End of training
Browse files- README.md +5 -5
- all_results.json +5 -5
- chat_template.jinja +54 -0
- config.json +1 -1
- generation_config.json +1 -1
- model.safetensors +1 -1
- tokenizer_config.json +0 -1
- train_results.json +5 -5
- trainer_state.json +115 -97
- training_args.bin +2 -2
README.md
CHANGED
|
@@ -37,10 +37,10 @@ This model was trained with GRPO, a method introduced in [DeepSeekMath: Pushing
|
|
| 37 |
|
| 38 |
### Framework versions
|
| 39 |
|
| 40 |
-
- TRL: 0.
|
| 41 |
-
- Transformers: 4.
|
| 42 |
-
- Pytorch: 2.
|
| 43 |
-
- Datasets: 3.
|
| 44 |
- Tokenizers: 0.21.1
|
| 45 |
|
| 46 |
## Citations
|
|
@@ -62,7 +62,7 @@ Cite TRL as:
|
|
| 62 |
```bibtex
|
| 63 |
@misc{vonwerra2022trl,
|
| 64 |
title = {{TRL: Transformer Reinforcement Learning}},
|
| 65 |
-
author = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin
|
| 66 |
year = 2020,
|
| 67 |
journal = {GitHub repository},
|
| 68 |
publisher = {GitHub},
|
|
|
|
| 37 |
|
| 38 |
### Framework versions
|
| 39 |
|
| 40 |
+
- TRL: 0.18.1
|
| 41 |
+
- Transformers: 4.52.4
|
| 42 |
+
- Pytorch: 2.7.0
|
| 43 |
+
- Datasets: 3.6.0
|
| 44 |
- Tokenizers: 0.21.1
|
| 45 |
|
| 46 |
## Citations
|
|
|
|
| 62 |
```bibtex
|
| 63 |
@misc{vonwerra2022trl,
|
| 64 |
title = {{TRL: Transformer Reinforcement Learning}},
|
| 65 |
+
author = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin Gallou{\'e}dec},
|
| 66 |
year = 2020,
|
| 67 |
journal = {GitHub repository},
|
| 68 |
publisher = {GitHub},
|
all_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
"total_flos": 0.0,
|
| 3 |
-
"train_loss": 0.
|
| 4 |
-
"train_runtime":
|
| 5 |
-
"train_samples":
|
| 6 |
-
"train_samples_per_second": 0.
|
| 7 |
-
"train_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"total_flos": 0.0,
|
| 3 |
+
"train_loss": 0.06442624032497406,
|
| 4 |
+
"train_runtime": 4300.2944,
|
| 5 |
+
"train_samples": 3,
|
| 6 |
+
"train_samples_per_second": 0.009,
|
| 7 |
+
"train_steps_per_second": 0.002
|
| 8 |
}
|
chat_template.jinja
ADDED
|
@@ -0,0 +1,54 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{%- if tools %}
|
| 2 |
+
{{- '<|im_start|>system\n' }}
|
| 3 |
+
{%- if messages[0]['role'] == 'system' %}
|
| 4 |
+
{{- messages[0]['content'] }}
|
| 5 |
+
{%- else %}
|
| 6 |
+
{{- 'You are Qwen, created by Alibaba Cloud. You are a helpful assistant.' }}
|
| 7 |
+
{%- endif %}
|
| 8 |
+
{{- "\n\n# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
|
| 9 |
+
{%- for tool in tools %}
|
| 10 |
+
{{- "\n" }}
|
| 11 |
+
{{- tool | tojson }}
|
| 12 |
+
{%- endfor %}
|
| 13 |
+
