Upload 12 files
Browse files- README.md +202 -1
- adapter_config.json +33 -0
- adapter_model.safetensors +3 -0
- optimizer.pt +3 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- special_tokens_map.json +24 -0
- tokenizer.json +0 -0
- tokenizer.model +3 -0
- tokenizer_config.json +43 -0
- trainer_state.json +2345 -0
- training_args.bin +3 -0
README.md
CHANGED
|
@@ -1,3 +1,204 @@
|
|
| 1 |
---
|
| 2 |
-
|
|
|
|
| 3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
---
|
| 2 |
+
library_name: peft
|
| 3 |
+
base_model: mistralai/Mistral-7B-Instruct-v0.2
|
| 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 0.9.0
|
adapter_config.json
ADDED
|
@@ -0,0 +1,33 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"alpha_pattern": {},
|
| 3 |
+
"auto_mapping": null,
|
| 4 |
+
"base_model_name_or_path": "mistralai/Mistral-7B-Instruct-v0.2",
|
| 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": {},
|
| 12 |
+
"lora_alpha": 16,
|
| 13 |
+
"lora_dropout": 0.05,
|
| 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": null,
|
| 21 |
+
"target_modules": [
|
| 22 |
+
"up_proj",
|
| 23 |
+
"o_proj",
|
| 24 |
+
"down_proj",
|
| 25 |
+
"v_proj",
|
| 26 |
+
"q_proj",
|
| 27 |
+
"gate_proj",
|
| 28 |
+
"k_proj"
|
| 29 |
+
],
|
| 30 |
+
"task_type": "CAUSAL_LM",
|
| 31 |
+
"use_dora": false,
|
| 32 |
+
"use_rslora": false
|
| 33 |
+
}
|
adapter_model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:597b1c776f6275896352bbaa6b1207c39d5de0115e56eeb9f58869200eb2d8cb
|
| 3 |
+
size 167832240
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c85fcdf8360ea42c84d0090bb22c20e439e2124e0616d6d1a2558c92f3ab57d4
|
| 3 |
+
size 84580630
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:bbd76872ac027eb39a9adfefe0495752f0bdba7a48edaf28359209de94e3e648
|
| 3 |
+
size 14244
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:225705c4e1aa7ad508c6caabd165fa312114485c0cc5734e167458f259296f6c
|
| 3 |
+
size 1064
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,24 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": {
|
| 3 |
+
"content": "<s>",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": false,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"eos_token": {
|
| 10 |
+
"content": "</s>",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": false,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"pad_token": "</s>",
|
| 17 |
+
"unk_token": {
|
| 18 |
+
"content": "<unk>",
|
| 19 |
+
"lstrip": false,
|
| 20 |
+
"normalized": false,
|
| 21 |
+
"rstrip": false,
|
| 22 |
+
"single_word": false
|
| 23 |
+
}
|
| 24 |
+
}
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer.model
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:dadfd56d766715c61d2ef780a525ab43b8e6da4de6865bda3d95fdef5e134055
|
| 3 |
+
size 493443
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,43 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_bos_token": true,
|
| 3 |
+
"add_eos_token": true,
|
| 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": "<s>",
|
| 15 |
+
"lstrip": false,
|
| 16 |
+
"normalized": false,
|
| 17 |
+
"rstrip": false,
|
| 18 |
+
"single_word": false,
|
| 19 |
+
"special": true
|
| 20 |
+
},
|
| 21 |
+
"2": {
|
| 22 |
+
"content": "</s>",
|
| 23 |
+
"lstrip": false,
|
| 24 |
+
"normalized": false,
|
| 25 |
+
"rstrip": false,
|
| 26 |
+
"single_word": false,
|
| 27 |
+
"special": true
|
| 28 |
+
}
|
| 29 |
+
},
|
| 30 |
+
"additional_special_tokens": [],
|
| 31 |
+
"bos_token": "<s>",
|
| 32 |
+
"chat_template": "{{ bos_token }}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% if message['role'] == 'user' %}{{ '[INST] ' + message['content'] + ' [/INST]' }}{% elif message['role'] == 'assistant' %}{{ message['content'] + eos_token}}{% else %}{{ raise_exception('Only user and assistant roles are supported!') }}{% endif %}{% endfor %}",
|
| 33 |
+
"clean_up_tokenization_spaces": false,
|
| 34 |
+
"eos_token": "</s>",
|
| 35 |
+
"legacy": true,
|
| 36 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 37 |
+
"pad_token": "</s>",
|
| 38 |
+
"sp_model_kwargs": {},
|
| 39 |
+
"spaces_between_special_tokens": false,
|
| 40 |
+
"tokenizer_class": "LlamaTokenizer",
|
| 41 |
+
"unk_token": "<unk>",
|
| 42 |
+
"use_default_system_prompt": false
|
| 43 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,2345 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 35.55555555555556,
|
| 5 |
+
"eval_steps": 40,
|
| 6 |
+
"global_step": 560,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.25,
|
| 13 |
+
"grad_norm": 7.928347587585449,
|
| 14 |
+
"learning_rate": 7.5e-08,
|
| 15 |
+
"logits/chosen": -2.7277705669403076,
|
| 16 |
+
"logits/rejected": -2.7679762840270996,
|
| 17 |
+
"logps/chosen": -128.4798583984375,
|
| 18 |
+
"logps/rejected": -98.36178588867188,
|
| 19 |
+
"loss": 0.6931,
|
| 20 |
+
"rewards/accuracies": 0.109375,
|
| 21 |
+
"rewards/chosen": -7.17043731128797e-05,
|
| 22 |
+
"rewards/margins": 5.325676465872675e-05,
|
| 23 |
+
"rewards/rejected": -0.0001249611668754369,
|
| 24 |
+
"step": 4
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.51,
|
| 28 |
+
"grad_norm": 9.242063522338867,
|
| 29 |
+
"learning_rate": 1.75e-07,
|
| 30 |
+
"logits/chosen": -2.6805009841918945,
|
| 31 |
+
"logits/rejected": -2.712625026702881,
|
| 32 |
+
"logps/chosen": -138.11959838867188,
|
| 33 |
+
"logps/rejected": -96.05882263183594,
|
| 34 |
+
"loss": 0.693,
|
| 35 |
+
"rewards/accuracies": 0.546875,
|
| 36 |
+
"rewards/chosen": 0.0008271098486147821,
|
| 37 |
+
"rewards/margins": 0.00034989125560969114,
|
| 38 |
+
"rewards/rejected": 0.00047721865121275187,
|
| 39 |
+
"step": 8
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.76,
|
| 43 |
+
"grad_norm": 8.156970977783203,
|
| 44 |
+
"learning_rate": 2.75e-07,
|
| 45 |
+
"logits/chosen": -2.7695400714874268,
|
| 46 |
+
"logits/rejected": -2.7420883178710938,
|
| 47 |
+
"logps/chosen": -133.04307556152344,
|
| 48 |
+
"logps/rejected": -104.19400024414062,
|
| 49 |
+
"loss": 0.6919,
|
| 50 |
+
"rewards/accuracies": 0.609375,
|
| 51 |
+
"rewards/chosen": 0.0034509659744799137,
|
| 52 |
+
"rewards/margins": 0.0025641382671892643,
|
| 53 |
+
"rewards/rejected": 0.0008868275908753276,
|
| 54 |
+
"step": 12
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 1.02,
|
| 58 |
+
"grad_norm": 8.723258018493652,
|
| 59 |
+
"learning_rate": 3.75e-07,
|
| 60 |
+
"logits/chosen": -2.6869003772735596,
|
| 61 |
+
"logits/rejected": -2.7081785202026367,
|
| 62 |
+
"logps/chosen": -126.2979736328125,
|
| 63 |
+
"logps/rejected": -98.4257583618164,
|
| 64 |
+
"loss": 0.6889,
|
| 65 |
+
"rewards/accuracies": 0.828125,
|
| 66 |
+
"rewards/chosen": 0.011365305632352829,
|
| 67 |
+
"rewards/margins": 0.008443659171462059,
|
| 68 |
+
"rewards/rejected": 0.0029216469265520573,
|
| 69 |
+
"step": 16
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 1.27,
|
| 73 |
+
"grad_norm": 10.34566593170166,
|
| 74 |
+
"learning_rate": 4.7499999999999995e-07,
|
| 75 |
+
"logits/chosen": -2.704402208328247,
|
| 76 |
+
"logits/rejected": -2.764737367630005,
|
| 77 |
+
"logps/chosen": -125.91941833496094,
|
| 78 |
+
"logps/rejected": -99.02344512939453,
|
| 79 |
+
"loss": 0.6832,
|
| 80 |
+
"rewards/accuracies": 0.9375,
|
| 81 |
+
"rewards/chosen": 0.02058127522468567,
|
| 82 |
+
"rewards/margins": 0.020110297948122025,
|
| 83 |
+
"rewards/rejected": 0.00047097797505557537,
|
| 84 |
+
"step": 20
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 1.52,
|
| 88 |
+
"grad_norm": 9.291991233825684,
|
| 89 |
+
"learning_rate": 4.999985633126757e-07,
|
| 90 |
+
"logits/chosen": -2.716165781021118,
|
| 91 |
+
"logits/rejected": -2.6845083236694336,
|
| 92 |
+
"logps/chosen": -137.2718505859375,
|
| 93 |
+
"logps/rejected": -97.50873565673828,
|
| 94 |
+
"loss": 0.6785,
|
| 95 |
+
"rewards/accuracies": 0.90625,
|
| 96 |
+
"rewards/chosen": 0.03298673778772354,
|
| 97 |
+
"rewards/margins": 0.029706846922636032,
|
| 98 |
+
"rewards/rejected": 0.0032798887696117163,
|
| 99 |
+
"step": 24
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 1.78,
|
| 103 |
+
"grad_norm": 9.161444664001465,
|
| 104 |
+
"learning_rate": 4.999921780689761e-07,
|
| 105 |
+
"logits/chosen": -2.735781192779541,
|
| 106 |
+
"logits/rejected": -2.7485365867614746,
|
| 107 |
+
"logps/chosen": -128.5098876953125,
|
| 108 |
+
"logps/rejected": -90.1419906616211,
|
| 109 |
+
"loss": 0.6693,
|
| 110 |
+
"rewards/accuracies": 0.953125,
|
| 111 |
+
"rewards/chosen": 0.04758748784661293,
|
| 112 |
+
"rewards/margins": 0.04847026616334915,
|
| 113 |
+
"rewards/rejected": -0.0008827749406918883,
|
| 114 |
+
"step": 28
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 2.03,
|
| 118 |
+
"grad_norm": 8.495955467224121,
|
| 119 |
+
"learning_rate": 4.99980684767309e-07,
|
| 120 |
+
"logits/chosen": -2.7206833362579346,
|
| 121 |
+
"logits/rejected": -2.750042676925659,
|
| 122 |
+
"logps/chosen": -133.54827880859375,
|
| 123 |
+
"logps/rejected": -108.23342895507812,
|
| 124 |
+
"loss": 0.6669,
|
| 125 |
+
"rewards/accuracies": 0.953125,
|
| 126 |
+
"rewards/chosen": 0.0631466805934906,
|
| 127 |
+
"rewards/margins": 0.05378420650959015,
|
| 128 |
+
"rewards/rejected": 0.009362474083900452,
|
| 129 |
+
"step": 32
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 2.29,
|
| 133 |
+
"grad_norm": 8.951277732849121,
|
| 134 |
+
"learning_rate": 4.999640836425158e-07,
|
| 135 |
+
"logits/chosen": -2.6706807613372803,
|
| 136 |
+
"logits/rejected": -2.6151680946350098,
|
| 137 |
+
"logps/chosen": -136.55226135253906,
|
| 138 |
+
"logps/rejected": -97.76315307617188,
|
| 139 |
+
"loss": 0.6578,
|
| 140 |
+
"rewards/accuracies": 0.953125,
|
| 141 |
+
"rewards/chosen": 0.08149851858615875,
|
| 142 |
+
"rewards/margins": 0.07302143424749374,
|
| 143 |
+
"rewards/rejected": 0.008477086201310158,
|
| 144 |
+
"step": 36
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 2.54,
|
| 148 |
+
"grad_norm": 8.608317375183105,
|
| 149 |
+
"learning_rate": 4.99942375033805e-07,
|
| 150 |
+
"logits/chosen": -2.7676639556884766,
|
| 151 |
+
"logits/rejected": -2.833590030670166,
|
| 152 |
+
"logps/chosen": -133.7703857421875,
|
| 153 |
+
"logps/rejected": -106.12245178222656,
|
| 154 |
+
"loss": 0.6488,
|
| 155 |
+
"rewards/accuracies": 0.9375,
|
| 156 |
+
"rewards/chosen": 0.10222765803337097,
|
| 157 |
+
"rewards/margins": 0.09170522540807724,
|
| 158 |
+
"rewards/rejected": 0.010522443801164627,
|
| 159 |
+
"step": 40
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 2.54,
|
| 163 |
+
"eval_logits/chosen": -2.806352138519287,
|
| 164 |
+
"eval_logits/rejected": -2.8898861408233643,
|
| 165 |
+
"eval_logps/chosen": -133.14093017578125,
|
| 166 |
+
"eval_logps/rejected": -97.8843002319336,
|
| 167 |
+
"eval_loss": 0.6592782139778137,
|
| 168 |
+
"eval_rewards/accuracies": 0.7857142686843872,
|
| 169 |
+
"eval_rewards/chosen": 0.09283973276615143,
|
| 170 |
+
"eval_rewards/margins": 0.07038124650716782,
|
| 171 |
+
"eval_rewards/rejected": 0.022458484396338463,
|
| 172 |
+
"eval_runtime": 19.3683,
|
| 173 |
+
"eval_samples_per_second": 1.446,
|
| 174 |
+
"eval_steps_per_second": 1.446,
|
| 175 |
+
"step": 40
|
| 176 |
+
},
|
| 177 |
+
{
|
| 178 |
+
"epoch": 2.79,
|
| 179 |
+
"grad_norm": 9.09485912322998,
|
| 180 |
+
"learning_rate": 4.999155593847457e-07,
|
| 181 |
+
"logits/chosen": -2.7024385929107666,
|
| 182 |
+
"logits/rejected": -2.7564737796783447,
|
| 183 |
+
"logps/chosen": -123.412353515625,
|
| 184 |
+
"logps/rejected": -91.91966247558594,
|
| 185 |
+
"loss": 0.6347,
|
| 186 |
+
"rewards/accuracies": 0.96875,
|
| 187 |
+
"rewards/chosen": 0.114817313849926,
|
| 188 |
+
"rewards/margins": 0.12200291454792023,
|
| 189 |
+
"rewards/rejected": -0.007185595575720072,
|
| 190 |
+
"step": 44
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 3.05,
|
| 194 |
+
"grad_norm": 9.453149795532227,
|
| 195 |
+
"learning_rate": 4.998836372432589e-07,
|
| 196 |
+
"logits/chosen": -2.736665725708008,
|
| 197 |
+
"logits/rejected": -2.757841110229492,
|
| 198 |
+
"logps/chosen": -127.0240478515625,
|
| 199 |
+
"logps/rejected": -103.66780853271484,
|
| 200 |
+
"loss": 0.636,
|
| 201 |
+
"rewards/accuracies": 0.96875,
|
| 202 |
+
"rewards/chosen": 0.13144327700138092,
|
| 203 |
+
"rewards/margins": 0.11972987651824951,
|
| 204 |
+
"rewards/rejected": 0.011713397689163685,
|
| 205 |
+
"step": 48
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 3.3,
|
| 209 |
+
"grad_norm": 8.674492835998535,
|
| 210 |
+
"learning_rate": 4.998466092616052e-07,
|
| 211 |
+
"logits/chosen": -2.710110664367676,
|
| 212 |
+
