commit dcb326bf6e60be7ee3c4ec3fef5f7c47835a7a1e Author: ModelHub XC Date: Tue Jun 16 06:15:16 2026 +0800 初始化项目，由ModelHub XC社区提供模型 Model: harsha070/expfinal-qwen-mbpp-s42-base Source: Original Platform diff --git a/.gitattributes b/.gitattributes new file mode 100644 index 0000000..52373fe --- /dev/null +++ b/.gitattributes @@ -0,0 +1,36 @@ +*.7z filter=lfs diff=lfs merge=lfs -text +*.arrow filter=lfs diff=lfs merge=lfs -text +*.bin filter=lfs diff=lfs merge=lfs -text +*.bz2 filter=lfs diff=lfs merge=lfs -text +*.ckpt filter=lfs diff=lfs merge=lfs -text +*.ftz filter=lfs diff=lfs merge=lfs -text +*.gz filter=lfs diff=lfs merge=lfs -text +*.h5 filter=lfs diff=lfs merge=lfs -text +*.joblib filter=lfs diff=lfs merge=lfs -text +*.lfs.* filter=lfs diff=lfs merge=lfs -text +*.mlmodel filter=lfs diff=lfs merge=lfs -text +*.model filter=lfs diff=lfs merge=lfs -text +*.msgpack filter=lfs diff=lfs merge=lfs -text +*.npy filter=lfs diff=lfs merge=lfs -text +*.npz filter=lfs diff=lfs merge=lfs -text +*.onnx filter=lfs diff=lfs merge=lfs -text +*.ot filter=lfs diff=lfs merge=lfs -text +*.parquet filter=lfs diff=lfs merge=lfs -text +*.pb filter=lfs diff=lfs merge=lfs -text +*.pickle filter=lfs diff=lfs merge=lfs -text +*.pkl filter=lfs diff=lfs merge=lfs -text +*.pt filter=lfs diff=lfs merge=lfs -text +*.pth filter=lfs diff=lfs merge=lfs -text +*.rar filter=lfs diff=lfs merge=lfs -text +*.safetensors filter=lfs diff=lfs merge=lfs -text +saved_model/**/* filter=lfs diff=lfs merge=lfs -text +*.tar.* filter=lfs diff=lfs merge=lfs -text +*.tar filter=lfs diff=lfs merge=lfs -text +*.tflite filter=lfs diff=lfs merge=lfs -text +*.tgz filter=lfs diff=lfs merge=lfs -text +*.wasm filter=lfs diff=lfs merge=lfs -text +*.xz filter=lfs diff=lfs merge=lfs -text +*.zip filter=lfs diff=lfs merge=lfs -text +*.zst filter=lfs diff=lfs merge=lfs -text +*tfevents* filter=lfs diff=lfs merge=lfs -text +tokenizer.json filter=lfs diff=lfs merge=lfs -text diff --git a/README.md b/README.md new file mode 100644 index 0000000..5e2fa7c --- /dev/null +++ b/README.md @@ -0,0 +1,67 @@ +--- +base_model: harsha070/sft-warmup-qwen-v1 +library_name: transformers +model_name: expfinal-qwen-mbpp-s42-base +tags: +- generated_from_trainer +- grpo +- trl +licence: license +--- + +# Model Card for expfinal-qwen-mbpp-s42-base + +This model is a fine-tuned version of [harsha070/sft-warmup-qwen-v1](https://huggingface.co/harsha070/sft-warmup-qwen-v1). +It has been trained using [TRL](https://github.com/huggingface/trl). + +## Quick start + +```python +from transformers import pipeline + +question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?" +generator = pipeline("text-generation", model="harsha070/expfinal-qwen-mbpp-s42-base", device="cuda") +output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0] +print(output["generated_text"]) +``` + +## Training procedure + +[Visualize in Weights & Biases](https://wandb.ai/models-self5933/obfuscation-early-warning/runs/xkjevc3j) + + + +This model was trained with GRPO, a method introduced in [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://huggingface.co/papers/2402.03300). + +### Framework versions + +- TRL: 1.3.0 +- Transformers: 5.7.0 +- Pytorch: 2.11.0 +- Datasets: 4.8.5 +- Tokenizers: 0.22.2 + +## Citations + +Cite GRPO as: + +```bibtex +@article{shao2024deepseekmath, + title = {{DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models}}, + author = {Zhihong Shao and Peiyi Wang and Qihao Zhu and Runxin Xu and Junxiao Song and Mingchuan Zhang and Y. K. Li and Y. Wu and Daya Guo}, + year = 2024, + eprint = {arXiv:2402.03300}, +} +``` + +Cite TRL as: + +```bibtex +@software{vonwerra2020trl, + title = {{TRL: Transformers Reinforcement Learning}}, + author = {von Werra, Leandro and Belkada, Younes and Tunstall, Lewis and Beeching, Edward and Thrush, Tristan and Lambert, Nathan and Huang, Shengyi and Rasul, Kashif and