Upload 12 files
Browse files- checkpoint-1600/README.md +202 -0
- checkpoint-1600/adapter_config.json +34 -0
- checkpoint-1600/adapter_model.safetensors +3 -0
- checkpoint-1600/optimizer.pt +3 -0
- checkpoint-1600/rng_state.pth +3 -0
- checkpoint-1600/scheduler.pt +3 -0
- checkpoint-1600/special_tokens_map.json +24 -0
- checkpoint-1600/tokenizer.json +0 -0
- checkpoint-1600/tokenizer.model +3 -0
- checkpoint-1600/tokenizer_config.json +45 -0
- checkpoint-1600/trainer_state.json +2434 -0
- checkpoint-1600/training_args.bin +3 -0
checkpoint-1600/README.md
ADDED
|
@@ -0,0 +1,202 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: mistralai/Mistral-7B-Instruct-v0.2
|
| 3 |
+
library_name: peft
|
| 4 |
+
---
|
| 5 |
+
|
| 6 |
+
# Model Card for Model ID
|
| 7 |
+
|
| 8 |
+
<!-- Provide a quick summary of what the model is/does. -->
|
| 9 |
+
|
| 10 |
+
|
| 11 |
+
|
| 12 |
+
## Model Details
|
| 13 |
+
|
| 14 |
+
### Model Description
|
| 15 |
+
|
| 16 |
+
<!-- Provide a longer summary of what this model is. -->
|
| 17 |
+
|
| 18 |
+
|
| 19 |
+
|
| 20 |
+
- **Developed by:** [More Information Needed]
|
| 21 |
+
- **Funded by [optional]:** [More Information Needed]
|
| 22 |
+
- **Shared by [optional]:** [More Information Needed]
|
| 23 |
+
- **Model type:** [More Information Needed]
|
| 24 |
+
- **Language(s) (NLP):** [More Information Needed]
|
| 25 |
+
- **License:** [More Information Needed]
|
| 26 |
+
- **Finetuned from model [optional]:** [More Information Needed]
|
| 27 |
+
|
| 28 |
+
### Model Sources [optional]
|
| 29 |
+
|
| 30 |
+
<!-- Provide the basic links for the model. -->
|
| 31 |
+
|
| 32 |
+
- **Repository:** [More Information Needed]
|
| 33 |
+
- **Paper [optional]:** [More Information Needed]
|
| 34 |
+
- **Demo [optional]:** [More Information Needed]
|
| 35 |
+
|
| 36 |
+
## Uses
|
| 37 |
+
|
| 38 |
+
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
|
| 39 |
+
|
| 40 |
+
### Direct Use
|
| 41 |
+
|
| 42 |
+
<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
|
| 43 |
+
|
| 44 |
+
[More Information Needed]
|
| 45 |
+
|
| 46 |
+
### Downstream Use [optional]
|
| 47 |
+
|
| 48 |
+
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
|
| 49 |
+
|
| 50 |
+
[More Information Needed]
|
| 51 |
+
|
| 52 |
+
### Out-of-Scope Use
|
| 53 |
+
|
| 54 |
+
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
|
| 55 |
+
|
| 56 |
+
[More Information Needed]
|
| 57 |
+
|
| 58 |
+
## Bias, Risks, and Limitations
|
| 59 |
+
|
| 60 |
+
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
|
| 61 |
+
|
| 62 |
+
[More Information Needed]
|
| 63 |
+
|
| 64 |
+
### Recommendations
|
| 65 |
+
|
| 66 |
+
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
|
| 67 |
+
|
| 68 |
+
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
|
| 69 |
+
|
| 70 |
+
## How to Get Started with the Model
|
| 71 |
+
|
| 72 |
+
Use the code below to get started with the model.
|
| 73 |
+
|
| 74 |
+
[More Information Needed]
|
| 75 |
+
|
| 76 |
+
## Training Details
|
| 77 |
+
|
| 78 |
+
### Training Data
|
| 79 |
+
|
| 80 |
+
<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
|
| 81 |
+
|
| 82 |
+
[More Information Needed]
|
| 83 |
+
|
| 84 |
+
### Training Procedure
|
| 85 |
+
|
| 86 |
+
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
|
| 87 |
+
|
| 88 |
+
#### Preprocessing [optional]
|
| 89 |
+
|
| 90 |
+
[More Information Needed]
|
| 91 |
+
|
| 92 |
+
|
| 93 |
+
#### Training Hyperparameters
|
| 94 |
+
|
| 95 |
+
- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
|
| 96 |
+
|
| 97 |
+
#### Speeds, Sizes, Times [optional]
|
| 98 |
+
|
| 99 |
+
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
|
| 100 |
+
|
| 101 |
+
[More Information Needed]
|
| 102 |
+
|
| 103 |
+
## Evaluation
|
| 104 |
+
|
| 105 |
+
<!-- This section describes the evaluation protocols and provides the results. -->
|
| 106 |
+
|
| 107 |
+
### Testing Data, Factors & Metrics
|
| 108 |
+
|
| 109 |
+
#### Testing Data
|
| 110 |
+
|
| 111 |
+
<!-- This should link to a Dataset Card if possible. -->
|
| 112 |
+
|
| 113 |
+
[More Information Needed]
|
| 114 |
+
|
| 115 |
+
#### Factors
|
| 116 |
+
|
| 117 |
+
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
|
| 118 |
+
|
| 119 |
+
[More Information Needed]
|
| 120 |
+
|
| 121 |
+
#### Metrics
|
| 122 |
+
|
| 123 |
+
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
|
| 124 |
+
|
| 125 |
+
[More Information Needed]
|
| 126 |
+
|
| 127 |
+
### Results
|
| 128 |
+
|
| 129 |
+
[More Information Needed]
|
| 130 |
+
|
| 131 |
+
#### Summary
|
| 132 |
+
|
| 133 |
+
|
| 134 |
+
|
| 135 |
+
## Model Examination [optional]
|
| 136 |
+
|
| 137 |
+
<!-- Relevant interpretability work for the model goes here -->
|
| 138 |
+
|
| 139 |
+
[More Information Needed]
|
| 140 |
+
|
| 141 |
+
## Environmental Impact
|
| 142 |
+
|
| 143 |
+
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
|
| 144 |
+
|
| 145 |
+
Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
|
| 146 |
+
|
| 147 |
+
- **Hardware Type:** [More Information Needed]
|
| 148 |
+
- **Hours used:** [More Information Needed]
|
| 149 |
+
- **Cloud Provider:** [More Information Needed]
|
| 150 |
+
- **Compute Region:** [More Information Needed]
|
| 151 |
+
- **Carbon Emitted:** [More Information Needed]
|
| 152 |
+
|
| 153 |
+
## Technical Specifications [optional]
|
| 154 |
+
|
| 155 |
+
### Model Architecture and Objective
|
| 156 |
+
|
| 157 |
+
[More Information Needed]
|
| 158 |
+
|
| 159 |
+
### Compute Infrastructure
|
| 160 |
+
|
| 161 |
+
[More Information Needed]
|
| 162 |
+
|
| 163 |
+
#### Hardware
|
| 164 |
+
|
| 165 |
+
[More Information Needed]
|
| 166 |
+
|
| 167 |
+
#### Software
|
| 168 |
+
|
| 169 |
+
[More Information Needed]
|
| 170 |
+
|
| 171 |
+
## Citation [optional]
|
| 172 |
+
|
| 173 |
+
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
|
| 174 |
+
|
| 175 |
+
**BibTeX:**
|
| 176 |
+
|
| 177 |
+
[More Information Needed]
|
| 178 |
+
|
| 179 |
+
**APA:**
|
| 180 |
+
|
| 181 |
+
[More Information Needed]
|
| 182 |
+
|
| 183 |
+
## Glossary [optional]
|
| 184 |
+
|
| 185 |
+
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
|
| 186 |
+
|
| 187 |
+
[More Information Needed]
|
| 188 |
+
|
| 189 |
+
## More Information [optional]
|
| 190 |
+
|
| 191 |
+
[More Information Needed]
|
| 192 |
+
|
| 193 |
+
## Model Card Authors [optional]
|
| 194 |
+
|
| 195 |
+
[More Information Needed]
|
| 196 |
+
|
| 197 |
+
## Model Card Contact
|
| 198 |
+
|
| 199 |
+
[More Information Needed]
|
| 200 |
+
### Framework versions
|
| 201 |
+
|
| 202 |
+
- PEFT 0.15.1
|
checkpoint-1600/adapter_config.json
ADDED
|
@@ -0,0 +1,34 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"alpha_pattern": {},
|
| 3 |
+
"auto_mapping": null,
|
| 4 |
+
"base_model_name_or_path": "mistralai/Mistral-7B-Instruct-v0.2",
|
| 5 |
+
"bias": "none",
|
| 6 |
+
"corda_config": null,
|
| 7 |
+
"eva_config": null,
|
| 8 |
+
"exclude_modules": null,
|
| 9 |
+
"fan_in_fan_out": false,
|
| 10 |
+
"inference_mode": true,
|
| 11 |
+
"init_lora_weights": true,
|
| 12 |
+
"layer_replication": null,
|
| 13 |
+
"layers_pattern": null,
|
| 14 |
+
"layers_to_transform": null,
|
| 15 |
+
"loftq_config": {},
|
| 16 |
+
"lora_alpha": 16,
|
| 17 |
+
"lora_bias": false,
|
| 18 |
+
"lora_dropout": 0.1,
|
| 19 |
+
"megatron_config": null,
|
| 20 |
+
"megatron_core": "megatron.core",
|
| 21 |
+
"modules_to_save": null,
|
| 22 |
+
"peft_type": "LORA",
|
| 23 |
+
"r": 8,
|
| 24 |
+
"rank_pattern": {},
|
| 25 |
+
"revision": null,
|
| 26 |
+
"target_modules": [
|
| 27 |
+
"v_proj",
|
| 28 |
+
"q_proj"
|
| 29 |
+
],
|
| 30 |
+
"task_type": "CAUSAL_LM",
|
| 31 |
+
"trainable_token_indices": null,
|
| 32 |
+
"use_dora": false,
|
| 33 |
+
"use_rslora": false
|
| 34 |
+
}
|
checkpoint-1600/adapter_model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fb878d00d540ea8298f72ef0e47c80d31d3d5c16795eccf461ec9e1b06d0d6b7
|
| 3 |
+
size 13648432
|
checkpoint-1600/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:79e355c6c356cfad3ca0d90382e6ba038fd1ce69830fbb320570a579f166c7b3
|
| 3 |
+
size 27370618
|
checkpoint-1600/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5ef05c858881145e0776217dd0bad8316dc1ad872cd5b3f91be59fc1e7ea25e8
|
| 3 |
+
size 14244
|
checkpoint-1600/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f75ff008f7073dd2cf931e55b2b4f8abaf96e3b6fc246ed9c29e8fd8110363a6
|
| 3 |
+
size 1064
|
checkpoint-1600/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 |
+
}
|
checkpoint-1600/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-1600/tokenizer.model
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:dadfd56d766715c61d2ef780a525ab43b8e6da4de6865bda3d95fdef5e134055
|
| 3 |
+
size 493443
|
checkpoint-1600/tokenizer_config.json
ADDED
|
@@ -0,0 +1,45 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_bos_token": true,
|
| 3 |
+
"add_eos_token": false,
|
| 4 |
+
"add_prefix_space": null,
|
| 5 |
+
"added_tokens_decoder": {
|
| 6 |
+
"0": {
|
| 7 |
+
"content": "<unk>",
|
| 8 |
+
"lstrip": false,
|
| 9 |
+
"normalized": false,
|
| 10 |
+
"rstrip": false,
|
| 11 |
+
"single_word": false,
|
| 12 |
+
"special": true
|
| 13 |
+
},
|
| 14 |
+
"1": {
|
| 15 |
+
"content": "<s>",
|
| 16 |
+
"lstrip": false,
|
| 17 |
+
"normalized": false,
|
| 18 |
+
"rstrip": false,
|
| 19 |
+
"single_word": false,
|
| 20 |
+
"special": true
|
| 21 |
+
},
|
| 22 |
+
"2": {
|
| 23 |
+
"content": "</s>",
|
| 24 |
+
"lstrip": false,
|
| 25 |
+
"normalized": false,
|
| 26 |
+
"rstrip": false,
|
| 27 |
+
"single_word": false,
|
| 28 |
+
"special": true
|
| 29 |
+
}
|
| 30 |
+
},
|
| 31 |
+
"additional_special_tokens": [],
|
| 32 |
+
"bos_token": "<s>",
|
| 33 |
+
"chat_template": "{%- if messages[0]['role'] == 'system' %}\n {%- set system_message = messages[0]['content'] %}\n {%- set loop_messages = messages[1:] %}\n{%- else %}\n {%- set loop_messages = messages %}\n{%- endif %}\n\n{{- bos_token }}\n{%- for message in loop_messages %}\n {%- if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}\n {{- raise_exception('After the optional system message, conversation roles must alternate user/assistant/user/assistant/...') }}\n {%- endif %}\n {%- if message['role'] == 'user' %}\n {%- if loop.first and system_message is defined %}\n {{- ' [INST] ' + system_message + '\\n\\n' + message['content'] + ' [/INST]' }}\n {%- else %}\n {{- ' [INST] ' + message['content'] + ' [/INST]' }}\n {%- endif %}\n {%- elif message['role'] == 'assistant' %}\n {{- ' ' + message['content'] + eos_token}}\n {%- else %}\n {{- raise_exception('Only user and assistant roles are supported, with the exception of an initial optional system message!') }}\n {%- endif %}\n{%- endfor %}\n",
|
| 34 |
+
"clean_up_tokenization_spaces": false,
|
| 35 |
+
"eos_token": "</s>",
|
| 36 |
+
"extra_special_tokens": {},
|
| 37 |
+
"legacy": false,
|
| 38 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 39 |
+
"pad_token": "</s>",
|
| 40 |
+
"sp_model_kwargs": {},
|
| 41 |
+
"spaces_between_special_tokens": false,
|
| 42 |
+
"tokenizer_class": "LlamaTokenizer",
|
| 43 |
+
"unk_token": "<unk>",
|
| 44 |
+
"use_default_system_prompt": false
|
| 45 |
+
}
|
checkpoint-1600/trainer_state.json
ADDED
|
@@ -0,0 +1,2434 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 40.0,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 1600,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.25,
|
| 14 |
+
"grad_norm": 26.865100860595703,
|
| 15 |
+
"learning_rate": 6.25e-07,
|
| 16 |
+
"logits/chosen": -3.0250370502471924,
|
| 17 |
+
"logits/rejected": -3.0283076763153076,
|
| 18 |
+
"logps/chosen": -291.21490478515625,
|
| 19 |
+
"logps/rejected": -308.63470458984375,
|
| 20 |
+
"loss": 0.6929,
|
| 21 |
+
"rewards/accuracies": 0.5,
|
| 22 |
+
"rewards/chosen": 0.0004953003372065723,
|
| 23 |
+
"rewards/margins": 0.00046472548274323344,
|
| 24 |
+
"rewards/rejected": 3.057479625567794e-05,
|
| 25 |
+
"step": 10
|
| 26 |
+
},
|
| 27 |
+
{
|
| 28 |
+
"epoch": 0.5,
|
| 29 |
+
"grad_norm": 30.28512191772461,
|
| 30 |
+
"learning_rate": 1.25e-06,
|
| 31 |
+
"logits/chosen": -3.051795482635498,
|
| 32 |
+
"logits/rejected": -3.0552544593811035,
|
| 33 |
+
"logps/chosen": -275.4926452636719,
|
| 34 |
+
"logps/rejected": -295.6332702636719,
|
| 35 |
+
"loss": 0.6913,
|
| 36 |
+
"rewards/accuracies": 0.824999988079071,
|
| 37 |
+
"rewards/chosen": 0.004581451416015625,
|
| 38 |
+
"rewards/margins": 0.003644981188699603,
|
| 39 |
+
"rewards/rejected": 0.0009364699944853783,
|
| 40 |
+
"step": 20
|
| 41 |
+
},
|
| 42 |
+
{
|
| 43 |
+
"epoch": 0.75,
|
| 44 |
+
"grad_norm": 22.003541946411133,
|
| 45 |
+
"learning_rate": 1.8750000000000003e-06,
|
| 46 |
+
"logits/chosen": -3.0227792263031006,
|
| 47 |
+
"logits/rejected": -3.037844657897949,
|
| 48 |
+
"logps/chosen": -288.5640563964844,
|
| 49 |
+
"logps/rejected": -305.0924072265625,
|
| 50 |
+
"loss": 0.6858,
|
| 51 |
+
"rewards/accuracies": 0.9750000238418579,
|
| 52 |
+
"rewards/chosen": 0.01881382055580616,
|
| 53 |
+
"rewards/margins": 0.014787979423999786,
|
| 54 |
+
"rewards/rejected": 0.004025841131806374,
|
| 55 |
+
"step": 30
|
| 56 |
+
},
|
| 57 |
+
{
|
| 58 |
+
"epoch": 1.0,
|
| 59 |
+
"grad_norm": 25.61740493774414,
|
| 60 |
+
"learning_rate": 2.5e-06,
|
| 61 |
+
"logits/chosen": -3.075989246368408,
|
| 62 |
+
"logits/rejected": -3.09385347366333,
|
| 63 |
+
"logps/chosen": -326.1593322753906,
|
| 64 |
+
"logps/rejected": -336.47247314453125,
|
| 65 |
+
"loss": 0.6792,
|
| 66 |
+
"rewards/accuracies": 1.0,
|
| 67 |
+
"rewards/chosen": 0.03061000630259514,
|
| 68 |
+
"rewards/margins": 0.02814296819269657,
|
| 69 |
+
"rewards/rejected": 0.0024670413695275784,
|
| 70 |
+
"step": 40