"logits/rejected": -2.7205581665039062,
|
| 213 |
+
"logps/chosen": -131.0464324951172,
|
| 214 |
+
"logps/rejected": -99.71279907226562,
|
| 215 |
+
"loss": 0.6166,
|
| 216 |
+
"rewards/accuracies": 0.96875,
|
| 217 |
+
"rewards/chosen": 0.15897569060325623,
|
| 218 |
+
"rewards/margins": 0.16229701042175293,
|
| 219 |
+
"rewards/rejected": -0.003321333322674036,
|
| 220 |
+
"step": 52
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 3.56,
|
| 224 |
+
"grad_norm": 9.434328079223633,
|
| 225 |
+
"learning_rate": 4.99804476196373e-07,
|
| 226 |
+
"logits/chosen": -2.685814380645752,
|
| 227 |
+
"logits/rejected": -2.736269474029541,
|
| 228 |
+
"logps/chosen": -125.750732421875,
|
| 229 |
+
"logps/rejected": -93.69428253173828,
|
| 230 |
+
"loss": 0.6131,
|
| 231 |
+
"rewards/accuracies": 0.96875,
|
| 232 |
+
"rewards/chosen": 0.16781488060951233,
|
| 233 |
+
"rewards/margins": 0.16998659074306488,
|
| 234 |
+
"rewards/rejected": -0.002171688713133335,
|
| 235 |
+
"step": 56
|
| 236 |
+
},
|
| 237 |
+
{
|
| 238 |
+
"epoch": 3.81,
|
| 239 |
+
"grad_norm": 8.643455505371094,
|
| 240 |
+
"learning_rate": 4.997572389084615e-07,
|
| 241 |
+
"logits/chosen": -2.7429587841033936,
|
| 242 |
+
"logits/rejected": -2.75490140914917,
|
| 243 |
+
"logps/chosen": -131.4405517578125,
|
| 244 |
+
"logps/rejected": -99.93754577636719,
|
| 245 |
+
"loss": 0.614,
|
| 246 |
+
"rewards/accuracies": 0.921875,
|
| 247 |
+
"rewards/chosen": 0.18873561918735504,
|
| 248 |
+
"rewards/margins": 0.17132429778575897,
|
| 249 |
+
"rewards/rejected": 0.01741132140159607,
|
| 250 |
+
"step": 60
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 4.06,
|
| 254 |
+
"grad_norm": 8.412778854370117,
|
| 255 |
+
"learning_rate": 4.997048983630643e-07,
|
| 256 |
+
"logits/chosen": -2.6886143684387207,
|
| 257 |
+
"logits/rejected": -2.710732936859131,
|
| 258 |
+
"logps/chosen": -123.77167510986328,
|
| 259 |
+
"logps/rejected": -97.30846405029297,
|
| 260 |
+
"loss": 0.6099,
|
| 261 |
+
"rewards/accuracies": 0.9375,
|
| 262 |
+
"rewards/chosen": 0.19313779473304749,
|
| 263 |
+
"rewards/margins": 0.1793561577796936,
|
| 264 |
+
"rewards/rejected": 0.013781649991869926,
|
| 265 |
+
"step": 64
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 4.32,
|
| 269 |
+
"grad_norm": 7.851288795471191,
|
| 270 |
+
"learning_rate": 4.99647455629649e-07,
|
| 271 |
+
"logits/chosen": -2.744528293609619,
|
| 272 |
+
"logits/rejected": -2.7117881774902344,
|
| 273 |
+
"logps/chosen": -136.25863647460938,
|
| 274 |
+
"logps/rejected": -95.11305236816406,
|
| 275 |
+
"loss": 0.5968,
|
| 276 |
+
"rewards/accuracies": 0.96875,
|
| 277 |
+
"rewards/chosen": 0.21463464200496674,
|
| 278 |
+
"rewards/margins": 0.20793242752552032,
|
| 279 |
+
"rewards/rejected": 0.0067022331058979034,
|
| 280 |
+
"step": 68
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"epoch": 4.57,
|
| 284 |
+
"grad_norm": 9.278959274291992,
|
| 285 |
+
"learning_rate": 4.995849118819353e-07,
|
| 286 |
+
"logits/chosen": -2.7611894607543945,
|
| 287 |
+
"logits/rejected": -2.7447633743286133,
|
| 288 |
+
"logps/chosen": -134.3462677001953,
|
| 289 |
+
"logps/rejected": -103.75216674804688,
|
| 290 |
+
"loss": 0.5828,
|
| 291 |
+
"rewards/accuracies": 0.984375,
|
| 292 |
+
"rewards/chosen": 0.25433310866355896,
|
| 293 |
+
"rewards/margins": 0.24298511445522308,
|
| 294 |
+
"rewards/rejected": 0.011347985826432705,
|
| 295 |
+
"step": 72
|
| 296 |
+
},
|
| 297 |
+
{
|
| 298 |
+
"epoch": 4.83,
|
| 299 |
+
"grad_norm": 7.795536994934082,
|
| 300 |
+
"learning_rate": 4.995172683978719e-07,
|
| 301 |
+
"logits/chosen": -2.698429584503174,
|
| 302 |
+
"logits/rejected": -2.7458348274230957,
|
| 303 |
+
"logps/chosen": -127.65770721435547,
|
| 304 |
+
"logps/rejected": -104.23495483398438,
|
| 305 |
+
"loss": 0.5796,
|
| 306 |
+
"rewards/accuracies": 0.9375,
|
| 307 |
+
"rewards/chosen": 0.25434932112693787,
|
| 308 |
+
"rewards/margins": 0.25114190578460693,
|
| 309 |
+
"rewards/rejected": 0.0032073920592665672,
|
| 310 |
+
"step": 76
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 5.08,
|
| 314 |
+
"grad_norm": 7.788575649261475,
|
| 315 |
+
"learning_rate": 4.994445265596091e-07,
|
| 316 |
+
"logits/chosen": -2.6734955310821533,
|
| 317 |
+
"logits/rejected": -2.7292165756225586,
|
| 318 |
+
"logps/chosen": -123.53501892089844,
|
| 319 |
+
"logps/rejected": -95.23420715332031,
|
| 320 |
+
"loss": 0.5676,
|
| 321 |
+
"rewards/accuracies": 0.96875,
|
| 322 |
+
"rewards/chosen": 0.2950035631656647,
|
| 323 |
+
"rewards/margins": 0.2805778682231903,
|
| 324 |
+
"rewards/rejected": 0.014425676316022873,
|
| 325 |
+
"step": 80
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 5.08,
|
| 329 |
+
"eval_logits/chosen": -2.8074097633361816,
|
| 330 |
+
"eval_logits/rejected": -2.890533685684204,
|
| 331 |
+
"eval_logps/chosen": -131.55104064941406,
|
| 332 |
+
"eval_logps/rejected": -97.56663513183594,
|
| 333 |
+
"eval_loss": 0.6045485734939575,
|
| 334 |
+
"eval_rewards/accuracies": 0.7857142686843872,
|
| 335 |
+
"eval_rewards/chosen": 0.2518289387226105,
|
| 336 |
+
"eval_rewards/margins": 0.1976027488708496,
|
| 337 |
+
"eval_rewards/rejected": 0.05422618240118027,
|
| 338 |
+
"eval_runtime": 19.3586,
|
| 339 |
+
"eval_samples_per_second": 1.446,
|
| 340 |
+
"eval_steps_per_second": 1.446,
|
| 341 |
+
"step": 80
|
| 342 |
+
},
|
| 343 |
+
{
|
| 344 |
+
"epoch": 5.33,
|
| 345 |
+
"grad_norm": 6.709670543670654,
|
| 346 |
+
"learning_rate": 4.993666878534718e-07,
|
| 347 |
+
"logits/chosen": -2.736870527267456,
|
| 348 |
+
"logits/rejected": -2.778140068054199,
|
| 349 |
+
"logps/chosen": -128.86972045898438,
|
| 350 |
+
"logps/rejected": -104.70977020263672,
|
| 351 |
+
"loss": 0.5637,
|
| 352 |
+
"rewards/accuracies": 0.953125,
|
| 353 |
+
"rewards/chosen": 0.31255263090133667,
|
| 354 |
+
"rewards/margins": 0.2908375561237335,
|
| 355 |
+
"rewards/rejected": 0.021715058013796806,
|
| 356 |
+
"step": 84
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 5.59,
|
| 360 |
+
"grad_norm": 8.178019523620605,
|
| 361 |
+
"learning_rate": 4.99283753869928e-07,
|
| 362 |
+
"logits/chosen": -2.686039447784424,
|
| 363 |
+
"logits/rejected": -2.681716203689575,
|
| 364 |
+
"logps/chosen": -129.52798461914062,
|
| 365 |
+
"logps/rejected": -97.80433654785156,
|
| 366 |
+
"loss": 0.5475,
|
| 367 |
+
"rewards/accuracies": 1.0,
|
| 368 |
+
"rewards/chosen": 0.32084640860557556,
|
| 369 |
+
"rewards/margins": 0.33009591698646545,
|
| 370 |
+
"rewards/rejected": -0.00924946740269661,
|
| 371 |
+
"step": 88
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 5.84,
|
| 375 |
+
"grad_norm": 7.515178680419922,
|
| 376 |
+
"learning_rate": 4.991957263035573e-07,
|
| 377 |
+
"logits/chosen": -2.763765811920166,
|
| 378 |
+
"logits/rejected": -2.7660324573516846,
|
| 379 |
+
"logps/chosen": -128.16851806640625,
|
| 380 |
+
"logps/rejected": -95.72579956054688,
|
| 381 |
+
"loss": 0.5493,
|
| 382 |
+
"rewards/accuracies": 0.96875,
|
| 383 |
+
"rewards/chosen": 0.3080848455429077,
|
| 384 |
+
"rewards/margins": 0.32350555062294006,
|
| 385 |
+
"rewards/rejected": -0.015420722775161266,
|
| 386 |
+
"step": 92
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 6.1,
|
| 390 |
+
"grad_norm": 8.599517822265625,
|
| 391 |
+
"learning_rate": 4.991026069530156e-07,
|
| 392 |
+
"logits/chosen": -2.7107338905334473,
|
| 393 |
+
"logits/rejected": -2.745981216430664,
|
| 394 |
+
"logps/chosen": -131.5907745361328,
|
| 395 |
+
"logps/rejected": -102.65729522705078,
|
| 396 |
+
"loss": 0.5568,
|
| 397 |
+
"rewards/accuracies": 0.9375,
|
| 398 |
+
"rewards/chosen": 0.36698901653289795,
|
| 399 |
+
"rewards/margins": 0.3141450881958008,
|
| 400 |
+
"rewards/rejected": 0.05284389108419418,
|
| 401 |
+
"step": 96
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 6.35,
|
| 405 |
+
"grad_norm": 7.087887287139893,
|
| 406 |
+
"learning_rate": 4.990043977209984e-07,
|
| 407 |
+
"logits/chosen": -2.75876522064209,
|
| 408 |
+
"logits/rejected": -2.800281047821045,
|
| 409 |
+
"logps/chosen": -124.87574005126953,
|
| 410 |
+
"logps/rejected": -99.60267639160156,
|
| 411 |
+
"loss": 0.5332,
|
| 412 |
+
"rewards/accuracies": 0.96875,
|
| 413 |
+
"rewards/chosen": 0.3760470151901245,
|
| 414 |
+
"rewards/margins": 0.37043848633766174,
|
| 415 |
+
"rewards/rejected": 0.00560858054086566,
|
| 416 |
+
"step": 100
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 6.6,
|
| 420 |
+
"grad_norm": 6.458583831787109,
|
| 421 |
+
"learning_rate": 4.989011006142024e-07,
|
| 422 |
+
"logits/chosen": -2.70744252204895,
|
| 423 |
+
"logits/rejected": -2.7285239696502686,
|
| 424 |
+
"logps/chosen": -129.41262817382812,
|
| 425 |
+
"logps/rejected": -98.60986328125,
|
| 426 |
+
"loss": 0.5286,
|
| 427 |
+
"rewards/accuracies": 0.984375,
|
| 428 |
+
"rewards/chosen": 0.3909401297569275,
|
| 429 |
+
"rewards/margins": 0.3845981955528259,
|
| 430 |
+
"rewards/rejected": 0.006341893225908279,
|
| 431 |
+
"step": 104
|
| 432 |
+
},
|
| 433 |
+
{
|
| 434 |
+
"epoch": 6.86,
|
| 435 |
+
"grad_norm": 8.084393501281738,
|
| 436 |
+
"learning_rate": 4.987927177432842e-07,
|
| 437 |
+
"logits/chosen": -2.731127977371216,
|
| 438 |
+
"logits/rejected": -2.7352988719940186,
|
| 439 |
+
"logps/chosen": -123.85149383544922,
|
| 440 |
+
"logps/rejected": -97.99423217773438,
|
| 441 |
+
"loss": 0.544,
|
| 442 |
+
"rewards/accuracies": 0.890625,
|
| 443 |
+
"rewards/chosen": 0.38825392723083496,
|
| 444 |
+
"rewards/margins": 0.34452566504478455,
|
| 445 |
+
"rewards/rejected": 0.043728262186050415,
|
| 446 |
+
"step": 108
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 7.11,
|
| 450 |
+
"grad_norm": 7.936505317687988,
|
| 451 |
+
"learning_rate": 4.987080943856886e-07,
|
| 452 |
+
"logits/chosen": -2.6931982040405273,
|
| 453 |
+
"logits/rejected": -2.683288812637329,
|
| 454 |
+
"logps/chosen": -130.97227478027344,
|
| 455 |
+
"logps/rejected": -93.71176147460938,
|
| 456 |
+
"loss": 0.4903,
|
| 457 |
+
"rewards/accuracies": 0.953125,
|
| 458 |
+
"rewards/chosen": 0.45238468050956726,
|
| 459 |
+
"rewards/margins": 0.48204928636550903,
|
| 460 |
+
"rewards/rejected": -0.029664600268006325,
|
| 461 |
+
"step": 112
|
| 462 |
+
},
|
| 463 |
+
{
|
| 464 |
+
"epoch": 7.37,
|
| 465 |
+
"grad_norm": 6.827272415161133,
|
| 466 |
+
"learning_rate": 4.985908168188602e-07,
|
| 467 |
+
"logits/chosen": -2.697653293609619,
|
| 468 |
+
"logits/rejected": -2.7467117309570312,
|
| 469 |
+
"logps/chosen": -119.16637420654297,
|
| 470 |
+
"logps/rejected": -95.0647201538086,
|
| 471 |
+
"loss": 0.5088,
|
| 472 |
+
"rewards/accuracies": 0.953125,
|
| 473 |
+
"rewards/chosen": 0.4183903634548187,
|
| 474 |
+
"rewards/margins": 0.43436184525489807,
|
| 475 |
+
"rewards/rejected": -0.015971507877111435,
|
| 476 |
+
"step": 116
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 7.62,
|
| 480 |
+
"grad_norm": 6.599144458770752,
|
| 481 |
+
"learning_rate": 4.984684598278982e-07,
|
| 482 |
+
"logits/chosen": -2.7136449813842773,
|
| 483 |
+
"logits/rejected": -2.7430026531219482,
|
| 484 |
+
"logps/chosen": -126.032958984375,
|
| 485 |
+
"logps/rejected": -102.69947814941406,
|
| 486 |
+
"loss": 0.5231,
|
| 487 |
+
"rewards/accuracies": 0.9375,
|
| 488 |
+
"rewards/chosen": 0.4575308859348297,
|
| 489 |
+
"rewards/margins": 0.40544256567955017,
|
| 490 |
+
"rewards/rejected": 0.052088312804698944,
|
| 491 |
+
"step": 120
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 7.62,
|
| 495 |
+
"eval_logits/chosen": -2.807668447494507,
|
| 496 |
+
"eval_logits/rejected": -2.8901402950286865,
|
| 497 |
+
"eval_logps/chosen": -129.97122192382812,
|
| 498 |
+
"eval_logps/rejected": -97.28166961669922,
|
| 499 |
+
"eval_loss": 0.5570675134658813,
|
| 500 |
+
"eval_rewards/accuracies": 0.7857142686843872,
|
| 501 |
+
"eval_rewards/chosen": 0.4098115563392639,
|
| 502 |
+
"eval_rewards/margins": 0.3270883560180664,
|
| 503 |
+
"eval_rewards/rejected": 0.0827232226729393,
|
| 504 |
+
"eval_runtime": 19.4265,
|
| 505 |
+
"eval_samples_per_second": 1.441,
|
| 506 |
+
"eval_steps_per_second": 1.441,
|
| 507 |
+
"step": 120
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 7.87,
|
| 511 |
+
"grad_norm": 7.769857883453369,
|
| 512 |
+
"learning_rate": 4.983410259129073e-07,
|
| 513 |
+
"logits/chosen": -2.7467093467712402,
|
| 514 |
+
"logits/rejected": -2.7454986572265625,
|
| 515 |
+
"logps/chosen": -134.60739135742188,
|
| 516 |
+
"logps/rejected": -103.20536804199219,
|
| 517 |
+
"loss": 0.4995,
|
| 518 |
+
"rewards/accuracies": 0.984375,
|
| 519 |
+
"rewards/chosen": 0.4985056221485138,
|