Gallouédec, Quentin}, + license = {Apache-2.0}, + url = {https://github.com/huggingface/trl}, + year = {2020} +} +``` \ No newline at end of file diff --git a/chat_template.jinja b/chat_template.jinja new file mode 100644 index 0000000..bdf7919 --- /dev/null +++ b/chat_template.jinja @@ -0,0 +1,54 @@ +{%- if tools %} + {{- '<|im_start|>system\n' }} + {%- if messages[0]['role'] == 'system' %} + {{- messages[0]['content'] }} + {%- else %} + {{- 'You are Qwen, created by Alibaba Cloud. You are a helpful assistant.' }} + {%- endif %} + {{- "\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 XML tags:\n" }} + {%- for tool in tools %} + {{- "\n" }} + {{- tool | tojson }} + {%- endfor %} + {{- "\n\n\nFor each function call, return a json object with function name and arguments within XML tags:\n\n{\"name\": , \"arguments\": }\n<|im_end|>\n" }} +{%- else %} + {%- if messages[0]['role'] == 'system' %} + {{- '<|im_start|>system\n' + messages[0]['content'] + '<|im_end|>\n' }} + {%- else %} + {{- '<|im_start|>system\nYou are Qwen, created by Alibaba Cloud. You are a helpful assistant.<|im_end|>\n' }} + {%- endif %} +{%- endif %} +{%- for message in messages %} + {%- if (message.role == "user") or (message.role == "system" and not loop.first) or (message.role == "assistant" and not message.tool_calls) %} + {{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }} + {%- elif message.role == "assistant" %} + {{- '<|im_start|>' + message.role }} + {%- if message.content %} + {{- '\n' + message.content }} + {%- endif %} + {%- for tool_call in message.tool_calls %} + {%- if tool_call.function is defined %} + {%- set tool_call = tool_call.function %} + {%- endif %} + {{- '\n\n{"name": "' }} + {{- tool_call.name }} + {{- '", "arguments": ' }} + {{- tool_call.arguments | tojson }} + {{- '}\n' }} + {%- endfor %} + {{- '<|im_end|>\n' }} + {%- elif message.role == "tool" %} + {%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != "tool") %} + {{- '<|im_start|>user' }} + {%- endif %} + {{- '\n\n' }} + {{- message.content }} + {{- '\n' }} + {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %} + {{- '<|im_end|>\n' }} + {%- endif %} + {%- endif %} +{%- endfor %} +{%- if add_generation_prompt %} + {{- '<|im_start|>assistant\n' }} +{%- endif %} diff --git a/completions/completions_00010.parquet b/completions/completions_00010.parquet new file mode 100644 index 0000000..c9b6184 --- /dev/null +++ b/completions/completions_00010.parquet @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2a84247eb5ff891760068e5a566aaef5825a5e08c9e7dbdbafd5b190bb51e699 +size 18748 diff --git a/completions/completions_00020.parquet b/completions/completions_00020.parquet new file mode 100644 index 0000000..877377c --- /dev/null +++ b/completions/completions_00020.parquet @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0e34214b92c5229f19ce9874ccb955d996eb59b3c98f39f8ee92c788fcbe8b69 +size 17953 diff --git a/completions/completions_00030.parquet b/completions/completions_00030.parquet new file mode 100644 index 0000000..9c715d0 --- /dev/null +++ b/completions/completions_00030.parquet @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:696f1b8b61a852e1be37a0ceb3ba890b94a1501086583464fa07a030a1f42f77 +size 18316 diff --git a/completions/completions_00040.parquet b/completions/completions_00040.parquet new file mode 100644 index 0000000..1a742dc --- /dev/null +++ b/completions/completions_00040.parquet @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5a51b699434670ffcb0f5f456355809b6b9a0887d8b68fc75214da3e96e5fe47 +size 18415 diff --git a/completions/completions_00050.parquet b/completions/completions_00050.parquet new file mode 100644 index 0000000..e9ffa5d --- /dev/null +++ b/completions/completions_00050.parquet @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:aa179b472cbee312af0c2b8aa6f67f77dbcb20a308b674d272a8e45bbe12be09 +size 18867 diff --git