Upload 33 files
Browse files- checkpoint-100/README.md +204 -0
- checkpoint-100/adapter_config.json +31 -0
- checkpoint-100/adapter_model.safetensors +3 -0
- checkpoint-100/rng_state.pth +3 -0
- checkpoint-100/scheduler.pt +3 -0
- checkpoint-100/special_tokens_map.json +30 -0
- checkpoint-100/tokenizer.json +0 -0
- checkpoint-100/tokenizer.model +3 -0
- checkpoint-100/tokenizer_config.json +42 -0
- checkpoint-100/trainer_state.json +1421 -0
- checkpoint-100/training_args.bin +3 -0
- checkpoint-200/README.md +204 -0
- checkpoint-200/adapter_config.json +31 -0
- checkpoint-200/adapter_model.safetensors +3 -0
- checkpoint-200/rng_state.pth +3 -0
- checkpoint-200/scheduler.pt +3 -0
- checkpoint-200/special_tokens_map.json +30 -0
- checkpoint-200/tokenizer.json +0 -0
- checkpoint-200/tokenizer.model +3 -0
- checkpoint-200/tokenizer_config.json +42 -0
- checkpoint-200/trainer_state.json +2821 -0
- checkpoint-200/training_args.bin +3 -0
- checkpoint-300/README.md +204 -0
- checkpoint-300/adapter_config.json +31 -0
- checkpoint-300/adapter_model.safetensors +3 -0
- checkpoint-300/rng_state.pth +3 -0
- checkpoint-300/scheduler.pt +3 -0
- checkpoint-300/special_tokens_map.json +30 -0
- checkpoint-300/tokenizer.json +0 -0
- checkpoint-300/tokenizer.model +3 -0
- checkpoint-300/tokenizer_config.json +42 -0
- checkpoint-300/trainer_state.json +0 -0
- checkpoint-300/training_args.bin +3 -0
checkpoint-100/README.md
ADDED
|
@@ -0,0 +1,204 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: peft
|
| 3 |
+
base_model: /run/media/adamo1139/82142F79142F6EFB/ProgramData/Anaconda3/envs/qlora-jondurbin/axolotl-git-linux/axolotl/yi-34b-200k-llamafied
|
| 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 |
+
|
| 201 |
+
|
| 202 |
+
### Framework versions
|
| 203 |
+
|
| 204 |
+
- PEFT 0.7.1
|
checkpoint-100/adapter_config.json
ADDED
|
@@ -0,0 +1,31 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"alpha_pattern": {},
|
| 3 |
+
"auto_mapping": null,
|
| 4 |
+
"base_model_name_or_path": "yi-34b-200k-llamafied",
|
| 5 |
+
"bias": "none",
|
| 6 |
+
"fan_in_fan_out": false,
|
| 7 |
+
"inference_mode": true,
|
| 8 |
+
"init_lora_weights": true,
|
| 9 |
+
"layers_pattern": null,
|
| 10 |
+
"layers_to_transform": null,
|
| 11 |
+
"loftq_config": null,
|
| 12 |
+
"lora_alpha": 32,
|
| 13 |
+
"lora_dropout": 0,
|
| 14 |
+
"megatron_config": null,
|
| 15 |
+
"megatron_core": "megatron.core",
|
| 16 |
+
"modules_to_save": null,
|
| 17 |
+
"peft_type": "LORA",
|
| 18 |
+
"r": 16,
|
| 19 |
+
"rank_pattern": {},
|
| 20 |
+
"revision": "unsloth",
|
| 21 |
+
"target_modules": [
|
| 22 |
+
"v_proj",
|
| 23 |
+
"gate_proj",
|
| 24 |
+
"o_proj",
|
| 25 |
+
"down_proj",
|
| 26 |
+
"q_proj",
|
| 27 |
+
"up_proj",
|
| 28 |
+
"k_proj"
|
| 29 |
+
],
|
| 30 |
+
"task_type": "CAUSAL_LM"
|
| 31 |
+
}
|
checkpoint-100/adapter_model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:072ecc026ba448391bcebe410322e08314ed4bec47eebcec1cf79af5efbb1739
|
| 3 |
+
size 491633464
|
checkpoint-100/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1ff264f99d31b522cc7e2a4eac9d38606d0c58a34c0adc74d71e0ca8b371dc36
|
| 3 |
+
size 14244
|
checkpoint-100/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:26e7a331a766a7b1c3ce8e8d12fb2325171adaf35890d903a56a1ac6d3f0733f
|
| 3 |
+
size 1064
|
checkpoint-100/special_tokens_map.json
ADDED
|
@@ -0,0 +1,30 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": {
|
| 3 |
+
"content": "<|startoftext|>",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": false,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"eos_token": {
|
| 10 |
+
"content": "<|endoftext|>",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": false,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"pad_token": {
|
| 17 |
+
"content": "<unk>",
|
| 18 |
+
"lstrip": false,
|
| 19 |
+
"normalized": false,
|
| 20 |
+
"rstrip": false,
|
| 21 |
+
"single_word": false
|
| 22 |
+
},
|
| 23 |
+
"unk_token": {
|
| 24 |
+
"content": "<unk>",
|
| 25 |
+
"lstrip": false,
|
| 26 |
+
"normalized": false,
|
| 27 |
+
"rstrip": false,
|
| 28 |
+
"single_word": false
|
| 29 |
+
}
|
| 30 |
+
}
|
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 sha256:386c49cf943d71aa110361135338c50e38beeff0a66593480421f37b319e1a39
|
| 3 |
+
size 1033105
|
checkpoint-100/tokenizer_config.json
ADDED
|
@@ -0,0 +1,42 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_bos_token": false,
|
| 3 |
+
"add_eos_token": false,
|
| 4 |
+
"added_tokens_decoder": {
|
| 5 |
+
"0": {
|
| 6 |
+
"content": "<unk>",
|
| 7 |
+
"lstrip": false,
|
| 8 |
+
"normalized": false,
|
| 9 |
+
"rstrip": false,
|
| 10 |
+
"single_word": false,
|
| 11 |
+
"special": true
|
| 12 |
+
},
|
| 13 |
+
"1": {
|
| 14 |
+
"content": "<|startoftext|>",
|
| 15 |
+
"lstrip": false,
|
| 16 |
+
"normalized": false,
|
| 17 |
+
"rstrip": false,
|
| 18 |
+
"single_word": false,
|
| 19 |
+
"special": true
|
| 20 |
+
},
|
| 21 |
+
"2": {
|
| 22 |
+
"content": "<|endoftext|>",
|
| 23 |
+
"lstrip": false,
|
| 24 |
+
"normalized": false,
|
| 25 |
+
"rstrip": false,
|
| 26 |
+
"single_word": false,
|
| 27 |
+
"special": true
|
| 28 |
+
}
|
| 29 |
+
},
|
| 30 |
+
"bos_token": "<|startoftext|>",
|
| 31 |
+
"chat_template": "{% if not add_generation_prompt is defined %}{% set add_generation_prompt = false %}{% endif %}{% for message in messages %}{{'<|im_start|>' + message['role'] + '\n' + message['content'] + '<|im_end|>' + '\n'}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant\n' }}{% endif %}",
|
| 32 |
+
"clean_up_tokenization_spaces": false,
|
| 33 |
+
"eos_token": "<|endoftext|>",
|
| 34 |
+
"legacy": true,
|
| 35 |
+
"model_max_length": 4096,
|
| 36 |
+
"pad_token": "<unk>",
|
| 37 |
+
"padding_side": "right",
|
| 38 |
+
"sp_model_kwargs": {},
|
| 39 |
+
"tokenizer_class": "LlamaTokenizer",
|
| 40 |
+
"unk_token": "<unk>",
|
| 41 |
+
"use_default_system_prompt": false
|
| 42 |
+
}
|
checkpoint-100/trainer_state.json
ADDED
|
@@ -0,0 +1,1421 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.10908842980841345,
|
| 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": 0.0,
|
| 13 |
+
"learning_rate": 9.782608695652175e-07,
|
| 14 |
+
"logits/chosen": -1.8868483304977417,
|
| 15 |
+
"logits/rejected": -2.3036646842956543,
|
| 16 |
+
"logps/chosen": -466.7117004394531,
|
| 17 |
+
"logps/rejected": -99.3152084350586,
|
| 18 |
+
"loss": 0.6931,
|
| 19 |
+
"rewards/accuracies": 0.0,
|
| 20 |
+
"rewards/chosen": 0.0,
|
| 21 |
+
"rewards/margins": 0.0,
|
| 22 |
+
"rewards/rejected": 0.0,
|
| 23 |
+
"step": 1
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.0,
|
| 27 |
+
"learning_rate": 1.956521739130435e-06,
|
| 28 |
+
"logits/chosen": -2.2296698093414307,
|
| 29 |
+
"logits/rejected": -2.469517469406128,
|
| 30 |
+
"logps/chosen": -335.50140380859375,
|
| 31 |
+
"logps/rejected": -97.97496032714844,
|
| 32 |
+
"loss": 0.6931,
|
| 33 |
+
"rewards/accuracies": 0.0,
|
| 34 |
+
"rewards/chosen": 0.0,
|
| 35 |
+
"rewards/margins": 0.0,
|
| 36 |
+
"rewards/rejected": 0.0,
|
| 37 |
+
"step": 2
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.0,
|
| 41 |
+
"learning_rate": 2.9347826086956523e-06,
|
| 42 |
+
"logits/chosen": -2.125047206878662,
|
| 43 |
+
"logits/rejected": -2.445204019546509,
|
| 44 |
+
"logps/chosen": -484.7939758300781,
|
| 45 |
+
"logps/rejected": -125.93072509765625,
|
| 46 |
+
"loss": 1.7268,
|
| 47 |
+
"rewards/accuracies": 0.4375,
|
| 48 |
+
"rewards/chosen": -1.0339633226394653,
|
| 49 |
+
"rewards/margins": -0.9356753826141357,
|
| 50 |
+
"rewards/rejected": -0.09828801453113556,
|
| 51 |
+
"step": 3
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.0,
|
| 55 |
+
"learning_rate": 3.91304347826087e-06,
|
| 56 |
+
"logits/chosen": -1.7763867378234863,
|
| 57 |
+
"logits/rejected": -2.1961894035339355,
|
| 58 |
+
"logps/chosen": -386.8011779785156,
|
| 59 |
+
"logps/rejected": -106.04309844970703,
|
| 60 |
+
"loss": 2.3806,
|
| 61 |
+
"rewards/accuracies": 0.5625,
|
| 62 |
+
"rewards/chosen": -0.9238373041152954,
|
| 63 |
+
"rewards/margins": -1.2902063131332397,
|
| 64 |
+
"rewards/rejected": 0.36636874079704285,
|
| 65 |
+
"step": 4
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.01,
|
| 69 |
+
"learning_rate": 4.891304347826087e-06,
|
| 70 |
+
"logits/chosen": -1.7257812023162842,
|
| 71 |
+
"logits/rejected": -2.353095769882202,
|
| 72 |
+
"logps/chosen": -524.7261962890625,
|
| 73 |
+
"logps/rejected": -70.26946258544922,
|
| 74 |
+
"loss": 0.7488,
|
| 75 |
+
"rewards/accuracies": 0.625,
|
| 76 |
+
"rewards/chosen": 2.197690010070801,
|
| 77 |
+
"rewards/margins": 2.0008764266967773,
|
| 78 |
+
"rewards/rejected": 0.19681359827518463,
|
| 79 |
+
"step": 5
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.01,
|
| 83 |
+
"learning_rate": 5.869565217391305e-06,
|
| 84 |
+
"logits/chosen": -2.1483359336853027,
|
| 85 |
+
"logits/rejected": -2.4817819595336914,
|
| 86 |
+
"logps/chosen": -283.6588134765625,
|
| 87 |
+
"logps/rejected": -58.25455093383789,
|
| 88 |
+
"loss": 0.8712,
|
| 89 |
+
"rewards/accuracies": 0.5625,
|
| 90 |
+
"rewards/chosen": 0.4915351867675781,
|
| 91 |
+
"rewards/margins": 0.2005542814731598,
|
| 92 |
+
"rewards/rejected": 0.2909809350967407,
|
| 93 |
+
"step": 6
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.01,
|
| 97 |
+
"learning_rate": 6.847826086956523e-06,
|
| 98 |
+
"logits/chosen": -2.1019060611724854,
|
| 99 |
+
"logits/rejected": -2.300177812576294,
|
| 100 |
+
"logps/chosen": -297.35235595703125,
|
| 101 |
+
"logps/rejected": -175.5060272216797,
|
| 102 |
+
"loss": 1.1392,
|
| 103 |
+
"rewards/accuracies": 0.5625,
|
| 104 |
+
"rewards/chosen": 0.6018204689025879,
|
| 105 |
+
"rewards/margins": -0.036775827407836914,
|
| 106 |
+
"rewards/rejected": 0.6385962963104248,
|
| 107 |
+
"step": 7
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.01,
|
| 111 |
+
"learning_rate": 7.82608695652174e-06,
|
| 112 |
+
"logits/chosen": -2.0169122219085693,
|
| 113 |
+
"logits/rejected": -2.201702117919922,
|
| 114 |
+
"logps/chosen": -342.04852294921875,
|
| 115 |
+
"logps/rejected": -117.9761962890625,
|
| 116 |
+
"loss": 1.1799,
|
| 117 |
+
"rewards/accuracies": 0.5625,
|
| 118 |
+
"rewards/chosen": 1.111680507659912,
|
| 119 |
+
"rewards/margins": 0.17297500371932983,
|
| 120 |
+
"rewards/rejected": 0.9387054443359375,
|
| 121 |
+
"step": 8
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.01,
|
| 125 |
+
"learning_rate": 8.804347826086957e-06,
|
| 126 |
+
"logits/chosen": -1.6254596710205078,
|
| 127 |
+
"logits/rejected": -2.0724740028381348,
|
| 128 |
+
"logps/chosen": -480.1983642578125,
|
| 129 |
+
"logps/rejected": -108.05937194824219,
|
| 130 |
+
"loss": 0.5203,
|
| 131 |
+
"rewards/accuracies": 0.8125,
|
| 132 |
+
"rewards/chosen": 2.161550283432007,
|
| 133 |
+
"rewards/margins": 1.3515225648880005,
|
| 134 |
+
"rewards/rejected": 0.8100277185440063,
|
| 135 |
+
"step": 9
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.01,
|
| 139 |
+
"learning_rate": 9.782608695652175e-06,
|
| 140 |
+
"logits/chosen": -1.4482485055923462,
|
| 141 |
+
"logits/rejected": -2.1116294860839844,
|
| 142 |
+
"logps/chosen": -451.5759582519531,
|
| 143 |
+
"logps/rejected": -67.73944854736328,
|
| 144 |
+
"loss": 0.355,
|
| 145 |
+
"rewards/accuracies": 0.8125,
|
| 146 |
+
"rewards/chosen": 6.18800163269043,
|
| 147 |
+
"rewards/margins": 5.184471130371094,
|
| 148 |
+
"rewards/rejected": 1.0035302639007568,
|
| 149 |
+
"step": 10
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.01,
|
| 153 |
+
"learning_rate": 1.0760869565217392e-05,
|
| 154 |
+
"logits/chosen": -1.2660062313079834,
|
| 155 |
+
"logits/rejected": -1.8649226427078247,
|
| 156 |
+
"logps/chosen": -265.8935546875,
|
| 157 |
+
"logps/rejected": -57.866336822509766,
|
| 158 |
+
"loss": 0.3178,
|
| 159 |
+
"rewards/accuracies": 0.8125,
|
| 160 |
+
"rewards/chosen": 5.78374719619751,
|
| 161 |
+
"rewards/margins": 4.856299877166748,
|
| 162 |
+
"rewards/rejected": 0.9274475574493408,
|
| 163 |
+
"step": 11
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.01,
|
| 167 |
+
"learning_rate": 1.173913043478261e-05,
|
| 168 |
+
"logits/chosen": -1.1804625988006592,
|
| 169 |
+
"logits/rejected": -1.7009960412979126,
|
| 170 |
+
"logps/chosen": -247.2069854736328,
|
| 171 |
+
"logps/rejected": -79.96739196777344,
|
| 172 |
+
"loss": 0.6617,
|
| 173 |
+
"rewards/accuracies": 0.75,
|
| 174 |
+
"rewards/chosen": 6.5535197257995605,
|
| 175 |
+
"rewards/margins": 4.647799015045166,
|
| 176 |
+
"rewards/rejected": 1.9057204723358154,
|
| 177 |
+
"step": 12
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.01,
|
| 181 |
+
"learning_rate": 1.2717391304347827e-05,
|
| 182 |
+
"logits/chosen": -0.9693112373352051,
|
| 183 |
+
"logits/rejected": -1.6497248411178589,
|
| 184 |
+
"logps/chosen": -372.8963623046875,
|
| 185 |
+
"logps/rejected": -70.47633361816406,
|
| 186 |
+
"loss": 0.2498,
|
| 187 |
+
"rewards/accuracies": 0.8125,
|
| 188 |
+
"rewards/chosen": 9.02272891998291,
|
| 189 |
+
"rewards/margins": 7.325143814086914,
|
| 190 |
+
"rewards/rejected": 1.6975862979888916,
|
| 191 |
+
"step": 13
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.02,
|
| 195 |
+
"learning_rate": 1.3695652173913046e-05,
|
| 196 |
+
"logits/chosen": -0.7723280787467957,
|
| 197 |
+
"logits/rejected": -1.4176554679870605,
|
| 198 |
+
"logps/chosen": -318.0392150878906,
|
| 199 |
+
"logps/rejected": -79.94412994384766,
|
| 200 |
+
"loss": 0.6401,
|
| 201 |
+
"rewards/accuracies": 0.6875,
|
| 202 |
+
"rewards/chosen": 6.698611259460449,
|
| 203 |
+
"rewards/margins": 4.517416477203369,
|
| 204 |
+
"rewards/rejected": 2.18119478225708,
|
| 205 |
+
"step": 14
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.02,
|
| 209 |
+
"learning_rate": 1.4673913043478261e-05,
|
| 210 |
+
"logits/chosen": -0.8084051609039307,
|
| 211 |
+
"logits/rejected": -1.4428132772445679,
|
| 212 |
+
"logps/chosen": -363.15869140625,
|
| 213 |
+
"logps/rejected": -70.70079803466797,
|
| 214 |
+
"loss": 0.4963,
|
| 215 |
+
"rewards/accuracies": 0.75,
|
| 216 |
+
"rewards/chosen": 7.02784538269043,
|
| 217 |
+
"rewards/margins": 5.400259971618652,
|
| 218 |
+
"rewards/rejected": 1.6275854110717773,
|
| 219 |
+
"step": 15
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.02,
|
| 223 |
+
"learning_rate": 1.565217391304348e-05,
|
| 224 |
+
"logits/chosen": -0.8056433200836182,
|
| 225 |
+
"logits/rejected": -1.4629294872283936,
|
| 226 |
+
"logps/chosen": -328.90435791015625,
|
| 227 |
+
"logps/rejected": -77.36820983886719,
|
| 228 |
+
"loss": 0.544,
|
| 229 |
+
"rewards/accuracies": 0.75,
|
| 230 |
+
"rewards/chosen": 9.417024612426758,
|
| 231 |
+
"rewards/margins": 7.023505687713623,
|
| 232 |
+
"rewards/rejected": 2.393519163131714,
|
| 233 |
+
"step": 16
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.02,
|
| 237 |
+
"learning_rate": 1.6630434782608694e-05,
|
| 238 |
+
"logits/chosen": -1.0918309688568115,
|
| 239 |
+
"logits/rejected": -1.6613047122955322,
|
| 240 |
+
"logps/chosen": -277.37982177734375,
|
| 241 |
+
"logps/rejected": -56.75635528564453,
|
| 242 |
+
"loss": 0.8958,
|
| 243 |
+
"rewards/accuracies": 0.8125,
|
| 244 |
+
"rewards/chosen": 6.844330310821533,
|
| 245 |
+
"rewards/margins": 4.82295560836792,
|
| 246 |
+
"rewards/rejected": 2.0213749408721924,
|
| 247 |
+
"step": 17
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.02,
|
| 251 |
+
"learning_rate": 1.7608695652173915e-05,
|
| 252 |
+
"logits/chosen": -0.9810999631881714,
|
| 253 |
+
"logits/rejected": -1.975247859954834,
|
| 254 |
+
"logps/chosen": -465.6856689453125,
|
| 255 |
+
"logps/rejected": -56.440269470214844,
|
| 256 |
+
"loss": 0.2275,
|
| 257 |
+
"rewards/accuracies": 0.875,
|
| 258 |
+
"rewards/chosen": 7.767542362213135,
|
| 259 |
+
"rewards/margins": 6.572459697723389,
|
| 260 |
+
"rewards/rejected": 1.1950825452804565,
|
| 261 |
+
"step": 18
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.02,
|
| 265 |
+
"learning_rate": 1.8586956521739132e-05,
|
| 266 |
+
"logits/chosen": -1.2428462505340576,
|
| 267 |
+
"logits/rejected": -1.8064367771148682,
|
| 268 |
+
"logps/chosen": -322.7100830078125,
|
| 269 |
+
"logps/rejected": -108.71321105957031,
|
| 270 |
+
"loss": 0.3006,
|
| 271 |
+
"rewards/accuracies": 0.875,
|
| 272 |
+
"rewards/chosen": 6.459216594696045,
|
| 273 |
+
"rewards/margins": 3.9207711219787598,
|
| 274 |
+
"rewards/rejected": 2.5384464263916016,
|
| 275 |
+
"step": 19
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.02,
|
| 279 |
+
"learning_rate": 1.956521739130435e-05,
|
| 280 |
+
"logits/chosen": -1.7773098945617676,
|
| 281 |
+
"logits/rejected": -2.178387403488159,
|
| 282 |
+
"logps/chosen": -357.8656921386719,
|
| 283 |
+
"logps/rejected": -112.65576934814453,
|
| 284 |
+
"loss": 0.3837,
|
| 285 |
+
"rewards/accuracies": 0.9375,
|
| 286 |
+
"rewards/chosen": 4.684590816497803,
|
| 287 |
+
"rewards/margins": 4.16137170791626,
|
| 288 |
+
"rewards/rejected": 0.5232191681861877,
|
| 289 |
+
"step": 20
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.02,
|
| 293 |
+
"learning_rate": 2.0543478260869567e-05,
|
| 294 |
+
"logits/chosen": -1.6802027225494385,
|
| 295 |
+
"logits/rejected": -1.943274974822998,
|
| 296 |
+
"logps/chosen": -327.85693359375,
|
| 297 |
+
"logps/rejected": -87.53279113769531,
|
| 298 |
+
"loss": 0.4428,
|
| 299 |
+
"rewards/accuracies": 0.9375,
|
| 300 |
+
"rewards/chosen": 2.681103229522705,
|
| 301 |
+
"rewards/margins": 2.3518502712249756,
|
| 302 |
+
"rewards/rejected": 0.32925307750701904,
|
| 303 |
+
"step": 21
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.02,
|
| 307 |
+
"learning_rate": 2.1521739130434784e-05,
|
| 308 |
+
"logits/chosen": -1.5789223909378052,
|
| 309 |
+
"logits/rejected": -2.4593281745910645,
|
| 310 |
+
"logps/chosen": -590.5732421875,
|
| 311 |
+
"logps/rejected": -86.78819274902344,
|
| 312 |
+
"loss": 0.252,
|
| 313 |
+
"rewards/accuracies": 0.875,
|
| 314 |
+
"rewards/chosen": 3.6503655910491943,
|
| 315 |
+
"rewards/margins": 3.621542453765869,
|
| 316 |
+
"rewards/rejected": 0.028822531923651695,
|
| 317 |
+
"step": 22
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.03,
|
| 321 |
+
"learning_rate": 2.25e-05,
|
| 322 |
+
"logits/chosen": -1.8142305612564087,
|
| 323 |
+
"logits/rejected": -2.3507237434387207,
|
| 324 |
+
"logps/chosen": -296.42266845703125,
|
| 325 |
+
"logps/rejected": -60.37131118774414,
|
| 326 |
+
"loss": 0.5602,
|
| 327 |
+
"rewards/accuracies": 0.75,
|
| 328 |
+
"rewards/chosen": 2.170171022415161,
|
| 329 |
+
"rewards/margins": 2.5353941917419434,
|
| 330 |
+
"rewards/rejected": -0.36522313952445984,
|
| 331 |
+
"step": 23
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.03,
|
| 335 |
+
"learning_rate": 2.347826086956522e-05,
|
| 336 |
+
"logits/chosen": -1.7050156593322754,
|
| 337 |
+
"logits/rejected": -2.158708333969116,
|
| 338 |
+
"logps/chosen": -523.41748046875,
|
| 339 |
+
"logps/rejected": -81.6954345703125,
|
| 340 |
+
"loss": 0.7359,
|
| 341 |
+
"rewards/accuracies": 0.9375,
|
| 342 |
+
"rewards/chosen": 2.3799068927764893,
|
| 343 |
+
"rewards/margins": 2.963341236114502,
|
| 344 |
+
"rewards/rejected": -0.583433985710144,
|
| 345 |
+
"step": 24
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.03,
|
| 349 |
+
"learning_rate": 2.4456521739130436e-05,
|
| 350 |
+
"logits/chosen": -1.704190731048584,
|
| 351 |
+
"logits/rejected": -2.215942859649658,
|
| 352 |
+
"logps/chosen": -457.6519775390625,
|
| 353 |
+
"logps/rejected": -91.778076171875,
|
| 354 |
+
"loss": 0.0865,
|
| 355 |
+
"rewards/accuracies": 1.0,
|
| 356 |
+
"rewards/chosen": 4.57167911529541,
|
| 357 |
+
"rewards/margins": 4.6597795486450195,
|
| 358 |
+
"rewards/rejected": -0.08810039609670639,
|
| 359 |
+
"step": 25
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.03,
|
| 363 |
+
"learning_rate": 2.5434782608695653e-05,
|
| 364 |
+
"logits/chosen": -1.4195502996444702,
|
| 365 |
+
"logits/rejected": -2.109910488128662,
|
| 366 |
+
"logps/chosen": -445.05908203125,
|
| 367 |
+
"logps/rejected": -80.95398712158203,
|
| 368 |
+
"loss": 0.1265,
|
| 369 |
+
"rewards/accuracies": 0.9375,
|
| 370 |
+
"rewards/chosen": 5.362960338592529,
|
| 371 |
+
"rewards/margins": 5.412578582763672,
|
| 372 |
+
"rewards/rejected": -0.049618080258369446,
|
| 373 |
+
"step": 26
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.03,
|
| 377 |
+
"learning_rate": 2.6413043478260874e-05,
|
| 378 |
+
"logits/chosen": -1.0274971723556519,
|
| 379 |
+
"logits/rejected": -1.7681918144226074,
|
| 380 |
+
"logps/chosen": -393.85870361328125,
|
| 381 |
+
"logps/rejected": -77.79011535644531,
|
| 382 |
+
"loss": 0.1702,
|
| 383 |
+
"rewards/accuracies": 1.0,
|
| 384 |
+
"rewards/chosen": 4.572484016418457,
|
| 385 |
+
"rewards/margins": 4.946743488311768,
|
| 386 |
+
"rewards/rejected": -0.374259889125824,
|
| 387 |
+
"step": 27
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.03,
|
| 391 |
+
"learning_rate": 2.739130434782609e-05,
|
| 392 |
+
"logits/chosen": -0.7365273237228394,
|
| 393 |
+
"logits/rejected": -1.9988534450531006,
|
| 394 |
+
"logps/chosen": -569.8936767578125,
|
| 395 |
+
"logps/rejected": -79.71562194824219,
|
| 396 |
+
"loss": 0.0465,
|
| 397 |
+
"rewards/accuracies": 1.0,
|
| 398 |
+
"rewards/chosen": 12.751686096191406,
|
| 399 |
+
"rewards/margins": 12.474647521972656,
|
| 400 |
+
"rewards/rejected": 0.2770393192768097,
|
| 401 |
+
"step": 28
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.03,
|
| 405 |
+
"learning_rate": 2.836956521739131e-05,
|
| 406 |
+
"logits/chosen": -0.6424872279167175,
|
| 407 |
+
"logits/rejected": -1.6069284677505493,
|
| 408 |
+
"logps/chosen": -368.68292236328125,
|
| 409 |
+
"logps/rejected": -103.9141845703125,
|
| 410 |
+
"loss": 0.0351,
|
| 411 |
+
"rewards/accuracies": 1.0,
|
| 412 |
+
"rewards/chosen": 10.144453048706055,
|
| 413 |
+
"rewards/margins": 10.381340980529785,
|
| 414 |
+
"rewards/rejected": -0.23688830435276031,
|
| 415 |
+
"step": 29
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.03,
|
| 419 |
+
"learning_rate": 2.9347826086956523e-05,
|
| 420 |
+
"logits/chosen": -0.6061868071556091,
|
| 421 |
+
"logits/rejected": -1.4256294965744019,
|
| 422 |
+
"logps/chosen": -360.4029846191406,
|
| 423 |
+
"logps/rejected": -130.27224731445312,
|
| 424 |
+
"loss": 0.0414,
|
| 425 |
+
"rewards/accuracies": 1.0,
|
| 426 |
+
"rewards/chosen": 12.649539947509766,
|
| 427 |
+
"rewards/margins": 12.36604118347168,
|
| 428 |
+
"rewards/rejected": 0.2834990620613098,
|
| 429 |
+
"step": 30
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.03,
|
| 433 |
+
"learning_rate": 3.032608695652174e-05,
|
| 434 |
+
"logits/chosen": -1.0075111389160156,
|
| 435 |
+
"logits/rejected": -1.3918204307556152,
|
| 436 |
+
"logps/chosen": -227.05450439453125,
|
| 437 |
+
"logps/rejected": -172.67947387695312,
|
| 438 |
+
"loss": 1.1145,
|
| 439 |
+
"rewards/accuracies": 0.875,
|
| 440 |
+
"rewards/chosen": 4.628115653991699,
|
| 441 |
+
"rewards/margins": 3.9521055221557617,
|
| 442 |