| 520 |
+
"rewards/margins": 0.4699307978153229,
|
| 521 |
+
"rewards/rejected": 0.028574790805578232,
|
| 522 |
+
"step": 124
|
| 523 |
+
},
|
| 524 |
+
{
|
| 525 |
+
"epoch": 8.13,
|
| 526 |
+
"grad_norm": 7.224113941192627,
|
| 527 |
+
"learning_rate": 4.982085176777285e-07,
|
| 528 |
+
"logits/chosen": -2.725543975830078,
|
| 529 |
+
"logits/rejected": -2.7536404132843018,
|
| 530 |
+
"logps/chosen": -121.35929870605469,
|
| 531 |
+
"logps/rejected": -97.49042510986328,
|
| 532 |
+
"loss": 0.4988,
|
| 533 |
+
"rewards/accuracies": 0.921875,
|
| 534 |
+
"rewards/chosen": 0.4724448025226593,
|
| 535 |
+
"rewards/margins": 0.4719679057598114,
|
| 536 |
+
"rewards/rejected": 0.0004768962971866131,
|
| 537 |
+
"step": 128
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 8.38,
|
| 541 |
+
"grad_norm": 5.916596412658691,
|
| 542 |
+
"learning_rate": 4.980709378298851e-07,
|
| 543 |
+
"logits/chosen": -2.7414817810058594,
|
| 544 |
+
"logits/rejected": -2.7378463745117188,
|
| 545 |
+
"logps/chosen": -119.76469421386719,
|
| 546 |
+
"logps/rejected": -92.40216827392578,
|
| 547 |
+
"loss": 0.4946,
|
| 548 |
+
"rewards/accuracies": 0.984375,
|
| 549 |
+
"rewards/chosen": 0.48038381338119507,
|
| 550 |
+
"rewards/margins": 0.4809207320213318,
|
| 551 |
+
"rewards/rejected": -0.0005369335412979126,
|
| 552 |
+
"step": 132
|
| 553 |
+
},
|
| 554 |
+
{
|
| 555 |
+
"epoch": 8.63,
|
| 556 |
+
"grad_norm": 6.414724826812744,
|
| 557 |
+
"learning_rate": 4.979282891805286e-07,
|
| 558 |
+
"logits/chosen": -2.7029452323913574,
|
| 559 |
+
"logits/rejected": -2.737753391265869,
|
| 560 |
+
"logps/chosen": -130.59829711914062,
|
| 561 |
+
"logps/rejected": -99.5873794555664,
|
| 562 |
+
"loss": 0.4564,
|
| 563 |
+
"rewards/accuracies": 0.96875,
|
| 564 |
+
"rewards/chosen": 0.5609064698219299,
|
| 565 |
+
"rewards/margins": 0.5927521586418152,
|
| 566 |
+
"rewards/rejected": -0.03184571862220764,
|
| 567 |
+
"step": 136
|
| 568 |
+
},
|
| 569 |
+
{
|
| 570 |
+
"epoch": 8.89,
|
| 571 |
+
"grad_norm": 7.006284236907959,
|
| 572 |
+
"learning_rate": 4.977805746443806e-07,
|
| 573 |
+
"logits/chosen": -2.719618320465088,
|
| 574 |
+
"logits/rejected": -2.7250468730926514,
|
| 575 |
+
"logps/chosen": -128.04678344726562,
|
| 576 |
+
"logps/rejected": -101.81006622314453,
|
| 577 |
+
"loss": 0.4761,
|
| 578 |
+
"rewards/accuracies": 0.96875,
|
| 579 |
+
"rewards/chosen": 0.5477940440177917,
|
| 580 |
+
"rewards/margins": 0.5303957462310791,
|
| 581 |
+
"rewards/rejected": 0.017398254945874214,
|
| 582 |
+
"step": 140
|
| 583 |
+
},
|
| 584 |
+
{
|
| 585 |
+
"epoch": 9.14,
|
| 586 |
+
"grad_norm": 7.040762901306152,
|
| 587 |
+
"learning_rate": 4.97627797239673e-07,
|
| 588 |
+
"logits/chosen": -2.6895217895507812,
|
| 589 |
+
"logits/rejected": -2.7162435054779053,
|
| 590 |
+
"logps/chosen": -133.3381805419922,
|
| 591 |
+
"logps/rejected": -102.25975799560547,
|
| 592 |
+
"loss": 0.4403,
|
| 593 |
+
"rewards/accuracies": 0.953125,
|
| 594 |
+
"rewards/chosen": 0.6185128688812256,
|
| 595 |
+
"rewards/margins": 0.6498669981956482,
|
| 596 |
+
"rewards/rejected": -0.03135409206151962,
|
| 597 |
+
"step": 144
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 9.4,
|
| 601 |
+
"grad_norm": 6.93245267868042,
|
| 602 |
+
"learning_rate": 4.974699600880869e-07,
|
| 603 |
+
"logits/chosen": -2.6911263465881348,
|
| 604 |
+
"logits/rejected": -2.7075552940368652,
|
| 605 |
+
"logps/chosen": -126.95785522460938,
|
| 606 |
+
"logps/rejected": -102.03314971923828,
|
| 607 |
+
"loss": 0.4587,
|
| 608 |
+
"rewards/accuracies": 0.96875,
|
| 609 |
+
"rewards/chosen": 0.573100745677948,
|
| 610 |
+
"rewards/margins": 0.5804403424263,
|
| 611 |
+
"rewards/rejected": -0.0073395660147070885,
|
| 612 |
+
"step": 148
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 9.65,
|
| 616 |
+
"grad_norm": 7.084275245666504,
|
| 617 |
+
"learning_rate": 4.973070664146885e-07,
|
| 618 |
+
"logits/chosen": -2.7053210735321045,
|
| 619 |
+
"logits/rejected": -2.7612156867980957,
|
| 620 |
+
"logps/chosen": -119.92796325683594,
|
| 621 |
+
"logps/rejected": -97.49836730957031,
|
| 622 |
+
"loss": 0.483,
|
| 623 |
+
"rewards/accuracies": 0.96875,
|
| 624 |
+
"rewards/chosen": 0.5761281251907349,
|
| 625 |
+
"rewards/margins": 0.5190825462341309,
|
| 626 |
+
"rewards/rejected": 0.057045597583055496,
|
| 627 |
+
"step": 152
|
| 628 |
+
},
|
| 629 |
+
{
|
| 630 |
+
"epoch": 9.9,
|
| 631 |
+
"grad_norm": 5.3888258934021,
|
| 632 |
+
"learning_rate": 4.971391195478631e-07,
|
| 633 |
+
"logits/chosen": -2.750107765197754,
|
| 634 |
+
"logits/rejected": -2.7442452907562256,
|
| 635 |
+
"logps/chosen": -128.85630798339844,
|
| 636 |
+
"logps/rejected": -96.89845275878906,
|
| 637 |
+
"loss": 0.4517,
|
| 638 |
+
"rewards/accuracies": 0.9375,
|
| 639 |
+
"rewards/chosen": 0.6420415043830872,
|
| 640 |
+
"rewards/margins": 0.6197397708892822,
|
| 641 |
+
"rewards/rejected": 0.022301698103547096,
|
| 642 |
+
"step": 156
|
| 643 |
+
},
|
| 644 |
+
{
|
| 645 |
+
"epoch": 10.16,
|
| 646 |
+
"grad_norm": 6.74313497543335,
|
| 647 |
+
"learning_rate": 4.969661229192477e-07,
|
| 648 |
+
"logits/chosen": -2.7252540588378906,
|
| 649 |
+
"logits/rejected": -2.709249258041382,
|
| 650 |
+
"logps/chosen": -126.21170043945312,
|
| 651 |
+
"logps/rejected": -105.25879669189453,
|
| 652 |
+
"loss": 0.4607,
|
| 653 |
+
"rewards/accuracies": 0.984375,
|
| 654 |
+
"rewards/chosen": 0.6133719682693481,
|
| 655 |
+
"rewards/margins": 0.6076129674911499,
|
| 656 |
+
"rewards/rejected": 0.005758981220424175,
|
| 657 |
+
"step": 160
|
| 658 |
+
},
|
| 659 |
+
{
|
| 660 |
+
"epoch": 10.16,
|
| 661 |
+
"eval_logits/chosen": -2.807128429412842,
|
| 662 |
+
"eval_logits/rejected": -2.8889076709747314,
|
| 663 |
+
"eval_logps/chosen": -128.48606872558594,
|
| 664 |
+
"eval_logps/rejected": -97.06517028808594,
|
| 665 |
+
"eval_loss": 0.5175719857215881,
|
| 666 |
+
"eval_rewards/accuracies": 0.8214285969734192,
|
| 667 |
+
"eval_rewards/chosen": 0.5583271384239197,
|
| 668 |
+
"eval_rewards/margins": 0.4539553225040436,
|
| 669 |
+
"eval_rewards/rejected": 0.10437185317277908,
|
| 670 |
+
"eval_runtime": 19.3848,
|
| 671 |
+
"eval_samples_per_second": 1.444,
|
| 672 |
+
"eval_steps_per_second": 1.444,
|
| 673 |
+
"step": 160
|
| 674 |
+
},
|
| 675 |
+
{
|
| 676 |
+
"epoch": 10.41,
|
| 677 |
+
"grad_norm": 5.443469524383545,
|
| 678 |
+
"learning_rate": 4.967880800636598e-07,
|
| 679 |
+
"logits/chosen": -2.7496747970581055,
|
| 680 |
+
"logits/rejected": -2.768813133239746,
|
| 681 |
+
"logps/chosen": -122.11544799804688,
|
| 682 |
+
"logps/rejected": -94.54540252685547,
|
| 683 |
+
"loss": 0.436,
|
| 684 |
+
"rewards/accuracies": 0.96875,
|
| 685 |
+
"rewards/chosen": 0.6616504788398743,
|
| 686 |
+
"rewards/margins": 0.67127525806427,
|
| 687 |
+
"rewards/rejected": -0.009624744765460491,
|
| 688 |
+
"step": 164
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 10.67,
|
| 692 |
+
"grad_norm": 5.551913738250732,
|
| 693 |
+
"learning_rate": 4.966049946190263e-07,
|
| 694 |
+
"logits/chosen": -2.7052814960479736,
|
| 695 |
+
"logits/rejected": -2.7216036319732666,
|
| 696 |
+
"logps/chosen": -131.8173065185547,
|
| 697 |
+
"logps/rejected": -102.79641723632812,
|
| 698 |
+
"loss": 0.4326,
|
| 699 |
+
"rewards/accuracies": 0.953125,
|
| 700 |
+
"rewards/chosen": 0.7013098001480103,
|
| 701 |
+
"rewards/margins": 0.6718720197677612,
|
| 702 |
+
"rewards/rejected": 0.02943783439695835,
|
| 703 |
+
"step": 168
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 10.92,
|
| 707 |
+
"grad_norm": 5.826549530029297,
|
| 708 |
+
"learning_rate": 4.964168703263086e-07,
|
| 709 |
+
"logits/chosen": -2.701007604598999,
|
| 710 |
+
"logits/rejected": -2.741464614868164,
|
| 711 |
+
"logps/chosen": -116.08711242675781,
|
| 712 |
+
"logps/rejected": -94.73246765136719,
|
| 713 |
+
"loss": 0.446,
|
| 714 |
+
"rewards/accuracies": 0.953125,
|
| 715 |
+
"rewards/chosen": 0.6390998363494873,
|
| 716 |
+
"rewards/margins": 0.6307864785194397,
|
| 717 |
+
"rewards/rejected": 0.008313396014273167,
|
| 718 |
+
"step": 172
|
| 719 |
+
},
|
| 720 |
+
{
|
| 721 |
+
"epoch": 11.17,
|
| 722 |
+
"grad_norm": 6.066051483154297,
|
| 723 |
+
"learning_rate": 4.962237110294259e-07,
|
| 724 |
+
"logits/chosen": -2.7041149139404297,
|
| 725 |
+
"logits/rejected": -2.731684684753418,
|
| 726 |
+
"logps/chosen": -124.84115600585938,
|
| 727 |
+
"logps/rejected": -103.8166732788086,
|
| 728 |
+
"loss": 0.4294,
|
| 729 |
+
"rewards/accuracies": 0.953125,
|
| 730 |
+
"rewards/chosen": 0.6835889220237732,
|
| 731 |
+
"rewards/margins": 0.6921644806861877,
|
| 732 |
+
"rewards/rejected": -0.00857558287680149,
|
| 733 |
+
"step": 176
|
| 734 |
+
},
|
| 735 |
+
{
|
| 736 |
+
"epoch": 11.43,
|
| 737 |
+
"grad_norm": 5.8145012855529785,
|
| 738 |
+
"learning_rate": 4.960255206751773e-07,
|
| 739 |
+
"logits/chosen": -2.7352046966552734,
|
| 740 |
+
"logits/rejected": -2.7856600284576416,
|
| 741 |
+
"logps/chosen": -123.47644805908203,
|
| 742 |
+
"logps/rejected": -93.96688079833984,
|
| 743 |
+
"loss": 0.4056,
|
| 744 |
+
"rewards/accuracies": 0.96875,
|
| 745 |
+
"rewards/chosen": 0.7594637274742126,
|
| 746 |
+
"rewards/margins": 0.7790957093238831,
|
| 747 |
+
"rewards/rejected": -0.019631966948509216,
|
| 748 |
+
"step": 180
|
| 749 |
+
},
|
| 750 |
+
{
|
| 751 |
+
"epoch": 11.68,
|
| 752 |
+
"grad_norm": 5.962335109710693,
|
| 753 |
+
"learning_rate": 4.958223033131609e-07,
|
| 754 |
+
"logits/chosen": -2.740914821624756,
|
| 755 |
+
"logits/rejected": -2.7585391998291016,
|
| 756 |
+
"logps/chosen": -124.35005950927734,
|
| 757 |
+
"logps/rejected": -105.05398559570312,
|
| 758 |
+
"loss": 0.4293,
|
| 759 |
+
"rewards/accuracies": 0.984375,
|
| 760 |
+
"rewards/chosen": 0.6812430620193481,
|
| 761 |
+
"rewards/margins": 0.6962765455245972,
|
| 762 |
+
"rewards/rejected": -0.015033497475087643,
|
| 763 |
+
"step": 184
|
| 764 |
+
},
|
| 765 |
+
{
|
| 766 |
+
"epoch": 11.94,
|
| 767 |
+
"grad_norm": 5.529379367828369,
|
| 768 |
+
"learning_rate": 4.956140630956907e-07,
|
| 769 |
+
"logits/chosen": -2.7024734020233154,
|
| 770 |
+
"logits/rejected": -2.6746203899383545,
|
| 771 |
+
"logps/chosen": -125.29593658447266,
|
| 772 |
+
"logps/rejected": -94.42231750488281,
|
| 773 |
+
"loss": 0.3963,
|
| 774 |
+
"rewards/accuracies": 0.984375,
|
| 775 |
+
"rewards/chosen": 0.7892162799835205,
|
| 776 |
+
"rewards/margins": 0.8050926327705383,
|
| 777 |
+
"rewards/rejected": -0.01587628945708275,
|
| 778 |
+
"step": 188
|
| 779 |
+
},
|
| 780 |
+
{
|
| 781 |
+
"epoch": 12.19,
|
| 782 |
+
"grad_norm": 5.774294376373291,
|
| 783 |
+
"learning_rate": 4.954008042777125e-07,
|
| 784 |
+
"logits/chosen": -2.7662813663482666,
|
| 785 |
+
"logits/rejected": -2.8167452812194824,
|
| 786 |
+
"logps/chosen": -124.10614776611328,
|
| 787 |
+
"logps/rejected": -94.9889907836914,
|
| 788 |
+
"loss": 0.3728,
|
| 789 |
+
"rewards/accuracies": 0.96875,
|
| 790 |
+
"rewards/chosen": 0.8182777166366577,
|
| 791 |
+
"rewards/margins": 0.8937400579452515,
|
| 792 |
+
"rewards/rejected": -0.07546230405569077,
|
| 793 |
+
"step": 192
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 12.44,
|
| 797 |
+
"grad_norm": 4.990906715393066,
|
| 798 |
+
"learning_rate": 4.951825312167159e-07,
|
| 799 |
+
"logits/chosen": -2.7352771759033203,
|
| 800 |
+
"logits/rejected": -2.7206149101257324,
|
| 801 |
+
"logps/chosen": -116.98539733886719,
|
| 802 |
+
"logps/rejected": -96.68843078613281,
|
| 803 |
+
"loss": 0.4435,
|
| 804 |
+
"rewards/accuracies": 0.9375,
|
| 805 |
+
"rewards/chosen": 0.7208267450332642,
|
| 806 |
+
"rewards/margins": 0.6662319302558899,
|
| 807 |
+
"rewards/rejected": 0.05459484085440636,
|
| 808 |
+
"step": 196
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 12.7,
|
| 812 |
+
"grad_norm": 7.139024257659912,
|
| 813 |
+
"learning_rate": 4.949592483726464e-07,
|
| 814 |
+
"logits/chosen": -2.728516101837158,
|
| 815 |
+
"logits/rejected": -2.732177257537842,
|
| 816 |
+
"logps/chosen": -125.51838684082031,
|
| 817 |
+
"logps/rejected": -103.7790298461914,
|
| 818 |
+
"loss": 0.391,
|
| 819 |
+
"rewards/accuracies": 0.984375,
|
| 820 |
+
"rewards/chosen": 0.7955451011657715,
|
| 821 |
+
"rewards/margins": 0.8273122310638428,
|
| 822 |
+
"rewards/rejected": -0.03176717832684517,
|
| 823 |
+
"step": 200
|
| 824 |
+
},
|
| 825 |
+
{
|
| 826 |
+
"epoch": 12.7,
|
| 827 |
+
"eval_logits/chosen": -2.8053383827209473,
|
| 828 |
+
"eval_logits/rejected": -2.8861937522888184,
|
| 829 |
+
"eval_logps/chosen": -127.08057403564453,
|
| 830 |
+
