provider:** [More Information Needed]
|
| 150 |
+
- **Compute Region:** [More Information Needed]
|
| 151 |
+
- **Carbon Emitted:** [More Information Needed]
|
| 152 |
+
|
| 153 |
+
## Technical Specifications [optional]
|
| 154 |
+
|
| 155 |
+
### Model Architecture and Objective
|
| 156 |
+
|
| 157 |
+
[More Information Needed]
|
| 158 |
+
|
| 159 |
+
### Compute Infrastructure
|
| 160 |
+
|
| 161 |
+
[More Information Needed]
|
| 162 |
+
|
| 163 |
+
#### Hardware
|
| 164 |
+
|
| 165 |
+
[More Information Needed]
|
| 166 |
+
|
| 167 |
+
#### Software
|
| 168 |
+
|
| 169 |
+
[More Information Needed]
|
| 170 |
+
|
| 171 |
+
## Citation [optional]
|
| 172 |
+
|
| 173 |
+
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
|
| 174 |
+
|
| 175 |
+
**BibTeX:**
|
| 176 |
+
|
| 177 |
+
[More Information Needed]
|
| 178 |
+
|
| 179 |
+
**APA:**
|
| 180 |
+
|
| 181 |
+
[More Information Needed]
|
| 182 |
+
|
| 183 |
+
## Glossary [optional]
|
| 184 |
+
|
| 185 |
+
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
|
| 186 |
+
|
| 187 |
+
[More Information Needed]
|
| 188 |
+
|
| 189 |
+
## More Information [optional]
|
| 190 |
+
|
| 191 |
+
[More Information Needed]
|
| 192 |
+
|
| 193 |
+
## Model Card Authors [optional]
|
| 194 |
+
|
| 195 |
+
[More Information Needed]
|
| 196 |
+
|
| 197 |
+
## Model Card Contact
|
| 198 |
+
|
| 199 |
+
[More Information Needed]
|
| 200 |
+
### Framework versions
|
| 201 |
+
|
| 202 |
+
- PEFT 0.11.1
|
checkpoint-10/adapter_config.json
ADDED
|
@@ -0,0 +1,34 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"alpha_pattern": {},
|
| 3 |
+
"auto_mapping": null,
|
| 4 |
+
"base_model_name_or_path": "teknium/OpenHermes-2.5-Mistral-7B",
|
| 5 |
+
"bias": "none",
|
| 6 |
+
"fan_in_fan_out": false,
|
| 7 |
+
"inference_mode": true,
|
| 8 |
+
"init_lora_weights": true,
|
| 9 |
+
"layer_replication": null,
|
| 10 |
+
"layers_pattern": null,
|
| 11 |
+
"layers_to_transform": null,
|
| 12 |
+
"loftq_config": {},
|
| 13 |
+
"lora_alpha": 16,
|
| 14 |
+
"lora_dropout": 0.05,
|
| 15 |
+
"megatron_config": null,
|
| 16 |
+
"megatron_core": "megatron.core",
|
| 17 |
+
"modules_to_save": null,
|
| 18 |
+
"peft_type": "LORA",
|
| 19 |
+
"r": 16,
|
| 20 |
+
"rank_pattern": {},
|
| 21 |
+
"revision": null,
|
| 22 |
+
"target_modules": [
|
| 23 |
+
"o_proj",
|
| 24 |
+
"k_proj",
|
| 25 |
+
"gate_proj",
|
| 26 |
+
"down_proj",
|
| 27 |
+
"up_proj",
|
| 28 |
+
"q_proj",
|
| 29 |
+
"v_proj"
|
| 30 |
+
],
|
| 31 |
+
"task_type": "CAUSAL_LM",
|
| 32 |
+
"use_dora": false,
|
| 33 |
+
"use_rslora": false
|
| 34 |
+
}
|
checkpoint-10/adapter_model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9caaf22c145d1909979e6d3e771619c4e8e11f06897e1b276cf278b0db8c5b34
|
| 3 |
+
size 167832240
|
checkpoint-10/added_tokens.json
ADDED
|
@@ -0,0 +1,4 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"<|im_end|>": 32000,
|
| 3 |
+
"<|im_start|>": 32001
|
| 4 |
+
}
|
checkpoint-10/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cc4a7e00b8a51259bc0834b65664dd57222ec87dfedf6ddc77e290d2ccca33b2
|
| 3 |
+
size 335928658
|
checkpoint-10/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:386fcc8cc1089aade9450d86fb239ea3483f455fd2d78d8378645feecfec9d69
|
| 3 |
+
size 14244
|
checkpoint-10/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5c4f545e4ff7c8846960748836121a95ae234b27aebe973fd422fbc71be8304a
|
| 3 |
+
size 1064
|
checkpoint-10/special_tokens_map.json
ADDED
|
@@ -0,0 +1,24 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": {
|
| 3 |
+
"content": "<s>",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": false,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"eos_token": {
|
| 10 |
+
"content": "<|im_end|>",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": false,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"pad_token": "<|im_end|>",
|
| 17 |
+
"unk_token": {
|
| 18 |
+
"content": "<unk>",
|
| 19 |
+
"lstrip": false,
|
| 20 |
+
"normalized": false,
|
| 21 |
+
"rstrip": false,
|
| 22 |
+
"single_word": false
|
| 23 |
+
}
|
| 24 |
+
}
|
checkpoint-10/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-10/tokenizer.model
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:dadfd56d766715c61d2ef780a525ab43b8e6da4de6865bda3d95fdef5e134055
|
| 3 |
+
size 493443
|
checkpoint-10/tokenizer_config.json
ADDED
|
@@ -0,0 +1,62 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_bos_token": true,
|
| 3 |
+
"add_eos_token": false,
|
| 4 |
+
"add_prefix_space": true,
|
| 5 |
+
"added_tokens_decoder": {
|
| 6 |
+
"0": {
|
| 7 |
+
"content": "<unk>",
|
| 8 |
+
"lstrip": false,
|
| 9 |
+
"normalized": false,
|
| 10 |
+
"rstrip": false,
|
| 11 |
+
"single_word": false,
|
| 12 |
+
"special": true
|
| 13 |
+
},
|
| 14 |
+
"1": {
|
| 15 |
+
"content": "<s>",
|
| 16 |
+
"lstrip": false,
|
| 17 |
+
"normalized": false,
|
| 18 |
+
"rstrip": false,
|
| 19 |
+
"single_word": false,
|
| 20 |
+
"special": true
|
| 21 |
+
},
|
| 22 |
+
"2": {
|
| 23 |
+
"content": "</s>",
|
| 24 |
+
"lstrip": false,
|
| 25 |
+
"normalized": false,
|
| 26 |
+
"rstrip": false,
|
| 27 |
+
"single_word": false,
|
| 28 |
+
"special": true
|
| 29 |
+
},
|
| 30 |
+
"32000": {
|
| 31 |
+
"content": "<|im_end|>",
|
| 32 |
+
"lstrip": false,
|
| 33 |
+
"normalized": false,
|
| 34 |
+
"rstrip": false,
|
| 35 |
+
"single_word": false,
|
| 36 |
+
"special": true
|
| 37 |
+
},
|
| 38 |
+
"32001": {
|
| 39 |
+
"content": "<|im_start|>",
|
| 40 |
+
"lstrip": false,
|
| 41 |
+
"normalized": false,
|
| 42 |
+
"rstrip": false,
|
| 43 |
+
"single_word": false,
|
| 44 |
+
"special": true
|
| 45 |
+
}
|
| 46 |
+
},
|
| 47 |
+
"additional_special_tokens": [],
|
| 48 |
+
"bos_token": "<s>",
|
| 49 |
+
"chat_template": "{% for message in messages %}{{'<|im_start|>' + message['role'] + '\n' + message['content'] + '<|im_end|>' + '\n'}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant\n' }}{% endif %}",
|
| 50 |
+
"clean_up_tokenization_spaces": false,
|
| 51 |
+
"eos_token": "<|im_end|>",
|
| 52 |
+
"legacy": true,
|
| 53 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 54 |
+
"pad_token": "<|im_end|>",
|
| 55 |
+
"sp_model_kwargs": {},
|
| 56 |
+
"spaces_between_special_tokens": false,
|
| 57 |
+
"tokenizer_class": "LlamaTokenizer",
|
| 58 |
+
"trust_remote_code": false,
|
| 59 |
+
"unk_token": "<unk>",
|
| 60 |
+
"use_default_system_prompt": true,
|
| 61 |
+
"use_fast": true
|
| 62 |
+
}
|
checkpoint-10/trainer_state.json
ADDED
|
@@ -0,0 +1,183 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.00034021127119941485,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 10,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 3.4021127119941484e-05,
|
| 13 |
+
"grad_norm": 33.091712951660156,
|
| 14 |
+
"learning_rate": 1.0000000000000002e-06,
|
| 15 |
+
"logits/chosen": -3.133824586868286,
|
| 16 |
+
"logits/rejected": -3.1356313228607178,
|
| 17 |
+
"logps/chosen": -317.8847961425781,
|
| 18 |
+
"logps/rejected": -306.8866271972656,
|
| 19 |
+
"loss": 0.8544,
|
| 20 |
+
"rewards/accuracies": 0.75,
|
| 21 |
+
"rewards/chosen": -0.10576057434082031,
|
| 22 |
+
"rewards/margins": 0.2207910716533661,
|
| 23 |
+
"rewards/rejected": -0.3265516459941864,
|
| 24 |
+
"step": 1
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 6.804225423988297e-05,
|
| 28 |
+
"grad_norm": 38.49541473388672,
|
| 29 |
+
"learning_rate": 2.0000000000000003e-06,
|
| 30 |
+
"logits/chosen": -3.1433358192443848,
|
| 31 |
+
"logits/rejected": -3.1124918460845947,
|
| 32 |
+
"logps/chosen": -212.93820190429688,
|
| 33 |
+
"logps/rejected": -198.22511291503906,
|
| 34 |
+
"loss": 1.02,
|
| 35 |
+
"rewards/accuracies": 0.5,
|
| 36 |
+
"rewards/chosen": -0.28638574481010437,
|
| 37 |
+
"rewards/margins": -0.37813109159469604,
|
| 38 |
+
"rewards/rejected": 0.0917452871799469,
|
| 39 |
+
"step": 2
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.00010206338135982445,
|
| 43 |
+
"grad_norm": 27.203516006469727,
|
| 44 |
+
"learning_rate": 3e-06,
|
| 45 |
+
"logits/chosen": -3.2104451656341553,
|
| 46 |
+
"logits/rejected": -3.2103281021118164,
|
| 47 |
+
"logps/chosen": -235.89254760742188,
|
| 48 |
+
"logps/rejected": -206.0758514404297,
|
| 49 |
+
"loss": 0.605,
|
| 50 |
+
"rewards/accuracies": 0.75,
|
| 51 |
+
"rewards/chosen": 0.40017586946487427,
|
| 52 |
+
"rewards/margins": 0.24887371063232422,
|
| 53 |
+
"rewards/rejected": 0.15130215883255005,
|
| 54 |
+
"step": 3
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 0.00013608450847976594,
|
| 58 |
+
"grad_norm": 26.42736053466797,
|
| 59 |
+
"learning_rate": 4.000000000000001e-06,
|
| 60 |
+
"logits/chosen": -3.0741634368896484,
|
| 61 |
+
"logits/rejected": -2.953117847442627,
|
| 62 |
+
"logps/chosen": -428.7364196777344,
|
| 63 |
+
"logps/rejected": -192.357666015625,
|
| 64 |
+
"loss": 0.4914,
|
| 65 |
+
"rewards/accuracies": 0.75,
|
| 66 |
+
"rewards/chosen": 0.4332427978515625,
|
| 67 |
+
"rewards/margins": 0.5804362893104553,
|
| 68 |
+
"rewards/rejected": -0.1471935510635376,
|
| 69 |
+
"step": 4
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 0.00017010563559970743,
|
| 73 |
+
"grad_norm": 35.623634338378906,
|
| 74 |
+
"learning_rate": 5e-06,
|
| 75 |
+
"logits/chosen": -2.985285758972168,
|
| 76 |
+
"logits/rejected": -2.8854246139526367,
|
| 77 |
+
"logps/chosen": -342.38604736328125,
|
| 78 |
+
"logps/rejected": -211.78237915039062,
|
| 79 |
+
"loss": 0.8273,
|
| 80 |
+
"rewards/accuracies": 0.25,
|
| 81 |
+
"rewards/chosen": -0.3475755751132965,
|
| 82 |
+
"rewards/margins": -0.19371166825294495,
|
| 83 |
+
"rewards/rejected": -0.15386392176151276,
|
| 84 |
+
"step": 5
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 0.0002041267627196489,
|
| 88 |
+
"grad_norm": 31.70409393310547,
|
| 89 |
+
"learning_rate": 6e-06,
|
| 90 |
+
"logits/chosen": -3.074589967727661,
|
| 91 |
+
"logits/rejected": -2.85581636428833,
|
| 92 |
+
"logps/chosen": -593.6435546875,
|
| 93 |
+
"logps/rejected": -259.5284729003906,
|
| 94 |
+
"loss": 0.5473,
|
| 95 |
+
"rewards/accuracies": 0.75,
|
| 96 |
+
"rewards/chosen": 0.2495414763689041,
|
| 97 |
+
"rewards/margins": 0.35888367891311646,
|
| 98 |
+
"rewards/rejected": -0.10934218764305115,
|
| 99 |
+
"step": 6
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 0.00023814788983959038,
|
| 103 |
+
"grad_norm": 32.41214370727539,
|
| 104 |
+
"learning_rate": 7.000000000000001e-06,
|
| 105 |
+
"logits/chosen": -2.976675033569336,
|
| 106 |
+
"logits/rejected": -2.978133201599121,
|
| 107 |
+
"logps/chosen": -218.11688232421875,
|
| 108 |
+
"logps/rejected": -231.7698974609375,
|
| 109 |
+
"loss": 0.7124,
|
| 110 |
+
"rewards/accuracies": 0.25,
|
| 111 |
+
"rewards/chosen": -0.08671779930591583,
|
| 112 |
+
"rewards/margins": 0.21950224041938782,
|
| 113 |
+
"rewards/rejected": -0.30622005462646484,
|
| 114 |
+
"step": 7
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.00027216901695953187,
|
| 118 |
+
"grad_norm": 27.637351989746094,
|
| 119 |
+
"learning_rate": 8.000000000000001e-06,
|
| 120 |
+
"logits/chosen": -2.997600793838501,
|
| 121 |
+
"logits/rejected": -3.1078853607177734,
|
| 122 |
+
"logps/chosen": -165.71444702148438,
|
| 123 |
+
"logps/rejected": -335.5938720703125,
|
| 124 |
+
"loss": 0.6085,
|
| 125 |
+
"rewards/accuracies": 0.75,
|
| 126 |
+
"rewards/chosen": -0.057393088936805725,
|
| 127 |
+
"rewards/margins": 0.34109365940093994,
|
| 128 |
+
"rewards/rejected": -0.3984867036342621,
|
| 129 |
+
"step": 8
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.00030619014407947336,
|
| 133 |
+
"grad_norm": 32.246559143066406,
|
| 134 |
+
"learning_rate": 9e-06,
|
| 135 |
+
"logits/chosen": -3.0760390758514404,
|
| 136 |
+
"logits/rejected": -2.938197135925293,
|
| 137 |
+
"logps/chosen": -463.1333312988281,
|
| 138 |
+
"logps/rejected": -173.99951171875,
|
| 139 |
+
"loss": 0.6398,
|
| 140 |
+
"rewards/accuracies": 0.5,
|
| 141 |
+
"rewards/chosen": 0.5581258535385132,
|
| 142 |
+
"rewards/margins": 0.7653559446334839,
|
| 143 |
+
"rewards/rejected": -0.20723000168800354,
|
| 144 |
+
"step": 9
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 0.00034021127119941485,
|
| 148 |
+
"grad_norm": 39.84626388549805,
|
| 149 |
+
"learning_rate": 1e-05,
|
| 150 |
+
"logits/chosen": -3.192596673965454,
|
| 151 |
+
"logits/rejected": -2.9751501083374023,
|
| 152 |
+
"logps/chosen": -473.14532470703125,
|
| 153 |
+
"logps/rejected": -110.90155029296875,
|
| 154 |
+
"loss": 0.9794,
|
| 155 |
+
"rewards/accuracies": 0.25,
|
| 156 |
+
"rewards/chosen": -0.3494076132774353,
|
| 157 |
+
"rewards/margins": -0.31546899676322937,
|
| 158 |
+
"rewards/rejected": -0.03393859416246414,
|
| 159 |
+
"step": 10
|
| 160 |
+
}
|
| 161 |
+
],
|
| 162 |
+
"logging_steps": 1,
|
| 163 |
+
"max_steps": 500,
|
| 164 |
+
"num_input_tokens_seen": 0,
|
| 165 |
+
"num_train_epochs": 1,
|
| 166 |
+
"save_steps": 10,
|
| 167 |
+
"stateful_callbacks": {
|
| 168 |
+
"TrainerControl": {
|
| 169 |
+
"args": {
|
| 170 |
+
"should_epoch_stop": false,
|
| 171 |
+
"should_evaluate": false,
|
| 172 |
+
"should_log": false,
|
| 173 |
+
"should_save": true,
|
| 174 |
+
"should_training_stop": false
|
| 175 |
+
},
|
| 176 |
+
"attributes": {}
|
| 177 |
+
}
|
| 178 |
+
},
|
| 179 |
+
"total_flos": 0.0,
|
| 180 |
+
"train_batch_size": 1,
|
| 181 |
+
"trial_name": null,
|
| 182 |
+
"trial_params": null
|
| 183 |
+
}
|
checkpoint-10/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2aa185d46c9d5abcf13df6d50376870fd603b11ffa1fb959ec10e0609dfec047
|
| 3 |
+
size 5880
|
checkpoint-100/README.md
ADDED
|
@@ -0,0 +1,202 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: peft
|
| 3 |
+
base_model: teknium/OpenHermes-2.5-Mistral-7B
|
| 4 |
+
---
|
| 5 |
+
|
| 6 |
+
# Model Card for Model ID
|
| 7 |
+
|
| 8 |
+
<!-- Provide a quick summary of what the model is/does. -->
|
| 9 |
+
|
| 10 |
+
|
| 11 |
+
|
| 12 |
+
## Model Details
|
| 13 |
+
|
| 14 |
+
### Model Description
|
| 15 |
+
|
| 16 |
+
<!-- Provide a longer summary of what this model is. -->
|
| 17 |
+
|
| 18 |
+
|
| 19 |
+
|
| 20 |
+
- **Developed by:** [More Information Needed]
|
| 21 |
+
- **Funded by [optional]:** [More Information Needed]
|
| 22 |
+
- **Shared by [optional]:** [More Information Needed]
|
| 23 |
+
- **Model type:** [More Information Needed]
|
| 24 |
+
- **Language(s) (NLP):** [More Information Needed]
|
| 25 |
+
- **License:** [More Information Needed]
|
| 26 |
+
- **Finetuned from model [optional]:** [More Information Needed]
|
| 27 |
+
|
| 28 |
+
### Model Sources [optional]
|
| 29 |
+
|
| 30 |
+
<!-- Provide the basic links for the model. -->
|
| 31 |
+
|
| 32 |
+
- **Repository:** [More Information Needed]
|
| 33 |
+
- **Paper [optional]:** [More Information Needed]
|
| 34 |
+
- **Demo [optional]:** [More Information Needed]
|
| 35 |
+
|
| 36 |
+
## Uses
|
| 37 |
+
|
| 38 |
+
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
|
| 39 |
+
|
| 40 |
+
### Direct Use
|
| 41 |
+
|
| 42 |
+
<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
|
| 43 |
+
|
| 44 |
+
[More Information Needed]
|
| 45 |
+
|
| 46 |
+
### Downstream Use [optional]
|
| 47 |
+
|
| 48 |
+
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
|
| 49 |
+
|
| 50 |
+
[More Information Needed]
|
| 51 |
+
|
| 52 |
+
### Out-of-Scope Use
|
| 53 |
+
|
| 54 |
+
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
|
| 55 |
+
|
| 56 |
+
[More Information Needed]
|
| 57 |
+
|
| 58 |
+
## Bias, Risks, and Limitations
|
| 59 |
+
|
| 60 |
+
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
|
| 61 |
+
|
| 62 |
+
[More Information Needed]
|
| 63 |
+
|
| 64 |
+
### Recommendations
|
| 65 |
+
|
| 66 |
+
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
|
| 67 |
+
|
| 68 |
+
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
|
| 69 |
+
|
| 70 |
+
## How to Get Started with the Model
|
| 71 |
+
|
| 72 |
+
Use the code below to get started with the model.
|
| 73 |
+
|
| 74 |
+
[More Information Needed]
|
| 75 |
+
|
| 76 |
+
## Training Details
|
| 77 |
+
|
| 78 |
+
### Training Data
|
| 79 |
+
|
| 80 |
+
<!-- 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. -->
|
| 81 |
+
|
| 82 |
+
[More Information Needed]
|
| 83 |
+
|
| 84 |
+
### Training Procedure
|
| 85 |
+
|
| 86 |
+
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
|
| 87 |
+
|
| 88 |
+
#### Preprocessing [optional]
|
| 89 |
+
|
| 90 |
+
[More Information Needed]
|
| 91 |
+
|
| 92 |
+
|
| 93 |
+
#### Training Hyperparameters
|
| 94 |
+
|
| 95 |
+
- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
|
| 96 |
+
|
| 97 |
+
#### Speeds, Sizes, Times [optional]
|
| 98 |
+
|
| 99 |
+
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
|
| 100 |
+
|
| 101 |
+
[More Information Needed]
|
| 102 |
+
|
| 103 |
+
## Evaluation
|
| 104 |
+
|
| 105 |
+
<!-- This section describes the evaluation protocols and provides the results. -->
|
| 106 |
+
|
| 107 |
+
### Testing Data, Factors & Metrics
|
| 108 |
+
|
| 109 |
+
#### Testing Data
|
| 110 |
+
|
| 111 |
+
<!-- This should link to a Dataset Card if possible. -->
|
| 112 |
+
|
| 113 |
+
[More Information Needed]
|
| 114 |
+
|
| 115 |
+
#### Factors
|
| 116 |
+
|
| 117 |
+
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
|
| 118 |
+
|
| 119 |
+
[More Information Needed]
|
| 120 |
+
|
| 121 |
+
#### Metrics
|
| 122 |
+
|
| 123 |
+
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
|
| 124 |
+
|
| 125 |
+
[More Information Needed]
|
| 126 |
+
|
| 127 |
+
### Results
|
| 128 |
+
|
| 129 |
+
[More Information Needed]
|
| 130 |
+
|
| 131 |
+
#### Summary
|
| 132 |
+
|
| 133 |
+
|
| 134 |
+
|
| 135 |
+
## Model Examination [optional]
|
| 136 |
+
|
| 137 |
+
<!-- Relevant interpretability work for the model goes here -->
|
| 138 |
+
|
| 139 |
+
[More Information Needed]
|
| 140 |
+
|
| 141 |
+
## Environmental Impact
|
| 142 |
+
|
| 143 |
+
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
|
| 144 |
+
|
| 145 |
+
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).