{{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
|
| 14 |
+
{%- else %}
|
| 15 |
+
{%- if messages[0]['role'] == 'system' %}
|
| 16 |
+
{{- '<|im_start|>system\n' + messages[0]['content'] + '<|im_end|>\n' }}
|
| 17 |
+
{%- else %}
|
| 18 |
+
{{- '<|im_start|>system\nYou are Qwen, created by Alibaba Cloud. You are a helpful assistant.<|im_end|>\n' }}
|
| 19 |
+
{%- endif %}
|
| 20 |
+
{%- endif %}
|
| 21 |
+
{%- for message in messages %}
|
| 22 |
+
{%- if (message.role == "user") or (message.role == "system" and not loop.first) or (message.role == "assistant" and not message.tool_calls) %}
|
| 23 |
+
{{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }}
|
| 24 |
+
{%- elif message.role == "assistant" %}
|
| 25 |
+
{{- '<|im_start|>' + message.role }}
|
| 26 |
+
{%- if message.content %}
|
| 27 |
+
{{- '\n' + message.content }}
|
| 28 |
+
{%- endif %}
|
| 29 |
+
{%- for tool_call in message.tool_calls %}
|
| 30 |
+
{%- if tool_call.function is defined %}
|
| 31 |
+
{%- set tool_call = tool_call.function %}
|
| 32 |
+
{%- endif %}
|
| 33 |
+
{{- '\n<tool_call>\n{"name": "' }}
|
| 34 |
+
{{- tool_call.name }}
|
| 35 |
+
{{- '", "arguments": ' }}
|
| 36 |
+
{{- tool_call.arguments | tojson }}
|
| 37 |
+
{{- '}\n</tool_call>' }}
|
| 38 |
+
{%- endfor %}
|
| 39 |
+
{{- '<|im_end|>\n' }}
|
| 40 |
+
{%- elif message.role == "tool" %}
|
| 41 |
+
{%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != "tool") %}
|
| 42 |
+
{{- '<|im_start|>user' }}
|
| 43 |
+
{%- endif %}
|
| 44 |
+
{{- '\n<tool_response>\n' }}
|
| 45 |
+
{{- message.content }}
|
| 46 |
+
{{- '\n</tool_response>' }}
|
| 47 |
+
{%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
|
| 48 |
+
{{- '<|im_end|>\n' }}
|
| 49 |
+
{%- endif %}
|
| 50 |
+
{%- endif %}
|
| 51 |
+
{%- endfor %}
|
| 52 |
+
{%- if add_generation_prompt %}
|
| 53 |
+
{{- '<|im_start|>assistant\n' }}
|
| 54 |
+
{%- endif %}
|
config.json
CHANGED
|
@@ -21,7 +21,7 @@
|
|
| 21 |
"sliding_window": null,
|
| 22 |
"tie_word_embeddings": true,
|
| 23 |
"torch_dtype": "float32",
|
| 24 |
-
"transformers_version": "4.
|
| 25 |
"unsloth_fixed": true,
|
| 26 |
"use_cache": true,
|
| 27 |
"use_sliding_window": false,
|
|
|
|
| 21 |
"sliding_window": null,
|
| 22 |
"tie_word_embeddings": true,
|
| 23 |
"torch_dtype": "float32",
|
| 24 |
+
"transformers_version": "4.52.4",
|
| 25 |
"unsloth_fixed": true,
|
| 26 |
"use_cache": true,
|
| 27 |
"use_sliding_window": false,
|
generation_config.json
CHANGED
|
@@ -11,5 +11,5 @@
|
|
| 11 |
"temperature": 0.7,
|
| 12 |
"top_k": 20,
|
| 13 |
"top_p": 0.8,
|
| 14 |
-
"transformers_version": "4.