|
| 71 |
+
},
|
| 72 |
+
{
|
| 73 |
+
"epoch": 1.25,
|
| 74 |
+
"grad_norm": 18.726919174194336,
|
| 75 |
+
"learning_rate": 3.125e-06,
|
| 76 |
+
"logits/chosen": -3.045722246170044,
|
| 77 |
+
"logits/rejected": -3.021096706390381,
|
| 78 |
+
"logps/chosen": -300.9041748046875,
|
| 79 |
+
"logps/rejected": -342.43634033203125,
|
| 80 |
+
"loss": 0.6553,
|
| 81 |
+
"rewards/accuracies": 1.0,
|
| 82 |
+
"rewards/chosen": 0.06766779720783234,
|
| 83 |
+
"rewards/margins": 0.0775221437215805,
|
| 84 |
+
"rewards/rejected": -0.009854355826973915,
|
| 85 |
+
"step": 50
|
| 86 |
+
},
|
| 87 |
+
{
|
| 88 |
+
"epoch": 1.5,
|
| 89 |
+
"grad_norm": 23.078794479370117,
|
| 90 |
+
"learning_rate": 3.7500000000000005e-06,
|
| 91 |
+
"logits/chosen": -2.990593433380127,
|
| 92 |
+
"logits/rejected": -3.0149338245391846,
|
| 93 |
+
"logps/chosen": -277.4351501464844,
|
| 94 |
+
"logps/rejected": -289.46417236328125,
|
| 95 |
+
"loss": 0.6346,
|
| 96 |
+
"rewards/accuracies": 1.0,
|
| 97 |
+
"rewards/chosen": 0.10617438703775406,
|
| 98 |
+
"rewards/margins": 0.12149196863174438,
|
| 99 |
+
"rewards/rejected": -0.015317574143409729,
|
| 100 |
+
"step": 60
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 1.75,
|
| 104 |
+
"grad_norm": 27.955883026123047,
|
| 105 |
+
"learning_rate": 4.3750000000000005e-06,
|
| 106 |
+
"logits/chosen": -3.044400691986084,
|
| 107 |
+
"logits/rejected": -3.057063102722168,
|
| 108 |
+
"logps/chosen": -293.9222717285156,
|
| 109 |
+
"logps/rejected": -292.18939208984375,
|
| 110 |
+
"loss": 0.6276,
|
| 111 |
+
"rewards/accuracies": 1.0,
|
| 112 |
+
"rewards/chosen": 0.13079899549484253,
|
| 113 |
+
"rewards/margins": 0.13752365112304688,
|
| 114 |
+
"rewards/rejected": -0.006724664010107517,
|
| 115 |
+
"step": 70
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 2.0,
|
| 119 |
+
"grad_norm": 19.496564865112305,
|
| 120 |
+
"learning_rate": 5e-06,
|
| 121 |
+
"logits/chosen": -3.0755558013916016,
|
| 122 |
+
"logits/rejected": -3.110250473022461,
|
| 123 |
+
"logps/chosen": -304.8177795410156,
|
| 124 |
+
"logps/rejected": -323.0760803222656,
|
| 125 |
+
"loss": 0.596,
|
| 126 |
+
"rewards/accuracies": 1.0,
|
| 127 |
+
"rewards/chosen": 0.19854378700256348,
|
| 128 |
+
"rewards/margins": 0.2069414108991623,
|
| 129 |
+
"rewards/rejected": -0.008397617377340794,
|
| 130 |
+
"step": 80
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 2.25,
|
| 134 |
+
"grad_norm": 18.082189559936523,
|
| 135 |
+
"learning_rate": 5.625e-06,
|
| 136 |
+
"logits/chosen": -2.9778950214385986,
|
| 137 |
+
"logits/rejected": -2.996166944503784,
|
| 138 |
+
"logps/chosen": -300.85528564453125,
|
| 139 |
+
"logps/rejected": -329.5995178222656,
|
| 140 |
+
"loss": 0.5514,
|
| 141 |
+
"rewards/accuracies": 1.0,
|
| 142 |
+
"rewards/chosen": 0.2628921568393707,
|
| 143 |
+
"rewards/margins": 0.31406083703041077,
|
| 144 |
+
"rewards/rejected": -0.05116865038871765,
|
| 145 |
+
"step": 90
|
| 146 |
+
},
|
| 147 |
+
{
|
| 148 |
+
"epoch": 2.5,
|
| 149 |
+
"grad_norm": 21.41973114013672,
|
| 150 |
+
"learning_rate": 6.25e-06,
|
| 151 |
+
"logits/chosen": -3.006256580352783,
|
| 152 |
+
"logits/rejected": -3.015749454498291,
|
| 153 |
+
"logps/chosen": -285.61602783203125,
|
| 154 |
+
"logps/rejected": -294.6287536621094,
|
| 155 |
+
"loss": 0.5194,
|
| 156 |
+
"rewards/accuracies": 1.0,
|
| 157 |
+
"rewards/chosen": 0.33982062339782715,
|
| 158 |
+
"rewards/margins": 0.3938543200492859,
|
| 159 |
+
"rewards/rejected": -0.05403373762965202,
|
| 160 |
+
"step": 100
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"epoch": 2.75,
|
| 164 |
+
"grad_norm": 19.096920013427734,
|
| 165 |
+
"learning_rate": 6.875e-06,
|
| 166 |
+
"logits/chosen": -3.0787596702575684,
|
| 167 |
+
"logits/rejected": -3.0709357261657715,
|
| 168 |
+
"logps/chosen": -284.648193359375,
|
| 169 |
+
"logps/rejected": -322.32989501953125,
|
| 170 |
+
"loss": 0.4882,
|
| 171 |
+
"rewards/accuracies": 1.0,
|
| 172 |
+
"rewards/chosen": 0.4202825129032135,
|
| 173 |
+
"rewards/margins": 0.4840998649597168,
|
| 174 |
+
"rewards/rejected": -0.06381740421056747,
|
| 175 |
+
"step": 110
|
| 176 |
+
},
|
| 177 |
+
{
|
| 178 |
+
"epoch": 3.0,
|
| 179 |
+
"grad_norm": 20.947290420532227,
|
| 180 |
+
"learning_rate": 7.500000000000001e-06,
|
| 181 |
+
"logits/chosen": -3.1311302185058594,
|
| 182 |
+
"logits/rejected": -3.158562183380127,
|
| 183 |
+
"logps/chosen": -298.94390869140625,
|
| 184 |
+
"logps/rejected": -304.8675842285156,
|
| 185 |
+
"loss": 0.4482,
|
| 186 |
+
"rewards/accuracies": 1.0,
|
| 187 |
+
"rewards/chosen": 0.541038990020752,
|
| 188 |
+
"rewards/margins": 0.5876237750053406,
|
| 189 |
+
"rewards/rejected": -0.04658476263284683,
|
| 190 |
+
"step": 120
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 3.25,
|
| 194 |
+
"grad_norm": 15.755688667297363,
|
| 195 |
+
"learning_rate": 8.125000000000001e-06,
|
| 196 |
+
"logits/chosen": -3.037346124649048,
|
| 197 |
+
"logits/rejected": -3.062375783920288,
|
| 198 |
+
"logps/chosen": -275.81427001953125,
|
| 199 |
+
"logps/rejected": -284.2325744628906,
|
| 200 |
+
"loss": 0.3734,
|
| 201 |
+
"rewards/accuracies": 1.0,
|
| 202 |
+
"rewards/chosen": 0.7324575781822205,
|
| 203 |
+
"rewards/margins": 0.8292361497879028,
|
| 204 |
+
"rewards/rejected": -0.09677852690219879,
|
| 205 |
+
"step": 130
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 3.5,
|
| 209 |
+
"grad_norm": 16.660192489624023,
|
| 210 |
+
"learning_rate": 8.750000000000001e-06,
|
| 211 |
+
"logits/chosen": -3.072263240814209,
|
| 212 |
+
"logits/rejected": -3.080423355102539,
|
| 213 |
+
"logps/chosen": -301.5035095214844,
|
| 214 |
+
"logps/rejected": -340.21142578125,
|
| 215 |
+
"loss": 0.3471,
|
| 216 |
+
"rewards/accuracies": 1.0,
|
| 217 |
+
"rewards/chosen": 0.7654526829719543,
|
| 218 |
+
"rewards/margins": 0.9343374967575073,
|
| 219 |
+
"rewards/rejected": -0.16888491809368134,
|
| 220 |
+
"step": 140
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 3.75,
|
| 224 |
+
"grad_norm": 14.524090766906738,
|
| 225 |
+
"learning_rate": 9.375000000000001e-06,
|
| 226 |
+
"logits/chosen": -3.060041904449463,
|
| 227 |
+
"logits/rejected": -3.0739455223083496,
|
| 228 |
+
"logps/chosen": -307.4438171386719,
|
| 229 |
+
"logps/rejected": -317.702392578125,
|
| 230 |
+
"loss": 0.2956,
|
| 231 |
+
"rewards/accuracies": 1.0,
|
| 232 |
+
"rewards/chosen": 1.040612816810608,
|
| 233 |
+
"rewards/margins": 1.1534035205841064,
|
| 234 |
+
"rewards/rejected": -0.11279074847698212,
|
| 235 |
+
"step": 150
|
| 236 |
+
},
|
| 237 |
+
{
|
| 238 |
+
"epoch": 4.0,
|
| 239 |
+
"grad_norm": 16.867460250854492,
|
| 240 |
+
"learning_rate": 1e-05,
|
| 241 |
+
"logits/chosen": -3.006394863128662,
|
| 242 |
+
"logits/rejected": -3.005350112915039,
|
| 243 |
+
"logps/chosen": -261.86370849609375,
|
| 244 |
+
"logps/rejected": -307.3501892089844,
|
| 245 |
+
"loss": 0.2868,
|
| 246 |
+
"rewards/accuracies": 1.0,
|
| 247 |
+
"rewards/chosen": 1.0526248216629028,
|
| 248 |
+
"rewards/margins": 1.1865119934082031,
|
| 249 |
+
"rewards/rejected": -0.13388730585575104,
|
| 250 |
+
"step": 160
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 4.25,
|
| 254 |
+
"grad_norm": 12.979177474975586,
|
| 255 |
+
"learning_rate": 9.998810135399545e-06,
|
| 256 |
+
"logits/chosen": -3.0648646354675293,
|
| 257 |
+
"logits/rejected": -3.0790963172912598,
|
| 258 |
+
"logps/chosen": -300.56158447265625,
|
| 259 |
+
"logps/rejected": -350.8419189453125,
|
| 260 |
+
"loss": 0.1818,
|
| 261 |
+
"rewards/accuracies": 1.0,
|
| 262 |
+
"rewards/chosen": 1.3793233633041382,
|
| 263 |
+
"rewards/margins": 1.7721487283706665,
|
| 264 |
+
"rewards/rejected": -0.39282527565956116,
|
| 265 |
+
"step": 170
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 4.5,
|
| 269 |
+
"grad_norm": 17.333553314208984,
|
| 270 |
+
"learning_rate": 9.99524110790929e-06,
|
| 271 |
+
"logits/chosen": -2.979598045349121,
|
| 272 |
+
"logits/rejected": -3.0077521800994873,
|
| 273 |
+
"logps/chosen": -284.83245849609375,
|
| 274 |
+
"logps/rejected": -317.96502685546875,
|
| 275 |
+
"loss": 0.1726,
|
| 276 |
+
"rewards/accuracies": 1.0,
|
| 277 |
+
"rewards/chosen": 1.4392335414886475,
|
| 278 |
+
"rewards/margins": 1.8440691232681274,
|
| 279 |
+
"rewards/rejected": -0.4048355221748352,
|
| 280 |
+
"step": 180
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"epoch": 4.75,
|
| 284 |
+
"grad_norm": 10.082865715026855,
|
| 285 |
+
"learning_rate": 9.989294616193018e-06,
|
| 286 |
+
"logits/chosen": -3.0077311992645264,
|
| 287 |
+
"logits/rejected": -3.0012829303741455,
|
| 288 |
+
"logps/chosen": -264.1474609375,
|
| 289 |
+
"logps/rejected": -299.96075439453125,
|
| 290 |
+
"loss": 0.146,
|
| 291 |
+
"rewards/accuracies": 1.0,
|
| 292 |
+
"rewards/chosen": 1.6937824487686157,
|
| 293 |
+
"rewards/margins": 2.088609218597412,
|
| 294 |
+
"rewards/rejected": -0.39482688903808594,
|
| 295 |
+
"step": 190
|
| 296 |
+
},
|
| 297 |
+
{
|
| 298 |
+
"epoch": 5.0,
|
| 299 |
+
"grad_norm": 7.095228672027588,
|
| 300 |
+
"learning_rate": 9.980973490458728e-06,
|
| 301 |
+
"logits/chosen": -3.10638165473938,
|
| 302 |
+
"logits/rejected": -3.1016831398010254,
|
| 303 |
+
"logps/chosen": -267.3155212402344,
|
| 304 |
+
"logps/rejected": -291.3060607910156,
|
| 305 |
+
"loss": 0.1178,
|
| 306 |
+
"rewards/accuracies": 1.0,
|
| 307 |
+
"rewards/chosen": 1.9219974279403687,
|
| 308 |
+
"rewards/margins": 2.328953742980957,
|
| 309 |
+
"rewards/rejected": -0.40695637464523315,
|
| 310 |
+
"step": 200
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 5.25,
|
| 314 |
+
"grad_norm": 6.248754978179932,
|
| 315 |
+
"learning_rate": 9.970281691111598e-06,
|
| 316 |
+
"logits/chosen": -3.0690035820007324,
|
| 317 |
+
"logits/rejected": -3.080627918243408,
|
| 318 |
+
"logps/chosen": -260.17803955078125,
|
| 319 |
+
"logps/rejected": -301.79058837890625,
|
| 320 |
+
"loss": 0.0743,
|
| 321 |
+
"rewards/accuracies": 1.0,
|
| 322 |
+
"rewards/chosen": 2.114182233810425,
|
| 323 |
+
"rewards/margins": 2.8904576301574707,
|
| 324 |
+
"rewards/rejected": -0.7762753367424011,
|
| 325 |
+
"step": 210
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 5.5,
|
| 329 |
+
"grad_norm": 3.346534490585327,
|
| 330 |
+
"learning_rate": 9.957224306869053e-06,
|
| 331 |
+
"logits/chosen": -2.9923551082611084,
|
| 332 |
+
"logits/rejected": -2.9941794872283936,
|
| 333 |
+
"logps/chosen": -284.899658203125,
|
| 334 |
+
"logps/rejected": -333.97760009765625,
|
| 335 |
+
"loss": 0.0469,
|
| 336 |
+
"rewards/accuracies": 1.0,
|
| 337 |
+
"rewards/chosen": 2.5475454330444336,
|
| 338 |
+
"rewards/margins": 3.5171380043029785,
|
| 339 |
+
"rewards/rejected": -0.9695925712585449,
|
| 340 |
+
"step": 220
|
| 341 |
+
},
|
| 342 |
+
{
|
| 343 |
+
"epoch": 5.75,
|
| 344 |
+
"grad_norm": 4.401463985443115,
|
| 345 |
+
"learning_rate": 9.941807552338805e-06,
|
| 346 |
+
"logits/chosen": -2.996560573577881,
|
| 347 |
+
"logits/rejected": -3.0060484409332275,
|
| 348 |
+
"logps/chosen": -259.8013000488281,
|
| 349 |
+
"logps/rejected": -321.4071044921875,
|
| 350 |
+
"loss": 0.0313,
|
| 351 |
+
"rewards/accuracies": 1.0,
|
| 352 |
+
"rewards/chosen": 2.493565082550049,
|
| 353 |
+
"rewards/margins": 3.8989131450653076,
|
| 354 |
+
"rewards/rejected": -1.4053480625152588,
|
| 355 |
+
"step": 230
|
| 356 |
+
},
|
| 357 |
+
{
|
| 358 |
+
"epoch": 6.0,
|
| 359 |
+
"grad_norm": 1.5952703952789307,
|
| 360 |
+
"learning_rate": 9.924038765061042e-06,
|
| 361 |
+
"logits/chosen": -2.9383039474487305,
|
| 362 |
+
"logits/rejected": -3.001051664352417,
|
| 363 |
+
"logps/chosen": -289.7506103515625,
|
| 364 |
+
"logps/rejected": -342.30413818359375,
|
| 365 |
+
"loss": 0.0607,
|
| 366 |
+
"rewards/accuracies": 0.9750000238418579,
|
| 367 |
+
"rewards/chosen": 1.786336898803711,
|
| 368 |
+
"rewards/margins": 3.688387632369995,
|
| 369 |
+
"rewards/rejected": -1.9020507335662842,
|
| 370 |
+
"step": 240
|
| 371 |
+
},
|
| 372 |
+
{
|
| 373 |
+
"epoch": 6.25,
|
| 374 |
+
"grad_norm": 1.3278578519821167,
|
| 375 |
+
"learning_rate": 9.903926402016153e-06,
|
| 376 |
+
"logits/chosen": -2.9040274620056152,
|
| 377 |
+
"logits/rejected": -2.955465316772461,
|
| 378 |
+
"logps/chosen": -274.72857666015625,
|
| 379 |
+
"logps/rejected": -317.9063415527344,
|
| 380 |
+
"loss": 0.0151,
|
| 381 |
+
"rewards/accuracies": 1.0,
|
| 382 |
+
"rewards/chosen": 2.4853923320770264,
|
| 383 |
+
"rewards/margins": 4.935563087463379,
|
| 384 |
+
"rewards/rejected": -2.4501702785491943,
|
| 385 |
+
"step": 250
|
| 386 |
+
},
|
| 387 |
+
{
|
| 388 |
+
"epoch": 6.5,
|
| 389 |
+
"grad_norm": 3.8670496940612793,
|
| 390 |
+
"learning_rate": 9.881480035599667e-06,
|
| 391 |
+
"logits/chosen": -2.9413228034973145,
|
| 392 |
+
"logits/rejected": -2.986402988433838,
|
| 393 |
+
"logps/chosen": -241.05313110351562,
|
| 394 |
+
"logps/rejected": -335.13330078125,
|
| 395 |
+
"loss": 0.0131,
|
| 396 |
+
"rewards/accuracies": 1.0,
|
| 397 |
+
"rewards/chosen": 2.401592969894409,
|
| 398 |
+
"rewards/margins": 5.367430686950684,
|
| 399 |
+
"rewards/rejected": -2.965837240219116,
|
| 400 |
+
"step": 260
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 6.75,
|
| 404 |
+
"grad_norm": 0.7370750904083252,
|
| 405 |
+
"learning_rate": 9.856710349066307e-06,
|
| 406 |
+
"logits/chosen": -2.8481552600860596,
|
| 407 |
+
"logits/rejected": -2.871796131134033,
|
| 408 |
+
"logps/chosen": -281.4495849609375,
|
| 409 |
+
"logps/rejected": -375.6628112792969,
|
| 410 |
+
"loss": 0.0191,
|
| 411 |
+
"rewards/accuracies": 1.0,
|
| 412 |
+
"rewards/chosen": 2.0013809204101562,
|
| 413 |
+
"rewards/margins": 5.650525093078613,
|
| 414 |
+
"rewards/rejected": -3.6491446495056152,