|
| 146 |
+
|
| 147 |
+
- **Hardware Type:** [More Information Needed]
|
| 148 |
+
- **Hours used:** [More Information Needed]
|
| 149 |
+
- **Cloud Provider:** [More Information Needed]
|
| 150 |
+
- **Compute Region:** [More Information Needed]
|
| 151 |
+
- **Carbon Emitted:** [More Information Needed]
|
| 152 |
+
|
| 153 |
+
## Technical Specifications [optional]
|
| 154 |
+
|
| 155 |
+
### Model Architecture and Objective
|
| 156 |
+
|
| 157 |
+
[More Information Needed]
|
| 158 |
+
|
| 159 |
+
### Compute Infrastructure
|
| 160 |
+
|
| 161 |
+
[More Information Needed]
|
| 162 |
+
|
| 163 |
+
#### Hardware
|
| 164 |
+
|
| 165 |
+
[More Information Needed]
|
| 166 |
+
|
| 167 |
+
#### Software
|
| 168 |
+
|
| 169 |
+
[More Information Needed]
|
| 170 |
+
|
| 171 |
+
## Citation [optional]
|
| 172 |
+
|
| 173 |
+
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
|
| 174 |
+
|
| 175 |
+
**BibTeX:**
|
| 176 |
+
|
| 177 |
+
[More Information Needed]
|
| 178 |
+
|
| 179 |
+
**APA:**
|
| 180 |
+
|
| 181 |
+
[More Information Needed]
|
| 182 |
+
|
| 183 |
+
## Glossary [optional]
|
| 184 |
+
|
| 185 |
+
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
|
| 186 |
+
|
| 187 |
+
[More Information Needed]
|
| 188 |
+
|
| 189 |
+
## More Information [optional]
|
| 190 |
+
|
| 191 |
+
[More Information Needed]
|
| 192 |
+
|
| 193 |
+
## Model Card Authors [optional]
|
| 194 |
+
|
| 195 |
+
[More Information Needed]
|
| 196 |
+
|
| 197 |
+
## Model Card Contact
|
| 198 |
+
|
| 199 |
+
[More Information Needed]
|
| 200 |
+
### Framework versions
|
| 201 |
+
|
| 202 |
+
- PEFT 0.11.1
|
checkpoint-100/adapter_config.json
ADDED
|
@@ -0,0 +1,34 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"alpha_pattern": {},
|
| 3 |
+
"auto_mapping": null,
|
| 4 |
+
"base_model_name_or_path": "teknium/OpenHermes-2.5-Mistral-7B",
|
| 5 |
+
"bias": "none",
|
| 6 |
+
"fan_in_fan_out": false,
|
| 7 |
+
"inference_mode": true,
|
| 8 |
+
"init_lora_weights": true,
|
| 9 |
+
"layer_replication": null,
|
| 10 |
+
"layers_pattern": null,
|
| 11 |
+
"layers_to_transform": null,
|
| 12 |
+
"loftq_config": {},
|
| 13 |
+
"lora_alpha": 16,
|
| 14 |
+
"lora_dropout": 0.05,
|
| 15 |
+
"megatron_config": null,
|
| 16 |
+
"megatron_core": "megatron.core",
|
| 17 |
+
"modules_to_save": null,
|
| 18 |
+
"peft_type": "LORA",
|
| 19 |
+
"r": 16,
|
| 20 |
+
"rank_pattern": {},
|
| 21 |
+
"revision": null,
|
| 22 |
+
"target_modules": [
|
| 23 |
+
"o_proj",
|
| 24 |
+
"k_proj",
|
| 25 |
+
"gate_proj",
|
| 26 |
+
"down_proj",
|
| 27 |
+
"up_proj",
|
| 28 |
+
"q_proj",
|
| 29 |
+
"v_proj"
|
| 30 |
+
],
|
| 31 |
+
"task_type": "CAUSAL_LM",
|
| 32 |
+
"use_dora": false,
|
| 33 |
+
"use_rslora": false
|
| 34 |
+
}
|
checkpoint-100/adapter_model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:27257a8647c2a6ab40199cc0b1ebfb07fd9da22252e2a897817850ef9e70ec3a
|
| 3 |
+
size 167832240
|
checkpoint-100/added_tokens.json
ADDED
|
@@ -0,0 +1,4 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"<|im_end|>": 32000,
|
| 3 |
+
"<|im_start|>": 32001
|
| 4 |
+
}
|
checkpoint-100/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ef063f1392a318d66624273b1979415ca198903ed29a548ab70efecea63d1bed
|
| 3 |
+
size 335928658
|
checkpoint-100/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:386fcc8cc1089aade9450d86fb239ea3483f455fd2d78d8378645feecfec9d69
|
| 3 |
+
size 14244
|
checkpoint-100/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:62feb8b21bb232bc70eb5719effdc243c4b613a4b57b695112867ac36e62f413
|
| 3 |
+
size 1064
|
checkpoint-100/special_tokens_map.json
ADDED
|
@@ -0,0 +1,24 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": {
|
| 3 |
+
"content": "<s>",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": false,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"eos_token": {
|
| 10 |
+
"content": "<|im_end|>",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": false,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"pad_token": "<|im_end|>",
|
| 17 |
+
"unk_token": {
|
| 18 |
+
"content": "<unk>",
|
| 19 |
+
"lstrip": false,
|
| 20 |
+
"normalized": false,
|
| 21 |
+
"rstrip": false,
|
| 22 |
+
"single_word": false
|
| 23 |
+
}
|
| 24 |
+
}
|
checkpoint-100/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-100/tokenizer.model
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:dadfd56d766715c61d2ef780a525ab43b8e6da4de6865bda3d95fdef5e134055
|
| 3 |
+
size 493443
|
checkpoint-100/tokenizer_config.json
ADDED
|
@@ -0,0 +1,62 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_bos_token": true,
|
| 3 |
+
"add_eos_token": false,
|
| 4 |
+
"add_prefix_space": true,
|
| 5 |
+
"added_tokens_decoder": {
|
| 6 |
+
"0": {
|
| 7 |
+
"content": "<unk>",
|
| 8 |
+
"lstrip": false,
|
| 9 |
+
"normalized": false,
|
| 10 |
+
"rstrip": false,
|
| 11 |
+
"single_word": false,
|
| 12 |
+
"special": true
|
| 13 |
+
},
|
| 14 |
+
"1": {
|
| 15 |
+
"content": "<s>",
|
| 16 |
+
"lstrip": false,
|
| 17 |
+
"normalized": false,
|
| 18 |
+
"rstrip": false,
|
| 19 |
+
"single_word": false,
|
| 20 |
+
"special": true
|
| 21 |
+
},
|
| 22 |
+
"2": {
|
| 23 |
+
"content": "</s>",
|
| 24 |
+
"lstrip": false,
|
| 25 |
+
"normalized": false,
|
| 26 |
+
"rstrip": false,
|
| 27 |
+
"single_word": false,
|
| 28 |
+
"special": true
|
| 29 |
+
},
|
| 30 |
+
"32000": {
|
| 31 |
+
"content": "<|im_end|>",
|
| 32 |
+
"lstrip": false,
|
| 33 |
+
"normalized": false,
|
| 34 |
+
"rstrip": false,
|
| 35 |
+
"single_word": false,
|
| 36 |
+
"special": true
|
| 37 |
+
},
|
| 38 |
+
"32001": {
|
| 39 |
+
"content": "<|im_start|>",
|
| 40 |
+
"lstrip": false,
|
| 41 |
+
"normalized": false,
|
| 42 |
+
"rstrip": false,
|
| 43 |
+
"single_word": false,
|
| 44 |
+
"special": true
|
| 45 |
+
}
|
| 46 |
+
},
|
| 47 |
+
"additional_special_tokens": [],
|
| 48 |
+
"bos_token": "<s>",
|
| 49 |
+
"chat_template": "{% for message in messages %}{{'<|im_start|>' + message['role'] + '\n' + message['content'] + '<|im_end|>' + '\n'}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant\n' }}{% endif %}",
|
| 50 |
+
"clean_up_tokenization_spaces": false,
|
| 51 |
+
"eos_token": "<|im_end|>",
|
| 52 |
+
"legacy": true,
|
| 53 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 54 |
+
"pad_token": "<|im_end|>",
|
| 55 |
+
"sp_model_kwargs": {},
|
| 56 |
+
"spaces_between_special_tokens": false,
|
| 57 |
+
"tokenizer_class": "LlamaTokenizer",
|
| 58 |
+
"trust_remote_code": false,
|
| 59 |
+
"unk_token": "<unk>",
|
| 60 |
+
"use_default_system_prompt": true,
|
| 61 |
+
"use_fast": true
|
| 62 |
+
}
|
checkpoint-100/trainer_state.json
ADDED
|
@@ -0,0 +1,1533 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.003402112711994148,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 100,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 3.4021127119941484e-05,
|
| 13 |
+
"grad_norm": 33.091712951660156,
|
| 14 |
+
"learning_rate": 1.0000000000000002e-06,
|
| 15 |
+
"logits/chosen": -3.133824586868286,
|
| 16 |
+
"logits/rejected": -3.1356313228607178,
|
| 17 |
+
"logps/chosen": -317.8847961425781,
|
| 18 |
+
"logps/rejected": -306.8866271972656,
|
| 19 |
+
"loss": 0.8544,
|
| 20 |
+
"rewards/accuracies": 0.75,
|
| 21 |
+
"rewards/chosen": -0.10576057434082031,
|
| 22 |
+
"rewards/margins": 0.2207910716533661,
|
| 23 |
+
"rewards/rejected": -0.3265516459941864,
|
| 24 |
+
"step": 1
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 6.804225423988297e-05,
|
| 28 |
+
"grad_norm": 38.49541473388672,
|
| 29 |
+
"learning_rate": 2.0000000000000003e-06,
|
| 30 |
+
"logits/chosen": -3.1433358192443848,
|
| 31 |
+
"logits/rejected": -3.1124918460845947,
|
| 32 |
+
"logps/chosen": -212.93820190429688,
|
| 33 |
+
"logps/rejected": -198.22511291503906,
|
| 34 |
+
"loss": 1.02,
|
| 35 |
+
"rewards/accuracies": 0.5,
|
| 36 |
+
"rewards/chosen": -0.28638574481010437,
|
| 37 |
+
"rewards/margins": -0.37813109159469604,
|
| 38 |
+
"rewards/rejected": 0.0917452871799469,
|
| 39 |
+
"step": 2
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.00010206338135982445,
|
| 43 |
+
"grad_norm": 27.203516006469727,
|
| 44 |
+
"learning_rate": 3e-06,
|
| 45 |
+
"logits/chosen": -3.2104451656341553,
|
| 46 |
+
"logits/rejected": -3.2103281021118164,
|
| 47 |
+
"logps/chosen": -235.89254760742188,
|
| 48 |
+
"logps/rejected": -206.0758514404297,
|
| 49 |
+
"loss": 0.605,
|
| 50 |
+
"rewards/accuracies": 0.75,
|
| 51 |
+
"rewards/chosen": 0.40017586946487427,
|
| 52 |
+
"rewards/margins": 0.24887371063232422,
|
| 53 |
+
"rewards/rejected": 0.15130215883255005,
|
| 54 |
+
"step": 3
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 0.00013608450847976594,
|
| 58 |
+
"grad_norm": 26.42736053466797,
|
| 59 |
+
"learning_rate": 4.000000000000001e-06,
|
| 60 |
+
"logits/chosen": -3.0741634368896484,
|
| 61 |
+
"logits/rejected": -2.953117847442627,
|
| 62 |
+
"logps/chosen": -428.7364196777344,
|
| 63 |
+
"logps/rejected": -192.357666015625,
|
| 64 |
+
"loss": 0.4914,
|
| 65 |
+
"rewards/accuracies": 0.75,
|
| 66 |
+
"rewards/chosen": 0.4332427978515625,
|
| 67 |
+
"rewards/margins": 0.5804362893104553,
|
| 68 |
+
"rewards/rejected": -0.1471935510635376,
|
| 69 |
+
"step": 4
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 0.00017010563559970743,
|
| 73 |
+
"grad_norm": 35.623634338378906,
|
| 74 |
+
"learning_rate": 5e-06,
|
| 75 |
+
"logits/chosen": -2.985285758972168,
|
| 76 |
+
"logits/rejected": -2.8854246139526367,
|
| 77 |
+
"logps/chosen": -342.38604736328125,
|
| 78 |
+
"logps/rejected": -211.78237915039062,
|
| 79 |
+
"loss": 0.8273,
|
| 80 |
+
"rewards/accuracies": 0.25,
|
| 81 |
+
"rewards/chosen": -0.3475755751132965,
|
| 82 |
+
"rewards/margins": -0.19371166825294495,
|
| 83 |
+
"rewards/rejected": -0.15386392176151276,
|
| 84 |
+
"step": 5
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 0.0002041267627196489,
|
| 88 |
+
"grad_norm": 31.70409393310547,
|
| 89 |
+
"learning_rate": 6e-06,
|
| 90 |
+
"logits/chosen": -3.074589967727661,
|
| 91 |
+
"logits/rejected": -2.85581636428833,
|
| 92 |
+
"logps/chosen": -593.6435546875,
|
| 93 |
+
"logps/rejected": -259.5284729003906,
|
| 94 |
+
"loss": 0.5473,
|
| 95 |
+
"rewards/accuracies": 0.75,
|
| 96 |
+
"rewards/chosen": 0.2495414763689041,
|
| 97 |
+
"rewards/margins": 0.35888367891311646,
|
| 98 |
+
"rewards/rejected": -0.10934218764305115,
|
| 99 |
+
"step": 6
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 0.00023814788983959038,
|
| 103 |
+
"grad_norm": 32.41214370727539,
|
| 104 |
+
"learning_rate": 7.000000000000001e-06,
|
| 105 |
+
"logits/chosen": -2.976675033569336,
|
| 106 |
+
"logits/rejected": -2.978133201599121,
|
| 107 |
+
"logps/chosen": -218.11688232421875,
|
| 108 |
+
"logps/rejected": -231.7698974609375,
|
| 109 |
+
"loss": 0.7124,
|
| 110 |
+
"rewards/accuracies": 0.25,
|
| 111 |
+
"rewards/chosen": -0.08671779930591583,
|
| 112 |
+
"rewards/margins": 0.21950224041938782,
|
| 113 |
+
"rewards/rejected": -0.30622005462646484,
|
| 114 |
+
"step": 7
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.00027216901695953187,
|
| 118 |
+
"grad_norm": 27.637351989746094,
|
| 119 |
+
"learning_rate": 8.000000000000001e-06,
|
| 120 |
+
"logits/chosen": -2.997600793838501,
|
| 121 |
+
"logits/rejected": -3.1078853607177734,
|
| 122 |
+
"logps/chosen": -165.71444702148438,
|
| 123 |
+
"logps/rejected": -335.5938720703125,
|
| 124 |
+
"loss": 0.6085,
|
| 125 |
+
"rewards/accuracies": 0.75,
|
| 126 |
+
"rewards/chosen": -0.057393088936805725,
|
| 127 |
+
"rewards/margins": 0.34109365940093994,
|
| 128 |
+
"rewards/rejected": -0.3984867036342621,
|
| 129 |
+
"step": 8
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.00030619014407947336,
|
| 133 |
+
"grad_norm": 32.246559143066406,
|
| 134 |
+
"learning_rate": 9e-06,
|
| 135 |
+
"logits/chosen": -3.0760390758514404,
|
| 136 |
+
"logits/rejected": -2.938197135925293,
|
| 137 |
+
"logps/chosen": -463.1333312988281,
|
| 138 |
+
"logps/rejected": -173.99951171875,
|
| 139 |
+
"loss": 0.6398,
|
| 140 |
+
"rewards/accuracies": 0.5,
|
| 141 |
+
"rewards/chosen": 0.5581258535385132,
|
| 142 |
+
"rewards/margins": 0.7653559446334839,
|
| 143 |
+
"rewards/rejected": -0.20723000168800354,
|
| 144 |
+
"step": 9
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 0.00034021127119941485,
|
| 148 |
+
"grad_norm": 39.84626388549805,
|
| 149 |
+
"learning_rate": 1e-05,
|
| 150 |
+
"logits/chosen": -3.192596673965454,
|
| 151 |
+
"logits/rejected": -2.9751501083374023,
|
| 152 |
+
"logps/chosen": -473.14532470703125,
|
| 153 |
+
"logps/rejected": -110.90155029296875,
|
| 154 |
+
"loss": 0.9794,
|
| 155 |
+
"rewards/accuracies": 0.25,
|
| 156 |
+
"rewards/chosen": -0.3494076132774353,
|
| 157 |
+
"rewards/margins": -0.31546899676322937,
|
| 158 |
+
"rewards/rejected": -0.03393859416246414,
|
| 159 |
+
"step": 10
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 0.00037423239831935634,
|
| 163 |
+
"grad_norm": 39.444889068603516,
|
| 164 |
+
"learning_rate": 1.1000000000000001e-05,
|
| 165 |
+
"logits/chosen": -3.096883535385132,
|
| 166 |
+
"logits/rejected": -3.043616771697998,
|
| 167 |
+
"logps/chosen": -391.7703552246094,
|
| 168 |
+
"logps/rejected": -271.8331604003906,
|
| 169 |
+
"loss": 0.8029,
|
| 170 |
+
"rewards/accuracies": 0.75,
|
| 171 |
+
"rewards/chosen": 0.6240158081054688,
|
| 172 |
+
"rewards/margins": 0.4651091992855072,
|
| 173 |
+
"rewards/rejected": 0.15890654921531677,
|
| 174 |
+
"step": 11
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 0.0004082535254392978,
|
| 178 |
+
"grad_norm": 40.78356170654297,
|
| 179 |
+
"learning_rate": 1.2e-05,
|
| 180 |
+
"logits/chosen": -3.000021457672119,
|
| 181 |
+
"logits/rejected": -3.0893115997314453,
|
| 182 |
+
"logps/chosen": -255.94607543945312,
|
| 183 |
+
"logps/rejected": -405.3093566894531,
|
| 184 |
+
"loss": 0.6297,
|
| 185 |
+
"rewards/accuracies": 0.5,
|
| 186 |
+
"rewards/chosen": 0.33697396516799927,
|
| 187 |
+
"rewards/margins": 0.3315456509590149,
|
| 188 |
+
"rewards/rejected": 0.005428321659564972,
|
| 189 |
+
"step": 12
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 0.00044227465255923927,
|
| 193 |
+
"grad_norm": 35.69674301147461,
|
| 194 |
+
"learning_rate": 1.3000000000000001e-05,
|
| 195 |
+
"logits/chosen": -3.011014699935913,
|
| 196 |
+
"logits/rejected": -3.0214414596557617,
|
| 197 |
+
"logps/chosen": -223.4480438232422,
|
| 198 |
+
"logps/rejected": -261.62469482421875,
|
| 199 |
+
"loss": 0.815,
|
| 200 |
+
"rewards/accuracies": 0.5,
|
| 201 |
+
"rewards/chosen": 0.7416718006134033,
|
| 202 |
+
"rewards/margins": 0.19626998901367188,
|
| 203 |
+
"rewards/rejected": 0.5454018115997314,
|
| 204 |
+
"step": 13
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 0.00047629577967918076,
|
| 208 |
+
"grad_norm": 32.88405990600586,
|
| 209 |
+
"learning_rate": 1.4000000000000001e-05,
|
| 210 |
+
"logits/chosen": -3.065549373626709,
|
| 211 |
+
"logits/rejected": -3.1094508171081543,
|
| 212 |
+
"logps/chosen": -169.08047485351562,
|
| 213 |
+
"logps/rejected": -227.85162353515625,
|
| 214 |
+
"loss": 0.8115,
|
| 215 |
+
"rewards/accuracies": 0.25,
|
| 216 |
+
"rewards/chosen": 0.37659722566604614,
|
| 217 |
+
"rewards/margins": -0.11390496790409088,
|
| 218 |
+
"rewards/rejected": 0.4905022084712982,
|
| 219 |
+
"step": 14
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.0005103169067991223,
|
| 223 |
+
"grad_norm": 20.918012619018555,
|
| 224 |
+
"learning_rate": 1.5e-05,
|
| 225 |
+
"logits/chosen": -3.045846462249756,
|
| 226 |
+
"logits/rejected": -2.9114990234375,
|
| 227 |
+
"logps/chosen": -573.1253662109375,
|
| 228 |
+
"logps/rejected": -273.9326477050781,
|
| 229 |
+
"loss": 0.3105,
|
| 230 |
+
"rewards/accuracies": 1.0,
|
| 231 |
+
"rewards/chosen": 0.7755428552627563,
|
| 232 |
+
"rewards/margins": 1.082038164138794,
|
| 233 |
+
"rewards/rejected": -0.3064952790737152,
|
| 234 |
+
"step": 15
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.0005443380339190637,
|
| 238 |
+
"grad_norm": 38.434749603271484,
|
| 239 |
+
"learning_rate": 1.6000000000000003e-05,
|
| 240 |
+
"logits/chosen": -3.1250810623168945,
|
| 241 |
+
"logits/rejected": -2.96527099609375,
|
| 242 |
+
"logps/chosen": -396.9795837402344,
|
| 243 |
+
"logps/rejected": -174.92918395996094,
|
| 244 |
+
"loss": 0.9971,
|
| 245 |
+
"rewards/accuracies": 0.5,
|
| 246 |
+
"rewards/chosen": 0.2244594693183899,
|
| 247 |
+
"rewards/margins": -0.37520748376846313,
|
| 248 |
+
"rewards/rejected": 0.5996670126914978,
|
| 249 |
+
"step": 16
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 0.0005783591610390052,
|
| 253 |
+
"grad_norm": 37.71885681152344,
|
| 254 |
+
"learning_rate": 1.7000000000000003e-05,
|
| 255 |
+
"logits/chosen": -2.970050096511841,
|
| 256 |
+
"logits/rejected": -3.0346829891204834,
|
| 257 |
+
"logps/chosen": -236.67071533203125,
|
| 258 |
+
"logps/rejected": -403.0212097167969,
|
| 259 |
+
"loss": 1.0307,
|
| 260 |
+
"rewards/accuracies": 0.5,
|
| 261 |
+
"rewards/chosen": 0.9259212613105774,
|
| 262 |
+
"rewards/margins": -0.23367927968502045,
|
| 263 |
+
"rewards/rejected": 1.1596004962921143,
|
| 264 |
+
"step": 17
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 0.0006123802881589467,
|
| 268 |
+
"grad_norm": 25.964290618896484,
|
| 269 |
+
"learning_rate": 1.8e-05,
|
| 270 |
+
"logits/chosen": -3.0690722465515137,
|
| 271 |
+
"logits/rejected": -3.0348947048187256,
|
| 272 |
+
"logps/chosen": -312.81317138671875,
|
| 273 |
+
"logps/rejected": -259.230712890625,
|
| 274 |
+
"loss": 0.6989,
|
| 275 |
+
"rewards/accuracies": 0.75,
|
| 276 |
+
"rewards/chosen": 0.6348915100097656,
|
| 277 |
+
"rewards/margins": 0.6418477892875671,
|
| 278 |
+
"rewards/rejected": -0.006956290453672409,
|
| 279 |
+
"step": 18
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 0.0006464014152788882,
|
| 283 |
+
"grad_norm": 27.076108932495117,
|
| 284 |
+
"learning_rate": 1.9e-05,
|
| 285 |
+
"logits/chosen": -3.135136842727661,
|
| 286 |
+
"logits/rejected": -2.911625862121582,
|
| 287 |
+
"logps/chosen": -473.7462158203125,
|
| 288 |
+
"logps/rejected": -141.9491729736328,
|
| 289 |
+
"loss": 0.6027,
|
| 290 |
+
"rewards/accuracies": 0.5,
|
| 291 |
+
"rewards/chosen": 0.9424793720245361,
|
| 292 |
+
"rewards/margins": 0.9380611181259155,
|
| 293 |
+
"rewards/rejected": 0.00441819429397583,
|
| 294 |
+
"step": 19
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 0.0006804225423988297,
|
| 298 |
+
"grad_norm": 51.820159912109375,
|
| 299 |
+
"learning_rate": 2e-05,
|
| 300 |
+
"logits/chosen": -2.8403027057647705,
|
| 301 |
+
"logits/rejected": -3.0664803981781006,
|
| 302 |
+
"logps/chosen": -219.79153442382812,
|
| 303 |
+
"logps/rejected": -231.70004272460938,
|
| 304 |
+
"loss": 1.1022,
|
| 305 |
+
"rewards/accuracies": 0.0,
|
| 306 |
+
"rewards/chosen": 0.052263081073760986,
|
| 307 |
+
"rewards/margins": -0.6957410573959351,
|
| 308 |
+
"rewards/rejected": 0.7480041980743408,
|
| 309 |
+
"step": 20
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 0.0007144436695187712,
|
| 313 |
+
"grad_norm": 45.9146614074707,
|
| 314 |
+
"learning_rate": 2.1e-05,
|
| 315 |
+
"logits/chosen": -2.9764764308929443,
|
| 316 |
+
"logits/rejected": -2.9391565322875977,
|
| 317 |
+
"logps/chosen": -458.0147705078125,
|
| 318 |
+
"logps/rejected": -218.37648010253906,
|
| 319 |
+
"loss": 1.0772,
|
| 320 |
+
"rewards/accuracies": 0.5,
|
| 321 |
+
"rewards/chosen": 0.8049196004867554,
|
| 322 |
+
"rewards/margins": -0.448437362909317,
|
| 323 |
+
"rewards/rejected": 1.2533570528030396,
|
| 324 |
+
"step": 21
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.0007484647966387127,
|
| 328 |
+
"grad_norm": 22.854082107543945,
|
| 329 |
+
"learning_rate": 2.2000000000000003e-05,
|
| 330 |
+
"logits/chosen": -3.1029181480407715,
|
| 331 |
+
"logits/rejected": -3.044419527053833,
|
| 332 |
+
"logps/chosen": -154.8139190673828,
|
| 333 |
+
"logps/rejected": -155.980712890625,
|
| 334 |
+
"loss": 0.4065,
|
| 335 |
+
"rewards/accuracies": 1.0,
|
| 336 |
+
"rewards/chosen": 0.45841503143310547,
|
| 337 |
+
"rewards/margins": 0.8753025531768799,
|
| 338 |
+
"rewards/rejected": -0.41688746213912964,
|
| 339 |
+
"step": 22
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.0007824859237586541,
|
| 343 |
+
"grad_norm": 43.94650650024414,
|
| 344 |
+
"learning_rate": 2.3000000000000003e-05,
|
| 345 |
+
"logits/chosen": -2.9984076023101807,
|
| 346 |
+
"logits/rejected": -3.114562511444092,
|
| 347 |
+
"logps/chosen": -263.8316650390625,
|
| 348 |
+
"logps/rejected": -340.1427307128906,
|
| 349 |
+
"loss": 1.0427,
|
| 350 |
+
"rewards/accuracies": 0.25,
|
| 351 |
+
"rewards/chosen": 0.6933631896972656,
|
| 352 |
+
"rewards/margins": -0.14637237787246704,
|
| 353 |
+
"rewards/rejected": 0.8397356271743774,
|
| 354 |
+
"step": 23
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 0.0008165070508785956,
|
| 358 |
+
"grad_norm": 49.11759948730469,
|
| 359 |
+
"learning_rate": 2.4e-05,
|
| 360 |
+
"logits/chosen": -3.025609016418457,
|
| 361 |
+
"logits/rejected": -3.0236239433288574,
|
| 362 |
+
"logps/chosen": -412.2383117675781,
|
| 363 |
+
"logps/rejected": -643.138427734375,
|
| 364 |
+
"loss": 1.0374,
|
| 365 |
+
"rewards/accuracies": 0.5,
|
| 366 |
+
"rewards/chosen": 1.1175220012664795,
|
| 367 |
+
"rewards/margins": -0.31503716111183167,
|
| 368 |
+
"rewards/rejected": 1.4325592517852783,
|
| 369 |
+
"step": 24
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 0.000850528177998537,
|
| 373 |
+
"grad_norm": 37.57383728027344,
|
| 374 |
+
"learning_rate": 2.5e-05,
|
| 375 |
+
"logits/chosen": -3.024618625640869,
|
| 376 |
+
"logits/rejected": -3.123701572418213,
|
| 377 |
+
"logps/chosen": -201.70530700683594,
|
| 378 |
+
"logps/rejected": -224.99038696289062,
|
| 379 |
+
"loss": 0.7477,
|
| 380 |
+
"rewards/accuracies": 0.25,
|
| 381 |
+
"rewards/chosen": 0.9206753373146057,
|
| 382 |
+
"rewards/margins": 0.5947309732437134,
|
| 383 |
+
"rewards/rejected": 0.32594436407089233,
|
| 384 |
+
"step": 25
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 0.0008845493051184785,
|
| 388 |
+
"grad_norm": 40.297794342041016,
|
| 389 |
+
"learning_rate": 2.6000000000000002e-05,
|
| 390 |
+
"logits/chosen": -3.1100215911865234,
|
| 391 |
+
"logits/rejected": -3.0195024013519287,
|
| 392 |
+
"logps/chosen": -226.2320098876953,
|
| 393 |
+
"logps/rejected": -193.0676727294922,
|
| 394 |
+
"loss": 0.8151,
|
| 395 |
+
"rewards/accuracies": 0.5,
|
| 396 |
+
"rewards/chosen": 0.9122087359428406,
|
| 397 |
+
"rewards/margins": 0.0203973650932312,
|
| 398 |
+
"rewards/rejected": 0.8918113708496094,
|
| 399 |
+
"step": 26
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 0.00091857043223842,
|
| 403 |
+
"grad_norm": 43.394229888916016,
|
| 404 |
+
"learning_rate": 2.7000000000000002e-05,
|
| 405 |
+
"logits/chosen": -3.0187630653381348,
|
| 406 |
+
"logits/rejected": -3.053678035736084,
|
| 407 |
+
"logps/chosen": -153.02883911132812,
|
| 408 |
+
"logps/rejected": -260.0540771484375,
|
| 409 |
+
"loss": 1.3457,
|
| 410 |
+
"rewards/accuracies": 0.5,
|
| 411 |
+
"rewards/chosen": 0.7145153284072876,
|
| 412 |
+
"rewards/margins": -0.806611180305481,
|
| 413 |
+
"rewards/rejected": 1.5211265087127686,
|
| 414 |
+
"step": 27
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 0.0009525915593583615,
|
| 418 |
+
"grad_norm": 28.59680938720703,
|
| 419 |
+
"learning_rate": 2.8000000000000003e-05,
|
| 420 |
+
"logits/chosen": -3.032409191131592,
|
| 421 |
+
"logits/rejected": -3.0502161979675293,
|
| 422 |
+
"logps/chosen": -240.1644287109375,
|
| 423 |
+
"logps/rejected": -245.81436157226562,
|
| 424 |
+
"loss": 0.5501,
|
| 425 |
+
"rewards/accuracies": 0.75,
|
| 426 |
+
"rewards/chosen": 2.0826761722564697,
|
| 427 |
+
"rewards/margins": 0.3433869779109955,
|
| 428 |
+
"rewards/rejected": 1.7392891645431519,
|
| 429 |
+
"step": 28