|
| 15 |
}
|
|
|
|
| 11 |
"temperature": 0.7,
|
| 12 |
"top_k": 20,
|
| 13 |
"top_p": 0.8,
|
| 14 |
+
"transformers_version": "4.52.4"
|
| 15 |
}
|
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1976163472
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d15d32b58eef1825455f44b03d11d277609175a2c6d0e9191574931bd167407d
|
| 3 |
size 1976163472
|
tokenizer_config.json
CHANGED
|
@@ -195,7 +195,6 @@
|
|
| 195 |
"<|video_pad|>"
|
| 196 |
],
|
| 197 |
"bos_token": null,
|
| 198 |
-
"chat_template": "{%- if tools %}\n {{- '<|im_start|>system\\n' }}\n {%- if messages[0]['role'] == 'system' %}\n {{- messages[0]['content'] }}\n {%- else %}\n {{- 'You are Qwen, created by Alibaba Cloud. You are a helpful assistant.' }}\n {%- endif %}\n {{- \"\\n\\n# Tools\\n\\nYou may call one or more functions to assist with the user query.\\n\\nYou are provided with function signatures within <tools></tools> XML tags:\\n<tools>\" }}\n {%- for tool in tools %}\n {{- \"\\n\" }}\n {{- tool | tojson }}\n {%- endfor %}\n {{- \"\\n</tools>\\n\\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\\n<tool_call>\\n{\\\"name\\\": <function-name>, \\\"arguments\\\": <args-json-object>}\\n</tool_call><|im_end|>\\n\" }}\n{%- else %}\n {%- if messages[0]['role'] == 'system' %}\n {{- '<|im_start|>system\\n' + messages[0]['content'] + '<|im_end|>\\n' }}\n {%- else %}\n {{- '<|im_start|>system\\nYou are Qwen, created by Alibaba Cloud. You are a helpful assistant.<|im_end|>\\n' }}\n {%- endif %}\n{%- endif %}\n{%- for message in messages %}\n {%- if (message.role == \"user\") or (message.role == \"system\" and not loop.first) or (message.role == \"assistant\" and not message.tool_calls) %}\n {{- '<|im_start|>' + message.role + '\\n' + message.content + '<|im_end|>' + '\\n' }}\n {%- elif message.role == \"assistant\" %}\n {{- '<|im_start|>' + message.role }}\n {%- if message.content %}\n {{- '\\n' + message.content }}\n {%- endif %}\n {%- for tool_call in message.tool_calls %}\n {%- if tool_call.function is defined %}\n {%- set tool_call = tool_call.function %}\n {%- endif %}\n {{- '\\n<tool_call>\\n{\"name\": \"' }}\n {{- tool_call.name }}\n {{- '\", \"arguments\": ' }}\n {{- tool_call.arguments | tojson }}\n {{- '}\\n</tool_call>' }}\n {%- endfor %}\n {{- '<|im_end|>\\n' }}\n {%- elif message.role == \"tool\" %}\n {%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != \"tool\") %}\n {{- '<|im_start|>user' }}\n {%- endif %}\n {{- '\\n<tool_response>\\n' }}\n {{- message.content }}\n {{- '\\n</tool_response>' }}\n {%- if loop.last or (messages[loop.index0 + 1].role != \"tool\") %}\n {{- '<|im_end|>\\n' }}\n {%- endif %}\n {%- endif %}\n{%- endfor %}\n{%- if add_generation_prompt %}\n {{- '<|im_start|>assistant\\n' }}\n{%- endif %}\n",
|
| 199 |
"clean_up_tokenization_spaces": false,
|
| 200 |
"eos_token": "<|im_end|>",
|
| 201 |
"errors": "replace",
|
|
|
|
| 195 |
"<|video_pad|>"
|
| 196 |
],
|
| 197 |
"bos_token": null,
|
|
|
|
| 198 |
"clean_up_tokenization_spaces": false,
|
| 199 |
"eos_token": "<|im_end|>",
|
| 200 |
"errors": "replace",
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
"total_flos": 0.0,
|
| 3 |
-
"train_loss": 0.
|
| 4 |
-
"train_runtime":
|
| 5 |
-
"train_samples":
|
| 6 |
-
"train_samples_per_second": 0.