|
| 415 |
+
"step": 270
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 7.0,
|
| 419 |
+
"grad_norm": 0.38341209292411804,
|
| 420 |
+
"learning_rate": 9.829629131445342e-06,
|
| 421 |
+
"logits/chosen": -2.975663661956787,
|
| 422 |
+
"logits/rejected": -2.9795100688934326,
|
| 423 |
+
"logps/chosen": -297.9117431640625,
|
| 424 |
+
"logps/rejected": -355.7571716308594,
|
| 425 |
+
"loss": 0.005,
|
| 426 |
+
"rewards/accuracies": 1.0,
|
| 427 |
+
"rewards/chosen": 1.8948161602020264,
|
| 428 |
+
"rewards/margins": 6.713313102722168,
|
| 429 |
+
"rewards/rejected": -4.818497657775879,
|
| 430 |
+
"step": 280
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 7.25,
|
| 434 |
+
"grad_norm": 0.8936977982521057,
|
| 435 |
+
"learning_rate": 9.800249271929645e-06,
|
| 436 |
+
"logits/chosen": -2.7922892570495605,
|
| 437 |
+
"logits/rejected": -2.82979154586792,
|
| 438 |
+
"logps/chosen": -301.52105712890625,
|
| 439 |
+
"logps/rejected": -391.2060852050781,
|
| 440 |
+
"loss": 0.0077,
|
| 441 |
+
"rewards/accuracies": 1.0,
|
| 442 |
+
"rewards/chosen": 1.6500043869018555,
|
| 443 |
+
"rewards/margins": 6.588435173034668,
|
| 444 |
+
"rewards/rejected": -4.938431262969971,
|
| 445 |
+
"step": 290
|
| 446 |
+
},
|
| 447 |
+
{
|
| 448 |
+
"epoch": 7.5,
|
| 449 |
+
"grad_norm": 0.1153794676065445,
|
| 450 |
+
"learning_rate": 9.768584753741134e-06,
|
| 451 |
+
"logits/chosen": -2.7638492584228516,
|
| 452 |
+
"logits/rejected": -2.8244481086730957,
|
| 453 |
+
"logps/chosen": -250.18270874023438,
|
| 454 |
+
"logps/rejected": -330.7033386230469,
|
| 455 |
+
"loss": 0.0013,
|
| 456 |
+
"rewards/accuracies": 1.0,
|
| 457 |
+
"rewards/chosen": 2.1644153594970703,
|
| 458 |
+
"rewards/margins": 7.6036481857299805,
|
| 459 |
+
"rewards/rejected": -5.43923282623291,
|
| 460 |
+
"step": 300
|
| 461 |
+
},
|
| 462 |
+
{
|
| 463 |
+
"epoch": 7.75,
|
| 464 |
+
"grad_norm": 0.43729519844055176,
|
| 465 |
+
"learning_rate": 9.73465064747553e-06,
|
| 466 |
+
"logits/chosen": -2.9230093955993652,
|
| 467 |
+
"logits/rejected": -2.956589937210083,
|
| 468 |
+
"logps/chosen": -259.6827392578125,
|
| 469 |
+
"logps/rejected": -352.92327880859375,
|
| 470 |
+
"loss": 0.0019,
|
| 471 |
+
"rewards/accuracies": 1.0,
|
| 472 |
+
"rewards/chosen": 1.9091575145721436,
|
| 473 |
+
"rewards/margins": 7.56458044052124,
|
| 474 |
+
"rewards/rejected": -5.655422687530518,
|
| 475 |
+
"step": 310
|
| 476 |
+
},
|
| 477 |
+
{
|
| 478 |
+
"epoch": 8.0,
|
| 479 |
+
"grad_norm": 0.09286278486251831,
|
| 480 |
+
"learning_rate": 9.698463103929542e-06,
|
| 481 |
+
"logits/chosen": -2.9670042991638184,
|
| 482 |
+
"logits/rejected": -2.979112386703491,
|
| 483 |
+
"logps/chosen": -295.2704162597656,
|
| 484 |
+
"logps/rejected": -391.45263671875,
|
| 485 |
+
"loss": 0.0011,
|
| 486 |
+
"rewards/accuracies": 1.0,
|
| 487 |
+
"rewards/chosen": 1.921958327293396,
|
| 488 |
+
"rewards/margins": 7.987102508544922,
|
| 489 |
+
"rewards/rejected": -6.065144062042236,
|
| 490 |
+
"step": 320
|
| 491 |
+
},
|
| 492 |
+
{
|
| 493 |
+
"epoch": 8.25,
|
| 494 |
+
"grad_norm": 0.04435203596949577,
|
| 495 |
+
"learning_rate": 9.660039346413994e-06,
|
| 496 |
+
"logits/chosen": -2.816281795501709,
|
| 497 |
+
"logits/rejected": -2.8549602031707764,
|
| 498 |
+
"logps/chosen": -269.79345703125,
|
| 499 |
+
"logps/rejected": -387.1796875,
|
| 500 |
+
"loss": 0.0004,
|
| 501 |
+
"rewards/accuracies": 1.0,
|
| 502 |
+
"rewards/chosen": 1.8488849401474,
|
| 503 |
+
"rewards/margins": 8.709232330322266,
|
| 504 |
+
"rewards/rejected": -6.860346794128418,
|
| 505 |
+
"step": 330
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 8.5,
|
| 509 |
+
"grad_norm": 0.10658630728721619,
|
| 510 |
+
"learning_rate": 9.619397662556434e-06,
|
| 511 |
+
"logits/chosen": -2.8702447414398193,
|
| 512 |
+
"logits/rejected": -2.8935627937316895,
|
| 513 |
+
"logps/chosen": -267.3102111816406,
|
| 514 |
+
"logps/rejected": -386.0473937988281,
|
| 515 |
+
"loss": 0.0006,
|
| 516 |
+
"rewards/accuracies": 1.0,
|
| 517 |
+
"rewards/chosen": 1.496031641960144,
|
| 518 |
+
"rewards/margins": 8.856051445007324,
|
| 519 |
+
"rewards/rejected": -7.360020637512207,
|
| 520 |
+
"step": 340
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 8.75,
|
| 524 |
+
"grad_norm": 0.34437116980552673,
|
| 525 |
+
"learning_rate": 9.576557395597237e-06,
|
| 526 |
+
"logits/chosen": -2.9254894256591797,
|
| 527 |
+
"logits/rejected": -2.9604105949401855,
|
| 528 |
+
"logps/chosen": -292.0805969238281,
|
| 529 |
+
"logps/rejected": -366.92828369140625,
|
| 530 |
+
"loss": 0.0006,
|
| 531 |
+
"rewards/accuracies": 1.0,
|
| 532 |
+
"rewards/chosen": 1.7589994668960571,
|
| 533 |
+
"rewards/margins": 8.077742576599121,
|
| 534 |
+
"rewards/rejected": -6.3187432289123535,
|
| 535 |
+
"step": 350
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 9.0,
|
| 539 |
+
"grad_norm": 0.03675313666462898,
|
| 540 |
+
"learning_rate": 9.531538935183252e-06,
|
| 541 |
+
"logits/chosen": -2.7603306770324707,
|
| 542 |
+
"logits/rejected": -2.778339147567749,
|
| 543 |
+
"logps/chosen": -287.90899658203125,
|
| 544 |
+
"logps/rejected": -375.89959716796875,
|
| 545 |
+
"loss": 0.0211,
|
| 546 |
+
"rewards/accuracies": 0.9750000238418579,
|
| 547 |
+
"rewards/chosen": 1.1417970657348633,
|
| 548 |
+
"rewards/margins": 8.125410079956055,
|
| 549 |
+
"rewards/rejected": -6.983613014221191,
|
| 550 |
+
"step": 360
|
| 551 |
+
},
|
| 552 |
+
{
|
| 553 |
+
"epoch": 9.25,
|
| 554 |
+
"grad_norm": 0.03928977623581886,
|
| 555 |
+
"learning_rate": 9.484363707663443e-06,
|
| 556 |
+
"logits/chosen": -2.8726091384887695,
|
| 557 |
+
"logits/rejected": -2.914290189743042,
|
| 558 |
+
"logps/chosen": -277.1239929199219,
|
| 559 |
+
"logps/rejected": -403.6535949707031,
|
| 560 |
+
"loss": 0.0004,
|
| 561 |
+
"rewards/accuracies": 1.0,
|
| 562 |
+
"rewards/chosen": 1.2768957614898682,
|
| 563 |
+
"rewards/margins": 9.230466842651367,
|
| 564 |
+
"rewards/rejected": -7.953570365905762,
|
| 565 |
+
"step": 370
|
| 566 |
+
},
|
| 567 |
+
{
|
| 568 |
+
"epoch": 9.5,
|
| 569 |
+
"grad_norm": 0.07424739003181458,
|
| 570 |
+
"learning_rate": 9.43505416589111e-06,
|
| 571 |
+
"logits/chosen": -2.7936153411865234,
|
| 572 |
+
"logits/rejected": -2.8210067749023438,
|
| 573 |
+
"logps/chosen": -286.84423828125,
|
| 574 |
+
"logps/rejected": -358.430908203125,
|
| 575 |
+
"loss": 0.0002,
|
| 576 |
+
"rewards/accuracies": 1.0,
|
| 577 |
+
"rewards/chosen": 1.287247896194458,
|
| 578 |
+
"rewards/margins": 9.114188194274902,
|
| 579 |
+
"rewards/rejected": -7.826941013336182,
|
| 580 |
+
"step": 380
|
| 581 |
+
},
|
| 582 |
+
{
|
| 583 |
+
"epoch": 9.75,
|
| 584 |
+
"grad_norm": 0.16556577384471893,
|
| 585 |
+
"learning_rate": 9.38363377853754e-06,
|
| 586 |
+
"logits/chosen": -2.7510828971862793,
|
| 587 |
+
"logits/rejected": -2.810314655303955,
|
| 588 |
+
"logps/chosen": -279.12481689453125,
|
| 589 |
+
"logps/rejected": -388.2037658691406,
|
| 590 |
+
"loss": 0.0012,
|
| 591 |
+
"rewards/accuracies": 1.0,
|
| 592 |
+
"rewards/chosen": 1.3857409954071045,
|
| 593 |
+
"rewards/margins": 8.737149238586426,
|
| 594 |
+
"rewards/rejected": -7.3514084815979,
|
| 595 |
+
"step": 390
|
| 596 |
+
},
|
| 597 |
+
{
|
| 598 |
+
"epoch": 10.0,
|
| 599 |
+
"grad_norm": 0.04436686635017395,
|
| 600 |
+
"learning_rate": 9.330127018922195e-06,
|
| 601 |
+
"logits/chosen": -2.8653323650360107,
|
| 602 |
+
"logits/rejected": -2.8766775131225586,
|
| 603 |
+
"logps/chosen": -288.20709228515625,
|
| 604 |
+
"logps/rejected": -406.8284606933594,
|
| 605 |
+
"loss": 0.0003,
|
| 606 |
+
"rewards/accuracies": 1.0,
|
| 607 |
+
"rewards/chosen": 1.3324298858642578,
|
| 608 |
+
"rewards/margins": 9.166605949401855,
|
| 609 |
+
"rewards/rejected": -7.834176540374756,
|
| 610 |
+
"step": 400
|
| 611 |
+
},
|
| 612 |
+
{
|
| 613 |
+
"epoch": 10.25,
|
| 614 |
+
"grad_norm": 0.05012385919690132,
|
| 615 |
+
"learning_rate": 9.274559353364734e-06,
|
| 616 |
+
"logits/chosen": -2.864193916320801,
|
| 617 |
+
"logits/rejected": -2.883277177810669,
|
| 618 |
+
"logps/chosen": -267.9872131347656,
|
| 619 |
+
"logps/rejected": -386.8029479980469,
|
| 620 |
+
"loss": 0.0003,
|
| 621 |
+
"rewards/accuracies": 1.0,
|
| 622 |
+
"rewards/chosen": 0.9138404726982117,
|
| 623 |
+
"rewards/margins": 9.733392715454102,
|
| 624 |
+
"rewards/rejected": -8.819551467895508,
|
| 625 |
+
"step": 410
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 10.5,
|
| 629 |
+
"grad_norm": 0.43551239371299744,
|
| 630 |
+
"learning_rate": 9.21695722906443e-06,
|
| 631 |
+
"logits/chosen": -2.7738356590270996,
|
| 632 |
+
"logits/rejected": -2.8102309703826904,
|
| 633 |
+
"logps/chosen": -263.6973571777344,
|
| 634 |
+
"logps/rejected": -381.28717041015625,
|
| 635 |
+
"loss": 0.0003,
|
| 636 |
+
"rewards/accuracies": 1.0,
|
| 637 |
+
"rewards/chosen": 1.2245395183563232,
|
| 638 |
+
"rewards/margins": 9.739076614379883,
|
| 639 |
+
"rewards/rejected": -8.514535903930664,
|
| 640 |
+
"step": 420
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 10.75,
|
| 644 |
+
"grad_norm": 0.13912050426006317,
|
| 645 |
+
"learning_rate": 9.157348061512728e-06,
|
| 646 |
+
"logits/chosen": -2.7845568656921387,
|
| 647 |
+
"logits/rejected": -2.8310136795043945,
|
| 648 |
+
"logps/chosen": -286.35858154296875,
|
| 649 |
+
"logps/rejected": -424.60162353515625,
|
| 650 |
+
"loss": 0.0002,
|
| 651 |
+
"rewards/accuracies": 1.0,
|
| 652 |
+
"rewards/chosen": 0.6671217679977417,
|
| 653 |
+
"rewards/margins": 9.726836204528809,
|
| 654 |
+
"rewards/rejected": -9.059713363647461,
|
| 655 |
+
"step": 430
|
| 656 |
+
},
|
| 657 |
+
{
|
| 658 |
+
"epoch": 11.0,
|
| 659 |
+
"grad_norm": 0.06393156945705414,
|
| 660 |
+
"learning_rate": 9.09576022144496e-06,
|
| 661 |
+
"logits/chosen": -2.7752254009246826,
|
| 662 |
+
"logits/rejected": -2.788267135620117,
|
| 663 |
+
"logps/chosen": -336.1651306152344,
|
| 664 |
+
"logps/rejected": -411.5840759277344,
|
| 665 |
+
"loss": 0.0082,
|
| 666 |
+
"rewards/accuracies": 1.0,
|
| 667 |
+
"rewards/chosen": 0.1248699277639389,
|
| 668 |
+
"rewards/margins": 9.496256828308105,
|
| 669 |
+
"rewards/rejected": -9.37138843536377,
|
| 670 |
+
"step": 440
|
| 671 |
+
},
|
| 672 |
+
{
|
| 673 |
+
"epoch": 11.25,
|
| 674 |
+
"grad_norm": 0.03681237995624542,
|
| 675 |
+
"learning_rate": 9.032223021337415e-06,
|
| 676 |
+
"logits/chosen": -2.8105359077453613,
|
| 677 |
+
"logits/rejected": -2.839118242263794,
|
| 678 |
+
"logps/chosen": -295.88751220703125,
|
| 679 |
+
"logps/rejected": -395.9588928222656,
|
| 680 |
+
"loss": 0.0002,
|
| 681 |
+
"rewards/accuracies": 1.0,
|
| 682 |
+
"rewards/chosen": 1.1100804805755615,
|
| 683 |
+
"rewards/margins": 9.81136417388916,
|
| 684 |
+
"rewards/rejected": -8.701284408569336,
|
| 685 |
+
"step": 450
|
| 686 |
+
},
|
| 687 |
+
{
|
| 688 |
+
"epoch": 11.5,
|
| 689 |
+
"grad_norm": 0.03024367056787014,
|
| 690 |
+
"learning_rate": 8.966766701456177e-06,
|
| 691 |
+
"logits/chosen": -2.8440585136413574,
|
| 692 |
+
"logits/rejected": -2.879945993423462,
|
| 693 |
+
"logps/chosen": -313.1834411621094,
|
| 694 |
+
"logps/rejected": -407.3936462402344,
|
| 695 |
+
"loss": 0.0002,
|
| 696 |
+
"rewards/accuracies": 1.0,
|
| 697 |
+
"rewards/chosen": 0.3773784041404724,
|
| 698 |
+
"rewards/margins": 9.822793960571289,
|
| 699 |
+
"rewards/rejected": -9.445414543151855,
|
| 700 |
+
"step": 460
|
| 701 |
+
},
|
| 702 |
+
{
|
| 703 |
+
"epoch": 11.75,
|
| 704 |
+
"grad_norm": 0.017395537346601486,
|
| 705 |
+
"learning_rate": 8.899422415464409e-06,
|
| 706 |
+
"logits/chosen": -2.8262779712677,
|
| 707 |
+
"logits/rejected": -2.868819236755371,
|
| 708 |
+
"logps/chosen": -275.37896728515625,
|
| 709 |
+
"logps/rejected": -413.3304748535156,
|
| 710 |
+
"loss": 0.0004,
|
| 711 |
+
"rewards/accuracies": 1.0,
|
| 712 |
+
"rewards/chosen": 1.2674087285995483,
|
| 713 |
+
"rewards/margins": 9.833629608154297,
|
| 714 |
+
"rewards/rejected": -8.566221237182617,
|
| 715 |
+
"step": 470
|
| 716 |
+
},
|
| 717 |
+
{
|
| 718 |
+
"epoch": 12.0,
|
| 719 |
+
"grad_norm": 0.015860358253121376,
|
| 720 |
+
"learning_rate": 8.83022221559489e-06,
|
| 721 |
+
"logits/chosen": -2.698403835296631,
|
| 722 |
+
"logits/rejected": -2.7310996055603027,
|
| 723 |
+
"logps/chosen": -270.9880676269531,
|
| 724 |
+
"logps/rejected": -397.07965087890625,
|
| 725 |
+
"loss": 0.0135,
|
| 726 |
+
"rewards/accuracies": 1.0,
|
| 727 |
+
"rewards/chosen": 0.23779204487800598,
|
| 728 |
+
"rewards/margins": 10.131186485290527,
|
| 729 |
+
"rewards/rejected": -9.893393516540527,
|
| 730 |
+
"step": 480
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 12.25,
|
| 734 |
+
"grad_norm": 0.011888382956385612,
|
| 735 |
+
"learning_rate": 8.759199037394888e-06,
|
| 736 |
+
"logits/chosen": -2.770529270172119,
|
| 737 |
+
"logits/rejected": -2.8385539054870605,
|
| 738 |
+
"logps/chosen": -293.5708312988281,
|
| 739 |
+
"logps/rejected": -416.44195556640625,
|
| 740 |
+
"loss": 0.0001,
|
| 741 |
+
"rewards/accuracies": 1.0,
|
| 742 |
+
"rewards/chosen": 1.0723686218261719,
|
| 743 |
+
"rewards/margins": 10.268576622009277,
|
| 744 |
+
"rewards/rejected": -9.196208000183105,
|
| 745 |
+
"step": 490
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 12.5,
|
| 749 |
+
"grad_norm": 0.03234002739191055,
|
| 750 |
+
"learning_rate": 8.68638668405062e-06,
|
| 751 |
+
"logits/chosen": -2.7817087173461914,
|
| 752 |
+
"logits/rejected": -2.8032031059265137,
|
| 753 |
+
"logps/chosen": -312.52459716796875,
|
| 754 |
+
"logps/rejected": -415.45654296875,
|
| 755 |
+
"loss": 0.0128,
|
| 756 |
+
"rewards/accuracies": 1.0,
|
| 757 |
+
"rewards/chosen": 0.5771347880363464,
|
| 758 |