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.000986612686478303,
|
| 433 |
+
"grad_norm": 37.069583892822266,
|
| 434 |
+
"learning_rate": 2.9e-05,
|
| 435 |
+
"logits/chosen": -3.0516104698181152,
|
| 436 |
+
"logits/rejected": -3.144774913787842,
|
| 437 |
+
"logps/chosen": -296.03125,
|
| 438 |
+
"logps/rejected": -292.9052429199219,
|
| 439 |
+
"loss": 1.1087,
|
| 440 |
+
"rewards/accuracies": 0.75,
|
| 441 |
+
"rewards/chosen": 0.8324771523475647,
|
| 442 |
+
"rewards/margins": -0.34529268741607666,
|
| 443 |
+
"rewards/rejected": 1.1777698993682861,
|
| 444 |
+
"step": 29
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.0010206338135982445,
|
| 448 |
+
"grad_norm": 30.238658905029297,
|
| 449 |
+
"learning_rate": 3e-05,
|
| 450 |
+
"logits/chosen": -3.064671039581299,
|
| 451 |
+
"logits/rejected": -2.8963613510131836,
|
| 452 |
+
"logps/chosen": -415.6085510253906,
|
| 453 |
+
"logps/rejected": -149.9669952392578,
|
| 454 |
+
"loss": 0.4892,
|
| 455 |
+
"rewards/accuracies": 0.75,
|
| 456 |
+
"rewards/chosen": 1.0249770879745483,
|
| 457 |
+
"rewards/margins": 0.7549355030059814,
|
| 458 |
+
"rewards/rejected": 0.27004164457321167,
|
| 459 |
+
"step": 30
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 0.001054654940718186,
|
| 463 |
+
"grad_norm": 40.35402297973633,
|
| 464 |
+
"learning_rate": 3.1e-05,
|
| 465 |
+
"logits/chosen": -3.1762304306030273,
|
| 466 |
+
"logits/rejected": -3.061676025390625,
|
| 467 |
+
"logps/chosen": -585.9200439453125,
|
| 468 |
+
"logps/rejected": -238.16307067871094,
|
| 469 |
+
"loss": 0.587,
|
| 470 |
+
"rewards/accuracies": 0.5,
|
| 471 |
+
"rewards/chosen": 2.726571559906006,
|
| 472 |
+
"rewards/margins": 0.9808880090713501,
|
| 473 |
+
"rewards/rejected": 1.7456837892532349,
|
| 474 |
+
"step": 31
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 0.0010886760678381275,
|
| 478 |
+
"grad_norm": 33.78396987915039,
|
| 479 |
+
"learning_rate": 3.2000000000000005e-05,
|
| 480 |
+
"logits/chosen": -3.05906081199646,
|
| 481 |
+
"logits/rejected": -2.9525132179260254,
|
| 482 |
+
"logps/chosen": -580.87646484375,
|
| 483 |
+
"logps/rejected": -223.65457153320312,
|
| 484 |
+
"loss": 0.8173,
|
| 485 |
+
"rewards/accuracies": 0.5,
|
| 486 |
+
"rewards/chosen": 1.3172895908355713,
|
| 487 |
+
"rewards/margins": 0.07103176414966583,
|
| 488 |
+
"rewards/rejected": 1.2462577819824219,
|
| 489 |
+
"step": 32
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 0.001122697194958069,
|
| 493 |
+
"grad_norm": 40.97666549682617,
|
| 494 |
+
"learning_rate": 3.3e-05,
|
| 495 |
+
"logits/chosen": -3.084476947784424,
|
| 496 |
+
"logits/rejected": -3.11270809173584,
|
| 497 |
+
"logps/chosen": -285.040283203125,
|
| 498 |
+
"logps/rejected": -251.92686462402344,
|
| 499 |
+
"loss": 1.0645,
|
| 500 |
+
"rewards/accuracies": 0.5,
|
| 501 |
+
"rewards/chosen": 1.4826446771621704,
|
| 502 |
+
"rewards/margins": -0.13718131184577942,
|
| 503 |
+
"rewards/rejected": 1.6198259592056274,
|
| 504 |
+
"step": 33
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 0.0011567183220780105,
|
| 508 |
+
"grad_norm": 33.32075881958008,
|
| 509 |
+
"learning_rate": 3.4000000000000007e-05,
|
| 510 |
+
"logits/chosen": -3.037482500076294,
|
| 511 |
+
"logits/rejected": -3.0169787406921387,
|
| 512 |
+
"logps/chosen": -467.7403869628906,
|
| 513 |
+
"logps/rejected": -308.62274169921875,
|
| 514 |
+
"loss": 0.5226,
|
| 515 |
+
"rewards/accuracies": 0.5,
|
| 516 |
+
"rewards/chosen": 2.4038407802581787,
|
| 517 |
+
"rewards/margins": 0.8624236583709717,
|
| 518 |
+
"rewards/rejected": 1.541416883468628,
|
| 519 |
+
"step": 34
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 0.001190739449197952,
|
| 523 |
+
"grad_norm": 26.35866928100586,
|
| 524 |
+
"learning_rate": 3.5e-05,
|
| 525 |
+
"logits/chosen": -2.969816207885742,
|
| 526 |
+
"logits/rejected": -2.9069106578826904,
|
| 527 |
+
"logps/chosen": -312.856689453125,
|
| 528 |
+
"logps/rejected": -183.8594512939453,
|
| 529 |
+
"loss": 0.4495,
|
| 530 |
+
"rewards/accuracies": 0.75,
|
| 531 |
+
"rewards/chosen": 1.1757181882858276,
|
| 532 |
+
"rewards/margins": 0.803225576877594,
|
| 533 |
+
"rewards/rejected": 0.37249261140823364,
|
| 534 |
+
"step": 35
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.0012247605763178934,
|
| 538 |
+
"grad_norm": 44.118099212646484,
|
| 539 |
+
"learning_rate": 3.6e-05,
|
| 540 |
+
"logits/chosen": -3.091865301132202,
|
| 541 |
+
"logits/rejected": -3.0135951042175293,
|
| 542 |
+
"logps/chosen": -724.1255493164062,
|
| 543 |
+
"logps/rejected": -335.2945861816406,
|
| 544 |
+
"loss": 0.6341,
|
| 545 |
+
"rewards/accuracies": 0.75,
|
| 546 |
+
"rewards/chosen": 1.7523369789123535,
|
| 547 |
+
"rewards/margins": 0.3912923336029053,
|
| 548 |
+
"rewards/rejected": 1.3610445261001587,
|
| 549 |
+
"step": 36
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.001258781703437835,
|
| 553 |
+
"grad_norm": 26.80854606628418,
|
| 554 |
+
"learning_rate": 3.7e-05,
|
| 555 |
+
"logits/chosen": -3.0334794521331787,
|
| 556 |
+
"logits/rejected": -2.9878833293914795,
|
| 557 |
+
"logps/chosen": -406.20745849609375,
|
| 558 |
+
"logps/rejected": -377.2811584472656,
|
| 559 |
+
"loss": 0.5577,
|
| 560 |
+
"rewards/accuracies": 0.75,
|
| 561 |
+
"rewards/chosen": 2.6232261657714844,
|
| 562 |
+
"rewards/margins": 0.9015346765518188,
|
| 563 |
+
"rewards/rejected": 1.7216914892196655,
|
| 564 |
+
"step": 37
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 0.0012928028305577764,
|
| 568 |
+
"grad_norm": 31.345367431640625,
|
| 569 |
+
"learning_rate": 3.8e-05,
|
| 570 |
+
"logits/chosen": -3.0986504554748535,
|
| 571 |
+
"logits/rejected": -3.062534809112549,
|
| 572 |
+
"logps/chosen": -412.54498291015625,
|
| 573 |
+
"logps/rejected": -386.44158935546875,
|
| 574 |
+
"loss": 0.5479,
|
| 575 |
+
"rewards/accuracies": 0.75,
|
| 576 |
+
"rewards/chosen": 1.985120177268982,
|
| 577 |
+
"rewards/margins": 0.8609979748725891,
|
| 578 |
+
"rewards/rejected": 1.1241222620010376,
|
| 579 |
+
"step": 38
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 0.001326823957677718,
|
| 583 |
+
"grad_norm": 15.26794719696045,
|
| 584 |
+
"learning_rate": 3.9000000000000006e-05,
|
| 585 |
+
"logits/chosen": -3.0730984210968018,
|
| 586 |
+
"logits/rejected": -3.0951075553894043,
|
| 587 |
+
"logps/chosen": -305.12982177734375,
|
| 588 |
+
"logps/rejected": -285.45001220703125,
|
| 589 |
+
"loss": 0.2547,
|
| 590 |
+
"rewards/accuracies": 1.0,
|
| 591 |
+
"rewards/chosen": 2.55714750289917,
|
| 592 |
+
"rewards/margins": 1.3561341762542725,
|
| 593 |
+
"rewards/rejected": 1.2010130882263184,
|
| 594 |
+
"step": 39
|
| 595 |
+
},
|
| 596 |
+
{
|
| 597 |
+
"epoch": 0.0013608450847976594,
|
| 598 |
+
"grad_norm": 22.881174087524414,
|
| 599 |
+
"learning_rate": 4e-05,
|
| 600 |
+
"logits/chosen": -3.162843942642212,
|
| 601 |
+
"logits/rejected": -3.15628981590271,
|
| 602 |
+
"logps/chosen": -277.48626708984375,
|
| 603 |
+
"logps/rejected": -219.9507293701172,
|
| 604 |
+
"loss": 0.5427,
|
| 605 |
+
"rewards/accuracies": 0.5,
|
| 606 |
+
"rewards/chosen": 2.6990911960601807,
|
| 607 |
+
"rewards/margins": 0.5041837692260742,
|
| 608 |
+
"rewards/rejected": 2.1949074268341064,
|
| 609 |
+
"step": 40
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 0.001394866211917601,
|
| 613 |
+
"grad_norm": 15.179062843322754,
|
| 614 |
+
"learning_rate": 4.1e-05,
|
| 615 |
+
"logits/chosen": -3.138144016265869,
|
| 616 |
+
"logits/rejected": -3.1033153533935547,
|
| 617 |
+
"logps/chosen": -234.6917724609375,
|
| 618 |
+
"logps/rejected": -141.10888671875,
|
| 619 |
+
"loss": 0.2626,
|
| 620 |
+
"rewards/accuracies": 1.0,
|
| 621 |
+
"rewards/chosen": 2.799344539642334,
|
| 622 |
+
"rewards/margins": 1.4650464057922363,
|
| 623 |
+
"rewards/rejected": 1.3342978954315186,
|
| 624 |
+
"step": 41
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"epoch": 0.0014288873390375424,
|
| 628 |
+
"grad_norm": 21.09362030029297,
|
| 629 |
+
"learning_rate": 4.2e-05,
|
| 630 |
+
"logits/chosen": -3.0385375022888184,
|
| 631 |
+
"logits/rejected": -3.083676338195801,
|
| 632 |
+
"logps/chosen": -168.33697509765625,
|
| 633 |
+
"logps/rejected": -179.30813598632812,
|
| 634 |
+
"loss": 0.4002,
|
| 635 |
+
"rewards/accuracies": 1.0,
|
| 636 |
+
"rewards/chosen": 2.3925881385803223,
|
| 637 |
+
"rewards/margins": 0.8477088809013367,
|
| 638 |
+
"rewards/rejected": 1.5448791980743408,
|
| 639 |
+
"step": 42
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.0014629084661574839,
|
| 643 |
+
"grad_norm": 38.255069732666016,
|
| 644 |
+
"learning_rate": 4.3e-05,
|
| 645 |
+
"logits/chosen": -3.080360174179077,
|
| 646 |
+
"logits/rejected": -3.029470443725586,
|
| 647 |
+
"logps/chosen": -267.5482177734375,
|
| 648 |
+
"logps/rejected": -161.17111206054688,
|
| 649 |
+
"loss": 0.8857,
|
| 650 |
+
"rewards/accuracies": 0.5,
|
| 651 |
+
"rewards/chosen": 1.8575999736785889,
|
| 652 |
+
"rewards/margins": -0.24116593599319458,
|
| 653 |
+
"rewards/rejected": 2.0987658500671387,
|
| 654 |
+
"step": 43
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.0014969295932774254,
|
| 658 |
+
"grad_norm": 27.97095489501953,
|
| 659 |
+
"learning_rate": 4.4000000000000006e-05,
|
| 660 |
+
"logits/chosen": -3.1322476863861084,
|
| 661 |
+
"logits/rejected": -3.1334149837493896,
|
| 662 |
+
"logps/chosen": -94.13690185546875,
|
| 663 |
+
"logps/rejected": -78.98738098144531,
|
| 664 |
+
"loss": 0.9987,
|
| 665 |
+
"rewards/accuracies": 0.5,
|
| 666 |
+
"rewards/chosen": 1.6137428283691406,
|
| 667 |
+
"rewards/margins": -0.4127828776836395,
|
| 668 |
+
"rewards/rejected": 2.0265257358551025,
|
| 669 |
+
"step": 44
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 0.0015309507203973669,
|
| 673 |
+
"grad_norm": 45.25028991699219,
|
| 674 |
+
"learning_rate": 4.5e-05,
|
| 675 |
+
"logits/chosen": -3.121274709701538,
|
| 676 |
+
"logits/rejected": -3.1236798763275146,
|
| 677 |
+
"logps/chosen": -516.3858642578125,
|
| 678 |
+
"logps/rejected": -212.694580078125,
|
| 679 |
+
"loss": 0.9457,
|
| 680 |
+
"rewards/accuracies": 0.25,
|
| 681 |
+
"rewards/chosen": 2.419821262359619,
|
| 682 |
+
"rewards/margins": -0.38284340500831604,
|
| 683 |
+
"rewards/rejected": 2.8026647567749023,
|
| 684 |
+
"step": 45
|
| 685 |
+
},
|
| 686 |
+
{
|
| 687 |
+
"epoch": 0.0015649718475173081,
|
| 688 |
+
"grad_norm": 23.58992576599121,
|
| 689 |
+
"learning_rate": 4.600000000000001e-05,
|
| 690 |
+
"logits/chosen": -3.113922357559204,
|
| 691 |
+
"logits/rejected": -3.0506935119628906,
|
| 692 |
+
"logps/chosen": -318.288330078125,
|
| 693 |
+
"logps/rejected": -124.70735168457031,
|
| 694 |
+
"loss": 0.5671,
|
| 695 |
+
"rewards/accuracies": 0.75,
|
| 696 |
+
"rewards/chosen": 2.407118320465088,
|
| 697 |
+
"rewards/margins": 0.30923187732696533,
|
| 698 |
+
"rewards/rejected": 2.097886323928833,
|
| 699 |
+
"step": 46
|
| 700 |
+
},
|
| 701 |
+
{
|
| 702 |
+
"epoch": 0.0015989929746372496,
|
| 703 |
+
"grad_norm": 36.93040084838867,
|
| 704 |
+
"learning_rate": 4.7e-05,
|
| 705 |
+
"logits/chosen": -3.079291582107544,
|
| 706 |
+
"logits/rejected": -3.1158454418182373,
|
| 707 |
+
"logps/chosen": -246.26065063476562,
|
| 708 |
+
"logps/rejected": -336.46649169921875,
|
| 709 |
+
"loss": 0.9352,
|
| 710 |
+
"rewards/accuracies": 0.25,
|
| 711 |
+
"rewards/chosen": 2.214073657989502,
|
| 712 |
+
"rewards/margins": -0.2623659074306488,
|
| 713 |
+
"rewards/rejected": 2.4764394760131836,
|
| 714 |
+
"step": 47
|
| 715 |
+
},
|
| 716 |
+
{
|
| 717 |
+
"epoch": 0.0016330141017571911,
|
| 718 |
+
"grad_norm": 32.5983772277832,
|
| 719 |
+
"learning_rate": 4.8e-05,
|
| 720 |
+
"logits/chosen": -3.155709981918335,
|
| 721 |
+
"logits/rejected": -3.1829142570495605,
|
| 722 |
+
"logps/chosen": -204.55328369140625,
|
| 723 |
+
"logps/rejected": -219.52647399902344,
|
| 724 |
+
"loss": 0.8805,
|
| 725 |
+
"rewards/accuracies": 0.5,
|
| 726 |
+
"rewards/chosen": 2.155839681625366,
|
| 727 |
+
"rewards/margins": -0.002133697271347046,
|
| 728 |
+
"rewards/rejected": 2.157973289489746,
|
| 729 |
+
"step": 48
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 0.0016670352288771326,
|
| 733 |
+
"grad_norm": 35.626869201660156,
|
| 734 |
+
"learning_rate": 4.9e-05,
|
| 735 |
+
"logits/chosen": -3.10532546043396,
|
| 736 |
+
"logits/rejected": -3.113755464553833,
|
| 737 |
+
"logps/chosen": -390.5479736328125,
|
| 738 |
+
"logps/rejected": -351.7720031738281,
|
| 739 |
+
"loss": 0.7657,
|
| 740 |
+
"rewards/accuracies": 0.5,
|
| 741 |
+
"rewards/chosen": 2.5670056343078613,
|
| 742 |
+
"rewards/margins": -0.04349440336227417,
|
| 743 |
+
"rewards/rejected": 2.6105000972747803,
|
| 744 |
+
"step": 49
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.001701056355997074,
|
| 748 |
+
"grad_norm": 37.51042938232422,
|
| 749 |
+
"learning_rate": 5e-05,
|
| 750 |
+
"logits/chosen": -3.122190237045288,
|
| 751 |
+
"logits/rejected": -3.0837321281433105,
|
| 752 |
+
"logps/chosen": -251.68600463867188,
|
| 753 |
+
"logps/rejected": -242.4346923828125,
|
| 754 |
+
"loss": 0.9539,
|
| 755 |
+
"rewards/accuracies": 0.5,
|
| 756 |
+
"rewards/chosen": 2.0532169342041016,
|
| 757 |
+
"rewards/margins": -0.32518482208251953,
|
| 758 |
+
"rewards/rejected": 2.378401756286621,
|
| 759 |
+
"step": 50
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 0.0017350774831170156,
|
| 763 |
+
"grad_norm": 43.16423416137695,
|
| 764 |
+
"learning_rate": 4.999939076763487e-05,
|
| 765 |
+
"logits/chosen": -3.0424001216888428,
|
| 766 |
+
"logits/rejected": -3.0246453285217285,
|
| 767 |
+
"logps/chosen": -228.8073272705078,
|
| 768 |
+
"logps/rejected": -399.57781982421875,
|
| 769 |
+
"loss": 1.204,
|
| 770 |
+
"rewards/accuracies": 0.5,
|
| 771 |
+
"rewards/chosen": 2.1423535346984863,
|
| 772 |
+
"rewards/margins": -0.628890335559845,
|
| 773 |
+
"rewards/rejected": 2.7712438106536865,
|
| 774 |
+
"step": 51
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 0.001769098610236957,
|
| 778 |
+
"grad_norm": 32.16966247558594,
|
| 779 |
+
"learning_rate": 4.999756310023261e-05,
|
| 780 |
+
"logits/chosen": -3.049051284790039,
|
| 781 |
+
"logits/rejected": -3.061192750930786,
|
| 782 |
+
"logps/chosen": -219.27503967285156,
|
| 783 |
+
"logps/rejected": -238.98226928710938,
|
| 784 |
+
"loss": 0.9616,
|
| 785 |
+
"rewards/accuracies": 0.5,
|
| 786 |
+
"rewards/chosen": 2.1868481636047363,
|
| 787 |
+
"rewards/margins": -0.2618050277233124,
|
| 788 |
+
"rewards/rejected": 2.448652744293213,
|
| 789 |
+
"step": 52
|
| 790 |
+
},
|
| 791 |
+
{
|
| 792 |
+
"epoch": 0.0018031197373568986,
|
| 793 |
+
"grad_norm": 38.5218505859375,
|
| 794 |
+
"learning_rate": 4.999451708687114e-05,
|
| 795 |
+
"logits/chosen": -3.088409185409546,
|
| 796 |
+
"logits/rejected": -3.0845394134521484,
|
| 797 |
+
"logps/chosen": -131.70074462890625,
|
| 798 |
+
"logps/rejected": -205.93624877929688,
|
| 799 |
+
"loss": 1.0419,
|
| 800 |
+
"rewards/accuracies": 0.5,
|
| 801 |
+
"rewards/chosen": 1.822286605834961,
|
| 802 |
+
"rewards/margins": -0.18211710453033447,
|
| 803 |
+
"rewards/rejected": 2.004403829574585,
|
| 804 |
+
"step": 53
|
| 805 |
+
},
|
| 806 |
+
{
|
| 807 |
+
"epoch": 0.00183714086447684,
|
| 808 |
+
"grad_norm": 31.46576690673828,
|
| 809 |
+
"learning_rate": 4.999025287600886e-05,
|
| 810 |
+
"logits/chosen": -3.143968105316162,
|
| 811 |
+
"logits/rejected": -3.032649278640747,
|
| 812 |
+
"logps/chosen": -403.4815673828125,
|
| 813 |
+
"logps/rejected": -195.69281005859375,
|
| 814 |
+
"loss": 0.567,
|
| 815 |
+
"rewards/accuracies": 0.75,
|
| 816 |
+
"rewards/chosen": 2.2466557025909424,
|
| 817 |
+
"rewards/margins": 0.6339659094810486,
|
| 818 |
+
"rewards/rejected": 1.612689733505249,
|
| 819 |
+
"step": 54
|
| 820 |
+
},
|
| 821 |
+
{
|
| 822 |
+
"epoch": 0.0018711619915967816,
|
| 823 |
+
"grad_norm": 22.339370727539062,
|
| 824 |
+
"learning_rate": 4.99847706754774e-05,
|
| 825 |
+
"logits/chosen": -3.1364409923553467,
|
| 826 |
+
"logits/rejected": -3.0678646564483643,
|
| 827 |
+
"logps/chosen": -272.07330322265625,
|
| 828 |
+
"logps/rejected": -94.81082153320312,
|
| 829 |
+
"loss": 0.4856,
|
| 830 |
+
"rewards/accuracies": 1.0,
|
| 831 |
+
"rewards/chosen": 1.9992039203643799,
|
| 832 |
+
"rewards/margins": 0.5075141191482544,
|
| 833 |
+
"rewards/rejected": 1.491689682006836,
|
| 834 |
+
"step": 55
|
| 835 |
+
},
|
| 836 |
+
{
|
| 837 |
+
"epoch": 0.001905183118716723,
|
| 838 |
+
"grad_norm": 38.88309860229492,
|
| 839 |
+
"learning_rate": 4.997807075247146e-05,
|
| 840 |
+
"logits/chosen": -3.0754241943359375,
|
| 841 |
+
"logits/rejected": -3.1445722579956055,
|
| 842 |
+
"logps/chosen": -161.1916961669922,
|
| 843 |
+
"logps/rejected": -230.43589782714844,
|
| 844 |
+
"loss": 0.8925,
|
| 845 |
+
"rewards/accuracies": 0.5,
|
| 846 |
+
"rewards/chosen": 2.070554256439209,
|
| 847 |
+
"rewards/margins": -0.027551889419555664,
|
| 848 |
+
"rewards/rejected": 2.0981061458587646,
|
| 849 |
+
"step": 56
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.0019392042458366645,
|
| 853 |
+
"grad_norm": 29.006982803344727,
|
| 854 |
+
"learning_rate": 4.997015343353585e-05,
|
| 855 |
+
"logits/chosen": -3.079509973526001,
|
| 856 |
+
"logits/rejected": -3.119426727294922,
|
| 857 |
+
"logps/chosen": -236.40664672851562,
|
| 858 |
+
"logps/rejected": -259.257568359375,
|
| 859 |
+
"loss": 0.5812,
|
| 860 |
+
"rewards/accuracies": 0.75,
|
| 861 |
+
"rewards/chosen": 2.4261646270751953,
|
| 862 |
+
"rewards/margins": 0.7371253371238708,
|
| 863 |
+
"rewards/rejected": 1.6890392303466797,
|
| 864 |
+
"step": 57
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 0.001973225372956606,
|
| 868 |
+
"grad_norm": 18.193532943725586,
|
| 869 |
+
"learning_rate": 4.996101910454953e-05,
|
| 870 |
+
"logits/chosen": -3.070995569229126,
|
| 871 |
+
"logits/rejected": -2.8317549228668213,
|
| 872 |
+
"logps/chosen": -730.03759765625,
|
| 873 |
+
"logps/rejected": -133.17929077148438,
|
| 874 |
+
"loss": 0.3401,
|
| 875 |
+
"rewards/accuracies": 0.75,
|
| 876 |
+
"rewards/chosen": 3.711276054382324,
|
| 877 |
+
"rewards/margins": 1.7915010452270508,
|
| 878 |
+
"rewards/rejected": 1.9197750091552734,
|
| 879 |
+
"step": 58
|
| 880 |
+
},
|
| 881 |
+
{
|
| 882 |
+
"epoch": 0.0020072465000765477,
|
| 883 |
+
"grad_norm": 35.50777053833008,
|
| 884 |
+
"learning_rate": 4.995066821070679e-05,
|
| 885 |
+
"logits/chosen": -3.093824863433838,
|
| 886 |
+
"logits/rejected": -3.083796739578247,
|
| 887 |
+
"logps/chosen": -238.11883544921875,
|
| 888 |
+
"logps/rejected": -151.69790649414062,
|
| 889 |
+
"loss": 0.6251,
|
| 890 |
+
"rewards/accuracies": 0.75,
|
| 891 |
+
"rewards/chosen": 2.0124969482421875,
|
| 892 |
+
"rewards/margins": 0.640554666519165,
|
| 893 |
+
"rewards/rejected": 1.371942400932312,
|
| 894 |
+
"step": 59
|
| 895 |
+
},
|
| 896 |
+
{
|
| 897 |
+
"epoch": 0.002041267627196489,
|
| 898 |
+
"grad_norm": 40.34638977050781,
|
| 899 |
+
"learning_rate": 4.993910125649561e-05,
|
| 900 |
+
"logits/chosen": -3.1637063026428223,
|
| 901 |
+
"logits/rejected": -3.055285930633545,
|
| 902 |
+
"logps/chosen": -365.89434814453125,
|
| 903 |
+
"logps/rejected": -187.11392211914062,
|
| 904 |
+
"loss": 1.1514,
|
| 905 |
+
"rewards/accuracies": 0.75,
|
| 906 |
+
"rewards/chosen": 2.089510679244995,
|
| 907 |
+
"rewards/margins": -0.3508464992046356,
|
| 908 |
+
"rewards/rejected": 2.440357208251953,
|
| 909 |
+
"step": 60
|
| 910 |
+
},
|
| 911 |
+
{
|
| 912 |
+
"epoch": 0.0020752887543164307,
|
| 913 |
+
"grad_norm": 31.27332878112793,
|
| 914 |
+
"learning_rate": 4.992631880567301e-05,
|
| 915 |
+
"logits/chosen": -2.9931535720825195,
|
| 916 |
+
"logits/rejected": -2.994088649749756,
|
| 917 |
+
"logps/chosen": -127.87042236328125,
|
| 918 |
+
"logps/rejected": -236.82327270507812,
|
| 919 |
+
"loss": 0.8745,
|
| 920 |
+
"rewards/accuracies": 0.5,
|
| 921 |
+
"rewards/chosen": 1.9324662685394287,
|
| 922 |
+
"rewards/margins": -0.040864601731300354,
|
| 923 |
+
"rewards/rejected": 1.9733309745788574,
|
| 924 |
+
"step": 61
|
| 925 |
+
},
|
| 926 |
+
{
|
| 927 |
+
"epoch": 0.002109309881436372,
|
| 928 |
+
"grad_norm": 24.1131591796875,
|
| 929 |
+
"learning_rate": 4.991232148123761e-05,
|
| 930 |
+
"logits/chosen": -3.094635009765625,
|
| 931 |
+
"logits/rejected": -3.060533046722412,
|
| 932 |
+
"logps/chosen": -369.5483093261719,
|
| 933 |
+
"logps/rejected": -167.15155029296875,
|
| 934 |
+
"loss": 0.46,
|
| 935 |
+
"rewards/accuracies": 0.75,
|
| 936 |
+
"rewards/chosen": 2.38918399810791,
|
| 937 |
+
"rewards/margins": 1.1014902591705322,
|
| 938 |
+
"rewards/rejected": 1.2876936197280884,
|
| 939 |
+
"step": 62
|
| 940 |
+
},
|
| 941 |
+
{
|
| 942 |
+
"epoch": 0.0021433310085563133,
|
| 943 |
+
"grad_norm": 49.346961975097656,
|
| 944 |
+
"learning_rate": 4.989710996539926e-05,
|
| 945 |
+
"logits/chosen": -3.1134653091430664,
|
| 946 |
+
"logits/rejected": -3.0323257446289062,
|
| 947 |
+
"logps/chosen": -273.39361572265625,
|
| 948 |
+
"logps/rejected": -157.84652709960938,
|
| 949 |
+
"loss": 1.023,
|
| 950 |
+
"rewards/accuracies": 0.25,
|
| 951 |
+
"rewards/chosen": 1.1458964347839355,
|
| 952 |
+
"rewards/margins": -0.5081260204315186,
|
| 953 |
+
"rewards/rejected": 1.654022455215454,
|
| 954 |
+
"step": 63
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.002177352135676255,
|
| 958 |
+
"grad_norm": 41.145633697509766,
|
| 959 |
+
"learning_rate": 4.988068499954578e-05,
|
| 960 |
+
"logits/chosen": -2.9834160804748535,
|
| 961 |
+
"logits/rejected": -2.9958994388580322,
|
| 962 |
+
"logps/chosen": -334.5246887207031,
|
| 963 |
+
"logps/rejected": -386.3387145996094,
|
| 964 |
+
"loss": 1.5388,
|
| 965 |
+
"rewards/accuracies": 0.5,
|
| 966 |
+
"rewards/chosen": 2.305959939956665,
|
| 967 |
+
"rewards/margins": 0.07403802871704102,
|
| 968 |
+
"rewards/rejected": 2.231921911239624,
|
| 969 |
+
"step": 64
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 0.0022113732627961962,
|
| 973 |
+
"grad_norm": 33.07326126098633,
|
| 974 |
+
"learning_rate": 4.9863047384206835e-05,
|
| 975 |
+
"logits/chosen": -3.0006370544433594,
|
| 976 |
+
"logits/rejected": -3.020918846130371,
|
| 977 |
+
"logps/chosen": -344.7704162597656,
|
| 978 |
+
"logps/rejected": -200.45114135742188,
|
| 979 |
+
"loss": 0.5845,
|
| 980 |
+
"rewards/accuracies": 0.75,
|
| 981 |
+
"rewards/chosen": 1.643819808959961,
|
| 982 |
+
"rewards/margins": 0.6079027652740479,
|
| 983 |
+
"rewards/rejected": 1.035917043685913,
|
| 984 |
+
"step": 65
|
| 985 |
+
},
|
| 986 |
+
{
|
| 987 |
+
"epoch": 0.002245394389916138,
|
| 988 |
+
"grad_norm": 51.295345306396484,
|
| 989 |
+
"learning_rate": 4.984419797901491e-05,
|
| 990 |
+
"logits/chosen": -3.039350986480713,
|
| 991 |
+
"logits/rejected": -3.030729293823242,
|
| 992 |
+
"logps/chosen": -307.3952331542969,
|
| 993 |
+
"logps/rejected": -238.56224060058594,
|
| 994 |
+
"loss": 1.2331,
|
| 995 |
+
"rewards/accuracies": 0.25,
|
| 996 |
+
"rewards/chosen": 0.6686372756958008,
|
| 997 |
+
"rewards/margins": -0.6910279989242554,
|
| 998 |
+
"rewards/rejected": 1.3596653938293457,
|
| 999 |
+
"step": 66
|
| 1000 |
+
},
|
| 1001 |
+
{
|
| 1002 |
+
"epoch": 0.0022794155170360792,
|
| 1003 |
+
"grad_norm": 27.732166290283203,
|
| 1004 |
+
"learning_rate": 4.982413770266342e-05,
|
| 1005 |
+
"logits/chosen": -3.092078685760498,
|
| 1006 |
+
"logits/rejected": -3.1104204654693604,
|
| 1007 |
+
"logps/chosen": -206.7743682861328,
|
| 1008 |
+
"logps/rejected": -112.66790771484375,
|
| 1009 |
+
"loss": 0.5552,
|
| 1010 |
+
"rewards/accuracies": 0.5,
|
| 1011 |
+
"rewards/chosen": 1.273049235343933,
|
| 1012 |
+
"rewards/margins": 0.48671725392341614,
|
| 1013 |
+
"rewards/rejected": 0.7863319516181946,
|
| 1014 |
+
"step": 67
|
| 1015 |
+
},
|
| 1016 |
+
{
|
| 1017 |
+
"epoch": 0.002313436644156021,
|
| 1018 |
+
"grad_norm": 46.03068161010742,