"eval_logps/rejected": -96.8802261352539,
|
| 831 |
+
"eval_loss": 0.4858916103839874,
|
| 832 |
+
"eval_rewards/accuracies": 0.8214285969734192,
|
| 833 |
+
"eval_rewards/chosen": 0.6988765597343445,
|
| 834 |
+
"eval_rewards/margins": 0.5760089755058289,
|
| 835 |
+
"eval_rewards/rejected": 0.12286762148141861,
|
| 836 |
+
"eval_runtime": 19.4173,
|
| 837 |
+
"eval_samples_per_second": 1.442,
|
| 838 |
+
"eval_steps_per_second": 1.442,
|
| 839 |
+
"step": 200
|
| 840 |
+
},
|
| 841 |
+
{
|
| 842 |
+
"epoch": 12.95,
|
| 843 |
+
"grad_norm": 5.450782299041748,
|
| 844 |
+
"learning_rate": 4.947309603078138e-07,
|
| 845 |
+
"logits/chosen": -2.6719353199005127,
|
| 846 |
+
"logits/rejected": -2.7134647369384766,
|
| 847 |
+
"logps/chosen": -124.22660827636719,
|
| 848 |
+
"logps/rejected": -99.73461151123047,
|
| 849 |
+
"loss": 0.4062,
|
| 850 |
+
"rewards/accuracies": 0.96875,
|
| 851 |
+
"rewards/chosen": 0.7935754060745239,
|
| 852 |
+
"rewards/margins": 0.783437192440033,
|
| 853 |
+
"rewards/rejected": 0.010138224810361862,
|
| 854 |
+
"step": 204
|
| 855 |
+
},
|
| 856 |
+
{
|
| 857 |
+
"epoch": 13.21,
|
| 858 |
+
"grad_norm": 5.752992153167725,
|
| 859 |
+
"learning_rate": 4.944976716867984e-07,
|
| 860 |
+
"logits/chosen": -2.7026755809783936,
|
| 861 |
+
"logits/rejected": -2.698920726776123,
|
| 862 |
+
"logps/chosen": -123.63475036621094,
|
| 863 |
+
"logps/rejected": -107.06332397460938,
|
| 864 |
+
"loss": 0.3952,
|
| 865 |
+
"rewards/accuracies": 0.96875,
|
| 866 |
+
"rewards/chosen": 0.8116875886917114,
|
| 867 |
+
"rewards/margins": 0.8335397839546204,
|
| 868 |
+
"rewards/rejected": -0.02185220457613468,
|
| 869 |
+
"step": 208
|
| 870 |
+
},
|
| 871 |
+
{
|
| 872 |
+
"epoch": 13.46,
|
| 873 |
+
"grad_norm": 5.456747531890869,
|
| 874 |
+
"learning_rate": 4.942593872763566e-07,
|
| 875 |
+
"logits/chosen": -2.701968193054199,
|
| 876 |
+
"logits/rejected": -2.786785840988159,
|
| 877 |
+
"logps/chosen": -124.91243743896484,
|
| 878 |
+
"logps/rejected": -101.95451354980469,
|
| 879 |
+
"loss": 0.3673,
|
| 880 |
+
"rewards/accuracies": 0.96875,
|
| 881 |
+
"rewards/chosen": 0.8831655979156494,
|
| 882 |
+
"rewards/margins": 0.9259040951728821,
|
| 883 |
+
"rewards/rejected": -0.04273851588368416,
|
| 884 |
+
"step": 212
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 13.71,
|
| 888 |
+
"grad_norm": 4.753533363342285,
|
| 889 |
+
"learning_rate": 4.940161119453231e-07,
|
| 890 |
+
"logits/chosen": -2.7109193801879883,
|
| 891 |
+
"logits/rejected": -2.6961448192596436,
|
| 892 |
+
"logps/chosen": -123.21333312988281,
|
| 893 |
+
"logps/rejected": -92.74653625488281,
|
| 894 |
+
"loss": 0.3832,
|
| 895 |
+
"rewards/accuracies": 0.984375,
|
| 896 |
+
"rewards/chosen": 0.8407853245735168,
|
| 897 |
+
"rewards/margins": 0.8698095083236694,
|
| 898 |
+
"rewards/rejected": -0.02902432158589363,
|
| 899 |
+
"step": 216
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 13.97,
|
| 903 |
+
"grad_norm": 5.374657154083252,
|
| 904 |
+
"learning_rate": 4.937678506645115e-07,
|
| 905 |
+
"logits/chosen": -2.7288832664489746,
|
| 906 |
+
"logits/rejected": -2.7349352836608887,
|
| 907 |
+
"logps/chosen": -123.1498031616211,
|
| 908 |
+
"logps/rejected": -95.77909851074219,
|
| 909 |
+
"loss": 0.3644,
|
| 910 |
+
"rewards/accuracies": 0.984375,
|
| 911 |
+
"rewards/chosen": 0.8925550580024719,
|
| 912 |
+
"rewards/margins": 0.9183064699172974,
|
| 913 |
+
"rewards/rejected": -0.025751333683729172,
|
| 914 |
+
"step": 220
|
| 915 |
+
},
|
| 916 |
+
{
|
| 917 |
+
"epoch": 14.22,
|
| 918 |
+
"grad_norm": 5.842701435089111,
|
| 919 |
+
"learning_rate": 4.935146085066125e-07,
|
| 920 |
+
"logits/chosen": -2.7030577659606934,
|
| 921 |
+
"logits/rejected": -2.7200875282287598,
|
| 922 |
+
"logps/chosen": -120.81951904296875,
|
| 923 |
+
"logps/rejected": -100.53494262695312,
|
| 924 |
+
"loss": 0.3581,
|
| 925 |
+
"rewards/accuracies": 0.96875,
|
| 926 |
+
"rewards/chosen": 0.9156714677810669,
|
| 927 |
+
"rewards/margins": 0.9704564809799194,
|
| 928 |
+
"rewards/rejected": -0.05478499457240105,
|
| 929 |
+
"step": 224
|
| 930 |
+
},
|
| 931 |
+
{
|
| 932 |
+
"epoch": 14.48,
|
| 933 |
+
"grad_norm": 5.020945072174072,
|
| 934 |
+
"learning_rate": 4.932563906460904e-07,
|
| 935 |
+
"logits/chosen": -2.722069501876831,
|
| 936 |
+
"logits/rejected": -2.731358051300049,
|
| 937 |
+
"logps/chosen": -123.65387725830078,
|
| 938 |
+
"logps/rejected": -103.83390808105469,
|
| 939 |
+
"loss": 0.391,
|
| 940 |
+
"rewards/accuracies": 0.96875,
|
| 941 |
+
"rewards/chosen": 0.8787897229194641,
|
| 942 |
+
"rewards/margins": 0.8743190765380859,
|
| 943 |
+
"rewards/rejected": 0.004470685496926308,
|
| 944 |
+
"step": 228
|
| 945 |
+
},
|
| 946 |
+
{
|
| 947 |
+
"epoch": 14.73,
|
| 948 |
+
"grad_norm": 5.041106700897217,
|
| 949 |
+
"learning_rate": 4.929932023590776e-07,
|
| 950 |
+
"logits/chosen": -2.735006093978882,
|
| 951 |
+
"logits/rejected": -2.7546191215515137,
|
| 952 |
+
"logps/chosen": -124.65998840332031,
|
| 953 |
+
"logps/rejected": -101.49640655517578,
|
| 954 |
+
"loss": 0.3586,
|
| 955 |
+
"rewards/accuracies": 0.96875,
|
| 956 |
+
"rewards/chosen": 0.9206160306930542,
|
| 957 |
+
"rewards/margins": 0.9452594518661499,
|
| 958 |
+
"rewards/rejected": -0.024643439799547195,
|
| 959 |
+
"step": 232
|
| 960 |
+
},
|
| 961 |
+
{
|
| 962 |
+
"epoch": 14.98,
|
| 963 |
+
"grad_norm": 5.318466663360596,
|
| 964 |
+
"learning_rate": 4.927250490232664e-07,
|
| 965 |
+
"logits/chosen": -2.6883015632629395,
|
| 966 |
+
"logits/rejected": -2.718212842941284,
|
| 967 |
+
"logps/chosen": -120.88870239257812,
|
| 968 |
+
"logps/rejected": -94.79769134521484,
|
| 969 |
+
"loss": 0.3564,
|
| 970 |
+
"rewards/accuracies": 0.96875,
|
| 971 |
+
"rewards/chosen": 0.9339860677719116,
|
| 972 |
+
"rewards/margins": 0.9901734590530396,
|
| 973 |
+
"rewards/rejected": -0.05618742108345032,
|
| 974 |
+
"step": 236
|
| 975 |
+
},
|
| 976 |
+
{
|
| 977 |
+
"epoch": 15.24,
|
| 978 |
+
"grad_norm": 5.438621997833252,
|
| 979 |
+
"learning_rate": 4.924519361177992e-07,
|
| 980 |
+
"logits/chosen": -2.7316794395446777,
|
| 981 |
+
"logits/rejected": -2.812800407409668,
|
| 982 |
+
"logps/chosen": -120.06542205810547,
|
| 983 |
+
"logps/rejected": -97.61270141601562,
|
| 984 |
+
"loss": 0.3471,
|
| 985 |
+
"rewards/accuracies": 0.96875,
|
| 986 |
+
"rewards/chosen": 0.9851706624031067,
|
| 987 |
+
"rewards/margins": 1.039989948272705,
|
| 988 |
+
"rewards/rejected": -0.05481935292482376,
|
| 989 |
+
"step": 240
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 15.24,
|
| 993 |
+
"eval_logits/chosen": -2.8022568225860596,
|
| 994 |
+
"eval_logits/rejected": -2.882326602935791,
|
| 995 |
+
"eval_logps/chosen": -125.87542724609375,
|
| 996 |
+
"eval_logps/rejected": -96.83832550048828,
|
| 997 |
+
"eval_loss": 0.46029001474380493,
|
| 998 |
+
"eval_rewards/accuracies": 0.7857142686843872,
|
| 999 |
+
"eval_rewards/chosen": 0.8193913102149963,
|
| 1000 |
+
"eval_rewards/margins": 0.6923348307609558,
|
| 1001 |
+
"eval_rewards/rejected": 0.12705650925636292,
|
| 1002 |
+
"eval_runtime": 19.4065,
|
| 1003 |
+
"eval_samples_per_second": 1.443,
|
| 1004 |
+
"eval_steps_per_second": 1.443,
|
| 1005 |
+
"step": 240
|
| 1006 |
+
},
|
| 1007 |
+
{
|
| 1008 |
+
"epoch": 15.49,
|
| 1009 |
+
"grad_norm": 5.301855087280273,
|
| 1010 |
+
"learning_rate": 4.921738692231572e-07,
|
| 1011 |
+
"logits/chosen": -2.661013126373291,
|
| 1012 |
+
"logits/rejected": -2.666978359222412,
|
| 1013 |
+
"logps/chosen": -121.69336700439453,
|
| 1014 |
+
"logps/rejected": -101.86541748046875,
|
| 1015 |
+
"loss": 0.3574,
|
| 1016 |
+
"rewards/accuracies": 0.984375,
|
| 1017 |
+
"rewards/chosen": 0.9682250022888184,
|
| 1018 |
+
"rewards/margins": 0.9734476208686829,
|
| 1019 |
+
"rewards/rejected": -0.0052225226536393166,
|
| 1020 |
+
"step": 244
|
| 1021 |
+
},
|
| 1022 |
+
{
|
| 1023 |
+
"epoch": 15.75,
|
| 1024 |
+
"grad_norm": 4.478713512420654,
|
| 1025 |
+
"learning_rate": 4.918908540210451e-07,
|
| 1026 |
+
"logits/chosen": -2.702775239944458,
|
| 1027 |
+
"logits/rejected": -2.7098357677459717,
|
| 1028 |
+
"logps/chosen": -119.93941497802734,
|
| 1029 |
+
"logps/rejected": -94.76625061035156,
|
| 1030 |
+
"loss": 0.3517,
|
| 1031 |
+
"rewards/accuracies": 0.953125,
|
| 1032 |
+
"rewards/chosen": 0.8942417502403259,
|
| 1033 |
+
"rewards/margins": 0.9958410263061523,
|
| 1034 |
+
"rewards/rejected": -0.10159924626350403,
|
| 1035 |
+
"step": 248
|
| 1036 |
+
},
|
| 1037 |
+
{
|
| 1038 |
+
"epoch": 16.0,
|
| 1039 |
+
"grad_norm": 4.398434162139893,
|
| 1040 |
+
"learning_rate": 4.916028962942762e-07,
|
| 1041 |
+
"logits/chosen": -2.7399990558624268,
|
| 1042 |
+
"logits/rejected": -2.7476425170898438,
|
| 1043 |
+
"logps/chosen": -124.2032470703125,
|
| 1044 |
+
"logps/rejected": -101.86943054199219,
|
| 1045 |
+
"loss": 0.3372,
|
| 1046 |
+
"rewards/accuracies": 0.984375,
|
| 1047 |
+
"rewards/chosen": 1.0369476079940796,
|
| 1048 |
+
"rewards/margins": 1.0664609670639038,
|
| 1049 |
+
"rewards/rejected": -0.029513416811823845,
|
| 1050 |
+
"step": 252
|
| 1051 |
+
},
|
| 1052 |
+
{
|
| 1053 |
+
"epoch": 16.25,
|
| 1054 |
+
"grad_norm": 4.430797576904297,
|
| 1055 |
+
"learning_rate": 4.913100019266536e-07,
|
| 1056 |
+
"logits/chosen": -2.6950950622558594,
|
| 1057 |
+
"logits/rejected": -2.7181689739227295,
|
| 1058 |
+
"logps/chosen": -123.97085571289062,
|
| 1059 |
+
"logps/rejected": -97.85747528076172,
|
| 1060 |
+
"loss": 0.3226,
|
| 1061 |
+
"rewards/accuracies": 0.984375,
|
| 1062 |
+
"rewards/chosen": 1.0282835960388184,
|
| 1063 |
+
"rewards/margins": 1.1182292699813843,
|
| 1064 |
+
"rewards/rejected": -0.08994561433792114,
|
| 1065 |
+
"step": 256
|
| 1066 |
+
},
|
| 1067 |
+
{
|
| 1068 |
+
"epoch": 16.51,
|
| 1069 |
+
"grad_norm": 6.383476734161377,
|
| 1070 |
+
"learning_rate": 4.910121769028503e-07,
|
| 1071 |
+
"logits/chosen": -2.7224864959716797,
|
| 1072 |
+
"logits/rejected": -2.727613925933838,
|
| 1073 |
+
"logps/chosen": -117.91517639160156,
|
| 1074 |
+
"logps/rejected": -99.52539825439453,
|
| 1075 |
+
"loss": 0.3658,
|
| 1076 |
+
"rewards/accuracies": 0.953125,
|
| 1077 |
+
"rewards/chosen": 0.9342272877693176,
|
| 1078 |
+
"rewards/margins": 0.9777376651763916,
|
| 1079 |
+
"rewards/rejected": -0.04351034015417099,
|
| 1080 |
+
"step": 260
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 16.76,
|
| 1084 |
+
"grad_norm": 4.453210353851318,
|
| 1085 |
+
"learning_rate": 4.907094273082864e-07,
|
| 1086 |
+
"logits/chosen": -2.7149100303649902,
|
| 1087 |
+
"logits/rejected": -2.710146188735962,
|
| 1088 |
+
"logps/chosen": -126.07495880126953,
|
| 1089 |
+
"logps/rejected": -100.1856689453125,
|
| 1090 |
+
"loss": 0.3162,
|
| 1091 |
+
"rewards/accuracies": 0.96875,
|
| 1092 |
+
"rewards/chosen": 1.1231064796447754,
|
| 1093 |
+
"rewards/margins": 1.175417423248291,
|
| 1094 |
+
"rewards/rejected": -0.05231098458170891,
|
| 1095 |
+
"step": 264
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 17.02,
|
| 1099 |
+
"grad_norm": 4.718712329864502,
|
| 1100 |
+
"learning_rate": 4.904017593290055e-07,
|
| 1101 |
+
"logits/chosen": -2.710063934326172,
|
| 1102 |
+
"logits/rejected": -2.7535319328308105,
|
| 1103 |
+
"logps/chosen": -117.01451873779297,
|
| 1104 |
+
"logps/rejected": -99.95758819580078,
|
| 1105 |
+
"loss": 0.322,
|
| 1106 |
+
"rewards/accuracies": 0.984375,
|
| 1107 |
+
"rewards/chosen": 1.033459186553955,
|
| 1108 |
+
"rewards/margins": 1.1137871742248535,
|
| 1109 |
+
"rewards/rejected": -0.08032786101102829,
|
| 1110 |
+
"step": 268
|
| 1111 |
+
},
|
| 1112 |
+
{
|
| 1113 |
+
"epoch": 17.27,
|
| 1114 |
+
"grad_norm": 5.164140701293945,
|
| 1115 |
+
"learning_rate": 4.900891792515478e-07,
|
| 1116 |
+
"logits/chosen": -2.7342586517333984,
|
| 1117 |
+
"logits/rejected": -2.7561776638031006,
|
| 1118 |
+
"logps/chosen": -118.80335998535156,
|
| 1119 |
+
"logps/rejected": -104.33666229248047,
|
| 1120 |
+
"loss": 0.3314,
|
| 1121 |
+
"rewards/accuracies": 0.984375,
|
| 1122 |
+
"rewards/chosen": 1.078546404838562,
|
| 1123 |
+
"rewards/margins": 1.0902469158172607,
|
| 1124 |
+
"rewards/rejected": -0.011700518429279327,
|
| 1125 |
+
"step": 272
|
| 1126 |
+
},
|
| 1127 |
+
{
|
| 1128 |
+
"epoch": 17.52,
|
| 1129 |
+
"grad_norm": 4.392764568328857,
|
| 1130 |
+
"learning_rate": 4.897716934628217e-07,
|
| 1131 |
+
"logits/chosen": -2.6796340942382812,
|
| 1132 |
+
"logits/rejected": -2.6666793823242188,
|
| 1133 |
+
"logps/chosen": -122.01671600341797,
|
| 1134 |
+
"logps/rejected": -97.74385070800781,
|