+
"rewards/rejected": 0.6760098934173584,
|
| 443 |
+
"step": 31
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.03,
|
| 447 |
+
"learning_rate": 3.130434782608696e-05,
|
| 448 |
+
"logits/chosen": -0.9060839414596558,
|
| 449 |
+
"logits/rejected": -2.0123891830444336,
|
| 450 |
+
"logps/chosen": -403.06781005859375,
|
| 451 |
+
"logps/rejected": -75.35684967041016,
|
| 452 |
+
"loss": 0.0319,
|
| 453 |
+
"rewards/accuracies": 1.0,
|
| 454 |
+
"rewards/chosen": 7.352475643157959,
|
| 455 |
+
"rewards/margins": 8.778717041015625,
|
| 456 |
+
"rewards/rejected": -1.4262410402297974,
|
| 457 |
+
"step": 32
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.04,
|
| 461 |
+
"learning_rate": 3.228260869565217e-05,
|
| 462 |
+
"logits/chosen": -1.2984271049499512,
|
| 463 |
+
"logits/rejected": -1.9507720470428467,
|
| 464 |
+
"logps/chosen": -306.7958984375,
|
| 465 |
+
"logps/rejected": -107.01404571533203,
|
| 466 |
+
"loss": 0.0795,
|
| 467 |
+
"rewards/accuracies": 0.9375,
|
| 468 |
+
"rewards/chosen": 5.223363399505615,
|
| 469 |
+
"rewards/margins": 7.012631416320801,
|
| 470 |
+
"rewards/rejected": -1.7892680168151855,
|
| 471 |
+
"step": 33
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.04,
|
| 475 |
+
"learning_rate": 3.326086956521739e-05,
|
| 476 |
+
"logits/chosen": -1.1749062538146973,
|
| 477 |
+
"logits/rejected": -2.004805564880371,
|
| 478 |
+
"logps/chosen": -437.9205322265625,
|
| 479 |
+
"logps/rejected": -135.21376037597656,
|
| 480 |
+
"loss": 0.0021,
|
| 481 |
+
"rewards/accuracies": 1.0,
|
| 482 |
+
"rewards/chosen": 9.3555908203125,
|
| 483 |
+
"rewards/margins": 12.285521507263184,
|
| 484 |
+
"rewards/rejected": -2.9299299716949463,
|
| 485 |
+
"step": 34
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.04,
|
| 489 |
+
"learning_rate": 3.423913043478261e-05,
|
| 490 |
+
"logits/chosen": -0.9383536577224731,
|
| 491 |
+
"logits/rejected": -2.0366413593292236,
|
| 492 |
+
"logps/chosen": -457.8018493652344,
|
| 493 |
+
"logps/rejected": -87.7723617553711,
|
| 494 |
+
"loss": 0.041,
|
| 495 |
+
"rewards/accuracies": 1.0,
|
| 496 |
+
"rewards/chosen": 6.836911678314209,
|
| 497 |
+
"rewards/margins": 9.43667984008789,
|
| 498 |
+
"rewards/rejected": -2.599769115447998,
|
| 499 |
+
"step": 35
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.04,
|
| 503 |
+
"learning_rate": 3.521739130434783e-05,
|
| 504 |
+
"logits/chosen": -1.120687484741211,
|
| 505 |
+
"logits/rejected": -1.9933563470840454,
|
| 506 |
+
"logps/chosen": -439.6186828613281,
|
| 507 |
+
"logps/rejected": -94.24763488769531,
|
| 508 |
+
"loss": 0.1906,
|
| 509 |
+
"rewards/accuracies": 0.9375,
|
| 510 |
+
"rewards/chosen": 6.565365314483643,
|
| 511 |
+
"rewards/margins": 9.163484573364258,
|
| 512 |
+
"rewards/rejected": -2.5981192588806152,
|
| 513 |
+
"step": 36
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.04,
|
| 517 |
+
"learning_rate": 3.619565217391305e-05,
|
| 518 |
+
"logits/chosen": -1.338189959526062,
|
| 519 |
+
"logits/rejected": -1.9118335247039795,
|
| 520 |
+
"logps/chosen": -286.96728515625,
|
| 521 |
+
"logps/rejected": -127.22909545898438,
|
| 522 |
+
"loss": 0.0279,
|
| 523 |
+
"rewards/accuracies": 1.0,
|
| 524 |
+
"rewards/chosen": 2.9341468811035156,
|
| 525 |
+
"rewards/margins": 6.362301349639893,
|
| 526 |
+
"rewards/rejected": -3.4281537532806396,
|
| 527 |
+
"step": 37
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.04,
|
| 531 |
+
"learning_rate": 3.7173913043478264e-05,
|
| 532 |
+
"logits/chosen": -1.2371840476989746,
|
| 533 |
+
"logits/rejected": -1.7362310886383057,
|
| 534 |
+
"logps/chosen": -319.52911376953125,
|
| 535 |
+
"logps/rejected": -151.65084838867188,
|
| 536 |
+
"loss": 0.0939,
|
| 537 |
+
"rewards/accuracies": 0.9375,
|
| 538 |
+
"rewards/chosen": 6.216956615447998,
|
| 539 |
+
"rewards/margins": 9.073128700256348,
|
| 540 |
+
"rewards/rejected": -2.856172561645508,
|
| 541 |
+
"step": 38
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.04,
|
| 545 |
+
"learning_rate": 3.815217391304348e-05,
|
| 546 |
+
"logits/chosen": -0.8356418609619141,
|
| 547 |
+
"logits/rejected": -1.7252635955810547,
|
| 548 |
+
"logps/chosen": -391.63275146484375,
|
| 549 |
+
"logps/rejected": -154.64361572265625,
|
| 550 |
+
"loss": 0.2483,
|
| 551 |
+
"rewards/accuracies": 0.9375,
|
| 552 |
+
"rewards/chosen": 8.802470207214355,
|
| 553 |
+
"rewards/margins": 11.676048278808594,
|
| 554 |
+
"rewards/rejected": -2.873577356338501,
|
| 555 |
+
"step": 39
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.04,
|
| 559 |
+
"learning_rate": 3.91304347826087e-05,
|
| 560 |
+
"logits/chosen": -0.9311200380325317,
|
| 561 |
+
"logits/rejected": -1.8107151985168457,
|
| 562 |
+
"logps/chosen": -374.2525939941406,
|
| 563 |
+
"logps/rejected": -131.86715698242188,
|
| 564 |
+
"loss": 0.0058,
|
| 565 |
+
"rewards/accuracies": 1.0,
|
| 566 |
+
"rewards/chosen": 8.586356163024902,
|
| 567 |
+
"rewards/margins": 11.15583324432373,
|
| 568 |
+
"rewards/rejected": -2.569479465484619,
|
| 569 |
+
"step": 40
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.04,
|
| 573 |
+
"learning_rate": 4.0108695652173916e-05,
|
| 574 |
+
"logits/chosen": -0.6861732006072998,
|
| 575 |
+
"logits/rejected": -1.6295411586761475,
|
| 576 |
+
"logps/chosen": -311.7944641113281,
|
| 577 |
+
"logps/rejected": -132.4564666748047,
|
| 578 |
+
"loss": 0.0124,
|
| 579 |
+
"rewards/accuracies": 1.0,
|
| 580 |
+
"rewards/chosen": 5.333560943603516,
|
| 581 |
+
"rewards/margins": 10.023443222045898,
|
| 582 |
+
"rewards/rejected": -4.689882755279541,
|
| 583 |
+
"step": 41
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.05,
|
| 587 |
+
"learning_rate": 4.1086956521739134e-05,
|
| 588 |
+
"logits/chosen": -1.1133558750152588,
|
| 589 |
+
"logits/rejected": -2.0819039344787598,
|
| 590 |
+
"logps/chosen": -282.22332763671875,
|
| 591 |
+
"logps/rejected": -98.7315673828125,
|
| 592 |
+
"loss": 0.0156,
|
| 593 |
+
"rewards/accuracies": 1.0,
|
| 594 |
+
"rewards/chosen": 4.930261611938477,
|
| 595 |
+
"rewards/margins": 8.344033241271973,
|
| 596 |
+
"rewards/rejected": -3.413771629333496,
|
| 597 |
+
"step": 42
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.05,
|
| 601 |
+
"learning_rate": 4.206521739130435e-05,
|
| 602 |
+
"logits/chosen": -1.0311042070388794,
|
| 603 |
+
"logits/rejected": -2.0758776664733887,
|
| 604 |
+
"logps/chosen": -369.7279052734375,
|
| 605 |
+
"logps/rejected": -146.08364868164062,
|
| 606 |
+
"loss": 0.0127,
|
| 607 |
+
"rewards/accuracies": 1.0,
|
| 608 |
+
"rewards/chosen": 7.7552289962768555,
|
| 609 |
+
"rewards/margins": 13.249262809753418,
|
| 610 |
+
"rewards/rejected": -5.494032859802246,
|
| 611 |
+
"step": 43
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.05,
|
| 615 |
+
"learning_rate": 4.304347826086957e-05,
|
| 616 |
+
"logits/chosen": -1.0191632509231567,
|
| 617 |
+
"logits/rejected": -1.995982050895691,
|
| 618 |
+
"logps/chosen": -367.692626953125,
|
| 619 |
+
"logps/rejected": -152.65135192871094,
|
| 620 |
+
"loss": 0.0004,
|
| 621 |
+
"rewards/accuracies": 1.0,
|
| 622 |
+
"rewards/chosen": 7.9517822265625,
|
| 623 |
+
"rewards/margins": 13.869260787963867,
|
| 624 |
+
"rewards/rejected": -5.917478084564209,
|
| 625 |
+
"step": 44
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.05,
|
| 629 |
+
"learning_rate": 4.4021739130434786e-05,
|
| 630 |
+
"logits/chosen": -1.3430471420288086,
|
| 631 |
+
"logits/rejected": -2.0910849571228027,
|
| 632 |
+
"logps/chosen": -334.8968811035156,
|
| 633 |
+
"logps/rejected": -168.4519500732422,
|
| 634 |
+
"loss": 0.0057,
|
| 635 |
+
"rewards/accuracies": 1.0,
|
| 636 |
+
"rewards/chosen": 5.113959312438965,
|
| 637 |
+
"rewards/margins": 12.524876594543457,
|
| 638 |
+
"rewards/rejected": -7.41091775894165,
|
| 639 |
+
"step": 45
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.05,
|
| 643 |
+
"learning_rate": 4.5e-05,
|
| 644 |
+
"logits/chosen": -0.8516018986701965,
|
| 645 |
+
"logits/rejected": -1.9934579133987427,
|
| 646 |
+
"logps/chosen": -386.7164001464844,
|
| 647 |
+
"logps/rejected": -152.8460235595703,
|
| 648 |
+
"loss": 0.0253,
|
| 649 |
+
"rewards/accuracies": 1.0,
|
| 650 |
+
"rewards/chosen": 7.0803022384643555,
|
| 651 |
+
"rewards/margins": 13.202106475830078,
|
| 652 |
+
"rewards/rejected": -6.121803283691406,
|
| 653 |
+
"step": 46
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.05,
|
| 657 |
+
"learning_rate": 4.494827586206897e-05,
|
| 658 |
+
"logits/chosen": -0.9460695385932922,
|
| 659 |
+
"logits/rejected": -1.9993284940719604,
|
| 660 |
+
"logps/chosen": -355.2873229980469,
|
| 661 |
+
"logps/rejected": -188.5989532470703,
|
| 662 |
+
"loss": 0.1008,
|
| 663 |
+
"rewards/accuracies": 0.9375,
|
| 664 |
+
"rewards/chosen": 6.138028144836426,
|
| 665 |
+
"rewards/margins": 13.93982982635498,
|
| 666 |
+
"rewards/rejected": -7.801800727844238,
|
| 667 |
+
"step": 47
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.05,
|
| 671 |
+
"learning_rate": 4.489655172413793e-05,
|
| 672 |
+
"logits/chosen": -1.212510108947754,
|
| 673 |
+
"logits/rejected": -2.0202791690826416,
|
| 674 |
+
"logps/chosen": -293.0318908691406,
|
| 675 |
+
"logps/rejected": -194.92649841308594,
|
| 676 |
+
"loss": 0.0058,
|
| 677 |
+
"rewards/accuracies": 1.0,
|
| 678 |
+
"rewards/chosen": 6.742659091949463,
|
| 679 |
+
"rewards/margins": 17.209505081176758,
|
| 680 |
+
"rewards/rejected": -10.466848373413086,
|
| 681 |
+
"step": 48
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.05,
|
| 685 |
+
"learning_rate": 4.48448275862069e-05,
|
| 686 |
+
"logits/chosen": -1.388157606124878,
|
| 687 |
+
"logits/rejected": -2.1885251998901367,
|
| 688 |
+
"logps/chosen": -330.6868591308594,
|
| 689 |
+
"logps/rejected": -214.35423278808594,
|
| 690 |
+
"loss": 0.0003,
|
| 691 |
+
"rewards/accuracies": 1.0,
|
| 692 |
+
"rewards/chosen": 5.653369903564453,
|
| 693 |
+
"rewards/margins": 18.257244110107422,
|
| 694 |
+
"rewards/rejected": -12.603873252868652,
|
| 695 |
+
"step": 49
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.05,
|
| 699 |
+
"learning_rate": 4.479310344827587e-05,
|
| 700 |
+
"logits/chosen": -1.020527958869934,
|
| 701 |
+
"logits/rejected": -1.9433815479278564,
|
| 702 |
+
"logps/chosen": -347.48394775390625,
|
| 703 |
+
"logps/rejected": -191.15147399902344,
|
| 704 |
+
"loss": 0.0001,
|
| 705 |
+
"rewards/accuracies": 1.0,
|
| 706 |
+
"rewards/chosen": 7.30579137802124,
|
| 707 |
+
"rewards/margins": 18.734811782836914,
|
| 708 |
+
"rewards/rejected": -11.429021835327148,
|
| 709 |
+
"step": 50
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.06,
|
| 713 |
+
"learning_rate": 4.474137931034483e-05,
|
| 714 |
+
"logits/chosen": -1.2260757684707642,
|
| 715 |
+
"logits/rejected": -2.0841352939605713,
|
| 716 |
+
"logps/chosen": -243.5459747314453,
|
| 717 |
+
"logps/rejected": -188.00442504882812,
|
| 718 |
+
"loss": 0.0002,
|
| 719 |
+
"rewards/accuracies": 1.0,
|
| 720 |
+
"rewards/chosen": 3.675260543823242,
|
| 721 |
+
"rewards/margins": 14.774760246276855,
|
| 722 |
+
"rewards/rejected": -11.099499702453613,
|
| 723 |
+
"step": 51
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.06,
|
| 727 |
+
"learning_rate": 4.46896551724138e-05,
|
| 728 |
+
"logits/chosen": -1.0378711223602295,
|
| 729 |
+
"logits/rejected": -2.0174639225006104,
|
| 730 |
+
"logps/chosen": -375.9520263671875,
|
| 731 |
+
"logps/rejected": -182.97463989257812,
|
| 732 |
+
"loss": 0.0,
|
| 733 |
+
"rewards/accuracies": 1.0,
|
| 734 |
+
"rewards/chosen": 7.18237829208374,
|
| 735 |
+
"rewards/margins": 19.010208129882812,
|
| 736 |
+
"rewards/rejected": -11.82783317565918,
|
| 737 |
+
"step": 52
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.06,
|
| 741 |
+
"learning_rate": 4.4637931034482765e-05,
|
| 742 |
+
"logits/chosen": -1.1728301048278809,
|
| 743 |
+
"logits/rejected": -1.9667140245437622,
|
| 744 |
+
"logps/chosen": -309.5382995605469,
|
| 745 |
+
"logps/rejected": -241.13743591308594,
|
| 746 |
+
"loss": 0.0,
|
| 747 |
+
"rewards/accuracies": 1.0,
|
| 748 |
+
"rewards/chosen": 6.681103706359863,
|
| 749 |
+
"rewards/margins": 21.542236328125,
|
| 750 |
+
"rewards/rejected": -14.861133575439453,
|
| 751 |
+
"step": 53
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.06,
|
| 755 |
+
"learning_rate": 4.4586206896551726e-05,
|
| 756 |
+
"logits/chosen": -1.1491725444793701,
|
| 757 |
+
"logits/rejected": -1.760243535041809,
|
| 758 |
+
"logps/chosen": -250.09036254882812,
|
| 759 |
+
"logps/rejected": -226.221435546875,
|
| 760 |
+
"loss": 0.0001,
|
| 761 |
+
"rewards/accuracies": 1.0,
|
| 762 |
+
"rewards/chosen": 2.3173880577087402,
|
| 763 |
+
"rewards/margins": 16.417604446411133,
|
| 764 |
+
"rewards/rejected": -14.100215911865234,
|
| 765 |
+
"step": 54
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.06,
|
| 769 |
+
"learning_rate": 4.4534482758620694e-05,
|
| 770 |
+
"logits/chosen": -1.2637637853622437,
|
| 771 |
+
"logits/rejected": -1.8996949195861816,
|
| 772 |
+
"logps/chosen": -401.9977722167969,
|
| 773 |
+
"logps/rejected": -303.9198303222656,
|
| 774 |
+
"loss": 0.0011,
|
| 775 |
+
"rewards/accuracies": 1.0,
|
| 776 |
+
"rewards/chosen": 8.595758438110352,
|
| 777 |
+
"rewards/margins": 26.598325729370117,
|
| 778 |
+
"rewards/rejected": -18.002567291259766,
|
| 779 |
+
"step": 55
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.06,
|
| 783 |
+
"learning_rate": 4.4482758620689656e-05,
|
| 784 |
+
"logits/chosen": -0.9855178594589233,
|
| 785 |
+
"logits/rejected": -2.0190346240997314,
|
| 786 |
+
"logps/chosen": -394.40625,
|
| 787 |
+
"logps/rejected": -253.97720336914062,
|
| 788 |
+
"loss": 0.0,
|
| 789 |
+
"rewards/accuracies": 1.0,
|
| 790 |
+
"rewards/chosen": 6.395060062408447,
|
| 791 |
+
"rewards/margins": 23.680639266967773,
|
| 792 |
+
"rewards/rejected": -17.28557777404785,
|
| 793 |
+
"step": 56
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.06,
|
| 797 |
+
"learning_rate": 4.4431034482758624e-05,
|
| 798 |
+
"logits/chosen": -1.049959421157837,
|
| 799 |
+
"logits/rejected": -1.6531357765197754,
|
| 800 |
+
"logps/chosen": -205.54058837890625,
|
| 801 |
+
"logps/rejected": -250.66958618164062,
|
| 802 |
+
"loss": 0.0,
|
| 803 |
+
"rewards/accuracies": 1.0,
|
| 804 |
+
"rewards/chosen": 4.874389171600342,
|
| 805 |
+
"rewards/margins": 20.083953857421875,
|
| 806 |
+
"rewards/rejected": -15.209564208984375,
|
| 807 |
+
"step": 57
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.06,
|
| 811 |
+
"learning_rate": 4.4379310344827585e-05,
|
| 812 |
+
"logits/chosen": -0.8982111811637878,
|
| 813 |
+
"logits/rejected": -1.6001371145248413,
|
| 814 |
+
"logps/chosen": -241.91790771484375,
|
| 815 |
+
"logps/rejected": -252.10299682617188,
|
| 816 |
+
"loss": 0.0,
|
| 817 |
+
"rewards/accuracies": 1.0,
|
| 818 |
+
"rewards/chosen": 6.044814109802246,
|
| 819 |
+
"rewards/margins": 21.77715301513672,
|
| 820 |
+
"rewards/rejected": -15.732340812683105,
|
| 821 |
+
"step": 58
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.06,
|
| 825 |
+
"learning_rate": 4.432758620689655e-05,
|
| 826 |
+
"logits/chosen": -0.753044068813324,
|
| 827 |
+
"logits/rejected": -1.6325318813323975,
|
| 828 |
+
"logps/chosen": -315.30230712890625,
|
| 829 |
+
"logps/rejected": -245.91976928710938,
|
| 830 |
+
"loss": 0.0078,
|
| 831 |
+
"rewards/accuracies": 1.0,
|
| 832 |
+
"rewards/chosen": 5.801758289337158,
|
| 833 |
+
"rewards/margins": 20.689924240112305,
|
| 834 |
+
"rewards/rejected": -14.888164520263672,
|
| 835 |
+
"step": 59
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.07,
|
| 839 |
+
"learning_rate": 4.427586206896552e-05,
|
| 840 |
+
"logits/chosen": -0.5443448424339294,
|
| 841 |
+
"logits/rejected": -1.7224345207214355,
|
| 842 |
+
"logps/chosen": -388.339599609375,
|
| 843 |
+
"logps/rejected": -211.5500946044922,
|
| 844 |
+
"loss": 0.0,
|
| 845 |
+
"rewards/accuracies": 1.0,
|
| 846 |
+
"rewards/chosen": 11.5880708694458,
|
| 847 |
+
"rewards/margins": 24.889877319335938,
|
| 848 |
+
"rewards/rejected": -13.301810264587402,
|
| 849 |
+
"step": 60
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.07,
|
| 853 |
+
"learning_rate": 4.422413793103448e-05,
|
| 854 |
+
"logits/chosen": -0.6921290755271912,
|
| 855 |
+
"logits/rejected": -1.6139135360717773,
|
| 856 |
+
"logps/chosen": -242.47312927246094,
|
| 857 |
+
"logps/rejected": -214.73622131347656,
|
| 858 |
+
"loss": 0.0,
|
| 859 |
+
"rewards/accuracies": 1.0,
|
| 860 |
+
"rewards/chosen": 11.329044342041016,
|
| 861 |
+
"rewards/margins": 21.45581817626953,
|
| 862 |
+
"rewards/rejected": -10.1267728805542,
|
| 863 |
+
"step": 61
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.07,
|
| 867 |
+
"learning_rate": 4.417241379310345e-05,
|
| 868 |
+
"logits/chosen": -0.4828271269798279,
|
| 869 |
+
"logits/rejected": -1.649349570274353,
|
| 870 |
+
"logps/chosen": -306.6004638671875,
|
| 871 |
+
"logps/rejected": -172.2523956298828,
|
| 872 |
+
"loss": 0.0003,
|
| 873 |
+
"rewards/accuracies": 1.0,
|
| 874 |
+
"rewards/chosen": 11.373737335205078,
|
| 875 |
+
"rewards/margins": 20.521434783935547,
|
| 876 |
+
"rewards/rejected": -9.147698402404785,
|
| 877 |
+
"step": 62
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.07,
|
| 881 |
+
"learning_rate": 4.412068965517242e-05,
|
| 882 |
+
"logits/chosen": -0.4676312506198883,
|
| 883 |
+
"logits/rejected": -1.6393827199935913,
|
| 884 |
+
"logps/chosen": -357.127685546875,
|
| 885 |
+
"logps/rejected": -199.84524536132812,
|
| 886 |
+
"loss": 0.0,
|
| 887 |
+
"rewards/accuracies": 1.0,
|
| 888 |
+
"rewards/chosen": 16.730457305908203,
|
| 889 |
+
"rewards/margins": 26.409488677978516,
|
| 890 |
+
"rewards/rejected": -9.679031372070312,
|
| 891 |
+
"step": 63
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.07,
|
| 895 |
+
"learning_rate": 4.406896551724138e-05,
|
| 896 |
+
"logits/chosen": -0.7774850726127625,
|
| 897 |
+
"logits/rejected": -1.324657678604126,
|
| 898 |
+
"logps/chosen": -193.75833129882812,
|
| 899 |
+
"logps/rejected": -194.70233154296875,
|
| 900 |
+
"loss": 0.0015,
|
| 901 |
+
"rewards/accuracies": 1.0,
|
| 902 |
+
"rewards/chosen": 5.371121406555176,
|
| 903 |
+
"rewards/margins": 15.129293441772461,
|
| 904 |
+
"rewards/rejected": -9.758172988891602,
|
| 905 |
+
"step": 64
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.07,
|
| 909 |
+
"learning_rate": 4.401724137931035e-05,
|
| 910 |
+
"logits/chosen": -0.5502160787582397,
|
| 911 |
+
"logits/rejected": -1.466262936592102,
|
| 912 |
+
"logps/chosen": -240.96836853027344,
|
| 913 |
+
"logps/rejected": -174.6772003173828,
|
| 914 |
+
"loss": 0.0,
|
| 915 |
+
"rewards/accuracies": 1.0,
|
| 916 |
+
"rewards/chosen": 7.808657169342041,
|
| 917 |
+
"rewards/margins": 17.520769119262695,
|
| 918 |
+
"rewards/rejected": -9.712109565734863,
|
| 919 |
+
"step": 65
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.07,
|
| 923 |
+
"learning_rate": 4.3965517241379315e-05,
|
| 924 |
+
"logits/chosen": -0.5615445375442505,
|
| 925 |
+
"logits/rejected": -1.3243337869644165,
|
| 926 |
+
"logps/chosen": -282.65484619140625,
|
| 927 |
+
"logps/rejected": -211.33326721191406,
|
| 928 |
+
"loss": 0.4403,
|
| 929 |
+
"rewards/accuracies": 0.9375,
|
| 930 |
+
"rewards/chosen": 9.213937759399414,
|
| 931 |
+
"rewards/margins": 18.93706512451172,
|
| 932 |
+
"rewards/rejected": -9.723125457763672,
|
| 933 |
+
"step": 66
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.07,
|
| 937 |
+
"learning_rate": 4.3913793103448277e-05,
|
| 938 |
+
"logits/chosen": -0.5882927775382996,
|
| 939 |
+
"logits/rejected": -1.4038889408111572,
|
| 940 |
+
"logps/chosen": -244.26333618164062,
|
| 941 |
+
"logps/rejected": -192.05517578125,
|
| 942 |
+
"loss": 0.0007,
|
| 943 |
+
"rewards/accuracies": 1.0,
|
| 944 |
+
"rewards/chosen": 5.999594688415527,
|
| 945 |
+
"rewards/margins": 16.232702255249023,
|
| 946 |
+
"rewards/rejected": -10.233107566833496,
|
| 947 |
+
"step": 67
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.07,
|
| 951 |
+
"learning_rate": 4.3862068965517245e-05,
|
| 952 |
+
"logits/chosen": -0.7710579633712769,
|
| 953 |
+
"logits/rejected": -1.5811477899551392,
|
| 954 |
+
"logps/chosen": -274.4286804199219,
|
| 955 |
+
"logps/rejected": -189.31427001953125,
|
| 956 |
+
"loss": 0.0003,
|
| 957 |
+
"rewards/accuracies": 1.0,
|
| 958 |
+
"rewards/chosen": 10.483807563781738,
|
| 959 |
+
"rewards/margins": 22.807409286499023,
|
| 960 |
+
"rewards/rejected": -12.323600769042969,
|
| 961 |
+
"step": 68
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.08,
|
| 965 |
+
"learning_rate": 4.381034482758621e-05,
|
| 966 |
+
"logits/chosen": -0.2001597136259079,
|
| 967 |
+
"logits/rejected": -1.5477687120437622,
|
| 968 |
+
"logps/chosen": -398.04296875,
|
| 969 |
+
"logps/rejected": -208.8218536376953,
|
| 970 |
+
"loss": 0.0,
|
| 971 |
+
"rewards/accuracies": 1.0,
|
| 972 |
+
"rewards/chosen": 11.122913360595703,
|
| 973 |
+
"rewards/margins": 23.094343185424805,
|
| 974 |
+
"rewards/rejected": -11.971429824829102,
|
| 975 |
+
"step": 69
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.08,
|
| 979 |
+
"learning_rate": 4.3758620689655174e-05,
|
| 980 |
+
"logits/chosen": -0.5891858339309692,
|
| 981 |
+
"logits/rejected": -1.500455379486084,
|
| 982 |
+
"logps/chosen": -223.07235717773438,
|
| 983 |
+
"logps/rejected": -219.37405395507812,
|
| 984 |
+
"loss": 0.0,
|
| 985 |
+
"rewards/accuracies": 1.0,
|
| 986 |
+
"rewards/chosen": 7.498288154602051,
|
| 987 |
+
"rewards/margins": 19.65944480895996,
|
| 988 |
+
"rewards/rejected": -12.16115665435791,
|
| 989 |
+
"step": 70
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 0.08,
|
| 993 |
+
"learning_rate": 4.370689655172414e-05,
|
| 994 |
+
"logits/chosen": -0.5222135186195374,
|
| 995 |
+
"logits/rejected": -1.459835410118103,
|
| 996 |
+
"logps/chosen": -264.3045349121094,
|
| 997 |
+
"logps/rejected": -184.4599609375,
|
| 998 |
+
"loss": 0.0006,
|
| 999 |
+
"rewards/accuracies": 1.0,
|
| 1000 |
+
"rewards/chosen": 7.261447429656982,
|
| 1001 |
+
"rewards/margins": 17.37218475341797,
|
| 1002 |
+
"rewards/rejected": -10.110737800598145,
|
| 1003 |
+
"step": 71
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 0.08,
|
| 1007 |
+
"learning_rate": 4.365517241379311e-05,
|
| 1008 |
+
"logits/chosen": -0.4974746108055115,
|
| 1009 |
+
"logits/rejected": -1.5458664894104004,
|
| 1010 |
+
"logps/chosen": -272.7602233886719,
|
| 1011 |
+
"logps/rejected": -181.00247192382812,
|
| 1012 |
+
"loss": 0.0,
|
| 1013 |
+
"rewards/accuracies": 1.0,
|
| 1014 |
+
"rewards/chosen": 10.925564765930176,
|
| 1015 |
+
"rewards/margins": 22.81020736694336,
|
| 1016 |
+
"rewards/rejected": -11.88464069366455,
|
| 1017 |
+
"step": 72
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.08,
|
| 1021 |
+
"learning_rate": 4.360344827586207e-05,
|
| 1022 |
+
"logits/chosen": -0.3357059359550476,
|
| 1023 |
+
"logits/rejected": -1.4441150426864624,
|
| 1024 |
+
"logps/chosen": -414.8647766113281,