|
| 1019 |
+
"learning_rate": 4.980286753286195e-05,
|
| 1020 |
+
"logits/chosen": -2.9219119548797607,
|
| 1021 |
+
"logits/rejected": -2.973864793777466,
|
| 1022 |
+
"logps/chosen": -455.3492431640625,
|
| 1023 |
+
"logps/rejected": -422.7337646484375,
|
| 1024 |
+
"loss": 0.8967,
|
| 1025 |
+
"rewards/accuracies": 0.5,
|
| 1026 |
+
"rewards/chosen": 1.788267731666565,
|
| 1027 |
+
"rewards/margins": -0.24189908802509308,
|
| 1028 |
+
"rewards/rejected": 2.0301668643951416,
|
| 1029 |
+
"step": 68
|
| 1030 |
+
},
|
| 1031 |
+
{
|
| 1032 |
+
"epoch": 0.002347457771275962,
|
| 1033 |
+
"grad_norm": 24.625539779663086,
|
| 1034 |
+
"learning_rate": 4.978038850628854e-05,
|
| 1035 |
+
"logits/chosen": -2.9824914932250977,
|
| 1036 |
+
"logits/rejected": -2.923306703567505,
|
| 1037 |
+
"logps/chosen": -320.2785339355469,
|
| 1038 |
+
"logps/rejected": -199.95407104492188,
|
| 1039 |
+
"loss": 0.5542,
|
| 1040 |
+
"rewards/accuracies": 0.75,
|
| 1041 |
+
"rewards/chosen": 2.4030070304870605,
|
| 1042 |
+
"rewards/margins": 1.2227776050567627,
|
| 1043 |
+
"rewards/rejected": 1.180229663848877,
|
| 1044 |
+
"step": 69
|
| 1045 |
+
},
|
| 1046 |
+
{
|
| 1047 |
+
"epoch": 0.002381478898395904,
|
| 1048 |
+
"grad_norm": 52.87644958496094,
|
| 1049 |
+
"learning_rate": 4.975670171853926e-05,
|
| 1050 |
+
"logits/chosen": -2.9576680660247803,
|
| 1051 |
+
"logits/rejected": -3.113696336746216,
|
| 1052 |
+
"logps/chosen": -226.1188201904297,
|
| 1053 |
+
"logps/rejected": -472.0656433105469,
|
| 1054 |
+
"loss": 1.5976,
|
| 1055 |
+
"rewards/accuracies": 0.25,
|
| 1056 |
+
"rewards/chosen": 1.8890628814697266,
|
| 1057 |
+
"rewards/margins": -1.2377982139587402,
|
| 1058 |
+
"rewards/rejected": 3.126861095428467,
|
| 1059 |
+
"step": 70
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.002415500025515845,
|
| 1063 |
+
"grad_norm": 41.89684295654297,
|
| 1064 |
+
"learning_rate": 4.9731808324074717e-05,
|
| 1065 |
+
"logits/chosen": -2.963118553161621,
|
| 1066 |
+
"logits/rejected": -2.95782732963562,
|
| 1067 |
+
"logps/chosen": -376.9123229980469,
|
| 1068 |
+
"logps/rejected": -408.6007080078125,
|
| 1069 |
+
"loss": 0.8305,
|
| 1070 |
+
"rewards/accuracies": 0.5,
|
| 1071 |
+
"rewards/chosen": 2.2748968601226807,
|
| 1072 |
+
"rewards/margins": 1.4728538990020752,
|
| 1073 |
+
"rewards/rejected": 0.8020429611206055,
|
| 1074 |
+
"step": 71
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 0.002449521152635787,
|
| 1078 |
+
"grad_norm": 33.66620635986328,
|
| 1079 |
+
"learning_rate": 4.9705709536163824e-05,
|
| 1080 |
+
"logits/chosen": -2.951019287109375,
|
| 1081 |
+
"logits/rejected": -2.8525712490081787,
|
| 1082 |
+
"logps/chosen": -334.75689697265625,
|
| 1083 |
+
"logps/rejected": -256.43963623046875,
|
| 1084 |
+
"loss": 0.5852,
|
| 1085 |
+
"rewards/accuracies": 0.75,
|
| 1086 |
+
"rewards/chosen": 2.2168543338775635,
|
| 1087 |
+
"rewards/margins": 1.3764129877090454,
|
| 1088 |
+
"rewards/rejected": 0.8404413461685181,
|
| 1089 |
+
"step": 72
|
| 1090 |
+
},
|
| 1091 |
+
{
|
| 1092 |
+
"epoch": 0.002483542279755728,
|
| 1093 |
+
"grad_norm": 48.537376403808594,
|
| 1094 |
+
"learning_rate": 4.96784066268247e-05,
|
| 1095 |
+
"logits/chosen": -2.96649169921875,
|
| 1096 |
+
"logits/rejected": -2.877800703048706,
|
| 1097 |
+
"logps/chosen": -342.9018859863281,
|
| 1098 |
+
"logps/rejected": -166.73826599121094,
|
| 1099 |
+
"loss": 1.3147,
|
| 1100 |
+
"rewards/accuracies": 0.0,
|
| 1101 |
+
"rewards/chosen": 1.4513309001922607,
|
| 1102 |
+
"rewards/margins": -0.9881281852722168,
|
| 1103 |
+
"rewards/rejected": 2.4394590854644775,
|
| 1104 |
+
"step": 73
|
| 1105 |
+
},
|
| 1106 |
+
{
|
| 1107 |
+
"epoch": 0.00251756340687567,
|
| 1108 |
+
"grad_norm": 35.8758544921875,
|
| 1109 |
+
"learning_rate": 4.964990092676263e-05,
|
| 1110 |
+
"logits/chosen": -3.007376194000244,
|
| 1111 |
+
"logits/rejected": -3.0710248947143555,
|
| 1112 |
+
"logps/chosen": -322.75604248046875,
|
| 1113 |
+
"logps/rejected": -288.9136047363281,
|
| 1114 |
+
"loss": 1.0483,
|
| 1115 |
+
"rewards/accuracies": 0.5,
|
| 1116 |
+
"rewards/chosen": 1.328976035118103,
|
| 1117 |
+
"rewards/margins": -0.057621777057647705,
|
| 1118 |
+
"rewards/rejected": 1.3865978717803955,
|
| 1119 |
+
"step": 74
|
| 1120 |
+
},
|
| 1121 |
+
{
|
| 1122 |
+
"epoch": 0.002551584533995611,
|
| 1123 |
+
"grad_norm": 52.42377471923828,
|
| 1124 |
+
"learning_rate": 4.962019382530521e-05,
|
| 1125 |
+
"logits/chosen": -2.952801465988159,
|
| 1126 |
+
"logits/rejected": -2.9580321311950684,
|
| 1127 |
+
"logps/chosen": -272.15576171875,
|
| 1128 |
+
"logps/rejected": -230.2147674560547,
|
| 1129 |
+
"loss": 1.4341,
|
| 1130 |
+
"rewards/accuracies": 0.25,
|
| 1131 |
+
"rewards/chosen": 1.2654911279678345,
|
| 1132 |
+
"rewards/margins": -0.9303737282752991,
|
| 1133 |
+
"rewards/rejected": 2.1958649158477783,
|
| 1134 |
+
"step": 75
|
| 1135 |
+
},
|
| 1136 |
+
{
|
| 1137 |
+
"epoch": 0.002585605661115553,
|
| 1138 |
+
"grad_norm": 54.157386779785156,
|
| 1139 |
+
"learning_rate": 4.9589286770334654e-05,
|
| 1140 |
+
"logits/chosen": -2.997375965118408,
|
| 1141 |
+
"logits/rejected": -3.091611385345459,
|
| 1142 |
+
"logps/chosen": -219.73159790039062,
|
| 1143 |
+
"logps/rejected": -503.797119140625,
|
| 1144 |
+
"loss": 1.3568,
|
| 1145 |
+
"rewards/accuracies": 0.25,
|
| 1146 |
+
"rewards/chosen": 1.520772933959961,
|
| 1147 |
+
"rewards/margins": -0.4154503643512726,
|
| 1148 |
+
"rewards/rejected": 1.9362232685089111,
|
| 1149 |
+
"step": 76
|
| 1150 |
+
},
|
| 1151 |
+
{
|
| 1152 |
+
"epoch": 0.002619626788235494,
|
| 1153 |
+
"grad_norm": 40.647274017333984,
|
| 1154 |
+
"learning_rate": 4.9557181268217227e-05,
|
| 1155 |
+
"logits/chosen": -3.0128493309020996,
|
| 1156 |
+
"logits/rejected": -2.9981260299682617,
|
| 1157 |
+
"logps/chosen": -319.1977233886719,
|
| 1158 |
+
"logps/rejected": -250.11123657226562,
|
| 1159 |
+
"loss": 0.7922,
|
| 1160 |
+
"rewards/accuracies": 0.5,
|
| 1161 |
+
"rewards/chosen": 1.9633705615997314,
|
| 1162 |
+
"rewards/margins": -0.027412444353103638,
|
| 1163 |
+
"rewards/rejected": 1.9907830953598022,
|
| 1164 |
+
"step": 77
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.002653647915355436,
|
| 1168 |
+
"grad_norm": 46.99341583251953,
|
| 1169 |
+
"learning_rate": 4.952387888372979e-05,
|
| 1170 |
+
"logits/chosen": -3.0307843685150146,
|
| 1171 |
+
"logits/rejected": -3.070122480392456,
|
| 1172 |
+
"logps/chosen": -134.52757263183594,
|
| 1173 |
+
"logps/rejected": -225.51242065429688,
|
| 1174 |
+
"loss": 1.182,
|
| 1175 |
+
"rewards/accuracies": 0.0,
|
| 1176 |
+
"rewards/chosen": 1.2856706380844116,
|
| 1177 |
+
"rewards/margins": -0.7930065393447876,
|
| 1178 |
+
"rewards/rejected": 2.078677177429199,
|
| 1179 |
+
"step": 78
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 0.002687669042475377,
|
| 1183 |
+
"grad_norm": 26.92460823059082,
|
| 1184 |
+
"learning_rate": 4.94893812399836e-05,
|
| 1185 |
+
"logits/chosen": -2.995112895965576,
|
| 1186 |
+
"logits/rejected": -2.950528144836426,
|
| 1187 |
+
"logps/chosen": -560.2005615234375,
|
| 1188 |
+
"logps/rejected": -383.92523193359375,
|
| 1189 |
+
"loss": 0.5477,
|
| 1190 |
+
"rewards/accuracies": 0.75,
|
| 1191 |
+
"rewards/chosen": 1.8620712757110596,
|
| 1192 |
+
"rewards/margins": 0.3632301390171051,
|
| 1193 |
+
"rewards/rejected": 1.4988411664962769,
|
| 1194 |
+
"step": 79
|
| 1195 |
+
},
|
| 1196 |
+
{
|
| 1197 |
+
"epoch": 0.002721690169595319,
|
| 1198 |
+
"grad_norm": 18.80535125732422,
|
| 1199 |
+
"learning_rate": 4.9453690018345144e-05,
|
| 1200 |
+
"logits/chosen": -3.0887956619262695,
|
| 1201 |
+
"logits/rejected": -3.013925790786743,
|
| 1202 |
+
"logps/chosen": -201.7890167236328,
|
| 1203 |
+
"logps/rejected": -120.17092895507812,
|
| 1204 |
+
"loss": 0.6049,
|
| 1205 |
+
"rewards/accuracies": 0.5,
|
| 1206 |
+
"rewards/chosen": 2.207493305206299,
|
| 1207 |
+
"rewards/margins": 0.5624717473983765,
|
| 1208 |
+
"rewards/rejected": 1.645021677017212,
|
| 1209 |
+
"step": 80
|
| 1210 |
+
},
|
| 1211 |
+
{
|
| 1212 |
+
"epoch": 0.00275571129671526,
|
| 1213 |
+
"grad_norm": 35.741817474365234,
|
| 1214 |
+
"learning_rate": 4.94168069583542e-05,
|
| 1215 |
+
"logits/chosen": -3.087347984313965,
|
| 1216 |
+
"logits/rejected": -2.9956464767456055,
|
| 1217 |
+
"logps/chosen": -485.24749755859375,
|
| 1218 |
+
"logps/rejected": -269.6015930175781,
|
| 1219 |
+
"loss": 1.0357,
|
| 1220 |
+
"rewards/accuracies": 0.5,
|
| 1221 |
+
"rewards/chosen": 1.650765299797058,
|
| 1222 |
+
"rewards/margins": -0.448674738407135,
|
| 1223 |
+
"rewards/rejected": 2.099440097808838,
|
| 1224 |
+
"step": 81
|
| 1225 |
+
},
|
| 1226 |
+
{
|
| 1227 |
+
"epoch": 0.002789732423835202,
|
| 1228 |
+
"grad_norm": 25.36280059814453,
|
| 1229 |
+
"learning_rate": 4.937873385763908e-05,
|
| 1230 |
+
"logits/chosen": -2.9839279651641846,
|
| 1231 |
+
"logits/rejected": -3.0026051998138428,
|
| 1232 |
+
"logps/chosen": -221.44424438476562,
|
| 1233 |
+
"logps/rejected": -193.8582305908203,
|
| 1234 |
+
"loss": 0.9423,
|
| 1235 |
+
"rewards/accuracies": 0.5,
|
| 1236 |
+
"rewards/chosen": 1.4655239582061768,
|
| 1237 |
+
"rewards/margins": -0.0549183189868927,
|
| 1238 |
+
"rewards/rejected": 1.520442247390747,
|
| 1239 |
+
"step": 82
|
| 1240 |
+
},
|
| 1241 |
+
{
|
| 1242 |
+
"epoch": 0.002823753550955143,
|
| 1243 |
+
"grad_norm": 12.532795906066895,
|
| 1244 |
+
"learning_rate": 4.933947257182901e-05,
|
| 1245 |
+
"logits/chosen": -3.0249555110931396,
|
| 1246 |
+
"logits/rejected": -3.0148353576660156,
|
| 1247 |
+
"logps/chosen": -386.8382263183594,
|
| 1248 |
+
"logps/rejected": -106.46695709228516,
|
| 1249 |
+
"loss": 0.3462,
|
| 1250 |
+
"rewards/accuracies": 0.75,
|
| 1251 |
+
"rewards/chosen": 3.745790481567383,
|
| 1252 |
+
"rewards/margins": 1.7120976448059082,
|
| 1253 |
+
"rewards/rejected": 2.0336928367614746,
|
| 1254 |
+
"step": 83
|
| 1255 |
+
},
|
| 1256 |
+
{
|
| 1257 |
+
"epoch": 0.002857774678075085,
|
| 1258 |
+
"grad_norm": 39.60961151123047,
|
| 1259 |
+
"learning_rate": 4.929902501446366e-05,
|
| 1260 |
+
"logits/chosen": -3.058694362640381,
|
| 1261 |
+
"logits/rejected": -3.0021092891693115,
|
| 1262 |
+
"logps/chosen": -269.97113037109375,
|
| 1263 |
+
"logps/rejected": -247.43728637695312,
|
| 1264 |
+
"loss": 1.5676,
|
| 1265 |
+
"rewards/accuracies": 0.0,
|
| 1266 |
+
"rewards/chosen": 1.230128526687622,
|
| 1267 |
+
"rewards/margins": -1.297512412071228,
|
| 1268 |
+
"rewards/rejected": 2.5276408195495605,
|
| 1269 |
+
"step": 84
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 0.002891795805195026,
|
| 1273 |
+
"grad_norm": 19.17630958557129,
|
| 1274 |
+
"learning_rate": 4.925739315689991e-05,
|
| 1275 |
+
"logits/chosen": -3.1123204231262207,
|
| 1276 |
+
"logits/rejected": -3.076711416244507,
|
| 1277 |
+
"logps/chosen": -323.8536376953125,
|
| 1278 |
+
"logps/rejected": -180.48394775390625,
|
| 1279 |
+
"loss": 0.5561,
|
| 1280 |
+
"rewards/accuracies": 0.75,
|
| 1281 |
+
"rewards/chosen": 2.0740671157836914,
|
| 1282 |
+
"rewards/margins": 0.7356327772140503,
|
| 1283 |
+
"rewards/rejected": 1.3384342193603516,
|
| 1284 |
+
"step": 85
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 0.0029258169323149678,
|
| 1288 |
+
"grad_norm": 15.29592227935791,
|
| 1289 |
+
"learning_rate": 4.9214579028215776e-05,
|
| 1290 |
+
"logits/chosen": -2.995969772338867,
|
| 1291 |
+
"logits/rejected": -2.975437879562378,
|
| 1292 |
+
"logps/chosen": -198.14756774902344,
|
| 1293 |
+
"logps/rejected": -97.39586639404297,
|
| 1294 |
+
"loss": 0.4645,
|
| 1295 |
+
"rewards/accuracies": 1.0,
|
| 1296 |
+
"rewards/chosen": 2.481008529663086,
|
| 1297 |
+
"rewards/margins": 0.6242896914482117,
|
| 1298 |
+
"rewards/rejected": 1.8567187786102295,
|
| 1299 |
+
"step": 86
|
| 1300 |
+
},
|
| 1301 |
+
{
|
| 1302 |
+
"epoch": 0.002959838059434909,
|
| 1303 |
+
"grad_norm": 37.80276870727539,
|
| 1304 |
+
"learning_rate": 4.917058471511149e-05,
|
| 1305 |
+
"logits/chosen": -3.0480551719665527,
|
| 1306 |
+
"logits/rejected": -3.0806996822357178,
|
| 1307 |
+
"logps/chosen": -224.3466796875,
|
| 1308 |
+
"logps/rejected": -420.4508361816406,
|
| 1309 |
+
"loss": 1.1967,
|
| 1310 |
+
"rewards/accuracies": 0.25,
|
| 1311 |
+
"rewards/chosen": 1.6259909868240356,
|
| 1312 |
+
"rewards/margins": -0.6426498889923096,
|
| 1313 |
+
"rewards/rejected": 2.2686409950256348,
|
| 1314 |
+
"step": 87
|
| 1315 |
+
},
|
| 1316 |
+
{
|
| 1317 |
+
"epoch": 0.0029938591865548508,
|
| 1318 |
+
"grad_norm": 20.813907623291016,
|
| 1319 |
+
"learning_rate": 4.912541236180779e-05,
|
| 1320 |
+
"logits/chosen": -3.045354127883911,
|
| 1321 |
+
"logits/rejected": -3.0551252365112305,
|
| 1322 |
+
"logps/chosen": -190.39498901367188,
|
| 1323 |
+
"logps/rejected": -417.9426574707031,
|
| 1324 |
+
"loss": 0.4819,
|
| 1325 |
+
"rewards/accuracies": 0.75,
|
| 1326 |
+
"rewards/chosen": 2.733407497406006,
|
| 1327 |
+
"rewards/margins": 0.5519695281982422,
|
| 1328 |
+
"rewards/rejected": 2.1814379692077637,
|
| 1329 |
+
"step": 88
|
| 1330 |
+
},
|
| 1331 |
+
{
|
| 1332 |
+
"epoch": 0.003027880313674792,
|
| 1333 |
+
"grad_norm": 18.250917434692383,
|
| 1334 |
+
"learning_rate": 4.907906416994146e-05,
|
| 1335 |
+
"logits/chosen": -3.1118156909942627,
|
| 1336 |
+
"logits/rejected": -3.124022960662842,
|
| 1337 |
+
"logps/chosen": -423.84698486328125,
|
| 1338 |
+
"logps/rejected": -217.9359130859375,
|
| 1339 |
+
"loss": 0.3913,
|
| 1340 |
+
"rewards/accuracies": 0.75,
|
| 1341 |
+
"rewards/chosen": 2.9902727603912354,
|
| 1342 |
+
"rewards/margins": 1.3608616590499878,
|
| 1343 |
+
"rewards/rejected": 1.629410982131958,
|
| 1344 |
+
"step": 89
|
| 1345 |
+
},
|
| 1346 |
+
{
|
| 1347 |
+
"epoch": 0.0030619014407947337,
|
| 1348 |
+
"grad_norm": 18.630115509033203,
|
| 1349 |
+
"learning_rate": 4.9031542398457974e-05,
|
| 1350 |
+
"logits/chosen": -3.0090408325195312,
|
| 1351 |
+
"logits/rejected": -3.0369369983673096,
|
| 1352 |
+
"logps/chosen": -435.40478515625,
|
| 1353 |
+
"logps/rejected": -191.86627197265625,
|
| 1354 |
+
"loss": 0.7079,
|
| 1355 |
+
"rewards/accuracies": 0.75,
|
| 1356 |
+
"rewards/chosen": 2.947678327560425,
|
| 1357 |
+
"rewards/margins": 0.8492887020111084,
|
| 1358 |
+
"rewards/rejected": 2.0983893871307373,
|
| 1359 |
+
"step": 90
|
| 1360 |
+
},
|
| 1361 |
+
{
|
| 1362 |
+
"epoch": 0.003095922567914675,
|
| 1363 |
+
"grad_norm": 19.552978515625,
|
| 1364 |
+
"learning_rate": 4.898284936350144e-05,
|
| 1365 |
+
"logits/chosen": -3.0316624641418457,
|
| 1366 |
+
"logits/rejected": -3.011552572250366,
|
| 1367 |
+
"logps/chosen": -432.43402099609375,
|
| 1368 |
+
"logps/rejected": -214.85935974121094,
|
| 1369 |
+
"loss": 0.7816,
|
| 1370 |
+
"rewards/accuracies": 0.5,
|
| 1371 |
+
"rewards/chosen": 2.7960076332092285,
|
| 1372 |
+
"rewards/margins": 0.5724576115608215,
|
| 1373 |
+
"rewards/rejected": 2.2235498428344727,
|
| 1374 |
+
"step": 91
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 0.0031299436950346163,
|
| 1378 |
+
"grad_norm": 7.452019691467285,
|
| 1379 |
+
"learning_rate": 4.893298743830168e-05,
|
| 1380 |
+
"logits/chosen": -3.038140296936035,
|
| 1381 |
+
"logits/rejected": -3.00756573677063,
|
| 1382 |
+
"logps/chosen": -168.3916473388672,
|
| 1383 |
+
"logps/rejected": -119.79571533203125,
|
| 1384 |
+
"loss": 0.2097,
|
| 1385 |
+
"rewards/accuracies": 1.0,
|
| 1386 |
+
"rewards/chosen": 2.8402457237243652,
|
| 1387 |
+
"rewards/margins": 1.5049405097961426,
|
| 1388 |
+
"rewards/rejected": 1.3353054523468018,
|
| 1389 |
+
"step": 92
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 0.003163964822154558,
|
| 1393 |
+
"grad_norm": 16.27883529663086,
|
| 1394 |
+
"learning_rate": 4.888195905305859e-05,
|
| 1395 |
+
"logits/chosen": -3.030277729034424,
|
| 1396 |
+
"logits/rejected": -3.0196878910064697,
|
| 1397 |
+
"logps/chosen": -295.2430114746094,
|
| 1398 |
+
"logps/rejected": -229.96961975097656,
|
| 1399 |
+
"loss": 0.5378,
|
| 1400 |
+
"rewards/accuracies": 0.5,
|
| 1401 |
+
"rewards/chosen": 2.0215210914611816,
|
| 1402 |
+
"rewards/margins": 0.6857559084892273,
|
| 1403 |
+
"rewards/rejected": 1.3357652425765991,
|
| 1404 |
+
"step": 93
|
| 1405 |
+
},
|
| 1406 |
+
{
|
| 1407 |
+
"epoch": 0.0031979859492744993,
|
| 1408 |
+
"grad_norm": 27.870994567871094,
|
| 1409 |
+
"learning_rate": 4.882976669482367e-05,
|
| 1410 |
+
"logits/chosen": -3.0157508850097656,
|
| 1411 |
+
"logits/rejected": -3.062469005584717,
|
| 1412 |
+
"logps/chosen": -594.1408081054688,
|
| 1413 |
+
"logps/rejected": -284.8695068359375,
|
| 1414 |
+
"loss": 0.829,
|
| 1415 |
+
"rewards/accuracies": 0.25,
|
| 1416 |
+
"rewards/chosen": 2.395214796066284,
|
| 1417 |
+
"rewards/margins": -0.06876510381698608,
|
| 1418 |
+
"rewards/rejected": 2.463979959487915,
|
| 1419 |
+
"step": 94
|
| 1420 |
+
},
|
| 1421 |
+
{
|
| 1422 |
+
"epoch": 0.003232007076394441,
|
| 1423 |
+
"grad_norm": 34.57992172241211,
|
| 1424 |
+
"learning_rate": 4.877641290737884e-05,
|
| 1425 |
+
"logits/chosen": -3.0241267681121826,
|
| 1426 |
+
"logits/rejected": -3.072436809539795,
|
| 1427 |
+
"logps/chosen": -206.46072387695312,
|
| 1428 |
+
"logps/rejected": -490.99163818359375,
|
| 1429 |
+
"loss": 0.9507,
|
| 1430 |
+
"rewards/accuracies": 0.25,
|
| 1431 |
+
"rewards/chosen": 3.0148510932922363,
|
| 1432 |
+
"rewards/margins": -0.41296035051345825,
|
| 1433 |
+
"rewards/rejected": 3.427811622619629,
|
| 1434 |
+
"step": 95
|
| 1435 |
+
},
|
| 1436 |
+
{
|
| 1437 |
+
"epoch": 0.0032660282035143822,
|
| 1438 |
+
"grad_norm": 31.449804306030273,
|
| 1439 |
+
"learning_rate": 4.8721900291112415e-05,
|
| 1440 |
+
"logits/chosen": -2.978365659713745,
|
| 1441 |
+
"logits/rejected": -2.9975037574768066,
|
| 1442 |
+
"logps/chosen": -353.8739929199219,
|
| 1443 |
+
"logps/rejected": -265.1501159667969,
|
| 1444 |
+
"loss": 0.7224,
|
| 1445 |
+
"rewards/accuracies": 0.5,
|
| 1446 |
+
"rewards/chosen": 3.355400562286377,
|
| 1447 |
+
"rewards/margins": 0.41021838784217834,
|
| 1448 |
+
"rewards/rejected": 2.9451825618743896,
|
| 1449 |
+
"step": 96
|
| 1450 |
+
},
|
| 1451 |
+
{
|
| 1452 |
+
"epoch": 0.003300049330634324,
|
| 1453 |
+
"grad_norm": 20.54676628112793,
|
| 1454 |
+
"learning_rate": 4.8666231502892415e-05,
|
| 1455 |
+
"logits/chosen": -3.0715813636779785,
|
| 1456 |
+
"logits/rejected": -3.0223703384399414,
|
| 1457 |
+
"logps/chosen": -417.56280517578125,
|
| 1458 |
+
"logps/rejected": -126.94705963134766,
|
| 1459 |
+
"loss": 0.5149,
|
| 1460 |
+
"rewards/accuracies": 0.75,
|
| 1461 |
+
"rewards/chosen": 2.667041540145874,
|
| 1462 |
+
"rewards/margins": 0.9312877655029297,
|
| 1463 |
+
"rewards/rejected": 1.7357536554336548,
|
| 1464 |
+
"step": 97
|
| 1465 |
+
},
|
| 1466 |
+
{
|
| 1467 |
+
"epoch": 0.0033340704577542652,
|
| 1468 |
+
"grad_norm": 34.95238494873047,
|
| 1469 |
+
"learning_rate": 4.860940925593703e-05,
|
| 1470 |
+
"logits/chosen": -3.038242816925049,
|
| 1471 |
+
"logits/rejected": -3.026169776916504,
|
| 1472 |
+
"logps/chosen": -320.06005859375,
|
| 1473 |
+
"logps/rejected": -534.7603759765625,
|
| 1474 |
+
"loss": 0.6213,
|
| 1475 |
+
"rewards/accuracies": 0.75,
|
| 1476 |
+
"rewards/chosen": 3.1372852325439453,
|
| 1477 |
+
"rewards/margins": 0.3597479462623596,
|
| 1478 |
+
"rewards/rejected": 2.7775373458862305,
|
| 1479 |
+
"step": 98
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 0.003368091584874207,
|
| 1483 |
+
"grad_norm": 22.04984474182129,
|
| 1484 |
+
"learning_rate": 4.855143631968242e-05,
|
| 1485 |
+
"logits/chosen": -3.0394222736358643,
|
| 1486 |
+
"logits/rejected": -3.040712356567383,
|
| 1487 |
+
"logps/chosen": -405.9694519042969,
|
| 1488 |
+
"logps/rejected": -224.75143432617188,
|
| 1489 |
+
"loss": 0.472,
|
| 1490 |
+
"rewards/accuracies": 0.75,
|
| 1491 |
+
"rewards/chosen": 3.6535189151763916,
|
| 1492 |
+
"rewards/margins": 1.3992540836334229,
|
| 1493 |
+
"rewards/rejected": 2.2542648315429688,
|
| 1494 |
+
"step": 99
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 0.003402112711994148,
|
| 1498 |
+
"grad_norm": 10.456045150756836,
|
| 1499 |
+
"learning_rate": 4.849231551964771e-05,
|
| 1500 |
+
"logits/chosen": -3.0558083057403564,
|
| 1501 |
+
"logits/rejected": -3.0164895057678223,
|
| 1502 |
+
"logps/chosen": -215.77474975585938,
|
| 1503 |
+
"logps/rejected": -114.40177917480469,
|
| 1504 |
+
"loss": 0.3431,
|
| 1505 |
+
"rewards/accuracies": 0.75,
|
| 1506 |
+
"rewards/chosen": 3.2260184288024902,
|
| 1507 |
+
"rewards/margins": 1.2816791534423828,
|
| 1508 |
+
"rewards/rejected": 1.9443392753601074,
|
| 1509 |
+
"step": 100
|
| 1510 |
+
}
|
| 1511 |
+
],
|
| 1512 |
+
"logging_steps": 1,
|
| 1513 |
+
"max_steps": 500,
|
| 1514 |
+
"num_input_tokens_seen": 0,
|
| 1515 |
+
"num_train_epochs": 1,
|
| 1516 |
+
"save_steps": 10,
|
| 1517 |
+
"stateful_callbacks": {
|
| 1518 |
+
"TrainerControl": {
|
| 1519 |
+
"args": {
|
| 1520 |
+
"should_epoch_stop": false,
|
| 1521 |
+
"should_evaluate": false,
|
| 1522 |
+
"should_log": false,
|
| 1523 |
+
"should_save": true,
|
| 1524 |
+
"should_training_stop": false
|
| 1525 |
+
},
|
| 1526 |
+
"attributes": {}
|
| 1527 |
+
}
|
| 1528 |
+
},
|
| 1529 |
+
"total_flos": 0.0,
|
| 1530 |
+
"train_batch_size": 1,
|
| 1531 |
+
"trial_name": null,
|
| 1532 |
+
"trial_params": null
|
| 1533 |
+
}
|
checkpoint-100/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2aa185d46c9d5abcf13df6d50376870fd603b11ffa1fb959ec10e0609dfec047
|
| 3 |
+
size 5880
|
checkpoint-110/README.md
ADDED
|
@@ -0,0 +1,202 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: peft
|
| 3 |
+
base_model: teknium/OpenHermes-2.5-Mistral-7B
|
| 4 |
+
---
|
| 5 |
+
|
| 6 |
+
# Model Card for Model ID
|
| 7 |
+
|
| 8 |
+
<!-- Provide a quick summary of what the model is/does. -->
|
| 9 |
+
|
| 10 |
+
|
| 11 |
+
|
| 12 |
+
## Model Details
|
| 13 |
+
|
| 14 |
+
### Model Description
|
| 15 |
+
|
| 16 |
+
<!-- Provide a longer summary of what this model is. -->
|
| 17 |
+
|
| 18 |
+
|
| 19 |
+
|
| 20 |
+
- **Developed by:** [More Information Needed]
|
| 21 |
+
- **Funded by [optional]:** [More Information Needed]
|
| 22 |
+
- **Shared by [optional]:** [More Information Needed]
|
| 23 |
+
- **Model type:** [More Information Needed]
|
| 24 |
+
- **Language(s) (NLP):** [More Information Needed]
|
| 25 |
+
- **License:** [More Information Needed]
|
| 26 |
+
- **Finetuned from model [optional]:** [More Information Needed]
|
| 27 |
+
|
| 28 |
+
### Model Sources [optional]
|
| 29 |
+
|
| 30 |
+
<!-- Provide the basic links for the model. -->
|
| 31 |
+
|
| 32 |
+
- **Repository:** [More Information Needed]
|
| 33 |
+
- **Paper [optional]:** [More Information Needed]
|
| 34 |
+
- **Demo [optional]:** [More Information Needed]
|
| 35 |
+
|
| 36 |
+
## Uses
|
| 37 |
+
|
| 38 |
+
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
|
| 39 |
+
|
| 40 |
+
### Direct Use
|
| 41 |
+
|
| 42 |
+
<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
|
| 43 |
+
|
| 44 |
+
[More Information Needed]
|
| 45 |
+
|
| 46 |
+
### Downstream Use [optional]
|
| 47 |
+
|
| 48 |
+
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
|
| 49 |
+
|
| 50 |
+
[More Information Needed]
|
| 51 |
+
|
| 52 |
+
### Out-of-Scope Use
|
| 53 |
+
|
| 54 |
+
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
|
| 55 |
+
|
| 56 |
+
[More Information Needed]
|
| 57 |
+
|
| 58 |
+
## Bias, Risks, and Limitations
|
| 59 |
+
|
| 60 |
+
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
|
| 61 |
+
|
| 62 |
+
[More Information Needed]
|
| 63 |
+
|
| 64 |
+
### Recommendations
|
| 65 |
+
|
| 66 |
+
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
|
| 67 |
+
|
| 68 |
+
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
|
| 69 |
+
|
| 70 |
+
## How to Get Started with the Model
|
| 71 |
+
|
| 72 |
+
Use the code below to get started with the model.