+
"rewards/margins": 9.832048416137695,
|
| 759 |
+
"rewards/rejected": -9.254913330078125,
|
| 760 |
+
"step": 500
|
| 761 |
+
},
|
| 762 |
+
{
|
| 763 |
+
"epoch": 12.75,
|
| 764 |
+
"grad_norm": 0.02184719778597355,
|
| 765 |
+
"learning_rate": 8.611819810298778e-06,
|
| 766 |
+
"logits/chosen": -2.790955066680908,
|
| 767 |
+
"logits/rejected": -2.8012702465057373,
|
| 768 |
+
"logps/chosen": -270.79864501953125,
|
| 769 |
+
"logps/rejected": -401.427490234375,
|
| 770 |
+
"loss": 0.0001,
|
| 771 |
+
"rewards/accuracies": 1.0,
|
| 772 |
+
"rewards/chosen": 1.1068525314331055,
|
| 773 |
+
"rewards/margins": 10.041081428527832,
|
| 774 |
+
"rewards/rejected": -8.934228897094727,
|
| 775 |
+
"step": 510
|
| 776 |
+
},
|
| 777 |
+
{
|
| 778 |
+
"epoch": 13.0,
|
| 779 |
+
"grad_norm": 0.051279619336128235,
|
| 780 |
+
"learning_rate": 8.535533905932739e-06,
|
| 781 |
+
"logits/chosen": -2.8458592891693115,
|
| 782 |
+
"logits/rejected": -2.880188465118408,
|
| 783 |
+
"logps/chosen": -275.39886474609375,
|
| 784 |
+
"logps/rejected": -387.335693359375,
|
| 785 |
+
"loss": 0.0001,
|
| 786 |
+
"rewards/accuracies": 1.0,
|
| 787 |
+
"rewards/chosen": 0.7447389364242554,
|
| 788 |
+
"rewards/margins": 10.408342361450195,
|
| 789 |
+
"rewards/rejected": -9.663604736328125,
|
| 790 |
+
"step": 520
|
| 791 |
+
},
|
| 792 |
+
{
|
| 793 |
+
"epoch": 13.25,
|
| 794 |
+
"grad_norm": 0.044555239379405975,
|
| 795 |
+
"learning_rate": 8.457565278911349e-06,
|
| 796 |
+
"logits/chosen": -2.7830259799957275,
|
| 797 |
+
"logits/rejected": -2.8375871181488037,
|
| 798 |
+
"logps/chosen": -270.60235595703125,
|
| 799 |
+
"logps/rejected": -393.5851135253906,
|
| 800 |
+
"loss": 0.0001,
|
| 801 |
+
"rewards/accuracies": 1.0,
|
| 802 |
+
"rewards/chosen": 1.0359405279159546,
|
| 803 |
+
"rewards/margins": 10.306293487548828,
|
| 804 |
+
"rewards/rejected": -9.270353317260742,
|
| 805 |
+
"step": 530
|
| 806 |
+
},
|
| 807 |
+
{
|
| 808 |
+
"epoch": 13.5,
|
| 809 |
+
"grad_norm": 0.17373812198638916,
|
| 810 |
+
"learning_rate": 8.377951038078303e-06,
|
| 811 |
+
"logits/chosen": -2.7552318572998047,
|
| 812 |
+
"logits/rejected": -2.8055858612060547,
|
| 813 |
+
"logps/chosen": -307.6959533691406,
|
| 814 |
+
"logps/rejected": -438.8316955566406,
|
| 815 |
+
"loss": 0.0001,
|
| 816 |
+
"rewards/accuracies": 1.0,
|
| 817 |
+
"rewards/chosen": 0.6580104827880859,
|
| 818 |
+
"rewards/margins": 10.06629467010498,
|
| 819 |
+
"rewards/rejected": -9.408284187316895,
|
| 820 |
+
"step": 540
|
| 821 |
+
},
|
| 822 |
+
{
|
| 823 |
+
"epoch": 13.75,
|
| 824 |
+
"grad_norm": 0.007444189395755529,
|
| 825 |
+
"learning_rate": 8.296729075500345e-06,
|
| 826 |
+
"logits/chosen": -2.933079481124878,
|
| 827 |
+
"logits/rejected": -2.9362237453460693,
|
| 828 |
+
"logps/chosen": -296.7786560058594,
|
| 829 |
+
"logps/rejected": -411.70782470703125,
|
| 830 |
+
"loss": 0.0001,
|
| 831 |
+
"rewards/accuracies": 1.0,
|
| 832 |
+
"rewards/chosen": 1.005584955215454,
|
| 833 |
+
"rewards/margins": 10.901617050170898,
|
| 834 |
+
"rewards/rejected": -9.89603328704834,
|
| 835 |
+
"step": 550
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 14.0,
|
| 839 |
+
"grad_norm": 0.01406220905482769,
|
| 840 |
+
"learning_rate": 8.213938048432697e-06,
|
| 841 |
+
"logits/chosen": -2.6971287727355957,
|
| 842 |
+
"logits/rejected": -2.7503409385681152,
|
| 843 |
+
"logps/chosen": -270.8581237792969,
|
| 844 |
+
"logps/rejected": -379.1615295410156,
|
| 845 |
+
"loss": 0.0001,
|
| 846 |
+
"rewards/accuracies": 1.0,
|
| 847 |
+
"rewards/chosen": 1.0409095287322998,
|
| 848 |
+
"rewards/margins": 10.008934020996094,
|
| 849 |
+
"rewards/rejected": -8.968024253845215,
|
| 850 |
+
"step": 560
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 14.25,
|
| 854 |
+
"grad_norm": 0.012559090740978718,
|
| 855 |
+
"learning_rate": 8.129617360920297e-06,
|
| 856 |
+
"logits/chosen": -2.891970157623291,
|
| 857 |
+
"logits/rejected": -2.8930327892303467,
|
| 858 |
+
"logps/chosen": -257.08953857421875,
|
| 859 |
+
"logps/rejected": -426.03045654296875,
|
| 860 |
+
"loss": 0.0001,
|
| 861 |
+
"rewards/accuracies": 1.0,
|
| 862 |
+
"rewards/chosen": 1.4467694759368896,
|
| 863 |
+
"rewards/margins": 10.829238891601562,
|
| 864 |
+
"rewards/rejected": -9.382468223571777,
|
| 865 |
+
"step": 570
|
| 866 |
+
},
|
| 867 |
+
{
|
| 868 |
+
"epoch": 14.5,
|
| 869 |
+
"grad_norm": 0.00887581892311573,
|
| 870 |
+
"learning_rate": 8.043807145043604e-06,
|
| 871 |
+
"logits/chosen": -2.800017833709717,
|
| 872 |
+
"logits/rejected": -2.8574788570404053,
|
| 873 |
+
"logps/chosen": -300.4933166503906,
|
| 874 |
+
"logps/rejected": -405.90228271484375,
|
| 875 |
+
"loss": 0.0001,
|
| 876 |
+
"rewards/accuracies": 1.0,
|
| 877 |
+
"rewards/chosen": 0.45210862159729004,
|
| 878 |
+
"rewards/margins": 10.489278793334961,
|
| 879 |
+
"rewards/rejected": -10.037171363830566,
|
| 880 |
+
"step": 580
|
| 881 |
+
},
|
| 882 |
+
{
|
| 883 |
+
"epoch": 14.75,
|
| 884 |
+
"grad_norm": 0.03181132301688194,
|
| 885 |
+
"learning_rate": 7.956548241817914e-06,
|
| 886 |
+
"logits/chosen": -2.773524761199951,
|
| 887 |
+
"logits/rejected": -2.8105709552764893,
|
| 888 |
+
"logps/chosen": -297.21917724609375,
|
| 889 |
+
"logps/rejected": -382.3047790527344,
|
| 890 |
+
"loss": 0.0026,
|
| 891 |
+
"rewards/accuracies": 1.0,
|
| 892 |
+
"rewards/chosen": 0.5754307508468628,
|
| 893 |
+
"rewards/margins": 10.027421951293945,
|
| 894 |
+
"rewards/rejected": -9.451990127563477,
|
| 895 |
+
"step": 590
|
| 896 |
+
},
|
| 897 |
+
{
|
| 898 |
+
"epoch": 15.0,
|
| 899 |
+
"grad_norm": 0.01683628000319004,
|
| 900 |
+
"learning_rate": 7.86788218175523e-06,
|
| 901 |
+
"logits/chosen": -2.6840453147888184,
|
| 902 |
+
"logits/rejected": -2.7441771030426025,
|
| 903 |
+
"logps/chosen": -299.78656005859375,
|
| 904 |
+
"logps/rejected": -423.99713134765625,
|
| 905 |
+
"loss": 0.0001,
|
| 906 |
+
"rewards/accuracies": 1.0,
|
| 907 |
+
"rewards/chosen": 0.4864146113395691,
|
| 908 |
+
"rewards/margins": 10.617002487182617,
|
| 909 |
+
"rewards/rejected": -10.130589485168457,
|
| 910 |
+
"step": 600
|
| 911 |
+
},
|
| 912 |
+
{
|
| 913 |
+
"epoch": 15.25,
|
| 914 |
+
"grad_norm": 0.008264761418104172,
|
| 915 |
+
"learning_rate": 7.777851165098012e-06,
|
| 916 |
+
"logits/chosen": -2.758063793182373,
|
| 917 |
+
"logits/rejected": -2.8015329837799072,
|
| 918 |
+
"logps/chosen": -288.71295166015625,
|
| 919 |
+
"logps/rejected": -417.6280212402344,
|
| 920 |
+
"loss": 0.0001,
|
| 921 |
+
"rewards/accuracies": 1.0,
|
| 922 |
+
"rewards/chosen": 0.4645349085330963,
|
| 923 |
+
"rewards/margins": 10.90323257446289,
|
| 924 |
+
"rewards/rejected": -10.43869686126709,
|
| 925 |
+
"step": 610
|
| 926 |
+
},
|
| 927 |
+
{
|
| 928 |
+
"epoch": 15.5,
|
| 929 |
+
"grad_norm": 0.030291497707366943,
|
| 930 |
+
"learning_rate": 7.686498041734121e-06,
|
| 931 |
+
"logits/chosen": -2.782529830932617,
|
| 932 |
+
"logits/rejected": -2.8300535678863525,
|
| 933 |
+
"logps/chosen": -270.43450927734375,
|
| 934 |
+
"logps/rejected": -401.63824462890625,
|
| 935 |
+
"loss": 0.0001,
|
| 936 |
+
"rewards/accuracies": 1.0,
|
| 937 |
+
"rewards/chosen": 1.1531331539154053,
|
| 938 |
+
"rewards/margins": 10.583002090454102,
|
| 939 |
+
"rewards/rejected": -9.429868698120117,
|
| 940 |
+
"step": 620
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 15.75,
|
| 944 |
+
"grad_norm": 0.009538864716887474,
|
| 945 |
+
"learning_rate": 7.593866290802608e-06,
|
| 946 |
+
"logits/chosen": -2.7459161281585693,
|
| 947 |
+
"logits/rejected": -2.788944721221924,
|
| 948 |
+
"logps/chosen": -280.4390869140625,
|
| 949 |
+
"logps/rejected": -394.48883056640625,
|
| 950 |
+
"loss": 0.0001,
|
| 951 |
+
"rewards/accuracies": 1.0,
|
| 952 |
+
"rewards/chosen": 0.5682659149169922,
|
| 953 |
+
"rewards/margins": 10.8987455368042,
|
| 954 |
+
"rewards/rejected": -10.330477714538574,
|
| 955 |
+
"step": 630
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 16.0,
|
| 959 |
+
"grad_norm": 0.02041240781545639,
|
| 960 |
+
"learning_rate": 7.500000000000001e-06,
|
| 961 |
+
"logits/chosen": -2.795477867126465,
|
| 962 |
+
"logits/rejected": -2.820490598678589,
|
| 963 |
+
"logps/chosen": -318.35418701171875,
|
| 964 |
+
"logps/rejected": -434.80987548828125,
|
| 965 |
+
"loss": 0.0049,
|
| 966 |
+
"rewards/accuracies": 1.0,
|
| 967 |
+
"rewards/chosen": 0.37515130639076233,
|
| 968 |
+
"rewards/margins": 10.356420516967773,
|
| 969 |
+
"rewards/rejected": -9.981268882751465,
|
| 970 |
+
"step": 640
|
| 971 |
+
},
|
| 972 |
+
{
|
| 973 |
+
"epoch": 16.25,
|
| 974 |
+
"grad_norm": 0.012305938638746738,
|
| 975 |
+
"learning_rate": 7.404943844596939e-06,
|
| 976 |
+
"logits/chosen": -2.665245771408081,
|
| 977 |
+
"logits/rejected": -2.72597074508667,
|
| 978 |
+
"logps/chosen": -330.48480224609375,
|
| 979 |
+
"logps/rejected": -456.65411376953125,
|
| 980 |
+
"loss": 0.0067,
|
| 981 |
+
"rewards/accuracies": 1.0,
|
| 982 |
+
"rewards/chosen": 0.4235839247703552,
|
| 983 |
+
"rewards/margins": 9.822153091430664,
|
| 984 |
+
"rewards/rejected": -9.398569107055664,
|
| 985 |
+
"step": 650
|
| 986 |
+
},
|
| 987 |
+
{
|
| 988 |
+
"epoch": 16.5,
|
| 989 |
+
"grad_norm": 0.01604386977851391,
|
| 990 |
+
"learning_rate": 7.308743066175172e-06,
|
| 991 |
+
"logits/chosen": -2.8210482597351074,
|
| 992 |
+
"logits/rejected": -2.8381505012512207,
|
| 993 |
+
"logps/chosen": -291.52581787109375,
|
| 994 |
+
"logps/rejected": -412.03546142578125,
|
| 995 |
+
"loss": 0.0,
|
| 996 |
+
"rewards/accuracies": 1.0,
|
| 997 |
+
"rewards/chosen": 0.3690633773803711,
|
| 998 |
+
"rewards/margins": 11.374852180480957,
|
| 999 |
+
"rewards/rejected": -11.005789756774902,
|
| 1000 |
+
"step": 660
|
| 1001 |
+
},
|
| 1002 |
+
{
|
| 1003 |
+
"epoch": 16.75,
|
| 1004 |
+
"grad_norm": 0.019596703350543976,
|
| 1005 |
+
"learning_rate": 7.211443451095007e-06,
|
| 1006 |
+
"logits/chosen": -2.756493091583252,
|
| 1007 |
+
"logits/rejected": -2.818694829940796,
|
| 1008 |
+
"logps/chosen": -252.20590209960938,
|
| 1009 |
+
"logps/rejected": -381.78485107421875,
|
| 1010 |
+
"loss": 0.0001,
|
| 1011 |
+
"rewards/accuracies": 1.0,
|
| 1012 |
+
"rewards/chosen": 0.5581167936325073,
|
| 1013 |
+
"rewards/margins": 10.99728775024414,
|
| 1014 |
+
"rewards/rejected": -10.439170837402344,
|
| 1015 |
+
"step": 670
|
| 1016 |
+
},
|
| 1017 |
+
{
|
| 1018 |
+
"epoch": 17.0,
|
| 1019 |
+
"grad_norm": 0.0072501422837376595,
|
| 1020 |
+
"learning_rate": 7.113091308703498e-06,
|
| 1021 |
+
"logits/chosen": -2.8325161933898926,
|
| 1022 |
+
"logits/rejected": -2.866009473800659,
|
| 1023 |
+
"logps/chosen": -289.85833740234375,
|
| 1024 |
+
"logps/rejected": -412.3795471191406,
|
| 1025 |
+
"loss": 0.0001,
|
| 1026 |
+
"rewards/accuracies": 1.0,
|
| 1027 |
+
"rewards/chosen": 0.4296005368232727,
|
| 1028 |
+
"rewards/margins": 11.440092086791992,
|
| 1029 |
+
"rewards/rejected": -11.010492324829102,
|
| 1030 |
+
"step": 680
|
| 1031 |
+
},
|
| 1032 |
+
{
|
| 1033 |
+
"epoch": 17.25,
|
| 1034 |
+
"grad_norm": 0.020742328837513924,
|
| 1035 |
+
"learning_rate": 7.0137334492936875e-06,
|
| 1036 |
+
"logits/chosen": -2.702772378921509,
|
| 1037 |
+
"logits/rejected": -2.769951105117798,
|
| 1038 |
+
"logps/chosen": -286.2898254394531,
|
| 1039 |
+
"logps/rejected": -432.13653564453125,
|
| 1040 |
+
"loss": 0.0012,
|
| 1041 |
+
"rewards/accuracies": 1.0,
|
| 1042 |
+
"rewards/chosen": 0.33315685391426086,
|
| 1043 |
+
"rewards/margins": 10.437726974487305,
|
| 1044 |
+
"rewards/rejected": -10.104570388793945,
|
| 1045 |
+
"step": 690
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 17.5,
|
| 1049 |
+
"grad_norm": 0.025510217994451523,
|
| 1050 |
+
"learning_rate": 6.913417161825449e-06,
|
| 1051 |
+
"logits/chosen": -2.7644479274749756,
|
| 1052 |
+
"logits/rejected": -2.7879385948181152,
|
| 1053 |
+
"logps/chosen": -268.5244140625,
|
| 1054 |
+
"logps/rejected": -382.07073974609375,
|
| 1055 |
+
"loss": 0.0001,
|
| 1056 |
+
"rewards/accuracies": 1.0,
|
| 1057 |
+
"rewards/chosen": 0.7151921987533569,
|
| 1058 |
+
"rewards/margins": 11.053590774536133,
|
| 1059 |
+
"rewards/rejected": -10.338397979736328,
|
| 1060 |
+
"step": 700
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 17.75,
|
| 1064 |
+
"grad_norm": 0.002281380584463477,
|
| 1065 |
+
"learning_rate": 6.812190191418508e-06,
|
| 1066 |
+
"logits/chosen": -2.785850763320923,
|
| 1067 |
+
"logits/rejected": -2.8239619731903076,
|
| 1068 |
+
"logps/chosen": -306.57843017578125,
|
| 1069 |
+
"logps/rejected": -423.93157958984375,
|
| 1070 |
+
"loss": 0.0,
|
| 1071 |
+
"rewards/accuracies": 1.0,
|
| 1072 |
+
"rewards/chosen": 0.2157173603773117,
|
| 1073 |
+
"rewards/margins": 10.968465805053711,
|
| 1074 |
+
"rewards/rejected": -10.752748489379883,
|
| 1075 |
+
"step": 710
|
| 1076 |
+
},
|
| 1077 |
+
{
|
| 1078 |
+
"epoch": 18.0,
|
| 1079 |
+
"grad_norm": 0.003572087734937668,
|
| 1080 |
+
"learning_rate": 6.710100716628345e-06,
|
| 1081 |
+
"logits/chosen": -2.803100109100342,
|
| 1082 |
+
"logits/rejected": -2.836097240447998,
|
| 1083 |
+
"logps/chosen": -304.5684509277344,
|
| 1084 |
+
"logps/rejected": -437.28302001953125,
|
| 1085 |
+
"loss": 0.0,
|
| 1086 |
+
"rewards/accuracies": 1.0,
|
| 1087 |
+
"rewards/chosen": 0.6064807772636414,
|
| 1088 |
+
"rewards/margins": 12.001432418823242,
|
| 1089 |
+
"rewards/rejected": -11.394950866699219,
|
| 1090 |
+
"step": 720
|
| 1091 |
+
},
|
| 1092 |
+
{
|
| 1093 |
+
"epoch": 18.25,
|
| 1094 |
+
"grad_norm": 0.008005160838365555,
|
| 1095 |
+
"learning_rate": 6.607197326515808e-06,
|
| 1096 |
+
"logits/chosen": -2.7193379402160645,
|
| 1097 |
+
"logits/rejected": -2.7719767093658447,