|
| 1025 |
+
"logps/rejected": -223.109375,
|
| 1026 |
+
"loss": 0.0004,
|
| 1027 |
+
"rewards/accuracies": 1.0,
|
| 1028 |
+
"rewards/chosen": 13.181656837463379,
|
| 1029 |
+
"rewards/margins": 27.527368545532227,
|
| 1030 |
+
"rewards/rejected": -14.345712661743164,
|
| 1031 |
+
"step": 73
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.08,
|
| 1035 |
+
"learning_rate": 4.355172413793104e-05,
|
| 1036 |
+
"logits/chosen": -0.6531690359115601,
|
| 1037 |
+
"logits/rejected": -1.4611525535583496,
|
| 1038 |
+
"logps/chosen": -221.1251220703125,
|
| 1039 |
+
"logps/rejected": -226.30352783203125,
|
| 1040 |
+
"loss": 0.0,
|
| 1041 |
+
"rewards/accuracies": 1.0,
|
| 1042 |
+
"rewards/chosen": 7.807477951049805,
|
| 1043 |
+
"rewards/margins": 21.502174377441406,
|
| 1044 |
+
"rewards/rejected": -13.694696426391602,
|
| 1045 |
+
"step": 74
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.08,
|
| 1049 |
+
"learning_rate": 4.35e-05,
|
| 1050 |
+
"logits/chosen": -0.5135932564735413,
|
| 1051 |
+
"logits/rejected": -1.4995931386947632,
|
| 1052 |
+
"logps/chosen": -332.56842041015625,
|
| 1053 |
+
"logps/rejected": -270.9891662597656,
|
| 1054 |
+
"loss": 0.0,
|
| 1055 |
+
"rewards/accuracies": 1.0,
|
| 1056 |
+
"rewards/chosen": 12.301117897033691,
|
| 1057 |
+
"rewards/margins": 26.830467224121094,
|
| 1058 |
+
"rewards/rejected": -14.529346466064453,
|
| 1059 |
+
"step": 75
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.08,
|
| 1063 |
+
"learning_rate": 4.344827586206897e-05,
|
| 1064 |
+
"logits/chosen": -0.3770020604133606,
|
| 1065 |
+
"logits/rejected": -1.44219970703125,
|
| 1066 |
+
"logps/chosen": -317.96240234375,
|
| 1067 |
+
"logps/rejected": -250.3682098388672,
|
| 1068 |
+
"loss": 0.0,
|
| 1069 |
+
"rewards/accuracies": 1.0,
|
| 1070 |
+
"rewards/chosen": 7.8860931396484375,
|
| 1071 |
+
"rewards/margins": 23.53211784362793,
|
| 1072 |
+
"rewards/rejected": -15.646024703979492,
|
| 1073 |
+
"step": 76
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.08,
|
| 1077 |
+
"learning_rate": 4.339655172413793e-05,
|
| 1078 |
+
"logits/chosen": -0.49628371000289917,
|
| 1079 |
+
"logits/rejected": -1.4045476913452148,
|
| 1080 |
+
"logps/chosen": -367.6162414550781,
|
| 1081 |
+
"logps/rejected": -265.2467346191406,
|
| 1082 |
+
"loss": 0.0,
|
| 1083 |
+
"rewards/accuracies": 1.0,
|
| 1084 |
+
"rewards/chosen": 10.148900985717773,
|
| 1085 |
+
"rewards/margins": 24.313669204711914,
|
| 1086 |
+
"rewards/rejected": -14.164766311645508,
|
| 1087 |
+
"step": 77
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 0.09,
|
| 1091 |
+
"learning_rate": 4.33448275862069e-05,
|
| 1092 |
+
"logits/chosen": -0.5287263989448547,
|
| 1093 |
+
"logits/rejected": -1.3665097951889038,
|
| 1094 |
+
"logps/chosen": -341.87664794921875,
|
| 1095 |
+
"logps/rejected": -249.6795654296875,
|
| 1096 |
+
"loss": 0.1517,
|
| 1097 |
+
"rewards/accuracies": 0.9375,
|
| 1098 |
+
"rewards/chosen": 8.620326042175293,
|
| 1099 |
+
"rewards/margins": 19.850658416748047,
|
| 1100 |
+
"rewards/rejected": -11.230331420898438,
|
| 1101 |
+
"step": 78
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.09,
|
| 1105 |
+
"learning_rate": 4.3293103448275865e-05,
|
| 1106 |
+
"logits/chosen": -0.2743966281414032,
|
| 1107 |
+
"logits/rejected": -1.4255130290985107,
|
| 1108 |
+
"logps/chosen": -334.3340759277344,
|
| 1109 |
+
"logps/rejected": -265.15032958984375,
|
| 1110 |
+
"loss": 0.0,
|
| 1111 |
+
"rewards/accuracies": 1.0,
|
| 1112 |
+
"rewards/chosen": 10.872364044189453,
|
| 1113 |
+
"rewards/margins": 27.00977325439453,
|
| 1114 |
+
"rewards/rejected": -16.13741111755371,
|
| 1115 |
+
"step": 79
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 0.09,
|
| 1119 |
+
"learning_rate": 4.324137931034483e-05,
|
| 1120 |
+
"logits/chosen": -0.5563563108444214,
|
| 1121 |
+
"logits/rejected": -1.6934064626693726,
|
| 1122 |
+
"logps/chosen": -408.9612731933594,
|
| 1123 |
+
"logps/rejected": -220.00091552734375,
|
| 1124 |
+
"loss": 0.0,
|
| 1125 |
+
"rewards/accuracies": 1.0,
|
| 1126 |
+
"rewards/chosen": 10.237764358520508,
|
| 1127 |
+
"rewards/margins": 24.824390411376953,
|
| 1128 |
+
"rewards/rejected": -14.586627960205078,
|
| 1129 |
+
"step": 80
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.09,
|
| 1133 |
+
"learning_rate": 4.3189655172413795e-05,
|
| 1134 |
+
"logits/chosen": -0.617451548576355,
|
| 1135 |
+
"logits/rejected": -1.640676736831665,
|
| 1136 |
+
"logps/chosen": -363.1540222167969,
|
| 1137 |
+
"logps/rejected": -226.42376708984375,
|
| 1138 |
+
"loss": 0.0,
|
| 1139 |
+
"rewards/accuracies": 1.0,
|
| 1140 |
+
"rewards/chosen": 10.672024726867676,
|
| 1141 |
+
"rewards/margins": 24.347591400146484,
|
| 1142 |
+
"rewards/rejected": -13.675565719604492,
|
| 1143 |
+
"step": 81
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 0.09,
|
| 1147 |
+
"learning_rate": 4.313793103448276e-05,
|
| 1148 |
+
"logits/chosen": -0.45464372634887695,
|
| 1149 |
+
"logits/rejected": -1.4708175659179688,
|
| 1150 |
+
"logps/chosen": -328.35675048828125,
|
| 1151 |
+
"logps/rejected": -223.79766845703125,
|
| 1152 |
+
"loss": 0.0,
|
| 1153 |
+
"rewards/accuracies": 1.0,
|
| 1154 |
+
"rewards/chosen": 6.965122699737549,
|
| 1155 |
+
"rewards/margins": 21.22077178955078,
|
| 1156 |
+
"rewards/rejected": -14.25564956665039,
|
| 1157 |
+
"step": 82
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.09,
|
| 1161 |
+
"learning_rate": 4.3086206896551724e-05,
|
| 1162 |
+
"logits/chosen": -0.6079340577125549,
|
| 1163 |
+
"logits/rejected": -1.5333082675933838,
|
| 1164 |
+
"logps/chosen": -219.4099578857422,
|
| 1165 |
+
"logps/rejected": -230.07948303222656,
|
| 1166 |
+
"loss": 0.0,
|
| 1167 |
+
"rewards/accuracies": 1.0,
|
| 1168 |
+
"rewards/chosen": 6.278050899505615,
|
| 1169 |
+
"rewards/margins": 20.45656967163086,
|
| 1170 |
+
"rewards/rejected": -14.178520202636719,
|
| 1171 |
+
"step": 83
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.09,
|
| 1175 |
+
"learning_rate": 4.303448275862069e-05,
|
| 1176 |
+
"logits/chosen": -0.4874611794948578,
|
| 1177 |
+
"logits/rejected": -1.5135740041732788,
|
| 1178 |
+
"logps/chosen": -304.00274658203125,
|
| 1179 |
+
"logps/rejected": -263.0120849609375,
|
| 1180 |
+
"loss": 0.0,
|
| 1181 |
+
"rewards/accuracies": 1.0,
|
| 1182 |
+
"rewards/chosen": 9.557435989379883,
|
| 1183 |
+
"rewards/margins": 24.41012954711914,
|
| 1184 |
+
"rewards/rejected": -14.852693557739258,
|
| 1185 |
+
"step": 84
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 0.09,
|
| 1189 |
+
"learning_rate": 4.298275862068966e-05,
|
| 1190 |
+
"logits/chosen": -0.4802761375904083,
|
| 1191 |
+
"logits/rejected": -1.4294252395629883,
|
| 1192 |
+
"logps/chosen": -293.41192626953125,
|
| 1193 |
+
"logps/rejected": -283.2821350097656,
|
| 1194 |
+
"loss": 0.0,
|
| 1195 |
+
"rewards/accuracies": 1.0,
|
| 1196 |
+
"rewards/chosen": 7.370968818664551,
|
| 1197 |
+
"rewards/margins": 24.32880401611328,
|
| 1198 |
+
"rewards/rejected": -16.957834243774414,
|
| 1199 |
+
"step": 85
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 0.09,
|
| 1203 |
+
"learning_rate": 4.293103448275862e-05,
|
| 1204 |
+
"logits/chosen": -0.3580012321472168,
|
| 1205 |
+
"logits/rejected": -1.499201774597168,
|
| 1206 |
+
"logps/chosen": -350.5351257324219,
|
| 1207 |
+
"logps/rejected": -244.6523895263672,
|
| 1208 |
+
"loss": 0.0001,
|
| 1209 |
+
"rewards/accuracies": 1.0,
|
| 1210 |
+
"rewards/chosen": 12.237674713134766,
|
| 1211 |
+
"rewards/margins": 26.58084487915039,
|
| 1212 |
+
"rewards/rejected": -14.343170166015625,
|
| 1213 |
+
"step": 86
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 0.09,
|
| 1217 |
+
"learning_rate": 4.287931034482759e-05,
|
| 1218 |
+
"logits/chosen": -0.6917211413383484,
|
| 1219 |
+
"logits/rejected": -1.4503819942474365,
|
| 1220 |
+
"logps/chosen": -217.5175323486328,
|
| 1221 |
+
"logps/rejected": -228.9481201171875,
|
| 1222 |
+
"loss": 0.0,
|
| 1223 |
+
"rewards/accuracies": 1.0,
|
| 1224 |
+
"rewards/chosen": 6.635045051574707,
|
| 1225 |
+
"rewards/margins": 20.42833709716797,
|
| 1226 |
+
"rewards/rejected": -13.793292999267578,
|
| 1227 |
+
"step": 87
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 0.1,
|
| 1231 |
+
"learning_rate": 4.282758620689656e-05,
|
| 1232 |
+
"logits/chosen": -0.8255922794342041,
|
| 1233 |
+
"logits/rejected": -1.4748644828796387,
|
| 1234 |
+
"logps/chosen": -217.336669921875,
|
| 1235 |
+
"logps/rejected": -206.8959197998047,
|
| 1236 |
+
"loss": 0.0005,
|
| 1237 |
+
"rewards/accuracies": 1.0,
|
| 1238 |
+
"rewards/chosen": 5.22888708114624,
|
| 1239 |
+
"rewards/margins": 19.673564910888672,
|
| 1240 |
+
"rewards/rejected": -14.444679260253906,
|
| 1241 |
+
"step": 88
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 0.1,
|
| 1245 |
+
"learning_rate": 4.2775862068965525e-05,
|
| 1246 |
+
"logits/chosen": -0.626908004283905,
|
| 1247 |
+
"logits/rejected": -1.6663663387298584,
|
| 1248 |
+
"logps/chosen": -328.56231689453125,
|
| 1249 |
+
"logps/rejected": -235.42843627929688,
|
| 1250 |
+
"loss": 0.0,
|
| 1251 |
+
"rewards/accuracies": 1.0,
|
| 1252 |
+
"rewards/chosen": 6.891221046447754,
|
| 1253 |
+
"rewards/margins": 21.77893829345703,
|
| 1254 |
+
"rewards/rejected": -14.887716293334961,
|
| 1255 |
+
"step": 89
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 0.1,
|
| 1259 |
+
"learning_rate": 4.2724137931034486e-05,
|
| 1260 |
+
"logits/chosen": -0.5689429640769958,
|
| 1261 |
+
"logits/rejected": -1.618609070777893,
|
| 1262 |
+
"logps/chosen": -298.08770751953125,
|
| 1263 |
+
"logps/rejected": -237.06094360351562,
|
| 1264 |
+
"loss": 0.0,
|
| 1265 |
+
"rewards/accuracies": 1.0,
|
| 1266 |
+
"rewards/chosen": 9.328791618347168,
|
| 1267 |
+
"rewards/margins": 24.572046279907227,
|
| 1268 |
+
"rewards/rejected": -15.243253707885742,
|
| 1269 |
+
"step": 90
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 0.1,
|
| 1273 |
+
"learning_rate": 4.2672413793103454e-05,
|
| 1274 |
+
"logits/chosen": -0.5474239587783813,
|
| 1275 |
+
"logits/rejected": -1.6142076253890991,
|
| 1276 |
+
"logps/chosen": -295.00494384765625,
|
| 1277 |
+
"logps/rejected": -268.9505615234375,
|
| 1278 |
+
"loss": 0.0287,
|
| 1279 |
+
"rewards/accuracies": 1.0,
|
| 1280 |
+
"rewards/chosen": 9.715471267700195,
|
| 1281 |
+
"rewards/margins": 23.968780517578125,
|
| 1282 |
+
"rewards/rejected": -14.25330924987793,
|
| 1283 |
+
"step": 91
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 0.1,
|
| 1287 |
+
"learning_rate": 4.2620689655172416e-05,
|
| 1288 |
+
"logits/chosen": -0.5132906436920166,
|
| 1289 |
+
"logits/rejected": -1.6248977184295654,
|
| 1290 |
+
"logps/chosen": -314.5284423828125,
|
| 1291 |
+
"logps/rejected": -210.1342315673828,
|
| 1292 |
+
"loss": 0.0595,
|
| 1293 |
+
"rewards/accuracies": 0.9375,
|
| 1294 |
+
"rewards/chosen": 7.707674980163574,
|
| 1295 |
+
"rewards/margins": 20.18167495727539,
|
| 1296 |
+
"rewards/rejected": -12.474000930786133,
|
| 1297 |
+
"step": 92
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 0.1,
|
| 1301 |
+
"learning_rate": 4.2568965517241384e-05,
|
| 1302 |
+
"logits/chosen": -0.45244526863098145,
|
| 1303 |
+
"logits/rejected": -1.7539148330688477,
|
| 1304 |
+
"logps/chosen": -394.9619140625,
|
| 1305 |
+
"logps/rejected": -230.32550048828125,
|
| 1306 |
+
"loss": 0.0,
|
| 1307 |
+
"rewards/accuracies": 1.0,
|
| 1308 |
+
"rewards/chosen": 11.32716178894043,
|
| 1309 |
+
"rewards/margins": 26.326129913330078,
|
| 1310 |
+
"rewards/rejected": -14.998967170715332,
|
| 1311 |
+
"step": 93
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 0.1,
|
| 1315 |
+
"learning_rate": 4.2517241379310345e-05,
|
| 1316 |
+
"logits/chosen": -0.7538139224052429,
|
| 1317 |
+
"logits/rejected": -1.5796599388122559,
|
| 1318 |
+
"logps/chosen": -286.95281982421875,
|
| 1319 |
+
"logps/rejected": -266.7412414550781,
|
| 1320 |
+
"loss": 0.0,
|
| 1321 |
+
"rewards/accuracies": 1.0,
|
| 1322 |
+
"rewards/chosen": 8.003694534301758,
|
| 1323 |
+
"rewards/margins": 25.05299949645996,
|
| 1324 |
+
"rewards/rejected": -17.04930305480957,
|
| 1325 |
+
"step": 94
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 0.1,
|
| 1329 |
+
"learning_rate": 4.246551724137931e-05,
|
| 1330 |
+
"logits/chosen": -0.44376400113105774,
|
| 1331 |
+
"logits/rejected": -1.5493687391281128,
|
| 1332 |
+
"logps/chosen": -406.93951416015625,
|
| 1333 |
+
"logps/rejected": -302.9401550292969,
|
| 1334 |
+
"loss": 0.0,
|
| 1335 |
+
"rewards/accuracies": 1.0,
|
| 1336 |
+
"rewards/chosen": 7.246862888336182,
|
| 1337 |
+
"rewards/margins": 27.937149047851562,
|
| 1338 |
+
"rewards/rejected": -20.69028663635254,
|
| 1339 |
+
"step": 95
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 0.1,
|
| 1343 |
+
"learning_rate": 4.2413793103448274e-05,
|
| 1344 |
+
"logits/chosen": -0.7577255964279175,
|
| 1345 |
+
"logits/rejected": -1.77558171749115,
|
| 1346 |
+
"logps/chosen": -357.90838623046875,
|
| 1347 |
+
"logps/rejected": -246.73458862304688,
|
| 1348 |
+
"loss": 0.0,
|
| 1349 |
+
"rewards/accuracies": 1.0,
|
| 1350 |
+
"rewards/chosen": 7.267894744873047,
|
| 1351 |
+
"rewards/margins": 24.79804229736328,
|
| 1352 |
+
"rewards/rejected": -17.530147552490234,
|
| 1353 |
+
"step": 96
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 0.11,
|
| 1357 |
+
"learning_rate": 4.236206896551724e-05,
|
| 1358 |
+
"logits/chosen": -0.8565713763237,
|
| 1359 |
+
"logits/rejected": -1.7760589122772217,
|
| 1360 |
+
"logps/chosen": -338.751220703125,
|
| 1361 |
+
"logps/rejected": -304.101806640625,
|
| 1362 |
+
"loss": 0.0,
|
| 1363 |
+
"rewards/accuracies": 1.0,
|
| 1364 |
+
"rewards/chosen": 9.238365173339844,
|
| 1365 |
+
"rewards/margins": 27.781978607177734,
|
| 1366 |
+
"rewards/rejected": -18.54361343383789,
|
| 1367 |
+
"step": 97
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 0.11,
|
| 1371 |
+
"learning_rate": 4.231034482758621e-05,
|
| 1372 |
+
"logits/chosen": -0.40401750802993774,
|
| 1373 |
+
"logits/rejected": -1.7701493501663208,
|
| 1374 |
+
"logps/chosen": -537.5440063476562,
|
| 1375 |
+
"logps/rejected": -349.8548889160156,
|
| 1376 |
+
"loss": 0.0,
|
| 1377 |
+
"rewards/accuracies": 1.0,
|
| 1378 |
+
"rewards/chosen": 14.211512565612793,
|
| 1379 |
+
"rewards/margins": 37.746315002441406,
|
| 1380 |
+
"rewards/rejected": -23.53480339050293,
|
| 1381 |
+
"step": 98
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 0.11,
|
| 1385 |
+
"learning_rate": 4.225862068965517e-05,
|
| 1386 |
+
"logits/chosen": -0.9799883365631104,
|
| 1387 |
+
"logits/rejected": -1.8936114311218262,
|
| 1388 |
+
"logps/chosen": -403.7567138671875,
|
| 1389 |
+
"logps/rejected": -257.6401672363281,
|
| 1390 |
+
"loss": 0.0001,
|
| 1391 |
+
"rewards/accuracies": 1.0,
|
| 1392 |
+
"rewards/chosen": 4.710639476776123,
|
| 1393 |
+
"rewards/margins": 23.923961639404297,
|
| 1394 |
+
"rewards/rejected": -19.213321685791016,
|
| 1395 |
+
"step": 99
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 0.11,
|
| 1399 |
+
"learning_rate": 4.220689655172414e-05,
|
| 1400 |
+
"logits/chosen": -0.94623202085495,
|
| 1401 |
+
"logits/rejected": -1.7190004587173462,
|
| 1402 |
+
"logps/chosen": -360.8213806152344,
|
| 1403 |
+
"logps/rejected": -310.1994323730469,
|
| 1404 |
+
"loss": 0.0,
|
| 1405 |
+
"rewards/accuracies": 1.0,
|
| 1406 |
+
"rewards/chosen": 3.991710662841797,
|
| 1407 |
+
"rewards/margins": 26.59444808959961,
|
| 1408 |
+
"rewards/rejected": -22.60273551940918,
|
| 1409 |
+
"step": 100
|
| 1410 |
+
}
|
| 1411 |
+
],
|
| 1412 |
+
"logging_steps": 1,
|
| 1413 |
+
"max_steps": 916,
|
| 1414 |
+
"num_input_tokens_seen": 0,
|
| 1415 |
+
"num_train_epochs": 1,
|
| 1416 |
+
"save_steps": 100,
|
| 1417 |
+
"total_flos": 0.0,
|
| 1418 |
+
"train_batch_size": 1,
|
| 1419 |
+
"trial_name": null,
|
| 1420 |
+
"trial_params": null
|
| 1421 |
+
}
|
checkpoint-100/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6f8e778ebfdc8e189d4259f43aa8cc8438633b1407516e064a080cf26f579c11
|
| 3 |
+
size 4664
|
checkpoint-200/README.md
ADDED
|
@@ -0,0 +1,204 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: peft
|
| 3 |
+
base_model: /run/media/adamo1139/82142F79142F6EFB/ProgramData/Anaconda3/envs/qlora-jondurbin/axolotl-git-linux/axolotl/yi-34b-200k-llamafied
|
| 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 |
+
|
| 201 |
+
|
| 202 |
+
### Framework versions
|
| 203 |
+
|
| 204 |
+
- PEFT 0.7.1
|
checkpoint-200/adapter_config.json
ADDED
|
@@ -0,0 +1,31 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"alpha_pattern": {},
|
| 3 |
+
"auto_mapping": null,
|
| 4 |
+
"base_model_name_or_path": "yi-34b-200k-llamafied",
|
| 5 |
+
"bias": "none",
|
| 6 |
+
"fan_in_fan_out": false,
|
| 7 |
+
"inference_mode": true,
|
| 8 |
+
"init_lora_weights": true,
|
| 9 |
+
"layers_pattern": null,
|
| 10 |
+
"layers_to_transform": null,
|
| 11 |
+
"loftq_config": null,
|
| 12 |
+
"lora_alpha": 32,
|
| 13 |
+
"lora_dropout": 0,
|
| 14 |
+
"megatron_config": null,
|
| 15 |
+
"megatron_core": "megatron.core",
|
| 16 |
+
"modules_to_save": null,
|
| 17 |
+
"peft_type": "LORA",
|
| 18 |
+
"r": 16,
|
| 19 |
+
"rank_pattern": {},
|
| 20 |
+
"revision": "unsloth",
|
| 21 |
+
"target_modules": [
|
| 22 |
+
"v_proj",
|
| 23 |
+
"gate_proj",
|
| 24 |
+
"o_proj",
|
| 25 |
+
"down_proj",
|
| 26 |
+
"q_proj",
|
| 27 |
+
"up_proj",
|
| 28 |
+
"k_proj"
|
| 29 |
+
],
|
| 30 |
+
"task_type": "CAUSAL_LM"
|
| 31 |
+
}
|
checkpoint-200/adapter_model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cd17499f5bbc669e455927592886da54487e914d976a676ade29444f330fde90
|
| 3 |
+
size 491633464
|
checkpoint-200/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1ff264f99d31b522cc7e2a4eac9d38606d0c58a34c0adc74d71e0ca8b371dc36
|
| 3 |
+
size 14244
|
checkpoint-200/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:457295efea71f81baab628dd2db06366d3bb8a3c1850c55839208acbb61044ff
|
| 3 |
+
size 1064
|
checkpoint-200/special_tokens_map.json
ADDED
|
@@ -0,0 +1,30 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": {
|
| 3 |
+
"content": "<|startoftext|>",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": false,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"eos_token": {
|
| 10 |
+
"content": "<|endoftext|>",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": false,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"pad_token": {
|
| 17 |
+
"content": "<unk>",
|
| 18 |
+
"lstrip": false,
|
| 19 |
+
"normalized": false,
|
| 20 |
+
"rstrip": false,
|
| 21 |
+
"single_word": false
|
| 22 |
+
},
|
| 23 |
+
"unk_token": {
|
| 24 |
+
"content": "<unk>",
|
| 25 |
+
"lstrip": false,
|
| 26 |
+
"normalized": false,
|
| 27 |
+
"rstrip": false,
|
| 28 |
+
"single_word": false
|
| 29 |
+
}
|
| 30 |
+
}
|
checkpoint-200/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-200/tokenizer.model
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:386c49cf943d71aa110361135338c50e38beeff0a66593480421f37b319e1a39
|
| 3 |
+
size 1033105
|
checkpoint-200/tokenizer_config.json
ADDED
|
@@ -0,0 +1,42 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_bos_token": false,
|
| 3 |
+
"add_eos_token": false,
|
| 4 |
+
"added_tokens_decoder": {
|
| 5 |
+
"0": {
|
| 6 |
+
"content": "<unk>",
|
| 7 |
+
"lstrip": false,
|
| 8 |
+
"normalized": false,
|
| 9 |
+
"rstrip": false,
|
| 10 |
+
"single_word": false,
|
| 11 |
+
"special": true
|
| 12 |
+
},
|
| 13 |
+
"1": {
|
| 14 |
+
"content": "<|startoftext|>",
|
| 15 |
+
"lstrip": false,
|
| 16 |
+
"normalized": false,
|
| 17 |
+
"rstrip": false,
|
| 18 |
+
"single_word": false,
|
| 19 |
+
"special": true
|
| 20 |
+
},
|
| 21 |
+
"2": {
|
| 22 |
+
"content": "<|endoftext|>",
|
| 23 |
+
"lstrip": false,
|
| 24 |
+
"normalized": false,
|
| 25 |
+
"rstrip": false,
|
| 26 |
+
"single_word": false,
|
| 27 |
+
"special": true
|
| 28 |
+
}
|
| 29 |
+
},
|
| 30 |
+
"bos_token": "<|startoftext|>",
|
| 31 |
+
"chat_template": "{% if not add_generation_prompt is defined %}{% set add_generation_prompt = false %}{% endif %}{% for message in messages %}{{'<|im_start|>' + message['role'] + '\n' + message['content'] + '<|im_end|>' + '\n'}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant\n' }}{% endif %}",
|
| 32 |
+
"clean_up_tokenization_spaces": false,
|
| 33 |
+
"eos_token": "<|endoftext|>",
|
| 34 |
+
"legacy": true,
|
| 35 |
+
"model_max_length": 4096,
|
| 36 |
+
"pad_token": "<unk>",
|
| 37 |
+
"padding_side": "right",
|
| 38 |
+
"sp_model_kwargs": {},
|
| 39 |
+
"tokenizer_class": "LlamaTokenizer",
|
| 40 |
+
"unk_token": "<unk>",
|
| 41 |
+
"use_default_system_prompt": false
|
| 42 |
+
}
|
checkpoint-200/trainer_state.json
ADDED
|
@@ -0,0 +1,2821 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.2181768596168269,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 200,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.0,
|
| 13 |
+
"learning_rate": 9.782608695652175e-07,
|
| 14 |
+
"logits/chosen": -1.8868483304977417,
|
| 15 |
+
"logits/rejected": -2.3036646842956543,
|
| 16 |
+
"logps/chosen": -466.7117004394531,
|
| 17 |
+
"logps/rejected": -99.3152084350586,
|
| 18 |
+
"loss": 0.6931,
|
| 19 |
+
"rewards/accuracies": 0.0,
|
| 20 |
+
"rewards/chosen": 0.0,
|
| 21 |
+
"rewards/margins": 0.0,
|
| 22 |
+
"rewards/rejected": 0.0,
|
| 23 |
+
"step": 1
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.0,
|
| 27 |
+
"learning_rate": 1.956521739130435e-06,
|
| 28 |
+
"logits/chosen": -2.2296698093414307,
|
| 29 |
+
"logits/rejected": -2.469517469406128,
|
| 30 |
+
"logps/chosen": -335.50140380859375,
|
| 31 |
+
"logps/rejected": -97.97496032714844,
|
| 32 |
+
"loss": 0.6931,
|
| 33 |
+
"rewards/accuracies": 0.0,
|
| 34 |
+
"rewards/chosen": 0.0,
|
| 35 |
+
"rewards/margins": 0.0,
|
| 36 |
+
"rewards/rejected": 0.0,
|
| 37 |
+
"step": 2
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.0,
|
| 41 |
+
"learning_rate": 2.9347826086956523e-06,
|
| 42 |
+
"logits/chosen": -2.125047206878662,
|
| 43 |
+
"logits/rejected": -2.445204019546509,
|
| 44 |
+
"logps/chosen": -484.7939758300781,
|
| 45 |
+
"logps/rejected": -125.93072509765625,
|
| 46 |
+
"loss": 1.7268,
|
| 47 |
+
"rewards/accuracies": 0.4375,
|
| 48 |
+
"rewards/chosen": -1.0339633226394653,
|
| 49 |
+
"rewards/margins": -0.9356753826141357,
|
| 50 |
+
"rewards/rejected": -0.09828801453113556,
|
| 51 |
+
"step": 3
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.0,
|
| 55 |
+
"learning_rate": 3.91304347826087e-06,
|
| 56 |
+
"logits/chosen": -1.7763867378234863,
|
| 57 |
+
"logits/rejected": -2.1961894035339355,
|
| 58 |
+
"logps/chosen": -386.8011779785156,
|
| 59 |
+
"logps/rejected": -106.04309844970703,
|
| 60 |
+
"loss": 2.3806,
|
| 61 |
+
"rewards/accuracies": 0.5625,
|
| 62 |
+
"rewards/chosen": -0.9238373041152954,
|
| 63 |
+
"rewards/margins": -1.2902063131332397,
|
| 64 |
+
"rewards/rejected": 0.36636874079704285,
|
| 65 |
+
"step": 4
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.01,
|
| 69 |
+
"learning_rate": 4.891304347826087e-06,
|
| 70 |