|
| 73 |
+
|
| 74 |
+
[More Information Needed]
|
| 75 |
+
|
| 76 |
+
## Training Details
|
| 77 |
+
|
| 78 |
+
### Training Data
|
| 79 |
+
|
| 80 |
+
<!-- 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. -->
|
| 81 |
+
|
| 82 |
+
[More Information Needed]
|
| 83 |
+
|
| 84 |
+
### Training Procedure
|
| 85 |
+
|
| 86 |
+
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
|
| 87 |
+
|
| 88 |
+
#### Preprocessing [optional]
|
| 89 |
+
|
| 90 |
+
[More Information Needed]
|
| 91 |
+
|
| 92 |
+
|
| 93 |
+
#### Training Hyperparameters
|
| 94 |
+
|
| 95 |
+
- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
|
| 96 |
+
|
| 97 |
+
#### Speeds, Sizes, Times [optional]
|
| 98 |
+
|
| 99 |
+
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
|
| 100 |
+
|
| 101 |
+
[More Information Needed]
|
| 102 |
+
|
| 103 |
+
## Evaluation
|
| 104 |
+
|
| 105 |
+
<!-- This section describes the evaluation protocols and provides the results. -->
|
| 106 |
+
|
| 107 |
+
### Testing Data, Factors & Metrics
|
| 108 |
+
|
| 109 |
+
#### Testing Data
|
| 110 |
+
|
| 111 |
+
<!-- This should link to a Dataset Card if possible. -->
|
| 112 |
+
|
| 113 |
+
[More Information Needed]
|
| 114 |
+
|
| 115 |
+
#### Factors
|
| 116 |
+
|
| 117 |
+
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
|
| 118 |
+
|
| 119 |
+
[More Information Needed]
|
| 120 |
+
|
| 121 |
+
#### Metrics
|
| 122 |
+
|
| 123 |
+
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
|
| 124 |
+
|
| 125 |
+
[More Information Needed]
|
| 126 |
+
|
| 127 |
+
### Results
|
| 128 |
+
|
| 129 |
+
[More Information Needed]
|
| 130 |
+
|
| 131 |
+
#### Summary
|
| 132 |
+
|
| 133 |
+
|
| 134 |
+
|
| 135 |
+
## Model Examination [optional]
|
| 136 |
+
|
| 137 |
+
<!-- Relevant interpretability work for the model goes here -->
|
| 138 |
+
|
| 139 |
+
[More Information Needed]
|
| 140 |
+
|
| 141 |
+
## Environmental Impact
|
| 142 |
+
|
| 143 |
+
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
|
| 144 |
+
|
| 145 |
+
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).
|
| 146 |
+
|
| 147 |
+
- **Hardware Type:** [More Information Needed]
|
| 148 |
+
- **Hours used:** [More Information Needed]
|
| 149 |
+
- **Cloud Provider:** [More Information Needed]
|
| 150 |
+
- **Compute Region:** [More Information Needed]
|
| 151 |
+
- **Carbon Emitted:** [More Information Needed]
|
| 152 |
+
|
| 153 |
+
## Technical Specifications [optional]
|
| 154 |
+
|
| 155 |
+
### Model Architecture and Objective
|
| 156 |
+
|
| 157 |
+
[More Information Needed]
|
| 158 |
+
|
| 159 |
+
### Compute Infrastructure
|
| 160 |
+
|
| 161 |
+
[More Information Needed]
|
| 162 |
+
|
| 163 |
+
#### Hardware
|
| 164 |
+
|
| 165 |
+
[More Information Needed]
|
| 166 |
+
|
| 167 |
+
#### Software
|
| 168 |
+
|
| 169 |
+
[More Information Needed]
|
| 170 |
+
|
| 171 |
+
## Citation [optional]
|
| 172 |
+
|
| 173 |
+
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
|
| 174 |
+
|
| 175 |
+
**BibTeX:**
|
| 176 |
+
|
| 177 |
+
[More Information Needed]
|
| 178 |
+
|
| 179 |
+
**APA:**
|
| 180 |
+
|
| 181 |
+
[More Information Needed]
|
| 182 |
+
|
| 183 |
+
## Glossary [optional]
|
| 184 |
+
|
| 185 |
+
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
|
| 186 |
+
|
| 187 |
+
[More Information Needed]
|
| 188 |
+
|
| 189 |
+
## More Information [optional]
|
| 190 |
+
|
| 191 |
+
[More Information Needed]
|
| 192 |
+
|
| 193 |
+
## Model Card Authors [optional]
|
| 194 |
+
|
| 195 |
+
[More Information Needed]
|
| 196 |
+
|
| 197 |
+
## Model Card Contact
|
| 198 |
+
|
| 199 |
+
[More Information Needed]
|
| 200 |
+
### Framework versions
|
| 201 |
+
|
| 202 |
+
- PEFT 0.11.1
|
checkpoint-110/adapter_config.json
ADDED
|
@@ -0,0 +1,34 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"alpha_pattern": {},
|
| 3 |
+
"auto_mapping": null,
|
| 4 |
+
"base_model_name_or_path": "teknium/OpenHermes-2.5-Mistral-7B",
|
| 5 |
+
"bias": "none",
|
| 6 |
+
"fan_in_fan_out": false,
|
| 7 |
+
"inference_mode": true,
|
| 8 |
+
"init_lora_weights": true,
|
| 9 |
+
"layer_replication": null,
|
| 10 |
+
"layers_pattern": null,
|
| 11 |
+
"layers_to_transform": null,
|
| 12 |
+
"loftq_config": {},
|
| 13 |
+
"lora_alpha": 16,
|
| 14 |
+
"lora_dropout": 0.05,
|
| 15 |
+
"megatron_config": null,
|
| 16 |
+
"megatron_core": "megatron.core",
|
| 17 |
+
"modules_to_save": null,
|
| 18 |
+
"peft_type": "LORA",
|
| 19 |
+
"r": 16,
|
| 20 |
+
"rank_pattern": {},
|
| 21 |
+
"revision": null,
|
| 22 |
+
"target_modules": [
|
| 23 |
+
"o_proj",
|
| 24 |
+
"k_proj",
|
| 25 |
+
"gate_proj",
|
| 26 |
+
"down_proj",
|
| 27 |
+
"up_proj",
|
| 28 |
+
"q_proj",
|
| 29 |
+
"v_proj"
|
| 30 |
+
],
|
| 31 |
+
"task_type": "CAUSAL_LM",
|
| 32 |
+
"use_dora": false,
|
| 33 |
+
"use_rslora": false
|
| 34 |
+
}
|
checkpoint-110/adapter_model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d399b3be8a20c68a8303a3b34fd48d1c8ae20c271b966e780ea2de8783c878af
|
| 3 |
+
size 167832240
|
checkpoint-110/added_tokens.json
ADDED
|
@@ -0,0 +1,4 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"<|im_end|>": 32000,
|
| 3 |
+
"<|im_start|>": 32001
|
| 4 |
+
}
|
checkpoint-110/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:25bc1658d09ca26a073f28184c37aff91c617a8c5984abcdfed01fbd02a69158
|
| 3 |
+
size 335928658
|
checkpoint-110/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:386fcc8cc1089aade9450d86fb239ea3483f455fd2d78d8378645feecfec9d69
|
| 3 |
+
size 14244
|
checkpoint-110/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f902385e3d5b0744fa286fd11d80d0116c3d1c3f5a7de9091350299005fbcdd9
|
| 3 |
+
size 1064
|
checkpoint-110/special_tokens_map.json
ADDED
|
@@ -0,0 +1,24 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": {
|
| 3 |
+
"content": "<s>",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": false,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"eos_token": {
|
| 10 |
+
"content": "<|im_end|>",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": false,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"pad_token": "<|im_end|>",
|
| 17 |
+
"unk_token": {
|
| 18 |
+
"content": "<unk>",
|
| 19 |
+
"lstrip": false,
|
| 20 |
+
"normalized": false,
|
| 21 |
+
"rstrip": false,
|
| 22 |
+
"single_word": false
|
| 23 |
+
}
|
| 24 |
+
}
|
checkpoint-110/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-110/tokenizer.model
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:dadfd56d766715c61d2ef780a525ab43b8e6da4de6865bda3d95fdef5e134055
|
| 3 |
+
size 493443
|
checkpoint-110/tokenizer_config.json
ADDED
|
@@ -0,0 +1,62 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_bos_token": true,
|
| 3 |
+
"add_eos_token": false,
|
| 4 |
+
"add_prefix_space": true,
|
| 5 |
+
"added_tokens_decoder": {
|
| 6 |
+
"0": {
|
| 7 |
+
"content": "<unk>",
|
| 8 |
+
"lstrip": false,
|
| 9 |
+
"normalized": false,
|
| 10 |
+
"rstrip": false,
|
| 11 |
+
"single_word": false,
|
| 12 |
+
"special": true
|
| 13 |
+
},
|
| 14 |
+
"1": {
|
| 15 |
+
"content": "<s>",
|
| 16 |
+
"lstrip": false,
|
| 17 |
+
"normalized": false,
|
| 18 |
+
"rstrip": false,
|
| 19 |
+
"single_word": false,
|
| 20 |
+
"special": true
|
| 21 |
+
},
|
| 22 |
+
"2": {
|
| 23 |
+
"content": "</s>",
|
| 24 |
+
"lstrip": false,
|
| 25 |
+
"normalized": false,
|
| 26 |
+
"rstrip": false,
|
| 27 |
+
"single_word": false,
|
| 28 |
+
"special": true
|
| 29 |
+
},
|
| 30 |
+
"32000": {
|
| 31 |
+
"content": "<|im_end|>",
|
| 32 |
+
"lstrip": false,
|
| 33 |
+
"normalized": false,
|
| 34 |
+
"rstrip": false,
|
| 35 |
+
"single_word": false,
|
| 36 |
+
"special": true
|
| 37 |
+
},
|
| 38 |
+
"32001": {
|
| 39 |
+
"content": "<|im_start|>",
|
| 40 |
+
"lstrip": false,
|
| 41 |
+
"normalized": false,
|
| 42 |
+
"rstrip": false,
|
| 43 |
+
"single_word": false,
|
| 44 |
+
"special": true
|
| 45 |
+
}
|
| 46 |
+
},
|
| 47 |
+
"additional_special_tokens": [],
|
| 48 |
+
"bos_token": "<s>",
|
| 49 |
+
"chat_template": "{% for message in messages %}{{'<|im_start|>' + message['role'] + '\n' + message['content'] + '<|im_end|>' + '\n'}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant\n' }}{% endif %}",
|
| 50 |
+
"clean_up_tokenization_spaces": false,
|
| 51 |
+
"eos_token": "<|im_end|>",
|
| 52 |
+
"legacy": true,
|
| 53 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 54 |
+
"pad_token": "<|im_end|>",
|
| 55 |
+
"sp_model_kwargs": {},
|
| 56 |
+
"spaces_between_special_tokens": false,
|
| 57 |
+
"tokenizer_class": "LlamaTokenizer",
|
| 58 |
+
"trust_remote_code": false,
|
| 59 |
+
"unk_token": "<unk>",
|
| 60 |
+
"use_default_system_prompt": true,
|
| 61 |
+
"use_fast": true
|
| 62 |
+
}
|
checkpoint-110/trainer_state.json
ADDED
|
@@ -0,0 +1,1683 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.003742323983193563,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 110,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 3.4021127119941484e-05,
|
| 13 |
+
"grad_norm": 33.091712951660156,
|
| 14 |
+
"learning_rate": 1.0000000000000002e-06,
|
| 15 |
+
"logits/chosen": -3.133824586868286,
|
| 16 |
+
"logits/rejected": -3.1356313228607178,
|
| 17 |
+
"logps/chosen": -317.8847961425781,
|
| 18 |
+
"logps/rejected": -306.8866271972656,
|
| 19 |
+
"loss": 0.8544,
|
| 20 |
+
"rewards/accuracies": 0.75,
|
| 21 |
+
"rewards/chosen": -0.10576057434082031,
|
| 22 |
+
"rewards/margins": 0.2207910716533661,
|
| 23 |
+
"rewards/rejected": -0.3265516459941864,
|
| 24 |
+
"step": 1
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 6.804225423988297e-05,
|
| 28 |
+
"grad_norm": 38.49541473388672,
|
| 29 |
+
"learning_rate": 2.0000000000000003e-06,
|
| 30 |
+
"logits/chosen": -3.1433358192443848,
|
| 31 |
+
"logits/rejected": -3.1124918460845947,
|
| 32 |
+
"logps/chosen": -212.93820190429688,
|
| 33 |
+
"logps/rejected": -198.22511291503906,
|
| 34 |
+
"loss": 1.02,
|
| 35 |
+
"rewards/accuracies": 0.5,
|
| 36 |
+
"rewards/chosen": -0.28638574481010437,
|
| 37 |
+
"rewards/margins": -0.37813109159469604,
|
| 38 |
+
"rewards/rejected": 0.0917452871799469,
|
| 39 |
+
"step": 2
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.00010206338135982445,
|
| 43 |
+
"grad_norm": 27.203516006469727,
|
| 44 |
+
"learning_rate": 3e-06,
|
| 45 |
+
"logits/chosen": -3.2104451656341553,
|
| 46 |
+
"logits/rejected": -3.2103281021118164,
|
| 47 |
+
"logps/chosen": -235.89254760742188,
|
| 48 |
+
"logps/rejected": -206.0758514404297,
|
| 49 |
+
"loss": 0.605,
|
| 50 |
+
"rewards/accuracies": 0.75,
|
| 51 |
+
"rewards/chosen": 0.40017586946487427,
|
| 52 |
+
"rewards/margins": 0.24887371063232422,
|
| 53 |
+
"rewards/rejected": 0.15130215883255005,
|
| 54 |
+
"step": 3
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 0.00013608450847976594,
|
| 58 |
+
"grad_norm": 26.42736053466797,
|
| 59 |
+
"learning_rate": 4.000000000000001e-06,
|
| 60 |
+
"logits/chosen": -3.0741634368896484,
|
| 61 |
+
"logits/rejected": -2.953117847442627,
|
| 62 |
+
"logps/chosen": -428.7364196777344,
|
| 63 |
+
"logps/rejected": -192.357666015625,
|
| 64 |
+
"loss": 0.4914,
|
| 65 |
+
"rewards/accuracies": 0.75,
|
| 66 |
+
"rewards/chosen": 0.4332427978515625,
|
| 67 |
+
"rewards/margins": 0.5804362893104553,
|
| 68 |
+
"rewards/rejected": -0.1471935510635376,
|
| 69 |
+
"step": 4
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 0.00017010563559970743,
|
| 73 |
+
"grad_norm": 35.623634338378906,
|
| 74 |
+
"learning_rate": 5e-06,
|
| 75 |
+
"logits/chosen": -2.985285758972168,
|
| 76 |
+
"logits/rejected": -2.8854246139526367,
|
| 77 |
+
"logps/chosen": -342.38604736328125,
|
| 78 |
+
"logps/rejected": -211.78237915039062,
|
| 79 |
+
"loss": 0.8273,
|
| 80 |
+
"rewards/accuracies": 0.25,
|
| 81 |
+
"rewards/chosen": -0.3475755751132965,
|
| 82 |
+
"rewards/margins": -0.19371166825294495,
|
| 83 |
+
"rewards/rejected": -0.15386392176151276,
|
| 84 |
+
"step": 5
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 0.0002041267627196489,
|
| 88 |
+
"grad_norm": 31.70409393310547,
|
| 89 |
+
"learning_rate": 6e-06,
|
| 90 |
+
"logits/chosen": -3.074589967727661,
|
| 91 |
+
"logits/rejected": -2.85581636428833,
|
| 92 |
+
"logps/chosen": -593.6435546875,
|
| 93 |
+
"logps/rejected": -259.5284729003906,
|
| 94 |
+
"loss": 0.5473,
|
| 95 |
+
"rewards/accuracies": 0.75,
|
| 96 |
+
"rewards/chosen": 0.2495414763689041,
|
| 97 |
+
"rewards/margins": 0.35888367891311646,
|
| 98 |
+
"rewards/rejected": -0.10934218764305115,
|
| 99 |
+
"step": 6
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 0.00023814788983959038,
|
| 103 |
+
"grad_norm": 32.41214370727539,
|
| 104 |
+
"learning_rate": 7.000000000000001e-06,
|
| 105 |
+
"logits/chosen": -2.976675033569336,
|
| 106 |
+
"logits/rejected": -2.978133201599121,
|
| 107 |
+
"logps/chosen": -218.11688232421875,
|
| 108 |
+
"logps/rejected": -231.7698974609375,
|
| 109 |
+
"loss": 0.7124,
|
| 110 |
+
"rewards/accuracies": 0.25,
|
| 111 |
+
"rewards/chosen": -0.08671779930591583,
|
| 112 |
+
"rewards/margins": 0.21950224041938782,
|
| 113 |
+
"rewards/rejected": -0.30622005462646484,
|
| 114 |
+
"step": 7
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.00027216901695953187,
|
| 118 |
+
"grad_norm": 27.637351989746094,
|
| 119 |
+
"learning_rate": 8.000000000000001e-06,
|
| 120 |
+
"logits/chosen": -2.997600793838501,
|
| 121 |
+
"logits/rejected": -3.1078853607177734,
|
| 122 |
+
"logps/chosen": -165.71444702148438,
|
| 123 |
+
"logps/rejected": -335.5938720703125,
|
| 124 |
+
"loss": 0.6085,
|
| 125 |
+
"rewards/accuracies": 0.75,
|
| 126 |
+
"rewards/chosen": -0.057393088936805725,
|
| 127 |
+
"rewards/margins": 0.34109365940093994,
|
| 128 |
+
"rewards/rejected": -0.3984867036342621,
|
| 129 |
+
"step": 8
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.00030619014407947336,
|
| 133 |
+
"grad_norm": 32.246559143066406,
|
| 134 |
+
"learning_rate": 9e-06,
|
| 135 |
+
"logits/chosen": -3.0760390758514404,
|
| 136 |
+
"logits/rejected": -2.938197135925293,
|
| 137 |
+
"logps/chosen": -463.1333312988281,
|
| 138 |
+
"logps/rejected": -173.99951171875,
|
| 139 |
+
"loss": 0.6398,
|
| 140 |
+
"rewards/accuracies": 0.5,
|
| 141 |
+
"rewards/chosen": 0.5581258535385132,
|
| 142 |
+
"rewards/margins": 0.7653559446334839,
|
| 143 |
+
"rewards/rejected": -0.20723000168800354,
|
| 144 |
+
"step": 9
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 0.00034021127119941485,
|
| 148 |
+
"grad_norm": 39.84626388549805,
|
| 149 |
+
"learning_rate": 1e-05,
|
| 150 |
+
"logits/chosen": -3.192596673965454,
|
| 151 |
+
"logits/rejected": -2.9751501083374023,
|
| 152 |
+
"logps/chosen": -473.14532470703125,
|
| 153 |
+
"logps/rejected": -110.90155029296875,
|
| 154 |
+
"loss": 0.9794,
|
| 155 |
+
"rewards/accuracies": 0.25,
|
| 156 |
+
"rewards/chosen": -0.3494076132774353,
|
| 157 |
+
"rewards/margins": -0.31546899676322937,
|
| 158 |
+
"rewards/rejected": -0.03393859416246414,
|
| 159 |
+
"step": 10
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 0.00037423239831935634,
|
| 163 |
+
"grad_norm": 39.444889068603516,
|
| 164 |
+
"learning_rate": 1.1000000000000001e-05,
|
| 165 |
+
"logits/chosen": -3.096883535385132,
|
| 166 |
+
"logits/rejected": -3.043616771697998,
|
| 167 |
+
"logps/chosen": -391.7703552246094,
|
| 168 |
+
"logps/rejected": -271.8331604003906,
|
| 169 |
+
"loss": 0.8029,
|
| 170 |
+
"rewards/accuracies": 0.75,
|
| 171 |
+
"rewards/chosen": 0.6240158081054688,
|
| 172 |
+
"rewards/margins": 0.4651091992855072,
|
| 173 |
+
"rewards/rejected": 0.15890654921531677,
|
| 174 |
+
"step": 11
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 0.0004082535254392978,
|
| 178 |
+
"grad_norm": 40.78356170654297,
|
| 179 |
+
"learning_rate": 1.2e-05,
|
| 180 |
+
"logits/chosen": -3.000021457672119,
|
| 181 |
+
"logits/rejected": -3.0893115997314453,
|
| 182 |
+
"logps/chosen": -255.94607543945312,
|
| 183 |
+
"logps/rejected": -405.3093566894531,
|
| 184 |
+
"loss": 0.6297,
|
| 185 |
+
"rewards/accuracies": 0.5,
|
| 186 |
+
"rewards/chosen": 0.33697396516799927,
|
| 187 |
+
"rewards/margins": 0.3315456509590149,
|
| 188 |
+
"rewards/rejected": 0.005428321659564972,
|
| 189 |
+
"step": 12
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 0.00044227465255923927,
|
| 193 |
+
"grad_norm": 35.69674301147461,
|
| 194 |
+
"learning_rate": 1.3000000000000001e-05,
|
| 195 |
+
"logits/chosen": -3.011014699935913,
|
| 196 |
+
"logits/rejected": -3.0214414596557617,
|
| 197 |
+
"logps/chosen": -223.4480438232422,
|
| 198 |
+
"logps/rejected": -261.62469482421875,
|
| 199 |
+
"loss": 0.815,
|
| 200 |
+
"rewards/accuracies": 0.5,
|
| 201 |
+
"rewards/chosen": 0.7416718006134033,
|
| 202 |
+
"rewards/margins": 0.19626998901367188,
|
| 203 |
+
"rewards/rejected": 0.5454018115997314,
|
| 204 |
+
"step": 13
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 0.00047629577967918076,
|
| 208 |
+
"grad_norm": 32.88405990600586,
|
| 209 |
+
"learning_rate": 1.4000000000000001e-05,
|
| 210 |
+
"logits/chosen": -3.065549373626709,
|
| 211 |
+
"logits/rejected": -3.1094508171081543,
|
| 212 |
+
"logps/chosen": -169.08047485351562,
|
| 213 |
+
"logps/rejected": -227.85162353515625,
|
| 214 |
+
"loss": 0.8115,
|
| 215 |
+
"rewards/accuracies": 0.25,
|
| 216 |
+
"rewards/chosen": 0.37659722566604614,
|
| 217 |
+
"rewards/margins": -0.11390496790409088,
|
| 218 |
+
"rewards/rejected": 0.4905022084712982,
|
| 219 |
+
"step": 14
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.0005103169067991223,
|
| 223 |
+
"grad_norm": 20.918012619018555,
|
| 224 |
+
"learning_rate": 1.5e-05,
|
| 225 |
+
"logits/chosen": -3.045846462249756,
|
| 226 |
+
"logits/rejected": -2.9114990234375,
|
| 227 |
+
"logps/chosen": -573.1253662109375,
|
| 228 |
+
"logps/rejected": -273.9326477050781,
|
| 229 |
+
"loss": 0.3105,
|
| 230 |
+
"rewards/accuracies": 1.0,
|
| 231 |
+
"rewards/chosen": 0.7755428552627563,
|
| 232 |
+
"rewards/margins": 1.082038164138794,
|
| 233 |
+
"rewards/rejected": -0.3064952790737152,
|
| 234 |
+
"step": 15
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.0005443380339190637,
|
| 238 |
+
"grad_norm": 38.434749603271484,
|
| 239 |
+
"learning_rate": 1.6000000000000003e-05,
|
| 240 |
+
"logits/chosen": -3.1250810623168945,
|
| 241 |
+
"logits/rejected": -2.96527099609375,
|
| 242 |
+
"logps/chosen": -396.9795837402344,
|
| 243 |
+
"logps/rejected": -174.92918395996094,
|
| 244 |
+
"loss": 0.9971,
|
| 245 |
+
"rewards/accuracies": 0.5,
|
| 246 |
+
"rewards/chosen": 0.2244594693183899,
|
| 247 |
+
"rewards/margins": -0.37520748376846313,
|
| 248 |
+
"rewards/rejected": 0.5996670126914978,
|
| 249 |
+
"step": 16
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 0.0005783591610390052,
|
| 253 |
+
"grad_norm": 37.71885681152344,
|
| 254 |
+
"learning_rate": 1.7000000000000003e-05,
|
| 255 |
+
"logits/chosen": -2.970050096511841,
|
| 256 |
+
"logits/rejected": -3.0346829891204834,
|
| 257 |
+
"logps/chosen": -236.67071533203125,
|
| 258 |
+
"logps/rejected": -403.0212097167969,
|
| 259 |
+
"loss": 1.0307,
|
| 260 |
+
"rewards/accuracies": 0.5,
|
| 261 |
+
"rewards/chosen": 0.9259212613105774,
|
| 262 |
+
"rewards/margins": -0.23367927968502045,
|
| 263 |
+
"rewards/rejected": 1.1596004962921143,
|
| 264 |
+
"step": 17
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 0.0006123802881589467,
|
| 268 |
+
"grad_norm": 25.964290618896484,
|
| 269 |
+
"learning_rate": 1.8e-05,
|
| 270 |
+
"logits/chosen": -3.0690722465515137,
|
| 271 |
+
"logits/rejected": -3.0348947048187256,
|
| 272 |
+
"logps/chosen": -312.81317138671875,
|
| 273 |
+
"logps/rejected": -259.230712890625,
|
| 274 |
+
"loss": 0.6989,
|
| 275 |
+
"rewards/accuracies": 0.75,
|
| 276 |
+
"rewards/chosen": 0.6348915100097656,
|
| 277 |
+
"rewards/margins": 0.6418477892875671,
|
| 278 |
+
"rewards/rejected": -0.006956290453672409,
|
| 279 |
+
"step": 18
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 0.0006464014152788882,
|
| 283 |
+
"grad_norm": 27.076108932495117,
|
| 284 |
+
"learning_rate": 1.9e-05,
|
| 285 |
+
"logits/chosen": -3.135136842727661,
|
| 286 |
+
"logits/rejected": -2.911625862121582,
|
| 287 |
+
"logps/chosen": -473.7462158203125,
|
| 288 |
+
"logps/rejected": -141.9491729736328,
|
| 289 |
+
"loss": 0.6027,
|
| 290 |
+
"rewards/accuracies": 0.5,
|
| 291 |
+
"rewards/chosen": 0.9424793720245361,
|
| 292 |
+
"rewards/margins": 0.9380611181259155,
|
| 293 |
+
"rewards/rejected": 0.00441819429397583,
|
| 294 |
+
"step": 19
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 0.0006804225423988297,
|
| 298 |
+
"grad_norm": 51.820159912109375,
|
| 299 |
+
"learning_rate": 2e-05,
|
| 300 |
+
"logits/chosen": -2.8403027057647705,
|
| 301 |
+
"logits/rejected": -3.0664803981781006,
|
| 302 |
+
"logps/chosen": -219.79153442382812,
|
| 303 |
+
"logps/rejected": -231.70004272460938,
|
| 304 |
+
"loss": 1.1022,
|
| 305 |
+
"rewards/accuracies": 0.0,
|
| 306 |
+
"rewards/chosen": 0.052263081073760986,
|
| 307 |
+
"rewards/margins": -0.6957410573959351,
|
| 308 |
+
"rewards/rejected": 0.7480041980743408,
|
| 309 |
+
"step": 20
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 0.0007144436695187712,
|
| 313 |
+
"grad_norm": 45.9146614074707,
|
| 314 |
+
"learning_rate": 2.1e-05,
|
| 315 |
+
"logits/chosen": -2.9764764308929443,
|
| 316 |
+
"logits/rejected": -2.9391565322875977,
|
| 317 |
+
"logps/chosen": -458.0147705078125,
|
| 318 |
+
"logps/rejected": -218.37648010253906,
|
| 319 |
+
"loss": 1.0772,
|
| 320 |
+
"rewards/accuracies": 0.5,
|
| 321 |
+
"rewards/chosen": 0.8049196004867554,
|
| 322 |
+
"rewards/margins": -0.448437362909317,
|
| 323 |
+
"rewards/rejected": 1.2533570528030396,
|
| 324 |
+
"step": 21
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.0007484647966387127,
|
| 328 |
+
"grad_norm": 22.854082107543945,
|
| 329 |
+
"learning_rate": 2.2000000000000003e-05,
|
| 330 |
+
"logits/chosen": -3.1029181480407715,
|
| 331 |
+
"logits/rejected": -3.044419527053833,
|
| 332 |
+
"logps/chosen": -154.8139190673828,
|
| 333 |
+
"logps/rejected": -155.980712890625,
|
| 334 |
+
"loss": 0.4065,
|
| 335 |
+
"rewards/accuracies": 1.0,
|
| 336 |
+
"rewards/chosen": 0.45841503143310547,
|
| 337 |
+
"rewards/margins": 0.8753025531768799,