|
| 7 |
-
"train_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"total_flos": 0.0,
|
| 3 |
+
"train_loss": 0.06442624032497406,
|
| 4 |
+
"train_runtime": 4300.2944,
|
| 5 |
+
"train_samples": 3,
|
| 6 |
+
"train_samples_per_second": 0.009,
|
| 7 |
+
"train_steps_per_second": 0.002
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -2,121 +2,139 @@
|
|
| 2 |
"best_global_step": null,
|
| 3 |
"best_metric": null,
|
| 4 |
"best_model_checkpoint": null,
|
| 5 |
-
"epoch":
|
| 6 |
"eval_steps": 500,
|
| 7 |
-
"global_step":
|
| 8 |
"is_hyper_param_search": false,
|
| 9 |
"is_local_process_zero": true,
|
| 10 |
"is_world_process_zero": true,
|
| 11 |
"log_history": [
|
| 12 |
{
|
| 13 |
-
"
|
| 14 |
-
"
|
| 15 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 16 |
"kl": 0.0,
|
| 17 |
"learning_rate": 5e-07,
|
| 18 |
-
"loss": 0.
|
| 19 |
-
"
|
| 20 |
-
"
|
| 21 |
-
"
|
| 22 |
-
"rewards/
|
| 23 |
-
"rewards/
|
| 24 |
-
"rewards/
|
| 25 |
-
"rewards/
|
| 26 |
-
"rewards/
|
| 27 |
-
"rewards/
|
| 28 |
-
"rewards/
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 29 |
"step": 2
|
| 30 |
},
|
| 31 |
{
|
| 32 |
-
"
|
| 33 |
-
"
|
| 34 |
-
"
|
| 35 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 36 |
"learning_rate": 4.415111107797445e-07,
|
| 37 |
-
"loss": 0.
|
| 38 |
-
"
|
| 39 |
-
"
|
| 40 |
-
"
|
| 41 |
-
"rewards/
|
| 42 |
-
"rewards/
|
| 43 |
-
"rewards/
|
| 44 |
-
"rewards/
|
| 45 |
-
"rewards/
|
| 46 |
-
"rewards/
|
| 47 |
-
"rewards/
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 48 |
"step": 4
|
| 49 |
},
|
| 50 |
{
|
| 51 |
-
"
|
| 52 |
-
"
|
| 53 |
-
"
|
| 54 |
-
"
|
| 55 |
-
"
|
| 56 |
-
"
|
| 57 |
-
"
|
| 58 |
-
"
|
| 59 |
-
"
|
| 60 |
-
"
|
| 61 |
-
"
|
| 62 |
-
"
|
| 63 |
-
"
|
| 64 |
-
"
|
| 65 |
-
"
|
| 66 |
-
"
|
| 67 |
-
"
|
| 68 |
-
|
| 69 |
-
|
| 70 |
-
"
|
| 71 |
-
"
|
| 72 |
-
"
|
| 73 |
-
"
|
| 74 |
-
"
|
| 75 |
-
"
|
| 76 |
-
"
|
| 77 |
-
"
|
| 78 |
-
"rewards/
|
| 79 |
-
"rewards/
|
| 80 |
-
"rewards/
|
| 81 |
-
"rewards/
|
| 82 |
-
"rewards/
|
| 83 |
-
"rewards/
|
| 84 |
-
"rewards/
|
| 85 |
-
"
|
| 86 |
-
"step": 8
|
| 87 |
-
},
|
| 88 |
-
{
|
| 89 |
-
"completion_length": 4.0,
|
| 90 |
-
"epoch": 3.4,
|
| 91 |
-
"grad_norm": 0.0,
|
| 92 |
-
"kl": 0.0,
|
| 93 |
-
"learning_rate": 1.507684480352292e-08,
|
| 94 |
-
"loss": 0.0,
|
| 95 |
-
"reward": 3.9540088176727295,
|
| 96 |
-
"reward_std": 0.0,
|
| 97 |
-
"rewards/concensus_correctness_reward_func": 2.585333267847697,
|
| 98 |
-
"rewards/consensus_reward_func": 1.3333333333333333,
|
| 99 |
-
"rewards/cumulative_reward_2": 0.0,
|
| 100 |
-
"rewards/final_correctness_reward_func": 0.0,
|
| 101 |
-
"rewards/question_recreation_reward_func": 0.035342139502366386,
|
| 102 |
-
"rewards/soft_format_reward_func": 0.0,
|
| 103 |
-
"rewards/strict_format_reward_func": 0.0,
|
| 104 |
-
"rewards/xmlcount_reward_func": 0.0,
|
| 105 |
-
"step": 10
|
| 106 |
-
},
|
| 107 |
-
{
|
| 108 |
-
"epoch": 3.4,
|
| 109 |
-
"step": 10,
|
| 110 |
"total_flos": 0.0,
|
| 111 |
-
"train_loss": 0.