+
"logits/chosen": -1.7257812023162842,
|
| 71 |
+
"logits/rejected": -2.353095769882202,
|
| 72 |
+
"logps/chosen": -524.7261962890625,
|
| 73 |
+
"logps/rejected": -70.26946258544922,
|
| 74 |
+
"loss": 0.7488,
|
| 75 |
+
"rewards/accuracies": 0.625,
|
| 76 |
+
"rewards/chosen": 2.197690010070801,
|
| 77 |
+
"rewards/margins": 2.0008764266967773,
|
| 78 |
+
"rewards/rejected": 0.19681359827518463,
|
| 79 |
+
"step": 5
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.01,
|
| 83 |
+
"learning_rate": 5.869565217391305e-06,
|
| 84 |
+
"logits/chosen": -2.1483359336853027,
|
| 85 |
+
"logits/rejected": -2.4817819595336914,
|
| 86 |
+
"logps/chosen": -283.6588134765625,
|
| 87 |
+
"logps/rejected": -58.25455093383789,
|
| 88 |
+
"loss": 0.8712,
|
| 89 |
+
"rewards/accuracies": 0.5625,
|
| 90 |
+
"rewards/chosen": 0.4915351867675781,
|
| 91 |
+
"rewards/margins": 0.2005542814731598,
|
| 92 |
+
"rewards/rejected": 0.2909809350967407,
|
| 93 |
+
"step": 6
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.01,
|
| 97 |
+
"learning_rate": 6.847826086956523e-06,
|
| 98 |
+
"logits/chosen": -2.1019060611724854,
|
| 99 |
+
"logits/rejected": -2.300177812576294,
|
| 100 |
+
"logps/chosen": -297.35235595703125,
|
| 101 |
+
"logps/rejected": -175.5060272216797,
|
| 102 |
+
"loss": 1.1392,
|
| 103 |
+
"rewards/accuracies": 0.5625,
|
| 104 |
+
"rewards/chosen": 0.6018204689025879,
|
| 105 |
+
"rewards/margins": -0.036775827407836914,
|
| 106 |
+
"rewards/rejected": 0.6385962963104248,
|
| 107 |
+
"step": 7
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.01,
|
| 111 |
+
"learning_rate": 7.82608695652174e-06,
|
| 112 |
+
"logits/chosen": -2.0169122219085693,
|
| 113 |
+
"logits/rejected": -2.201702117919922,
|
| 114 |
+
"logps/chosen": -342.04852294921875,
|
| 115 |
+
"logps/rejected": -117.9761962890625,
|
| 116 |
+
"loss": 1.1799,
|
| 117 |
+
"rewards/accuracies": 0.5625,
|
| 118 |
+
"rewards/chosen": 1.111680507659912,
|
| 119 |
+
"rewards/margins": 0.17297500371932983,
|
| 120 |
+
"rewards/rejected": 0.9387054443359375,
|
| 121 |
+
"step": 8
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.01,
|
| 125 |
+
"learning_rate": 8.804347826086957e-06,
|
| 126 |
+
"logits/chosen": -1.6254596710205078,
|
| 127 |
+
"logits/rejected": -2.0724740028381348,
|
| 128 |
+
"logps/chosen": -480.1983642578125,
|
| 129 |
+
"logps/rejected": -108.05937194824219,
|
| 130 |
+
"loss": 0.5203,
|
| 131 |
+
"rewards/accuracies": 0.8125,
|
| 132 |
+
"rewards/chosen": 2.161550283432007,
|
| 133 |
+
"rewards/margins": 1.3515225648880005,
|
| 134 |
+
"rewards/rejected": 0.8100277185440063,
|
| 135 |
+
"step": 9
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.01,
|
| 139 |
+
"learning_rate": 9.782608695652175e-06,
|
| 140 |
+
"logits/chosen": -1.4482485055923462,
|
| 141 |
+
"logits/rejected": -2.1116294860839844,
|
| 142 |
+
"logps/chosen": -451.5759582519531,
|
| 143 |
+
"logps/rejected": -67.73944854736328,
|
| 144 |
+
"loss": 0.355,
|
| 145 |
+
"rewards/accuracies": 0.8125,
|
| 146 |
+
"rewards/chosen": 6.18800163269043,
|
| 147 |
+
"rewards/margins": 5.184471130371094,
|
| 148 |
+
"rewards/rejected": 1.0035302639007568,
|
| 149 |
+
"step": 10
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.01,
|
| 153 |
+
"learning_rate": 1.0760869565217392e-05,
|
| 154 |
+
"logits/chosen": -1.2660062313079834,
|
| 155 |
+
"logits/rejected": -1.8649226427078247,
|
| 156 |
+
"logps/chosen": -265.8935546875,
|
| 157 |
+
"logps/rejected": -57.866336822509766,
|
| 158 |
+
"loss": 0.3178,
|
| 159 |
+
"rewards/accuracies": 0.8125,
|
| 160 |
+
"rewards/chosen": 5.78374719619751,
|
| 161 |
+
"rewards/margins": 4.856299877166748,
|
| 162 |
+
"rewards/rejected": 0.9274475574493408,
|
| 163 |
+
"step": 11
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.01,
|
| 167 |
+
"learning_rate": 1.173913043478261e-05,
|
| 168 |
+
"logits/chosen": -1.1804625988006592,
|
| 169 |
+
"logits/rejected": -1.7009960412979126,
|
| 170 |
+
"logps/chosen": -247.2069854736328,
|
| 171 |
+
"logps/rejected": -79.96739196777344,
|
| 172 |
+
"loss": 0.6617,
|
| 173 |
+
"rewards/accuracies": 0.75,
|
| 174 |
+
"rewards/chosen": 6.5535197257995605,
|
| 175 |
+
"rewards/margins": 4.647799015045166,
|
| 176 |
+
"rewards/rejected": 1.9057204723358154,
|
| 177 |
+
"step": 12
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.01,
|
| 181 |
+
"learning_rate": 1.2717391304347827e-05,
|
| 182 |
+
"logits/chosen": -0.9693112373352051,
|
| 183 |
+
"logits/rejected": -1.6497248411178589,
|
| 184 |
+
"logps/chosen": -372.8963623046875,
|
| 185 |
+
"logps/rejected": -70.47633361816406,
|
| 186 |
+
"loss": 0.2498,
|
| 187 |
+
"rewards/accuracies": 0.8125,
|
| 188 |
+
"rewards/chosen": 9.02272891998291,
|
| 189 |
+
"rewards/margins": 7.325143814086914,
|
| 190 |
+
"rewards/rejected": 1.6975862979888916,
|
| 191 |
+
"step": 13
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.02,
|
| 195 |
+
"learning_rate": 1.3695652173913046e-05,
|
| 196 |
+
"logits/chosen": -0.7723280787467957,
|
| 197 |
+
"logits/rejected": -1.4176554679870605,
|
| 198 |
+
"logps/chosen": -318.0392150878906,
|
| 199 |
+
"logps/rejected": -79.94412994384766,
|
| 200 |
+
"loss": 0.6401,
|
| 201 |
+
"rewards/accuracies": 0.6875,
|
| 202 |
+
"rewards/chosen": 6.698611259460449,
|
| 203 |
+
"rewards/margins": 4.517416477203369,
|
| 204 |
+
"rewards/rejected": 2.18119478225708,
|
| 205 |
+
"step": 14
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.02,
|
| 209 |
+
"learning_rate": 1.4673913043478261e-05,
|
| 210 |
+
"logits/chosen": -0.8084051609039307,
|
| 211 |
+
"logits/rejected": -1.4428132772445679,
|
| 212 |
+
"logps/chosen": -363.15869140625,
|
| 213 |
+
"logps/rejected": -70.70079803466797,
|
| 214 |
+
"loss": 0.4963,
|
| 215 |
+
"rewards/accuracies": 0.75,
|
| 216 |
+
"rewards/chosen": 7.02784538269043,
|
| 217 |
+
"rewards/margins": 5.400259971618652,
|
| 218 |
+
"rewards/rejected": 1.6275854110717773,
|
| 219 |
+
"step": 15
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.02,
|
| 223 |
+
"learning_rate": 1.565217391304348e-05,
|
| 224 |
+
"logits/chosen": -0.8056433200836182,
|
| 225 |
+
"logits/rejected": -1.4629294872283936,
|
| 226 |
+
"logps/chosen": -328.90435791015625,
|
| 227 |
+
"logps/rejected": -77.36820983886719,
|
| 228 |
+
"loss": 0.544,
|
| 229 |
+
"rewards/accuracies": 0.75,
|
| 230 |
+
"rewards/chosen": 9.417024612426758,
|
| 231 |
+
"rewards/margins": 7.023505687713623,
|
| 232 |
+
"rewards/rejected": 2.393519163131714,
|
| 233 |
+
"step": 16
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.02,
|
| 237 |
+
"learning_rate": 1.6630434782608694e-05,
|
| 238 |
+
"logits/chosen": -1.0918309688568115,
|
| 239 |
+
"logits/rejected": -1.6613047122955322,
|
| 240 |
+
"logps/chosen": -277.37982177734375,
|
| 241 |
+
"logps/rejected": -56.75635528564453,
|
| 242 |
+
"loss": 0.8958,
|
| 243 |
+
"rewards/accuracies": 0.8125,
|
| 244 |
+
"rewards/chosen": 6.844330310821533,
|
| 245 |
+
"rewards/margins": 4.82295560836792,
|
| 246 |
+
"rewards/rejected": 2.0213749408721924,
|
| 247 |
+
"step": 17
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.02,
|
| 251 |
+
"learning_rate": 1.7608695652173915e-05,
|
| 252 |
+
"logits/chosen": -0.9810999631881714,
|
| 253 |
+
"logits/rejected": -1.975247859954834,
|
| 254 |
+
"logps/chosen": -465.6856689453125,
|
| 255 |
+
"logps/rejected": -56.440269470214844,
|
| 256 |
+
"loss": 0.2275,
|
| 257 |
+
"rewards/accuracies": 0.875,
|
| 258 |
+
"rewards/chosen": 7.767542362213135,
|
| 259 |
+
"rewards/margins": 6.572459697723389,
|
| 260 |
+
"rewards/rejected": 1.1950825452804565,
|
| 261 |
+
"step": 18
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.02,
|
| 265 |
+
"learning_rate": 1.8586956521739132e-05,
|
| 266 |
+
"logits/chosen": -1.2428462505340576,
|
| 267 |
+
"logits/rejected": -1.8064367771148682,
|
| 268 |
+
"logps/chosen": -322.7100830078125,
|
| 269 |
+
"logps/rejected": -108.71321105957031,
|
| 270 |
+
"loss": 0.3006,
|
| 271 |
+
"rewards/accuracies": 0.875,
|
| 272 |
+
"rewards/chosen": 6.459216594696045,
|
| 273 |
+
"rewards/margins": 3.9207711219787598,
|
| 274 |
+
"rewards/rejected": 2.5384464263916016,
|
| 275 |
+
"step": 19
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.02,
|
| 279 |
+
"learning_rate": 1.956521739130435e-05,
|
| 280 |
+
"logits/chosen": -1.7773098945617676,
|
| 281 |
+
"logits/rejected": -2.178387403488159,
|
| 282 |
+
"logps/chosen": -357.8656921386719,
|
| 283 |
+
"logps/rejected": -112.65576934814453,
|
| 284 |
+
"loss": 0.3837,
|
| 285 |
+
"rewards/accuracies": 0.9375,
|
| 286 |
+
"rewards/chosen": 4.684590816497803,
|
| 287 |
+
"rewards/margins": 4.16137170791626,
|
| 288 |
+
"rewards/rejected": 0.5232191681861877,
|
| 289 |
+
"step": 20
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.02,
|
| 293 |
+
"learning_rate": 2.0543478260869567e-05,
|
| 294 |
+
"logits/chosen": -1.6802027225494385,
|
| 295 |
+
"logits/rejected": -1.943274974822998,
|
| 296 |
+
"logps/chosen": -327.85693359375,
|
| 297 |
+
"logps/rejected": -87.53279113769531,
|
| 298 |
+
"loss": 0.4428,
|
| 299 |
+
"rewards/accuracies": 0.9375,
|
| 300 |
+
"rewards/chosen": 2.681103229522705,
|
| 301 |
+
"rewards/margins": 2.3518502712249756,
|
| 302 |
+
"rewards/rejected": 0.32925307750701904,
|
| 303 |
+
"step": 21
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.02,
|
| 307 |
+
"learning_rate": 2.1521739130434784e-05,
|
| 308 |
+
"logits/chosen": -1.5789223909378052,
|
| 309 |
+
"logits/rejected": -2.4593281745910645,
|
| 310 |
+
"logps/chosen": -590.5732421875,
|
| 311 |
+
"logps/rejected": -86.78819274902344,
|
| 312 |
+
"loss": 0.252,
|
| 313 |
+
"rewards/accuracies": 0.875,
|
| 314 |
+
"rewards/chosen": 3.6503655910491943,
|
| 315 |
+
"rewards/margins": 3.621542453765869,
|
| 316 |
+
"rewards/rejected": 0.028822531923651695,
|
| 317 |
+
"step": 22
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.03,
|
| 321 |
+
"learning_rate": 2.25e-05,
|
| 322 |
+
"logits/chosen": -1.8142305612564087,
|
| 323 |
+
"logits/rejected": -2.3507237434387207,
|
| 324 |
+
"logps/chosen": -296.42266845703125,
|
| 325 |
+
"logps/rejected": -60.37131118774414,
|
| 326 |
+
"loss": 0.5602,
|
| 327 |
+
"rewards/accuracies": 0.75,
|
| 328 |
+
"rewards/chosen": 2.170171022415161,
|
| 329 |
+
"rewards/margins": 2.5353941917419434,
|
| 330 |
+
"rewards/rejected": -0.36522313952445984,
|
| 331 |
+
"step": 23
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.03,
|
| 335 |
+
"learning_rate": 2.347826086956522e-05,
|
| 336 |
+
"logits/chosen": -1.7050156593322754,
|
| 337 |
+
"logits/rejected": -2.158708333969116,
|
| 338 |
+
"logps/chosen": -523.41748046875,
|
| 339 |
+
"logps/rejected": -81.6954345703125,
|
| 340 |
+
"loss": 0.7359,
|
| 341 |
+
"rewards/accuracies": 0.9375,
|
| 342 |
+
"rewards/chosen": 2.3799068927764893,
|
| 343 |
+
"rewards/margins": 2.963341236114502,
|
| 344 |
+
"rewards/rejected": -0.583433985710144,
|
| 345 |
+
"step": 24
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.03,
|
| 349 |
+
"learning_rate": 2.4456521739130436e-05,
|
| 350 |
+
"logits/chosen": -1.704190731048584,
|
| 351 |
+
"logits/rejected": -2.215942859649658,
|
| 352 |
+
"logps/chosen": -457.6519775390625,
|
| 353 |
+
"logps/rejected": -91.778076171875,
|
| 354 |
+
"loss": 0.0865,
|
| 355 |
+
"rewards/accuracies": 1.0,
|
| 356 |
+
"rewards/chosen": 4.57167911529541,
|
| 357 |
+
"rewards/margins": 4.6597795486450195,
|
| 358 |
+
"rewards/rejected": -0.08810039609670639,
|
| 359 |
+
"step": 25
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.03,
|
| 363 |
+
"learning_rate": 2.5434782608695653e-05,
|
| 364 |
+
"logits/chosen": -1.4195502996444702,
|
| 365 |
+
"logits/rejected": -2.109910488128662,
|
| 366 |
+
"logps/chosen": -445.05908203125,
|
| 367 |
+
"logps/rejected": -80.95398712158203,
|
| 368 |
+
"loss": 0.1265,
|
| 369 |
+
"rewards/accuracies": 0.9375,
|
| 370 |
+
"rewards/chosen": 5.362960338592529,
|
| 371 |
+
"rewards/margins": 5.412578582763672,
|
| 372 |
+
"rewards/rejected": -0.049618080258369446,
|
| 373 |
+
"step": 26
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.03,
|
| 377 |
+
"learning_rate": 2.6413043478260874e-05,
|
| 378 |
+
"logits/chosen": -1.0274971723556519,
|
| 379 |
+
"logits/rejected": -1.7681918144226074,
|
| 380 |
+
"logps/chosen": -393.85870361328125,
|
| 381 |
+
"logps/rejected": -77.79011535644531,
|
| 382 |
+
"loss": 0.1702,
|
| 383 |
+
"rewards/accuracies": 1.0,
|
| 384 |
+
"rewards/chosen": 4.572484016418457,
|
| 385 |
+
"rewards/margins": 4.946743488311768,
|
| 386 |
+
"rewards/rejected": -0.374259889125824,
|
| 387 |
+
"step": 27
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.03,
|
| 391 |
+
"learning_rate": 2.739130434782609e-05,
|
| 392 |
+
"logits/chosen": -0.7365273237228394,
|
| 393 |
+
"logits/rejected": -1.9988534450531006,
|
| 394 |
+
"logps/chosen": -569.8936767578125,
|
| 395 |
+
"logps/rejected": -79.71562194824219,
|
| 396 |
+
"loss": 0.0465,
|
| 397 |
+
"rewards/accuracies": 1.0,
|
| 398 |
+
"rewards/chosen": 12.751686096191406,
|
| 399 |
+
"rewards/margins": 12.474647521972656,
|
| 400 |
+
"rewards/rejected": 0.2770393192768097,
|
| 401 |
+
"step": 28
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.03,
|
| 405 |
+
"learning_rate": 2.836956521739131e-05,
|
| 406 |
+
"logits/chosen": -0.6424872279167175,
|
| 407 |
+
"logits/rejected": -1.6069284677505493,
|
| 408 |
+
"logps/chosen": -368.68292236328125,
|
| 409 |
+
"logps/rejected": -103.9141845703125,
|
| 410 |
+
"loss": 0.0351,
|
| 411 |
+
"rewards/accuracies": 1.0,
|
| 412 |
+
"rewards/chosen": 10.144453048706055,
|
| 413 |
+
"rewards/margins": 10.381340980529785,
|
| 414 |
+
"rewards/rejected": -0.23688830435276031,
|
| 415 |
+
"step": 29
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.03,
|
| 419 |
+
"learning_rate": 2.9347826086956523e-05,
|
| 420 |
+
"logits/chosen": -0.6061868071556091,
|
| 421 |
+
"logits/rejected": -1.4256294965744019,
|
| 422 |
+
"logps/chosen": -360.4029846191406,
|
| 423 |
+
"logps/rejected": -130.27224731445312,
|
| 424 |
+
"loss": 0.0414,
|
| 425 |
+
"rewards/accuracies": 1.0,
|
| 426 |
+
"rewards/chosen": 12.649539947509766,
|
| 427 |
+
"rewards/margins": 12.36604118347168,
|
| 428 |
+
"rewards/rejected": 0.2834990620613098,
|
| 429 |
+
"step": 30
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.03,
|
| 433 |
+
"learning_rate": 3.032608695652174e-05,
|
| 434 |
+
"logits/chosen": -1.0075111389160156,
|
| 435 |
+
"logits/rejected": -1.3918204307556152,
|
| 436 |
+
"logps/chosen": -227.05450439453125,
|
| 437 |
+
"logps/rejected": -172.67947387695312,
|
| 438 |
+
"loss": 1.1145,
|
| 439 |
+
"rewards/accuracies": 0.875,
|
| 440 |
+
"rewards/chosen": 4.628115653991699,
|
| 441 |
+
"rewards/margins": 3.9521055221557617,
|
| 442 |
+
"rewards/rejected": 0.6760098934173584,
|
| 443 |
+
"step": 31
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.03,
|
| 447 |
+
"learning_rate": 3.130434782608696e-05,
|
| 448 |
+
"logits/chosen": -0.9060839414596558,
|
| 449 |
+
"logits/rejected": -2.0123891830444336,
|
| 450 |
+
"logps/chosen": -403.06781005859375,
|
| 451 |
+
"logps/rejected": -75.35684967041016,
|
| 452 |
+
"loss": 0.0319,
|
| 453 |
+
"rewards/accuracies": 1.0,
|
| 454 |
+
"rewards/chosen": 7.352475643157959,
|
| 455 |
+
"rewards/margins": 8.778717041015625,
|
| 456 |
+
"rewards/rejected": -1.4262410402297974,
|
| 457 |
+
"step": 32
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.04,
|
| 461 |
+
"learning_rate": 3.228260869565217e-05,
|
| 462 |
+
"logits/chosen": -1.2984271049499512,
|
| 463 |
+
"logits/rejected": -1.9507720470428467,
|
| 464 |
+
"logps/chosen": -306.7958984375,
|
| 465 |
+
"logps/rejected": -107.01404571533203,
|
| 466 |
+
"loss": 0.0795,
|
| 467 |
+
"rewards/accuracies": 0.9375,
|
| 468 |
+
"rewards/chosen": 5.223363399505615,
|
| 469 |
+
"rewards/margins": 7.012631416320801,
|
| 470 |
+
"rewards/rejected": -1.7892680168151855,
|
| 471 |
+
"step": 33
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.04,
|
| 475 |
+
"learning_rate": 3.326086956521739e-05,
|
| 476 |
+
"logits/chosen": -1.1749062538146973,
|
| 477 |
+
"logits/rejected": -2.004805564880371,
|
| 478 |
+
"logps/chosen": -437.9205322265625,
|
| 479 |
+
"logps/rejected": -135.21376037597656,
|
| 480 |
+
"loss": 0.0021,
|
| 481 |
+
"rewards/accuracies": 1.0,
|
| 482 |
+
"rewards/chosen": 9.3555908203125,
|
| 483 |
+
"rewards/margins": 12.285521507263184,
|
| 484 |
+
"rewards/rejected": -2.9299299716949463,
|
| 485 |
+
"step": 34
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.04,
|
| 489 |
+
"learning_rate": 3.423913043478261e-05,
|
| 490 |
+
"logits/chosen": -0.9383536577224731,
|
| 491 |
+
"logits/rejected": -2.0366413593292236,
|
| 492 |
+
"logps/chosen": -457.8018493652344,
|
| 493 |
+
"logps/rejected": -87.7723617553711,
|
| 494 |
+
"loss": 0.041,
|
| 495 |
+
"rewards/accuracies": 1.0,
|
| 496 |
+
"rewards/chosen": 6.836911678314209,
|
| 497 |
+
"rewards/margins": 9.43667984008789,
|
| 498 |
+
"rewards/rejected": -2.599769115447998,
|
| 499 |
+
"step": 35
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.04,
|
| 503 |
+
"learning_rate": 3.521739130434783e-05,
|
| 504 |
+
"logits/chosen": -1.120687484741211,
|
| 505 |
+
"logits/rejected": -1.9933563470840454,
|
| 506 |
+
"logps/chosen": -439.6186828613281,
|
| 507 |
+
"logps/rejected": -94.24763488769531,
|
| 508 |
+
"loss": 0.1906,
|
| 509 |
+
"rewards/accuracies": 0.9375,
|
| 510 |
+
"rewards/chosen": 6.565365314483643,
|
| 511 |
+
"rewards/margins": 9.163484573364258,
|
| 512 |
+
"rewards/rejected": -2.5981192588806152,
|
| 513 |
+
"step": 36
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.04,
|
| 517 |
+
"learning_rate": 3.619565217391305e-05,
|
| 518 |
+
"logits/chosen": -1.338189959526062,
|
| 519 |
+
"logits/rejected": -1.9118335247039795,
|
| 520 |
+
"logps/chosen": -286.96728515625,
|
| 521 |
+
"logps/rejected": -127.22909545898438,
|
| 522 |
+
"loss": 0.0279,
|
| 523 |
+
"rewards/accuracies": 1.0,
|
| 524 |
+
"rewards/chosen": 2.9341468811035156,
|
| 525 |
+
"rewards/margins": 6.362301349639893,
|
| 526 |
+
"rewards/rejected": -3.4281537532806396,
|
| 527 |
+
"step": 37
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.04,
|
| 531 |
+
"learning_rate": 3.7173913043478264e-05,
|
| 532 |
+
"logits/chosen": -1.2371840476989746,
|
| 533 |
+
"logits/rejected": -1.7362310886383057,
|
| 534 |
+
"logps/chosen": -319.52911376953125,
|
| 535 |
+
"logps/rejected": -151.65084838867188,
|
| 536 |
+
"loss": 0.0939,
|
| 537 |
+
"rewards/accuracies": 0.9375,
|
| 538 |
+
"rewards/chosen": 6.216956615447998,
|
| 539 |
+
"rewards/margins": 9.073128700256348,
|
| 540 |
+
"rewards/rejected": -2.856172561645508,
|
| 541 |
+
"step": 38
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.04,
|
| 545 |
+
"learning_rate": 3.815217391304348e-05,
|
| 546 |
+
"logits/chosen": -0.8356418609619141,
|
| 547 |
+
"logits/rejected": -1.7252635955810547,
|
| 548 |
+
"logps/chosen": -391.63275146484375,
|
| 549 |
+
"logps/rejected": -154.64361572265625,
|
| 550 |
+
"loss": 0.2483,
|
| 551 |
+
"rewards/accuracies": 0.9375,
|
| 552 |
+
"rewards/chosen": 8.802470207214355,
|
| 553 |
+
"rewards/margins": 11.676048278808594,
|
| 554 |
+
"rewards/rejected": -2.873577356338501,
|
| 555 |
+
"step": 39
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.04,
|
| 559 |
+
"learning_rate": 3.91304347826087e-05,
|
| 560 |
+
"logits/chosen": -0.9311200380325317,
|
| 561 |
+
"logits/rejected": -1.8107151985168457,
|
| 562 |
+
"logps/chosen": -374.2525939941406,
|
| 563 |
+
"logps/rejected": -131.86715698242188,
|
| 564 |
+
"loss": 0.0058,
|
| 565 |
+
"rewards/accuracies": 1.0,
|
| 566 |
+
"rewards/chosen": 8.586356163024902,
|
| 567 |
+
"rewards/margins": 11.15583324432373,
|
| 568 |
+
"rewards/rejected": -2.569479465484619,
|
| 569 |
+
"step": 40
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.04,
|
| 573 |
+
"learning_rate": 4.0108695652173916e-05,
|
| 574 |
+
"logits/chosen": -0.6861732006072998,
|
| 575 |
+
"logits/rejected": -1.6295411586761475,
|
| 576 |
+
"logps/chosen": -311.7944641113281,
|
| 577 |
+
"logps/rejected": -132.4564666748047,
|
| 578 |
+
"loss": 0.0124,
|
| 579 |
+
"rewards/accuracies": 1.0,
|
| 580 |
+
"rewards/chosen": 5.333560943603516,
|
| 581 |
+
"rewards/margins": 10.023443222045898,
|
| 582 |
+
"rewards/rejected": -4.689882755279541,
|
| 583 |
+
"step": 41
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.05,
|
| 587 |
+
"learning_rate": 4.1086956521739134e-05,
|
| 588 |
+
"logits/chosen": -1.1133558750152588,
|
| 589 |
+
"logits/rejected": -2.0819039344787598,
|
| 590 |
+
"logps/chosen": -282.22332763671875,
|
| 591 |
+
"logps/rejected": -98.7315673828125,
|
| 592 |
+
"loss": 0.0156,
|
| 593 |
+
"rewards/accuracies": 1.0,
|
| 594 |
+
"rewards/chosen": 4.930261611938477,
|
| 595 |
+
"rewards/margins": 8.344033241271973,
|
| 596 |
+
"rewards/rejected": -3.413771629333496,
|
| 597 |
+
"step": 42
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.05,
|
| 601 |
+
"learning_rate": 4.206521739130435e-05,
|
| 602 |
+
"logits/chosen": -1.0311042070388794,
|
| 603 |
+
"logits/rejected": -2.0758776664733887,
|
| 604 |
+
"logps/chosen": -369.7279052734375,
|
| 605 |
+
"logps/rejected": -146.08364868164062,
|
| 606 |
+
"loss": 0.0127,
|
| 607 |
+
"rewards/accuracies": 1.0,
|
| 608 |
+
"rewards/chosen": 7.7552289962768555,
|
| 609 |
+
"rewards/margins": 13.249262809753418,
|
| 610 |
+
"rewards/rejected": -5.494032859802246,
|
| 611 |
+
"step": 43
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.05,
|
| 615 |
+
"learning_rate": 4.304347826086957e-05,
|
| 616 |
+
"logits/chosen": -1.0191632509231567,
|
| 617 |
+
"logits/rejected": -1.995982050895691,
|
| 618 |
+
"logps/chosen": -367.692626953125,
|
| 619 |
+
"logps/rejected": -152.65135192871094,
|
| 620 |
+
"loss": 0.0004,
|
| 621 |
+
"rewards/accuracies": 1.0,
|
| 622 |
+
"rewards/chosen": 7.9517822265625,
|
| 623 |
+
"rewards/margins": 13.869260787963867,
|
| 624 |
+
"rewards/rejected": -5.917478084564209,
|
| 625 |
+
"step": 44
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.05,
|
| 629 |
+
"learning_rate": 4.4021739130434786e-05,
|
| 630 |
+
"logits/chosen": -1.3430471420288086,
|
| 631 |
+
"logits/rejected": -2.0910849571228027,
|
| 632 |
+
"logps/chosen": -334.8968811035156,
|
| 633 |
+
"logps/rejected": -168.4519500732422,
|
| 634 |
+
"loss": 0.0057,
|
| 635 |
+
"rewards/accuracies": 1.0,
|
| 636 |
+
"rewards/chosen": 5.113959312438965,
|
| 637 |
+
"rewards/margins": 12.524876594543457,
|
| 638 |
+
"rewards/rejected": -7.41091775894165,
|
| 639 |
+
"step": 45
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.05,
|
| 643 |
+
"learning_rate": 4.5e-05,
|
| 644 |
+
"logits/chosen": -0.8516018986701965,
|
| 645 |
+
"logits/rejected": -1.9934579133987427,
|
| 646 |
+
"logps/chosen": -386.7164001464844,
|
| 647 |
+
"logps/rejected": -152.8460235595703,
|
| 648 |
+
"loss": 0.0253,
|
| 649 |
+
"rewards/accuracies": 1.0,
|
| 650 |
+
"rewards/chosen": 7.0803022384643555,
|
| 651 |
+