| 1135 |
+
"loss": 0.3047,
|
| 1136 |
+
"rewards/accuracies": 1.0,
|
| 1137 |
+
"rewards/chosen": 1.0594760179519653,
|
| 1138 |
+
"rewards/margins": 1.1904582977294922,
|
| 1139 |
+
"rewards/rejected": -0.1309822052717209,
|
| 1140 |
+
"step": 276
|
| 1141 |
+
},
|
| 1142 |
+
{
|
| 1143 |
+
"epoch": 17.78,
|
| 1144 |
+
"grad_norm": 4.647542953491211,
|
| 1145 |
+
"learning_rate": 4.894493084499736e-07,
|
| 1146 |
+
"logits/chosen": -2.7203075885772705,
|
| 1147 |
+
"logits/rejected": -2.7299890518188477,
|
| 1148 |
+
"logps/chosen": -120.08210754394531,
|
| 1149 |
+
"logps/rejected": -98.8072738647461,
|
| 1150 |
+
"loss": 0.3496,
|
| 1151 |
+
"rewards/accuracies": 0.921875,
|
| 1152 |
+
"rewards/chosen": 1.0658477544784546,
|
| 1153 |
+
"rewards/margins": 1.065575122833252,
|
| 1154 |
+
"rewards/rejected": 0.0002726661041378975,
|
| 1155 |
+
"step": 280
|
| 1156 |
+
},
|
| 1157 |
+
{
|
| 1158 |
+
"epoch": 17.78,
|
| 1159 |
+
"eval_logits/chosen": -2.798058271408081,
|
| 1160 |
+
"eval_logits/rejected": -2.8771989345550537,
|
| 1161 |
+
"eval_logps/chosen": -124.81001281738281,
|
| 1162 |
+
"eval_logps/rejected": -96.83968353271484,
|
| 1163 |
+
"eval_loss": 0.44096532464027405,
|
| 1164 |
+
"eval_rewards/accuracies": 0.7857142686843872,
|
| 1165 |
+
"eval_rewards/chosen": 0.9259332418441772,
|
| 1166 |
+
"eval_rewards/margins": 0.799013078212738,
|
| 1167 |
+
"eval_rewards/rejected": 0.12692023813724518,
|
| 1168 |
+
"eval_runtime": 19.3925,
|
| 1169 |
+
"eval_samples_per_second": 1.444,
|
| 1170 |
+
"eval_steps_per_second": 1.444,
|
| 1171 |
+
"step": 280
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 18.03,
|
| 1175 |
+
"grad_norm": 4.158952236175537,
|
| 1176 |
+
"learning_rate": 4.891220308002547e-07,
|
| 1177 |
+
"logits/chosen": -2.6795365810394287,
|
| 1178 |
+
"logits/rejected": -2.7270519733428955,
|
| 1179 |
+
"logps/chosen": -120.39765167236328,
|
| 1180 |
+
"logps/rejected": -100.02141571044922,
|
| 1181 |
+
"loss": 0.319,
|
| 1182 |
+
"rewards/accuracies": 0.96875,
|
| 1183 |
+
"rewards/chosen": 1.0584629774093628,
|
| 1184 |
+
"rewards/margins": 1.19419264793396,
|
| 1185 |
+
"rewards/rejected": -0.13572952151298523,
|
| 1186 |
+
"step": 284
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 18.29,
|
| 1190 |
+
"grad_norm": 4.605132579803467,
|
| 1191 |
+
"learning_rate": 4.887898672008871e-07,
|
| 1192 |
+
"logits/chosen": -2.696953296661377,
|
| 1193 |
+
"logits/rejected": -2.718316078186035,
|
| 1194 |
+
"logps/chosen": -120.32958984375,
|
| 1195 |
+
"logps/rejected": -101.76454162597656,
|
| 1196 |
+
"loss": 0.331,
|
| 1197 |
+
"rewards/accuracies": 0.953125,
|
| 1198 |
+
"rewards/chosen": 1.0935906171798706,
|
| 1199 |
+
"rewards/margins": 1.1072776317596436,
|
| 1200 |
+
"rewards/rejected": -0.013687020167708397,
|
| 1201 |
+
"step": 288
|
| 1202 |
+
},
|
| 1203 |
+
{
|
| 1204 |
+
"epoch": 18.54,
|
| 1205 |
+
"grad_norm": 5.824638843536377,
|
| 1206 |
+
"learning_rate": 4.884528244389268e-07,
|
| 1207 |
+
"logits/chosen": -2.6889617443084717,
|
| 1208 |
+
"logits/rejected": -2.690692663192749,
|
| 1209 |
+
"logps/chosen": -120.37040710449219,
|
| 1210 |
+
"logps/rejected": -98.27632141113281,
|
| 1211 |
+
"loss": 0.3204,
|
| 1212 |
+
"rewards/accuracies": 0.953125,
|
| 1213 |
+
"rewards/chosen": 1.1049582958221436,
|
| 1214 |
+
"rewards/margins": 1.175376057624817,
|
| 1215 |
+
"rewards/rejected": -0.07041767239570618,
|
| 1216 |
+
"step": 292
|
| 1217 |
+
},
|
| 1218 |
+
{
|
| 1219 |
+
"epoch": 18.79,
|
| 1220 |
+
"grad_norm": 5.431331634521484,
|
| 1221 |
+
"learning_rate": 4.881109094011253e-07,
|
| 1222 |
+
"logits/chosen": -2.740906238555908,
|
| 1223 |
+
"logits/rejected": -2.725914478302002,
|
| 1224 |
+
"logps/chosen": -124.35327911376953,
|
| 1225 |
+
"logps/rejected": -100.71537780761719,
|
| 1226 |
+
"loss": 0.2884,
|
| 1227 |
+
"rewards/accuracies": 1.0,
|
| 1228 |
+
"rewards/chosen": 1.2057325839996338,
|
| 1229 |
+
"rewards/margins": 1.3113868236541748,
|
| 1230 |
+
"rewards/rejected": -0.10565409809350967,
|
| 1231 |
+
"step": 296
|
| 1232 |
+
},
|
| 1233 |
+
{
|
| 1234 |
+
"epoch": 19.05,
|
| 1235 |
+
"grad_norm": 4.33558988571167,
|
| 1236 |
+
"learning_rate": 4.877641290737883e-07,
|
| 1237 |
+
"logits/chosen": -2.700420379638672,
|
| 1238 |
+
"logits/rejected": -2.7527971267700195,
|
| 1239 |
+
"logps/chosen": -117.01264953613281,
|
| 1240 |
+
"logps/rejected": -101.50834655761719,
|
| 1241 |
+
"loss": 0.2757,
|
| 1242 |
+
"rewards/accuracies": 1.0,
|
| 1243 |
+
"rewards/chosen": 1.150115728378296,
|
| 1244 |
+
"rewards/margins": 1.34402596950531,
|
| 1245 |
+
"rewards/rejected": -0.19391010701656342,
|
| 1246 |
+
"step": 300
|
| 1247 |
+
},
|
| 1248 |
+
{
|
| 1249 |
+
"epoch": 19.3,
|
| 1250 |
+
"grad_norm": 4.987298488616943,
|
| 1251 |
+
"learning_rate": 4.874124905426338e-07,
|
| 1252 |
+
"logits/chosen": -2.715477705001831,
|
| 1253 |
+
"logits/rejected": -2.760146141052246,
|
| 1254 |
+
"logps/chosen": -117.20336151123047,
|
| 1255 |
+
"logps/rejected": -94.89369201660156,
|
| 1256 |
+
"loss": 0.3244,
|
| 1257 |
+
"rewards/accuracies": 0.953125,
|
| 1258 |
+
"rewards/chosen": 1.107708215713501,
|
| 1259 |
+
"rewards/margins": 1.1569105386734009,
|
| 1260 |
+
"rewards/rejected": -0.04920227825641632,
|
| 1261 |
+
"step": 304
|
| 1262 |
+
},
|
| 1263 |
+
{
|
| 1264 |
+
"epoch": 19.56,
|
| 1265 |
+
"grad_norm": 4.090747833251953,
|
| 1266 |
+
"learning_rate": 4.870560009926464e-07,
|
| 1267 |
+
"logits/chosen": -2.6785225868225098,
|
| 1268 |
+
"logits/rejected": -2.6984167098999023,
|
| 1269 |
+
"logps/chosen": -112.27914428710938,
|
| 1270 |
+
"logps/rejected": -94.59768676757812,
|
| 1271 |
+
"loss": 0.2833,
|
| 1272 |
+
"rewards/accuracies": 0.984375,
|
| 1273 |
+
"rewards/chosen": 1.2028326988220215,
|
| 1274 |
+
"rewards/margins": 1.334948182106018,
|
| 1275 |
+
"rewards/rejected": -0.13211557269096375,
|
| 1276 |
+
"step": 308
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 19.81,
|
| 1280 |
+
"grad_norm": 5.0767083168029785,
|
| 1281 |
+
"learning_rate": 4.866946677079314e-07,
|
| 1282 |
+
"logits/chosen": -2.712416410446167,
|
| 1283 |
+
"logits/rejected": -2.7296035289764404,
|
| 1284 |
+
"logps/chosen": -121.65861511230469,
|
| 1285 |
+
"logps/rejected": -106.11611938476562,
|
| 1286 |
+
"loss": 0.2893,
|
| 1287 |
+
"rewards/accuracies": 0.984375,
|
| 1288 |
+
"rewards/chosen": 1.1842504739761353,
|
| 1289 |
+
"rewards/margins": 1.3191744089126587,
|
| 1290 |
+
"rewards/rejected": -0.13492396473884583,
|
| 1291 |
+
"step": 312
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 20.06,
|
| 1295 |
+
"grad_norm": 4.1403326988220215,
|
| 1296 |
+
"learning_rate": 4.863284980715648e-07,
|
| 1297 |
+
"logits/chosen": -2.658827781677246,
|
| 1298 |
+
"logits/rejected": -2.631199836730957,
|
| 1299 |
+
"logps/chosen": -129.3588409423828,
|
| 1300 |
+
"logps/rejected": -103.25465393066406,
|
| 1301 |
+
"loss": 0.2678,
|
| 1302 |
+
"rewards/accuracies": 0.984375,
|
| 1303 |
+
"rewards/chosen": 1.291969895362854,
|
| 1304 |
+
"rewards/margins": 1.4036282300949097,
|
| 1305 |
+
"rewards/rejected": -0.11165820807218552,
|
| 1306 |
+
"step": 316
|
| 1307 |
+
},
|
| 1308 |
+
{
|
| 1309 |
+
"epoch": 20.32,
|
| 1310 |
+
"grad_norm": 4.350497245788574,
|
| 1311 |
+
"learning_rate": 4.859574995654441e-07,
|
| 1312 |
+
"logits/chosen": -2.761063814163208,
|
| 1313 |
+
"logits/rejected": -2.7751922607421875,
|
| 1314 |
+
"logps/chosen": -114.76087951660156,
|
| 1315 |
+
"logps/rejected": -100.04124450683594,
|
| 1316 |
+
"loss": 0.298,
|
| 1317 |
+
"rewards/accuracies": 0.984375,
|
| 1318 |
+
"rewards/chosen": 1.1642746925354004,
|
| 1319 |
+
"rewards/margins": 1.3020979166030884,
|
| 1320 |
+
"rewards/rejected": -0.13782335817813873,
|
| 1321 |
+
"step": 320
|
| 1322 |
+
},
|
| 1323 |
+
{
|
| 1324 |
+
"epoch": 20.32,
|
| 1325 |
+
"eval_logits/chosen": -2.792503595352173,
|
| 1326 |
+
"eval_logits/rejected": -2.870570421218872,
|
| 1327 |
+
"eval_logps/chosen": -123.9360122680664,
|
| 1328 |
+
"eval_logps/rejected": -96.95860290527344,
|
| 1329 |
+
"eval_loss": 0.4256473481655121,
|
| 1330 |
+
"eval_rewards/accuracies": 0.8214285969734192,
|
| 1331 |
+
"eval_rewards/chosen": 1.0133329629898071,
|
| 1332 |
+
"eval_rewards/margins": 0.8983038663864136,
|
| 1333 |
+
"eval_rewards/rejected": 0.11502902954816818,
|
| 1334 |
+
"eval_runtime": 19.3701,
|
| 1335 |
+
"eval_samples_per_second": 1.446,
|
| 1336 |
+
"eval_steps_per_second": 1.446,
|
| 1337 |
+
"step": 320
|
| 1338 |
+
},
|
| 1339 |
+
{
|
| 1340 |
+
"epoch": 20.57,
|
| 1341 |
+
"grad_norm": 4.334249973297119,
|
| 1342 |
+
"learning_rate": 4.855816797701336e-07,
|
| 1343 |
+
"logits/chosen": -2.690117359161377,
|
| 1344 |
+
"logits/rejected": -2.7377476692199707,
|
| 1345 |
+
"logps/chosen": -116.01565551757812,
|
| 1346 |
+
"logps/rejected": -102.1231689453125,
|
| 1347 |
+
"loss": 0.2872,
|
| 1348 |
+
"rewards/accuracies": 0.953125,
|
| 1349 |
+
"rewards/chosen": 1.2394335269927979,
|
| 1350 |
+
"rewards/margins": 1.3586478233337402,
|
| 1351 |
+
"rewards/rejected": -0.11921422928571701,
|
| 1352 |
+
"step": 324
|
| 1353 |
+
},
|
| 1354 |
+
{
|
| 1355 |
+
"epoch": 20.83,
|
| 1356 |
+
"grad_norm": 4.597221851348877,
|
| 1357 |
+
"learning_rate": 4.85201046364711e-07,
|
| 1358 |
+
"logits/chosen": -2.741600751876831,
|
| 1359 |
+
"logits/rejected": -2.712954044342041,
|
| 1360 |
+
"logps/chosen": -124.32752990722656,
|
| 1361 |
+
"logps/rejected": -101.2073974609375,
|
| 1362 |
+
"loss": 0.2676,
|
| 1363 |
+
"rewards/accuracies": 0.984375,
|
| 1364 |
+
"rewards/chosen": 1.22231924533844,
|
| 1365 |
+
"rewards/margins": 1.4293614625930786,
|
| 1366 |
+
"rewards/rejected": -0.20704227685928345,
|
| 1367 |
+
"step": 328
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 21.08,
|
| 1371 |
+
"grad_norm": 3.682258367538452,
|
| 1372 |
+
"learning_rate": 4.848156071266095e-07,
|
| 1373 |
+
"logits/chosen": -2.6585566997528076,
|
| 1374 |
+
"logits/rejected": -2.704587459564209,
|
| 1375 |
+
"logps/chosen": -120.91087341308594,
|
| 1376 |
+
"logps/rejected": -98.83493041992188,
|
| 1377 |
+
"loss": 0.2705,
|
| 1378 |
+
"rewards/accuracies": 1.0,
|
| 1379 |
+
"rewards/chosen": 1.2277172803878784,
|
| 1380 |
+
"rewards/margins": 1.3790152072906494,
|
| 1381 |
+
"rewards/rejected": -0.15129785239696503,
|
| 1382 |
+
"step": 332
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 21.33,
|
| 1386 |
+
"grad_norm": 3.7775039672851562,
|
| 1387 |
+
"learning_rate": 4.844253699314596e-07,
|
| 1388 |
+
"logits/chosen": -2.7104837894439697,
|
| 1389 |
+
"logits/rejected": -2.7378451824188232,
|
| 1390 |
+
"logps/chosen": -117.22598266601562,
|
| 1391 |
+
"logps/rejected": -97.59563446044922,
|
| 1392 |
+
"loss": 0.2755,
|
| 1393 |
+
"rewards/accuracies": 1.0,
|
| 1394 |
+
"rewards/chosen": 1.2548604011535645,
|
| 1395 |
+
"rewards/margins": 1.3694878816604614,
|
| 1396 |
+
"rewards/rejected": -0.11462761461734772,
|
| 1397 |
+
"step": 336
|
| 1398 |
+
},
|
| 1399 |
+
{
|
| 1400 |
+
"epoch": 21.59,
|
| 1401 |
+
"grad_norm": 4.84611701965332,
|
| 1402 |
+
"learning_rate": 4.840303427529273e-07,
|
| 1403 |
+
"logits/chosen": -2.6837968826293945,
|
| 1404 |
+
"logits/rejected": -2.7071595191955566,
|
| 1405 |
+
"logps/chosen": -121.88581085205078,
|
| 1406 |
+
"logps/rejected": -102.45454406738281,
|
| 1407 |
+
"loss": 0.286,
|
| 1408 |
+
"rewards/accuracies": 0.96875,
|
| 1409 |
+
"rewards/chosen": 1.2159500122070312,
|
| 1410 |
+
"rewards/margins": 1.3315218687057495,
|
| 1411 |
+
"rewards/rejected": -0.11557181179523468,
|
| 1412 |
+
"step": 340
|
| 1413 |
+
},
|
| 1414 |
+
{
|
| 1415 |
+
"epoch": 21.84,
|
| 1416 |
+
"grad_norm": 3.760068655014038,
|
| 1417 |
+
"learning_rate": 4.836305336625523e-07,
|
| 1418 |
+
"logits/chosen": -2.709066390991211,
|
| 1419 |
+
"logits/rejected": -2.73175048828125,
|
| 1420 |
+
"logps/chosen": -117.1901626586914,
|
| 1421 |
+
"logps/rejected": -105.62377166748047,
|
| 1422 |
+
"loss": 0.2661,
|
| 1423 |
+
"rewards/accuracies": 0.984375,
|
| 1424 |
+
"rewards/chosen": 1.3225661516189575,
|
| 1425 |
+
"rewards/margins": 1.445369005203247,
|
| 1426 |
+
"rewards/rejected": -0.12280277907848358,
|
| 1427 |
+
"step": 344
|
| 1428 |
+
},
|
| 1429 |
+
{
|
| 1430 |
+
"epoch": 22.1,
|
| 1431 |
+
"grad_norm": 3.8488616943359375,
|
| 1432 |
+
"learning_rate": 4.832259508295822e-07,
|
| 1433 |
+
"logits/chosen": -2.7598538398742676,
|
| 1434 |
+
"logits/rejected": -2.776313066482544,
|
| 1435 |
+
"logps/chosen": -117.15644836425781,
|
| 1436 |
+
"logps/rejected": -93.95286560058594,
|
| 1437 |
+