|
| 112 |
-
"train_runtime":
|
| 113 |
-
"train_samples_per_second": 0.
|
| 114 |
-
"train_steps_per_second": 0.
|
| 115 |
}
|
| 116 |
],
|
| 117 |
"logging_steps": 2,
|
| 118 |
"max_steps": 10,
|
| 119 |
-
"num_input_tokens_seen":
|
| 120 |
"num_train_epochs": 5,
|
| 121 |
"save_steps": 10,
|
| 122 |
"stateful_callbacks": {
|
|
@@ -125,8 +143,8 @@
|
|
| 125 |
"should_epoch_stop": false,
|
| 126 |
"should_evaluate": false,
|
| 127 |
"should_log": false,
|
| 128 |
-
"should_save":
|
| 129 |
-
"should_training_stop":
|
| 130 |
},
|
| 131 |
"attributes": {}
|
| 132 |
}
|
|
|
|
| 2 |
"best_global_step": null,
|
| 3 |
"best_metric": null,
|
| 4 |
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 4.666666666666667,
|
| 6 |
"eval_steps": 500,
|
| 7 |
+
"global_step": 5,
|
| 8 |
"is_hyper_param_search": false,
|
| 9 |
"is_local_process_zero": true,
|
| 10 |
"is_world_process_zero": true,
|
| 11 |
"log_history": [
|
| 12 |
{
|
| 13 |
+
"clip_ratio/high_max": 0.0,
|
| 14 |
+
"clip_ratio/high_mean": 0.0,
|
| 15 |
+
"clip_ratio/low_mean": 0.0,
|
| 16 |
+
"clip_ratio/low_min": 0.0,
|
| 17 |
+
"clip_ratio/region_mean": 0.0,
|
| 18 |
+
"completions/clipped_ratio": 0.5,
|
| 19 |
+
"completions/max_length": 256.0,
|
| 20 |
+
"completions/max_terminated_length": 182.5,
|
| 21 |
+
"completions/mean_length": 191.125,
|
| 22 |
+
"completions/mean_terminated_length": 126.25,
|
| 23 |
+
"completions/min_length": 70.0,
|
| 24 |
+
"completions/min_terminated_length": 70.0,
|
| 25 |
+
"epoch": 1.6666666666666665,
|
| 26 |
+
"frac_reward_zero_std": 0.0,
|
| 27 |
+
"grad_norm": 9.101638793945312,
|
| 28 |
"kl": 0.0,
|
| 29 |
"learning_rate": 5e-07,
|
| 30 |
+
"loss": 0.1492,
|
| 31 |
+
"num_tokens": 2553.0,
|
| 32 |
+
"reward": 0.03208251763135195,
|
| 33 |
+
"reward_std": 0.00460250279866159,
|
| 34 |
+
"rewards/concensus_correctness_reward_func/mean": 0.0,
|
| 35 |
+
"rewards/concensus_correctness_reward_func/std": 0.0,
|
| 36 |
+
"rewards/consensus_reward_func/mean": 0.0,
|
| 37 |
+
"rewards/consensus_reward_func/std": 0.0,
|
| 38 |
+
"rewards/cumulative_reward_2/mean": 0.0,
|
| 39 |
+
"rewards/cumulative_reward_2/std": 0.0,
|
| 40 |
+
"rewards/final_correctness_reward_func/mean": 0.0,
|
| 41 |
+
"rewards/final_correctness_reward_func/std": 0.0,
|
| 42 |
+
"rewards/question_recreation_reward_func/mean": 0.0320825157687068,
|
| 43 |
+
"rewards/question_recreation_reward_func/std": 0.012270803097635508,
|
| 44 |
+