|
| 1098 |
+
"logps/chosen": -296.9802551269531,
|
| 1099 |
+
"logps/rejected": -437.65142822265625,
|
| 1100 |
+
"loss": 0.0008,
|
| 1101 |
+
"rewards/accuracies": 1.0,
|
| 1102 |
+
"rewards/chosen": 0.9086353182792664,
|
| 1103 |
+
"rewards/margins": 10.867486000061035,
|
| 1104 |
+
"rewards/rejected": -9.958850860595703,
|
| 1105 |
+
"step": 730
|
| 1106 |
+
},
|
| 1107 |
+
{
|
| 1108 |
+
"epoch": 18.5,
|
| 1109 |
+
"grad_norm": 0.009184204041957855,
|
| 1110 |
+
"learning_rate": 6.503528997521365e-06,
|
| 1111 |
+
"logits/chosen": -2.7671329975128174,
|
| 1112 |
+
"logits/rejected": -2.8140933513641357,
|
| 1113 |
+
"logps/chosen": -288.43389892578125,
|
| 1114 |
+
"logps/rejected": -418.6834411621094,
|
| 1115 |
+
"loss": 0.0,
|
| 1116 |
+
"rewards/accuracies": 1.0,
|
| 1117 |
+
"rewards/chosen": 0.7397573590278625,
|
| 1118 |
+
"rewards/margins": 11.15060806274414,
|
| 1119 |
+
"rewards/rejected": -10.410850524902344,
|
| 1120 |
+
"step": 740
|
| 1121 |
+
},
|
| 1122 |
+
{
|
| 1123 |
+
"epoch": 18.75,
|
| 1124 |
+
"grad_norm": 0.008301551453769207,
|
| 1125 |
+
"learning_rate": 6.399145070154962e-06,
|
| 1126 |
+
"logits/chosen": -2.791762113571167,
|
| 1127 |
+
"logits/rejected": -2.8275296688079834,
|
| 1128 |
+
"logps/chosen": -266.60394287109375,
|
| 1129 |
+
"logps/rejected": -389.6926574707031,
|
| 1130 |
+
"loss": 0.0001,
|
| 1131 |
+
"rewards/accuracies": 1.0,
|
| 1132 |
+
"rewards/chosen": 0.2849633991718292,
|
| 1133 |
+
"rewards/margins": 11.423474311828613,
|
| 1134 |
+
"rewards/rejected": -11.138509750366211,
|
| 1135 |
+
"step": 750
|
| 1136 |
+
},
|
| 1137 |
+
{
|
| 1138 |
+
"epoch": 19.0,
|
| 1139 |
+
"grad_norm": 0.015563180670142174,
|
| 1140 |
+
"learning_rate": 6.294095225512604e-06,
|
| 1141 |
+
"logits/chosen": -2.7435309886932373,
|
| 1142 |
+
"logits/rejected": -2.7662229537963867,
|
| 1143 |
+
"logps/chosen": -313.7065734863281,
|
| 1144 |
+
"logps/rejected": -435.94647216796875,
|
| 1145 |
+
"loss": 0.0,
|
| 1146 |
+
"rewards/accuracies": 1.0,
|
| 1147 |
+
"rewards/chosen": -0.47863197326660156,
|
| 1148 |
+
"rewards/margins": 11.74044418334961,
|
| 1149 |
+
"rewards/rejected": -12.219076156616211,
|
| 1150 |
+
"step": 760
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 19.25,
|
| 1154 |
+
"grad_norm": 0.010865326970815659,
|
| 1155 |
+
"learning_rate": 6.188429461630866e-06,
|
| 1156 |
+
"logits/chosen": -2.7403242588043213,
|
| 1157 |
+
"logits/rejected": -2.7853188514709473,
|
| 1158 |
+
"logps/chosen": -310.6920166015625,
|
| 1159 |
+
"logps/rejected": -450.71630859375,
|
| 1160 |
+
"loss": 0.0147,
|
| 1161 |
+
"rewards/accuracies": 1.0,
|
| 1162 |
+
"rewards/chosen": 0.3375871777534485,
|
| 1163 |
+
"rewards/margins": 11.067597389221191,
|
| 1164 |
+
"rewards/rejected": -10.730010032653809,
|
| 1165 |
+
"step": 770
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 19.5,
|
| 1169 |
+
"grad_norm": 0.011109190993010998,
|
| 1170 |
+
"learning_rate": 6.0821980696905145e-06,
|
| 1171 |
+
"logits/chosen": -2.7623090744018555,
|
| 1172 |
+
"logits/rejected": -2.8018345832824707,
|
| 1173 |
+
"logps/chosen": -275.25665283203125,
|
| 1174 |
+
"logps/rejected": -388.6085205078125,
|
| 1175 |
+
"loss": 0.0,
|
| 1176 |
+
"rewards/accuracies": 1.0,
|
| 1177 |
+
"rewards/chosen": 0.39552536606788635,
|
| 1178 |
+
"rewards/margins": 11.198080062866211,
|
| 1179 |
+
"rewards/rejected": -10.802556037902832,
|
| 1180 |
+
"step": 780
|
| 1181 |
+
},
|
| 1182 |
+
{
|
| 1183 |
+
"epoch": 19.75,
|
| 1184 |
+
"grad_norm": 0.13688191771507263,
|
| 1185 |
+
"learning_rate": 5.975451610080643e-06,
|
| 1186 |
+
"logits/chosen": -2.792748212814331,
|
| 1187 |
+
"logits/rejected": -2.8350272178649902,
|
| 1188 |
+
"logps/chosen": -280.85430908203125,
|
| 1189 |
+
"logps/rejected": -433.8482971191406,
|
| 1190 |
+
"loss": 0.0001,
|
| 1191 |
+
"rewards/accuracies": 1.0,
|
| 1192 |
+
"rewards/chosen": 0.1213034838438034,
|
| 1193 |
+
"rewards/margins": 11.554020881652832,
|
| 1194 |
+
"rewards/rejected": -11.432718276977539,
|
| 1195 |
+
"step": 790
|
| 1196 |
+
},
|
| 1197 |
+
{
|
| 1198 |
+
"epoch": 20.0,
|
| 1199 |
+
"grad_norm": 0.012268920429050922,
|
| 1200 |
+
"learning_rate": 5.8682408883346535e-06,
|
| 1201 |
+
"logits/chosen": -2.7681257724761963,
|
| 1202 |
+
"logits/rejected": -2.7897846698760986,
|
| 1203 |
+
"logps/chosen": -304.73272705078125,
|
| 1204 |
+
"logps/rejected": -409.9903869628906,
|
| 1205 |
+
"loss": 0.0,
|
| 1206 |
+
"rewards/accuracies": 1.0,
|
| 1207 |
+
"rewards/chosen": 0.6212397813796997,
|
| 1208 |
+
"rewards/margins": 11.226862907409668,
|
| 1209 |
+
"rewards/rejected": -10.605623245239258,
|
| 1210 |
+
"step": 800
|
| 1211 |
+
},
|
| 1212 |
+
{
|
| 1213 |
+
"epoch": 20.25,
|
| 1214 |
+
"grad_norm": 0.012932064011693,
|
| 1215 |
+
"learning_rate": 5.760616930949584e-06,
|
| 1216 |
+
"logits/chosen": -2.770331859588623,
|
| 1217 |
+
"logits/rejected": -2.7890381813049316,
|
| 1218 |
+
"logps/chosen": -306.2134704589844,
|
| 1219 |
+
"logps/rejected": -418.7654724121094,
|
| 1220 |
+
"loss": 0.0,
|
| 1221 |
+
"rewards/accuracies": 1.0,
|
| 1222 |
+
"rewards/chosen": 0.3504090905189514,
|
| 1223 |
+
"rewards/margins": 11.633293151855469,
|
| 1224 |
+
"rewards/rejected": -11.28288459777832,
|
| 1225 |
+
"step": 810
|
| 1226 |
+
},
|
| 1227 |
+
{
|
| 1228 |
+
"epoch": 20.5,
|
| 1229 |
+
"grad_norm": 0.006929120514541864,
|
| 1230 |
+
"learning_rate": 5.65263096110026e-06,
|
| 1231 |
+
"logits/chosen": -2.7397124767303467,
|
| 1232 |
+
"logits/rejected": -2.7778358459472656,
|
| 1233 |
+
"logps/chosen": -308.2324523925781,
|
| 1234 |
+
"logps/rejected": -470.67791748046875,
|
| 1235 |
+
"loss": 0.0006,
|
| 1236 |
+
"rewards/accuracies": 1.0,
|
| 1237 |
+
"rewards/chosen": 0.12212464958429337,
|
| 1238 |
+
"rewards/margins": 10.697164535522461,
|
| 1239 |
+
"rewards/rejected": -10.575040817260742,
|
| 1240 |
+
"step": 820
|
| 1241 |
+
},
|
| 1242 |
+
{
|
| 1243 |
+
"epoch": 20.75,
|
| 1244 |
+
"grad_norm": 0.0058187334798276424,
|
| 1245 |
+
"learning_rate": 5.544334374259823e-06,
|
| 1246 |
+
"logits/chosen": -2.820895195007324,
|
| 1247 |
+
"logits/rejected": -2.853609085083008,
|
| 1248 |
+
"logps/chosen": -296.0104064941406,
|
| 1249 |
+
"logps/rejected": -399.3705749511719,
|
| 1250 |
+
"loss": 0.0,
|
| 1251 |
+
"rewards/accuracies": 1.0,
|
| 1252 |
+
"rewards/chosen": 0.6735066175460815,
|
| 1253 |
+
"rewards/margins": 11.60952091217041,
|
| 1254 |
+
"rewards/rejected": -10.936014175415039,
|
| 1255 |
+
"step": 830
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 21.0,
|
| 1259 |
+
"grad_norm": 0.00335478107444942,
|
| 1260 |
+
"learning_rate": 5.435778713738292e-06,
|
| 1261 |
+
"logits/chosen": -2.7048516273498535,
|
| 1262 |
+
"logits/rejected": -2.756258726119995,
|
| 1263 |
+
"logps/chosen": -254.4524688720703,
|
| 1264 |
+
"logps/rejected": -396.0684814453125,
|
| 1265 |
+
"loss": 0.0,
|
| 1266 |
+
"rewards/accuracies": 1.0,
|
| 1267 |
+
"rewards/chosen": 0.6581499576568604,
|
| 1268 |
+
"rewards/margins": 11.730459213256836,
|
| 1269 |
+
"rewards/rejected": -11.072309494018555,
|
| 1270 |
+
"step": 840
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 21.25,
|
| 1274 |
+
"grad_norm": 0.018709037452936172,
|
| 1275 |
+
"learning_rate": 5.327015646150716e-06,
|
| 1276 |
+
"logits/chosen": -2.6166112422943115,
|
| 1277 |
+
"logits/rejected": -2.677870273590088,
|
| 1278 |
+
"logps/chosen": -297.18731689453125,
|
| 1279 |
+
"logps/rejected": -451.8585510253906,
|
| 1280 |
+
"loss": 0.0059,
|
| 1281 |
+
"rewards/accuracies": 1.0,
|
| 1282 |
+
"rewards/chosen": 0.0066245137713849545,
|
| 1283 |
+
"rewards/margins": 10.767498970031738,
|
| 1284 |
+
"rewards/rejected": -10.760875701904297,
|
| 1285 |
+
"step": 850
|
| 1286 |
+
},
|
| 1287 |
+
{
|
| 1288 |
+
"epoch": 21.5,
|
| 1289 |
+
"grad_norm": 0.00640977593138814,
|
| 1290 |
+
"learning_rate": 5.218096936826681e-06,
|
| 1291 |
+
"logits/chosen": -2.7767691612243652,
|
| 1292 |
+
"logits/rejected": -2.782492160797119,
|
| 1293 |
+
"logps/chosen": -281.61224365234375,
|
| 1294 |
+
"logps/rejected": -401.4766845703125,
|
| 1295 |
+
"loss": 0.0001,
|
| 1296 |
+
"rewards/accuracies": 1.0,
|
| 1297 |
+
"rewards/chosen": 0.07375653833150864,
|
| 1298 |
+
"rewards/margins": 11.642614364624023,
|
| 1299 |
+
"rewards/rejected": -11.568860054016113,
|
| 1300 |
+
"step": 860
|
| 1301 |
+
},
|
| 1302 |
+
{
|
| 1303 |
+
"epoch": 21.75,
|
| 1304 |
+
"grad_norm": 0.018667172640562057,
|
| 1305 |
+
"learning_rate": 5.109074425172806e-06,
|
| 1306 |
+
"logits/chosen": -2.754488468170166,
|
| 1307 |
+
"logits/rejected": -2.797163486480713,
|
| 1308 |
+
"logps/chosen": -284.8846435546875,
|
| 1309 |
+
"logps/rejected": -415.6575622558594,
|
| 1310 |
+
"loss": 0.0,
|
| 1311 |
+
"rewards/accuracies": 1.0,
|
| 1312 |
+
"rewards/chosen": 0.540486216545105,
|
| 1313 |
+
"rewards/margins": 11.744099617004395,
|
| 1314 |
+
"rewards/rejected": -11.20361328125,
|
| 1315 |
+
"step": 870
|
| 1316 |
+
},
|
| 1317 |
+
{
|
| 1318 |
+
"epoch": 22.0,
|
| 1319 |
+
"grad_norm": 0.003822013735771179,
|
| 1320 |
+
"learning_rate": 5e-06,
|
| 1321 |
+
"logits/chosen": -2.859811782836914,
|
| 1322 |
+
"logits/rejected": -2.891792058944702,
|
| 1323 |
+
"logps/chosen": -310.1468200683594,
|
| 1324 |
+
"logps/rejected": -433.6126403808594,
|
| 1325 |
+
"loss": 0.0,
|
| 1326 |
+
"rewards/accuracies": 1.0,
|
| 1327 |
+
"rewards/chosen": 0.2926860451698303,
|
| 1328 |
+
"rewards/margins": 12.372417449951172,
|
| 1329 |
+
"rewards/rejected": -12.079731941223145,
|
| 1330 |
+
"step": 880
|
| 1331 |
+
},
|
| 1332 |
+
{
|
| 1333 |
+
"epoch": 22.25,
|
| 1334 |
+
"grad_norm": 0.010801389813423157,
|
| 1335 |
+
"learning_rate": 4.890925574827195e-06,
|
| 1336 |
+
"logits/chosen": -2.7312240600585938,
|
| 1337 |
+
"logits/rejected": -2.7544617652893066,
|
| 1338 |
+
"logps/chosen": -293.9632568359375,
|
| 1339 |
+
"logps/rejected": -422.1595764160156,
|
| 1340 |
+
"loss": 0.0,
|
| 1341 |
+
"rewards/accuracies": 1.0,
|
| 1342 |
+
"rewards/chosen": 0.030094634741544724,
|
| 1343 |
+
"rewards/margins": 11.75137996673584,
|
| 1344 |
+
"rewards/rejected": -11.721285820007324,
|
| 1345 |
+
"step": 890
|
| 1346 |
+
},
|
| 1347 |
+
{
|
| 1348 |
+
"epoch": 22.5,
|
| 1349 |
+
"grad_norm": 0.00307653215713799,
|
| 1350 |
+
"learning_rate": 4.781903063173321e-06,
|
| 1351 |
+
"logits/chosen": -2.7293999195098877,
|
| 1352 |
+
"logits/rejected": -2.801996946334839,
|
| 1353 |
+
"logps/chosen": -290.69244384765625,
|
| 1354 |
+
"logps/rejected": -423.0887756347656,
|
| 1355 |
+
"loss": 0.0,
|
| 1356 |
+
"rewards/accuracies": 1.0,
|
| 1357 |
+
"rewards/chosen": 0.5350134372711182,
|
| 1358 |
+
"rewards/margins": 12.045049667358398,
|
| 1359 |
+
"rewards/rejected": -11.510037422180176,
|
| 1360 |
+
"step": 900
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 22.75,
|
| 1364 |
+
"grad_norm": 0.0032190086785703897,
|
| 1365 |
+
"learning_rate": 4.672984353849285e-06,
|
| 1366 |
+
"logits/chosen": -2.7523961067199707,
|
| 1367 |
+
"logits/rejected": -2.7586894035339355,
|
| 1368 |
+
"logps/chosen": -295.1702575683594,
|
| 1369 |
+
"logps/rejected": -440.543701171875,
|
| 1370 |
+
"loss": 0.0,
|
| 1371 |
+
"rewards/accuracies": 1.0,
|
| 1372 |
+
"rewards/chosen": 0.17444385588169098,
|
| 1373 |
+
"rewards/margins": 12.515422821044922,
|
| 1374 |
+
"rewards/rejected": -12.340980529785156,
|
| 1375 |
+
"step": 910
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 23.0,
|
| 1379 |
+
"grad_norm": 0.013608491979539394,
|
| 1380 |
+
"learning_rate": 4.564221286261709e-06,
|
| 1381 |
+
"logits/chosen": -2.7768361568450928,
|
| 1382 |
+
"logits/rejected": -2.829939126968384,
|
| 1383 |
+
"logps/chosen": -296.3355712890625,
|
| 1384 |
+
"logps/rejected": -421.0843811035156,
|
| 1385 |
+
"loss": 0.0216,
|
| 1386 |
+
"rewards/accuracies": 0.9750000238418579,
|
| 1387 |
+
"rewards/chosen": -0.010329616256058216,
|
| 1388 |
+
"rewards/margins": 10.561881065368652,
|
| 1389 |
+
"rewards/rejected": -10.572213172912598,
|
| 1390 |
+
"step": 920
|
| 1391 |
+
},
|
| 1392 |
+
{
|
| 1393 |
+
"epoch": 23.25,
|
| 1394 |
+
"grad_norm": 0.010324417613446712,
|
| 1395 |
+
"learning_rate": 4.4556656257401786e-06,
|
| 1396 |
+
"logits/chosen": -2.814089298248291,
|
| 1397 |
+
"logits/rejected": -2.8441967964172363,
|
| 1398 |
+
"logps/chosen": -245.9521026611328,
|
| 1399 |
+
"logps/rejected": -392.0475158691406,
|
| 1400 |
+
"loss": 0.0,
|
| 1401 |
+
"rewards/accuracies": 1.0,
|
| 1402 |
+
"rewards/chosen": 0.9622662663459778,
|
| 1403 |
+
"rewards/margins": 11.689537048339844,
|
| 1404 |
+
"rewards/rejected": -10.72727108001709,
|
| 1405 |
+
"step": 930
|
| 1406 |
+
},
|
| 1407 |
+
{
|
| 1408 |
+
"epoch": 23.5,
|
| 1409 |
+
"grad_norm": 0.00877166073769331,
|
| 1410 |
+
"learning_rate": 4.347369038899744e-06,
|
| 1411 |
+
"logits/chosen": -2.763094186782837,
|
| 1412 |
+
"logits/rejected": -2.803144931793213,
|
| 1413 |
+
"logps/chosen": -303.9397277832031,
|
| 1414 |
+
"logps/rejected": -402.18841552734375,
|
| 1415 |
+
"loss": 0.0,
|
| 1416 |
+
"rewards/accuracies": 1.0,
|
| 1417 |
+
"rewards/chosen": 0.5226415991783142,
|
| 1418 |
+
"rewards/margins": 11.711947441101074,
|
| 1419 |
+
"rewards/rejected": -11.18930721282959,
|
| 1420 |
+
"step": 940
|
| 1421 |
+
},
|
| 1422 |
+
{
|
| 1423 |
+
"epoch": 23.75,
|
| 1424 |
+
"grad_norm": 0.012196135707199574,
|
| 1425 |
+
"learning_rate": 4.239383069050417e-06,
|
| 1426 |
+
"logits/chosen": -2.7214579582214355,
|
| 1427 |
+
"logits/rejected": -2.751098155975342,
|
| 1428 |
+
"logps/chosen": -308.78851318359375,
|
| 1429 |
+
"logps/rejected": -460.69281005859375,
|
| 1430 |
+
"loss": 0.0,
|
| 1431 |
+
"rewards/accuracies": 1.0,
|