"loss": 0.2711,
|
| 1438 |
+
"rewards/accuracies": 0.953125,
|
| 1439 |
+
"rewards/chosen": 1.3003535270690918,
|
| 1440 |
+
"rewards/margins": 1.5021978616714478,
|
| 1441 |
+
"rewards/rejected": -0.20184442400932312,
|
| 1442 |
+
"step": 348
|
| 1443 |
+
},
|
| 1444 |
+
{
|
| 1445 |
+
"epoch": 22.35,
|
| 1446 |
+
"grad_norm": 3.152104377746582,
|
| 1447 |
+
"learning_rate": 4.828166025208058e-07,
|
| 1448 |
+
"logits/chosen": -2.6503114700317383,
|
| 1449 |
+
"logits/rejected": -2.6512064933776855,
|
| 1450 |
+
"logps/chosen": -118.29179382324219,
|
| 1451 |
+
"logps/rejected": -97.83879852294922,
|
| 1452 |
+
"loss": 0.2394,
|
| 1453 |
+
"rewards/accuracies": 0.984375,
|
| 1454 |
+
"rewards/chosen": 1.2750773429870605,
|
| 1455 |
+
"rewards/margins": 1.585729718208313,
|
| 1456 |
+
"rewards/rejected": -0.31065240502357483,
|
| 1457 |
+
"step": 352
|
| 1458 |
+
},
|
| 1459 |
+
{
|
| 1460 |
+
"epoch": 22.6,
|
| 1461 |
+
"grad_norm": 4.1737165451049805,
|
| 1462 |
+
"learning_rate": 4.824024971003845e-07,
|
| 1463 |
+
"logits/chosen": -2.708979845046997,
|
| 1464 |
+
"logits/rejected": -2.718264579772949,
|
| 1465 |
+
"logps/chosen": -113.75922393798828,
|
| 1466 |
+
"logps/rejected": -100.5313720703125,
|
| 1467 |
+
"loss": 0.2626,
|
| 1468 |
+
"rewards/accuracies": 1.0,
|
| 1469 |
+
"rewards/chosen": 1.3227206468582153,
|
| 1470 |
+
"rewards/margins": 1.4637939929962158,
|
| 1471 |
+
"rewards/rejected": -0.1410733461380005,
|
| 1472 |
+
"step": 356
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 22.86,
|
| 1476 |
+
"grad_norm": 4.2186737060546875,
|
| 1477 |
+
"learning_rate": 4.819836430296808e-07,
|
| 1478 |
+
"logits/chosen": -2.688643217086792,
|
| 1479 |
+
"logits/rejected": -2.7126376628875732,
|
| 1480 |
+
"logps/chosen": -119.3924560546875,
|
| 1481 |
+
"logps/rejected": -102.88687896728516,
|
| 1482 |
+
"loss": 0.272,
|
| 1483 |
+
"rewards/accuracies": 0.984375,
|
| 1484 |
+
"rewards/chosen": 1.3319997787475586,
|
| 1485 |
+
"rewards/margins": 1.3775808811187744,
|
| 1486 |
+
"rewards/rejected": -0.04558102786540985,
|
| 1487 |
+
"step": 360
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 22.86,
|
| 1491 |
+
"eval_logits/chosen": -2.7863781452178955,
|
| 1492 |
+
"eval_logits/rejected": -2.8632652759552,
|
| 1493 |
+
"eval_logps/chosen": -123.22322082519531,
|
| 1494 |
+
"eval_logps/rejected": -97.17105865478516,
|
| 1495 |
+
"eval_loss": 0.4147447645664215,
|
| 1496 |
+
"eval_rewards/accuracies": 0.8214285969734192,
|
| 1497 |
+
"eval_rewards/chosen": 1.0846121311187744,
|
| 1498 |
+
"eval_rewards/margins": 0.9908290505409241,
|
| 1499 |
+
"eval_rewards/rejected": 0.09378316253423691,
|
| 1500 |
+
"eval_runtime": 19.3953,
|
| 1501 |
+
"eval_samples_per_second": 1.444,
|
| 1502 |
+
"eval_steps_per_second": 1.444,
|
| 1503 |
+
"step": 360
|
| 1504 |
+
},
|
| 1505 |
+
{
|
| 1506 |
+
"epoch": 23.11,
|
| 1507 |
+
"grad_norm": 3.8801398277282715,
|
| 1508 |
+
"learning_rate": 4.815600488670862e-07,
|
| 1509 |
+
"logits/chosen": -2.661937713623047,
|
| 1510 |
+
"logits/rejected": -2.69406795501709,
|
| 1511 |
+
"logps/chosen": -120.42941284179688,
|
| 1512 |
+
"logps/rejected": -107.56249237060547,
|
| 1513 |
+
"loss": 0.2463,
|
| 1514 |
+
"rewards/accuracies": 1.0,
|
| 1515 |
+
"rewards/chosen": 1.303568959236145,
|
| 1516 |
+
"rewards/margins": 1.5796161890029907,
|
| 1517 |
+
"rewards/rejected": -0.2760472595691681,
|
| 1518 |
+
"step": 364
|
| 1519 |
+
},
|
| 1520 |
+
{
|
| 1521 |
+
"epoch": 23.37,
|
| 1522 |
+
"grad_norm": 3.6953530311584473,
|
| 1523 |
+
"learning_rate": 4.811317232678456e-07,
|
| 1524 |
+
"logits/chosen": -2.696389675140381,
|
| 1525 |
+
"logits/rejected": -2.7162017822265625,
|
| 1526 |
+
"logps/chosen": -115.95695495605469,
|
| 1527 |
+
"logps/rejected": -95.26593017578125,
|
| 1528 |
+
"loss": 0.2557,
|
| 1529 |
+
"rewards/accuracies": 0.984375,
|
| 1530 |
+
"rewards/chosen": 1.3928656578063965,
|
| 1531 |
+
"rewards/margins": 1.5144150257110596,
|
| 1532 |
+
"rewards/rejected": -0.12154923379421234,
|
| 1533 |
+
"step": 368
|
| 1534 |
+
},
|
| 1535 |
+
{
|
| 1536 |
+
"epoch": 23.62,
|
| 1537 |
+
"grad_norm": 4.2311553955078125,
|
| 1538 |
+
"learning_rate": 4.806986749838806e-07,
|
| 1539 |
+
"logits/chosen": -2.735870599746704,
|
| 1540 |
+
"logits/rejected": -2.7359132766723633,
|
| 1541 |
+
"logps/chosen": -120.84663391113281,
|
| 1542 |
+
"logps/rejected": -107.17831420898438,
|
| 1543 |
+
"loss": 0.2922,
|
| 1544 |
+
"rewards/accuracies": 0.984375,
|
| 1545 |
+
"rewards/chosen": 1.2136850357055664,
|
| 1546 |
+
"rewards/margins": 1.3251327276229858,
|
| 1547 |
+
"rewards/rejected": -0.11144763976335526,
|
| 1548 |
+
"step": 372
|
| 1549 |
+
},
|
| 1550 |
+
{
|
| 1551 |
+
"epoch": 23.87,
|
| 1552 |
+
"grad_norm": 3.250326633453369,
|
| 1553 |
+
"learning_rate": 4.802609128636112e-07,
|
| 1554 |
+
"logits/chosen": -2.6809797286987305,
|
| 1555 |
+
"logits/rejected": -2.700901508331299,
|
| 1556 |
+
"logps/chosen": -121.92070007324219,
|
| 1557 |
+
"logps/rejected": -99.91931915283203,
|
| 1558 |
+
"loss": 0.2062,
|
| 1559 |
+
"rewards/accuracies": 0.984375,
|
| 1560 |
+
"rewards/chosen": 1.4374094009399414,
|
| 1561 |
+
"rewards/margins": 1.8276853561401367,
|
| 1562 |
+
"rewards/rejected": -0.3902759850025177,
|
| 1563 |
+
"step": 376
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 24.13,
|
| 1567 |
+
"grad_norm": 4.59958553314209,
|
| 1568 |
+
"learning_rate": 4.798184458517744e-07,
|
| 1569 |
+
"logits/chosen": -2.7332396507263184,
|
| 1570 |
+
"logits/rejected": -2.768554210662842,
|
| 1571 |
+
"logps/chosen": -114.1657943725586,
|
| 1572 |
+
"logps/rejected": -100.7848892211914,
|
| 1573 |
+
"loss": 0.2353,
|
| 1574 |
+
"rewards/accuracies": 1.0,
|
| 1575 |
+
"rewards/chosen": 1.4016469717025757,
|
| 1576 |
+
"rewards/margins": 1.5988764762878418,
|
| 1577 |
+
"rewards/rejected": -0.19722968339920044,
|
| 1578 |
+
"step": 380
|
| 1579 |
+
},
|
| 1580 |
+
{
|
| 1581 |
+
"epoch": 24.38,
|
| 1582 |
+
"grad_norm": 3.6598093509674072,
|
| 1583 |
+
"learning_rate": 4.793712829892414e-07,
|
| 1584 |
+
"logits/chosen": -2.7024924755096436,
|
| 1585 |
+
"logits/rejected": -2.7085440158843994,
|
| 1586 |
+
"logps/chosen": -114.50930786132812,
|
| 1587 |
+
"logps/rejected": -104.57618713378906,
|
| 1588 |
+
"loss": 0.2538,
|
| 1589 |
+
"rewards/accuracies": 1.0,
|
| 1590 |
+
"rewards/chosen": 1.4238942861557007,
|
| 1591 |
+
"rewards/margins": 1.5259473323822021,
|
| 1592 |
+
"rewards/rejected": -0.10205309092998505,
|
| 1593 |
+
"step": 384
|
| 1594 |
+
},
|
| 1595 |
+
{
|
| 1596 |
+
"epoch": 24.63,
|
| 1597 |
+
"grad_norm": 3.3857998847961426,
|
| 1598 |
+
"learning_rate": 4.789194334128338e-07,
|
| 1599 |
+
"logits/chosen": -2.7139151096343994,
|
| 1600 |
+
"logits/rejected": -2.7087197303771973,
|
| 1601 |
+
"logps/chosen": -121.57815551757812,
|
| 1602 |
+
"logps/rejected": -102.66044616699219,
|
| 1603 |
+
"loss": 0.2324,
|
| 1604 |
+
"rewards/accuracies": 1.0,
|
| 1605 |
+
"rewards/chosen": 1.3619202375411987,
|
| 1606 |
+
"rewards/margins": 1.667473316192627,
|
| 1607 |
+
"rewards/rejected": -0.305553138256073,
|
| 1608 |
+
"step": 388
|
| 1609 |
+
},
|
| 1610 |
+
{
|
| 1611 |
+
"epoch": 24.89,
|
| 1612 |
+
"grad_norm": 4.038039207458496,
|
| 1613 |
+
"learning_rate": 4.784629063551354e-07,
|
| 1614 |
+
"logits/chosen": -2.651282548904419,
|
| 1615 |
+
"logits/rejected": -2.679689884185791,
|
| 1616 |
+
"logps/chosen": -118.9875717163086,
|
| 1617 |
+
"logps/rejected": -97.62678527832031,
|
| 1618 |
+
"loss": 0.2304,
|
| 1619 |
+
"rewards/accuracies": 0.96875,
|
| 1620 |
+
"rewards/chosen": 1.3316906690597534,
|
| 1621 |
+
"rewards/margins": 1.7062404155731201,
|
| 1622 |
+
"rewards/rejected": -0.3745496869087219,
|
| 1623 |
+
"step": 392
|
| 1624 |
+
},
|
| 1625 |
+
{
|
| 1626 |
+
"epoch": 25.14,
|
| 1627 |
+
"grad_norm": 3.0911149978637695,
|
| 1628 |
+
"learning_rate": 4.780017111443047e-07,
|
| 1629 |
+
"logits/chosen": -2.718074083328247,
|
| 1630 |
+
"logits/rejected": -2.7447762489318848,
|
| 1631 |
+
"logps/chosen": -121.20558166503906,
|
| 1632 |
+
"logps/rejected": -105.63926696777344,
|
| 1633 |
+
"loss": 0.2411,
|
| 1634 |
+
"rewards/accuracies": 1.0,
|
| 1635 |
+
"rewards/chosen": 1.4981815814971924,
|
| 1636 |
+
"rewards/margins": 1.644660472869873,
|
| 1637 |
+
"rewards/rejected": -0.14647886157035828,
|
| 1638 |
+
"step": 396
|
| 1639 |
+
},
|
| 1640 |
+
{
|
| 1641 |
+
"epoch": 25.4,
|
| 1642 |
+
"grad_norm": 3.236288070678711,
|
| 1643 |
+
"learning_rate": 4.775358572038845e-07,
|
| 1644 |
+
"logits/chosen": -2.693341016769409,
|
| 1645 |
+
"logits/rejected": -2.681511878967285,
|
| 1646 |
+
"logps/chosen": -114.28794860839844,
|
| 1647 |
+
"logps/rejected": -91.69850158691406,
|
| 1648 |
+
"loss": 0.259,
|
| 1649 |
+
"rewards/accuracies": 0.984375,
|
| 1650 |
+
"rewards/chosen": 1.3284053802490234,
|
| 1651 |
+
"rewards/margins": 1.5037521123886108,
|
| 1652 |
+
"rewards/rejected": -0.17534679174423218,
|
| 1653 |
+
"step": 400
|
| 1654 |
+
},
|
| 1655 |
+
{
|
| 1656 |
+
"epoch": 25.4,
|
| 1657 |
+
"eval_logits/chosen": -2.7796428203582764,
|
| 1658 |
+
"eval_logits/rejected": -2.8552372455596924,
|
| 1659 |
+
"eval_logps/chosen": -122.60997009277344,
|
| 1660 |
+
"eval_logps/rejected": -97.46685028076172,
|
| 1661 |
+
"eval_loss": 0.4054030478000641,
|
| 1662 |
+
"eval_rewards/accuracies": 0.7857142686843872,
|
| 1663 |
+
"eval_rewards/chosen": 1.1459368467330933,
|
| 1664 |
+
"eval_rewards/margins": 1.0817322731018066,
|
| 1665 |
+
"eval_rewards/rejected": 0.06420455127954483,
|
| 1666 |
+
"eval_runtime": 19.3983,
|
| 1667 |
+
"eval_samples_per_second": 1.443,
|
| 1668 |
+
"eval_steps_per_second": 1.443,
|
| 1669 |
+
"step": 400
|
| 1670 |
+
},
|
| 1671 |
+
{
|
| 1672 |
+
"epoch": 25.65,
|
| 1673 |
+
"grad_norm": 3.2894155979156494,
|
| 1674 |
+
"learning_rate": 4.770653540526078e-07,
|
| 1675 |
+
"logits/chosen": -2.652012825012207,
|
| 1676 |
+
"logits/rejected": -2.690704345703125,
|
| 1677 |
+
"logps/chosen": -118.52864837646484,
|
| 1678 |
+
"logps/rejected": -102.0433578491211,
|
| 1679 |
+
"loss": 0.1918,
|
| 1680 |
+
"rewards/accuracies": 1.0,
|
| 1681 |
+
"rewards/chosen": 1.5184407234191895,
|
| 1682 |
+
"rewards/margins": 1.8730953931808472,
|
| 1683 |
+
"rewards/rejected": -0.3546547293663025,
|
| 1684 |
+
"step": 404
|
| 1685 |
+
},
|
| 1686 |
+
{
|
| 1687 |
+
"epoch": 25.9,
|
| 1688 |
+
"grad_norm": 4.958095550537109,
|
| 1689 |
+
"learning_rate": 4.765902113042053e-07,
|
| 1690 |
+
"logits/chosen": -2.7191715240478516,
|
| 1691 |
+
"logits/rejected": -2.701702833175659,
|
| 1692 |
+
"logps/chosen": -115.93360900878906,
|
| 1693 |
+
"logps/rejected": -100.49573516845703,
|
| 1694 |
+
"loss": 0.2536,
|
| 1695 |
+
"rewards/accuracies": 1.0,
|
| 1696 |
+
"rewards/chosen": 1.357507348060608,
|
| 1697 |
+
"rewards/margins": 1.542129635810852,
|
| 1698 |
+
"rewards/rejected": -0.18462230265140533,
|
| 1699 |
+
"step": 408
|
| 1700 |
+
},
|
| 1701 |
+
{
|
| 1702 |
+
"epoch": 26.16,
|
| 1703 |
+
"grad_norm": 3.1129214763641357,
|
| 1704 |
+
"learning_rate": 4.7611043866720737e-07,
|
| 1705 |
+
"logits/chosen": -2.6934893131256104,
|
| 1706 |
+
"logits/rejected": -2.7433109283447266,
|
| 1707 |
+
"logps/chosen": -117.78741455078125,
|
| 1708 |
+
"logps/rejected": -102.46746063232422,
|
| 1709 |
+
"loss": 0.2137,
|
| 1710 |
+
"rewards/accuracies": 0.96875,
|
| 1711 |
+
"rewards/chosen": 1.484704613685608,
|
| 1712 |
+
"rewards/margins": 1.8401193618774414,
|
| 1713 |
+
"rewards/rejected": -0.3554147481918335,
|
| 1714 |
+
"step": 412
|
| 1715 |
+
},
|
| 1716 |
+
{
|
| 1717 |
+
"epoch": 26.41,
|
| 1718 |
+
"grad_norm": 3.2823047637939453,
|
| 1719 |
+
"learning_rate": 4.7562604594474643e-07,
|
| 1720 |
+
"logits/chosen": -2.6928577423095703,
|
| 1721 |
+
"logits/rejected": -2.708937644958496,
|
| 1722 |
+
"logps/chosen": -124.94667053222656,
|
| 1723 |
+
"logps/rejected": -106.61329650878906,
|
| 1724 |
+
"loss": 0.2137,
|
| 1725 |
+
"rewards/accuracies": 1.0,
|
| 1726 |
+
"rewards/chosen": 1.471519112586975,
|
| 1727 |
+
"rewards/margins": 1.7852158546447754,
|
| 1728 |
+
"rewards/rejected": -0.31369656324386597,
|
| 1729 |
+
"step": 416
|
| 1730 |
+
},
|
| 1731 |
+
{
|
| 1732 |
+
"epoch": 26.67,
|
| 1733 |
+
"grad_norm": 3.7071120738983154,
|
| 1734 |
+
"learning_rate": 4.7513704303435674e-07,
|
| 1735 |
+
"logits/chosen": -2.670248031616211,
|
| 1736 |
+
"logits/rejected": -2.6764914989471436,
|
| 1737 |
+
"logps/chosen": -110.71532440185547,
|
| 1738 |
+
"logps/rejected": -101.84510803222656,
|
| 1739 |
+
"loss": 0.229,
|