"rewards/margins": 13.202106475830078,
|
| 652 |
+
"rewards/rejected": -6.121803283691406,
|
| 653 |
+
"step": 46
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.05,
|
| 657 |
+
"learning_rate": 4.494827586206897e-05,
|
| 658 |
+
"logits/chosen": -0.9460695385932922,
|
| 659 |
+
"logits/rejected": -1.9993284940719604,
|
| 660 |
+
"logps/chosen": -355.2873229980469,
|
| 661 |
+
"logps/rejected": -188.5989532470703,
|
| 662 |
+
"loss": 0.1008,
|
| 663 |
+
"rewards/accuracies": 0.9375,
|
| 664 |
+
"rewards/chosen": 6.138028144836426,
|
| 665 |
+
"rewards/margins": 13.93982982635498,
|
| 666 |
+
"rewards/rejected": -7.801800727844238,
|
| 667 |
+
"step": 47
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.05,
|
| 671 |
+
"learning_rate": 4.489655172413793e-05,
|
| 672 |
+
"logits/chosen": -1.212510108947754,
|
| 673 |
+
"logits/rejected": -2.0202791690826416,
|
| 674 |
+
"logps/chosen": -293.0318908691406,
|
| 675 |
+
"logps/rejected": -194.92649841308594,
|
| 676 |
+
"loss": 0.0058,
|
| 677 |
+
"rewards/accuracies": 1.0,
|
| 678 |
+
"rewards/chosen": 6.742659091949463,
|
| 679 |
+
"rewards/margins": 17.209505081176758,
|
| 680 |
+
"rewards/rejected": -10.466848373413086,
|
| 681 |
+
"step": 48
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.05,
|
| 685 |
+
"learning_rate": 4.48448275862069e-05,
|
| 686 |
+
"logits/chosen": -1.388157606124878,
|
| 687 |
+
"logits/rejected": -2.1885251998901367,
|
| 688 |
+
"logps/chosen": -330.6868591308594,
|
| 689 |
+
"logps/rejected": -214.35423278808594,
|
| 690 |
+
"loss": 0.0003,
|
| 691 |
+
"rewards/accuracies": 1.0,
|
| 692 |
+
"rewards/chosen": 5.653369903564453,
|
| 693 |
+
"rewards/margins": 18.257244110107422,
|
| 694 |
+
"rewards/rejected": -12.603873252868652,
|
| 695 |
+
"step": 49
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.05,
|
| 699 |
+
"learning_rate": 4.479310344827587e-05,
|
| 700 |
+
"logits/chosen": -1.020527958869934,
|
| 701 |
+
"logits/rejected": -1.9433815479278564,
|
| 702 |
+
"logps/chosen": -347.48394775390625,
|
| 703 |
+
"logps/rejected": -191.15147399902344,
|
| 704 |
+
"loss": 0.0001,
|
| 705 |
+
"rewards/accuracies": 1.0,
|
| 706 |
+
"rewards/chosen": 7.30579137802124,
|
| 707 |
+
"rewards/margins": 18.734811782836914,
|
| 708 |
+
"rewards/rejected": -11.429021835327148,
|
| 709 |
+
"step": 50
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.06,
|
| 713 |
+
"learning_rate": 4.474137931034483e-05,
|
| 714 |
+
"logits/chosen": -1.2260757684707642,
|
| 715 |
+
"logits/rejected": -2.0841352939605713,
|
| 716 |
+
"logps/chosen": -243.5459747314453,
|
| 717 |
+
"logps/rejected": -188.00442504882812,
|
| 718 |
+
"loss": 0.0002,
|
| 719 |
+
"rewards/accuracies": 1.0,
|
| 720 |
+
"rewards/chosen": 3.675260543823242,
|
| 721 |
+
"rewards/margins": 14.774760246276855,
|
| 722 |
+
"rewards/rejected": -11.099499702453613,
|
| 723 |
+
"step": 51
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.06,
|
| 727 |
+
"learning_rate": 4.46896551724138e-05,
|
| 728 |
+
"logits/chosen": -1.0378711223602295,
|
| 729 |
+
"logits/rejected": -2.0174639225006104,
|
| 730 |
+
"logps/chosen": -375.9520263671875,
|
| 731 |
+
"logps/rejected": -182.97463989257812,
|
| 732 |
+
"loss": 0.0,
|
| 733 |
+
"rewards/accuracies": 1.0,
|
| 734 |
+
"rewards/chosen": 7.18237829208374,
|
| 735 |
+
"rewards/margins": 19.010208129882812,
|
| 736 |
+
"rewards/rejected": -11.82783317565918,
|
| 737 |
+
"step": 52
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.06,
|
| 741 |
+
"learning_rate": 4.4637931034482765e-05,
|
| 742 |
+
"logits/chosen": -1.1728301048278809,
|
| 743 |
+
"logits/rejected": -1.9667140245437622,
|
| 744 |
+
"logps/chosen": -309.5382995605469,
|
| 745 |
+
"logps/rejected": -241.13743591308594,
|
| 746 |
+
"loss": 0.0,
|
| 747 |
+
"rewards/accuracies": 1.0,
|
| 748 |
+
"rewards/chosen": 6.681103706359863,
|
| 749 |
+
"rewards/margins": 21.542236328125,
|
| 750 |
+
"rewards/rejected": -14.861133575439453,
|
| 751 |
+
"step": 53
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.06,
|
| 755 |
+
"learning_rate": 4.4586206896551726e-05,
|
| 756 |
+
"logits/chosen": -1.1491725444793701,
|
| 757 |
+
"logits/rejected": -1.760243535041809,
|
| 758 |
+
"logps/chosen": -250.09036254882812,
|
| 759 |
+
"logps/rejected": -226.221435546875,
|
| 760 |
+
"loss": 0.0001,
|
| 761 |
+
"rewards/accuracies": 1.0,
|
| 762 |
+
"rewards/chosen": 2.3173880577087402,
|
| 763 |
+
"rewards/margins": 16.417604446411133,
|
| 764 |
+
"rewards/rejected": -14.100215911865234,
|
| 765 |
+
"step": 54
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.06,
|
| 769 |
+
"learning_rate": 4.4534482758620694e-05,
|
| 770 |
+
"logits/chosen": -1.2637637853622437,
|
| 771 |
+
"logits/rejected": -1.8996949195861816,
|
| 772 |
+
"logps/chosen": -401.9977722167969,
|
| 773 |
+
"logps/rejected": -303.9198303222656,
|
| 774 |
+
"loss": 0.0011,
|
| 775 |
+
"rewards/accuracies": 1.0,
|
| 776 |
+
"rewards/chosen": 8.595758438110352,
|
| 777 |
+
"rewards/margins": 26.598325729370117,
|
| 778 |
+
"rewards/rejected": -18.002567291259766,
|
| 779 |
+
"step": 55
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.06,
|
| 783 |
+
"learning_rate": 4.4482758620689656e-05,
|
| 784 |
+
"logits/chosen": -0.9855178594589233,
|
| 785 |
+
"logits/rejected": -2.0190346240997314,
|
| 786 |
+
"logps/chosen": -394.40625,
|
| 787 |
+
"logps/rejected": -253.97720336914062,
|
| 788 |
+
"loss": 0.0,
|
| 789 |
+
"rewards/accuracies": 1.0,
|
| 790 |
+
"rewards/chosen": 6.395060062408447,
|
| 791 |
+
"rewards/margins": 23.680639266967773,
|
| 792 |
+
"rewards/rejected": -17.28557777404785,
|
| 793 |
+
"step": 56
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.06,
|
| 797 |
+
"learning_rate": 4.4431034482758624e-05,
|
| 798 |
+
"logits/chosen": -1.049959421157837,
|
| 799 |
+
"logits/rejected": -1.6531357765197754,
|
| 800 |
+
"logps/chosen": -205.54058837890625,
|
| 801 |
+
"logps/rejected": -250.66958618164062,
|
| 802 |
+
"loss": 0.0,
|
| 803 |
+
"rewards/accuracies": 1.0,
|
| 804 |
+
"rewards/chosen": 4.874389171600342,
|
| 805 |
+
"rewards/margins": 20.083953857421875,
|
| 806 |
+
"rewards/rejected": -15.209564208984375,
|
| 807 |
+
"step": 57
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.06,
|
| 811 |
+
"learning_rate": 4.4379310344827585e-05,
|
| 812 |
+
"logits/chosen": -0.8982111811637878,
|
| 813 |
+
"logits/rejected": -1.6001371145248413,
|
| 814 |
+
"logps/chosen": -241.91790771484375,
|
| 815 |
+
"logps/rejected": -252.10299682617188,
|
| 816 |
+
"loss": 0.0,
|
| 817 |
+
"rewards/accuracies": 1.0,
|
| 818 |
+
"rewards/chosen": 6.044814109802246,
|
| 819 |
+
"rewards/margins": 21.77715301513672,
|
| 820 |
+
"rewards/rejected": -15.732340812683105,
|
| 821 |
+
"step": 58
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.06,
|
| 825 |
+
"learning_rate": 4.432758620689655e-05,
|
| 826 |
+
"logits/chosen": -0.753044068813324,
|
| 827 |
+
"logits/rejected": -1.6325318813323975,
|
| 828 |
+
"logps/chosen": -315.30230712890625,
|
| 829 |
+
"logps/rejected": -245.91976928710938,
|
| 830 |
+
"loss": 0.0078,
|
| 831 |
+
"rewards/accuracies": 1.0,
|
| 832 |
+
"rewards/chosen": 5.801758289337158,
|
| 833 |
+
"rewards/margins": 20.689924240112305,
|
| 834 |
+
"rewards/rejected": -14.888164520263672,
|
| 835 |
+
"step": 59
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.07,
|
| 839 |
+
"learning_rate": 4.427586206896552e-05,
|
| 840 |
+
"logits/chosen": -0.5443448424339294,
|
| 841 |
+
"logits/rejected": -1.7224345207214355,
|
| 842 |
+
"logps/chosen": -388.339599609375,
|
| 843 |
+
"logps/rejected": -211.5500946044922,
|
| 844 |
+
"loss": 0.0,
|
| 845 |
+
"rewards/accuracies": 1.0,
|
| 846 |
+
"rewards/chosen": 11.5880708694458,
|
| 847 |
+
"rewards/margins": 24.889877319335938,
|
| 848 |
+
"rewards/rejected": -13.301810264587402,
|
| 849 |
+
"step": 60
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.07,
|
| 853 |
+
"learning_rate": 4.422413793103448e-05,
|
| 854 |
+
"logits/chosen": -0.6921290755271912,
|
| 855 |
+
"logits/rejected": -1.6139135360717773,
|
| 856 |
+
"logps/chosen": -242.47312927246094,
|
| 857 |
+
"logps/rejected": -214.73622131347656,
|
| 858 |
+
"loss": 0.0,
|
| 859 |
+
"rewards/accuracies": 1.0,
|
| 860 |
+
"rewards/chosen": 11.329044342041016,
|
| 861 |
+
"rewards/margins": 21.45581817626953,
|
| 862 |
+
"rewards/rejected": -10.1267728805542,
|
| 863 |
+
"step": 61
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.07,
|
| 867 |
+
"learning_rate": 4.417241379310345e-05,
|
| 868 |
+
"logits/chosen": -0.4828271269798279,
|
| 869 |
+
"logits/rejected": -1.649349570274353,
|
| 870 |
+
"logps/chosen": -306.6004638671875,
|
| 871 |
+
"logps/rejected": -172.2523956298828,
|
| 872 |
+
"loss": 0.0003,
|
| 873 |
+
"rewards/accuracies": 1.0,
|
| 874 |
+
"rewards/chosen": 11.373737335205078,
|
| 875 |
+
"rewards/margins": 20.521434783935547,
|
| 876 |
+
"rewards/rejected": -9.147698402404785,
|
| 877 |
+
"step": 62
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.07,
|
| 881 |
+
"learning_rate": 4.412068965517242e-05,
|
| 882 |
+
"logits/chosen": -0.4676312506198883,
|
| 883 |
+
"logits/rejected": -1.6393827199935913,
|
| 884 |
+
"logps/chosen": -357.127685546875,
|
| 885 |
+
"logps/rejected": -199.84524536132812,
|
| 886 |
+
"loss": 0.0,
|
| 887 |
+
"rewards/accuracies": 1.0,
|
| 888 |
+
"rewards/chosen": 16.730457305908203,
|
| 889 |
+
"rewards/margins": 26.409488677978516,
|
| 890 |
+
"rewards/rejected": -9.679031372070312,
|
| 891 |
+
"step": 63
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.07,
|
| 895 |
+
"learning_rate": 4.406896551724138e-05,
|
| 896 |
+
"logits/chosen": -0.7774850726127625,
|
| 897 |
+
"logits/rejected": -1.324657678604126,
|
| 898 |
+
"logps/chosen": -193.75833129882812,
|
| 899 |
+
"logps/rejected": -194.70233154296875,
|
| 900 |
+
"loss": 0.0015,
|
| 901 |
+
"rewards/accuracies": 1.0,
|
| 902 |
+
"rewards/chosen": 5.371121406555176,
|
| 903 |
+
"rewards/margins": 15.129293441772461,
|
| 904 |
+
"rewards/rejected": -9.758172988891602,
|
| 905 |
+
"step": 64
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.07,
|
| 909 |
+
"learning_rate": 4.401724137931035e-05,
|
| 910 |
+
"logits/chosen": -0.5502160787582397,
|
| 911 |
+
"logits/rejected": -1.466262936592102,
|
| 912 |
+
"logps/chosen": -240.96836853027344,
|
| 913 |
+
"logps/rejected": -174.6772003173828,
|
| 914 |
+
"loss": 0.0,
|
| 915 |
+
"rewards/accuracies": 1.0,
|
| 916 |
+
"rewards/chosen": 7.808657169342041,
|
| 917 |
+
"rewards/margins": 17.520769119262695,
|
| 918 |
+
"rewards/rejected": -9.712109565734863,
|
| 919 |
+
"step": 65
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.07,
|
| 923 |
+
"learning_rate": 4.3965517241379315e-05,
|
| 924 |
+
"logits/chosen": -0.5615445375442505,
|
| 925 |
+
"logits/rejected": -1.3243337869644165,
|
| 926 |
+
"logps/chosen": -282.65484619140625,
|
| 927 |
+
"logps/rejected": -211.33326721191406,
|
| 928 |
+
"loss": 0.4403,
|
| 929 |
+
"rewards/accuracies": 0.9375,
|
| 930 |
+
"rewards/chosen": 9.213937759399414,
|
| 931 |
+
"rewards/margins": 18.93706512451172,
|
| 932 |
+
"rewards/rejected": -9.723125457763672,
|
| 933 |
+
"step": 66
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.07,
|
| 937 |
+
"learning_rate": 4.3913793103448277e-05,
|
| 938 |
+
"logits/chosen": -0.5882927775382996,
|
| 939 |
+
"logits/rejected": -1.4038889408111572,
|
| 940 |
+
"logps/chosen": -244.26333618164062,
|
| 941 |
+
"logps/rejected": -192.05517578125,
|
| 942 |
+
"loss": 0.0007,
|
| 943 |
+
"rewards/accuracies": 1.0,
|
| 944 |
+
"rewards/chosen": 5.999594688415527,
|
| 945 |
+
"rewards/margins": 16.232702255249023,
|
| 946 |
+
"rewards/rejected": -10.233107566833496,
|
| 947 |
+
"step": 67
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.07,
|
| 951 |
+
"learning_rate": 4.3862068965517245e-05,
|
| 952 |
+
"logits/chosen": -0.7710579633712769,
|
| 953 |
+
"logits/rejected": -1.5811477899551392,
|
| 954 |
+
"logps/chosen": -274.4286804199219,
|
| 955 |
+
"logps/rejected": -189.31427001953125,
|
| 956 |
+
"loss": 0.0003,
|
| 957 |
+
"rewards/accuracies": 1.0,
|
| 958 |
+
"rewards/chosen": 10.483807563781738,
|
| 959 |
+
"rewards/margins": 22.807409286499023,
|
| 960 |
+
"rewards/rejected": -12.323600769042969,
|
| 961 |
+
"step": 68
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.08,
|
| 965 |
+
"learning_rate": 4.381034482758621e-05,
|
| 966 |
+
"logits/chosen": -0.2001597136259079,
|
| 967 |
+
"logits/rejected": -1.5477687120437622,
|
| 968 |
+
"logps/chosen": -398.04296875,
|
| 969 |
+
"logps/rejected": -208.8218536376953,
|
| 970 |
+
"loss": 0.0,
|
| 971 |
+
"rewards/accuracies": 1.0,
|
| 972 |
+
"rewards/chosen": 11.122913360595703,
|
| 973 |
+
"rewards/margins": 23.094343185424805,
|
| 974 |
+
"rewards/rejected": -11.971429824829102,
|
| 975 |
+
"step": 69
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.08,
|
| 979 |
+
"learning_rate": 4.3758620689655174e-05,
|
| 980 |
+
"logits/chosen": -0.5891858339309692,
|
| 981 |
+
"logits/rejected": -1.500455379486084,
|
| 982 |
+
"logps/chosen": -223.07235717773438,
|
| 983 |
+
"logps/rejected": -219.37405395507812,
|
| 984 |
+
"loss": 0.0,
|
| 985 |
+
"rewards/accuracies": 1.0,
|
| 986 |
+
"rewards/chosen": 7.498288154602051,
|
| 987 |
+
"rewards/margins": 19.65944480895996,
|
| 988 |
+
"rewards/rejected": -12.16115665435791,
|
| 989 |
+
"step": 70
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 0.08,
|
| 993 |
+
"learning_rate": 4.370689655172414e-05,
|
| 994 |
+
"logits/chosen": -0.5222135186195374,
|
| 995 |
+
"logits/rejected": -1.459835410118103,
|
| 996 |
+
"logps/chosen": -264.3045349121094,
|
| 997 |
+
"logps/rejected": -184.4599609375,
|
| 998 |
+
"loss": 0.0006,
|
| 999 |
+
"rewards/accuracies": 1.0,
|
| 1000 |
+
"rewards/chosen": 7.261447429656982,
|
| 1001 |
+
"rewards/margins": 17.37218475341797,
|
| 1002 |
+
"rewards/rejected": -10.110737800598145,
|
| 1003 |
+
"step": 71
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 0.08,
|
| 1007 |
+
"learning_rate": 4.365517241379311e-05,
|
| 1008 |
+
"logits/chosen": -0.4974746108055115,
|
| 1009 |
+
"logits/rejected": -1.5458664894104004,
|
| 1010 |
+
"logps/chosen": -272.7602233886719,
|
| 1011 |
+
"logps/rejected": -181.00247192382812,
|
| 1012 |
+
"loss": 0.0,
|
| 1013 |
+
"rewards/accuracies": 1.0,
|
| 1014 |
+
"rewards/chosen": 10.925564765930176,
|
| 1015 |
+
"rewards/margins": 22.81020736694336,
|
| 1016 |
+
"rewards/rejected": -11.88464069366455,
|
| 1017 |
+
"step": 72
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.08,
|
| 1021 |
+
"learning_rate": 4.360344827586207e-05,
|
| 1022 |
+
"logits/chosen": -0.3357059359550476,
|
| 1023 |
+
"logits/rejected": -1.4441150426864624,
|
| 1024 |
+
"logps/chosen": -414.8647766113281,
|
| 1025 |
+
"logps/rejected": -223.109375,
|
| 1026 |
+
"loss": 0.0004,
|
| 1027 |
+
"rewards/accuracies": 1.0,
|
| 1028 |
+
"rewards/chosen": 13.181656837463379,
|
| 1029 |
+
"rewards/margins": 27.527368545532227,
|
| 1030 |
+
"rewards/rejected": -14.345712661743164,
|
| 1031 |
+
"step": 73
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.08,
|
| 1035 |
+
"learning_rate": 4.355172413793104e-05,
|
| 1036 |
+
"logits/chosen": -0.6531690359115601,
|
| 1037 |
+
"logits/rejected": -1.4611525535583496,
|
| 1038 |
+
"logps/chosen": -221.1251220703125,
|
| 1039 |
+
"logps/rejected": -226.30352783203125,
|
| 1040 |
+
"loss": 0.0,
|
| 1041 |
+
"rewards/accuracies": 1.0,
|
| 1042 |
+
"rewards/chosen": 7.807477951049805,
|
| 1043 |
+
"rewards/margins": 21.502174377441406,
|
| 1044 |
+
"rewards/rejected": -13.694696426391602,
|
| 1045 |
+
"step": 74
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.08,
|
| 1049 |
+
"learning_rate": 4.35e-05,
|
| 1050 |
+
"logits/chosen": -0.5135932564735413,
|
| 1051 |
+
"logits/rejected": -1.4995931386947632,
|
| 1052 |
+
"logps/chosen": -332.56842041015625,
|
| 1053 |
+
"logps/rejected": -270.9891662597656,
|
| 1054 |
+
"loss": 0.0,
|
| 1055 |
+
"rewards/accuracies": 1.0,
|
| 1056 |
+
"rewards/chosen": 12.301117897033691,
|
| 1057 |
+
"rewards/margins": 26.830467224121094,
|
| 1058 |
+
"rewards/rejected": -14.529346466064453,
|
| 1059 |
+
"step": 75
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.08,
|
| 1063 |
+
"learning_rate": 4.344827586206897e-05,
|
| 1064 |
+
"logits/chosen": -0.3770020604133606,
|
| 1065 |
+
"logits/rejected": -1.44219970703125,
|
| 1066 |
+
"logps/chosen": -317.96240234375,
|
| 1067 |
+
"logps/rejected": -250.3682098388672,
|
| 1068 |
+
"loss": 0.0,
|
| 1069 |
+
"rewards/accuracies": 1.0,
|
| 1070 |
+
"rewards/chosen": 7.8860931396484375,
|
| 1071 |
+
"rewards/margins": 23.53211784362793,
|
| 1072 |
+
"rewards/rejected": -15.646024703979492,
|
| 1073 |
+
"step": 76
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.08,
|
| 1077 |
+
"learning_rate": 4.339655172413793e-05,
|
| 1078 |
+
"logits/chosen": -0.49628371000289917,
|
| 1079 |
+
"logits/rejected": -1.4045476913452148,
|
| 1080 |
+
"logps/chosen": -367.6162414550781,
|
| 1081 |
+
"logps/rejected": -265.2467346191406,
|
| 1082 |
+
"loss": 0.0,
|
| 1083 |
+
"rewards/accuracies": 1.0,
|
| 1084 |
+
"rewards/chosen": 10.148900985717773,
|
| 1085 |
+
"rewards/margins": 24.313669204711914,
|
| 1086 |
+
"rewards/rejected": -14.164766311645508,
|
| 1087 |
+
"step": 77
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 0.09,
|
| 1091 |
+
"learning_rate": 4.33448275862069e-05,
|
| 1092 |
+
"logits/chosen": -0.5287263989448547,
|
| 1093 |
+
"logits/rejected": -1.3665097951889038,
|
| 1094 |
+
"logps/chosen": -341.87664794921875,
|
| 1095 |
+
"logps/rejected": -249.6795654296875,
|
| 1096 |
+
"loss": 0.1517,
|
| 1097 |
+
"rewards/accuracies": 0.9375,
|
| 1098 |
+
"rewards/chosen": 8.620326042175293,
|
| 1099 |
+
"rewards/margins": 19.850658416748047,
|
| 1100 |
+
"rewards/rejected": -11.230331420898438,
|
| 1101 |
+
"step": 78
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.09,
|
| 1105 |
+
"learning_rate": 4.3293103448275865e-05,
|
| 1106 |
+
"logits/chosen": -0.2743966281414032,
|
| 1107 |
+
"logits/rejected": -1.4255130290985107,
|
| 1108 |
+
"logps/chosen": -334.3340759277344,
|
| 1109 |
+
"logps/rejected": -265.15032958984375,
|
| 1110 |
+
"loss": 0.0,
|
| 1111 |
+
"rewards/accuracies": 1.0,
|
| 1112 |
+
"rewards/chosen": 10.872364044189453,
|
| 1113 |
+
"rewards/margins": 27.00977325439453,
|
| 1114 |
+
"rewards/rejected": -16.13741111755371,
|
| 1115 |
+
"step": 79
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 0.09,
|
| 1119 |
+
"learning_rate": 4.324137931034483e-05,
|
| 1120 |
+
"logits/chosen": -0.5563563108444214,
|
| 1121 |
+
"logits/rejected": -1.6934064626693726,
|
| 1122 |
+
"logps/chosen": -408.9612731933594,
|
| 1123 |
+
"logps/rejected": -220.00091552734375,
|
| 1124 |
+
"loss": 0.0,
|
| 1125 |
+
"rewards/accuracies": 1.0,
|
| 1126 |
+
"rewards/chosen": 10.237764358520508,
|
| 1127 |
+
"rewards/margins": 24.824390411376953,
|
| 1128 |
+
"rewards/rejected": -14.586627960205078,
|
| 1129 |
+
"step": 80
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.09,
|
| 1133 |
+
"learning_rate": 4.3189655172413795e-05,
|
| 1134 |
+
"logits/chosen": -0.617451548576355,
|
| 1135 |
+
"logits/rejected": -1.640676736831665,
|
| 1136 |
+
"logps/chosen": -363.1540222167969,
|
| 1137 |
+
"logps/rejected": -226.42376708984375,
|
| 1138 |
+
"loss": 0.0,
|
| 1139 |
+
"rewards/accuracies": 1.0,
|
| 1140 |
+
"rewards/chosen": 10.672024726867676,
|
| 1141 |
+
"rewards/margins": 24.347591400146484,
|
| 1142 |
+
"rewards/rejected": -13.675565719604492,
|
| 1143 |
+
"step": 81
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 0.09,
|
| 1147 |
+
"learning_rate": 4.313793103448276e-05,
|
| 1148 |
+
"logits/chosen": -0.45464372634887695,
|
| 1149 |
+
"logits/rejected": -1.4708175659179688,
|
| 1150 |
+
"logps/chosen": -328.35675048828125,
|
| 1151 |
+
"logps/rejected": -223.79766845703125,
|
| 1152 |
+
"loss": 0.0,
|
| 1153 |
+
"rewards/accuracies": 1.0,
|
| 1154 |
+
"rewards/chosen": 6.965122699737549,
|
| 1155 |
+
"rewards/margins": 21.22077178955078,
|
| 1156 |
+
"rewards/rejected": -14.25564956665039,
|
| 1157 |
+
"step": 82
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.09,
|
| 1161 |
+
"learning_rate": 4.3086206896551724e-05,
|
| 1162 |
+
"logits/chosen": -0.6079340577125549,
|
| 1163 |
+
"logits/rejected": -1.5333082675933838,
|
| 1164 |
+
"logps/chosen": -219.4099578857422,
|
| 1165 |
+
"logps/rejected": -230.07948303222656,
|
| 1166 |
+
"loss": 0.0,
|
| 1167 |
+
"rewards/accuracies": 1.0,
|
| 1168 |
+
"rewards/chosen": 6.278050899505615,
|
| 1169 |
+
"rewards/margins": 20.45656967163086,
|
| 1170 |
+
"rewards/rejected": -14.178520202636719,
|
| 1171 |
+
"step": 83
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.09,
|
| 1175 |
+
"learning_rate": 4.303448275862069e-05,
|
| 1176 |
+
"logits/chosen": -0.4874611794948578,
|
| 1177 |
+
"logits/rejected": -1.5135740041732788,
|
| 1178 |
+
"logps/chosen": -304.00274658203125,
|
| 1179 |
+
"logps/rejected": -263.0120849609375,
|
| 1180 |
+
"loss": 0.0,
|
| 1181 |
+
"rewards/accuracies": 1.0,
|
| 1182 |
+
"rewards/chosen": 9.557435989379883,
|
| 1183 |
+
"rewards/margins": 24.41012954711914,
|
| 1184 |
+
"rewards/rejected": -14.852693557739258,
|
| 1185 |
+
"step": 84
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 0.09,
|
| 1189 |
+
"learning_rate": 4.298275862068966e-05,
|
| 1190 |
+
"logits/chosen": -0.4802761375904083,
|
| 1191 |
+
"logits/rejected": -1.4294252395629883,
|
| 1192 |
+
"logps/chosen": -293.41192626953125,
|
| 1193 |
+
"logps/rejected": -283.2821350097656,
|
| 1194 |
+
"loss": 0.0,
|
| 1195 |
+
"rewards/accuracies": 1.0,
|
| 1196 |
+
"rewards/chosen": 7.370968818664551,
|
| 1197 |
+
"rewards/margins": 24.32880401611328,
|
| 1198 |
+
"rewards/rejected": -16.957834243774414,
|
| 1199 |
+
"step": 85
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 0.09,
|
| 1203 |
+
"learning_rate": 4.293103448275862e-05,
|
| 1204 |
+
"logits/chosen": -0.3580012321472168,
|
| 1205 |
+
"logits/rejected": -1.499201774597168,
|
| 1206 |
+
"logps/chosen": -350.5351257324219,
|
| 1207 |
+
"logps/rejected": -244.6523895263672,
|
| 1208 |
+
"loss": 0.0001,
|
| 1209 |
+
"rewards/accuracies": 1.0,
|
| 1210 |
+
"rewards/chosen": 12.237674713134766,
|
| 1211 |
+
"rewards/margins": 26.58084487915039,
|
| 1212 |
+
"rewards/rejected": -14.343170166015625,
|
| 1213 |
+
"step": 86
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 0.09,
|
| 1217 |
+
"learning_rate": 4.287931034482759e-05,
|
| 1218 |
+
"logits/chosen": -0.6917211413383484,
|
| 1219 |
+
"logits/rejected": -1.4503819942474365,
|
| 1220 |
+
"logps/chosen": -217.5175323486328,
|
| 1221 |
+
"logps/rejected": -228.9481201171875,
|
| 1222 |
+
"loss": 0.0,
|
| 1223 |
+
"rewards/accuracies": 1.0,
|
| 1224 |
+
"rewards/chosen": 6.635045051574707,
|
| 1225 |
+
"rewards/margins": 20.42833709716797,
|
| 1226 |
+
"rewards/rejected": -13.793292999267578,
|
| 1227 |
+
"step": 87
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 0.1,
|
| 1231 |