| 1432 |
+
"rewards/chosen": -0.11721920967102051,
|
| 1433 |
+
"rewards/margins": 12.034148216247559,
|
| 1434 |
+
"rewards/rejected": -12.151368141174316,
|
| 1435 |
+
"step": 950
|
| 1436 |
+
},
|
| 1437 |
+
{
|
| 1438 |
+
"epoch": 24.0,
|
| 1439 |
+
"grad_norm": 0.016689885407686234,
|
| 1440 |
+
"learning_rate": 4.131759111665349e-06,
|
| 1441 |
+
"logits/chosen": -2.7126431465148926,
|
| 1442 |
+
"logits/rejected": -2.766502618789673,
|
| 1443 |
+
"logps/chosen": -310.998291015625,
|
| 1444 |
+
"logps/rejected": -442.939697265625,
|
| 1445 |
+
"loss": 0.0001,
|
| 1446 |
+
"rewards/accuracies": 1.0,
|
| 1447 |
+
"rewards/chosen": 0.4348395764827728,
|
| 1448 |
+
"rewards/margins": 11.327308654785156,
|
| 1449 |
+
"rewards/rejected": -10.892468452453613,
|
| 1450 |
+
"step": 960
|
| 1451 |
+
},
|
| 1452 |
+
{
|
| 1453 |
+
"epoch": 24.25,
|
| 1454 |
+
"grad_norm": 0.7392864227294922,
|
| 1455 |
+
"learning_rate": 4.02454838991936e-06,
|
| 1456 |
+
"logits/chosen": -2.6969475746154785,
|
| 1457 |
+
"logits/rejected": -2.702312469482422,
|
| 1458 |
+
"logps/chosen": -306.0939025878906,
|
| 1459 |
+
"logps/rejected": -427.08074951171875,
|
| 1460 |
+
"loss": 0.0002,
|
| 1461 |
+
"rewards/accuracies": 1.0,
|
| 1462 |
+
"rewards/chosen": 0.31883805990219116,
|
| 1463 |
+
"rewards/margins": 11.24242115020752,
|
| 1464 |
+
"rewards/rejected": -10.923582077026367,
|
| 1465 |
+
"step": 970
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 24.5,
|
| 1469 |
+
"grad_norm": 0.003211956238374114,
|
| 1470 |
+
"learning_rate": 3.917801930309486e-06,
|
| 1471 |
+
"logits/chosen": -2.782787322998047,
|
| 1472 |
+
"logits/rejected": -2.808108329772949,
|
| 1473 |
+
"logps/chosen": -258.2384033203125,
|
| 1474 |
+
"logps/rejected": -389.11798095703125,
|
| 1475 |
+
"loss": 0.0,
|
| 1476 |
+
"rewards/accuracies": 1.0,
|
| 1477 |
+
"rewards/chosen": 1.0415832996368408,
|
| 1478 |
+
"rewards/margins": 11.802536010742188,
|
| 1479 |
+
"rewards/rejected": -10.760952949523926,
|
| 1480 |
+
"step": 980
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"epoch": 24.75,
|
| 1484 |
+
"grad_norm": 0.003792324336245656,
|
| 1485 |
+
"learning_rate": 3.8115705383691354e-06,
|
| 1486 |
+
"logits/chosen": -2.756258010864258,
|
| 1487 |
+
"logits/rejected": -2.803229331970215,
|
| 1488 |
+
"logps/chosen": -304.2614440917969,
|
| 1489 |
+
"logps/rejected": -457.0116271972656,
|
| 1490 |
+
"loss": 0.0,
|
| 1491 |
+
"rewards/accuracies": 1.0,
|
| 1492 |
+
"rewards/chosen": -0.01773083209991455,
|
| 1493 |
+
"rewards/margins": 12.328390121459961,
|
| 1494 |
+
"rewards/rejected": -12.346120834350586,
|
| 1495 |
+
"step": 990
|
| 1496 |
+
},
|
| 1497 |
+
{
|
| 1498 |
+
"epoch": 25.0,
|
| 1499 |
+
"grad_norm": 0.00849173590540886,
|
| 1500 |
+
"learning_rate": 3.705904774487396e-06,
|
| 1501 |
+
"logits/chosen": -2.7826552391052246,
|
| 1502 |
+
"logits/rejected": -2.8513171672821045,
|
| 1503 |
+
"logps/chosen": -300.17535400390625,
|
| 1504 |
+
"logps/rejected": -427.40234375,
|
| 1505 |
+
"loss": 0.0,
|
| 1506 |
+
"rewards/accuracies": 1.0,
|
| 1507 |
+
"rewards/chosen": 0.5104937553405762,
|
| 1508 |
+
"rewards/margins": 11.4827880859375,
|
| 1509 |
+
"rewards/rejected": -10.972294807434082,
|
| 1510 |
+
"step": 1000
|
| 1511 |
+
},
|
| 1512 |
+
{
|
| 1513 |
+
"epoch": 25.25,
|
| 1514 |
+
"grad_norm": 0.00654919259250164,
|
| 1515 |
+
"learning_rate": 3.6008549298450403e-06,
|
| 1516 |
+
"logits/chosen": -2.7531707286834717,
|
| 1517 |
+
"logits/rejected": -2.7878074645996094,
|
| 1518 |
+
"logps/chosen": -284.87396240234375,
|
| 1519 |
+
"logps/rejected": -390.9798583984375,
|
| 1520 |
+
"loss": 0.0,
|
| 1521 |
+
"rewards/accuracies": 1.0,
|
| 1522 |
+
"rewards/chosen": 0.6618450880050659,
|
| 1523 |
+
"rewards/margins": 11.684257507324219,
|
| 1524 |
+
"rewards/rejected": -11.022411346435547,
|
| 1525 |
+
"step": 1010
|
| 1526 |
+
},
|
| 1527 |
+
{
|
| 1528 |
+
"epoch": 25.5,
|
| 1529 |
+
"grad_norm": 0.0041053337045013905,
|
| 1530 |
+
"learning_rate": 3.4964710024786354e-06,
|
| 1531 |
+
"logits/chosen": -2.6924521923065186,
|
| 1532 |
+
"logits/rejected": -2.7480525970458984,
|
| 1533 |
+
"logps/chosen": -282.3283386230469,
|
| 1534 |
+
"logps/rejected": -394.1541442871094,
|
| 1535 |
+
"loss": 0.0,
|
| 1536 |
+
"rewards/accuracies": 1.0,
|
| 1537 |
+
"rewards/chosen": 0.988944411277771,
|
| 1538 |
+
"rewards/margins": 11.823431015014648,
|
| 1539 |
+
"rewards/rejected": -10.83448600769043,
|
| 1540 |
+
"step": 1020
|
| 1541 |
+
},
|
| 1542 |
+
{
|
| 1543 |
+
"epoch": 25.75,
|
| 1544 |
+
"grad_norm": 0.004295062739402056,
|
| 1545 |
+
"learning_rate": 3.3928026734841935e-06,
|
| 1546 |
+
"logits/chosen": -2.7896180152893066,
|
| 1547 |
+
"logits/rejected": -2.833456516265869,
|
| 1548 |
+
"logps/chosen": -286.898193359375,
|
| 1549 |
+
"logps/rejected": -469.9169006347656,
|
| 1550 |
+
"loss": 0.0005,
|
| 1551 |
+
"rewards/accuracies": 1.0,
|
| 1552 |
+
"rewards/chosen": 0.3953208327293396,
|
| 1553 |
+
"rewards/margins": 11.806564331054688,
|
| 1554 |
+
"rewards/rejected": -11.41124153137207,
|
| 1555 |
+
"step": 1030
|
| 1556 |
+
},
|
| 1557 |
+
{
|
| 1558 |
+
"epoch": 26.0,
|
| 1559 |
+
"grad_norm": 0.007940116338431835,
|
| 1560 |
+
"learning_rate": 3.289899283371657e-06,
|
| 1561 |
+
"logits/chosen": -2.7841601371765137,
|
| 1562 |
+
"logits/rejected": -2.8034167289733887,
|
| 1563 |
+
"logps/chosen": -310.6112060546875,
|
| 1564 |
+
"logps/rejected": -443.7947692871094,
|
| 1565 |
+
"loss": 0.0001,
|
| 1566 |
+
"rewards/accuracies": 1.0,
|
| 1567 |
+
"rewards/chosen": -0.23492594063282013,
|
| 1568 |
+
"rewards/margins": 11.810868263244629,
|
| 1569 |
+
"rewards/rejected": -12.045793533325195,
|
| 1570 |
+
"step": 1040
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 26.25,
|
| 1574 |
+
"grad_norm": 0.006178665440529585,
|
| 1575 |
+
"learning_rate": 3.1878098085814926e-06,
|
| 1576 |
+
"logits/chosen": -2.7314186096191406,
|
| 1577 |
+
"logits/rejected": -2.7590575218200684,
|
| 1578 |
+
"logps/chosen": -291.75933837890625,
|
| 1579 |
+
"logps/rejected": -405.439208984375,
|
| 1580 |
+
"loss": 0.0,
|
| 1581 |
+
"rewards/accuracies": 1.0,
|
| 1582 |
+
"rewards/chosen": 0.7959985136985779,
|
| 1583 |
+
"rewards/margins": 11.849149703979492,
|
| 1584 |
+
"rewards/rejected": -11.05315113067627,
|
| 1585 |
+
"step": 1050
|
| 1586 |
+
},
|
| 1587 |
+
{
|
| 1588 |
+
"epoch": 26.5,
|
| 1589 |
+
"grad_norm": 0.0034364075399935246,
|
| 1590 |
+
"learning_rate": 3.0865828381745515e-06,
|
| 1591 |
+
"logits/chosen": -2.6799168586730957,
|
| 1592 |
+
"logits/rejected": -2.752761125564575,
|
| 1593 |
+
"logps/chosen": -320.683837890625,
|
| 1594 |
+
"logps/rejected": -476.5531311035156,
|
| 1595 |
+
"loss": 0.0017,
|
| 1596 |
+
"rewards/accuracies": 1.0,
|
| 1597 |
+
"rewards/chosen": -0.32047194242477417,
|
| 1598 |
+
"rewards/margins": 11.475412368774414,
|
| 1599 |
+
"rewards/rejected": -11.79588508605957,
|
| 1600 |
+
"step": 1060
|
| 1601 |
+
},
|
| 1602 |
+
{
|
| 1603 |
+
"epoch": 26.75,
|
| 1604 |
+
"grad_norm": 0.012047665193676949,
|
| 1605 |
+
"learning_rate": 2.986266550706315e-06,
|
| 1606 |
+
"logits/chosen": -2.823770523071289,
|
| 1607 |
+
"logits/rejected": -2.831331729888916,
|
| 1608 |
+
"logps/chosen": -292.18621826171875,
|
| 1609 |
+
"logps/rejected": -409.607177734375,
|
| 1610 |
+
"loss": 0.0,
|
| 1611 |
+
"rewards/accuracies": 1.0,
|
| 1612 |
+
"rewards/chosen": 0.3296182155609131,
|
| 1613 |
+
"rewards/margins": 12.108281135559082,
|
| 1614 |
+
"rewards/rejected": -11.77866268157959,
|
| 1615 |
+
"step": 1070
|
| 1616 |
+
},
|
| 1617 |
+
{
|
| 1618 |
+
"epoch": 27.0,
|
| 1619 |
+
"grad_norm": 0.00699230283498764,
|
| 1620 |
+
"learning_rate": 2.886908691296504e-06,
|
| 1621 |
+
"logits/chosen": -2.768615961074829,
|
| 1622 |
+
"logits/rejected": -2.8066773414611816,
|
| 1623 |
+
"logps/chosen": -266.96826171875,
|
| 1624 |
+
"logps/rejected": -419.3490295410156,
|
| 1625 |
+
"loss": 0.0,
|
| 1626 |
+
"rewards/accuracies": 1.0,
|
| 1627 |
+
"rewards/chosen": 0.5609016418457031,
|
| 1628 |
+
"rewards/margins": 12.091551780700684,
|
| 1629 |
+
"rewards/rejected": -11.530649185180664,
|
| 1630 |
+
"step": 1080
|
| 1631 |
+
},
|
| 1632 |
+
{
|
| 1633 |
+
"epoch": 27.25,
|
| 1634 |
+
"grad_norm": 0.0021058099810034037,
|
| 1635 |
+
"learning_rate": 2.7885565489049948e-06,
|
| 1636 |
+
"logits/chosen": -2.813713550567627,
|
| 1637 |
+
"logits/rejected": -2.8499045372009277,
|
| 1638 |
+
"logps/chosen": -264.84381103515625,
|
| 1639 |
+
"logps/rejected": -419.009765625,
|
| 1640 |
+
"loss": 0.0,
|
| 1641 |
+
"rewards/accuracies": 1.0,
|
| 1642 |
+
"rewards/chosen": 0.655460000038147,
|
| 1643 |
+
"rewards/margins": 12.120909690856934,
|
| 1644 |
+
"rewards/rejected": -11.465449333190918,
|
| 1645 |
+
"step": 1090
|
| 1646 |
+
},
|
| 1647 |
+
{
|
| 1648 |
+
"epoch": 27.5,
|
| 1649 |
+
"grad_norm": 0.011556684039533138,
|
| 1650 |
+
"learning_rate": 2.6912569338248317e-06,
|
| 1651 |
+
"logits/chosen": -2.765392780303955,
|
| 1652 |
+
"logits/rejected": -2.78332781791687,
|
| 1653 |
+
"logps/chosen": -313.57452392578125,
|
| 1654 |
+
"logps/rejected": -453.7864685058594,
|
| 1655 |
+
"loss": 0.0002,
|
| 1656 |
+
"rewards/accuracies": 1.0,
|
| 1657 |
+
"rewards/chosen": -0.08190927654504776,
|
| 1658 |
+
"rewards/margins": 12.152857780456543,
|
| 1659 |
+
"rewards/rejected": -12.23476791381836,
|
| 1660 |
+
"step": 1100
|
| 1661 |
+
},
|
| 1662 |
+
{
|
| 1663 |
+
"epoch": 27.75,
|
| 1664 |
+
"grad_norm": 0.01144924946129322,
|
| 1665 |
+
"learning_rate": 2.595056155403063e-06,
|
| 1666 |
+
"logits/chosen": -2.7482686042785645,
|
| 1667 |
+
"logits/rejected": -2.7814669609069824,
|
| 1668 |
+
"logps/chosen": -288.89404296875,
|
| 1669 |
+
"logps/rejected": -417.9188537597656,
|
| 1670 |
+
"loss": 0.0,
|
| 1671 |
+
"rewards/accuracies": 1.0,
|
| 1672 |
+
"rewards/chosen": 0.6152423620223999,
|
| 1673 |
+
"rewards/margins": 12.10200309753418,
|
| 1674 |
+
"rewards/rejected": -11.486761093139648,
|
| 1675 |
+
"step": 1110
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 28.0,
|
| 1679 |
+
"grad_norm": 0.007494707591831684,
|
| 1680 |
+
"learning_rate": 2.5000000000000015e-06,
|
| 1681 |
+
"logits/chosen": -2.678860902786255,
|
| 1682 |
+
"logits/rejected": -2.7341535091400146,
|
| 1683 |
+
"logps/chosen": -302.2644958496094,
|
| 1684 |
+
"logps/rejected": -424.6878967285156,
|
| 1685 |
+
"loss": 0.0,
|
| 1686 |
+
"rewards/accuracies": 1.0,
|
| 1687 |
+
"rewards/chosen": 0.14390479028224945,
|
| 1688 |
+
"rewards/margins": 11.76207160949707,
|
| 1689 |
+
"rewards/rejected": -11.618168830871582,
|
| 1690 |
+
"step": 1120
|
| 1691 |
+
},
|
| 1692 |
+
{
|
| 1693 |
+
"epoch": 28.25,
|
| 1694 |
+
"grad_norm": 0.00261974660679698,
|
| 1695 |
+
"learning_rate": 2.406133709197392e-06,
|
| 1696 |
+
"logits/chosen": -2.8185958862304688,
|
| 1697 |
+
"logits/rejected": -2.8240275382995605,
|
| 1698 |
+
"logps/chosen": -288.8084411621094,
|
| 1699 |
+
"logps/rejected": -427.37762451171875,
|
| 1700 |
+
"loss": 0.0049,
|
| 1701 |
+
"rewards/accuracies": 1.0,
|
| 1702 |
+
"rewards/chosen": 0.27108827233314514,
|
| 1703 |
+
"rewards/margins": 11.560935020446777,
|
| 1704 |
+
"rewards/rejected": -11.289847373962402,
|
| 1705 |
+
"step": 1130
|
| 1706 |
+
},
|
| 1707 |
+
{
|
| 1708 |
+
"epoch": 28.5,
|
| 1709 |
+
"grad_norm": 0.006601645611226559,
|
| 1710 |
+
"learning_rate": 2.3135019582658803e-06,
|
| 1711 |
+
"logits/chosen": -2.759382486343384,
|
| 1712 |
+
"logits/rejected": -2.8069920539855957,
|
| 1713 |
+
"logps/chosen": -298.5120544433594,
|
| 1714 |
+
"logps/rejected": -424.13360595703125,
|
| 1715 |
+
"loss": 0.0,
|
| 1716 |
+
"rewards/accuracies": 1.0,
|
| 1717 |
+
"rewards/chosen": 0.5905256271362305,
|
| 1718 |
+
"rewards/margins": 12.292265892028809,
|
| 1719 |
+
"rewards/rejected": -11.701741218566895,
|
| 1720 |
+
"step": 1140
|
| 1721 |
+
},
|
| 1722 |
+
{
|
| 1723 |
+
"epoch": 28.75,
|
| 1724 |
+
"grad_norm": 0.006943762768059969,
|
| 1725 |
+
"learning_rate": 2.2221488349019903e-06,
|
| 1726 |
+
"logits/chosen": -2.6638615131378174,
|
| 1727 |
+
"logits/rejected": -2.73032546043396,
|
| 1728 |
+
"logps/chosen": -294.6612548828125,
|
| 1729 |
+
"logps/rejected": -444.9859313964844,
|
| 1730 |
+
"loss": 0.0,
|
| 1731 |
+
"rewards/accuracies": 1.0,
|
| 1732 |
+
"rewards/chosen": -0.28661757707595825,
|
| 1733 |
+
"rewards/margins": 12.06439208984375,
|
| 1734 |
+
"rewards/rejected": -12.3510103225708,
|
| 1735 |
+
"step": 1150
|
| 1736 |
+
},
|
| 1737 |
+
{
|
| 1738 |
+
"epoch": 29.0,
|
| 1739 |
+
"grad_norm": 0.0011426012497395277,
|
| 1740 |
+
"learning_rate": 2.132117818244771e-06,
|
| 1741 |
+
"logits/chosen": -2.7360968589782715,
|
| 1742 |
+
"logits/rejected": -2.767688512802124,
|
| 1743 |
+
"logps/chosen": -291.3235168457031,
|
| 1744 |
+
"logps/rejected": -420.4112854003906,
|
| 1745 |
+
"loss": 0.0,
|
| 1746 |
+
"rewards/accuracies": 1.0,
|
| 1747 |
+
"rewards/chosen": 0.6397272348403931,
|
| 1748 |
+
"rewards/margins": 12.181781768798828,
|
| 1749 |
+
"rewards/rejected": -11.542055130004883,
|
| 1750 |
+
"step": 1160
|
| 1751 |
+
},
|
| 1752 |
+
{
|
| 1753 |
+
"epoch": 29.25,
|
| 1754 |
+
"grad_norm": 0.003964864648878574,
|
| 1755 |
+
"learning_rate": 2.0434517581820893e-06,
|
| 1756 |
+
"logits/chosen": -2.6658082008361816,
|
| 1757 |
+
"logits/rejected": -2.717425584793091,
|
| 1758 |
+
"logps/chosen": -343.0668640136719,
|
| 1759 |
+
"logps/rejected": -475.0995178222656,
|
| 1760 |
+
"loss": 0.0002,
|
| 1761 |
+
"rewards/accuracies": 1.0,
|
| 1762 |
+
"rewards/chosen": 0.19656455516815186,
|
| 1763 |
+
"rewards/margins": 11.766244888305664,
|
| 1764 |
+
"rewards/rejected": -11.569681167602539,
|
| 1765 |