| 1740 |
+
"rewards/accuracies": 0.984375,
|
| 1741 |
+
"rewards/chosen": 1.4525586366653442,
|
| 1742 |
+
"rewards/margins": 1.6690635681152344,
|
| 1743 |
+
"rewards/rejected": -0.21650481224060059,
|
| 1744 |
+
"step": 420
|
| 1745 |
+
},
|
| 1746 |
+
{
|
| 1747 |
+
"epoch": 26.92,
|
| 1748 |
+
"grad_norm": 3.449319362640381,
|
| 1749 |
+
"learning_rate": 4.7464343992777165e-07,
|
| 1750 |
+
"logits/chosen": -2.688829183578491,
|
| 1751 |
+
"logits/rejected": -2.7286489009857178,
|
| 1752 |
+
"logps/chosen": -115.69140625,
|
| 1753 |
+
"logps/rejected": -103.97221374511719,
|
| 1754 |
+
"loss": 0.2231,
|
| 1755 |
+
"rewards/accuracies": 1.0,
|
| 1756 |
+
"rewards/chosen": 1.4910831451416016,
|
| 1757 |
+
"rewards/margins": 1.7472798824310303,
|
| 1758 |
+
"rewards/rejected": -0.2561967670917511,
|
| 1759 |
+
"step": 424
|
| 1760 |
+
},
|
| 1761 |
+
{
|
| 1762 |
+
"epoch": 27.17,
|
| 1763 |
+
"grad_norm": 2.4958574771881104,
|
| 1764 |
+
"learning_rate": 4.741452467107199e-07,
|
| 1765 |
+
"logits/chosen": -2.6915664672851562,
|
| 1766 |
+
"logits/rejected": -2.7021422386169434,
|
| 1767 |
+
"logps/chosen": -110.91499328613281,
|
| 1768 |
+
"logps/rejected": -98.70269775390625,
|
| 1769 |
+
"loss": 0.233,
|
| 1770 |
+
"rewards/accuracies": 0.984375,
|
| 1771 |
+
"rewards/chosen": 1.4354866743087769,
|
| 1772 |
+
"rewards/margins": 1.7168078422546387,
|
| 1773 |
+
"rewards/rejected": -0.28132107853889465,
|
| 1774 |
+
"step": 428
|
| 1775 |
+
},
|
| 1776 |
+
{
|
| 1777 |
+
"epoch": 27.43,
|
| 1778 |
+
"grad_norm": 3.7154319286346436,
|
| 1779 |
+
"learning_rate": 4.7364247356271927e-07,
|
| 1780 |
+
"logits/chosen": -2.6695199012756348,
|
| 1781 |
+
"logits/rejected": -2.6698789596557617,
|
| 1782 |
+
"logps/chosen": -118.46913146972656,
|
| 1783 |
+
"logps/rejected": -102.86263275146484,
|
| 1784 |
+
"loss": 0.2337,
|
| 1785 |
+
"rewards/accuracies": 1.0,
|
| 1786 |
+
"rewards/chosen": 1.4410640001296997,
|
| 1787 |
+
"rewards/margins": 1.6770535707473755,
|
| 1788 |
+
"rewards/rejected": -0.23598948121070862,
|
| 1789 |
+
"step": 432
|
| 1790 |
+
},
|
| 1791 |
+
{
|
| 1792 |
+
"epoch": 27.68,
|
| 1793 |
+
"grad_norm": 3.101527690887451,
|
| 1794 |
+
"learning_rate": 4.731351307568687e-07,
|
| 1795 |
+
"logits/chosen": -2.6910083293914795,
|
| 1796 |
+
"logits/rejected": -2.6801414489746094,
|
| 1797 |
+
"logps/chosen": -115.44275665283203,
|
| 1798 |
+
"logps/rejected": -97.7877197265625,
|
| 1799 |
+
"loss": 0.2,
|
| 1800 |
+
"rewards/accuracies": 1.0,
|
| 1801 |
+
"rewards/chosen": 1.5193870067596436,
|
| 1802 |
+
"rewards/margins": 1.8750090599060059,
|
| 1803 |
+
"rewards/rejected": -0.35562214255332947,
|
| 1804 |
+
"step": 436
|
| 1805 |
+
},
|
| 1806 |
+
{
|
| 1807 |
+
"epoch": 27.94,
|
| 1808 |
+
"grad_norm": 3.9017672538757324,
|
| 1809 |
+
"learning_rate": 4.7262322865963846e-07,
|
| 1810 |
+
"logits/chosen": -2.7122743129730225,
|
| 1811 |
+
"logits/rejected": -2.723294258117676,
|
| 1812 |
+
"logps/chosen": -116.90724182128906,
|
| 1813 |
+
"logps/rejected": -104.45236206054688,
|
| 1814 |
+
"loss": 0.1959,
|
| 1815 |
+
"rewards/accuracies": 1.0,
|
| 1816 |
+
"rewards/chosen": 1.5187357664108276,
|
| 1817 |
+
"rewards/margins": 1.918578863143921,
|
| 1818 |
+
"rewards/rejected": -0.39984312653541565,
|
| 1819 |
+
"step": 440
|
| 1820 |
+
},
|
| 1821 |
+
{
|
| 1822 |
+
"epoch": 27.94,
|
| 1823 |
+
"eval_logits/chosen": -2.7724335193634033,
|
| 1824 |
+
"eval_logits/rejected": -2.8468399047851562,
|
| 1825 |
+
"eval_logps/chosen": -122.11509704589844,
|
| 1826 |
+
"eval_logps/rejected": -97.87715148925781,
|
| 1827 |
+
"eval_loss": 0.39711207151412964,
|
| 1828 |
+
"eval_rewards/accuracies": 0.7857142686843872,
|
| 1829 |
+
"eval_rewards/chosen": 1.1954243183135986,
|
| 1830 |
+
"eval_rewards/margins": 1.1722500324249268,
|
| 1831 |
+
"eval_rewards/rejected": 0.023174213245511055,
|
| 1832 |
+
"eval_runtime": 19.4446,
|
| 1833 |
+
"eval_samples_per_second": 1.44,
|
| 1834 |
+
"eval_steps_per_second": 1.44,
|
| 1835 |
+
"step": 440
|
| 1836 |
+
},
|
| 1837 |
+
{
|
| 1838 |
+
"epoch": 28.19,
|
| 1839 |
+
"grad_norm": 4.0338592529296875,
|
| 1840 |
+
"learning_rate": 4.721067777306581e-07,
|
| 1841 |
+
"logits/chosen": -2.6722593307495117,
|
| 1842 |
+
"logits/rejected": -2.682727575302124,
|
| 1843 |
+
"logps/chosen": -119.6982650756836,
|
| 1844 |
+
"logps/rejected": -105.52352905273438,
|
| 1845 |
+
"loss": 0.2108,
|
| 1846 |
+
"rewards/accuracies": 0.984375,
|
| 1847 |
+
"rewards/chosen": 1.5048611164093018,
|
| 1848 |
+
"rewards/margins": 1.8753631114959717,
|
| 1849 |
+
"rewards/rejected": -0.3705020546913147,
|
| 1850 |
+
"step": 444
|
| 1851 |
+
},
|
| 1852 |
+
{
|
| 1853 |
+
"epoch": 28.44,
|
| 1854 |
+
"grad_norm": 3.167048692703247,
|
| 1855 |
+
"learning_rate": 4.71585788522503e-07,
|
| 1856 |
+
"logits/chosen": -2.652644157409668,
|
| 1857 |
+
"logits/rejected": -2.6818065643310547,
|
| 1858 |
+
"logps/chosen": -113.42478942871094,
|
| 1859 |
+
"logps/rejected": -104.37007141113281,
|
| 1860 |
+
"loss": 0.2076,
|
| 1861 |
+
"rewards/accuracies": 1.0,
|
| 1862 |
+
"rewards/chosen": 1.474198341369629,
|
| 1863 |
+
"rewards/margins": 1.7973999977111816,
|
| 1864 |
+
"rewards/rejected": -0.32320165634155273,
|
| 1865 |
+
"step": 448
|
| 1866 |
+
},
|
| 1867 |
+
{
|
| 1868 |
+
"epoch": 28.7,
|
| 1869 |
+
"grad_norm": 3.4248368740081787,
|
| 1870 |
+
"learning_rate": 4.7106027168047833e-07,
|
| 1871 |
+
"logits/chosen": -2.7117884159088135,
|
| 1872 |
+
"logits/rejected": -2.77234148979187,
|
| 1873 |
+
"logps/chosen": -116.26769256591797,
|
| 1874 |
+
"logps/rejected": -101.22008514404297,
|
| 1875 |
+
"loss": 0.1858,
|
| 1876 |
+
"rewards/accuracies": 1.0,
|
| 1877 |
+
"rewards/chosen": 1.5492609739303589,
|
| 1878 |
+
"rewards/margins": 1.9960644245147705,
|
| 1879 |
+
"rewards/rejected": -0.4468035101890564,
|
| 1880 |
+
"step": 452
|
| 1881 |
+
},
|
| 1882 |
+
{
|
| 1883 |
+
"epoch": 28.95,
|
| 1884 |
+
"grad_norm": 2.855328321456909,
|
| 1885 |
+
"learning_rate": 4.705302379424022e-07,
|
| 1886 |
+
"logits/chosen": -2.6717069149017334,
|
| 1887 |
+
"logits/rejected": -2.672755718231201,
|
| 1888 |
+
"logps/chosen": -116.75204467773438,
|
| 1889 |
+
"logps/rejected": -106.22608184814453,
|
| 1890 |
+
"loss": 0.2065,
|
| 1891 |
+
"rewards/accuracies": 1.0,
|
| 1892 |
+
"rewards/chosen": 1.5143406391143799,
|
| 1893 |
+
"rewards/margins": 1.8520642518997192,
|
| 1894 |
+
"rewards/rejected": -0.33772367238998413,
|
| 1895 |
+
"step": 456
|
| 1896 |
+
},
|
| 1897 |
+
{
|
| 1898 |
+
"epoch": 29.21,
|
| 1899 |
+
"grad_norm": 2.8747594356536865,
|
| 1900 |
+
"learning_rate": 4.699956981383857e-07,
|
| 1901 |
+
"logits/chosen": -2.693591356277466,
|
| 1902 |
+
"logits/rejected": -2.662400960922241,
|
| 1903 |
+
"logps/chosen": -120.1160888671875,
|
| 1904 |
+
"logps/rejected": -91.72486877441406,
|
| 1905 |
+
"loss": 0.1874,
|
| 1906 |
+
"rewards/accuracies": 1.0,
|
| 1907 |
+
"rewards/chosen": 1.6478428840637207,
|
| 1908 |
+
"rewards/margins": 2.0628483295440674,
|
| 1909 |
+
"rewards/rejected": -0.4150054454803467,
|
| 1910 |
+
"step": 460
|
| 1911 |
+
},
|
| 1912 |
+
{
|
| 1913 |
+
"epoch": 29.46,
|
| 1914 |
+
"grad_norm": 2.8170723915100098,
|
| 1915 |
+
"learning_rate": 4.694566631906116e-07,
|
| 1916 |
+
"logits/chosen": -2.6773173809051514,
|
| 1917 |
+
"logits/rejected": -2.672825574874878,
|
| 1918 |
+
"logps/chosen": -113.59075927734375,
|
| 1919 |
+
"logps/rejected": -103.25709533691406,
|
| 1920 |
+
"loss": 0.1984,
|
| 1921 |
+
"rewards/accuracies": 1.0,
|
| 1922 |
+
"rewards/chosen": 1.4863029718399048,
|
| 1923 |
+
"rewards/margins": 1.892244577407837,
|
| 1924 |
+
"rewards/rejected": -0.405941367149353,
|
| 1925 |
+
"step": 464
|
| 1926 |
+
},
|
| 1927 |
+
{
|
| 1928 |
+
"epoch": 29.71,
|
| 1929 |
+
"grad_norm": 3.132080078125,
|
| 1930 |
+
"learning_rate": 4.689131441131118e-07,
|
| 1931 |
+
"logits/chosen": -2.6520447731018066,
|
| 1932 |
+
"logits/rejected": -2.6872832775115967,
|
| 1933 |
+
"logps/chosen": -118.07131958007812,
|
| 1934 |
+
"logps/rejected": -107.88346862792969,
|
| 1935 |
+
"loss": 0.2029,
|
| 1936 |
+
"rewards/accuracies": 0.984375,
|
| 1937 |
+
"rewards/chosen": 1.6008590459823608,
|
| 1938 |
+
"rewards/margins": 1.909461259841919,
|
| 1939 |
+
"rewards/rejected": -0.30860215425491333,
|
| 1940 |
+
"step": 468
|
| 1941 |
+
},
|
| 1942 |
+
{
|
| 1943 |
+
"epoch": 29.97,
|
| 1944 |
+
"grad_norm": 3.000974416732788,
|
| 1945 |
+
"learning_rate": 4.6836515201154135e-07,
|
| 1946 |
+
"logits/chosen": -2.709559202194214,
|
| 1947 |
+
"logits/rejected": -2.723712205886841,
|
| 1948 |
+
"logps/chosen": -112.09906768798828,
|
| 1949 |
+
"logps/rejected": -106.09660339355469,
|
| 1950 |
+
"loss": 0.2037,
|
| 1951 |
+
"rewards/accuracies": 1.0,
|
| 1952 |
+
"rewards/chosen": 1.5191339254379272,
|
| 1953 |
+
"rewards/margins": 1.8524271249771118,
|
| 1954 |
+
"rewards/rejected": -0.3332933187484741,
|
| 1955 |
+
"step": 472
|
| 1956 |
+
},
|
| 1957 |
+
{
|
| 1958 |
+
"epoch": 30.22,
|
| 1959 |
+
"grad_norm": 2.544147491455078,
|
| 1960 |
+
"learning_rate": 4.678126980829524e-07,
|
| 1961 |
+
"logits/chosen": -2.678325653076172,
|
| 1962 |
+
"logits/rejected": -2.699833631515503,
|
| 1963 |
+
"logps/chosen": -115.4670639038086,
|
| 1964 |
+
"logps/rejected": -101.5027847290039,
|
| 1965 |
+
"loss": 0.1813,
|
| 1966 |
+
"rewards/accuracies": 1.0,
|
| 1967 |
+
"rewards/chosen": 1.4956222772598267,
|
| 1968 |
+
"rewards/margins": 1.9856685400009155,
|
| 1969 |
+
"rewards/rejected": -0.4900462329387665,
|
| 1970 |
+
"step": 476
|
| 1971 |
+
},
|
| 1972 |
+
{
|
| 1973 |
+
"epoch": 30.48,
|
| 1974 |
+
"grad_norm": 2.9770796298980713,
|
| 1975 |
+
"learning_rate": 4.672557936155649e-07,
|
| 1976 |
+
"logits/chosen": -2.708721160888672,
|
| 1977 |
+
"logits/rejected": -2.728306770324707,
|
| 1978 |
+
"logps/chosen": -109.71602630615234,
|
| 1979 |
+
"logps/rejected": -98.1273422241211,
|
| 1980 |
+
"loss": 0.1946,
|
| 1981 |
+
"rewards/accuracies": 0.984375,
|
| 1982 |
+
"rewards/chosen": 1.5404164791107178,
|
| 1983 |
+
"rewards/margins": 1.9511140584945679,
|
| 1984 |
+
"rewards/rejected": -0.4106977581977844,
|
| 1985 |
+
"step": 480
|
| 1986 |
+
},
|
| 1987 |
+
{
|
| 1988 |
+
"epoch": 30.48,
|
| 1989 |
+
"eval_logits/chosen": -2.764019012451172,
|
| 1990 |
+
"eval_logits/rejected": -2.8370797634124756,
|
| 1991 |
+
"eval_logps/chosen": -121.74720001220703,
|
| 1992 |
+
"eval_logps/rejected": -98.30638122558594,
|
| 1993 |
+
"eval_loss": 0.393496572971344,
|
| 1994 |
+
"eval_rewards/accuracies": 0.7857142686843872,
|
| 1995 |
+
"eval_rewards/chosen": 1.232213020324707,
|
| 1996 |
+
"eval_rewards/margins": 1.2519627809524536,
|
| 1997 |
+
"eval_rewards/rejected": -0.019749613478779793,
|
| 1998 |
+
"eval_runtime": 19.4183,
|
| 1999 |
+
"eval_samples_per_second": 1.442,
|
| 2000 |
+
"eval_steps_per_second": 1.442,
|
| 2001 |
+
"step": 480
|
| 2002 |
+
},
|
| 2003 |
+
{
|
| 2004 |
+
"epoch": 30.73,
|
| 2005 |
+
"grad_norm": 3.826716661453247,
|
| 2006 |
+
"learning_rate": 4.6669444998853605e-07,
|
| 2007 |
+
"logits/chosen": -2.6732892990112305,
|
| 2008 |
+
"logits/rejected": -2.700220823287964,
|
| 2009 |
+
"logps/chosen": -117.65553283691406,
|
| 2010 |
+
"logps/rejected": -111.19263458251953,
|
| 2011 |
+
"loss": 0.1876,
|
| 2012 |
+
"rewards/accuracies": 1.0,
|
| 2013 |
+
"rewards/chosen": 1.6948100328445435,
|
| 2014 |
+
"rewards/margins": 2.0582737922668457,
|
| 2015 |
+
"rewards/rejected": -0.3634639382362366,
|
| 2016 |
+
"step": 484
|
| 2017 |
+
},
|
| 2018 |
+
{
|
| 2019 |
+
"epoch": 30.98,
|
| 2020 |
+
"grad_norm": 3.326704740524292,
|
| 2021 |
+
"learning_rate": 4.661286786717278e-07,
|
| 2022 |
+
"logits/chosen": -2.6515111923217773,
|
| 2023 |
+
"logits/rejected": -2.6504664421081543,
|
| 2024 |
+
"logps/chosen": -120.06623840332031,
|
| 2025 |
+
"logps/rejected": -101.47470092773438,
|
| 2026 |
+
"loss": 0.1828,
|
| 2027 |
+
"rewards/accuracies": 1.0,
|
| 2028 |
+
"rewards/chosen": 1.5970648527145386,
|
| 2029 |
+
"rewards/margins": 2.0543787479400635,
|
| 2030 |
+
"rewards/rejected": -0.45731377601623535,
|
| 2031 |
+
"step": 488
|
| 2032 |
+
},
|
| 2033 |
+
{
|
| 2034 |
+
"epoch": 31.24,
|
| 2035 |
+
"grad_norm": 2.700209856033325,
|
| 2036 |
+
"learning_rate": 4.6555849122547263e-07,
|
| 2037 |
+
"logits/chosen": -2.6469173431396484,
|
| 2038 |
+
"logits/rejected": -2.666424036026001,
|
| 2039 |
+
"logps/chosen": -111.76505279541016,
|
| 2040 |
+
"logps/rejected": -103.40081024169922,
|
| 2041 |
+
"loss": 0.185,
|
| 2042 |
+
"rewards/accuracies": 1.0,
|