|
| 338 |
+
"rewards/rejected": -0.41688746213912964,
|
| 339 |
+
"step": 22
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.0007824859237586541,
|
| 343 |
+
"grad_norm": 43.94650650024414,
|
| 344 |
+
"learning_rate": 2.3000000000000003e-05,
|
| 345 |
+
"logits/chosen": -2.9984076023101807,
|
| 346 |
+
"logits/rejected": -3.114562511444092,
|
| 347 |
+
"logps/chosen": -263.8316650390625,
|
| 348 |
+
"logps/rejected": -340.1427307128906,
|
| 349 |
+
"loss": 1.0427,
|
| 350 |
+
"rewards/accuracies": 0.25,
|
| 351 |
+
"rewards/chosen": 0.6933631896972656,
|
| 352 |
+
"rewards/margins": -0.14637237787246704,
|
| 353 |
+
"rewards/rejected": 0.8397356271743774,
|
| 354 |
+
"step": 23
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 0.0008165070508785956,
|
| 358 |
+
"grad_norm": 49.11759948730469,
|
| 359 |
+
"learning_rate": 2.4e-05,
|
| 360 |
+
"logits/chosen": -3.025609016418457,
|
| 361 |
+
"logits/rejected": -3.0236239433288574,
|
| 362 |
+
"logps/chosen": -412.2383117675781,
|
| 363 |
+
"logps/rejected": -643.138427734375,
|
| 364 |
+
"loss": 1.0374,
|
| 365 |
+
"rewards/accuracies": 0.5,
|
| 366 |
+
"rewards/chosen": 1.1175220012664795,
|
| 367 |
+
"rewards/margins": -0.31503716111183167,
|
| 368 |
+
"rewards/rejected": 1.4325592517852783,
|
| 369 |
+
"step": 24
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 0.000850528177998537,
|
| 373 |
+
"grad_norm": 37.57383728027344,
|
| 374 |
+
"learning_rate": 2.5e-05,
|
| 375 |
+
"logits/chosen": -3.024618625640869,
|
| 376 |
+
"logits/rejected": -3.123701572418213,
|
| 377 |
+
"logps/chosen": -201.70530700683594,
|
| 378 |
+
"logps/rejected": -224.99038696289062,
|
| 379 |
+
"loss": 0.7477,
|
| 380 |
+
"rewards/accuracies": 0.25,
|
| 381 |
+
"rewards/chosen": 0.9206753373146057,
|
| 382 |
+
"rewards/margins": 0.5947309732437134,
|
| 383 |
+
"rewards/rejected": 0.32594436407089233,
|
| 384 |
+
"step": 25
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 0.0008845493051184785,
|
| 388 |
+
"grad_norm": 40.297794342041016,
|
| 389 |
+
"learning_rate": 2.6000000000000002e-05,
|
| 390 |
+
"logits/chosen": -3.1100215911865234,
|
| 391 |
+
"logits/rejected": -3.0195024013519287,
|
| 392 |
+
"logps/chosen": -226.2320098876953,
|
| 393 |
+
"logps/rejected": -193.0676727294922,
|
| 394 |
+
"loss": 0.8151,
|
| 395 |
+
"rewards/accuracies": 0.5,
|
| 396 |
+
"rewards/chosen": 0.9122087359428406,
|
| 397 |
+
"rewards/margins": 0.0203973650932312,
|
| 398 |
+
"rewards/rejected": 0.8918113708496094,
|
| 399 |
+
"step": 26
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 0.00091857043223842,
|
| 403 |
+
"grad_norm": 43.394229888916016,
|
| 404 |
+
"learning_rate": 2.7000000000000002e-05,
|
| 405 |
+
"logits/chosen": -3.0187630653381348,
|
| 406 |
+
"logits/rejected": -3.053678035736084,
|
| 407 |
+
"logps/chosen": -153.02883911132812,
|
| 408 |
+
"logps/rejected": -260.0540771484375,
|
| 409 |
+
"loss": 1.3457,
|
| 410 |
+
"rewards/accuracies": 0.5,
|
| 411 |
+
"rewards/chosen": 0.7145153284072876,
|
| 412 |
+
"rewards/margins": -0.806611180305481,
|
| 413 |
+
"rewards/rejected": 1.5211265087127686,
|
| 414 |
+
"step": 27
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 0.0009525915593583615,
|
| 418 |
+
"grad_norm": 28.59680938720703,
|
| 419 |
+
"learning_rate": 2.8000000000000003e-05,
|
| 420 |
+
"logits/chosen": -3.032409191131592,
|
| 421 |
+
"logits/rejected": -3.0502161979675293,
|
| 422 |
+
"logps/chosen": -240.1644287109375,
|
| 423 |
+
"logps/rejected": -245.81436157226562,
|
| 424 |
+
"loss": 0.5501,
|
| 425 |
+
"rewards/accuracies": 0.75,
|
| 426 |
+
"rewards/chosen": 2.0826761722564697,
|
| 427 |
+
"rewards/margins": 0.3433869779109955,
|
| 428 |
+
"rewards/rejected": 1.7392891645431519,
|
| 429 |
+
"step": 28
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.000986612686478303,
|
| 433 |
+
"grad_norm": 37.069583892822266,
|
| 434 |
+
"learning_rate": 2.9e-05,
|
| 435 |
+
"logits/chosen": -3.0516104698181152,
|
| 436 |
+
"logits/rejected": -3.144774913787842,
|
| 437 |
+
"logps/chosen": -296.03125,
|
| 438 |
+
"logps/rejected": -292.9052429199219,
|
| 439 |
+
"loss": 1.1087,
|
| 440 |
+
"rewards/accuracies": 0.75,
|
| 441 |
+
"rewards/chosen": 0.8324771523475647,
|
| 442 |
+
"rewards/margins": -0.34529268741607666,
|
| 443 |
+
"rewards/rejected": 1.1777698993682861,
|
| 444 |
+
"step": 29
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.0010206338135982445,
|
| 448 |
+
"grad_norm": 30.238658905029297,
|
| 449 |
+
"learning_rate": 3e-05,
|
| 450 |
+
"logits/chosen": -3.064671039581299,
|
| 451 |
+
"logits/rejected": -2.8963613510131836,
|
| 452 |
+
"logps/chosen": -415.6085510253906,
|
| 453 |
+
"logps/rejected": -149.9669952392578,
|
| 454 |
+
"loss": 0.4892,
|
| 455 |
+
"rewards/accuracies": 0.75,
|
| 456 |
+
"rewards/chosen": 1.0249770879745483,
|
| 457 |
+
"rewards/margins": 0.7549355030059814,
|
| 458 |
+
"rewards/rejected": 0.27004164457321167,
|
| 459 |
+
"step": 30
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 0.001054654940718186,
|
| 463 |
+
"grad_norm": 40.35402297973633,
|
| 464 |
+
"learning_rate": 3.1e-05,
|
| 465 |
+
"logits/chosen": -3.1762304306030273,
|
| 466 |
+
"logits/rejected": -3.061676025390625,
|
| 467 |
+
"logps/chosen": -585.9200439453125,
|
| 468 |
+
"logps/rejected": -238.16307067871094,
|
| 469 |
+
"loss": 0.587,
|
| 470 |
+
"rewards/accuracies": 0.5,
|
| 471 |
+
"rewards/chosen": 2.726571559906006,
|
| 472 |
+
"rewards/margins": 0.9808880090713501,
|
| 473 |
+
"rewards/rejected": 1.7456837892532349,
|
| 474 |
+
"step": 31
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 0.0010886760678381275,
|
| 478 |
+
"grad_norm": 33.78396987915039,
|
| 479 |
+
"learning_rate": 3.2000000000000005e-05,
|
| 480 |
+
"logits/chosen": -3.05906081199646,
|
| 481 |
+
"logits/rejected": -2.9525132179260254,
|
| 482 |
+
"logps/chosen": -580.87646484375,
|
| 483 |
+
"logps/rejected": -223.65457153320312,
|
| 484 |
+
"loss": 0.8173,
|
| 485 |
+
"rewards/accuracies": 0.5,
|
| 486 |
+
"rewards/chosen": 1.3172895908355713,
|
| 487 |
+
"rewards/margins": 0.07103176414966583,
|
| 488 |
+
"rewards/rejected": 1.2462577819824219,
|
| 489 |
+
"step": 32
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 0.001122697194958069,
|
| 493 |
+
"grad_norm": 40.97666549682617,
|
| 494 |
+
"learning_rate": 3.3e-05,
|
| 495 |
+
"logits/chosen": -3.084476947784424,
|
| 496 |
+
"logits/rejected": -3.11270809173584,
|
| 497 |
+
"logps/chosen": -285.040283203125,
|
| 498 |
+
"logps/rejected": -251.92686462402344,
|
| 499 |
+
"loss": 1.0645,
|
| 500 |
+
"rewards/accuracies": 0.5,
|
| 501 |
+
"rewards/chosen": 1.4826446771621704,
|
| 502 |
+
"rewards/margins": -0.13718131184577942,
|
| 503 |
+
"rewards/rejected": 1.6198259592056274,
|
| 504 |
+
"step": 33
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 0.0011567183220780105,
|
| 508 |
+
"grad_norm": 33.32075881958008,
|
| 509 |
+
"learning_rate": 3.4000000000000007e-05,
|
| 510 |
+
"logits/chosen": -3.037482500076294,
|
| 511 |
+
"logits/rejected": -3.0169787406921387,
|
| 512 |
+
"logps/chosen": -467.7403869628906,
|
| 513 |
+
"logps/rejected": -308.62274169921875,
|
| 514 |
+
"loss": 0.5226,
|
| 515 |
+
"rewards/accuracies": 0.5,
|
| 516 |
+
"rewards/chosen": 2.4038407802581787,
|
| 517 |
+
"rewards/margins": 0.8624236583709717,
|
| 518 |
+
"rewards/rejected": 1.541416883468628,
|
| 519 |
+
"step": 34
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 0.001190739449197952,
|
| 523 |
+
"grad_norm": 26.35866928100586,
|
| 524 |
+
"learning_rate": 3.5e-05,
|
| 525 |
+
"logits/chosen": -2.969816207885742,
|
| 526 |
+
"logits/rejected": -2.9069106578826904,
|
| 527 |
+
"logps/chosen": -312.856689453125,
|
| 528 |
+
"logps/rejected": -183.8594512939453,
|
| 529 |
+
"loss": 0.4495,
|
| 530 |
+
"rewards/accuracies": 0.75,
|
| 531 |
+
"rewards/chosen": 1.1757181882858276,
|
| 532 |
+
"rewards/margins": 0.803225576877594,
|
| 533 |
+
"rewards/rejected": 0.37249261140823364,
|
| 534 |
+
"step": 35
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.0012247605763178934,
|
| 538 |
+
"grad_norm": 44.118099212646484,
|
| 539 |
+
"learning_rate": 3.6e-05,
|
| 540 |
+
"logits/chosen": -3.091865301132202,
|
| 541 |
+
"logits/rejected": -3.0135951042175293,
|
| 542 |
+
"logps/chosen": -724.1255493164062,
|
| 543 |
+
"logps/rejected": -335.2945861816406,
|
| 544 |
+
"loss": 0.6341,
|
| 545 |
+
"rewards/accuracies": 0.75,
|
| 546 |
+
"rewards/chosen": 1.7523369789123535,
|
| 547 |
+
"rewards/margins": 0.3912923336029053,
|
| 548 |
+
"rewards/rejected": 1.3610445261001587,
|
| 549 |
+
"step": 36
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.001258781703437835,
|
| 553 |
+
"grad_norm": 26.80854606628418,
|
| 554 |
+
"learning_rate": 3.7e-05,
|
| 555 |
+
"logits/chosen": -3.0334794521331787,
|
| 556 |
+
"logits/rejected": -2.9878833293914795,
|
| 557 |
+
"logps/chosen": -406.20745849609375,
|
| 558 |
+
"logps/rejected": -377.2811584472656,
|
| 559 |
+
"loss": 0.5577,
|
| 560 |
+
"rewards/accuracies": 0.75,
|
| 561 |
+
"rewards/chosen": 2.6232261657714844,
|
| 562 |
+
"rewards/margins": 0.9015346765518188,
|
| 563 |
+
"rewards/rejected": 1.7216914892196655,
|
| 564 |
+
"step": 37
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 0.0012928028305577764,
|
| 568 |
+
"grad_norm": 31.345367431640625,
|
| 569 |
+
"learning_rate": 3.8e-05,
|
| 570 |
+
"logits/chosen": -3.0986504554748535,
|
| 571 |
+
"logits/rejected": -3.062534809112549,
|
| 572 |
+
"logps/chosen": -412.54498291015625,
|
| 573 |
+
"logps/rejected": -386.44158935546875,
|
| 574 |
+
"loss": 0.5479,
|
| 575 |
+
"rewards/accuracies": 0.75,
|
| 576 |
+
"rewards/chosen": 1.985120177268982,
|
| 577 |
+
"rewards/margins": 0.8609979748725891,
|
| 578 |
+
"rewards/rejected": 1.1241222620010376,
|
| 579 |
+
"step": 38
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 0.001326823957677718,
|
| 583 |
+
"grad_norm": 15.26794719696045,
|
| 584 |
+
"learning_rate": 3.9000000000000006e-05,
|
| 585 |
+
"logits/chosen": -3.0730984210968018,
|
| 586 |
+
"logits/rejected": -3.0951075553894043,
|
| 587 |
+
"logps/chosen": -305.12982177734375,
|
| 588 |
+
"logps/rejected": -285.45001220703125,
|
| 589 |
+
"loss": 0.2547,
|
| 590 |
+
"rewards/accuracies": 1.0,
|
| 591 |
+
"rewards/chosen": 2.55714750289917,
|
| 592 |
+
"rewards/margins": 1.3561341762542725,
|
| 593 |
+
"rewards/rejected": 1.2010130882263184,
|
| 594 |
+
"step": 39
|
| 595 |
+
},
|
| 596 |
+
{
|
| 597 |
+
"epoch": 0.0013608450847976594,
|
| 598 |
+
"grad_norm": 22.881174087524414,
|
| 599 |
+
"learning_rate": 4e-05,
|
| 600 |
+
"logits/chosen": -3.162843942642212,
|
| 601 |
+
"logits/rejected": -3.15628981590271,
|
| 602 |
+
"logps/chosen": -277.48626708984375,
|
| 603 |
+
"logps/rejected": -219.9507293701172,
|
| 604 |
+
"loss": 0.5427,
|
| 605 |
+
"rewards/accuracies": 0.5,
|
| 606 |
+
"rewards/chosen": 2.6990911960601807,
|
| 607 |
+
"rewards/margins": 0.5041837692260742,
|
| 608 |
+
"rewards/rejected": 2.1949074268341064,
|
| 609 |
+
"step": 40
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 0.001394866211917601,
|
| 613 |
+
"grad_norm": 15.179062843322754,
|
| 614 |
+
"learning_rate": 4.1e-05,
|
| 615 |
+
"logits/chosen": -3.138144016265869,
|
| 616 |
+
"logits/rejected": -3.1033153533935547,
|
| 617 |
+
"logps/chosen": -234.6917724609375,
|
| 618 |
+
"logps/rejected": -141.10888671875,
|
| 619 |
+
"loss": 0.2626,
|
| 620 |
+
"rewards/accuracies": 1.0,
|
| 621 |
+
"rewards/chosen": 2.799344539642334,
|
| 622 |
+
"rewards/margins": 1.4650464057922363,
|
| 623 |
+
"rewards/rejected": 1.3342978954315186,
|
| 624 |
+
"step": 41
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"epoch": 0.0014288873390375424,
|
| 628 |
+
"grad_norm": 21.09362030029297,
|
| 629 |
+
"learning_rate": 4.2e-05,
|
| 630 |
+
"logits/chosen": -3.0385375022888184,
|
| 631 |
+
"logits/rejected": -3.083676338195801,
|
| 632 |
+
"logps/chosen": -168.33697509765625,
|
| 633 |
+
"logps/rejected": -179.30813598632812,
|
| 634 |
+
"loss": 0.4002,
|
| 635 |
+
"rewards/accuracies": 1.0,
|
| 636 |
+
"rewards/chosen": 2.3925881385803223,
|
| 637 |
+
"rewards/margins": 0.8477088809013367,
|
| 638 |
+
"rewards/rejected": 1.5448791980743408,
|
| 639 |
+
"step": 42
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.0014629084661574839,
|
| 643 |
+
"grad_norm": 38.255069732666016,
|
| 644 |
+
"learning_rate": 4.3e-05,
|
| 645 |
+
"logits/chosen": -3.080360174179077,
|
| 646 |
+
"logits/rejected": -3.029470443725586,
|
| 647 |
+
"logps/chosen": -267.5482177734375,
|
| 648 |
+
"logps/rejected": -161.17111206054688,
|
| 649 |
+
"loss": 0.8857,
|
| 650 |
+
"rewards/accuracies": 0.5,
|
| 651 |
+
"rewards/chosen": 1.8575999736785889,
|
| 652 |
+
"rewards/margins": -0.24116593599319458,
|
| 653 |
+
"rewards/rejected": 2.0987658500671387,
|
| 654 |
+
"step": 43
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.0014969295932774254,
|
| 658 |
+
"grad_norm": 27.97095489501953,
|
| 659 |
+
"learning_rate": 4.4000000000000006e-05,
|
| 660 |
+
"logits/chosen": -3.1322476863861084,
|
| 661 |
+
"logits/rejected": -3.1334149837493896,
|
| 662 |
+
"logps/chosen": -94.13690185546875,
|
| 663 |
+
"logps/rejected": -78.98738098144531,
|
| 664 |
+
"loss": 0.9987,
|
| 665 |
+
"rewards/accuracies": 0.5,
|
| 666 |
+
"rewards/chosen": 1.6137428283691406,
|
| 667 |
+
"rewards/margins": -0.4127828776836395,
|
| 668 |
+
"rewards/rejected": 2.0265257358551025,
|
| 669 |
+
"step": 44
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 0.0015309507203973669,
|
| 673 |
+
"grad_norm": 45.25028991699219,
|
| 674 |
+
"learning_rate": 4.5e-05,
|
| 675 |
+
"logits/chosen": -3.121274709701538,
|
| 676 |
+
"logits/rejected": -3.1236798763275146,
|
| 677 |
+
"logps/chosen": -516.3858642578125,
|
| 678 |
+
"logps/rejected": -212.694580078125,
|
| 679 |
+
"loss": 0.9457,
|
| 680 |
+
"rewards/accuracies": 0.25,
|
| 681 |
+
"rewards/chosen": 2.419821262359619,
|
| 682 |
+
"rewards/margins": -0.38284340500831604,
|
| 683 |
+
"rewards/rejected": 2.8026647567749023,
|
| 684 |
+
"step": 45
|
| 685 |
+
},
|
| 686 |
+
{
|
| 687 |
+
"epoch": 0.0015649718475173081,
|
| 688 |
+
"grad_norm": 23.58992576599121,
|
| 689 |
+
"learning_rate": 4.600000000000001e-05,
|
| 690 |
+
"logits/chosen": -3.113922357559204,
|
| 691 |
+
"logits/rejected": -3.0506935119628906,
|
| 692 |
+
"logps/chosen": -318.288330078125,
|
| 693 |
+
"logps/rejected": -124.70735168457031,
|
| 694 |
+
"loss": 0.5671,
|
| 695 |
+
"rewards/accuracies": 0.75,
|
| 696 |
+
"rewards/chosen": 2.407118320465088,
|
| 697 |
+
"rewards/margins": 0.30923187732696533,
|
| 698 |
+
"rewards/rejected": 2.097886323928833,
|
| 699 |
+
"step": 46
|
| 700 |
+
},
|
| 701 |
+
{
|
| 702 |
+
"epoch": 0.0015989929746372496,
|
| 703 |
+
"grad_norm": 36.93040084838867,
|
| 704 |
+
"learning_rate": 4.7e-05,
|
| 705 |
+
"logits/chosen": -3.079291582107544,
|
| 706 |
+
"logits/rejected": -3.1158454418182373,
|
| 707 |
+
"logps/chosen": -246.26065063476562,
|
| 708 |
+
"logps/rejected": -336.46649169921875,
|
| 709 |
+
"loss": 0.9352,
|
| 710 |
+
"rewards/accuracies": 0.25,
|
| 711 |
+
"rewards/chosen": 2.214073657989502,
|
| 712 |
+
"rewards/margins": -0.2623659074306488,
|
| 713 |
+
"rewards/rejected": 2.4764394760131836,
|
| 714 |
+
"step": 47
|
| 715 |
+
},
|
| 716 |
+
{
|
| 717 |
+
"epoch": 0.0016330141017571911,
|
| 718 |
+
"grad_norm": 32.5983772277832,
|
| 719 |
+
"learning_rate": 4.8e-05,
|
| 720 |
+
"logits/chosen": -3.155709981918335,
|
| 721 |
+
"logits/rejected": -3.1829142570495605,
|
| 722 |
+
"logps/chosen": -204.55328369140625,
|
| 723 |
+
"logps/rejected": -219.52647399902344,
|
| 724 |
+
"loss": 0.8805,
|
| 725 |
+
"rewards/accuracies": 0.5,
|
| 726 |
+
"rewards/chosen": 2.155839681625366,
|
| 727 |
+
"rewards/margins": -0.002133697271347046,
|
| 728 |
+
"rewards/rejected": 2.157973289489746,
|
| 729 |
+
"step": 48
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 0.0016670352288771326,
|
| 733 |
+
"grad_norm": 35.626869201660156,
|
| 734 |
+
"learning_rate": 4.9e-05,
|
| 735 |
+
"logits/chosen": -3.10532546043396,
|
| 736 |
+
"logits/rejected": -3.113755464553833,
|
| 737 |
+
"logps/chosen": -390.5479736328125,
|
| 738 |
+
"logps/rejected": -351.7720031738281,
|
| 739 |
+
"loss": 0.7657,
|
| 740 |
+
"rewards/accuracies": 0.5,
|
| 741 |
+
"rewards/chosen": 2.5670056343078613,
|
| 742 |
+
"rewards/margins": -0.04349440336227417,
|
| 743 |
+
"rewards/rejected": 2.6105000972747803,
|
| 744 |
+
"step": 49
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.001701056355997074,
|
| 748 |
+
"grad_norm": 37.51042938232422,
|
| 749 |
+
"learning_rate": 5e-05,
|
| 750 |
+
"logits/chosen": -3.122190237045288,
|
| 751 |
+
"logits/rejected": -3.0837321281433105,
|
| 752 |
+
"logps/chosen": -251.68600463867188,
|
| 753 |
+
"logps/rejected": -242.4346923828125,
|
| 754 |
+
"loss": 0.9539,
|
| 755 |
+
"rewards/accuracies": 0.5,
|
| 756 |
+
"rewards/chosen": 2.0532169342041016,
|
| 757 |
+
"rewards/margins": -0.32518482208251953,
|
| 758 |
+
"rewards/rejected": 2.378401756286621,
|
| 759 |
+
"step": 50
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 0.0017350774831170156,
|
| 763 |
+
"grad_norm": 43.16423416137695,
|
| 764 |
+
"learning_rate": 4.999939076763487e-05,
|
| 765 |
+
"logits/chosen": -3.0424001216888428,
|
| 766 |
+
"logits/rejected": -3.0246453285217285,
|
| 767 |
+
"logps/chosen": -228.8073272705078,
|
| 768 |
+
"logps/rejected": -399.57781982421875,
|
| 769 |
+
"loss": 1.204,
|
| 770 |
+
"rewards/accuracies": 0.5,
|
| 771 |
+
"rewards/chosen": 2.1423535346984863,
|
| 772 |
+
"rewards/margins": -0.628890335559845,
|
| 773 |
+
"rewards/rejected": 2.7712438106536865,
|
| 774 |
+
"step": 51
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 0.001769098610236957,
|
| 778 |
+
"grad_norm": 32.16966247558594,
|
| 779 |
+
"learning_rate": 4.999756310023261e-05,
|
| 780 |
+
"logits/chosen": -3.049051284790039,
|
| 781 |
+
"logits/rejected": -3.061192750930786,
|
| 782 |
+
"logps/chosen": -219.27503967285156,
|
| 783 |
+
"logps/rejected": -238.98226928710938,
|
| 784 |
+
"loss": 0.9616,
|
| 785 |
+
"rewards/accuracies": 0.5,
|
| 786 |
+
"rewards/chosen": 2.1868481636047363,
|
| 787 |
+
"rewards/margins": -0.2618050277233124,
|
| 788 |
+
"rewards/rejected": 2.448652744293213,
|
| 789 |
+
"step": 52
|
| 790 |
+
},
|
| 791 |
+
{
|
| 792 |
+
"epoch": 0.0018031197373568986,
|
| 793 |
+
"grad_norm": 38.5218505859375,
|
| 794 |
+
"learning_rate": 4.999451708687114e-05,
|
| 795 |
+
"logits/chosen": -3.088409185409546,
|
| 796 |
+
"logits/rejected": -3.0845394134521484,
|
| 797 |
+
"logps/chosen": -131.70074462890625,
|
| 798 |
+
"logps/rejected": -205.93624877929688,
|
| 799 |
+
"loss": 1.0419,
|
| 800 |
+
"rewards/accuracies": 0.5,
|
| 801 |
+
"rewards/chosen": 1.822286605834961,
|
| 802 |
+
"rewards/margins": -0.18211710453033447,
|
| 803 |
+
"rewards/rejected": 2.004403829574585,
|
| 804 |
+
"step": 53
|
| 805 |
+
},
|
| 806 |
+
{
|
| 807 |
+
"epoch": 0.00183714086447684,
|
| 808 |
+
"grad_norm": 31.46576690673828,
|
| 809 |
+
"learning_rate": 4.999025287600886e-05,
|
| 810 |
+
"logits/chosen": -3.143968105316162,
|
| 811 |
+
"logits/rejected": -3.032649278640747,
|
| 812 |
+
"logps/chosen": -403.4815673828125,
|
| 813 |
+
"logps/rejected": -195.69281005859375,
|
| 814 |
+
"loss": 0.567,
|
| 815 |
+
"rewards/accuracies": 0.75,
|
| 816 |
+
"rewards/chosen": 2.2466557025909424,
|
| 817 |
+
"rewards/margins": 0.6339659094810486,
|
| 818 |
+
"rewards/rejected": 1.612689733505249,
|
| 819 |
+
"step": 54
|
| 820 |
+
},
|
| 821 |
+
{
|
| 822 |
+
"epoch": 0.0018711619915967816,
|
| 823 |
+
"grad_norm": 22.339370727539062,
|
| 824 |
+
"learning_rate": 4.99847706754774e-05,
|
| 825 |
+
"logits/chosen": -3.1364409923553467,
|
| 826 |
+
"logits/rejected": -3.0678646564483643,
|
| 827 |
+
"logps/chosen": -272.07330322265625,
|
| 828 |
+
"logps/rejected": -94.81082153320312,
|
| 829 |
+
"loss": 0.4856,
|
| 830 |
+
"rewards/accuracies": 1.0,
|
| 831 |
+
"rewards/chosen": 1.9992039203643799,
|
| 832 |
+
"rewards/margins": 0.5075141191482544,
|
| 833 |
+
"rewards/rejected": 1.491689682006836,
|
| 834 |
+
"step": 55
|
| 835 |
+
},
|
| 836 |
+
{
|
| 837 |
+
"epoch": 0.001905183118716723,
|
| 838 |
+
"grad_norm": 38.88309860229492,
|
| 839 |
+
"learning_rate": 4.997807075247146e-05,
|
| 840 |
+
"logits/chosen": -3.0754241943359375,
|
| 841 |
+
"logits/rejected": -3.1445722579956055,
|
| 842 |
+
"logps/chosen": -161.1916961669922,
|
| 843 |
+
"logps/rejected": -230.43589782714844,
|
| 844 |
+
"loss": 0.8925,
|
| 845 |
+
"rewards/accuracies": 0.5,
|
| 846 |
+
"rewards/chosen": 2.070554256439209,
|
| 847 |
+
"rewards/margins": -0.027551889419555664,
|
| 848 |
+
"rewards/rejected": 2.0981061458587646,
|
| 849 |
+
"step": 56
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.0019392042458366645,
|
| 853 |
+
"grad_norm": 29.006982803344727,
|
| 854 |
+
"learning_rate": 4.997015343353585e-05,
|
| 855 |
+
"logits/chosen": -3.079509973526001,
|
| 856 |
+
"logits/rejected": -3.119426727294922,
|
| 857 |
+
"logps/chosen": -236.40664672851562,
|
| 858 |
+
"logps/rejected": -259.257568359375,
|
| 859 |
+
"loss": 0.5812,
|
| 860 |
+
"rewards/accuracies": 0.75,
|
| 861 |
+
"rewards/chosen": 2.4261646270751953,
|
| 862 |
+
"rewards/margins": 0.7371253371238708,
|
| 863 |
+
"rewards/rejected": 1.6890392303466797,
|
| 864 |
+
"step": 57
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 0.001973225372956606,
|
| 868 |
+
"grad_norm": 18.193532943725586,
|
| 869 |
+
"learning_rate": 4.996101910454953e-05,
|
| 870 |
+
"logits/chosen": -3.070995569229126,
|
| 871 |
+
"logits/rejected": -2.8317549228668213,
|
| 872 |
+
"logps/chosen": -730.03759765625,
|
| 873 |
+
"logps/rejected": -133.17929077148438,
|
| 874 |
+
"loss": 0.3401,
|
| 875 |
+
"rewards/accuracies": 0.75,
|
| 876 |
+
"rewards/chosen": 3.711276054382324,
|
| 877 |
+
"rewards/margins": 1.7915010452270508,
|
| 878 |
+
"rewards/rejected": 1.9197750091552734,
|
| 879 |
+
"step": 58
|
| 880 |
+
},
|
| 881 |
+
{
|
| 882 |
+
"epoch": 0.0020072465000765477,
|
| 883 |
+
"grad_norm": 35.50777053833008,
|
| 884 |
+
"learning_rate": 4.995066821070679e-05,