+
"learning_rate": 4.282758620689656e-05,
|
| 1232 |
+
"logits/chosen": -0.8255922794342041,
|
| 1233 |
+
"logits/rejected": -1.4748644828796387,
|
| 1234 |
+
"logps/chosen": -217.336669921875,
|
| 1235 |
+
"logps/rejected": -206.8959197998047,
|
| 1236 |
+
"loss": 0.0005,
|
| 1237 |
+
"rewards/accuracies": 1.0,
|
| 1238 |
+
"rewards/chosen": 5.22888708114624,
|
| 1239 |
+
"rewards/margins": 19.673564910888672,
|
| 1240 |
+
"rewards/rejected": -14.444679260253906,
|
| 1241 |
+
"step": 88
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 0.1,
|
| 1245 |
+
"learning_rate": 4.2775862068965525e-05,
|
| 1246 |
+
"logits/chosen": -0.626908004283905,
|
| 1247 |
+
"logits/rejected": -1.6663663387298584,
|
| 1248 |
+
"logps/chosen": -328.56231689453125,
|
| 1249 |
+
"logps/rejected": -235.42843627929688,
|
| 1250 |
+
"loss": 0.0,
|
| 1251 |
+
"rewards/accuracies": 1.0,
|
| 1252 |
+
"rewards/chosen": 6.891221046447754,
|
| 1253 |
+
"rewards/margins": 21.77893829345703,
|
| 1254 |
+
"rewards/rejected": -14.887716293334961,
|
| 1255 |
+
"step": 89
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 0.1,
|
| 1259 |
+
"learning_rate": 4.2724137931034486e-05,
|
| 1260 |
+
"logits/chosen": -0.5689429640769958,
|
| 1261 |
+
"logits/rejected": -1.618609070777893,
|
| 1262 |
+
"logps/chosen": -298.08770751953125,
|
| 1263 |
+
"logps/rejected": -237.06094360351562,
|
| 1264 |
+
"loss": 0.0,
|
| 1265 |
+
"rewards/accuracies": 1.0,
|
| 1266 |
+
"rewards/chosen": 9.328791618347168,
|
| 1267 |
+
"rewards/margins": 24.572046279907227,
|
| 1268 |
+
"rewards/rejected": -15.243253707885742,
|
| 1269 |
+
"step": 90
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 0.1,
|
| 1273 |
+
"learning_rate": 4.2672413793103454e-05,
|
| 1274 |
+
"logits/chosen": -0.5474239587783813,
|
| 1275 |
+
"logits/rejected": -1.6142076253890991,
|
| 1276 |
+
"logps/chosen": -295.00494384765625,
|
| 1277 |
+
"logps/rejected": -268.9505615234375,
|
| 1278 |
+
"loss": 0.0287,
|
| 1279 |
+
"rewards/accuracies": 1.0,
|
| 1280 |
+
"rewards/chosen": 9.715471267700195,
|
| 1281 |
+
"rewards/margins": 23.968780517578125,
|
| 1282 |
+
"rewards/rejected": -14.25330924987793,
|
| 1283 |
+
"step": 91
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 0.1,
|
| 1287 |
+
"learning_rate": 4.2620689655172416e-05,
|
| 1288 |
+
"logits/chosen": -0.5132906436920166,
|
| 1289 |
+
"logits/rejected": -1.6248977184295654,
|
| 1290 |
+
"logps/chosen": -314.5284423828125,
|
| 1291 |
+
"logps/rejected": -210.1342315673828,
|
| 1292 |
+
"loss": 0.0595,
|
| 1293 |
+
"rewards/accuracies": 0.9375,
|
| 1294 |
+
"rewards/chosen": 7.707674980163574,
|
| 1295 |
+
"rewards/margins": 20.18167495727539,
|
| 1296 |
+
"rewards/rejected": -12.474000930786133,
|
| 1297 |
+
"step": 92
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 0.1,
|
| 1301 |
+
"learning_rate": 4.2568965517241384e-05,
|
| 1302 |
+
"logits/chosen": -0.45244526863098145,
|
| 1303 |
+
"logits/rejected": -1.7539148330688477,
|
| 1304 |
+
"logps/chosen": -394.9619140625,
|
| 1305 |
+
"logps/rejected": -230.32550048828125,
|
| 1306 |
+
"loss": 0.0,
|
| 1307 |
+
"rewards/accuracies": 1.0,
|
| 1308 |
+
"rewards/chosen": 11.32716178894043,
|
| 1309 |
+
"rewards/margins": 26.326129913330078,
|
| 1310 |
+
"rewards/rejected": -14.998967170715332,
|
| 1311 |
+
"step": 93
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 0.1,
|
| 1315 |
+
"learning_rate": 4.2517241379310345e-05,
|
| 1316 |
+
"logits/chosen": -0.7538139224052429,
|
| 1317 |
+
"logits/rejected": -1.5796599388122559,
|
| 1318 |
+
"logps/chosen": -286.95281982421875,
|
| 1319 |
+
"logps/rejected": -266.7412414550781,
|
| 1320 |
+
"loss": 0.0,
|
| 1321 |
+
"rewards/accuracies": 1.0,
|
| 1322 |
+
"rewards/chosen": 8.003694534301758,
|
| 1323 |
+
"rewards/margins": 25.05299949645996,
|
| 1324 |
+
"rewards/rejected": -17.04930305480957,
|
| 1325 |
+
"step": 94
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 0.1,
|
| 1329 |
+
"learning_rate": 4.246551724137931e-05,
|
| 1330 |
+
"logits/chosen": -0.44376400113105774,
|
| 1331 |
+
"logits/rejected": -1.5493687391281128,
|
| 1332 |
+
"logps/chosen": -406.93951416015625,
|
| 1333 |
+
"logps/rejected": -302.9401550292969,
|
| 1334 |
+
"loss": 0.0,
|
| 1335 |
+
"rewards/accuracies": 1.0,
|
| 1336 |
+
"rewards/chosen": 7.246862888336182,
|
| 1337 |
+
"rewards/margins": 27.937149047851562,
|
| 1338 |
+
"rewards/rejected": -20.69028663635254,
|
| 1339 |
+
"step": 95
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 0.1,
|
| 1343 |
+
"learning_rate": 4.2413793103448274e-05,
|
| 1344 |
+
"logits/chosen": -0.7577255964279175,
|
| 1345 |
+
"logits/rejected": -1.77558171749115,
|
| 1346 |
+
"logps/chosen": -357.90838623046875,
|
| 1347 |
+
"logps/rejected": -246.73458862304688,
|
| 1348 |
+
"loss": 0.0,
|
| 1349 |
+
"rewards/accuracies": 1.0,
|
| 1350 |
+
"rewards/chosen": 7.267894744873047,
|
| 1351 |
+
"rewards/margins": 24.79804229736328,
|
| 1352 |
+
"rewards/rejected": -17.530147552490234,
|
| 1353 |
+
"step": 96
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 0.11,
|
| 1357 |
+
"learning_rate": 4.236206896551724e-05,
|
| 1358 |
+
"logits/chosen": -0.8565713763237,
|
| 1359 |
+
"logits/rejected": -1.7760589122772217,
|
| 1360 |
+
"logps/chosen": -338.751220703125,
|
| 1361 |
+
"logps/rejected": -304.101806640625,
|
| 1362 |
+
"loss": 0.0,
|
| 1363 |
+
"rewards/accuracies": 1.0,
|
| 1364 |
+
"rewards/chosen": 9.238365173339844,
|
| 1365 |
+
"rewards/margins": 27.781978607177734,
|
| 1366 |
+
"rewards/rejected": -18.54361343383789,
|
| 1367 |
+
"step": 97
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 0.11,
|
| 1371 |
+
"learning_rate": 4.231034482758621e-05,
|
| 1372 |
+
"logits/chosen": -0.40401750802993774,
|
| 1373 |
+
"logits/rejected": -1.7701493501663208,
|
| 1374 |
+
"logps/chosen": -537.5440063476562,
|
| 1375 |
+
"logps/rejected": -349.8548889160156,
|
| 1376 |
+
"loss": 0.0,
|
| 1377 |
+
"rewards/accuracies": 1.0,
|
| 1378 |
+
"rewards/chosen": 14.211512565612793,
|
| 1379 |
+
"rewards/margins": 37.746315002441406,
|
| 1380 |
+
"rewards/rejected": -23.53480339050293,
|
| 1381 |
+
"step": 98
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 0.11,
|
| 1385 |
+
"learning_rate": 4.225862068965517e-05,
|
| 1386 |
+
"logits/chosen": -0.9799883365631104,
|
| 1387 |
+
"logits/rejected": -1.8936114311218262,
|
| 1388 |
+
"logps/chosen": -403.7567138671875,
|
| 1389 |
+
"logps/rejected": -257.6401672363281,
|
| 1390 |
+
"loss": 0.0001,
|
| 1391 |
+
"rewards/accuracies": 1.0,
|
| 1392 |
+
"rewards/chosen": 4.710639476776123,
|
| 1393 |
+
"rewards/margins": 23.923961639404297,
|
| 1394 |
+
"rewards/rejected": -19.213321685791016,
|
| 1395 |
+
"step": 99
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 0.11,
|
| 1399 |
+
"learning_rate": 4.220689655172414e-05,
|
| 1400 |
+
"logits/chosen": -0.94623202085495,
|
| 1401 |
+
"logits/rejected": -1.7190004587173462,
|
| 1402 |
+
"logps/chosen": -360.8213806152344,
|
| 1403 |
+
"logps/rejected": -310.1994323730469,
|
| 1404 |
+
"loss": 0.0,
|
| 1405 |
+
"rewards/accuracies": 1.0,
|
| 1406 |
+
"rewards/chosen": 3.991710662841797,
|
| 1407 |
+
"rewards/margins": 26.59444808959961,
|
| 1408 |
+
"rewards/rejected": -22.60273551940918,
|
| 1409 |
+
"step": 100
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 0.11,
|
| 1413 |
+
"learning_rate": 4.215517241379311e-05,
|
| 1414 |
+
"logits/chosen": -0.6102348566055298,
|
| 1415 |
+
"logits/rejected": -1.8472094535827637,
|
| 1416 |
+
"logps/chosen": -460.48699951171875,
|
| 1417 |
+
"logps/rejected": -369.1959228515625,
|
| 1418 |
+
"loss": 0.0002,
|
| 1419 |
+
"rewards/accuracies": 1.0,
|
| 1420 |
+
"rewards/chosen": 6.696294784545898,
|
| 1421 |
+
"rewards/margins": 32.79343795776367,
|
| 1422 |
+
"rewards/rejected": -26.097143173217773,
|
| 1423 |
+
"step": 101
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 0.11,
|
| 1427 |
+
"learning_rate": 4.210344827586207e-05,
|
| 1428 |
+
"logits/chosen": -0.9649127721786499,
|
| 1429 |
+
"logits/rejected": -1.8413913249969482,
|
| 1430 |
+
"logps/chosen": -455.1119079589844,
|
| 1431 |
+
"logps/rejected": -365.0443420410156,
|
| 1432 |
+
"loss": 0.0,
|
| 1433 |
+
"rewards/accuracies": 1.0,
|
| 1434 |
+
"rewards/chosen": 7.14402437210083,
|
| 1435 |
+
"rewards/margins": 32.23406219482422,
|
| 1436 |
+
"rewards/rejected": -25.090038299560547,
|
| 1437 |
+
"step": 102
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 0.11,
|
| 1441 |
+
"learning_rate": 4.2051724137931036e-05,
|
| 1442 |
+
"logits/chosen": -1.176360011100769,
|
| 1443 |
+
"logits/rejected": -1.6914632320404053,
|
| 1444 |
+
"logps/chosen": -283.3887023925781,
|
| 1445 |
+
"logps/rejected": -343.1808166503906,
|
| 1446 |
+
"loss": 0.0008,
|
| 1447 |
+
"rewards/accuracies": 1.0,
|
| 1448 |
+
"rewards/chosen": -0.03441965579986572,
|
| 1449 |
+
"rewards/margins": 24.389293670654297,
|
| 1450 |
+
"rewards/rejected": -24.423709869384766,
|
| 1451 |
+
"step": 103
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 0.11,
|
| 1455 |
+
"learning_rate": 4.2000000000000004e-05,
|
| 1456 |
+
"logits/chosen": -0.7725791335105896,
|
| 1457 |
+
"logits/rejected": -1.943673014640808,
|
| 1458 |
+
"logps/chosen": -484.6883544921875,
|
| 1459 |
+
"logps/rejected": -338.5666198730469,
|
| 1460 |
+
"loss": 0.0,
|
| 1461 |
+
"rewards/accuracies": 1.0,
|
| 1462 |
+
"rewards/chosen": 7.784733295440674,
|
| 1463 |
+
"rewards/margins": 31.62430763244629,
|
| 1464 |
+
"rewards/rejected": -23.839576721191406,
|
| 1465 |
+
"step": 104
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 0.11,
|
| 1469 |
+
"learning_rate": 4.194827586206897e-05,
|
| 1470 |
+
"logits/chosen": -0.6000739336013794,
|
| 1471 |
+
"logits/rejected": -1.681613802909851,
|
| 1472 |
+
"logps/chosen": -342.3570556640625,
|
| 1473 |
+
"logps/rejected": -256.06829833984375,
|
| 1474 |
+
"loss": 0.0,
|
| 1475 |
+
"rewards/accuracies": 1.0,
|
| 1476 |
+
"rewards/chosen": 7.42122745513916,
|
| 1477 |
+
"rewards/margins": 25.636981964111328,
|
| 1478 |
+
"rewards/rejected": -18.21575355529785,
|
| 1479 |
+
"step": 105
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 0.12,
|
| 1483 |
+
"learning_rate": 4.1896551724137934e-05,
|
| 1484 |
+
"logits/chosen": -0.5727726221084595,
|
| 1485 |
+
"logits/rejected": -1.5278774499893188,
|
| 1486 |
+
"logps/chosen": -340.2919921875,
|
| 1487 |
+
"logps/rejected": -246.67628479003906,
|
| 1488 |
+
"loss": 0.0183,
|
| 1489 |
+
"rewards/accuracies": 1.0,
|
| 1490 |
+
"rewards/chosen": 7.279099464416504,
|
| 1491 |
+
"rewards/margins": 23.8295841217041,
|
| 1492 |
+
"rewards/rejected": -16.550485610961914,
|
| 1493 |
+
"step": 106
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 0.12,
|
| 1497 |
+
"learning_rate": 4.18448275862069e-05,
|
| 1498 |
+
"logits/chosen": -0.588262677192688,
|
| 1499 |
+
"logits/rejected": -1.4113011360168457,
|
| 1500 |
+
"logps/chosen": -239.1262664794922,
|
| 1501 |
+
"logps/rejected": -323.8817443847656,
|
| 1502 |
+
"loss": 0.0,
|
| 1503 |
+
"rewards/accuracies": 1.0,
|
| 1504 |
+
"rewards/chosen": 4.26235294342041,
|
| 1505 |
+
"rewards/margins": 25.60938262939453,
|
| 1506 |
+
"rewards/rejected": -21.34703254699707,
|
| 1507 |
+
"step": 107
|
| 1508 |
+
},
|
| 1509 |
+
{
|
| 1510 |
+
"epoch": 0.12,
|
| 1511 |
+
"learning_rate": 4.179310344827587e-05,
|
| 1512 |
+
"logits/chosen": -0.825312077999115,
|
| 1513 |
+
"logits/rejected": -1.6536335945129395,
|
| 1514 |
+
"logps/chosen": -245.26333618164062,
|
| 1515 |
+
"logps/rejected": -275.9490966796875,
|
| 1516 |
+
"loss": 0.0,
|
| 1517 |
+
"rewards/accuracies": 1.0,
|
| 1518 |
+
"rewards/chosen": 4.488056182861328,
|
| 1519 |
+
"rewards/margins": 24.79592514038086,
|
| 1520 |
+
"rewards/rejected": -20.30786895751953,
|
| 1521 |
+
"step": 108
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 0.12,
|
| 1525 |
+
"learning_rate": 4.174137931034483e-05,
|
| 1526 |
+
"logits/chosen": -0.6509658694267273,
|
| 1527 |
+
"logits/rejected": -1.7768073081970215,
|
| 1528 |
+
"logps/chosen": -494.1702575683594,
|
| 1529 |
+
"logps/rejected": -269.7856140136719,
|
| 1530 |
+
"loss": 0.0,
|
| 1531 |
+
"rewards/accuracies": 1.0,
|
| 1532 |
+
"rewards/chosen": 11.041641235351562,
|
| 1533 |
+
"rewards/margins": 30.055639266967773,
|
| 1534 |
+
"rewards/rejected": -19.013999938964844,
|
| 1535 |
+
"step": 109
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 0.12,
|
| 1539 |
+
"learning_rate": 4.16896551724138e-05,
|
| 1540 |
+
"logits/chosen": -0.813829243183136,
|
| 1541 |
+
"logits/rejected": -1.7704083919525146,
|
| 1542 |
+
"logps/chosen": -370.6927795410156,
|
| 1543 |
+
"logps/rejected": -288.31463623046875,
|
| 1544 |
+
"loss": 0.0,
|
| 1545 |
+
"rewards/accuracies": 1.0,
|
| 1546 |
+
"rewards/chosen": 10.1367826461792,
|
| 1547 |
+
"rewards/margins": 30.452945709228516,
|
| 1548 |
+
"rewards/rejected": -20.316162109375,
|
| 1549 |
+
"step": 110
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 0.12,
|
| 1553 |
+
"learning_rate": 4.163793103448276e-05,
|
| 1554 |
+
"logits/chosen": -0.450665146112442,
|
| 1555 |
+
"logits/rejected": -1.4787647724151611,
|
| 1556 |
+
"logps/chosen": -336.4142761230469,
|
| 1557 |
+
"logps/rejected": -275.8489990234375,
|
| 1558 |
+
"loss": 0.0,
|
| 1559 |
+
"rewards/accuracies": 1.0,
|
| 1560 |
+
"rewards/chosen": 6.380362510681152,
|
| 1561 |
+
"rewards/margins": 27.303016662597656,
|
| 1562 |
+
"rewards/rejected": -20.922651290893555,
|
| 1563 |
+
"step": 111
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 0.12,
|
| 1567 |
+
"learning_rate": 4.158620689655173e-05,
|
| 1568 |
+
"logits/chosen": -0.7610673904418945,
|
| 1569 |
+
"logits/rejected": -1.5079166889190674,
|
| 1570 |
+
"logps/chosen": -290.8649597167969,
|
| 1571 |
+
"logps/rejected": -391.47772216796875,
|
| 1572 |
+
"loss": 0.4671,
|
| 1573 |
+
"rewards/accuracies": 0.9375,
|
| 1574 |
+
"rewards/chosen": 4.447336196899414,
|
| 1575 |
+
"rewards/margins": 27.807403564453125,
|
| 1576 |
+
"rewards/rejected": -23.36006736755371,
|
| 1577 |
+
"step": 112
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 0.12,
|
| 1581 |
+
"learning_rate": 4.153448275862069e-05,
|
| 1582 |
+
"logits/chosen": -0.9873843193054199,
|
| 1583 |
+
"logits/rejected": -1.6280449628829956,
|
| 1584 |
+
"logps/chosen": -244.6575164794922,
|
| 1585 |
+
"logps/rejected": -324.1699523925781,
|
| 1586 |
+
"loss": 0.0036,
|
| 1587 |
+
"rewards/accuracies": 1.0,
|
| 1588 |
+
"rewards/chosen": 2.1840460300445557,
|
| 1589 |
+
"rewards/margins": 25.491558074951172,
|
| 1590 |
+
"rewards/rejected": -23.307514190673828,
|
| 1591 |
+
"step": 113
|
| 1592 |
+
},
|
| 1593 |
+
{
|
| 1594 |
+
"epoch": 0.12,
|
| 1595 |
+
"learning_rate": 4.148275862068966e-05,
|
| 1596 |
+
"logits/chosen": -1.0418487787246704,
|
| 1597 |
+
"logits/rejected": -1.7689330577850342,
|
| 1598 |
+
"logps/chosen": -309.9642028808594,
|
| 1599 |
+
"logps/rejected": -433.8427734375,
|
| 1600 |
+
"loss": 0.0,
|
| 1601 |
+
"rewards/accuracies": 1.0,
|
| 1602 |
+
"rewards/chosen": 3.8272902965545654,
|
| 1603 |
+
"rewards/margins": 33.52722930908203,
|
| 1604 |
+
"rewards/rejected": -29.69993782043457,
|
| 1605 |
+
"step": 114
|
| 1606 |
+
},
|
| 1607 |
+
{
|
| 1608 |
+
"epoch": 0.13,
|
| 1609 |
+
"learning_rate": 4.143103448275862e-05,
|
| 1610 |
+
"logits/chosen": -0.9574589133262634,
|
| 1611 |
+
"logits/rejected": -1.9435979127883911,
|
| 1612 |
+
"logps/chosen": -485.497314453125,
|
| 1613 |
+
"logps/rejected": -301.46038818359375,
|
| 1614 |
+
"loss": 0.0002,
|
| 1615 |
+
"rewards/accuracies": 1.0,
|
| 1616 |
+
"rewards/chosen": 3.4343650341033936,
|
| 1617 |
+
"rewards/margins": 27.89614486694336,
|
| 1618 |
+
"rewards/rejected": -24.46177864074707,
|
| 1619 |
+
"step": 115
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 0.13,
|
| 1623 |
+
"learning_rate": 4.1379310344827587e-05,
|
| 1624 |
+
"logits/chosen": -1.2192739248275757,
|
| 1625 |
+
"logits/rejected": -1.9555035829544067,
|
| 1626 |
+
"logps/chosen": -485.7822265625,
|
| 1627 |
+
"logps/rejected": -452.36566162109375,
|
| 1628 |
+
"loss": 0.0,
|
| 1629 |
+
"rewards/accuracies": 1.0,
|
| 1630 |
+
"rewards/chosen": -3.4936442375183105,
|
| 1631 |
+
"rewards/margins": 31.84035873413086,
|
| 1632 |
+
"rewards/rejected": -35.33400344848633,
|
| 1633 |
+
"step": 116
|
| 1634 |
+
},
|
| 1635 |
+
{
|
| 1636 |
+
"epoch": 0.13,
|
| 1637 |
+
"learning_rate": 4.1327586206896555e-05,
|
| 1638 |
+
"logits/chosen": -1.3901960849761963,
|
| 1639 |
+
"logits/rejected": -2.194340944290161,
|
| 1640 |
+
"logps/chosen": -556.4442749023438,
|
| 1641 |
+
"logps/rejected": -403.41094970703125,
|
| 1642 |
+
"loss": 0.4842,
|
| 1643 |
+
"rewards/accuracies": 0.9375,
|
| 1644 |
+
"rewards/chosen": -6.604147434234619,
|
| 1645 |
+
"rewards/margins": 25.96053695678711,
|
| 1646 |
+
"rewards/rejected": -32.5646858215332,
|
| 1647 |
+
"step": 117
|
| 1648 |
+
},
|
| 1649 |
+
{
|
| 1650 |
+
"epoch": 0.13,
|
| 1651 |
+
"learning_rate": 4.1275862068965516e-05,
|
| 1652 |
+
"logits/chosen": -1.10233473777771,
|
| 1653 |
+
"logits/rejected": -1.911782145500183,
|
| 1654 |
+
"logps/chosen": -319.7232666015625,
|
| 1655 |
+
"logps/rejected": -366.8639221191406,
|
| 1656 |
+
"loss": 0.0,
|
| 1657 |
+
"rewards/accuracies": 1.0,
|
| 1658 |
+
"rewards/chosen": 0.5800843238830566,
|
| 1659 |
+
"rewards/margins": 30.92066192626953,
|
| 1660 |
+
"rewards/rejected": -30.3405818939209,
|
| 1661 |
+
"step": 118
|
| 1662 |
+
},
|
| 1663 |
+
{
|
| 1664 |
+
"epoch": 0.13,
|
| 1665 |
+
"learning_rate": 4.1224137931034484e-05,
|
| 1666 |
+
"logits/chosen": -0.7030006051063538,
|
| 1667 |
+
"logits/rejected": -1.7387290000915527,
|
| 1668 |
+
"logps/chosen": -291.53961181640625,
|
| 1669 |
+
"logps/rejected": -350.3901062011719,
|
| 1670 |
+
"loss": 0.0,
|
| 1671 |
+
"rewards/accuracies": 1.0,
|
| 1672 |
+
"rewards/chosen": 1.4014167785644531,
|
| 1673 |
+
"rewards/margins": 28.830717086791992,
|
| 1674 |
+
"rewards/rejected": -27.429298400878906,
|
| 1675 |
+
"step": 119
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 0.13,
|
| 1679 |
+
"learning_rate": 4.117241379310345e-05,
|
| 1680 |
+
"logits/chosen": -0.5296112895011902,
|
| 1681 |
+
"logits/rejected": -1.5072706937789917,
|
| 1682 |
+
"logps/chosen": -286.94879150390625,
|
| 1683 |
+
"logps/rejected": -346.3030700683594,
|
| 1684 |
+
"loss": 0.0,
|
| 1685 |
+
"rewards/accuracies": 1.0,
|
| 1686 |
+
"rewards/chosen": 1.8979191780090332,
|
| 1687 |
+
"rewards/margins": 28.363784790039062,
|
| 1688 |
+
"rewards/rejected": -26.465864181518555,
|
| 1689 |
+
"step": 120
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 0.13,
|
| 1693 |
+
"learning_rate": 4.112068965517242e-05,
|
| 1694 |
+
"logits/chosen": -0.42470210790634155,
|
| 1695 |
+
"logits/rejected": -1.7659740447998047,
|
| 1696 |
+
"logps/chosen": -477.8876953125,
|
| 1697 |
+
"logps/rejected": -315.16717529296875,
|
| 1698 |
+
"loss": 0.0,
|
| 1699 |
+
"rewards/accuracies": 1.0,
|
| 1700 |
+
"rewards/chosen": 9.757699966430664,
|
| 1701 |
+
"rewards/margins": 34.151214599609375,
|
| 1702 |
+
"rewards/rejected": -24.393516540527344,
|
| 1703 |
+
"step": 121
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 0.13,
|
| 1707 |
+
"learning_rate": 4.106896551724138e-05,
|
| 1708 |
+
"logits/chosen": -0.37895911931991577,
|
| 1709 |
+
"logits/rejected": -1.4872817993164062,
|
| 1710 |
+
"logps/chosen": -360.4671936035156,
|
| 1711 |
+
"logps/rejected": -311.3260803222656,
|
| 1712 |
+
"loss": 0.0,
|
| 1713 |
+
"rewards/accuracies": 1.0,
|
| 1714 |
+
"rewards/chosen": 10.024797439575195,
|
| 1715 |
+
"rewards/margins": 30.34416961669922,
|
| 1716 |
+
"rewards/rejected": -20.31937026977539,
|
| 1717 |
+
"step": 122
|
| 1718 |
+
},
|
| 1719 |
+
{
|
| 1720 |
+
"epoch": 0.13,
|
| 1721 |
+
"learning_rate": 4.101724137931035e-05,
|
| 1722 |
+
"logits/chosen": -0.4127517342567444,
|
| 1723 |
+
"logits/rejected": -1.4179290533065796,
|
| 1724 |
+
"logps/chosen": -396.7817077636719,
|
| 1725 |
+
"logps/rejected": -257.0691833496094,
|
| 1726 |
+
"loss": 0.0,
|
| 1727 |
+
"rewards/accuracies": 1.0,
|
| 1728 |
+
"rewards/chosen": 8.115121841430664,
|
| 1729 |
+
"rewards/margins": 25.882369995117188,
|
| 1730 |
+
"rewards/rejected": -17.767250061035156,
|
| 1731 |
+
"step": 123
|
| 1732 |
+
},
|
| 1733 |
+
{
|
| 1734 |
+
"epoch": 0.14,
|
| 1735 |
+
"learning_rate": 4.096551724137932e-05,
|
| 1736 |
+
"logits/chosen": -0.4867606461048126,
|
| 1737 |
+
"logits/rejected": -1.433240532875061,
|
| 1738 |
+
"logps/chosen": -307.6831359863281,
|
| 1739 |
+
"logps/rejected": -285.9851379394531,
|
| 1740 |
+
"loss": 0.0006,
|
| 1741 |
+
"rewards/accuracies": 1.0,
|
| 1742 |
+
"rewards/chosen": 9.156608581542969,
|
| 1743 |
+
"rewards/margins": 24.28235626220703,
|
| 1744 |
+
"rewards/rejected": -15.12574577331543,
|
| 1745 |
+
"step": 124
|
| 1746 |
+
},
|
| 1747 |
+
{
|
| 1748 |
+
"epoch": 0.14,
|
| 1749 |
+
"learning_rate": 4.091379310344828e-05,
|
| 1750 |
+
"logits/chosen": -0.5465874671936035,
|
| 1751 |
+
"logits/rejected": -1.398258924484253,
|
| 1752 |
+
"logps/chosen": -244.0363311767578,
|
| 1753 |
+
"logps/rejected": -240.68551635742188,
|
| 1754 |
+
"loss": 0.0,
|
| 1755 |
+
"rewards/accuracies": 1.0,
|
| 1756 |
+
"rewards/chosen": 4.43398904800415,
|
| 1757 |
+
"rewards/margins": 23.1319522857666,
|
| 1758 |
+
"rewards/rejected": -18.69796371459961,
|
| 1759 |
+
"step": 125
|
| 1760 |
+
},
|
| 1761 |
+
{
|
| 1762 |
+
"epoch": 0.14,
|
| 1763 |
+
"learning_rate": 4.0862068965517246e-05,
|
| 1764 |
+
"logits/chosen": -0.38893836736679077,
|
| 1765 |
+
"logits/rejected": -1.427646279335022,
|
| 1766 |
+
"logps/chosen": -351.76348876953125,
|
| 1767 |
+
"logps/rejected": -289.0392150878906,
|
| 1768 |
+
"loss": 0.0,
|
| 1769 |
+
"rewards/accuracies": 1.0,
|
| 1770 |
+
"rewards/chosen": 7.574392795562744,
|
| 1771 |
+
"rewards/margins": 25.838218688964844,
|
| 1772 |
+
"rewards/rejected": -18.26382827758789,
|
| 1773 |
+
"step": 126
|
| 1774 |
+
},
|
| 1775 |
+
{
|
| 1776 |
+
"epoch": 0.14,
|
| 1777 |
+
"learning_rate": 4.0810344827586214e-05,
|
| 1778 |
+
"logits/chosen": -0.3710274398326874,
|
| 1779 |
+
"logits/rejected": -1.5999269485473633,
|
| 1780 |
+
"logps/chosen": -394.9821472167969,
|
| 1781 |
+
"logps/rejected": -246.48887634277344,
|
| 1782 |
+
"loss": 0.0,
|
| 1783 |
+
"rewards/accuracies": 1.0,
|
| 1784 |
+
"rewards/chosen": 8.973166465759277,
|
| 1785 |
+
"rewards/margins": 26.636192321777344,
|
| 1786 |
+
"rewards/rejected": -17.66302490234375,
|
| 1787 |
+
"step": 127
|
| 1788 |
+
},
|
| 1789 |
+
{
|
| 1790 |
+
"epoch": 0.14,
|
| 1791 |
+
"learning_rate": 4.0758620689655175e-05,
|
| 1792 |
+
"logits/chosen": -0.673173725605011,
|
| 1793 |
+
"logits/rejected": -1.2474948167800903,
|
| 1794 |
+
"logps/chosen": -257.3236083984375,
|
| 1795 |
+
"logps/rejected": -264.7505187988281,
|
| 1796 |
+
"loss": 0.0,
|
| 1797 |
+
"rewards/accuracies": 1.0,
|
| 1798 |
+
"rewards/chosen": 2.773941993713379,
|
| 1799 |
+
"rewards/margins": 19.319332122802734,