| 2043 |
+
"rewards/chosen": 1.524863839149475,
|
| 2044 |
+
"rewards/margins": 1.9786373376846313,
|
| 2045 |
+
"rewards/rejected": -0.4537736773490906,
|
| 2046 |
+
"step": 492
|
| 2047 |
+
},
|
| 2048 |
+
{
|
| 2049 |
+
"epoch": 31.49,
|
| 2050 |
+
"grad_norm": 2.592606544494629,
|
| 2051 |
+
"learning_rate": 4.649838993003372e-07,
|
| 2052 |
+
"logits/chosen": -2.7192447185516357,
|
| 2053 |
+
"logits/rejected": -2.726841926574707,
|
| 2054 |
+
"logps/chosen": -112.72746276855469,
|
| 2055 |
+
"logps/rejected": -100.51376342773438,
|
| 2056 |
+
"loss": 0.1678,
|
| 2057 |
+
"rewards/accuracies": 0.984375,
|
| 2058 |
+
"rewards/chosen": 1.5847443342208862,
|
| 2059 |
+
"rewards/margins": 2.0699594020843506,
|
| 2060 |
+
"rewards/rejected": -0.4852150082588196,
|
| 2061 |
+
"step": 496
|
| 2062 |
+
},
|
| 2063 |
+
{
|
| 2064 |
+
"epoch": 31.75,
|
| 2065 |
+
"grad_norm": 2.768076181411743,
|
| 2066 |
+
"learning_rate": 4.644049146368843e-07,
|
| 2067 |
+
"logits/chosen": -2.6624693870544434,
|
| 2068 |
+
"logits/rejected": -2.677476406097412,
|
| 2069 |
+
"logps/chosen": -119.13687896728516,
|
| 2070 |
+
"logps/rejected": -103.49778747558594,
|
| 2071 |
+
"loss": 0.1873,
|
| 2072 |
+
"rewards/accuracies": 1.0,
|
| 2073 |
+
"rewards/chosen": 1.7677032947540283,
|
| 2074 |
+
"rewards/margins": 2.089017868041992,
|
| 2075 |
+
"rewards/rejected": -0.32131490111351013,
|
| 2076 |
+
"step": 500
|
| 2077 |
+
},
|
| 2078 |
+
{
|
| 2079 |
+
"epoch": 32.0,
|
| 2080 |
+
"grad_norm": 2.4647607803344727,
|
| 2081 |
+
"learning_rate": 4.638215490654329e-07,
|
| 2082 |
+
"logits/chosen": -2.6946330070495605,
|
| 2083 |
+
"logits/rejected": -2.6913187503814697,
|
| 2084 |
+
"logps/chosen": -118.64775085449219,
|
| 2085 |
+
"logps/rejected": -107.40900421142578,
|
| 2086 |
+
"loss": 0.1847,
|
| 2087 |
+
"rewards/accuracies": 1.0,
|
| 2088 |
+
"rewards/chosen": 1.5691322088241577,
|
| 2089 |
+
"rewards/margins": 2.105942964553833,
|
| 2090 |
+
"rewards/rejected": -0.5368106365203857,
|
| 2091 |
+
"step": 504
|
| 2092 |
+
},
|
| 2093 |
+
{
|
| 2094 |
+
"epoch": 32.25,
|
| 2095 |
+
"grad_norm": 3.723209857940674,
|
| 2096 |
+
"learning_rate": 4.6323381450581665e-07,
|
| 2097 |
+
"logits/chosen": -2.6984715461730957,
|
| 2098 |
+
"logits/rejected": -2.686072826385498,
|
| 2099 |
+
"logps/chosen": -118.19576263427734,
|
| 2100 |
+
"logps/rejected": -104.87760162353516,
|
| 2101 |
+
"loss": 0.1853,
|
| 2102 |
+
"rewards/accuracies": 1.0,
|
| 2103 |
+
"rewards/chosen": 1.504220724105835,
|
| 2104 |
+
"rewards/margins": 2.023249864578247,
|
| 2105 |
+
"rewards/rejected": -0.519028902053833,
|
| 2106 |
+
"step": 508
|
| 2107 |
+
},
|
| 2108 |
+
{
|
| 2109 |
+
"epoch": 32.51,
|
| 2110 |
+
"grad_norm": 2.6810081005096436,
|
| 2111 |
+
"learning_rate": 4.6264172296714e-07,
|
| 2112 |
+
"logits/chosen": -2.6358213424682617,
|
| 2113 |
+
"logits/rejected": -2.6452157497406006,
|
| 2114 |
+
"logps/chosen": -116.47501373291016,
|
| 2115 |
+
"logps/rejected": -104.9029769897461,
|
| 2116 |
+
"loss": 0.1948,
|
| 2117 |
+
"rewards/accuracies": 0.984375,
|
| 2118 |
+
"rewards/chosen": 1.6027023792266846,
|
| 2119 |
+
"rewards/margins": 2.105417251586914,
|
| 2120 |
+
"rewards/rejected": -0.5027150511741638,
|
| 2121 |
+
"step": 512
|
| 2122 |
+
},
|
| 2123 |
+
{
|
| 2124 |
+
"epoch": 32.76,
|
| 2125 |
+
"grad_norm": 2.1455435752868652,
|
| 2126 |
+
"learning_rate": 4.6204528654753304e-07,
|
| 2127 |
+
"logits/chosen": -2.692194700241089,
|
| 2128 |
+
"logits/rejected": -2.705289125442505,
|
| 2129 |
+
"logps/chosen": -113.38731384277344,
|
| 2130 |
+
"logps/rejected": -103.49237823486328,
|
| 2131 |
+
"loss": 0.1617,
|
| 2132 |
+
"rewards/accuracies": 1.0,
|
| 2133 |
+
"rewards/chosen": 1.7136329412460327,
|
| 2134 |
+
"rewards/margins": 2.1900761127471924,
|
| 2135 |
+
"rewards/rejected": -0.47644317150115967,
|
| 2136 |
+
"step": 516
|
| 2137 |
+
},
|
| 2138 |
+
{
|
| 2139 |
+
"epoch": 33.02,
|
| 2140 |
+
"grad_norm": 3.499128818511963,
|
| 2141 |
+
"learning_rate": 4.614445174339045e-07,
|
| 2142 |
+
"logits/chosen": -2.661409616470337,
|
| 2143 |
+
"logits/rejected": -2.700990915298462,
|
| 2144 |
+
"logps/chosen": -113.71514892578125,
|
| 2145 |
+
"logps/rejected": -103.91769409179688,
|
| 2146 |
+
"loss": 0.1507,
|
| 2147 |
+
"rewards/accuracies": 1.0,
|
| 2148 |
+
"rewards/chosen": 1.7427312135696411,
|
| 2149 |
+
"rewards/margins": 2.1958351135253906,
|
| 2150 |
+
"rewards/rejected": -0.45310381054878235,
|
| 2151 |
+
"step": 520
|
| 2152 |
+
},
|
| 2153 |
+
{
|
| 2154 |
+
"epoch": 33.02,
|
| 2155 |
+
"eval_logits/chosen": -2.755352020263672,
|
| 2156 |
+
"eval_logits/rejected": -2.826873540878296,
|
| 2157 |
+
"eval_logps/chosen": -121.47989654541016,
|
| 2158 |
+
"eval_logps/rejected": -98.8564453125,
|
| 2159 |
+
"eval_loss": 0.3899853527545929,
|
| 2160 |
+
"eval_rewards/accuracies": 0.7857142686843872,
|
| 2161 |
+
"eval_rewards/chosen": 1.2589442729949951,
|
| 2162 |
+
"eval_rewards/margins": 1.333700180053711,
|
| 2163 |
+
"eval_rewards/rejected": -0.07475597411394119,
|
| 2164 |
+
"eval_runtime": 19.3899,
|
| 2165 |
+
"eval_samples_per_second": 1.444,
|
| 2166 |
+
"eval_steps_per_second": 1.444,
|
| 2167 |
+
"step": 520
|
| 2168 |
+
},
|
| 2169 |
+
{
|
| 2170 |
+
"epoch": 33.27,
|
| 2171 |
+
"grad_norm": 2.419506549835205,
|
| 2172 |
+
"learning_rate": 4.6083942790169206e-07,
|
| 2173 |
+
"logits/chosen": -2.6735386848449707,
|
| 2174 |
+
"logits/rejected": -2.696384906768799,
|
| 2175 |
+
"logps/chosen": -111.79097747802734,
|
| 2176 |
+
"logps/rejected": -98.7368392944336,
|
| 2177 |
+
"loss": 0.1581,
|
| 2178 |
+
"rewards/accuracies": 0.984375,
|
| 2179 |
+
"rewards/chosen": 1.5749679803848267,
|
| 2180 |
+
"rewards/margins": 2.265718460083008,
|
| 2181 |
+
"rewards/rejected": -0.690750777721405,
|
| 2182 |
+
"step": 524
|
| 2183 |
+
},
|
| 2184 |
+
{
|
| 2185 |
+
"epoch": 33.52,
|
| 2186 |
+
"grad_norm": 2.1046078205108643,
|
| 2187 |
+
"learning_rate": 4.6023003031461227e-07,
|
| 2188 |
+
"logits/chosen": -2.6465351581573486,
|
| 2189 |
+
"logits/rejected": -2.6622884273529053,
|
| 2190 |
+
"logps/chosen": -114.73282623291016,
|
| 2191 |
+
"logps/rejected": -103.12973022460938,
|
| 2192 |
+
"loss": 0.1778,
|
| 2193 |
+
"rewards/accuracies": 1.0,
|
| 2194 |
+
"rewards/chosen": 1.6710339784622192,
|
| 2195 |
+
"rewards/margins": 2.119300365447998,
|
| 2196 |
+
"rewards/rejected": -0.4482663571834564,
|
| 2197 |
+
"step": 528
|
| 2198 |
+
},
|
| 2199 |
+
{
|
| 2200 |
+
"epoch": 33.78,
|
| 2201 |
+
"grad_norm": 3.0457003116607666,
|
| 2202 |
+
"learning_rate": 4.596163371244076e-07,
|
| 2203 |
+
"logits/chosen": -2.6422927379608154,
|
| 2204 |
+
"logits/rejected": -2.6462604999542236,
|
| 2205 |
+
"logps/chosen": -114.708740234375,
|
| 2206 |
+
"logps/rejected": -105.07784271240234,
|
| 2207 |
+
"loss": 0.16,
|
| 2208 |
+
"rewards/accuracies": 1.0,
|
| 2209 |
+
"rewards/chosen": 1.7936890125274658,
|
| 2210 |
+
"rewards/margins": 2.2034826278686523,
|
| 2211 |
+
"rewards/rejected": -0.40979355573654175,
|
| 2212 |
+
"step": 532
|
| 2213 |
+
},
|
| 2214 |
+
{
|
| 2215 |
+
"epoch": 34.03,
|
| 2216 |
+
"grad_norm": 2.8257133960723877,
|
| 2217 |
+
"learning_rate": 4.5899836087059174e-07,
|
| 2218 |
+
"logits/chosen": -2.6918063163757324,
|
| 2219 |
+
"logits/rejected": -2.696005344390869,
|
| 2220 |
+
"logps/chosen": -117.54481506347656,
|
| 2221 |
+
"logps/rejected": -111.74267578125,
|
| 2222 |
+
"loss": 0.1753,
|
| 2223 |
+
"rewards/accuracies": 1.0,
|
| 2224 |
+
"rewards/chosen": 1.625847339630127,
|
| 2225 |
+
"rewards/margins": 2.2450764179229736,
|
| 2226 |
+
"rewards/rejected": -0.6192291975021362,
|
| 2227 |
+
"step": 536
|
| 2228 |
+
},
|
| 2229 |
+
{
|
| 2230 |
+
"epoch": 34.29,
|
| 2231 |
+
"grad_norm": 2.5381155014038086,
|
| 2232 |
+
"learning_rate": 4.58376114180194e-07,
|
| 2233 |
+
"logits/chosen": -2.684847116470337,
|
| 2234 |
+
"logits/rejected": -2.6875247955322266,
|
| 2235 |
+
"logps/chosen": -121.1004867553711,
|
| 2236 |
+
"logps/rejected": -101.69570922851562,
|
| 2237 |
+
"loss": 0.1562,
|
| 2238 |
+
"rewards/accuracies": 0.984375,
|
| 2239 |
+
"rewards/chosen": 1.6668047904968262,
|
| 2240 |
+
"rewards/margins": 2.3632285594940186,
|
| 2241 |
+
"rewards/rejected": -0.6964237689971924,
|
| 2242 |
+
"step": 540
|
| 2243 |
+
},
|
| 2244 |
+
{
|
| 2245 |
+
"epoch": 34.54,
|
| 2246 |
+
"grad_norm": 2.519352912902832,
|
| 2247 |
+
"learning_rate": 4.577496097675009e-07,
|
| 2248 |
+
"logits/chosen": -2.6664843559265137,
|
| 2249 |
+
"logits/rejected": -2.681034803390503,
|
| 2250 |
+
"logps/chosen": -115.94261169433594,
|
| 2251 |
+
"logps/rejected": -108.1679458618164,
|
| 2252 |
+
"loss": 0.1516,
|
| 2253 |
+
"rewards/accuracies": 1.0,
|
| 2254 |
+
"rewards/chosen": 1.7702218294143677,
|
| 2255 |
+
"rewards/margins": 2.2215099334716797,
|
| 2256 |
+
"rewards/rejected": -0.4512881338596344,
|
| 2257 |
+
"step": 544
|
| 2258 |
+
},
|
| 2259 |
+
{
|
| 2260 |
+
"epoch": 34.79,
|
| 2261 |
+
"grad_norm": 2.789398431777954,
|
| 2262 |
+
"learning_rate": 4.571188604337962e-07,
|
| 2263 |
+
"logits/chosen": -2.668377637863159,
|
| 2264 |
+
"logits/rejected": -2.688797950744629,
|
| 2265 |
+
"logps/chosen": -108.2844467163086,
|
| 2266 |
+
"logps/rejected": -104.94053649902344,
|
| 2267 |
+
"loss": 0.1838,
|
| 2268 |
+
"rewards/accuracies": 1.0,
|
| 2269 |
+
"rewards/chosen": 1.559061050415039,
|
| 2270 |
+
"rewards/margins": 2.064011812210083,
|
| 2271 |
+
"rewards/rejected": -0.504950761795044,
|
| 2272 |
+
"step": 548
|
| 2273 |
+
},
|
| 2274 |
+
{
|
| 2275 |
+
"epoch": 35.05,
|
| 2276 |
+
"grad_norm": 3.570829153060913,
|
| 2277 |
+
"learning_rate": 4.5648387906709995e-07,
|
| 2278 |
+
"logits/chosen": -2.6559903621673584,
|
| 2279 |
+
"logits/rejected": -2.650508165359497,
|
| 2280 |
+
"logps/chosen": -112.94985961914062,
|
| 2281 |
+
"logps/rejected": -101.36627197265625,
|
| 2282 |
+
"loss": 0.1484,
|
| 2283 |
+
"rewards/accuracies": 1.0,
|
| 2284 |
+
"rewards/chosen": 1.709464192390442,
|
| 2285 |
+
"rewards/margins": 2.303464889526367,
|
| 2286 |
+
"rewards/rejected": -0.5940006375312805,
|
| 2287 |
+
"step": 552
|
| 2288 |
+
},
|
| 2289 |
+
{
|
| 2290 |
+
"epoch": 35.3,
|
| 2291 |
+
"grad_norm": 2.487671136856079,
|
| 2292 |
+
"learning_rate": 4.558446786419045e-07,
|
| 2293 |
+
"logits/chosen": -2.6454193592071533,
|
| 2294 |
+
"logits/rejected": -2.667191982269287,
|
| 2295 |
+
"logps/chosen": -112.91888427734375,
|
| 2296 |
+
"logps/rejected": -100.34318542480469,
|
| 2297 |
+
"loss": 0.1571,
|
| 2298 |
+
"rewards/accuracies": 0.984375,
|
| 2299 |
+
"rewards/chosen": 1.6041696071624756,
|
| 2300 |
+
"rewards/margins": 2.3108744621276855,
|
| 2301 |
+
"rewards/rejected": -0.7067050337791443,
|
| 2302 |
+
"step": 556
|
| 2303 |
+
},
|
| 2304 |
+
{
|
| 2305 |
+
"epoch": 35.56,
|
| 2306 |
+
"grad_norm": 2.396897315979004,
|
| 2307 |
+
"learning_rate": 4.552012722189099e-07,
|
| 2308 |
+
"logits/chosen": -2.6444239616394043,
|
| 2309 |
+
"logits/rejected": -2.6464180946350098,
|
| 2310 |
+
"logps/chosen": -115.7559814453125,
|
| 2311 |
+
"logps/rejected": -111.63291931152344,
|
| 2312 |
+
"loss": 0.1752,
|
| 2313 |
+
"rewards/accuracies": 1.0,
|
| 2314 |
+
"rewards/chosen": 1.7378344535827637,
|
| 2315 |
+
"rewards/margins": 2.1783785820007324,
|
| 2316 |
+
"rewards/rejected": -0.4405441880226135,
|
| 2317 |
+
"step": 560
|
| 2318 |
+
},
|
| 2319 |
+
{
|
| 2320 |
+
"epoch": 35.56,
|
| 2321 |
+
"eval_logits/chosen": -2.746135711669922,
|
| 2322 |
+
"eval_logits/rejected": -2.8161184787750244,
|
| 2323 |
+
"eval_logps/chosen": -121.27652740478516,
|
| 2324 |
+
"eval_logps/rejected": -99.45245361328125,
|
| 2325 |
+
"eval_loss": 0.38749733567237854,
|
| 2326 |
+
"eval_rewards/accuracies": 0.7857142686843872,
|
| 2327 |
+
"eval_rewards/chosen": 1.2792812585830688,
|
| 2328 |
+
"eval_rewards/margins": 1.4136368036270142,
|
| 2329 |
+
"eval_rewards/rejected": -0.1343555599451065,
|
| 2330 |
+
"eval_runtime": 19.3831,
|
| 2331 |
+
"eval_samples_per_second": 1.445,
|
| 2332 |
+
"eval_steps_per_second": 1.445,
|
| 2333 |
+
"step": 560
|
| 2334 |
+
}
|
| 2335 |
+
],
|
| 2336 |
+
"logging_steps": 4,
|
| 2337 |
+
"max_steps": 2800,
|
| 2338 |
+
"num_input_tokens_seen": 0,
|
| 2339 |
+
"num_train_epochs": 187,
|
| 2340 |
+
"save_steps": 40,
|
| 2341 |
+
"total_flos": 0.0,
|
| 2342 |
+
"train_batch_size": 4,
|
| 2343 |
+
"trial_name": null,
|
| 2344 |
+
"trial_params": null
|
| 2345 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3e8c673aef5ce58206d2d5a710f6e8ef65196ef39b6515dd2d04f3242c26dcc4
|
| 3 |
+
size 4856
|