"rewards/soft_format_reward_func/mean": 0.0,
|
| 45 |
+
"rewards/soft_format_reward_func/std": 0.0,
|
| 46 |
+
"rewards/strict_format_reward_func/mean": 0.0,
|
| 47 |
+
"rewards/strict_format_reward_func/std": 0.0,
|
| 48 |
+
"rewards/xmlcount_reward_func/mean": 0.0,
|
| 49 |
+
"rewards/xmlcount_reward_func/std": 0.0,
|
| 50 |
"step": 2
|
| 51 |
},
|
| 52 |
{
|
| 53 |
+
"clip_ratio/high_max": 0.0,
|
| 54 |
+
"clip_ratio/high_mean": 0.0,
|
| 55 |
+
"clip_ratio/low_mean": 0.0,
|
| 56 |
+
"clip_ratio/low_min": 0.0,
|
| 57 |
+
"clip_ratio/region_mean": 0.0,
|
| 58 |
+
"completions/clipped_ratio": 0.625,
|
| 59 |
+
"completions/max_length": 256.0,
|
| 60 |
+
"completions/max_terminated_length": 166.5,
|
| 61 |
+
"completions/mean_length": 217.75,
|
| 62 |
+
"completions/mean_terminated_length": 158.75,
|
| 63 |
+
"completions/min_length": 151.0,
|
| 64 |
+
"completions/min_terminated_length": 151.0,
|
| 65 |
+
"epoch": 3.6666666666666665,
|
| 66 |
+
"frac_reward_zero_std": 0.0,
|
| 67 |
+
"grad_norm": 8.234423637390137,
|
| 68 |
+
"kl": 0.0014661487402918283,
|
| 69 |
"learning_rate": 4.415111107797445e-07,
|
| 70 |
+
"loss": 0.0119,
|
| 71 |
+
"num_tokens": 5319.0,
|
| 72 |
+
"reward": 0.032989093102514744,
|
| 73 |
+
"reward_std": 0.0026227867347188294,
|
| 74 |
+
"rewards/concensus_correctness_reward_func/mean": 0.0,
|
| 75 |
+
"rewards/concensus_correctness_reward_func/std": 0.0,
|
| 76 |
+
"rewards/consensus_reward_func/mean": 0.0,
|
| 77 |
+
"rewards/consensus_reward_func/std": 0.0,
|
| 78 |
+
"rewards/cumulative_reward_2/mean": 0.0,
|
| 79 |
+
"rewards/cumulative_reward_2/std": 0.0,
|
| 80 |
+
"rewards/final_correctness_reward_func/mean": 0.0,
|
| 81 |
+
"rewards/final_correctness_reward_func/std": 0.0,
|
| 82 |
+
"rewards/question_recreation_reward_func/mean": 0.032989093102514744,
|
| 83 |
+
"rewards/question_recreation_reward_func/std": 0.014430227922275662,
|
| 84 |
+
"rewards/soft_format_reward_func/mean": 0.0,
|
| 85 |
+
"rewards/soft_format_reward_func/std": 0.0,
|
| 86 |
+
"rewards/strict_format_reward_func/mean": 0.0,
|
| 87 |
+
"rewards/strict_format_reward_func/std": 0.0,
|
| 88 |
+
"rewards/xmlcount_reward_func/mean": 0.0,
|
| 89 |
+
"rewards/xmlcount_reward_func/std": 0.0,
|
| 90 |
"step": 4
|
| 91 |
},
|
| 92 |
{
|
| 93 |
+
"clip_ratio/high_max": 0.0,
|
| 94 |
+
"clip_ratio/high_mean": 0.0,
|
| 95 |
+
"clip_ratio/low_mean": 0.0,
|
| 96 |
+
"clip_ratio/low_min": 0.0,
|
| 97 |
+
"clip_ratio/region_mean": 0.0,