|
| 1800 |
+
"rewards/rejected": -16.545391082763672,
|
| 1801 |
+
"step": 128
|
| 1802 |
+
},
|
| 1803 |
+
{
|
| 1804 |
+
"epoch": 0.14,
|
| 1805 |
+
"learning_rate": 4.0706896551724143e-05,
|
| 1806 |
+
"logits/chosen": -0.560947060585022,
|
| 1807 |
+
"logits/rejected": -1.374908447265625,
|
| 1808 |
+
"logps/chosen": -301.8670959472656,
|
| 1809 |
+
"logps/rejected": -232.6251983642578,
|
| 1810 |
+
"loss": 0.0,
|
| 1811 |
+
"rewards/accuracies": 1.0,
|
| 1812 |
+
"rewards/chosen": 10.369309425354004,
|
| 1813 |
+
"rewards/margins": 26.155746459960938,
|
| 1814 |
+
"rewards/rejected": -15.786436080932617,
|
| 1815 |
+
"step": 129
|
| 1816 |
+
},
|
| 1817 |
+
{
|
| 1818 |
+
"epoch": 0.14,
|
| 1819 |
+
"learning_rate": 4.0655172413793105e-05,
|
| 1820 |
+
"logits/chosen": -0.6484676003456116,
|
| 1821 |
+
"logits/rejected": -1.3061511516571045,
|
| 1822 |
+
"logps/chosen": -240.1058807373047,
|
| 1823 |
+
"logps/rejected": -292.108154296875,
|
| 1824 |
+
"loss": 0.0,
|
| 1825 |
+
"rewards/accuracies": 1.0,
|
| 1826 |
+
"rewards/chosen": 4.469289302825928,
|
| 1827 |
+
"rewards/margins": 25.45884132385254,
|
| 1828 |
+
"rewards/rejected": -20.989551544189453,
|
| 1829 |
+
"step": 130
|
| 1830 |
+
},
|
| 1831 |
+
{
|
| 1832 |
+
"epoch": 0.14,
|
| 1833 |
+
"learning_rate": 4.060344827586207e-05,
|
| 1834 |
+
"logits/chosen": -0.4261731505393982,
|
| 1835 |
+
"logits/rejected": -1.5151227712631226,
|
| 1836 |
+
"logps/chosen": -402.2912292480469,
|
| 1837 |
+
"logps/rejected": -289.1283264160156,
|
| 1838 |
+
"loss": 0.0,
|
| 1839 |
+
"rewards/accuracies": 1.0,
|
| 1840 |
+
"rewards/chosen": 10.882993698120117,
|
| 1841 |
+
"rewards/margins": 30.517051696777344,
|
| 1842 |
+
"rewards/rejected": -19.63405990600586,
|
| 1843 |
+
"step": 131
|
| 1844 |
+
},
|
| 1845 |
+
{
|
| 1846 |
+
"epoch": 0.14,
|
| 1847 |
+
"learning_rate": 4.0551724137931034e-05,
|
| 1848 |
+
"logits/chosen": -0.3702995181083679,
|
| 1849 |
+
"logits/rejected": -1.5198004245758057,
|
| 1850 |
+
"logps/chosen": -377.6826477050781,
|
| 1851 |
+
"logps/rejected": -287.36566162109375,
|
| 1852 |
+
"loss": 0.0,
|
| 1853 |
+
"rewards/accuracies": 1.0,
|
| 1854 |
+
"rewards/chosen": 10.187080383300781,
|
| 1855 |
+
"rewards/margins": 30.941383361816406,
|
| 1856 |
+
"rewards/rejected": -20.754301071166992,
|
| 1857 |
+
"step": 132
|
| 1858 |
+
},
|
| 1859 |
+
{
|
| 1860 |
+
"epoch": 0.15,
|
| 1861 |
+
"learning_rate": 4.05e-05,
|
| 1862 |
+
"logits/chosen": -0.394854873418808,
|
| 1863 |
+
"logits/rejected": -1.641805648803711,
|
| 1864 |
+
"logps/chosen": -462.9255676269531,
|
| 1865 |
+
"logps/rejected": -254.52276611328125,
|
| 1866 |
+
"loss": 0.0,
|
| 1867 |
+
"rewards/accuracies": 1.0,
|
| 1868 |
+
"rewards/chosen": 10.504643440246582,
|
| 1869 |
+
"rewards/margins": 27.564743041992188,
|
| 1870 |
+
"rewards/rejected": -17.060100555419922,
|
| 1871 |
+
"step": 133
|
| 1872 |
+
},
|
| 1873 |
+
{
|
| 1874 |
+
"epoch": 0.15,
|
| 1875 |
+
"learning_rate": 4.044827586206896e-05,
|
| 1876 |
+
"logits/chosen": -0.3886590600013733,
|
| 1877 |
+
"logits/rejected": -1.4597467184066772,
|
| 1878 |
+
"logps/chosen": -378.5803527832031,
|
| 1879 |
+
"logps/rejected": -263.2305908203125,
|
| 1880 |
+
"loss": 0.0,
|
| 1881 |
+
"rewards/accuracies": 1.0,
|
| 1882 |
+
"rewards/chosen": 13.531683921813965,
|
| 1883 |
+
"rewards/margins": 31.480838775634766,
|
| 1884 |
+
"rewards/rejected": -17.949153900146484,
|
| 1885 |
+
"step": 134
|
| 1886 |
+
},
|
| 1887 |
+
{
|
| 1888 |
+
"epoch": 0.15,
|
| 1889 |
+
"learning_rate": 4.039655172413793e-05,
|
| 1890 |
+
"logits/chosen": -0.5102599859237671,
|
| 1891 |
+
"logits/rejected": -1.250558614730835,
|
| 1892 |
+
"logps/chosen": -295.6455078125,
|
| 1893 |
+
"logps/rejected": -345.3118591308594,
|
| 1894 |
+
"loss": 0.0,
|
| 1895 |
+
"rewards/accuracies": 1.0,
|
| 1896 |
+
"rewards/chosen": 7.247247695922852,
|
| 1897 |
+
"rewards/margins": 26.292499542236328,
|
| 1898 |
+
"rewards/rejected": -19.04525375366211,
|
| 1899 |
+
"step": 135
|
| 1900 |
+
},
|
| 1901 |
+
{
|
| 1902 |
+
"epoch": 0.15,
|
| 1903 |
+
"learning_rate": 4.03448275862069e-05,
|
| 1904 |
+
"logits/chosen": -0.39377278089523315,
|
| 1905 |
+
"logits/rejected": -1.5313684940338135,
|
| 1906 |
+
"logps/chosen": -441.0325622558594,
|
| 1907 |
+
"logps/rejected": -223.12750244140625,
|
| 1908 |
+
"loss": 0.0,
|
| 1909 |
+
"rewards/accuracies": 1.0,
|
| 1910 |
+
"rewards/chosen": 12.754582405090332,
|
| 1911 |
+
"rewards/margins": 28.21717071533203,
|
| 1912 |
+
"rewards/rejected": -15.462587356567383,
|
| 1913 |
+
"step": 136
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 0.15,
|
| 1917 |
+
"learning_rate": 4.029310344827587e-05,
|
| 1918 |
+
"logits/chosen": -0.45350462198257446,
|
| 1919 |
+
"logits/rejected": -1.3594186305999756,
|
| 1920 |
+
"logps/chosen": -313.7057189941406,
|
| 1921 |
+
"logps/rejected": -304.0419616699219,
|
| 1922 |
+
"loss": 0.0,
|
| 1923 |
+
"rewards/accuracies": 1.0,
|
| 1924 |
+
"rewards/chosen": 6.3996429443359375,
|
| 1925 |
+
"rewards/margins": 23.64727020263672,
|
| 1926 |
+
"rewards/rejected": -17.24762725830078,
|
| 1927 |
+
"step": 137
|
| 1928 |
+
},
|
| 1929 |
+
{
|
| 1930 |
+
"epoch": 0.15,
|
| 1931 |
+
"learning_rate": 4.024137931034483e-05,
|
| 1932 |
+
"logits/chosen": -0.17602750658988953,
|
| 1933 |
+
"logits/rejected": -1.2745662927627563,
|
| 1934 |
+
"logps/chosen": -371.0513916015625,
|
| 1935 |
+
"logps/rejected": -299.74481201171875,
|
| 1936 |
+
"loss": 0.0,
|
| 1937 |
+
"rewards/accuracies": 1.0,
|
| 1938 |
+
"rewards/chosen": 7.569783687591553,
|
| 1939 |
+
"rewards/margins": 26.724822998046875,
|
| 1940 |
+
"rewards/rejected": -19.155040740966797,
|
| 1941 |
+
"step": 138
|
| 1942 |
+
},
|
| 1943 |
+
{
|
| 1944 |
+
"epoch": 0.15,
|
| 1945 |
+
"learning_rate": 4.0189655172413796e-05,
|
| 1946 |
+
"logits/chosen": -0.44520992040634155,
|
| 1947 |
+
"logits/rejected": -1.4310128688812256,
|
| 1948 |
+
"logps/chosen": -334.8761901855469,
|
| 1949 |
+
"logps/rejected": -278.89349365234375,
|
| 1950 |
+
"loss": 0.0,
|
| 1951 |
+
"rewards/accuracies": 1.0,
|
| 1952 |
+
"rewards/chosen": 3.9243433475494385,
|
| 1953 |
+
"rewards/margins": 22.313913345336914,
|
| 1954 |
+
"rewards/rejected": -18.389568328857422,
|
| 1955 |
+
"step": 139
|
| 1956 |
+
},
|
| 1957 |
+
{
|
| 1958 |
+
"epoch": 0.15,
|
| 1959 |
+
"learning_rate": 4.0137931034482764e-05,
|
| 1960 |
+
"logits/chosen": -0.3496861755847931,
|
| 1961 |
+
"logits/rejected": -1.480123519897461,
|
| 1962 |
+
"logps/chosen": -345.8548583984375,
|
| 1963 |
+
"logps/rejected": -295.12042236328125,
|
| 1964 |
+
"loss": 0.0,
|
| 1965 |
+
"rewards/accuracies": 1.0,
|
| 1966 |
+
"rewards/chosen": 9.938793182373047,
|
| 1967 |
+
"rewards/margins": 28.614458084106445,
|
| 1968 |
+
"rewards/rejected": -18.67566680908203,
|
| 1969 |
+
"step": 140
|
| 1970 |
+
},
|
| 1971 |
+
{
|
| 1972 |
+
"epoch": 0.15,
|
| 1973 |
+
"learning_rate": 4.0086206896551726e-05,
|
| 1974 |
+
"logits/chosen": -0.4202902913093567,
|
| 1975 |
+
"logits/rejected": -1.5481184720993042,
|
| 1976 |
+
"logps/chosen": -408.18121337890625,
|
| 1977 |
+
"logps/rejected": -273.8868408203125,
|
| 1978 |
+
"loss": 0.0,
|
| 1979 |
+
"rewards/accuracies": 1.0,
|
| 1980 |
+
"rewards/chosen": 9.119111061096191,
|
| 1981 |
+
"rewards/margins": 29.09032440185547,
|
| 1982 |
+
"rewards/rejected": -19.971214294433594,
|
| 1983 |
+
"step": 141
|
| 1984 |
+
},
|
| 1985 |
+
{
|
| 1986 |
+
"epoch": 0.15,
|
| 1987 |
+
"learning_rate": 4.0034482758620694e-05,
|
| 1988 |
+
"logits/chosen": -0.5120177268981934,
|
| 1989 |
+
"logits/rejected": -1.4507927894592285,
|
| 1990 |
+
"logps/chosen": -271.6726989746094,
|
| 1991 |
+
"logps/rejected": -297.4678955078125,
|
| 1992 |
+
"loss": 0.0,
|
| 1993 |
+
"rewards/accuracies": 1.0,
|
| 1994 |
+
"rewards/chosen": 5.688560962677002,
|
| 1995 |
+
"rewards/margins": 26.492656707763672,
|
| 1996 |
+
"rewards/rejected": -20.804094314575195,
|
| 1997 |
+
"step": 142
|
| 1998 |
+
},
|
| 1999 |
+
{
|
| 2000 |
+
"epoch": 0.16,
|
| 2001 |
+
"learning_rate": 3.998275862068966e-05,
|
| 2002 |
+
"logits/chosen": -0.41148996353149414,
|
| 2003 |
+
"logits/rejected": -1.4985872507095337,
|
| 2004 |
+
"logps/chosen": -343.8206481933594,
|
| 2005 |
+
"logps/rejected": -241.0497589111328,
|
| 2006 |
+
"loss": 0.0,
|
| 2007 |
+
"rewards/accuracies": 1.0,
|
| 2008 |
+
"rewards/chosen": 9.314858436584473,
|
| 2009 |
+
"rewards/margins": 26.18862533569336,
|
| 2010 |
+
"rewards/rejected": -16.87376594543457,
|
| 2011 |
+
"step": 143
|
| 2012 |
+
},
|
| 2013 |
+
{
|
| 2014 |
+
"epoch": 0.16,
|
| 2015 |
+
"learning_rate": 3.993103448275862e-05,
|
| 2016 |
+
"logits/chosen": -0.1992441862821579,
|
| 2017 |
+
"logits/rejected": -1.8458625078201294,
|
| 2018 |
+
"logps/chosen": -504.0348205566406,
|
| 2019 |
+
"logps/rejected": -318.8588562011719,
|
| 2020 |
+
"loss": 0.0,
|
| 2021 |
+
"rewards/accuracies": 1.0,
|
| 2022 |
+
"rewards/chosen": 12.68671989440918,
|
| 2023 |
+
"rewards/margins": 34.4812126159668,
|
| 2024 |
+
"rewards/rejected": -21.79449462890625,
|
| 2025 |
+
"step": 144
|
| 2026 |
+
},
|
| 2027 |
+
{
|
| 2028 |
+
"epoch": 0.16,
|
| 2029 |
+
"learning_rate": 3.987931034482759e-05,
|
| 2030 |
+
"logits/chosen": -0.5546131730079651,
|
| 2031 |
+
"logits/rejected": -1.527896761894226,
|
| 2032 |
+
"logps/chosen": -333.9448547363281,
|
| 2033 |
+
"logps/rejected": -303.9145202636719,
|
| 2034 |
+
"loss": 0.0,
|
| 2035 |
+
"rewards/accuracies": 1.0,
|
| 2036 |
+
"rewards/chosen": 10.168367385864258,
|
| 2037 |
+
"rewards/margins": 30.0524959564209,
|
| 2038 |
+
"rewards/rejected": -19.88412857055664,
|
| 2039 |
+
"step": 145
|
| 2040 |
+
},
|
| 2041 |
+
{
|
| 2042 |
+
"epoch": 0.16,
|
| 2043 |
+
"learning_rate": 3.982758620689656e-05,
|
| 2044 |
+
"logits/chosen": -0.3886173367500305,
|
| 2045 |
+
"logits/rejected": -1.621660828590393,
|
| 2046 |
+
"logps/chosen": -369.8122253417969,
|
| 2047 |
+
"logps/rejected": -272.4569396972656,
|
| 2048 |
+
"loss": 0.0016,
|
| 2049 |
+
"rewards/accuracies": 1.0,
|
| 2050 |
+
"rewards/chosen": 8.674747467041016,
|
| 2051 |
+
"rewards/margins": 26.678119659423828,
|
| 2052 |
+
"rewards/rejected": -18.003372192382812,
|
| 2053 |
+
"step": 146
|
| 2054 |
+
},
|
| 2055 |
+
{
|
| 2056 |
+
"epoch": 0.16,
|
| 2057 |
+
"learning_rate": 3.977586206896552e-05,
|
| 2058 |
+
"logits/chosen": -0.61821049451828,
|
| 2059 |
+
"logits/rejected": -1.6792049407958984,
|
| 2060 |
+
"logps/chosen": -345.2456359863281,
|
| 2061 |
+
"logps/rejected": -307.6291809082031,
|
| 2062 |
+
"loss": 0.0,
|
| 2063 |
+
"rewards/accuracies": 1.0,
|
| 2064 |
+
"rewards/chosen": 7.595407485961914,
|
| 2065 |
+
"rewards/margins": 29.619413375854492,
|
| 2066 |
+
"rewards/rejected": -22.024009704589844,
|
| 2067 |
+
"step": 147
|
| 2068 |
+
},
|
| 2069 |
+
{
|
| 2070 |
+
"epoch": 0.16,
|
| 2071 |
+
"learning_rate": 3.972413793103449e-05,
|
| 2072 |
+
"logits/chosen": -0.23747998476028442,
|
| 2073 |
+
"logits/rejected": -1.287044644355774,
|
| 2074 |
+
"logps/chosen": -361.4759521484375,
|
| 2075 |
+
"logps/rejected": -267.63592529296875,
|
| 2076 |
+
"loss": 0.2105,
|
| 2077 |
+
"rewards/accuracies": 0.9375,
|
| 2078 |
+
"rewards/chosen": 7.393239974975586,
|
| 2079 |
+
"rewards/margins": 23.982959747314453,
|
| 2080 |
+
"rewards/rejected": -16.5897216796875,
|
| 2081 |
+
"step": 148
|
| 2082 |
+
},
|
| 2083 |
+
{
|
| 2084 |
+
"epoch": 0.16,
|
| 2085 |
+
"learning_rate": 3.967241379310345e-05,
|
| 2086 |
+
"logits/chosen": -0.3899969160556793,
|
| 2087 |
+
"logits/rejected": -1.6496773958206177,
|
| 2088 |
+
"logps/chosen": -331.9876403808594,
|
| 2089 |
+
"logps/rejected": -278.04522705078125,
|
| 2090 |
+
"loss": 0.0,
|
| 2091 |
+
"rewards/accuracies": 1.0,
|
| 2092 |
+
"rewards/chosen": 8.100687026977539,
|
| 2093 |
+
"rewards/margins": 29.64002227783203,
|
| 2094 |
+
"rewards/rejected": -21.539335250854492,
|
| 2095 |
+
"step": 149
|
| 2096 |
+
},
|
| 2097 |
+
{
|
| 2098 |
+
"epoch": 0.16,
|
| 2099 |
+
"learning_rate": 3.962068965517242e-05,
|
| 2100 |
+
"logits/chosen": -0.4621184766292572,
|
| 2101 |
+
"logits/rejected": -1.3823509216308594,
|
| 2102 |
+
"logps/chosen": -291.25494384765625,
|
| 2103 |
+
"logps/rejected": -313.21734619140625,
|
| 2104 |
+
"loss": 0.0,
|
| 2105 |
+
"rewards/accuracies": 1.0,
|
| 2106 |
+
"rewards/chosen": 4.036273002624512,
|
| 2107 |
+
"rewards/margins": 25.113258361816406,
|
| 2108 |
+
"rewards/rejected": -21.076988220214844,
|
| 2109 |
+
"step": 150
|
| 2110 |
+
},
|
| 2111 |
+
{
|
| 2112 |
+
"epoch": 0.16,
|
| 2113 |
+
"learning_rate": 3.956896551724138e-05,
|
| 2114 |
+
"logits/chosen": -0.3050623834133148,
|
| 2115 |
+
"logits/rejected": -1.3223061561584473,
|
| 2116 |
+
"logps/chosen": -388.243896484375,
|
| 2117 |
+
"logps/rejected": -266.01983642578125,
|
| 2118 |
+
"loss": 0.0,
|
| 2119 |
+
"rewards/accuracies": 1.0,
|
| 2120 |
+
"rewards/chosen": 7.008649826049805,
|
| 2121 |
+
"rewards/margins": 24.918262481689453,
|
| 2122 |
+
"rewards/rejected": -17.90961265563965,
|
| 2123 |
+
"step": 151
|
| 2124 |
+
},
|
| 2125 |
+
{
|
| 2126 |
+
"epoch": 0.17,
|
| 2127 |
+
"learning_rate": 3.9517241379310346e-05,
|
| 2128 |
+
"logits/chosen": -0.26263201236724854,
|
| 2129 |
+
"logits/rejected": -1.262205958366394,
|
| 2130 |
+
"logps/chosen": -300.5107421875,
|
| 2131 |
+
"logps/rejected": -266.810302734375,
|
| 2132 |
+
"loss": 0.0007,
|
| 2133 |
+
"rewards/accuracies": 1.0,
|
| 2134 |
+
"rewards/chosen": 5.9579243659973145,
|
| 2135 |
+
"rewards/margins": 23.850536346435547,
|
| 2136 |
+
"rewards/rejected": -17.892614364624023,
|
| 2137 |
+
"step": 152
|
| 2138 |
+
},
|
| 2139 |
+
{
|
| 2140 |
+
"epoch": 0.17,
|
| 2141 |
+
"learning_rate": 3.9465517241379314e-05,
|
| 2142 |
+
"logits/chosen": -0.34424301981925964,
|
| 2143 |
+
"logits/rejected": -1.2062197923660278,
|
| 2144 |
+
"logps/chosen": -273.7149353027344,
|
| 2145 |
+
"logps/rejected": -266.54150390625,
|
| 2146 |
+
"loss": 0.0,
|
| 2147 |
+
"rewards/accuracies": 1.0,
|
| 2148 |
+
"rewards/chosen": 5.983701705932617,
|
| 2149 |
+
"rewards/margins": 26.13591957092285,
|
| 2150 |
+
"rewards/rejected": -20.1522216796875,
|
| 2151 |
+
"step": 153
|
| 2152 |
+
},
|
| 2153 |
+
{
|
| 2154 |
+
"epoch": 0.17,
|
| 2155 |
+
"learning_rate": 3.9413793103448276e-05,
|
| 2156 |
+
"logits/chosen": -0.1809622049331665,
|
| 2157 |
+
"logits/rejected": -1.4350645542144775,
|
| 2158 |
+
"logps/chosen": -418.6148376464844,
|
| 2159 |
+
"logps/rejected": -297.8706359863281,
|
| 2160 |
+
"loss": 0.0,
|
| 2161 |
+
"rewards/accuracies": 1.0,
|
| 2162 |
+
"rewards/chosen": 8.36530876159668,
|
| 2163 |
+
"rewards/margins": 30.20969581604004,
|
| 2164 |
+
"rewards/rejected": -21.84438705444336,
|
| 2165 |
+
"step": 154
|
| 2166 |
+
},
|
| 2167 |
+
{
|
| 2168 |
+
"epoch": 0.17,
|
| 2169 |
+
"learning_rate": 3.9362068965517244e-05,
|
| 2170 |
+
"logits/chosen": -0.3474291265010834,
|
| 2171 |
+
"logits/rejected": -1.3261463642120361,
|
| 2172 |
+
"logps/chosen": -392.76068115234375,
|
| 2173 |
+
"logps/rejected": -282.137451171875,
|
| 2174 |
+
"loss": 0.12,
|
| 2175 |
+
"rewards/accuracies": 0.9375,
|
| 2176 |
+
"rewards/chosen": 8.480328559875488,
|
| 2177 |
+
"rewards/margins": 24.708953857421875,
|
| 2178 |
+
"rewards/rejected": -16.228626251220703,
|
| 2179 |
+
"step": 155
|
| 2180 |
+
},
|
| 2181 |
+
{
|
| 2182 |
+
"epoch": 0.17,
|
| 2183 |
+
"learning_rate": 3.931034482758621e-05,
|
| 2184 |
+
"logits/chosen": -0.29881787300109863,
|
| 2185 |
+
"logits/rejected": -1.59579598903656,
|
| 2186 |
+
"logps/chosen": -491.68280029296875,
|
| 2187 |
+
"logps/rejected": -329.0652160644531,
|
| 2188 |
+
"loss": 0.0,
|
| 2189 |
+
"rewards/accuracies": 1.0,
|
| 2190 |
+
"rewards/chosen": 9.562919616699219,
|
| 2191 |
+
"rewards/margins": 31.969928741455078,
|
| 2192 |
+
"rewards/rejected": -22.407007217407227,
|
| 2193 |
+
"step": 156
|
| 2194 |
+
},
|
| 2195 |
+
{
|
| 2196 |
+
"epoch": 0.17,
|
| 2197 |
+
"learning_rate": 3.925862068965517e-05,
|
| 2198 |
+
"logits/chosen": -0.24624793231487274,
|
| 2199 |
+
"logits/rejected": -1.2760369777679443,
|
| 2200 |
+
"logps/chosen": -309.42950439453125,
|
| 2201 |
+
"logps/rejected": -337.400634765625,
|
| 2202 |
+
"loss": 0.0,
|
| 2203 |
+
"rewards/accuracies": 1.0,
|
| 2204 |
+
"rewards/chosen": 5.126012802124023,
|
| 2205 |
+
"rewards/margins": 27.284088134765625,
|
| 2206 |
+
"rewards/rejected": -22.1580753326416,
|
| 2207 |
+
"step": 157
|
| 2208 |
+
},
|
| 2209 |
+
{
|
| 2210 |
+
"epoch": 0.17,
|
| 2211 |
+
"learning_rate": 3.920689655172414e-05,
|
| 2212 |
+
"logits/chosen": -0.24486477673053741,
|
| 2213 |
+
"logits/rejected": -1.4145996570587158,
|
| 2214 |
+
"logps/chosen": -435.2896423339844,
|
| 2215 |
+
"logps/rejected": -347.84698486328125,
|
| 2216 |
+
"loss": 0.0,
|
| 2217 |
+
"rewards/accuracies": 1.0,
|
| 2218 |
+
"rewards/chosen": 3.421405076980591,
|
| 2219 |
+
"rewards/margins": 30.74762725830078,
|
| 2220 |
+
"rewards/rejected": -27.32622528076172,
|
| 2221 |
+
"step": 158
|
| 2222 |
+
},
|
| 2223 |
+
{
|
| 2224 |
+
"epoch": 0.17,
|
| 2225 |
+
"learning_rate": 3.915517241379311e-05,
|
| 2226 |
+
"logits/chosen": -0.4706140458583832,
|
| 2227 |
+
"logits/rejected": -1.4215468168258667,
|
| 2228 |
+
"logps/chosen": -427.6326904296875,
|
| 2229 |
+
"logps/rejected": -399.431640625,
|
| 2230 |
+
"loss": 0.0,
|
| 2231 |
+
"rewards/accuracies": 1.0,
|
| 2232 |
+
"rewards/chosen": 4.1375885009765625,
|
| 2233 |
+
"rewards/margins": 31.478370666503906,
|
| 2234 |
+
"rewards/rejected": -27.340782165527344,
|
| 2235 |
+
"step": 159
|
| 2236 |
+
},
|
| 2237 |
+
{
|
| 2238 |
+
"epoch": 0.17,
|
| 2239 |
+
"learning_rate": 3.910344827586207e-05,
|
| 2240 |
+
"logits/chosen": -0.25800830125808716,
|
| 2241 |
+
"logits/rejected": -1.604130744934082,
|
| 2242 |
+
"logps/chosen": -430.5594482421875,
|
| 2243 |
+
"logps/rejected": -431.40301513671875,
|
| 2244 |
+
"loss": 0.0,
|
| 2245 |
+
"rewards/accuracies": 1.0,
|
| 2246 |
+
"rewards/chosen": 1.8110466003417969,
|
| 2247 |
+
"rewards/margins": 36.45166015625,
|
| 2248 |
+
"rewards/rejected": -34.64060974121094,
|
| 2249 |
+
"step": 160
|
| 2250 |
+
},
|
| 2251 |
+
{
|
| 2252 |
+
"epoch": 0.18,
|
| 2253 |
+
"learning_rate": 3.905172413793104e-05,
|
| 2254 |
+
"logits/chosen": -0.3043404221534729,
|
| 2255 |
+
"logits/rejected": -1.5278654098510742,
|
| 2256 |
+
"logps/chosen": -465.7344970703125,
|
| 2257 |
+
"logps/rejected": -436.65899658203125,
|
| 2258 |
+
"loss": 0.0,
|
| 2259 |
+
"rewards/accuracies": 1.0,
|
| 2260 |
+
"rewards/chosen": -1.1547170877456665,
|
| 2261 |
+
"rewards/margins": 33.50754928588867,
|
| 2262 |
+
"rewards/rejected": -34.66226577758789,
|
| 2263 |
+
"step": 161
|
| 2264 |
+
},
|
| 2265 |
+
{
|
| 2266 |
+
"epoch": 0.18,
|
| 2267 |
+
"learning_rate": 3.9000000000000006e-05,
|
| 2268 |
+
"logits/chosen": -0.22293318808078766,
|
| 2269 |
+
"logits/rejected": -1.2031610012054443,
|
| 2270 |
+
"logps/chosen": -428.4457092285156,
|
| 2271 |
+
"logps/rejected": -419.1228332519531,
|
| 2272 |
+
"loss": 0.0,
|
| 2273 |
+
"rewards/accuracies": 1.0,
|
| 2274 |
+
"rewards/chosen": -3.0903587341308594,
|
| 2275 |
+
"rewards/margins": 28.546253204345703,
|
| 2276 |
+
"rewards/rejected": -31.636613845825195,
|
| 2277 |
+
"step": 162
|
| 2278 |
+
},
|
| 2279 |
+
{
|
| 2280 |
+
"epoch": 0.18,
|
| 2281 |
+
"learning_rate": 3.894827586206897e-05,
|
| 2282 |
+
"logits/chosen": -0.18892326951026917,
|
| 2283 |
+
"logits/rejected": -1.481589913368225,
|
| 2284 |
+
"logps/chosen": -513.7359008789062,
|
| 2285 |
+
"logps/rejected": -456.95208740234375,
|
| 2286 |
+
"loss": 0.0,
|
| 2287 |
+
"rewards/accuracies": 1.0,
|
| 2288 |
+
"rewards/chosen": 2.6842398643493652,
|
| 2289 |
+
"rewards/margins": 39.1934928894043,
|
| 2290 |
+
"rewards/rejected": -36.509254455566406,
|
| 2291 |
+
"step": 163
|
| 2292 |
+
},
|
| 2293 |
+
{
|
| 2294 |
+
"epoch": 0.18,
|
| 2295 |
+
"learning_rate": 3.8896551724137935e-05,
|
| 2296 |
+
"logits/chosen": -0.48873329162597656,
|
| 2297 |
+
"logits/rejected": -1.4820678234100342,
|
| 2298 |
+
"logps/chosen": -393.00079345703125,
|
| 2299 |
+
"logps/rejected": -426.32696533203125,
|
| 2300 |
+
"loss": 0.0194,
|
| 2301 |
+
"rewards/accuracies": 1.0,
|
| 2302 |
+
"rewards/chosen": -5.972713470458984,
|
| 2303 |
+
"rewards/margins": 28.33656120300293,
|
| 2304 |
+
"rewards/rejected": -34.30927276611328,
|
| 2305 |
+
"step": 164
|
| 2306 |
+
},
|
| 2307 |
+
{
|
| 2308 |
+
"epoch": 0.18,
|
| 2309 |
+
"learning_rate": 3.88448275862069e-05,
|
| 2310 |
+
"logits/chosen": -0.4431387484073639,
|
| 2311 |
+
"logits/rejected": -1.3382742404937744,
|
| 2312 |
+
"logps/chosen": -392.2378845214844,
|
| 2313 |
+
"logps/rejected": -503.30194091796875,
|
| 2314 |
+
"loss": 0.0,
|
| 2315 |
+
"rewards/accuracies": 1.0,
|
| 2316 |
+
"rewards/chosen": -5.949856758117676,
|
| 2317 |
+
"rewards/margins": 34.904537200927734,
|
| 2318 |
+
"rewards/rejected": -40.854393005371094,
|
| 2319 |
+
"step": 165
|
| 2320 |
+
},
|
| 2321 |
+
{
|
| 2322 |
+
"epoch": 0.18,
|
| 2323 |
+
"learning_rate": 3.8793103448275865e-05,
|
| 2324 |
+
"logits/chosen": -0.2996464669704437,
|
| 2325 |
+
"logits/rejected": -1.4847010374069214,
|
| 2326 |
+
"logps/chosen": -483.34161376953125,
|
| 2327 |
+
"logps/rejected": -548.205078125,
|
| 2328 |
+
"loss": 0.0,
|
| 2329 |
+
"rewards/accuracies": 1.0,
|
| 2330 |
+
"rewards/chosen": 3.7714486122131348,
|
| 2331 |
+
"rewards/margins": 47.965972900390625,
|
| 2332 |
+
"rewards/rejected": -44.19452667236328,
|
| 2333 |
+
"step": 166
|
| 2334 |
+
},
|
| 2335 |
+
{
|
| 2336 |
+
"epoch": 0.18,
|
| 2337 |
+
"learning_rate": 3.874137931034483e-05,
|
| 2338 |
+
"logits/chosen": -0.3217467963695526,
|
| 2339 |
+
"logits/rejected": -1.2237169742584229,
|
| 2340 |
+
"logps/chosen": -396.9007568359375,
|
| 2341 |
+
"logps/rejected": -509.6751708984375,
|
| 2342 |
+
"loss": 0.0,
|
| 2343 |
+
"rewards/accuracies": 1.0,
|
| 2344 |
+
"rewards/chosen": 0.9992933869361877,
|
| 2345 |
+
"rewards/margins": 39.428592681884766,
|
| 2346 |
+
"rewards/rejected": -38.429298400878906,
|
| 2347 |
+
"step": 167
|
| 2348 |
+
},
|
| 2349 |
+
{
|
| 2350 |
+
"epoch": 0.18,
|
| 2351 |
+
"learning_rate": 3.8689655172413794e-05,
|
| 2352 |
+
"logits/chosen": -0.26858553290367126,
|
| 2353 |
+
"logits/rejected": -1.3162765502929688,
|
| 2354 |
+
"logps/chosen": -436.47088623046875,
|
| 2355 |
+
"logps/rejected": -461.6593933105469,
|
| 2356 |
+
"loss": 0.0,
|
| 2357 |
+
"rewards/accuracies": 1.0,
|
| 2358 |
+
"rewards/chosen": 2.6044464111328125,
|
| 2359 |
+
"rewards/margins": 38.4824104309082,
|
| 2360 |
+
"rewards/rejected": -35.87796401977539,
|
| 2361 |
+
"step": 168
|
| 2362 |
+
},
|
| 2363 |
+
{
|
| 2364 |
+
"epoch": 0.18,
|
| 2365 |
+
"learning_rate": 3.863793103448276e-05,
|
| 2366 |
+
"logits/chosen": -0.343127965927124,
|
| 2367 |
+
"logits/rejected": -1.390782117843628,
|
| 2368 |
+
"logps/chosen": -392.77978515625,
|
| 2369 |
+
"logps/rejected": -443.3890686035156,
|
| 2370 |
+
"loss": 0.0,
|
| 2371 |
+