a/completions/completions_00060.parquet b/completions/completions_00060.parquet new file mode 100644 index 0000000..5e2cdbd --- /dev/null +++ b/completions/completions_00060.parquet @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:177dbfd6dc4543d33be598a7323eab0cb0b8f83bb10f6115825aa7547f6d3f80 +size 16484 diff --git a/completions/completions_00070.parquet b/completions/completions_00070.parquet new file mode 100644 index 0000000..0d57871 --- /dev/null +++ b/completions/completions_00070.parquet @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:05f13e120e190c2f3227b19aa7e2ccf1c7b150a62314106dd34a9e95a82abce0 +size 17109 diff --git a/completions/completions_00080.parquet b/completions/completions_00080.parquet new file mode 100644 index 0000000..f78d646 --- /dev/null +++ b/completions/completions_00080.parquet @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:571fe84fff524679c48601932c5f5fa47a4d1e77e8b98d55e53c2038fdb8de8a +size 15914 diff --git a/completions/completions_00090.parquet b/completions/completions_00090.parquet new file mode 100644 index 0000000..c4a65ac --- /dev/null +++ b/completions/completions_00090.parquet @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:60da4365cd0b80087827a53eeaf870b15bc2499f718c9218b23c08bbe3500545 +size 15746 diff --git a/completions/completions_00100.parquet b/completions/completions_00100.parquet new file mode 100644 index 0000000..6ce49e6 --- /dev/null +++ b/completions/completions_00100.parquet @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ef161e938548ca873752d4c8e6f77996317a8148654ac93ccb0603d5a9b1f80f +size 18037 diff --git a/config.json b/config.json new file mode 100644 index 0000000..64c100d --- /dev/null +++ b/config.json @@ -0,0 +1,69 @@ +{ + "architectures": [ + "Qwen2ForCausalLM" + ], + "attention_dropout": 0.0, + "bos_token_id": null, + "dtype": "bfloat16", + "eos_token_id": 151645, + "hidden_act": "silu", + "hidden_size": 2048, + "initializer_range": 0.02, + "intermediate_size": 11008, + "layer_types": [ + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention" + ], + "max_position_embeddings": 32768, + "max_window_layers": 70, + "model_type": "qwen2", + "num_attention_heads": 16, + "num_hidden_layers": 36, + "num_key_value_heads": 2, + "pad_token_id": 151643, + "rms_norm_eps": 1e-06, + "rope_parameters": { + "rope_theta": 1000000.0, + "rope_type": "default" + }, + "sliding_window": null, + "tie_word_embeddings": true, + "transformers_version": "5.7.0", + "use_cache": false, + "use_sliding_window": false, + "vocab_size": 151936 +} diff --git a/generation_config.json b/generation_config.json new file mode 100644 index 0000000..aaf8639 --- /dev/null +++ b/generation_config.json @@ -0,0 +1,13 @@ +{ + "do_sample": true, + "eos_token_id": [ + 151645, + 151643 + ], + "pad_token_id": 151643, + "repetition_penalty": 1.05, + "temperature": 0.7, + "top_k": 20, + "top_p": 0.8, + "transformers_version": "5.7.0" +} diff --git a/last-checkpoint/chat_template.jinja b/last-checkpoint/chat_template.jinja new file mode 100644 index 0000000..bdf7919 --- /dev/null +++ b/last-checkpoint/chat_template.jinja @@ -0,0 +1,54 @@ +{%- if tools %} + {{- '<|im_start|>system\n' }} + {%- if messages[0]['role'] == 'system' %} + {{- messages[0]['content'] }} + {%- else %} + {{- 'You are Qwen, created by Alibaba Cloud. You are a helpful assistant.' }} + {%- endif %} + {{- "\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 XML tags:\n" }} + {%- for tool in tools %} + {{- "\n" }} + {{- tool | tojson }} + {%- endfor %} + {{- "\n\n\nFor each function call, return a json object with function name and arguments within XML tags:\n\n{\"name\": , \"arguments\": }\n<|im_end|>\n" }} +{%- else %} + {%- if messages[0]['role'] == 'system' %} + {{- '<|im_start|>system\n' + messages[0]['content'] + '<|im_end|>\n' }} + {%- else %} + {{- '<|im_start|>system\nYou are Qwen, created by Alibaba Cloud. You are a helpful assistant.