|
| 885 |
+
"logits/chosen": -3.093824863433838,
|
| 886 |
+
"logits/rejected": -3.083796739578247,
|
| 887 |
+
"logps/chosen": -238.11883544921875,
|
| 888 |
+
"logps/rejected": -151.69790649414062,
|
| 889 |
+
"loss": 0.6251,
|
| 890 |
+
"rewards/accuracies": 0.75,
|
| 891 |
+
"rewards/chosen": 2.0124969482421875,
|
| 892 |
+
"rewards/margins": 0.640554666519165,
|
| 893 |
+
"rewards/rejected": 1.371942400932312,
|
| 894 |
+
"step": 59
|
| 895 |
+
},
|
| 896 |
+
{
|
| 897 |
+
"epoch": 0.002041267627196489,
|
| 898 |
+
"grad_norm": 40.34638977050781,
|
| 899 |
+
"learning_rate": 4.993910125649561e-05,
|
| 900 |
+
"logits/chosen": -3.1637063026428223,
|
| 901 |
+
"logits/rejected": -3.055285930633545,
|
| 902 |
+
"logps/chosen": -365.89434814453125,
|
| 903 |
+
"logps/rejected": -187.11392211914062,
|
| 904 |
+
"loss": 1.1514,
|
| 905 |
+
"rewards/accuracies": 0.75,
|
| 906 |
+
"rewards/chosen": 2.089510679244995,
|
| 907 |
+
"rewards/margins": -0.3508464992046356,
|
| 908 |
+
"rewards/rejected": 2.440357208251953,
|
| 909 |
+
"step": 60
|
| 910 |
+
},
|
| 911 |
+
{
|
| 912 |
+
"epoch": 0.0020752887543164307,
|
| 913 |
+
"grad_norm": 31.27332878112793,
|
| 914 |
+
"learning_rate": 4.992631880567301e-05,
|
| 915 |
+
"logits/chosen": -2.9931535720825195,
|
| 916 |
+
"logits/rejected": -2.994088649749756,
|
| 917 |
+
"logps/chosen": -127.87042236328125,
|
| 918 |
+
"logps/rejected": -236.82327270507812,
|
| 919 |
+
"loss": 0.8745,
|
| 920 |
+
"rewards/accuracies": 0.5,
|
| 921 |
+
"rewards/chosen": 1.9324662685394287,
|
| 922 |
+
"rewards/margins": -0.040864601731300354,
|
| 923 |
+
"rewards/rejected": 1.9733309745788574,
|
| 924 |
+
"step": 61
|
| 925 |
+
},
|
| 926 |
+
{
|
| 927 |
+
"epoch": 0.002109309881436372,
|
| 928 |
+
"grad_norm": 24.1131591796875,
|
| 929 |
+
"learning_rate": 4.991232148123761e-05,
|
| 930 |
+
"logits/chosen": -3.094635009765625,
|
| 931 |
+
"logits/rejected": -3.060533046722412,
|
| 932 |
+
"logps/chosen": -369.5483093261719,
|
| 933 |
+
"logps/rejected": -167.15155029296875,
|
| 934 |
+
"loss": 0.46,
|
| 935 |
+
"rewards/accuracies": 0.75,
|
| 936 |
+
"rewards/chosen": 2.38918399810791,
|
| 937 |
+
"rewards/margins": 1.1014902591705322,
|
| 938 |
+
"rewards/rejected": 1.2876936197280884,
|
| 939 |
+
"step": 62
|
| 940 |
+
},
|
| 941 |
+
{
|
| 942 |
+
"epoch": 0.0021433310085563133,
|
| 943 |
+
"grad_norm": 49.346961975097656,
|
| 944 |
+
"learning_rate": 4.989710996539926e-05,
|
| 945 |
+
"logits/chosen": -3.1134653091430664,
|
| 946 |
+
"logits/rejected": -3.0323257446289062,
|
| 947 |
+
"logps/chosen": -273.39361572265625,
|
| 948 |
+
"logps/rejected": -157.84652709960938,
|
| 949 |
+
"loss": 1.023,
|
| 950 |
+
"rewards/accuracies": 0.25,
|
| 951 |
+
"rewards/chosen": 1.1458964347839355,
|
| 952 |
+
"rewards/margins": -0.5081260204315186,
|
| 953 |
+
"rewards/rejected": 1.654022455215454,
|
| 954 |
+
"step": 63
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.002177352135676255,
|
| 958 |
+
"grad_norm": 41.145633697509766,
|
| 959 |
+
"learning_rate": 4.988068499954578e-05,
|
| 960 |
+
"logits/chosen": -2.9834160804748535,
|
| 961 |
+
"logits/rejected": -2.9958994388580322,
|
| 962 |
+
"logps/chosen": -334.5246887207031,
|
| 963 |
+
"logps/rejected": -386.3387145996094,
|
| 964 |
+
"loss": 1.5388,
|
| 965 |
+
"rewards/accuracies": 0.5,
|
| 966 |
+
"rewards/chosen": 2.305959939956665,
|
| 967 |
+
"rewards/margins": 0.07403802871704102,
|
| 968 |
+
"rewards/rejected": 2.231921911239624,
|
| 969 |
+
"step": 64
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 0.0022113732627961962,
|
| 973 |
+
"grad_norm": 33.07326126098633,
|
| 974 |
+
"learning_rate": 4.9863047384206835e-05,
|
| 975 |
+
"logits/chosen": -3.0006370544433594,
|
| 976 |
+
"logits/rejected": -3.020918846130371,
|
| 977 |
+
"logps/chosen": -344.7704162597656,
|
| 978 |
+
"logps/rejected": -200.45114135742188,
|
| 979 |
+
"loss": 0.5845,
|
| 980 |
+
"rewards/accuracies": 0.75,
|
| 981 |
+
"rewards/chosen": 1.643819808959961,
|
| 982 |
+
"rewards/margins": 0.6079027652740479,
|
| 983 |
+
"rewards/rejected": 1.035917043685913,
|
| 984 |
+
"step": 65
|
| 985 |
+
},
|
| 986 |
+
{
|
| 987 |
+
"epoch": 0.002245394389916138,
|
| 988 |
+
"grad_norm": 51.295345306396484,
|
| 989 |
+
"learning_rate": 4.984419797901491e-05,
|
| 990 |
+
"logits/chosen": -3.039350986480713,
|
| 991 |
+
"logits/rejected": -3.030729293823242,
|
| 992 |
+
"logps/chosen": -307.3952331542969,
|
| 993 |
+
"logps/rejected": -238.56224060058594,
|
| 994 |
+
"loss": 1.2331,
|
| 995 |
+
"rewards/accuracies": 0.25,
|
| 996 |
+
"rewards/chosen": 0.6686372756958008,
|
| 997 |
+
"rewards/margins": -0.6910279989242554,
|
| 998 |
+
"rewards/rejected": 1.3596653938293457,
|
| 999 |
+
"step": 66
|
| 1000 |
+
},
|
| 1001 |
+
{
|
| 1002 |
+
"epoch": 0.0022794155170360792,
|
| 1003 |
+
"grad_norm": 27.732166290283203,
|
| 1004 |
+
"learning_rate": 4.982413770266342e-05,
|
| 1005 |
+
"logits/chosen": -3.092078685760498,
|
| 1006 |
+
"logits/rejected": -3.1104204654693604,
|
| 1007 |
+
"logps/chosen": -206.7743682861328,
|
| 1008 |
+
"logps/rejected": -112.66790771484375,
|
| 1009 |
+
"loss": 0.5552,
|
| 1010 |
+
"rewards/accuracies": 0.5,
|
| 1011 |
+
"rewards/chosen": 1.273049235343933,
|
| 1012 |
+
"rewards/margins": 0.48671725392341614,
|
| 1013 |
+
"rewards/rejected": 0.7863319516181946,
|
| 1014 |
+
"step": 67
|
| 1015 |
+
},
|
| 1016 |
+
{
|
| 1017 |
+
"epoch": 0.002313436644156021,
|
| 1018 |
+
"grad_norm": 46.03068161010742,
|
| 1019 |
+
"learning_rate": 4.980286753286195e-05,
|
| 1020 |
+
"logits/chosen": -2.9219119548797607,
|
| 1021 |
+
"logits/rejected": -2.973864793777466,
|
| 1022 |
+
"logps/chosen": -455.3492431640625,
|
| 1023 |
+
"logps/rejected": -422.7337646484375,
|
| 1024 |
+
"loss": 0.8967,
|
| 1025 |
+
"rewards/accuracies": 0.5,
|
| 1026 |
+
"rewards/chosen": 1.788267731666565,
|
| 1027 |
+
"rewards/margins": -0.24189908802509308,
|
| 1028 |
+
"rewards/rejected": 2.0301668643951416,
|
| 1029 |
+
"step": 68
|
| 1030 |
+
},
|
| 1031 |
+
{
|
| 1032 |
+
"epoch": 0.002347457771275962,
|
| 1033 |
+
"grad_norm": 24.625539779663086,
|
| 1034 |
+
"learning_rate": 4.978038850628854e-05,
|
| 1035 |
+
"logits/chosen": -2.9824914932250977,
|
| 1036 |
+
"logits/rejected": -2.923306703567505,
|
| 1037 |
+
"logps/chosen": -320.2785339355469,
|
| 1038 |
+
"logps/rejected": -199.95407104492188,
|
| 1039 |
+
"loss": 0.5542,
|
| 1040 |
+
"rewards/accuracies": 0.75,
|
| 1041 |
+
"rewards/chosen": 2.4030070304870605,
|
| 1042 |
+
"rewards/margins": 1.2227776050567627,
|
| 1043 |
+
"rewards/rejected": 1.180229663848877,
|
| 1044 |
+
"step": 69
|
| 1045 |
+
},
|
| 1046 |
+
{
|
| 1047 |
+
"epoch": 0.002381478898395904,
|
| 1048 |
+
"grad_norm": 52.87644958496094,
|
| 1049 |
+
"learning_rate": 4.975670171853926e-05,
|
| 1050 |
+
"logits/chosen": -2.9576680660247803,
|
| 1051 |
+
"logits/rejected": -3.113696336746216,
|
| 1052 |
+
"logps/chosen": -226.1188201904297,
|
| 1053 |
+
"logps/rejected": -472.0656433105469,
|
| 1054 |
+
"loss": 1.5976,
|
| 1055 |
+
"rewards/accuracies": 0.25,
|
| 1056 |
+
"rewards/chosen": 1.8890628814697266,
|
| 1057 |
+
"rewards/margins": -1.2377982139587402,
|
| 1058 |
+
"rewards/rejected": 3.126861095428467,
|
| 1059 |
+
"step": 70
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.002415500025515845,
|
| 1063 |
+
"grad_norm": 41.89684295654297,
|
| 1064 |
+
"learning_rate": 4.9731808324074717e-05,
|
| 1065 |
+
"logits/chosen": -2.963118553161621,
|
| 1066 |
+
"logits/rejected": -2.95782732963562,
|
| 1067 |
+
"logps/chosen": -376.9123229980469,
|
| 1068 |
+
"logps/rejected": -408.6007080078125,
|
| 1069 |
+
"loss": 0.8305,
|
| 1070 |
+
"rewards/accuracies": 0.5,
|
| 1071 |
+
"rewards/chosen": 2.2748968601226807,
|
| 1072 |
+
"rewards/margins": 1.4728538990020752,
|
| 1073 |
+
"rewards/rejected": 0.8020429611206055,
|
| 1074 |
+
"step": 71
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 0.002449521152635787,
|
| 1078 |
+
"grad_norm": 33.66620635986328,
|
| 1079 |
+
"learning_rate": 4.9705709536163824e-05,
|
| 1080 |
+
"logits/chosen": -2.951019287109375,
|
| 1081 |
+
"logits/rejected": -2.8525712490081787,
|
| 1082 |
+
"logps/chosen": -334.75689697265625,
|
| 1083 |
+
"logps/rejected": -256.43963623046875,
|
| 1084 |
+
"loss": 0.5852,
|
| 1085 |
+
"rewards/accuracies": 0.75,
|
| 1086 |
+
"rewards/chosen": 2.2168543338775635,
|
| 1087 |
+
"rewards/margins": 1.3764129877090454,
|
| 1088 |
+
"rewards/rejected": 0.8404413461685181,
|
| 1089 |
+
"step": 72
|
| 1090 |
+
},
|
| 1091 |
+
{
|
| 1092 |
+
"epoch": 0.002483542279755728,
|
| 1093 |
+
"grad_norm": 48.537376403808594,
|
| 1094 |
+
"learning_rate": 4.96784066268247e-05,
|
| 1095 |
+
"logits/chosen": -2.96649169921875,
|
| 1096 |
+
"logits/rejected": -2.877800703048706,
|
| 1097 |
+
"logps/chosen": -342.9018859863281,
|
| 1098 |
+
"logps/rejected": -166.73826599121094,
|
| 1099 |
+
"loss": 1.3147,
|
| 1100 |
+
"rewards/accuracies": 0.0,
|
| 1101 |
+
"rewards/chosen": 1.4513309001922607,
|
| 1102 |
+
"rewards/margins": -0.9881281852722168,
|
| 1103 |
+
"rewards/rejected": 2.4394590854644775,
|
| 1104 |
+
"step": 73
|
| 1105 |
+
},
|
| 1106 |
+
{
|
| 1107 |
+
"epoch": 0.00251756340687567,
|
| 1108 |
+
"grad_norm": 35.8758544921875,
|
| 1109 |
+
"learning_rate": 4.964990092676263e-05,
|
| 1110 |
+
"logits/chosen": -3.007376194000244,
|
| 1111 |
+
"logits/rejected": -3.0710248947143555,
|
| 1112 |
+
"logps/chosen": -322.75604248046875,
|
| 1113 |
+
"logps/rejected": -288.9136047363281,
|
| 1114 |
+
"loss": 1.0483,
|
| 1115 |
+
"rewards/accuracies": 0.5,
|
| 1116 |
+
"rewards/chosen": 1.328976035118103,
|
| 1117 |
+
"rewards/margins": -0.057621777057647705,
|
| 1118 |
+
"rewards/rejected": 1.3865978717803955,
|
| 1119 |
+
"step": 74
|
| 1120 |
+
},
|
| 1121 |
+
{
|
| 1122 |
+
"epoch": 0.002551584533995611,
|
| 1123 |
+
"grad_norm": 52.42377471923828,
|
| 1124 |
+
"learning_rate": 4.962019382530521e-05,
|
| 1125 |
+
"logits/chosen": -2.952801465988159,
|
| 1126 |
+
"logits/rejected": -2.9580321311950684,
|
| 1127 |
+
"logps/chosen": -272.15576171875,
|
| 1128 |
+
"logps/rejected": -230.2147674560547,
|
| 1129 |
+
"loss": 1.4341,
|
| 1130 |
+
"rewards/accuracies": 0.25,
|
| 1131 |
+
"rewards/chosen": 1.2654911279678345,
|
| 1132 |
+
"rewards/margins": -0.9303737282752991,
|
| 1133 |
+
"rewards/rejected": 2.1958649158477783,
|
| 1134 |
+
"step": 75
|
| 1135 |
+
},
|
| 1136 |
+
{
|
| 1137 |
+
"epoch": 0.002585605661115553,
|
| 1138 |
+
"grad_norm": 54.157386779785156,
|
| 1139 |
+
"learning_rate": 4.9589286770334654e-05,
|
| 1140 |
+
"logits/chosen": -2.997375965118408,
|
| 1141 |
+
"logits/rejected": -3.091611385345459,
|
| 1142 |
+
"logps/chosen": -219.73159790039062,
|
| 1143 |
+
"logps/rejected": -503.797119140625,
|
| 1144 |
+
"loss": 1.3568,
|
| 1145 |
+
"rewards/accuracies": 0.25,
|
| 1146 |
+
"rewards/chosen": 1.520772933959961,
|
| 1147 |
+
"rewards/margins": -0.4154503643512726,
|
| 1148 |
+
"rewards/rejected": 1.9362232685089111,
|
| 1149 |
+
"step": 76
|
| 1150 |
+
},
|
| 1151 |
+
{
|
| 1152 |
+
"epoch": 0.002619626788235494,
|
| 1153 |
+
"grad_norm": 40.647274017333984,
|
| 1154 |
+
"learning_rate": 4.9557181268217227e-05,
|
| 1155 |
+
"logits/chosen": -3.0128493309020996,
|
| 1156 |
+
"logits/rejected": -2.9981260299682617,
|
| 1157 |
+
"logps/chosen": -319.1977233886719,
|
| 1158 |
+
"logps/rejected": -250.11123657226562,
|
| 1159 |
+
"loss": 0.7922,
|
| 1160 |
+
"rewards/accuracies": 0.5,
|
| 1161 |
+
"rewards/chosen": 1.9633705615997314,
|
| 1162 |
+
"rewards/margins": -0.027412444353103638,
|
| 1163 |
+
"rewards/rejected": 1.9907830953598022,
|
| 1164 |
+
"step": 77
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.002653647915355436,
|
| 1168 |
+
"grad_norm": 46.99341583251953,
|
| 1169 |
+
"learning_rate": 4.952387888372979e-05,
|
| 1170 |
+
"logits/chosen": -3.0307843685150146,
|
| 1171 |
+
"logits/rejected": -3.070122480392456,
|
| 1172 |
+
"logps/chosen": -134.52757263183594,
|
| 1173 |
+
"logps/rejected": -225.51242065429688,
|
| 1174 |
+
"loss": 1.182,
|
| 1175 |
+
"rewards/accuracies": 0.0,
|
| 1176 |
+
"rewards/chosen": 1.2856706380844116,
|
| 1177 |
+
"rewards/margins": -0.7930065393447876,
|
| 1178 |
+
"rewards/rejected": 2.078677177429199,
|
| 1179 |
+
"step": 78
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 0.002687669042475377,
|
| 1183 |
+
"grad_norm": 26.92460823059082,
|
| 1184 |
+
"learning_rate": 4.94893812399836e-05,
|
| 1185 |
+
"logits/chosen": -2.995112895965576,
|
| 1186 |
+
"logits/rejected": -2.950528144836426,
|
| 1187 |
+
"logps/chosen": -560.2005615234375,
|
| 1188 |
+
"logps/rejected": -383.92523193359375,
|
| 1189 |
+
"loss": 0.5477,
|
| 1190 |
+
"rewards/accuracies": 0.75,
|
| 1191 |
+
"rewards/chosen": 1.8620712757110596,
|
| 1192 |
+
"rewards/margins": 0.3632301390171051,
|
| 1193 |
+
"rewards/rejected": 1.4988411664962769,
|
| 1194 |
+
"step": 79
|
| 1195 |
+
},
|
| 1196 |
+
{
|
| 1197 |
+
"epoch": 0.002721690169595319,
|
| 1198 |
+
"grad_norm": 18.80535125732422,
|
| 1199 |
+
"learning_rate": 4.9453690018345144e-05,
|
| 1200 |
+
"logits/chosen": -3.0887956619262695,
|
| 1201 |
+
"logits/rejected": -3.013925790786743,
|
| 1202 |
+
"logps/chosen": -201.7890167236328,
|
| 1203 |
+
"logps/rejected": -120.17092895507812,
|
| 1204 |
+
"loss": 0.6049,
|
| 1205 |
+
"rewards/accuracies": 0.5,
|
| 1206 |
+
"rewards/chosen": 2.207493305206299,
|
| 1207 |
+
"rewards/margins": 0.5624717473983765,
|
| 1208 |
+
"rewards/rejected": 1.645021677017212,
|
| 1209 |
+
"step": 80
|
| 1210 |
+
},
|
| 1211 |
+
{
|
| 1212 |
+
"epoch": 0.00275571129671526,
|
| 1213 |
+
"grad_norm": 35.741817474365234,
|
| 1214 |
+
"learning_rate": 4.94168069583542e-05,
|
| 1215 |
+
"logits/chosen": -3.087347984313965,
|
| 1216 |
+
"logits/rejected": -2.9956464767456055,
|
| 1217 |
+
"logps/chosen": -485.24749755859375,
|
| 1218 |
+
"logps/rejected": -269.6015930175781,
|
| 1219 |
+
"loss": 1.0357,
|
| 1220 |
+
"rewards/accuracies": 0.5,
|
| 1221 |
+
"rewards/chosen": 1.650765299797058,
|
| 1222 |
+
"rewards/margins": -0.448674738407135,
|
| 1223 |
+
"rewards/rejected": 2.099440097808838,
|
| 1224 |
+
"step": 81
|
| 1225 |
+
},
|
| 1226 |
+
{
|
| 1227 |
+
"epoch": 0.002789732423835202,
|
| 1228 |
+
"grad_norm": 25.36280059814453,
|
| 1229 |
+
"learning_rate": 4.937873385763908e-05,
|
| 1230 |
+
"logits/chosen": -2.9839279651641846,
|
| 1231 |
+
"logits/rejected": -3.0026051998138428,
|
| 1232 |
+
"logps/chosen": -221.44424438476562,
|
| 1233 |
+
"logps/rejected": -193.8582305908203,
|
| 1234 |
+
"loss": 0.9423,
|
| 1235 |
+
"rewards/accuracies": 0.5,
|
| 1236 |
+
"rewards/chosen": 1.4655239582061768,
|
| 1237 |
+
"rewards/margins": -0.0549183189868927,
|
| 1238 |
+
"rewards/rejected": 1.520442247390747,
|
| 1239 |
+
"step": 82
|
| 1240 |
+
},
|
| 1241 |
+
{
|
| 1242 |
+
"epoch": 0.002823753550955143,
|
| 1243 |
+
"grad_norm": 12.532795906066895,
|
| 1244 |
+
"learning_rate": 4.933947257182901e-05,
|
| 1245 |
+
"logits/chosen": -3.0249555110931396,
|
| 1246 |
+
"logits/rejected": -3.0148353576660156,
|
| 1247 |
+
"logps/chosen": -386.8382263183594,
|
| 1248 |
+
"logps/rejected": -106.46695709228516,
|
| 1249 |
+
"loss": 0.3462,
|
| 1250 |
+
"rewards/accuracies": 0.75,
|
| 1251 |
+
"rewards/chosen": 3.745790481567383,
|
| 1252 |
+
"rewards/margins": 1.7120976448059082,
|
| 1253 |
+
"rewards/rejected": 2.0336928367614746,
|
| 1254 |
+
"step": 83
|
| 1255 |
+
},
|
| 1256 |
+
{
|
| 1257 |
+
"epoch": 0.002857774678075085,
|
| 1258 |
+
"grad_norm": 39.60961151123047,
|
| 1259 |
+
"learning_rate": 4.929902501446366e-05,
|
| 1260 |
+
"logits/chosen": -3.058694362640381,
|
| 1261 |
+
"logits/rejected": -3.0021092891693115,
|
| 1262 |
+
"logps/chosen": -269.97113037109375,
|
| 1263 |
+
"logps/rejected": -247.43728637695312,
|
| 1264 |
+
"loss": 1.5676,
|
| 1265 |
+
"rewards/accuracies": 0.0,
|
| 1266 |
+
"rewards/chosen": 1.230128526687622,
|
| 1267 |
+
"rewards/margins": -1.297512412071228,
|
| 1268 |
+
"rewards/rejected": 2.5276408195495605,
|
| 1269 |
+
"step": 84
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 0.002891795805195026,
|
| 1273 |
+
"grad_norm": 19.17630958557129,
|
| 1274 |
+
"learning_rate": 4.925739315689991e-05,
|
| 1275 |
+
"logits/chosen": -3.1123204231262207,
|
| 1276 |
+
"logits/rejected": -3.076711416244507,
|
| 1277 |
+
"logps/chosen": -323.8536376953125,
|
| 1278 |
+
"logps/rejected": -180.48394775390625,
|
| 1279 |
+
"loss": 0.5561,
|
| 1280 |
+
"rewards/accuracies": 0.75,
|
| 1281 |
+
"rewards/chosen": 2.0740671157836914,
|
| 1282 |
+
"rewards/margins": 0.7356327772140503,
|
| 1283 |
+
"rewards/rejected": 1.3384342193603516,
|
| 1284 |
+
"step": 85
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 0.0029258169323149678,
|
| 1288 |
+
"grad_norm": 15.29592227935791,
|
| 1289 |
+
"learning_rate": 4.9214579028215776e-05,
|
| 1290 |
+
"logits/chosen": -2.995969772338867,
|
| 1291 |
+
"logits/rejected": -2.975437879562378,
|
| 1292 |
+
"logps/chosen": -198.14756774902344,
|
| 1293 |
+
"logps/rejected": -97.39586639404297,
|
| 1294 |
+
"loss": 0.4645,
|
| 1295 |
+
"rewards/accuracies": 1.0,
|
| 1296 |
+
"rewards/chosen": 2.481008529663086,
|
| 1297 |
+
"rewards/margins": 0.6242896914482117,
|
| 1298 |
+
"rewards/rejected": 1.8567187786102295,
|
| 1299 |
+
"step": 86
|
| 1300 |
+
},
|
| 1301 |
+
{
|
| 1302 |
+
"epoch": 0.002959838059434909,
|
| 1303 |
+
"grad_norm": 37.80276870727539,
|
| 1304 |
+
"learning_rate": 4.917058471511149e-05,
|
| 1305 |
+
"logits/chosen": -3.0480551719665527,
|
| 1306 |
+
"logits/rejected": -3.0806996822357178,
|
| 1307 |
+
"logps/chosen": -224.3466796875,
|
| 1308 |
+
"logps/rejected": -420.4508361816406,
|
| 1309 |
+
"loss": 1.1967,
|
| 1310 |
+
"rewards/accuracies": 0.25,
|
| 1311 |
+
"rewards/chosen": 1.6259909868240356,
|
| 1312 |
+
"rewards/margins": -0.6426498889923096,
|
| 1313 |
+
"rewards/rejected": 2.2686409950256348,
|
| 1314 |
+
"step": 87
|
| 1315 |
+
},
|
| 1316 |
+
{
|
| 1317 |
+
"epoch": 0.0029938591865548508,
|
| 1318 |
+
"grad_norm": 20.813907623291016,
|
| 1319 |
+
"learning_rate": 4.912541236180779e-05,
|
| 1320 |
+
"logits/chosen": -3.045354127883911,
|
| 1321 |
+
"logits/rejected": -3.0551252365112305,
|
| 1322 |
+
"logps/chosen": -190.39498901367188,
|
| 1323 |
+
"logps/rejected": -417.9426574707031,
|
| 1324 |
+
"loss": 0.4819,
|
| 1325 |
+
"rewards/accuracies": 0.75,
|
| 1326 |
+
"rewards/chosen": 2.733407497406006,
|
| 1327 |
+
"rewards/margins": 0.5519695281982422,
|
| 1328 |
+
"rewards/rejected": 2.1814379692077637,
|
| 1329 |
+
"step": 88
|
| 1330 |
+
},
|
| 1331 |
+
{
|
| 1332 |
+
"epoch": 0.003027880313674792,
|
| 1333 |
+
"grad_norm": 18.250917434692383,
|
| 1334 |
+
"learning_rate": 4.907906416994146e-05,
|
| 1335 |
+
"logits/chosen": -3.1118156909942627,
|
| 1336 |
+
"logits/rejected": -3.124022960662842,
|
| 1337 |
+
"logps/chosen": -423.84698486328125,
|
| 1338 |
+
"logps/rejected": -217.9359130859375,
|
| 1339 |
+
"loss": 0.3913,
|
| 1340 |
+
"rewards/accuracies": 0.75,
|
| 1341 |
+
"rewards/chosen": 2.9902727603912354,
|
| 1342 |
+
"rewards/margins": 1.3608616590499878,
|
| 1343 |
+
"rewards/rejected": 1.629410982131958,
|
| 1344 |
+
"step": 89
|
| 1345 |
+
},
|
| 1346 |
+
{
|
| 1347 |
+
"epoch": 0.0030619014407947337,
|
| 1348 |
+
"grad_norm": 18.630115509033203,
|
| 1349 |
+
"learning_rate": 4.9031542398457974e-05,
|
| 1350 |
+
"logits/chosen": -3.0090408325195312,
|
| 1351 |
+
"logits/rejected": -3.0369369983673096,
|
| 1352 |
+
"logps/chosen": -435.40478515625,
|
| 1353 |
+
"logps/rejected": -191.86627197265625,
|
| 1354 |
+
"loss": 0.7079,
|
| 1355 |
+
"rewards/accuracies": 0.75,
|
| 1356 |
+
"rewards/chosen": 2.947678327560425,
|
| 1357 |
+
"rewards/margins": 0.8492887020111084,
|
| 1358 |
+
"rewards/rejected": 2.0983893871307373,
|
| 1359 |
+
"step": 90
|
| 1360 |
+
},
|
| 1361 |
+
{
|
| 1362 |
+
"epoch": 0.003095922567914675,
|
| 1363 |
+
"grad_norm": 19.552978515625,
|
| 1364 |
+
"learning_rate": 4.898284936350144e-05,
|
| 1365 |
+
"logits/chosen": -3.0316624641418457,
|
| 1366 |
+
"logits/rejected": -3.011552572250366,
|
| 1367 |
+
"logps/chosen": -432.43402099609375,
|
| 1368 |
+
"logps/rejected": -214.85935974121094,
|
| 1369 |
+
"loss": 0.7816,
|
| 1370 |
+
"rewards/accuracies": 0.5,
|
| 1371 |
+
"rewards/chosen": 2.7960076332092285,
|
| 1372 |
+
"rewards/margins": 0.5724576115608215,
|
| 1373 |
+
"rewards/rejected": 2.2235498428344727,
|
| 1374 |
+
"step": 91
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 0.0031299436950346163,
|
| 1378 |
+
"grad_norm": 7.452019691467285,
|
| 1379 |
+
"learning_rate": 4.893298743830168e-05,
|
| 1380 |
+
"logits/chosen": -3.038140296936035,
|
| 1381 |
+
"logits/rejected": -3.00756573677063,
|
| 1382 |
+
"logps/chosen": -168.3916473388672,
|
| 1383 |
+
"logps/rejected": -119.79571533203125,
|
| 1384 |
+
"loss": 0.2097,
|
| 1385 |
+
"rewards/accuracies": 1.0,
|
| 1386 |
+
"rewards/chosen": 2.8402457237243652,
|
| 1387 |
+
"rewards/margins": 1.5049405097961426,
|
| 1388 |
+
"rewards/rejected": 1.3353054523468018,
|
| 1389 |
+
"step": 92
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 0.003163964822154558,
|
| 1393 |
+
"grad_norm": 16.27883529663086,
|
| 1394 |
+
"learning_rate": 4.888195905305859e-05,
|
| 1395 |
+
"logits/chosen": -3.030277729034424,
|
| 1396 |
+
"logits/rejected": -3.0196878910064697,
|
| 1397 |
+
"logps/chosen": -295.2430114746094,
|
| 1398 |
+
"logps/rejected": -229.96961975097656,
|
| 1399 |
+
"loss": 0.5378,
|
| 1400 |
+
"rewards/accuracies": 0.5,
|
| 1401 |
+
"rewards/chosen": 2.0215210914611816,
|
| 1402 |
+
"rewards/margins": 0.6857559084892273,
|
| 1403 |
+
"rewards/rejected": 1.3357652425765991,
|
| 1404 |
+
"step": 93
|
| 1405 |
+
},
|
| 1406 |
+
{
|
| 1407 |
+
"epoch": 0.0031979859492744993,
|
| 1408 |
+
"grad_norm": 27.870994567871094,
|
| 1409 |
+
"learning_rate": 4.882976669482367e-05,
|
| 1410 |
+
"logits/chosen": -3.0157508850097656,
|
| 1411 |
+
"logits/rejected": -3.062469005584717,
|
| 1412 |
+
"logps/chosen": -594.1408081054688,
|
| 1413 |
+
"logps/rejected": -284.8695068359375,
|
| 1414 |
+
"loss": 0.829,
|
| 1415 |
+