+
"step": 1170
|
| 1766 |
+
},
|
| 1767 |
+
{
|
| 1768 |
+
"epoch": 29.5,
|
| 1769 |
+
"grad_norm": 0.003464967478066683,
|
| 1770 |
+
"learning_rate": 1.956192854956397e-06,
|
| 1771 |
+
"logits/chosen": -2.7914059162139893,
|
| 1772 |
+
"logits/rejected": -2.8226091861724854,
|
| 1773 |
+
"logps/chosen": -284.6141052246094,
|
| 1774 |
+
"logps/rejected": -396.912109375,
|
| 1775 |
+
"loss": 0.0,
|
| 1776 |
+
"rewards/accuracies": 1.0,
|
| 1777 |
+
"rewards/chosen": 0.4676376283168793,
|
| 1778 |
+
"rewards/margins": 12.429913520812988,
|
| 1779 |
+
"rewards/rejected": -11.962274551391602,
|
| 1780 |
+
"step": 1180
|
| 1781 |
+
},
|
| 1782 |
+
{
|
| 1783 |
+
"epoch": 29.75,
|
| 1784 |
+
"grad_norm": 0.0023241571616381407,
|
| 1785 |
+
"learning_rate": 1.8703826390797047e-06,
|
| 1786 |
+
"logits/chosen": -2.740171432495117,
|
| 1787 |
+
"logits/rejected": -2.7767977714538574,
|
| 1788 |
+
"logps/chosen": -265.77099609375,
|
| 1789 |
+
"logps/rejected": -446.67431640625,
|
| 1790 |
+
"loss": 0.0,
|
| 1791 |
+
"rewards/accuracies": 1.0,
|
| 1792 |
+
"rewards/chosen": 0.4891475737094879,
|
| 1793 |
+
"rewards/margins": 12.44828987121582,
|
| 1794 |
+
"rewards/rejected": -11.959141731262207,
|
| 1795 |
+
"step": 1190
|
| 1796 |
+
},
|
| 1797 |
+
{
|
| 1798 |
+
"epoch": 30.0,
|
| 1799 |
+
"grad_norm": 0.0072005875408649445,
|
| 1800 |
+
"learning_rate": 1.7860619515673034e-06,
|
| 1801 |
+
"logits/chosen": -2.7672040462493896,
|
| 1802 |
+
"logits/rejected": -2.8059945106506348,
|
| 1803 |
+
"logps/chosen": -281.13922119140625,
|
| 1804 |
+
"logps/rejected": -401.81390380859375,
|
| 1805 |
+
"loss": 0.0,
|
| 1806 |
+
"rewards/accuracies": 1.0,
|
| 1807 |
+
"rewards/chosen": -0.1551550179719925,
|
| 1808 |
+
"rewards/margins": 11.701306343078613,
|
| 1809 |
+
"rewards/rejected": -11.856461524963379,
|
| 1810 |
+
"step": 1200
|
| 1811 |
+
},
|
| 1812 |
+
{
|
| 1813 |
+
"epoch": 30.25,
|
| 1814 |
+
"grad_norm": 0.002375040901824832,
|
| 1815 |
+
"learning_rate": 1.7032709244996559e-06,
|
| 1816 |
+
"logits/chosen": -2.78956937789917,
|
| 1817 |
+
"logits/rejected": -2.806525468826294,
|
| 1818 |
+
"logps/chosen": -312.67327880859375,
|
| 1819 |
+
"logps/rejected": -466.92034912109375,
|
| 1820 |
+
"loss": 0.0,
|
| 1821 |
+
"rewards/accuracies": 1.0,
|
| 1822 |
+
"rewards/chosen": -0.4753616452217102,
|
| 1823 |
+
"rewards/margins": 12.765235900878906,
|
| 1824 |
+
"rewards/rejected": -13.24059772491455,
|
| 1825 |
+
"step": 1210
|
| 1826 |
+
},
|
| 1827 |
+
{
|
| 1828 |
+
"epoch": 30.5,
|
| 1829 |
+
"grad_norm": 0.007024294696748257,
|
| 1830 |
+
"learning_rate": 1.6220489619216988e-06,
|
| 1831 |
+
"logits/chosen": -2.6667380332946777,
|
| 1832 |
+
"logits/rejected": -2.7224926948547363,
|
| 1833 |
+
"logps/chosen": -296.1967468261719,
|
| 1834 |
+
"logps/rejected": -395.61114501953125,
|
| 1835 |
+
"loss": 0.0,
|
| 1836 |
+
"rewards/accuracies": 1.0,
|
| 1837 |
+
"rewards/chosen": 0.27107173204421997,
|
| 1838 |
+
"rewards/margins": 11.369800567626953,
|
| 1839 |
+
"rewards/rejected": -11.098730087280273,
|
| 1840 |
+
"step": 1220
|
| 1841 |
+
},
|
| 1842 |
+
{
|
| 1843 |
+
"epoch": 30.75,
|
| 1844 |
+
"grad_norm": 0.002501589013263583,
|
| 1845 |
+
"learning_rate": 1.5424347210886538e-06,
|
| 1846 |
+
"logits/chosen": -2.7434449195861816,
|
| 1847 |
+
"logits/rejected": -2.7846179008483887,
|
| 1848 |
+
"logps/chosen": -303.44464111328125,
|
| 1849 |
+
"logps/rejected": -444.24127197265625,
|
| 1850 |
+
"loss": 0.0,
|
| 1851 |
+
"rewards/accuracies": 1.0,
|
| 1852 |
+
"rewards/chosen": 0.6443105340003967,
|
| 1853 |
+
"rewards/margins": 12.676385879516602,
|
| 1854 |
+
"rewards/rejected": -12.032075881958008,
|
| 1855 |
+
"step": 1230
|
| 1856 |
+
},
|
| 1857 |
+
{
|
| 1858 |
+
"epoch": 31.0,
|
| 1859 |
+
"grad_norm": 0.005482690874487162,
|
| 1860 |
+
"learning_rate": 1.4644660940672628e-06,
|
| 1861 |
+
"logits/chosen": -2.7875418663024902,
|
| 1862 |
+
"logits/rejected": -2.7959892749786377,
|
| 1863 |
+
"logps/chosen": -263.8461608886719,
|
| 1864 |
+
"logps/rejected": -416.28662109375,
|
| 1865 |
+
"loss": 0.0002,
|
| 1866 |
+
"rewards/accuracies": 1.0,
|
| 1867 |
+
"rewards/chosen": 0.32634425163269043,
|
| 1868 |
+
"rewards/margins": 11.619983673095703,
|
| 1869 |
+
"rewards/rejected": -11.293639183044434,
|
| 1870 |
+
"step": 1240
|
| 1871 |
+
},
|
| 1872 |
+
{
|
| 1873 |
+
"epoch": 31.25,
|
| 1874 |
+
"grad_norm": 0.006386056076735258,
|
| 1875 |
+
"learning_rate": 1.3881801897012225e-06,
|
| 1876 |
+
"logits/chosen": -2.666128635406494,
|
| 1877 |
+
"logits/rejected": -2.6980326175689697,
|
| 1878 |
+
"logps/chosen": -307.4803771972656,
|
| 1879 |
+
"logps/rejected": -438.99853515625,
|
| 1880 |
+
"loss": 0.0002,
|
| 1881 |
+
"rewards/accuracies": 1.0,
|
| 1882 |
+
"rewards/chosen": 0.005978202912956476,
|
| 1883 |
+
"rewards/margins": 11.530052185058594,
|
| 1884 |
+
"rewards/rejected": -11.524072647094727,
|
| 1885 |
+
"step": 1250
|
| 1886 |
+
},
|
| 1887 |
+
{
|
| 1888 |
+
"epoch": 31.5,
|
| 1889 |
+
"grad_norm": 0.010775255970656872,
|
| 1890 |
+
"learning_rate": 1.3136133159493803e-06,
|
| 1891 |
+
"logits/chosen": -2.7379791736602783,
|
| 1892 |
+
"logits/rejected": -2.7717361450195312,
|
| 1893 |
+
"logps/chosen": -262.4990539550781,
|
| 1894 |
+
"logps/rejected": -398.7112731933594,
|
| 1895 |
+
"loss": 0.0,
|
| 1896 |
+
"rewards/accuracies": 1.0,
|
| 1897 |
+
"rewards/chosen": 0.6824764609336853,
|
| 1898 |
+
"rewards/margins": 12.771679878234863,
|
| 1899 |
+
"rewards/rejected": -12.089202880859375,
|
| 1900 |
+
"step": 1260
|
| 1901 |
+
},
|
| 1902 |
+
{
|
| 1903 |
+
"epoch": 31.75,
|
| 1904 |
+
"grad_norm": 0.005956208799034357,
|
| 1905 |
+
"learning_rate": 1.2408009626051137e-06,
|
| 1906 |
+
"logits/chosen": -2.8342833518981934,
|
| 1907 |
+
"logits/rejected": -2.8574931621551514,
|
| 1908 |
+
"logps/chosen": -311.24884033203125,
|
| 1909 |
+
"logps/rejected": -467.5516662597656,
|
| 1910 |
+
"loss": 0.0,
|
| 1911 |
+
"rewards/accuracies": 1.0,
|
| 1912 |
+
"rewards/chosen": -0.5595420002937317,
|
| 1913 |
+
"rewards/margins": 12.868734359741211,
|
| 1914 |
+
"rewards/rejected": -13.428276062011719,
|
| 1915 |
+
"step": 1270
|
| 1916 |
+
},
|
| 1917 |
+
{
|
| 1918 |
+
"epoch": 32.0,
|
| 1919 |
+
"grad_norm": 0.00918492116034031,
|
| 1920 |
+
"learning_rate": 1.1697777844051105e-06,
|
| 1921 |
+
"logits/chosen": -2.725320339202881,
|
| 1922 |
+
"logits/rejected": -2.775146245956421,
|
| 1923 |
+
"logps/chosen": -297.0390319824219,
|
| 1924 |
+
"logps/rejected": -427.73724365234375,
|
| 1925 |
+
"loss": 0.0001,
|
| 1926 |
+
"rewards/accuracies": 1.0,
|
| 1927 |
+
"rewards/chosen": 0.23792123794555664,
|
| 1928 |
+
"rewards/margins": 11.733413696289062,
|
| 1929 |
+
"rewards/rejected": -11.495491981506348,
|
| 1930 |
+
"step": 1280
|
| 1931 |
+
},
|
| 1932 |
+
{
|
| 1933 |
+
"epoch": 32.25,
|
| 1934 |
+
"grad_norm": 0.0025609093718230724,
|
| 1935 |
+
"learning_rate": 1.100577584535592e-06,
|
| 1936 |
+
"logits/chosen": -2.6968932151794434,
|
| 1937 |
+
"logits/rejected": -2.7568459510803223,
|
| 1938 |
+
"logps/chosen": -276.00592041015625,
|
| 1939 |
+
"logps/rejected": -420.32989501953125,
|
| 1940 |
+
"loss": 0.0,
|
| 1941 |
+
"rewards/accuracies": 1.0,
|
| 1942 |
+
"rewards/chosen": 0.28203171491622925,
|
| 1943 |
+
"rewards/margins": 12.333124160766602,
|
| 1944 |
+
"rewards/rejected": -12.051092147827148,
|
| 1945 |
+
"step": 1290
|
| 1946 |
+
},
|
| 1947 |
+
{
|
| 1948 |
+
"epoch": 32.5,
|
| 1949 |
+
"grad_norm": 0.002115601906552911,
|
| 1950 |
+
"learning_rate": 1.0332332985438248e-06,
|
| 1951 |
+
"logits/chosen": -2.7094054222106934,
|
| 1952 |
+
"logits/rejected": -2.7540245056152344,
|
| 1953 |
+
"logps/chosen": -278.0862121582031,
|
| 1954 |
+
"logps/rejected": -412.1825256347656,
|
| 1955 |
+
"loss": 0.0,
|
| 1956 |
+
"rewards/accuracies": 1.0,
|
| 1957 |
+
"rewards/chosen": 0.8267415165901184,
|
| 1958 |
+
"rewards/margins": 12.288800239562988,
|
| 1959 |
+
"rewards/rejected": -11.462059020996094,
|
| 1960 |
+
"step": 1300
|
| 1961 |
+
},
|
| 1962 |
+
{
|
| 1963 |
+
"epoch": 32.75,
|
| 1964 |
+
"grad_norm": 0.010958091355860233,
|
| 1965 |
+
"learning_rate": 9.677769786625869e-07,
|
| 1966 |
+
"logits/chosen": -2.8227927684783936,
|
| 1967 |
+
"logits/rejected": -2.853522539138794,
|
| 1968 |
+
"logps/chosen": -327.1927490234375,
|
| 1969 |
+
"logps/rejected": -470.41546630859375,
|
| 1970 |
+
"loss": 0.0002,
|
| 1971 |
+
"rewards/accuracies": 1.0,
|
| 1972 |
+
"rewards/chosen": -0.44436079263687134,
|
| 1973 |
+
"rewards/margins": 11.060840606689453,
|
| 1974 |
+
"rewards/rejected": -11.505200386047363,
|
| 1975 |
+
"step": 1310
|
| 1976 |
+
},
|
| 1977 |
+
{
|
| 1978 |
+
"epoch": 33.0,
|
| 1979 |
+
"grad_norm": 0.001769970403984189,
|
| 1980 |
+
"learning_rate": 9.042397785550405e-07,
|
| 1981 |
+
"logits/chosen": -2.719733715057373,
|
| 1982 |
+
"logits/rejected": -2.772249460220337,
|
| 1983 |
+
"logps/chosen": -299.26641845703125,
|
| 1984 |
+
"logps/rejected": -427.08612060546875,
|
| 1985 |
+
"loss": 0.0001,
|
| 1986 |
+
"rewards/accuracies": 1.0,
|
| 1987 |
+
"rewards/chosen": -0.377460777759552,
|
| 1988 |
+
"rewards/margins": 13.097183227539062,
|
| 1989 |
+
"rewards/rejected": -13.474644660949707,
|
| 1990 |
+
"step": 1320
|
| 1991 |
+
},
|
| 1992 |
+
{
|
| 1993 |
+
"epoch": 33.25,
|
| 1994 |
+
"grad_norm": 0.017327116802334785,
|
| 1995 |
+
"learning_rate": 8.426519384872733e-07,
|
| 1996 |
+
"logits/chosen": -2.7387938499450684,
|
| 1997 |
+
"logits/rejected": -2.7465500831604004,
|
| 1998 |
+
"logps/chosen": -316.33941650390625,
|
| 1999 |
+
"logps/rejected": -463.70294189453125,
|
| 2000 |
+
"loss": 0.0002,
|
| 2001 |
+
"rewards/accuracies": 1.0,
|
| 2002 |
+
"rewards/chosen": -0.29718923568725586,
|
| 2003 |
+
"rewards/margins": 12.517646789550781,
|
| 2004 |
+
"rewards/rejected": -12.814834594726562,
|
| 2005 |
+
"step": 1330
|
| 2006 |
+
},
|
| 2007 |
+
{
|
| 2008 |
+
"epoch": 33.5,
|
| 2009 |
+
"grad_norm": 0.015168960206210613,
|
| 2010 |
+
"learning_rate": 7.830427709355726e-07,
|
| 2011 |
+
"logits/chosen": -2.7458043098449707,
|
| 2012 |
+
"logits/rejected": -2.7986690998077393,
|
| 2013 |
+
"logps/chosen": -282.633056640625,
|
| 2014 |
+
"logps/rejected": -415.61865234375,
|
| 2015 |
+
"loss": 0.0,
|
| 2016 |
+
"rewards/accuracies": 1.0,
|
| 2017 |
+
"rewards/chosen": 0.6882933378219604,
|
| 2018 |
+
"rewards/margins": 12.167409896850586,
|
| 2019 |
+
"rewards/rejected": -11.479117393493652,
|
| 2020 |
+
"step": 1340
|
| 2021 |
+
},
|
| 2022 |
+
{
|
| 2023 |
+
"epoch": 33.75,
|
| 2024 |
+
"grad_norm": 0.0029403967782855034,
|
| 2025 |
+
"learning_rate": 7.254406466352682e-07,
|
| 2026 |
+
"logits/chosen": -2.751441240310669,
|
| 2027 |
+
"logits/rejected": -2.7536258697509766,
|
| 2028 |
+
"logps/chosen": -271.40545654296875,
|
| 2029 |
+
"logps/rejected": -417.96044921875,
|
| 2030 |
+
"loss": 0.0,
|
| 2031 |
+
"rewards/accuracies": 1.0,
|
| 2032 |
+
"rewards/chosen": -0.024983251467347145,
|
| 2033 |
+
"rewards/margins": 12.195524215698242,
|
| 2034 |
+
"rewards/rejected": -12.220507621765137,
|
| 2035 |
+
"step": 1350
|
| 2036 |
+
},
|
| 2037 |
+
{
|
| 2038 |
+
"epoch": 34.0,
|
| 2039 |
+
"grad_norm": 0.008198041468858719,
|
| 2040 |
+
"learning_rate": 6.698729810778065e-07,
|
| 2041 |
+
"logits/chosen": -2.764251470565796,
|
| 2042 |
+
"logits/rejected": -2.802863836288452,
|
| 2043 |
+
"logps/chosen": -311.47430419921875,
|
| 2044 |
+
"logps/rejected": -441.355224609375,
|
| 2045 |
+
"loss": 0.0,
|
| 2046 |
+
"rewards/accuracies": 1.0,
|
| 2047 |
+
"rewards/chosen": -0.037785936146974564,
|
| 2048 |
+
"rewards/margins": 12.428993225097656,
|
| 2049 |
+
"rewards/rejected": -12.466779708862305,
|
| 2050 |
+
"step": 1360
|
| 2051 |
+
},
|
| 2052 |
+
{
|
| 2053 |
+
"epoch": 34.25,
|
| 2054 |
+
"grad_norm": 0.01094425655901432,
|
| 2055 |
+
"learning_rate": 6.163662214624616e-07,
|
| 2056 |
+
"logits/chosen": -2.7676146030426025,
|
| 2057 |
+
"logits/rejected": -2.7861106395721436,
|
| 2058 |
+
"logps/chosen": -317.7796936035156,
|
| 2059 |
+
"logps/rejected": -462.99884033203125,
|
| 2060 |
+
"loss": 0.0,
|
| 2061 |
+
"rewards/accuracies": 1.0,
|
| 2062 |
+
"rewards/chosen": -0.8035726547241211,
|
| 2063 |
+
"rewards/margins": 13.26586627960205,
|
| 2064 |
+
"rewards/rejected": -14.069439888000488,
|
| 2065 |
+
"step": 1370
|
| 2066 |
+
},
|
| 2067 |
+
{
|
| 2068 |
+
"epoch": 34.5,
|
| 2069 |
+
"grad_norm": 0.014173522591590881,
|
| 2070 |
+
"learning_rate": 5.649458341088915e-07,
|
| 2071 |
+
"logits/chosen": -2.7046866416931152,
|
| 2072 |
+
"logits/rejected": -2.780492067337036,
|
| 2073 |
+
"logps/chosen": -266.5248718261719,
|
| 2074 |
+
"logps/rejected": -408.0611267089844,
|
| 2075 |
+
"loss": 0.0,
|
| 2076 |
+
"rewards/accuracies": 1.0,
|
| 2077 |
+
"rewards/chosen": 0.9178122282028198,
|
| 2078 |
+
"rewards/margins": 11.700557708740234,
|
| 2079 |
+
"rewards/rejected": -10.782745361328125,
|
| 2080 |
+
"step": 1380
|
| 2081 |
+
},
|
| 2082 |
+
{
|
| 2083 |
+
"epoch": 34.75,
|
| 2084 |
+
"grad_norm": 0.005380965769290924,
|
| 2085 |
+
"learning_rate": 5.156362923365587e-07,
|
| 2086 |
+
"logits/chosen": -2.804652214050293,
|
| 2087 |
+
"logits/rejected": -2.8121368885040283,
|
| 2088 |
+
"logps/chosen": -317.160888671875,
|
| 2089 |
+
"logps/rejected": -424.66339111328125,
|
| 2090 |
+
"loss": 0.0,
|
| 2091 |
+
"rewards/accuracies": 1.0,
|
| 2092 |
+
"rewards/chosen": -0.029048467054963112,
|
| 2093 |
+
"rewards/margins": 12.284951210021973,
|
| 2094 |
+
"rewards/rejected": -12.314001083374023,
|
| 2095 |
+
"step": 1390
|
| 2096 |
+
},
|
| 2097 |
+
{
|
| 2098 |
+
"epoch": 35.0,
|
| 2099 |
+
"grad_norm": 0.0023389519192278385,
|
| 2100 |