<|im_end|>\n' }} + {%- endif %} +{%- endif %} +{%- for message in messages %} + {%- if (message.role == "user") or (message.role == "system" and not loop.first) or (message.role == "assistant" and not message.tool_calls) %} + {{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }} + {%- elif message.role == "assistant" %} + {{- '<|im_start|>' + message.role }} + {%- if message.content %} + {{- '\n' + message.content }} + {%- endif %} + {%- for tool_call in message.tool_calls %} + {%- if tool_call.function is defined %} + {%- set tool_call = tool_call.function %} + {%- endif %} + {{- '\n\n{"name": "' }} + {{- tool_call.name }} + {{- '", "arguments": ' }} + {{- tool_call.arguments | tojson }} + {{- '}\n' }} + {%- endfor %} + {{- '<|im_end|>\n' }} + {%- elif message.role == "tool" %} + {%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != "tool") %} + {{- '<|im_start|>user' }} + {%- endif %} + {{- '\n\n' }} + {{- message.content }} + {{- '\n' }} + {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %} + {{- '<|im_end|>\n' }} + {%- endif %} + {%- endif %} +{%- endfor %} +{%- if add_generation_prompt %} + {{- '<|im_start|>assistant\n' }} +{%- endif %} diff --git a/last-checkpoint/config.json b/last-checkpoint/config.json new file mode 100644 index 0000000..64c100d --- /dev/null +++ b/last-checkpoint/config.json @@ -0,0 +1,69 @@ +{ + "architectures": [ + "Qwen2ForCausalLM" + ], + "attention_dropout": 0.0, + "bos_token_id": null, + "dtype": "bfloat16", + "eos_token_id": 151645, + "hidden_act": "silu", + "hidden_size": 2048, + "initializer_range": 0.02, + "intermediate_size": 11008, + "layer_types": [ + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention" + ], + "max_position_embeddings": 32768, + "max_window_layers": 70, + "model_type": "qwen2", + "num_attention_heads": 16, + "num_hidden_layers": 36, + "num_key_value_heads": 2, + "pad_token_id": 151643, + "rms_norm_eps": 1e-06, + "rope_parameters": { + "rope_theta": 1000000.0, + "rope_type": "default" + }, + "sliding_window": null, + "tie_word_embeddings": true, + "transformers_version": "5.7.0", + "use_cache": false, + "use_sliding_window": false, + "vocab_size": 151936 +} diff --git a/last-checkpoint/generation_config.json b/last-checkpoint/generation_config.json new file mode 100644 index 0000000..aaf8639 --- /dev/null +++ b/last-checkpoint/generation_config.json @@ -0,0 +1,13 @@ +{ + "do_sample": true, + "eos_token_id": [ + 151645, + 151643 + ], + "pad_token_id": 151643, + "repetition_penalty": 1.05, + "temperature": 0.7, + "top_k": 20, + "top_p": 0.8, + "transformers_version": "5.7.0" +} diff --git a/last-checkpoint/model.safetensors b/last-checkpoint/model.safetensors new file mode 100644 index 0000000..43ca29d --- /dev/null +++ b/last-checkpoint/model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:141a7a20f546a1ac816a72981736a5a085805cd0eb72586ee102a1557cf9f193 +size 6171927112 diff --git a/last-checkpoint/tokenizer.json b/last-checkpoint/tokenizer.json new file mode 100644 index 0000000..34510ff --- /dev/null +++ b/last-checkpoint/tokenizer.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3fd169731d2cbde95e10bf356d66d5997fd885dd8dbb6fb4684da3f23b2585d8 +size 11421892 diff --git a/last-checkpoint/tokenizer_config.json b/last-checkpoint/tokenizer_config.json new file mode 100644 index 0000000..770e41d --- /dev/null +++ b/last-checkpoint/tokenizer_config.json @@ -0,0 +1,30 @@ +{ + "add_prefix_space": false, + "backend": "tokenizers", + "bos_token": null, + "clean_up_tokenization_spaces": false, + "eos_token": "<|im_end|>", + "errors": "replace", + "extra_special_tokens": [ + "<|im_start|>", + "<|im_end|>", + "<|object_ref_start|>", + "<|object_ref_end|>", + "<|box_start|>", + "<|box_end|>", + "<|quad_start|>", + "<|quad_end|>", + "<|vision_start|>", + "<|vision_end|>", + "<|vision_pad|>", + "<|image_pad|>", + "<|video_pad|>" + ], + "is_local": false, + "local_files_only": false, + "model_max_length": 