|
| 98 |
+
"completions/clipped_ratio": 0.5,
|
| 99 |
+
"completions/max_length": 256.0,
|
| 100 |
+
"completions/max_terminated_length": 4.0,
|
| 101 |
+
"completions/mean_length": 130.0,
|
| 102 |
+
"completions/mean_terminated_length": 4.0,
|
| 103 |
+
"completions/min_length": 4.0,
|
| 104 |
+
"completions/min_terminated_length": 4.0,
|
| 105 |
+
"epoch": 4.666666666666667,
|
| 106 |
+
"frac_reward_zero_std": 0.5,
|
| 107 |
+
"kl": 0.0002789091595332138,
|
| 108 |
+
"num_tokens": 6351.0,
|
| 109 |
+
"reward": 0.05653843283653259,
|
| 110 |
+
"reward_std": 3.0632854759460315e-05,
|
| 111 |
+
"rewards/concensus_correctness_reward_func/mean": 0.0,
|
| 112 |
+
"rewards/concensus_correctness_reward_func/std": 0.0,
|
| 113 |
+
"rewards/consensus_reward_func/mean": 0.0,
|
| 114 |
+
"rewards/consensus_reward_func/std": 0.0,
|
| 115 |
+
"rewards/cumulative_reward_2/mean": 0.0,
|
| 116 |
+
"rewards/cumulative_reward_2/std": 0.0,
|
| 117 |
+
"rewards/final_correctness_reward_func/mean": 0.0,
|
| 118 |
+
"rewards/final_correctness_reward_func/std": 0.0,
|
| 119 |
+
"rewards/question_recreation_reward_func/mean": 0.05653843283653259,
|
| 120 |
+
"rewards/question_recreation_reward_func/std": 0.03623591363430023,
|
| 121 |
+
"rewards/soft_format_reward_func/mean": 0.0,
|
| 122 |
+
"rewards/soft_format_reward_func/std": 0.0,
|
| 123 |
+
"rewards/strict_format_reward_func/mean": 0.0,
|
| 124 |
+
"rewards/strict_format_reward_func/std": 0.0,
|
| 125 |
+
"rewards/xmlcount_reward_func/mean": 0.0,
|
| 126 |
+
"rewards/xmlcount_reward_func/std": 0.0,
|
| 127 |
+
"step": 5,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 128 |
"total_flos": 0.0,
|
| 129 |
+
"train_loss": 0.06442624032497406,
|
| 130 |
+
"train_runtime": 4300.2944,
|
| 131 |
+
"train_samples_per_second": 0.009,
|
| 132 |
+
"train_steps_per_second": 0.002
|
| 133 |
}
|
| 134 |
],
|
| 135 |
"logging_steps": 2,
|
| 136 |
"max_steps": 10,
|
| 137 |
+
"num_input_tokens_seen": 6351,
|
| 138 |
"num_train_epochs": 5,
|
| 139 |
"save_steps": 10,
|
| 140 |
"stateful_callbacks": {
|
|
|
|
| 143 |
"should_epoch_stop": false,
|
| 144 |
"should_evaluate": false,
|
| 145 |
"should_log": false,
|
| 146 |
+
"should_save": false,
|
| 147 |
+
"should_training_stop": false
|
| 148 |
},
|
| 149 |
"attributes": {}
|
| 150 |
}
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:28ee1a580d965a76bcf2cf92000418dfd16e154a05afb479ec9bbddcc7bd6995
|
| 3 |
+
size 6993
|