"rewards/accuracies": 1.0,
|
| 2372 |
+
"rewards/chosen": 0.19043821096420288,
|
| 2373 |
+
"rewards/margins": 34.40040969848633,
|
| 2374 |
+
"rewards/rejected": -34.20996856689453,
|
| 2375 |
+
"step": 169
|
| 2376 |
+
},
|
| 2377 |
+
{
|
| 2378 |
+
"epoch": 0.19,
|
| 2379 |
+
"learning_rate": 3.858620689655172e-05,
|
| 2380 |
+
"logits/chosen": -0.43945077061653137,
|
| 2381 |
+
"logits/rejected": -1.3458408117294312,
|
| 2382 |
+
"logps/chosen": -386.0972595214844,
|
| 2383 |
+
"logps/rejected": -481.8536071777344,
|
| 2384 |
+
"loss": 0.0,
|
| 2385 |
+
"rewards/accuracies": 1.0,
|
| 2386 |
+
"rewards/chosen": 1.2730538845062256,
|
| 2387 |
+
"rewards/margins": 35.0412483215332,
|
| 2388 |
+
"rewards/rejected": -33.76819610595703,
|
| 2389 |
+
"step": 170
|
| 2390 |
+
},
|
| 2391 |
+
{
|
| 2392 |
+
"epoch": 0.19,
|
| 2393 |
+
"learning_rate": 3.853448275862069e-05,
|
| 2394 |
+
"logits/chosen": -0.28766676783561707,
|
| 2395 |
+
"logits/rejected": -1.314267873764038,
|
| 2396 |
+
"logps/chosen": -423.24981689453125,
|
| 2397 |
+
"logps/rejected": -416.8682861328125,
|
| 2398 |
+
"loss": 0.0,
|
| 2399 |
+
"rewards/accuracies": 1.0,
|
| 2400 |
+
"rewards/chosen": 1.9727609157562256,
|
| 2401 |
+
"rewards/margins": 34.98452377319336,
|
| 2402 |
+
"rewards/rejected": -33.01176834106445,
|
| 2403 |
+
"step": 171
|
| 2404 |
+
},
|
| 2405 |
+
{
|
| 2406 |
+
"epoch": 0.19,
|
| 2407 |
+
"learning_rate": 3.848275862068966e-05,
|
| 2408 |
+
"logits/chosen": -0.3893183767795563,
|
| 2409 |
+
"logits/rejected": -1.443713903427124,
|
| 2410 |
+
"logps/chosen": -474.935546875,
|
| 2411 |
+
"logps/rejected": -453.4057922363281,
|
| 2412 |
+
"loss": 0.0,
|
| 2413 |
+
"rewards/accuracies": 1.0,
|
| 2414 |
+
"rewards/chosen": 2.1096689701080322,
|
| 2415 |
+
"rewards/margins": 38.80006790161133,
|
| 2416 |
+
"rewards/rejected": -36.690399169921875,
|
| 2417 |
+
"step": 172
|
| 2418 |
+
},
|
| 2419 |
+
{
|
| 2420 |
+
"epoch": 0.19,
|
| 2421 |
+
"learning_rate": 3.843103448275862e-05,
|
| 2422 |
+
"logits/chosen": -0.34813588857650757,
|
| 2423 |
+
"logits/rejected": -1.271755337715149,
|
| 2424 |
+
"logps/chosen": -392.284423828125,
|
| 2425 |
+
"logps/rejected": -496.442138671875,
|
| 2426 |
+
"loss": 0.0,
|
| 2427 |
+
"rewards/accuracies": 1.0,
|
| 2428 |
+
"rewards/chosen": 2.8468620777130127,
|
| 2429 |
+
"rewards/margins": 41.135921478271484,
|
| 2430 |
+
"rewards/rejected": -38.289058685302734,
|
| 2431 |
+
"step": 173
|
| 2432 |
+
},
|
| 2433 |
+
{
|
| 2434 |
+
"epoch": 0.19,
|
| 2435 |
+
"learning_rate": 3.837931034482759e-05,
|
| 2436 |
+
"logits/chosen": -0.3464352786540985,
|
| 2437 |
+
"logits/rejected": -1.223905324935913,
|
| 2438 |
+
"logps/chosen": -372.1242370605469,
|
| 2439 |
+
"logps/rejected": -446.0730285644531,
|
| 2440 |
+
"loss": 0.0073,
|
| 2441 |
+
"rewards/accuracies": 1.0,
|
| 2442 |
+
"rewards/chosen": 0.18368953466415405,
|
| 2443 |
+
"rewards/margins": 33.218013763427734,
|
| 2444 |
+
"rewards/rejected": -33.034324645996094,
|
| 2445 |
+
"step": 174
|
| 2446 |
+
},
|
| 2447 |
+
{
|
| 2448 |
+
"epoch": 0.19,
|
| 2449 |
+
"learning_rate": 3.8327586206896556e-05,
|
| 2450 |
+
"logits/chosen": -0.42792245745658875,
|
| 2451 |
+
"logits/rejected": -1.189011812210083,
|
| 2452 |
+
"logps/chosen": -324.2301940917969,
|
| 2453 |
+
"logps/rejected": -503.5957946777344,
|
| 2454 |
+
"loss": 0.0,
|
| 2455 |
+
"rewards/accuracies": 1.0,
|
| 2456 |
+
"rewards/chosen": -1.7187633514404297,
|
| 2457 |
+
"rewards/margins": 37.19697189331055,
|
| 2458 |
+
"rewards/rejected": -38.91573715209961,
|
| 2459 |
+
"step": 175
|
| 2460 |
+
},
|
| 2461 |
+
{
|
| 2462 |
+
"epoch": 0.19,
|
| 2463 |
+
"learning_rate": 3.827586206896552e-05,
|
| 2464 |
+
"logits/chosen": -0.16668730974197388,
|
| 2465 |
+
"logits/rejected": -1.5985019207000732,
|
| 2466 |
+
"logps/chosen": -482.8082580566406,
|
| 2467 |
+
"logps/rejected": -449.761474609375,
|
| 2468 |
+
"loss": 0.0,
|
| 2469 |
+
"rewards/accuracies": 1.0,
|
| 2470 |
+
"rewards/chosen": 3.7022485733032227,
|
| 2471 |
+
"rewards/margins": 42.01506805419922,
|
| 2472 |
+
"rewards/rejected": -38.31282043457031,
|
| 2473 |
+
"step": 176
|
| 2474 |
+
},
|
| 2475 |
+
{
|
| 2476 |
+
"epoch": 0.19,
|
| 2477 |
+
"learning_rate": 3.8224137931034485e-05,
|
| 2478 |
+
"logits/chosen": -0.26949506998062134,
|
| 2479 |
+
"logits/rejected": -1.2986763715744019,
|
| 2480 |
+
"logps/chosen": -348.618896484375,
|
| 2481 |
+
"logps/rejected": -442.8262023925781,
|
| 2482 |
+
"loss": 0.0,
|
| 2483 |
+
"rewards/accuracies": 1.0,
|
| 2484 |
+
"rewards/chosen": 2.1358022689819336,
|
| 2485 |
+
"rewards/margins": 35.771461486816406,
|
| 2486 |
+
"rewards/rejected": -33.63566207885742,
|
| 2487 |
+
"step": 177
|
| 2488 |
+
},
|
| 2489 |
+
{
|
| 2490 |
+
"epoch": 0.19,
|
| 2491 |
+
"learning_rate": 3.8172413793103453e-05,
|
| 2492 |
+
"logits/chosen": -0.3438015580177307,
|
| 2493 |
+
"logits/rejected": -1.3001600503921509,
|
| 2494 |
+
"logps/chosen": -333.9195251464844,
|
| 2495 |
+
"logps/rejected": -457.9849853515625,
|
| 2496 |
+
"loss": 0.0,
|
| 2497 |
+
"rewards/accuracies": 1.0,
|
| 2498 |
+
"rewards/chosen": 2.145051956176758,
|
| 2499 |
+
"rewards/margins": 41.207923889160156,
|
| 2500 |
+
"rewards/rejected": -39.06287384033203,
|
| 2501 |
+
"step": 178
|
| 2502 |
+
},
|
| 2503 |
+
{
|
| 2504 |
+
"epoch": 0.2,
|
| 2505 |
+
"learning_rate": 3.8120689655172415e-05,
|
| 2506 |
+
"logits/chosen": -0.10503964871168137,
|
| 2507 |
+
"logits/rejected": -1.2384865283966064,
|
| 2508 |
+
"logps/chosen": -451.48663330078125,
|
| 2509 |
+
"logps/rejected": -479.8037414550781,
|
| 2510 |
+
"loss": 0.0,
|
| 2511 |
+
"rewards/accuracies": 1.0,
|
| 2512 |
+
"rewards/chosen": 0.525763988494873,
|
| 2513 |
+
"rewards/margins": 38.3972282409668,
|
| 2514 |
+
"rewards/rejected": -37.8714599609375,
|
| 2515 |
+
"step": 179
|
| 2516 |
+
},
|
| 2517 |
+
{
|
| 2518 |
+
"epoch": 0.2,
|
| 2519 |
+
"learning_rate": 3.806896551724138e-05,
|
| 2520 |
+
"logits/chosen": -0.30969691276550293,
|
| 2521 |
+
"logits/rejected": -1.1278965473175049,
|
| 2522 |
+
"logps/chosen": -298.7955322265625,
|
| 2523 |
+
"logps/rejected": -499.42279052734375,
|
| 2524 |
+
"loss": 0.0,
|
| 2525 |
+
"rewards/accuracies": 1.0,
|
| 2526 |
+
"rewards/chosen": -2.1820168495178223,
|
| 2527 |
+
"rewards/margins": 37.28181457519531,
|
| 2528 |
+
"rewards/rejected": -39.463829040527344,
|
| 2529 |
+
"step": 180
|
| 2530 |
+
},
|
| 2531 |
+
{
|
| 2532 |
+
"epoch": 0.2,
|
| 2533 |
+
"learning_rate": 3.801724137931035e-05,
|
| 2534 |
+
"logits/chosen": -0.3709387183189392,
|
| 2535 |
+
"logits/rejected": -1.5942573547363281,
|
| 2536 |
+
"logps/chosen": -422.3631286621094,
|
| 2537 |
+
"logps/rejected": -511.52008056640625,
|
| 2538 |
+
"loss": 0.0,
|
| 2539 |
+
"rewards/accuracies": 1.0,
|
| 2540 |
+
"rewards/chosen": 2.8516347408294678,
|
| 2541 |
+
"rewards/margins": 41.39206314086914,
|
| 2542 |
+
"rewards/rejected": -38.540428161621094,
|
| 2543 |
+
"step": 181
|
| 2544 |
+
},
|
| 2545 |
+
{
|
| 2546 |
+
"epoch": 0.2,
|
| 2547 |
+
"learning_rate": 3.796551724137931e-05,
|
| 2548 |
+
"logits/chosen": -0.2693447470664978,
|
| 2549 |
+
"logits/rejected": -1.5490005016326904,
|
| 2550 |
+
"logps/chosen": -480.881591796875,
|
| 2551 |
+
"logps/rejected": -414.71917724609375,
|
| 2552 |
+
"loss": 0.0,
|
| 2553 |
+
"rewards/accuracies": 1.0,
|
| 2554 |
+
"rewards/chosen": 2.7573628425598145,
|
| 2555 |
+
"rewards/margins": 39.35446548461914,
|
| 2556 |
+
"rewards/rejected": -36.59709930419922,
|
| 2557 |
+
"step": 182
|
| 2558 |
+
},
|
| 2559 |
+
{
|
| 2560 |
+
"epoch": 0.2,
|
| 2561 |
+
"learning_rate": 3.791379310344828e-05,
|
| 2562 |
+
"logits/chosen": -0.39544716477394104,
|
| 2563 |
+
"logits/rejected": -1.4232298135757446,
|
| 2564 |
+
"logps/chosen": -400.05267333984375,
|
| 2565 |
+
"logps/rejected": -465.1136169433594,
|
| 2566 |
+
"loss": 0.0018,
|
| 2567 |
+
"rewards/accuracies": 1.0,
|
| 2568 |
+
"rewards/chosen": 3.5475125312805176,
|
| 2569 |
+
"rewards/margins": 38.094913482666016,
|
| 2570 |
+
"rewards/rejected": -34.547393798828125,
|
| 2571 |
+
"step": 183
|
| 2572 |
+
},
|
| 2573 |
+
{
|
| 2574 |
+
"epoch": 0.2,
|
| 2575 |
+
"learning_rate": 3.786206896551725e-05,
|
| 2576 |
+
"logits/chosen": -0.3474569320678711,
|
| 2577 |
+
"logits/rejected": -1.4260185956954956,
|
| 2578 |
+
"logps/chosen": -334.5423889160156,
|
| 2579 |
+
"logps/rejected": -438.87408447265625,
|
| 2580 |
+
"loss": 0.0,
|
| 2581 |
+
"rewards/accuracies": 1.0,
|
| 2582 |
+
"rewards/chosen": -1.5236696004867554,
|
| 2583 |
+
"rewards/margins": 33.86663055419922,
|
| 2584 |
+
"rewards/rejected": -35.390296936035156,
|
| 2585 |
+
"step": 184
|
| 2586 |
+
},
|
| 2587 |
+
{
|
| 2588 |
+
"epoch": 0.2,
|
| 2589 |
+
"learning_rate": 3.781034482758621e-05,
|
| 2590 |
+
"logits/chosen": -0.3836357593536377,
|
| 2591 |
+
"logits/rejected": -1.554406762123108,
|
| 2592 |
+
"logps/chosen": -372.57830810546875,
|
| 2593 |
+
"logps/rejected": -439.4316101074219,
|
| 2594 |
+
"loss": 0.0,
|
| 2595 |
+
"rewards/accuracies": 1.0,
|
| 2596 |
+
"rewards/chosen": 1.840342402458191,
|
| 2597 |
+
"rewards/margins": 34.46831512451172,
|
| 2598 |
+
"rewards/rejected": -32.627967834472656,
|
| 2599 |
+
"step": 185
|
| 2600 |
+
},
|
| 2601 |
+
{
|
| 2602 |
+
"epoch": 0.2,
|
| 2603 |
+
"learning_rate": 3.775862068965517e-05,
|
| 2604 |
+
"logits/chosen": -0.18503640592098236,
|
| 2605 |
+
"logits/rejected": -1.800915241241455,
|
| 2606 |
+
"logps/chosen": -485.19354248046875,
|
| 2607 |
+
"logps/rejected": -432.1221923828125,
|
| 2608 |
+
"loss": 0.0,
|
| 2609 |
+
"rewards/accuracies": 1.0,
|
| 2610 |
+
"rewards/chosen": 4.203826427459717,
|
| 2611 |
+
"rewards/margins": 39.722900390625,
|
| 2612 |
+
"rewards/rejected": -35.51907730102539,
|
| 2613 |
+
"step": 186
|
| 2614 |
+
},
|
| 2615 |
+
{
|
| 2616 |
+
"epoch": 0.2,
|
| 2617 |
+
"learning_rate": 3.770689655172414e-05,
|
| 2618 |
+
"logits/chosen": -0.3300974667072296,
|
| 2619 |
+
"logits/rejected": -1.5052634477615356,
|
| 2620 |
+
"logps/chosen": -418.87017822265625,
|
| 2621 |
+
"logps/rejected": -408.99554443359375,
|
| 2622 |
+
"loss": 0.0,
|
| 2623 |
+
"rewards/accuracies": 1.0,
|
| 2624 |
+
"rewards/chosen": 2.9931721687316895,
|
| 2625 |
+
"rewards/margins": 35.2595100402832,
|
| 2626 |
+
"rewards/rejected": -32.266334533691406,
|
| 2627 |
+
"step": 187
|
| 2628 |
+
},
|
| 2629 |
+
{
|
| 2630 |
+
"epoch": 0.21,
|
| 2631 |
+
"learning_rate": 3.7655172413793106e-05,
|
| 2632 |
+
"logits/chosen": -0.27791130542755127,
|
| 2633 |
+
"logits/rejected": -1.4637497663497925,
|
| 2634 |
+
"logps/chosen": -370.642822265625,
|
| 2635 |
+
"logps/rejected": -457.7041320800781,
|
| 2636 |
+
"loss": 0.0,
|
| 2637 |
+
"rewards/accuracies": 1.0,
|
| 2638 |
+
"rewards/chosen": 4.304169178009033,
|
| 2639 |
+
"rewards/margins": 40.587310791015625,
|
| 2640 |
+
"rewards/rejected": -36.283138275146484,
|
| 2641 |
+
"step": 188
|
| 2642 |
+
},
|
| 2643 |
+
{
|
| 2644 |
+
"epoch": 0.21,
|
| 2645 |
+
"learning_rate": 3.760344827586207e-05,
|
| 2646 |
+
"logits/chosen": -0.4701173007488251,
|
| 2647 |
+
"logits/rejected": -1.403714895248413,
|
| 2648 |
+
"logps/chosen": -333.7972717285156,
|
| 2649 |
+
"logps/rejected": -436.09967041015625,
|
| 2650 |
+
"loss": 0.0,
|
| 2651 |
+
"rewards/accuracies": 1.0,
|
| 2652 |
+
"rewards/chosen": 1.8245915174484253,
|
| 2653 |
+
"rewards/margins": 36.37276077270508,
|
| 2654 |
+
"rewards/rejected": -34.54816818237305,
|
| 2655 |
+
"step": 189
|
| 2656 |
+
},
|
| 2657 |
+
{
|
| 2658 |
+
"epoch": 0.21,
|
| 2659 |
+
"learning_rate": 3.7551724137931035e-05,
|
| 2660 |
+
"logits/chosen": -0.08819374442100525,
|
| 2661 |
+
"logits/rejected": -1.395701289176941,
|
| 2662 |
+
"logps/chosen": -401.96673583984375,
|
| 2663 |
+
"logps/rejected": -440.0989685058594,
|
| 2664 |
+
"loss": 0.0,
|
| 2665 |
+
"rewards/accuracies": 1.0,
|
| 2666 |
+
"rewards/chosen": 2.565300941467285,
|
| 2667 |
+
"rewards/margins": 33.236175537109375,
|
| 2668 |
+
"rewards/rejected": -30.670875549316406,
|
| 2669 |
+
"step": 190
|
| 2670 |
+
},
|
| 2671 |
+
{
|
| 2672 |
+
"epoch": 0.21,
|
| 2673 |
+
"learning_rate": 3.7500000000000003e-05,
|
| 2674 |
+
"logits/chosen": -0.17811371386051178,
|
| 2675 |
+
"logits/rejected": -1.501448154449463,
|
| 2676 |
+
"logps/chosen": -447.5896301269531,
|
| 2677 |
+
"logps/rejected": -375.3368835449219,
|
| 2678 |
+
"loss": 0.0,
|
| 2679 |
+
"rewards/accuracies": 1.0,
|
| 2680 |
+
"rewards/chosen": 2.7833425998687744,
|
| 2681 |
+
"rewards/margins": 31.601383209228516,
|
| 2682 |
+
"rewards/rejected": -28.818038940429688,
|
| 2683 |
+
"step": 191
|
| 2684 |
+
},
|
| 2685 |
+
{
|
| 2686 |
+
"epoch": 0.21,
|
| 2687 |
+
"learning_rate": 3.7448275862068965e-05,
|
| 2688 |
+
"logits/chosen": -0.44955122470855713,
|
| 2689 |
+
"logits/rejected": -1.550938606262207,
|
| 2690 |
+
"logps/chosen": -289.89959716796875,
|
| 2691 |
+
"logps/rejected": -400.880859375,
|
| 2692 |
+
"loss": 0.0,
|
| 2693 |
+
"rewards/accuracies": 1.0,
|
| 2694 |
+
"rewards/chosen": -0.4812740385532379,
|
| 2695 |
+
"rewards/margins": 33.061256408691406,
|
| 2696 |
+
"rewards/rejected": -33.54253005981445,
|
| 2697 |
+
"step": 192
|
| 2698 |
+
},
|
| 2699 |
+
{
|
| 2700 |
+
"epoch": 0.21,
|
| 2701 |
+
"learning_rate": 3.739655172413793e-05,
|
| 2702 |
+
"logits/chosen": -0.18104656040668488,
|
| 2703 |
+
"logits/rejected": -1.4971003532409668,
|
| 2704 |
+
"logps/chosen": -418.8290710449219,
|
| 2705 |
+
"logps/rejected": -424.74859619140625,
|
| 2706 |
+
"loss": 0.0,
|
| 2707 |
+
"rewards/accuracies": 1.0,
|
| 2708 |
+
"rewards/chosen": 6.926714897155762,
|
| 2709 |
+
"rewards/margins": 39.01382064819336,
|
| 2710 |
+
"rewards/rejected": -32.08710861206055,
|
| 2711 |
+
"step": 193
|
| 2712 |
+
},
|
| 2713 |
+
{
|
| 2714 |
+
"epoch": 0.21,
|
| 2715 |
+
"learning_rate": 3.73448275862069e-05,
|
| 2716 |
+
"logits/chosen": -0.16005319356918335,
|
| 2717 |
+
"logits/rejected": -1.4771628379821777,
|
| 2718 |
+
"logps/chosen": -437.9535827636719,
|
| 2719 |
+
"logps/rejected": -435.7216796875,
|
| 2720 |
+
"loss": 0.0,
|
| 2721 |
+
"rewards/accuracies": 1.0,
|
| 2722 |
+
"rewards/chosen": 5.183934688568115,
|
| 2723 |
+
"rewards/margins": 40.51701354980469,
|
| 2724 |
+
"rewards/rejected": -35.33308410644531,
|
| 2725 |
+
"step": 194
|
| 2726 |
+
},
|
| 2727 |
+
{
|
| 2728 |
+
"epoch": 0.21,
|
| 2729 |
+
"learning_rate": 3.729310344827586e-05,
|
| 2730 |
+
"logits/chosen": -0.35886967182159424,
|
| 2731 |
+
"logits/rejected": -1.6980268955230713,
|
| 2732 |
+
"logps/chosen": -404.4248962402344,
|
| 2733 |
+
"logps/rejected": -412.4001159667969,
|
| 2734 |
+
"loss": 0.0068,
|
| 2735 |
+
"rewards/accuracies": 1.0,
|
| 2736 |
+
"rewards/chosen": 7.873406887054443,
|
| 2737 |
+
"rewards/margins": 38.6983757019043,
|
| 2738 |
+
"rewards/rejected": -30.824968338012695,
|
| 2739 |
+
"step": 195
|
| 2740 |
+
},
|
| 2741 |
+
{
|
| 2742 |
+
"epoch": 0.21,
|
| 2743 |
+
"learning_rate": 3.724137931034483e-05,
|
| 2744 |
+
"logits/chosen": -0.4241984486579895,
|
| 2745 |
+
"logits/rejected": -1.4724136590957642,
|
| 2746 |
+
"logps/chosen": -353.877197265625,
|
| 2747 |
+
"logps/rejected": -414.2342529296875,
|
| 2748 |
+
"loss": 0.0,
|
| 2749 |
+
"rewards/accuracies": 1.0,
|
| 2750 |
+
"rewards/chosen": 2.698596954345703,
|
| 2751 |
+
"rewards/margins": 36.431583404541016,
|
| 2752 |
+
"rewards/rejected": -33.73298645019531,
|
| 2753 |
+
"step": 196
|
| 2754 |
+
},
|
| 2755 |
+
{
|
| 2756 |
+
"epoch": 0.21,
|
| 2757 |
+
"learning_rate": 3.71896551724138e-05,
|
| 2758 |
+
"logits/chosen": -0.38190075755119324,
|
| 2759 |
+
"logits/rejected": -1.520878553390503,
|
| 2760 |
+
"logps/chosen": -303.9383544921875,
|
| 2761 |
+
"logps/rejected": -443.4150390625,
|
| 2762 |
+
"loss": 0.0,
|
| 2763 |
+
"rewards/accuracies": 1.0,
|
| 2764 |
+
"rewards/chosen": 5.500533103942871,
|
| 2765 |
+
"rewards/margins": 41.76572036743164,
|
| 2766 |
+
"rewards/rejected": -36.26518630981445,
|
| 2767 |
+
"step": 197
|
| 2768 |
+
},
|
| 2769 |
+
{
|
| 2770 |
+
"epoch": 0.22,
|
| 2771 |
+
"learning_rate": 3.713793103448276e-05,
|
| 2772 |
+
"logits/chosen": -0.5008329153060913,
|
| 2773 |
+
"logits/rejected": -1.4159965515136719,
|
| 2774 |
+
"logps/chosen": -298.4696044921875,
|
| 2775 |
+
"logps/rejected": -496.4792175292969,
|
| 2776 |
+
"loss": 0.0,
|
| 2777 |
+
"rewards/accuracies": 1.0,
|
| 2778 |
+
"rewards/chosen": 3.6865413188934326,
|
| 2779 |
+
"rewards/margins": 43.41154098510742,
|
| 2780 |
+
"rewards/rejected": -39.72499465942383,
|
| 2781 |
+
"step": 198
|
| 2782 |
+
},
|
| 2783 |
+
{
|
| 2784 |
+
"epoch": 0.22,
|
| 2785 |
+
"learning_rate": 3.708620689655173e-05,
|
| 2786 |
+
"logits/chosen": -0.30573636293411255,
|
| 2787 |
+
"logits/rejected": -1.4757778644561768,
|
| 2788 |
+
"logps/chosen": -404.08349609375,
|
| 2789 |
+
"logps/rejected": -536.4090576171875,
|
| 2790 |
+
"loss": 0.0994,
|
| 2791 |
+
"rewards/accuracies": 0.9375,
|
| 2792 |
+
"rewards/chosen": 2.837350606918335,
|
| 2793 |
+
"rewards/margins": 47.625877380371094,
|
| 2794 |
+
"rewards/rejected": -44.78852844238281,
|
| 2795 |
+
"step": 199
|
| 2796 |
+
},
|
| 2797 |
+
{
|
| 2798 |
+
"epoch": 0.22,
|
| 2799 |
+
"learning_rate": 3.7034482758620695e-05,
|
| 2800 |
+
"logits/chosen": -0.08167792111635208,
|
| 2801 |
+
"logits/rejected": -1.6543066501617432,
|
| 2802 |
+
"logps/chosen": -478.5865173339844,
|
| 2803 |
+
"logps/rejected": -535.1455078125,
|
| 2804 |
+
"loss": 0.0,
|
| 2805 |
+
"rewards/accuracies": 1.0,
|
| 2806 |
+
"rewards/chosen": 6.218913555145264,
|
| 2807 |
+
"rewards/margins": 51.752662658691406,
|
| 2808 |
+
"rewards/rejected": -45.533748626708984,
|
| 2809 |
+
"step": 200
|
| 2810 |
+
}
|
| 2811 |
+
],
|
| 2812 |
+
"logging_steps": 1,
|
| 2813 |
+
"max_steps": 916,
|
| 2814 |
+
"num_input_tokens_seen": 0,
|
| 2815 |
+
"num_train_epochs": 1,
|
| 2816 |
+
"save_steps": 100,
|
| 2817 |
+
"total_flos": 0.0,
|
| 2818 |
+
"train_batch_size": 1,
|
| 2819 |
+
"trial_name": null,
|
| 2820 |
+
"trial_params": null
|
| 2821 |
+
}
|
checkpoint-200/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6f8e778ebfdc8e189d4259f43aa8cc8438633b1407516e064a080cf26f579c11
|
| 3 |
+
size 4664
|
checkpoint-300/README.md
ADDED
|
@@ -0,0 +1,204 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: peft
|
| 3 |
+
base_model: /run/media/adamo1139/82142F79142F6EFB/ProgramData/Anaconda3/envs/qlora-jondurbin/axolotl-git-linux/axolotl/yi-34b-200k-llamafied
|
| 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 |
+
|
| 201 |
+
|
| 202 |
+
### Framework versions
|
| 203 |
+
|
| 204 |
+
- PEFT 0.7.1
|
checkpoint-300/adapter_config.json
ADDED
|
@@ -0,0 +1,31 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"alpha_pattern": {},
|
| 3 |
+
"auto_mapping": null,
|
| 4 |
+
"base_model_name_or_path": "yi-34b-200k-llamafied",
|
| 5 |
+
"bias": "none",
|
| 6 |
+
"fan_in_fan_out": false,
|
| 7 |
+
"inference_mode": true,
|
| 8 |
+
"init_lora_weights": true,
|
| 9 |
+
"layers_pattern": null,
|
| 10 |
+
"layers_to_transform": null,
|
| 11 |
+
"loftq_config": null,
|
| 12 |
+
"lora_alpha": 32,
|
| 13 |
+
"lora_dropout": 0,
|
| 14 |
+
"megatron_config": null,
|
| 15 |
+
"megatron_core": "megatron.core",
|
| 16 |
+
"modules_to_save": null,
|
| 17 |
+
"peft_type": "LORA",
|
| 18 |
+
"r": 16,
|
| 19 |
+
"rank_pattern": {},
|
| 20 |
+
"revision": "unsloth",
|
| 21 |
+
"target_modules": [
|
| 22 |
+
"v_proj",
|
| 23 |
+
"gate_proj",
|
| 24 |
+
"o_proj",
|
| 25 |
+
"down_proj",
|
| 26 |
+
"q_proj",
|
| 27 |
+
"up_proj",
|
| 28 |
+
"k_proj"
|
| 29 |
+
],
|
| 30 |
+
"task_type": "CAUSAL_LM"
|
| 31 |
+
}
|
checkpoint-300/adapter_model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4911591b593c045ffdc4ecde4f0a7805cd441302516c3b4642ead2cf658ddd74
|
| 3 |
+
size 491633464
|
checkpoint-300/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1ff264f99d31b522cc7e2a4eac9d38606d0c58a34c0adc74d71e0ca8b371dc36
|
| 3 |
+
size 14244
|
checkpoint-300/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:79ebb2f81c622f336b43b3df591ecc6208eda3036d54d729534e15955533e108
|
| 3 |
+
size 1064
|
checkpoint-300/special_tokens_map.json
ADDED
|
@@ -0,0 +1,30 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": {
|
| 3 |
+
"content": "<|startoftext|>",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": false,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"eos_token": {
|
| 10 |
+
"content": "<|endoftext|>",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": false,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"pad_token": {
|
| 17 |
+
"content": "<unk>",
|
| 18 |
+
"lstrip": false,
|
| 19 |
+
"normalized": false,
|
| 20 |
+
"rstrip": false,
|
| 21 |
+
"single_word": false
|
| 22 |
+
},
|
| 23 |
+
"unk_token": {
|
| 24 |
+
"content": "<unk>",
|
| 25 |
+
"lstrip": false,
|
| 26 |
+
"normalized": false,
|
| 27 |
+
"rstrip": false,
|
| 28 |
+
"single_word": false
|
| 29 |
+
}
|
| 30 |
+
}
|
checkpoint-300/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-300/tokenizer.model
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:386c49cf943d71aa110361135338c50e38beeff0a66593480421f37b319e1a39
|
| 3 |
+
size 1033105
|
checkpoint-300/tokenizer_config.json
ADDED
|
@@ -0,0 +1,42 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_bos_token": false,
|
| 3 |
+
"add_eos_token": false,
|
| 4 |
+
"added_tokens_decoder": {
|
| 5 |
+
"0": {
|
| 6 |
+
"content": "<unk>",
|
| 7 |
+
"lstrip": false,
|
| 8 |
+
"normalized": false,
|
| 9 |
+
"rstrip": false,
|
| 10 |
+
"single_word": false,
|
| 11 |
+
"special": true
|
| 12 |
+
},
|
| 13 |
+
"1": {
|
| 14 |
+
"content": "<|startoftext|>",
|
| 15 |
+
"lstrip": false,
|
| 16 |
+
"normalized": false,
|
| 17 |
+
"rstrip": false,
|
| 18 |
+
"single_word": false,
|
| 19 |
+
"special": true
|
| 20 |
+
},
|
| 21 |
+
"2": {
|
| 22 |
+
"content": "<|endoftext|>",
|
| 23 |
+
"lstrip": false,
|
| 24 |
+
"normalized": false,
|
| 25 |
+
"rstrip": false,
|
| 26 |
+
"single_word": false,
|
| 27 |
+
"special": true
|
| 28 |
+
}
|
| 29 |
+
},
|
| 30 |
+
"bos_token": "<|startoftext|>",
|
| 31 |
+
"chat_template": "{% if not add_generation_prompt is defined %}{% set add_generation_prompt = false %}{% endif %}{% for message in messages %}{{'<|im_start|>' + message['role'] + '\n' + message['content'] + '<|im_end|>' + '\n'}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant\n' }}{% endif %}",
|
| 32 |
+
"clean_up_tokenization_spaces": false,
|
| 33 |
+
"eos_token": "<|endoftext|>",
|
| 34 |
+
"legacy": true,
|
| 35 |
+
"model_max_length": 4096,
|
| 36 |
+
"pad_token": "<unk>",
|
| 37 |
+
"padding_side": "right",
|
| 38 |
+
"sp_model_kwargs": {},
|
| 39 |
+
"tokenizer_class": "LlamaTokenizer",
|
| 40 |
+
"unk_token": "<unk>",
|
| 41 |
+
"use_default_system_prompt": false
|
| 42 |
+
}
|
checkpoint-300/trainer_state.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-300/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6f8e778ebfdc8e189d4259f43aa8cc8438633b1407516e064a080cf26f579c11
|
| 3 |
+
size 4664
|