+
"learning_rate": 4.6846106481675035e-07,
|
| 2101 |
+
"logits/chosen": -2.6911251544952393,
|
| 2102 |
+
"logits/rejected": -2.7417197227478027,
|
| 2103 |
+
"logps/chosen": -279.67474365234375,
|
| 2104 |
+
"logps/rejected": -443.27264404296875,
|
| 2105 |
+
"loss": 0.0001,
|
| 2106 |
+
"rewards/accuracies": 1.0,
|
| 2107 |
+
"rewards/chosen": 0.21463599801063538,
|
| 2108 |
+
"rewards/margins": 12.136545181274414,
|
| 2109 |
+
"rewards/rejected": -11.921908378601074,
|
| 2110 |
+
"step": 1400
|
| 2111 |
+
},
|
| 2112 |
+
{
|
| 2113 |
+
"epoch": 35.25,
|
| 2114 |
+
"grad_norm": 0.017130231484770775,
|
| 2115 |
+
"learning_rate": 4.2344260440276455e-07,
|
| 2116 |
+
"logits/chosen": -2.705173969268799,
|
| 2117 |
+
"logits/rejected": -2.755000591278076,
|
| 2118 |
+
"logps/chosen": -279.12811279296875,
|
| 2119 |
+
"logps/rejected": -430.97509765625,
|
| 2120 |
+
"loss": 0.0,
|
| 2121 |
+
"rewards/accuracies": 1.0,
|
| 2122 |
+
"rewards/chosen": 0.5486839413642883,
|
| 2123 |
+
"rewards/margins": 12.646631240844727,
|
| 2124 |
+
"rewards/rejected": -12.09794807434082,
|
| 2125 |
+
"step": 1410
|
| 2126 |
+
},
|
| 2127 |
+
{
|
| 2128 |
+
"epoch": 35.5,
|
| 2129 |
+
"grad_norm": 0.007070457097142935,
|
| 2130 |
+
"learning_rate": 3.8060233744356634e-07,
|
| 2131 |
+
"logits/chosen": -2.706882953643799,
|
| 2132 |
+
"logits/rejected": -2.758007287979126,
|
| 2133 |
+
"logps/chosen": -290.89569091796875,
|
| 2134 |
+
"logps/rejected": -439.9056091308594,
|
| 2135 |
+
"loss": 0.0,
|
| 2136 |
+
"rewards/accuracies": 1.0,
|
| 2137 |
+
"rewards/chosen": 0.09179888665676117,
|
| 2138 |
+
"rewards/margins": 11.889338493347168,
|
| 2139 |
+
"rewards/rejected": -11.797539710998535,
|
| 2140 |
+
"step": 1420
|
| 2141 |
+
},
|
| 2142 |
+
{
|
| 2143 |
+
"epoch": 35.75,
|
| 2144 |
+
"grad_norm": 0.0010205187136307359,
|
| 2145 |
+
"learning_rate": 3.399606535860078e-07,
|
| 2146 |
+
"logits/chosen": -2.761723279953003,
|
| 2147 |
+
"logits/rejected": -2.7938809394836426,
|
| 2148 |
+
"logps/chosen": -297.20263671875,
|
| 2149 |
+
"logps/rejected": -408.91900634765625,
|
| 2150 |
+
"loss": 0.0,
|
| 2151 |
+
"rewards/accuracies": 1.0,
|
| 2152 |
+
"rewards/chosen": -0.1678989827632904,
|
| 2153 |
+
"rewards/margins": 12.045710563659668,
|
| 2154 |
+
"rewards/rejected": -12.21360969543457,
|
| 2155 |
+
"step": 1430
|
| 2156 |
+
},
|
| 2157 |
+
{
|
| 2158 |
+
"epoch": 36.0,
|
| 2159 |
+
"grad_norm": 0.0035463483072817326,
|
| 2160 |
+
"learning_rate": 3.015368960704584e-07,
|
| 2161 |
+
"logits/chosen": -2.7846503257751465,
|
| 2162 |
+
"logits/rejected": -2.818844795227051,
|
| 2163 |
+
"logps/chosen": -316.3269958496094,
|
| 2164 |
+
"logps/rejected": -458.6512145996094,
|
| 2165 |
+
"loss": 0.0003,
|
| 2166 |
+
"rewards/accuracies": 1.0,
|
| 2167 |
+
"rewards/chosen": -0.409940242767334,
|
| 2168 |
+
"rewards/margins": 12.515970230102539,
|
| 2169 |
+
"rewards/rejected": -12.925909042358398,
|
| 2170 |
+
"step": 1440
|
| 2171 |
+
},
|
| 2172 |
+
{
|
| 2173 |
+
"epoch": 36.25,
|
| 2174 |
+
"grad_norm": 0.002209126716479659,
|
| 2175 |
+
"learning_rate": 2.653493525244721e-07,
|
| 2176 |
+
"logits/chosen": -2.8237838745117188,
|
| 2177 |
+
"logits/rejected": -2.8457577228546143,
|
| 2178 |
+
"logps/chosen": -266.9129333496094,
|
| 2179 |
+
"logps/rejected": -445.2333984375,
|
| 2180 |
+
"loss": 0.0,
|
| 2181 |
+
"rewards/accuracies": 1.0,
|
| 2182 |
+
"rewards/chosen": 0.5144069790840149,
|
| 2183 |
+
"rewards/margins": 12.918063163757324,
|
| 2184 |
+
"rewards/rejected": -12.403654098510742,
|
| 2185 |
+
"step": 1450
|
| 2186 |
+
},
|
| 2187 |
+
{
|
| 2188 |
+
"epoch": 36.5,
|
| 2189 |
+
"grad_norm": 0.006079660728573799,
|
| 2190 |
+
"learning_rate": 2.314152462588659e-07,
|
| 2191 |
+
"logits/chosen": -2.733881711959839,
|
| 2192 |
+
"logits/rejected": -2.7485454082489014,
|
| 2193 |
+
"logps/chosen": -292.0540466308594,
|
| 2194 |
+
"logps/rejected": -413.36590576171875,
|
| 2195 |
+
"loss": 0.0191,
|
| 2196 |
+
"rewards/accuracies": 0.9750000238418579,
|
| 2197 |
+
"rewards/chosen": 0.3418157994747162,
|
| 2198 |
+
"rewards/margins": 11.966304779052734,
|
| 2199 |
+
"rewards/rejected": -11.624489784240723,
|
| 2200 |
+
"step": 1460
|
| 2201 |
+
},
|
| 2202 |
+
{
|
| 2203 |
+
"epoch": 36.75,
|
| 2204 |
+
"grad_norm": 0.004013635218143463,
|
| 2205 |
+
"learning_rate": 1.99750728070357e-07,
|
| 2206 |
+
"logits/chosen": -2.6816630363464355,
|
| 2207 |
+
"logits/rejected": -2.748070001602173,
|
| 2208 |
+
"logps/chosen": -331.69964599609375,
|
| 2209 |
+
"logps/rejected": -471.43798828125,
|
| 2210 |
+
"loss": 0.0,
|
| 2211 |
+
"rewards/accuracies": 1.0,
|
| 2212 |
+
"rewards/chosen": -0.7932103872299194,
|
| 2213 |
+
"rewards/margins": 11.995916366577148,
|
| 2214 |
+
"rewards/rejected": -12.789125442504883,
|
| 2215 |
+
"step": 1470
|
| 2216 |
+
},
|
| 2217 |
+
{
|
| 2218 |
+
"epoch": 37.0,
|
| 2219 |
+
"grad_norm": 0.010304308496415615,
|
| 2220 |
+
"learning_rate": 1.7037086855465902e-07,
|
| 2221 |
+
"logits/chosen": -2.7218892574310303,
|
| 2222 |
+
"logits/rejected": -2.7555019855499268,
|
| 2223 |
+
"logps/chosen": -291.52740478515625,
|
| 2224 |
+
"logps/rejected": -406.4840087890625,
|
| 2225 |
+
"loss": 0.0,
|
| 2226 |
+
"rewards/accuracies": 1.0,
|
| 2227 |
+
"rewards/chosen": 0.12983234226703644,
|
| 2228 |
+
"rewards/margins": 12.319366455078125,
|
| 2229 |
+
"rewards/rejected": -12.189533233642578,
|
| 2230 |
+
"step": 1480
|
| 2231 |
+
},
|
| 2232 |
+
{
|
| 2233 |
+
"epoch": 37.25,
|
| 2234 |
+
"grad_norm": 0.0003127313684672117,
|
| 2235 |
+
"learning_rate": 1.4328965093369284e-07,
|
| 2236 |
+
"logits/chosen": -2.6925010681152344,
|
| 2237 |
+
"logits/rejected": -2.7279093265533447,
|
| 2238 |
+
"logps/chosen": -284.3682556152344,
|
| 2239 |
+
"logps/rejected": -427.28729248046875,
|
| 2240 |
+
"loss": 0.0,
|
| 2241 |
+
"rewards/accuracies": 1.0,
|
| 2242 |
+
"rewards/chosen": 0.09465457499027252,
|
| 2243 |
+
"rewards/margins": 12.547832489013672,
|
| 2244 |
+
"rewards/rejected": -12.453178405761719,
|
| 2245 |
+
"step": 1490
|
| 2246 |
+
},
|
| 2247 |
+
{
|
| 2248 |
+
"epoch": 37.5,
|
| 2249 |
+
"grad_norm": 0.007795688696205616,
|
| 2250 |
+
"learning_rate": 1.185199644003332e-07,
|
| 2251 |
+
"logits/chosen": -2.79750657081604,
|
| 2252 |
+
"logits/rejected": -2.8254566192626953,
|
| 2253 |
+
"logps/chosen": -292.19561767578125,
|
| 2254 |
+
"logps/rejected": -425.38128662109375,
|
| 2255 |
+
"loss": 0.0,
|
| 2256 |
+
"rewards/accuracies": 1.0,
|
| 2257 |
+
"rewards/chosen": 0.0007894396549090743,
|
| 2258 |
+
"rewards/margins": 12.570943832397461,
|
| 2259 |
+
"rewards/rejected": -12.570155143737793,
|
| 2260 |
+
"step": 1500
|
| 2261 |
+
},
|
| 2262 |
+
{
|
| 2263 |
+
"epoch": 37.75,
|
| 2264 |
+
"grad_norm": 0.004515082575380802,
|
| 2265 |
+
"learning_rate": 9.607359798384785e-08,
|
| 2266 |
+
"logits/chosen": -2.8021745681762695,
|
| 2267 |
+
"logits/rejected": -2.8447844982147217,
|
| 2268 |
+
"logps/chosen": -306.9266357421875,
|
| 2269 |
+
"logps/rejected": -404.08831787109375,
|
| 2270 |
+
"loss": 0.0,
|
| 2271 |
+
"rewards/accuracies": 1.0,
|
| 2272 |
+
"rewards/chosen": -0.2106831967830658,
|
| 2273 |
+
"rewards/margins": 12.148092269897461,
|
| 2274 |
+
"rewards/rejected": -12.358774185180664,
|
| 2275 |
+
"step": 1510
|
| 2276 |
+
},
|
| 2277 |
+
{
|
| 2278 |
+
"epoch": 38.0,
|
| 2279 |
+
"grad_norm": 0.2741380035877228,
|
| 2280 |
+
"learning_rate": 7.59612349389599e-08,
|
| 2281 |
+
"logits/chosen": -2.702080488204956,
|
| 2282 |
+
"logits/rejected": -2.7362682819366455,
|
| 2283 |
+
"logps/chosen": -303.89068603515625,
|
| 2284 |
+
"logps/rejected": -479.22119140625,
|
| 2285 |
+
"loss": 0.0012,
|
| 2286 |
+
"rewards/accuracies": 1.0,
|
| 2287 |
+
"rewards/chosen": 0.2993970811367035,
|
| 2288 |
+
"rewards/margins": 11.771306991577148,
|
| 2289 |
+
"rewards/rejected": -11.47191047668457,
|
| 2290 |
+
"step": 1520
|
| 2291 |
+
},
|
| 2292 |
+
{
|
| 2293 |
+
"epoch": 38.25,
|
| 2294 |
+
"grad_norm": 0.005564592778682709,
|
| 2295 |
+
"learning_rate": 5.8192447661196694e-08,
|
| 2296 |
+
"logits/chosen": -2.7832558155059814,
|
| 2297 |
+
"logits/rejected": -2.825497627258301,
|
| 2298 |
+
"logps/chosen": -267.9512939453125,
|
| 2299 |
+
"logps/rejected": -420.0101623535156,
|
| 2300 |
+
"loss": 0.0046,
|
| 2301 |
+
"rewards/accuracies": 1.0,
|
| 2302 |
+
"rewards/chosen": 0.6914939284324646,
|
| 2303 |
+
"rewards/margins": 11.328948020935059,
|
| 2304 |
+
"rewards/rejected": -10.63745403289795,
|
| 2305 |
+
"step": 1530
|
| 2306 |
+
},
|
| 2307 |
+
{
|
| 2308 |
+
"epoch": 38.5,
|
| 2309 |
+
"grad_norm": 0.006241649389266968,
|
| 2310 |
+
"learning_rate": 4.2775693130948094e-08,
|
| 2311 |
+
"logits/chosen": -2.7838566303253174,
|
| 2312 |
+
"logits/rejected": -2.7983555793762207,
|
| 2313 |
+
"logps/chosen": -312.9120788574219,
|
| 2314 |
+
"logps/rejected": -482.3827209472656,
|
| 2315 |
+
"loss": 0.0,
|
| 2316 |
+
"rewards/accuracies": 1.0,
|
| 2317 |
+
"rewards/chosen": -0.37507203221321106,
|
| 2318 |
+
"rewards/margins": 13.01745319366455,
|
| 2319 |
+
"rewards/rejected": -13.392524719238281,
|
| 2320 |
+
"step": 1540
|
| 2321 |
+
},
|
| 2322 |
+
{
|
| 2323 |
+
"epoch": 38.75,
|
| 2324 |
+
"grad_norm": 0.0031968543771654367,
|
| 2325 |
+
"learning_rate": 2.971830888840177e-08,
|
| 2326 |
+
"logits/chosen": -2.751046657562256,
|
| 2327 |
+
"logits/rejected": -2.774620532989502,
|
| 2328 |
+
"logps/chosen": -289.15106201171875,
|
| 2329 |
+
"logps/rejected": -391.25018310546875,
|
| 2330 |
+
"loss": 0.0,
|
| 2331 |
+
"rewards/accuracies": 1.0,
|
| 2332 |
+
"rewards/chosen": -0.36087581515312195,
|
| 2333 |
+
"rewards/margins": 12.353649139404297,
|
| 2334 |
+
"rewards/rejected": -12.71452522277832,
|
| 2335 |
+
"step": 1550
|
| 2336 |
+
},
|
| 2337 |
+
{
|
| 2338 |
+
"epoch": 39.0,
|
| 2339 |
+
"grad_norm": 0.0021725529804825783,
|
| 2340 |
+
"learning_rate": 1.9026509541272276e-08,
|
| 2341 |
+
"logits/chosen": -2.6909258365631104,
|
| 2342 |
+
"logits/rejected": -2.746695041656494,
|
| 2343 |
+
"logps/chosen": -310.7889709472656,
|
| 2344 |
+
"logps/rejected": -442.99005126953125,
|
| 2345 |
+
"loss": 0.0,
|
| 2346 |
+
"rewards/accuracies": 1.0,
|
| 2347 |
+
"rewards/chosen": 0.3208140730857849,
|
| 2348 |
+
"rewards/margins": 12.539921760559082,
|
| 2349 |
+
"rewards/rejected": -12.219108581542969,
|
| 2350 |
+
"step": 1560
|
| 2351 |
+
},
|
| 2352 |
+
{
|
| 2353 |
+
"epoch": 39.25,
|
| 2354 |
+
"grad_norm": 0.005507487803697586,
|
| 2355 |
+
"learning_rate": 1.0705383806982606e-08,
|
| 2356 |
+
"logits/chosen": -2.7035202980041504,
|
| 2357 |
+
"logits/rejected": -2.7346391677856445,
|
| 2358 |
+
"logps/chosen": -311.97955322265625,
|
| 2359 |
+
"logps/rejected": -452.98773193359375,
|
| 2360 |
+
"loss": 0.0,
|
| 2361 |
+
"rewards/accuracies": 1.0,
|
| 2362 |
+
"rewards/chosen": -0.07614202797412872,
|
| 2363 |
+
"rewards/margins": 12.285137176513672,
|
| 2364 |
+
"rewards/rejected": -12.361278533935547,
|
| 2365 |
+
"step": 1570
|
| 2366 |
+
},
|
| 2367 |
+
{
|
| 2368 |
+
"epoch": 39.5,
|
| 2369 |
+
"grad_norm": 0.007295460440218449,
|
| 2370 |
+
"learning_rate": 4.758892090711009e-09,
|
| 2371 |
+
"logits/chosen": -2.7759146690368652,
|
| 2372 |
+
"logits/rejected": -2.8210902214050293,
|
| 2373 |
+
"logps/chosen": -297.54779052734375,
|
| 2374 |
+
"logps/rejected": -436.24365234375,
|
| 2375 |
+
"loss": 0.0018,
|
| 2376 |
+
"rewards/accuracies": 1.0,
|
| 2377 |
+
"rewards/chosen": -0.14304715394973755,
|
| 2378 |
+
"rewards/margins": 11.92789077758789,
|
| 2379 |
+
"rewards/rejected": -12.070939064025879,
|
| 2380 |
+
"step": 1580
|
| 2381 |
+
},
|
| 2382 |
+
{
|
| 2383 |
+
"epoch": 39.75,
|
| 2384 |
+
"grad_norm": 0.003202753607183695,
|
| 2385 |
+
"learning_rate": 1.189864600454338e-09,
|
| 2386 |
+
"logits/chosen": -2.77467679977417,
|
| 2387 |
+
"logits/rejected": -2.812164783477783,
|
| 2388 |
+
"logps/chosen": -293.31451416015625,
|
| 2389 |
+
"logps/rejected": -415.6048889160156,
|
| 2390 |
+
"loss": 0.0,
|
| 2391 |
+
"rewards/accuracies": 1.0,
|
| 2392 |
+
"rewards/chosen": -0.15772318840026855,
|
| 2393 |
+
"rewards/margins": 12.923543930053711,
|
| 2394 |
+
"rewards/rejected": -13.081266403198242,
|
| 2395 |
+
"step": 1590
|
| 2396 |
+
},
|
| 2397 |
+
{
|
| 2398 |
+
"epoch": 40.0,
|
| 2399 |
+
"grad_norm": 0.004265669733285904,
|
| 2400 |
+
"learning_rate": 0.0,
|
| 2401 |
+
"logits/chosen": -2.672964572906494,
|
| 2402 |
+
"logits/rejected": -2.7406535148620605,
|
| 2403 |
+
"logps/chosen": -279.08856201171875,
|
| 2404 |
+
"logps/rejected": -434.34405517578125,
|
| 2405 |
+
"loss": 0.0,
|
| 2406 |
+
"rewards/accuracies": 1.0,
|
| 2407 |
+
"rewards/chosen": 0.5602867603302002,
|
| 2408 |
+
"rewards/margins": 12.123626708984375,
|
| 2409 |
+
"rewards/rejected": -11.56334114074707,
|
| 2410 |
+
"step": 1600
|
| 2411 |
+
}
|
| 2412 |
+
],
|
| 2413 |
+
"logging_steps": 10,
|
| 2414 |
+
"max_steps": 1600,
|
| 2415 |
+
"num_input_tokens_seen": 0,
|
| 2416 |
+
"num_train_epochs": 40,
|
| 2417 |
+
"save_steps": 500,
|
| 2418 |
+
"stateful_callbacks": {
|
| 2419 |
+
"TrainerControl": {
|
| 2420 |
+
"args": {
|
| 2421 |
+
"should_epoch_stop": false,
|
| 2422 |
+
"should_evaluate": false,
|
| 2423 |
+
"should_log": false,
|
| 2424 |
+
"should_save": true,
|
| 2425 |
+
"should_training_stop": true
|
| 2426 |
+
},
|
| 2427 |
+
"attributes": {}
|
| 2428 |
+
}
|
| 2429 |
+
},
|
| 2430 |
+
"total_flos": 0.0,
|
| 2431 |
+
"train_batch_size": 4,
|
| 2432 |
+
"trial_name": null,
|
| 2433 |
+
"trial_params": null
|
| 2434 |
+
}
|
checkpoint-1600/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:939e2927d0c1eb1a5271851f6d9052ddfadc6757223e2e3e0263bde749dba65c
|
| 3 |
+
size 6200
|