"rewards/accuracies": 0.25,
|
| 1416 |
+
"rewards/chosen": 2.395214796066284,
|
| 1417 |
+
"rewards/margins": -0.06876510381698608,
|
| 1418 |
+
"rewards/rejected": 2.463979959487915,
|
| 1419 |
+
"step": 94
|
| 1420 |
+
},
|
| 1421 |
+
{
|
| 1422 |
+
"epoch": 0.003232007076394441,
|
| 1423 |
+
"grad_norm": 34.57992172241211,
|
| 1424 |
+
"learning_rate": 4.877641290737884e-05,
|
| 1425 |
+
"logits/chosen": -3.0241267681121826,
|
| 1426 |
+
"logits/rejected": -3.072436809539795,
|
| 1427 |
+
"logps/chosen": -206.46072387695312,
|
| 1428 |
+
"logps/rejected": -490.99163818359375,
|
| 1429 |
+
"loss": 0.9507,
|
| 1430 |
+
"rewards/accuracies": 0.25,
|
| 1431 |
+
"rewards/chosen": 3.0148510932922363,
|
| 1432 |
+
"rewards/margins": -0.41296035051345825,
|
| 1433 |
+
"rewards/rejected": 3.427811622619629,
|
| 1434 |
+
"step": 95
|
| 1435 |
+
},
|
| 1436 |
+
{
|
| 1437 |
+
"epoch": 0.0032660282035143822,
|
| 1438 |
+
"grad_norm": 31.449804306030273,
|
| 1439 |
+
"learning_rate": 4.8721900291112415e-05,
|
| 1440 |
+
"logits/chosen": -2.978365659713745,
|
| 1441 |
+
"logits/rejected": -2.9975037574768066,
|
| 1442 |
+
"logps/chosen": -353.8739929199219,
|
| 1443 |
+
"logps/rejected": -265.1501159667969,
|
| 1444 |
+
"loss": 0.7224,
|
| 1445 |
+
"rewards/accuracies": 0.5,
|
| 1446 |
+
"rewards/chosen": 3.355400562286377,
|
| 1447 |
+
"rewards/margins": 0.41021838784217834,
|
| 1448 |
+
"rewards/rejected": 2.9451825618743896,
|
| 1449 |
+
"step": 96
|
| 1450 |
+
},
|
| 1451 |
+
{
|
| 1452 |
+
"epoch": 0.003300049330634324,
|
| 1453 |
+
"grad_norm": 20.54676628112793,
|
| 1454 |
+
"learning_rate": 4.8666231502892415e-05,
|
| 1455 |
+
"logits/chosen": -3.0715813636779785,
|
| 1456 |
+
"logits/rejected": -3.0223703384399414,
|
| 1457 |
+
"logps/chosen": -417.56280517578125,
|
| 1458 |
+
"logps/rejected": -126.94705963134766,
|
| 1459 |
+
"loss": 0.5149,
|
| 1460 |
+
"rewards/accuracies": 0.75,
|
| 1461 |
+
"rewards/chosen": 2.667041540145874,
|
| 1462 |
+
"rewards/margins": 0.9312877655029297,
|
| 1463 |
+
"rewards/rejected": 1.7357536554336548,
|
| 1464 |
+
"step": 97
|
| 1465 |
+
},
|
| 1466 |
+
{
|
| 1467 |
+
"epoch": 0.0033340704577542652,
|
| 1468 |
+
"grad_norm": 34.95238494873047,
|
| 1469 |
+
"learning_rate": 4.860940925593703e-05,
|
| 1470 |
+
"logits/chosen": -3.038242816925049,
|
| 1471 |
+
"logits/rejected": -3.026169776916504,
|
| 1472 |
+
"logps/chosen": -320.06005859375,
|
| 1473 |
+
"logps/rejected": -534.7603759765625,
|
| 1474 |
+
"loss": 0.6213,
|
| 1475 |
+
"rewards/accuracies": 0.75,
|
| 1476 |
+
"rewards/chosen": 3.1372852325439453,
|
| 1477 |
+
"rewards/margins": 0.3597479462623596,
|
| 1478 |
+
"rewards/rejected": 2.7775373458862305,
|
| 1479 |
+
"step": 98
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 0.003368091584874207,
|
| 1483 |
+
"grad_norm": 22.04984474182129,
|
| 1484 |
+
"learning_rate": 4.855143631968242e-05,
|
| 1485 |
+
"logits/chosen": -3.0394222736358643,
|
| 1486 |
+
"logits/rejected": -3.040712356567383,
|
| 1487 |
+
"logps/chosen": -405.9694519042969,
|
| 1488 |
+
"logps/rejected": -224.75143432617188,
|
| 1489 |
+
"loss": 0.472,
|
| 1490 |
+
"rewards/accuracies": 0.75,
|
| 1491 |
+
"rewards/chosen": 3.6535189151763916,
|
| 1492 |
+
"rewards/margins": 1.3992540836334229,
|
| 1493 |
+
"rewards/rejected": 2.2542648315429688,
|
| 1494 |
+
"step": 99
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 0.003402112711994148,
|
| 1498 |
+
"grad_norm": 10.456045150756836,
|
| 1499 |
+
"learning_rate": 4.849231551964771e-05,
|
| 1500 |
+
"logits/chosen": -3.0558083057403564,
|
| 1501 |
+
"logits/rejected": -3.0164895057678223,
|
| 1502 |
+
"logps/chosen": -215.77474975585938,
|
| 1503 |
+
"logps/rejected": -114.40177917480469,
|
| 1504 |
+
"loss": 0.3431,
|
| 1505 |
+
"rewards/accuracies": 0.75,
|
| 1506 |
+
"rewards/chosen": 3.2260184288024902,
|
| 1507 |
+
"rewards/margins": 1.2816791534423828,
|
| 1508 |
+
"rewards/rejected": 1.9443392753601074,
|
| 1509 |
+
"step": 100
|
| 1510 |
+
},
|
| 1511 |
+
{
|
| 1512 |
+
"epoch": 0.00343613383911409,
|
| 1513 |
+
"grad_norm": 26.428396224975586,
|
| 1514 |
+
"learning_rate": 4.843204973729729e-05,
|
| 1515 |
+
"logits/chosen": -2.9279229640960693,
|
| 1516 |
+
"logits/rejected": -2.941478729248047,
|
| 1517 |
+
"logps/chosen": -344.0908203125,
|
| 1518 |
+
"logps/rejected": -369.3677978515625,
|
| 1519 |
+
"loss": 0.851,
|
| 1520 |
+
"rewards/accuracies": 0.5,
|
| 1521 |
+
"rewards/chosen": 4.361503601074219,
|
| 1522 |
+
"rewards/margins": 2.4121603965759277,
|
| 1523 |
+
"rewards/rejected": 1.949343204498291,
|
| 1524 |
+
"step": 101
|
| 1525 |
+
},
|
| 1526 |
+
{
|
| 1527 |
+
"epoch": 0.003470154966234031,
|
| 1528 |
+
"grad_norm": 20.163902282714844,
|
| 1529 |
+
"learning_rate": 4.837064190990036e-05,
|
| 1530 |
+
"logits/chosen": -2.994917869567871,
|
| 1531 |
+
"logits/rejected": -2.9446635246276855,
|
| 1532 |
+
"logps/chosen": -348.66937255859375,
|
| 1533 |
+
"logps/rejected": -345.49578857421875,
|
| 1534 |
+
"loss": 0.4779,
|
| 1535 |
+
"rewards/accuracies": 0.75,
|
| 1536 |
+
"rewards/chosen": 3.1643588542938232,
|
| 1537 |
+
"rewards/margins": 1.304919958114624,
|
| 1538 |
+
"rewards/rejected": 1.8594388961791992,
|
| 1539 |
+
"step": 102
|
| 1540 |
+
},
|
| 1541 |
+
{
|
| 1542 |
+
"epoch": 0.003504176093353973,
|
| 1543 |
+
"grad_norm": 10.98054027557373,
|
| 1544 |
+
"learning_rate": 4.830809503038781e-05,
|
| 1545 |
+
"logits/chosen": -2.9521918296813965,
|
| 1546 |
+
"logits/rejected": -2.9645028114318848,
|
| 1547 |
+
"logps/chosen": -226.67506408691406,
|
| 1548 |
+
"logps/rejected": -326.85455322265625,
|
| 1549 |
+
"loss": 0.1768,
|
| 1550 |
+
"rewards/accuracies": 1.0,
|
| 1551 |
+
"rewards/chosen": 3.5958943367004395,
|
| 1552 |
+
"rewards/margins": 1.691298007965088,
|
| 1553 |
+
"rewards/rejected": 1.904596209526062,
|
| 1554 |
+
"step": 103
|
| 1555 |
+
},
|
| 1556 |
+
{
|
| 1557 |
+
"epoch": 0.003538197220473914,
|
| 1558 |
+
"grad_norm": 10.843141555786133,
|
| 1559 |
+
"learning_rate": 4.8244412147206284e-05,
|
| 1560 |
+
"logits/chosen": -2.9416680335998535,
|
| 1561 |
+
"logits/rejected": -2.935615062713623,
|
| 1562 |
+
"logps/chosen": -242.29098510742188,
|
| 1563 |
+
"logps/rejected": -196.54017639160156,
|
| 1564 |
+
"loss": 0.4134,
|
| 1565 |
+
"rewards/accuracies": 0.75,
|
| 1566 |
+
"rewards/chosen": 2.624265670776367,
|
| 1567 |
+
"rewards/margins": 1.845750331878662,
|
| 1568 |
+
"rewards/rejected": 0.778515100479126,
|
| 1569 |
+
"step": 104
|
| 1570 |
+
},
|
| 1571 |
+
{
|
| 1572 |
+
"epoch": 0.003572218347593856,
|
| 1573 |
+
"grad_norm": 51.45275115966797,
|
| 1574 |
+
"learning_rate": 4.817959636416969e-05,
|
| 1575 |
+
"logits/chosen": -2.987889289855957,
|
| 1576 |
+
"logits/rejected": -2.9992475509643555,
|
| 1577 |
+
"logps/chosen": -265.65509033203125,
|
| 1578 |
+
"logps/rejected": -504.4512023925781,
|
| 1579 |
+
"loss": 2.0434,
|
| 1580 |
+
"rewards/accuracies": 0.25,
|
| 1581 |
+
"rewards/chosen": 1.9749891757965088,
|
| 1582 |
+
"rewards/margins": -1.6412996053695679,
|
| 1583 |
+
"rewards/rejected": 3.6162891387939453,
|
| 1584 |
+
"step": 105
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 0.003606239474713797,
|
| 1588 |
+
"grad_norm": 34.740169525146484,
|
| 1589 |
+
"learning_rate": 4.8113650840307834e-05,
|
| 1590 |
+
"logits/chosen": -2.9495849609375,
|
| 1591 |
+
"logits/rejected": -2.9650793075561523,
|
| 1592 |
+
"logps/chosen": -267.8492126464844,
|
| 1593 |
+
"logps/rejected": -229.05166625976562,
|
| 1594 |
+
"loss": 1.4427,
|
| 1595 |
+
"rewards/accuracies": 0.25,
|
| 1596 |
+
"rewards/chosen": 1.6815515756607056,
|
| 1597 |
+
"rewards/margins": -1.025644063949585,
|
| 1598 |
+
"rewards/rejected": 2.70719575881958,
|
| 1599 |
+
"step": 106
|
| 1600 |
+
},
|
| 1601 |
+
{
|
| 1602 |
+
"epoch": 0.003640260601833739,
|
| 1603 |
+
"grad_norm": 33.90592956542969,
|
| 1604 |
+
"learning_rate": 4.8046578789712515e-05,
|
| 1605 |
+
"logits/chosen": -2.9517102241516113,
|
| 1606 |
+
"logits/rejected": -2.892892360687256,
|
| 1607 |
+
"logps/chosen": -271.4212646484375,
|
| 1608 |
+
"logps/rejected": -432.76318359375,
|
| 1609 |
+
"loss": 0.9206,
|
| 1610 |
+
"rewards/accuracies": 0.5,
|
| 1611 |
+
"rewards/chosen": 2.004805564880371,
|
| 1612 |
+
"rewards/margins": -0.28335341811180115,
|
| 1613 |
+
"rewards/rejected": 2.288158893585205,
|
| 1614 |
+
"step": 107
|
| 1615 |
+
},
|
| 1616 |
+
{
|
| 1617 |
+
"epoch": 0.00367428172895368,
|
| 1618 |
+
"grad_norm": 26.77821922302246,
|
| 1619 |
+
"learning_rate": 4.797838348138086e-05,
|
| 1620 |
+
"logits/chosen": -2.961052656173706,
|
| 1621 |
+
"logits/rejected": -2.963571548461914,
|
| 1622 |
+
"logps/chosen": -235.24127197265625,
|
| 1623 |
+
"logps/rejected": -195.713623046875,
|
| 1624 |
+
"loss": 0.9046,
|
| 1625 |
+
"rewards/accuracies": 0.25,
|
| 1626 |
+
"rewards/chosen": 0.960392951965332,
|
| 1627 |
+
"rewards/margins": -0.3249235153198242,
|
| 1628 |
+
"rewards/rejected": 1.2853164672851562,
|
| 1629 |
+
"step": 108
|
| 1630 |
+
},
|
| 1631 |
+
{
|
| 1632 |
+
"epoch": 0.003708302856073622,
|
| 1633 |
+
"grad_norm": 20.291223526000977,
|
| 1634 |
+
"learning_rate": 4.790906823905599e-05,
|
| 1635 |
+
"logits/chosen": -2.922367811203003,
|
| 1636 |
+
"logits/rejected": -2.9089303016662598,
|
| 1637 |
+
"logps/chosen": -432.6478576660156,
|
| 1638 |
+
"logps/rejected": -147.66995239257812,
|
| 1639 |
+
"loss": 0.5442,
|
| 1640 |
+
"rewards/accuracies": 0.75,
|
| 1641 |
+
"rewards/chosen": 2.312082052230835,
|
| 1642 |
+
"rewards/margins": 0.7899919748306274,
|
| 1643 |
+
"rewards/rejected": 1.522089958190918,
|
| 1644 |
+
"step": 109
|
| 1645 |
+
},
|
| 1646 |
+
{
|
| 1647 |
+
"epoch": 0.003742323983193563,
|
| 1648 |
+
"grad_norm": 21.555723190307617,
|
| 1649 |
+
"learning_rate": 4.783863644106502e-05,
|
| 1650 |
+
"logits/chosen": -2.893828868865967,
|
| 1651 |
+
"logits/rejected": -2.9130947589874268,
|
| 1652 |
+
"logps/chosen": -188.0413360595703,
|
| 1653 |
+
"logps/rejected": -163.48797607421875,
|
| 1654 |
+
"loss": 0.7379,
|
| 1655 |
+
"rewards/accuracies": 0.25,
|
| 1656 |
+
"rewards/chosen": 1.8718959093093872,
|
| 1657 |
+
"rewards/margins": 0.09508019685745239,
|
| 1658 |
+
"rewards/rejected": 1.77681565284729,
|
| 1659 |
+
"step": 110
|
| 1660 |
+
}
|
| 1661 |
+
],
|
| 1662 |
+
"logging_steps": 1,
|
| 1663 |
+
"max_steps": 500,
|
| 1664 |
+
"num_input_tokens_seen": 0,
|
| 1665 |
+
"num_train_epochs": 1,
|
| 1666 |
+
"save_steps": 10,
|
| 1667 |
+
"stateful_callbacks": {
|
| 1668 |
+
"TrainerControl": {
|
| 1669 |
+
"args": {
|
| 1670 |
+
"should_epoch_stop": false,
|
| 1671 |
+
"should_evaluate": false,
|
| 1672 |
+
"should_log": false,
|
| 1673 |
+
"should_save": true,
|
| 1674 |
+
"should_training_stop": false
|
| 1675 |
+
},
|
| 1676 |
+
"attributes": {}
|
| 1677 |
+
}
|
| 1678 |
+
},
|
| 1679 |
+
"total_flos": 0.0,
|
| 1680 |
+
"train_batch_size": 1,
|
| 1681 |
+
"trial_name": null,
|
| 1682 |
+
"trial_params": null
|
| 1683 |
+
}
|
checkpoint-110/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2aa185d46c9d5abcf13df6d50376870fd603b11ffa1fb959ec10e0609dfec047
|
| 3 |
+
size 5880
|
checkpoint-120/README.md
ADDED
|
@@ -0,0 +1,202 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: peft
|
| 3 |
+
base_model: teknium/OpenHermes-2.5-Mistral-7B
|
| 4 |
+
---
|
| 5 |
+
|
| 6 |
+
# Model Card for Model ID
|
| 7 |
+
|
| 8 |
+
<!-- Provide a quick summary of what the model is/does. -->
|
| 9 |
+
|
| 10 |
+
|
| 11 |
+
|
| 12 |
+
## Model Details
|
| 13 |
+
|
| 14 |
+
### Model Description
|
| 15 |
+
|
| 16 |
+
<!-- Provide a longer summary of what this model is. -->
|
| 17 |
+
|
| 18 |
+
|
| 19 |
+
|
| 20 |
+
- **Developed by:** [More Information Needed]
|
| 21 |
+
- **Funded by [optional]:** [More Information Needed]
|
| 22 |
+
- **Shared by [optional]:** [More Information Needed]
|
| 23 |
+
- **Model type:** [More Information Needed]
|
| 24 |
+
- **Language(s) (NLP):** [More Information Needed]
|
| 25 |
+
- **License:** [More Information Needed]
|
| 26 |
+
- **Finetuned from model [optional]:** [More Information Needed]
|
| 27 |
+
|
| 28 |
+
### Model Sources [optional]
|
| 29 |
+
|
| 30 |
+
<!-- Provide the basic links for the model. -->
|
| 31 |
+
|
| 32 |
+
- **Repository:** [More Information Needed]
|
| 33 |
+
- **Paper [optional]:** [More Information Needed]
|
| 34 |
+
- **Demo [optional]:** [More Information Needed]
|
| 35 |
+
|
| 36 |
+
## Uses
|
| 37 |
+
|
| 38 |
+
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
|
| 39 |
+
|
| 40 |
+
### Direct Use
|
| 41 |
+
|
| 42 |
+
<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
|
| 43 |
+
|
| 44 |
+
[More Information Needed]
|
| 45 |
+
|
| 46 |
+
### Downstream Use [optional]
|
| 47 |
+
|
| 48 |
+
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
|
| 49 |
+
|
| 50 |
+
[More Information Needed]
|
| 51 |
+
|
| 52 |
+
### Out-of-Scope Use
|
| 53 |
+
|
| 54 |
+
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
|
| 55 |
+
|
| 56 |
+
[More Information Needed]
|
| 57 |
+
|
| 58 |
+
## Bias, Risks, and Limitations
|
| 59 |
+
|
| 60 |
+
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
|
| 61 |
+
|
| 62 |
+
[More Information Needed]
|
| 63 |
+
|
| 64 |
+
### Recommendations
|
| 65 |
+
|
| 66 |
+
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
|
| 67 |
+
|
| 68 |
+
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
|
| 69 |
+
|
| 70 |
+
## How to Get Started with the Model
|
| 71 |
+
|
| 72 |
+
Use the code below to get started with the model.
|
| 73 |
+
|
| 74 |
+
[More Information Needed]
|
| 75 |
+
|
| 76 |
+
## Training Details
|
| 77 |
+
|
| 78 |
+
### Training Data
|
| 79 |
+
|
| 80 |
+
<!-- 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. -->
|
| 81 |
+
|
| 82 |
+
[More Information Needed]
|
| 83 |
+
|
| 84 |
+
### Training Procedure
|
| 85 |
+
|
| 86 |
+
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
|
| 87 |
+
|
| 88 |
+
#### Preprocessing [optional]
|
| 89 |
+
|
| 90 |
+
[More Information Needed]
|
| 91 |
+
|
| 92 |
+
|
| 93 |
+
#### Training Hyperparameters
|
| 94 |
+
|
| 95 |
+
- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
|
| 96 |
+
|
| 97 |
+
#### Speeds, Sizes, Times [optional]
|
| 98 |
+
|
| 99 |
+
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
|
| 100 |
+
|
| 101 |
+
[More Information Needed]
|
| 102 |
+
|
| 103 |
+
## Evaluation
|
| 104 |
+
|
| 105 |
+
<!-- This section describes the evaluation protocols and provides the results. -->
|
| 106 |
+
|
| 107 |
+
### Testing Data, Factors & Metrics
|
| 108 |
+
|
| 109 |
+
#### Testing Data
|
| 110 |
+
|
| 111 |
+
<!-- This should link to a Dataset Card if possible. -->
|
| 112 |
+
|
| 113 |
+
[More Information Needed]
|
| 114 |
+
|
| 115 |
+
#### Factors
|
| 116 |
+
|
| 117 |
+
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
|
| 118 |
+
|
| 119 |
+
[More Information Needed]
|
| 120 |
+
|
| 121 |
+
#### Metrics
|
| 122 |
+
|
| 123 |
+
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
|
| 124 |
+
|
| 125 |
+
[More Information Needed]
|
| 126 |
+
|
| 127 |
+
### Results
|
| 128 |
+
|
| 129 |
+
[More Information Needed]
|
| 130 |
+
|
| 131 |
+
#### Summary
|
| 132 |
+
|
| 133 |
+
|
| 134 |
+
|
| 135 |
+
## Model Examination [optional]
|
| 136 |
+
|
| 137 |
+
<!-- Relevant interpretability work for the model goes here -->
|
| 138 |
+
|
| 139 |
+
[More Information Needed]
|
| 140 |
+
|
| 141 |
+
## Environmental Impact
|
| 142 |
+
|
| 143 |
+
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
|
| 144 |
+
|
| 145 |
+
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).
|
| 146 |
+
|
| 147 |
+
- **Hardware Type:** [More Information Needed]
|
| 148 |
+
- **Hours used:** [More Information Needed]
|
| 149 |
+
- **Cloud Provider:** [More Information Needed]
|
| 150 |
+
- **Compute Region:** [More Information Needed]
|
| 151 |
+
- **Carbon Emitted:** [More Information Needed]
|
| 152 |
+
|
| 153 |
+
## Technical Specifications [optional]
|
| 154 |
+
|
| 155 |
+
### Model Architecture and Objective
|
| 156 |
+
|
| 157 |
+
[More Information Needed]
|
| 158 |
+
|
| 159 |
+
### Compute Infrastructure
|
| 160 |
+
|
| 161 |
+
[More Information Needed]
|
| 162 |
+
|
| 163 |
+
#### Hardware
|
| 164 |
+
|
| 165 |
+
[More Information Needed]
|
| 166 |
+
|
| 167 |
+
#### Software
|
| 168 |
+
|
| 169 |
+
[More Information Needed]
|
| 170 |
+
|
| 171 |
+
## Citation [optional]
|
| 172 |
+
|
| 173 |
+
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
|
| 174 |
+
|
| 175 |
+
**BibTeX:**
|
| 176 |
+
|
| 177 |
+
[More Information Needed]
|
| 178 |
+
|
| 179 |
+
**APA:**
|
| 180 |
+
|
| 181 |
+
[More Information Needed]
|
| 182 |
+
|
| 183 |
+
## Glossary [optional]
|
| 184 |
+
|
| 185 |
+
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
|
| 186 |
+
|
| 187 |
+
[More Information Needed]
|
| 188 |
+
|
| 189 |
+
## More Information [optional]
|
| 190 |
+
|
| 191 |
+
[More Information Needed]
|
| 192 |
+
|
| 193 |
+
## Model Card Authors [optional]
|
| 194 |
+
|
| 195 |
+
[More Information Needed]
|
| 196 |
+
|
| 197 |
+
## Model Card Contact
|
| 198 |
+
|
| 199 |
+
[More Information Needed]
|
| 200 |
+
### Framework versions
|
| 201 |
+
|
| 202 |
+
- PEFT 0.11.1
|
checkpoint-120/adapter_config.json
ADDED
|
@@ -0,0 +1,34 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"alpha_pattern": {},
|
| 3 |
+
"auto_mapping": null,
|
| 4 |
+
"base_model_name_or_path": "teknium/OpenHermes-2.5-Mistral-7B",
|
| 5 |
+
"bias": "none",
|
| 6 |
+
"fan_in_fan_out": false,
|
| 7 |
+
"inference_mode": true,
|
| 8 |
+
"init_lora_weights": true,
|
| 9 |
+
"layer_replication": null,
|
| 10 |
+
"layers_pattern": null,
|
| 11 |
+
"layers_to_transform": null,
|
| 12 |
+
"loftq_config": {},
|
| 13 |
+
"lora_alpha": 16,
|
| 14 |
+
"lora_dropout": 0.05,
|
| 15 |
+
"megatron_config": null,
|
| 16 |
+
"megatron_core": "megatron.core",
|
| 17 |
+
"modules_to_save": null,
|
| 18 |
+
"peft_type": "LORA",
|
| 19 |
+
"r": 16,
|
| 20 |
+
"rank_pattern": {},
|
| 21 |
+
"revision": null,
|
| 22 |
+
"target_modules": [
|
| 23 |
+
"o_proj",
|
| 24 |
+
"k_proj",
|
| 25 |
+
"gate_proj",
|
| 26 |
+
"down_proj",
|
| 27 |
+
"up_proj",
|
| 28 |
+
"q_proj",
|
| 29 |
+
"v_proj"
|
| 30 |
+
],
|
| 31 |
+
"task_type": "CAUSAL_LM",
|
| 32 |
+
"use_dora": false,
|
| 33 |
+
"use_rslora": false
|
| 34 |
+
}
|
checkpoint-120/adapter_model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:633e1f5f973013180d6a30ab16fda9087f79d5034e03d978c66fe52326a17c0d
|
| 3 |
+
size 167832240
|
checkpoint-120/added_tokens.json
ADDED
|
@@ -0,0 +1,4 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"<|im_end|>": 32000,
|
| 3 |
+
"<|im_start|>": 32001
|
| 4 |
+
}
|
checkpoint-120/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:80bb04b3fb3f16dd77493aecf9bfd50f6544c6abff037dd7756ff4971eb56325
|
| 3 |
+
size 335928658
|
checkpoint-120/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:386fcc8cc1089aade9450d86fb239ea3483f455fd2d78d8378645feecfec9d69
|
| 3 |
+
size 14244
|
checkpoint-120/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:755ecfc0f97ef143497432f27fc797e37ec91da9377c2759e28dd231d773e09b
|
| 3 |
+
size 1064
|
checkpoint-120/special_tokens_map.json
ADDED
|
@@ -0,0 +1,24 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": {
|
| 3 |
+
"content": "<s>",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": false,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"eos_token": {
|
| 10 |
+
"content": "<|im_end|>",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": false,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"pad_token": "<|im_end|>",
|
| 17 |
+
"unk_token": {
|
| 18 |
+
"content": "<unk>",
|
| 19 |
+
"lstrip": false,
|
| 20 |
+
"normalized": false,
|
| 21 |
+
"rstrip": false,
|
| 22 |
+
"single_word": false
|
| 23 |
+
}
|
| 24 |
+
}
|
checkpoint-120/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-120/tokenizer.model
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:dadfd56d766715c61d2ef780a525ab43b8e6da4de6865bda3d95fdef5e134055
|
| 3 |
+
size 493443
|
checkpoint-120/tokenizer_config.json
ADDED
|
@@ -0,0 +1,62 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_bos_token": true,
|
| 3 |
+
"add_eos_token": false,
|
| 4 |
+
"add_prefix_space": true,
|
| 5 |
+
"added_tokens_decoder": {
|
| 6 |
+
"0": {
|
| 7 |
+
"content": "<unk>",
|
| 8 |
+
"lstrip": false,
|
| 9 |
+
"normalized": false,
|
| 10 |
+
"rstrip": false,
|
| 11 |
+
"single_word": false,
|
| 12 |
+
"special": true
|
| 13 |
+
},
|
| 14 |
+
"1": {
|
| 15 |
+
"content": "<s>",
|
| 16 |
+
"lstrip": false,
|
| 17 |
+
"normalized": false,
|
| 18 |
+
"rstrip": false,
|
| 19 |
+
"single_word": false,
|
| 20 |
+
"special": true
|
| 21 |
+
},
|
| 22 |
+
"2": {
|
| 23 |
+
"content": "</s>",
|
| 24 |
+
"lstrip": false,
|
| 25 |
+
"normalized": false,
|
| 26 |
+
"rstrip": false,
|
| 27 |
+
"single_word": false,
|
| 28 |
+
"special": true
|
| 29 |
+
},
|
| 30 |
+
"32000": {
|
| 31 |
+
"content": "<|im_end|>",
|
| 32 |
+
"lstrip": false,
|
| 33 |
+
"normalized": false,
|
| 34 |
+
"rstrip": false,
|
| 35 |
+
"single_word": false,
|
| 36 |
+
"special": true
|
| 37 |
+
},
|
| 38 |
+
"32001": {
|
| 39 |
+
"content": "<|im_start|>",
|
| 40 |
+
"lstrip": false,
|
| 41 |
+
"normalized": false,
|
| 42 |
+
"rstrip": false,
|
| 43 |
+
"single_word": false,
|
| 44 |
+
"special": true
|
| 45 |
+
}
|
| 46 |
+
},
|
| 47 |
+
"additional_special_tokens": [],
|
| 48 |
+
"bos_token": "<s>",
|
| 49 |
+
"chat_template": "{% for message in messages %}{{'<|im_start|>' + message['role'] + '\n' + message['content'] + '<|im_end|>' + '\n'}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant\n' }}{% endif %}",
|
| 50 |
+
"clean_up_tokenization_spaces": false,
|
| 51 |
+
"eos_token": "<|im_end|>",
|
| 52 |
+
"legacy": true,
|
| 53 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 54 |
+
"pad_token": "<|im_end|>",
|
| 55 |
+
"sp_model_kwargs": {},
|
| 56 |
+
"spaces_between_special_tokens": false,
|
| 57 |
+
"tokenizer_class": "LlamaTokenizer",
|
| 58 |
+
"trust_remote_code": false,
|
| 59 |
+
"unk_token": "<unk>",
|
| 60 |
+
"use_default_system_prompt": true,
|
| 61 |
+
"use_fast": true
|
| 62 |
+
}
|