131072, + "pad_token": "<|endoftext|>", + "split_special_tokens": false, + "tokenizer_class": "Qwen2Tokenizer", + "unk_token": null +} diff --git a/last-checkpoint/trainer_state.json b/last-checkpoint/trainer_state.json new file mode 100644 index 0000000..3b36161 --- /dev/null +++ b/last-checkpoint/trainer_state.json @@ -0,0 +1,314 @@ +{ + "best_global_step": null, + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 0.6666666666666666, + "eval_steps": 500, + "global_step": 100, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "clip_ratio/high_max": 0.0, + "clip_ratio/high_mean": 0.0, + "clip_ratio/low_mean": 0.0, + "clip_ratio/low_min": 0.0, + "clip_ratio/region_mean": 0.0, + "completions/clipped_ratio": 0.7375, + "completions/max_length": 512.0, + "completions/max_terminated_length": 284.9, + "completions/mean_length": 491.525, + "completions/mean_terminated_length": 261.77666931152345, + "completions/min_length": 438.2, + "completions/min_terminated_length": 233.4, + "entropy": 0.23946007741615177, + "epoch": 0.06666666666666667, + "frac_reward_zero_std": 0.45, + "grad_norm": 1.453125, + "kl": 0.05540826761134667, + "learning_rate": 9.100000000000001e-06, + "loss": 0.0011166661977767944, + "num_tokens": 52190.0, + "reward": 0.7, + "reward_std": 0.35391277372837066, + "rewards/JointRewardFunction/mean": 0.7, + "rewards/JointRewardFunction/std": 0.3539127916097641, + "step": 10, + "step_time": 22.393758795900045 + }, + { + "clip_ratio/high_max": 0.0, + "clip_ratio/high_mean": 0.0, + "clip_ratio/low_mean": 0.0, + "clip_ratio/low_min": 0.0, + "clip_ratio/region_mean": 0.0, + "completions/clipped_ratio": 0.7625, + "completions/max_length": 512.0, + "completions/max_terminated_length": 380.1, + "completions/mean_length": 495.3625, + "completions/mean_terminated_length": 357.5166717529297, + "completions/min_length": 439.9, + "completions/min_terminated_length": 337.5, + "entropy": 0.15637785932049156, + "epoch": 0.13333333333333333, + "frac_reward_zero_std": 0.5, + "grad_norm": 0.024169921875, + "kl": 0.07800487959757448, + "learning_rate": 8.1e-06, + "loss": -0.0069251880049705505, + "num_tokens": 105579.0, + "reward": 0.775, + "reward_std": 0.3446744382381439, + "rewards/JointRewardFunction/mean": 0.775, + "rewards/JointRewardFunction/std": 0.3446744501590729, + "step": 20, + "step_time": 22.42510473759985 + }, + { + "clip_ratio/high_max": 0.0, + "clip_ratio/high_mean": 0.0, + "clip_ratio/low_mean": 0.0, + "clip_ratio/low_min": 0.0, + "clip_ratio/region_mean": 0.0, + "completions/clipped_ratio": 0.8125, + "completions/max_length": 512.0, + "completions/max_terminated_length": 382.0, + "completions/mean_length": 500.8125, + "completions/mean_terminated_length": 368.27000122070314, + "completions/min_length": 451.0, + "completions/min_terminated_length": 348.6, + "entropy": 0.16634994922205806, + "epoch": 0.2, + "frac_reward_zero_std": 0.65, + "grad_norm": 1.5390625, + "kl": 0.10815342329442501, + "learning_rate": 7.100000000000001e-06, + "loss": 0.005232430994510651, + "num_tokens": 157060.0, + "reward": 0.8875, + "reward_std": 0.2394672751426697, + "rewards/JointRewardFunction/mean": 0.8875, + "rewards/JointRewardFunction/std": 0.23946728110313414, + "step": 30, + "step_time": 22.14527411999982 + }, + { + "clip_ratio/high_max": 0.0, + "clip_ratio/high_mean": 0.0, + "clip_ratio/low_mean": 0.0, + "clip_ratio/low_min": 0.0, + "clip_ratio/region_mean": 0.0, + "completions/clipped_ratio": 0.8875, + "completions/max_length": 512.0, + "completions/max_terminated_length": 262.4, + "completions/mean_length": 502.0125, + "completions/mean_terminated_length": 256.325, + "completions/min_length": 451.3, + "completions/min_terminated_length": 246.5, + "entropy": 0.15448929518461227, + "epoch": 0.26666666666666666, + "frac_reward_zero_std": 0.45, + "grad_norm": 0.02001953125, + "kl": 0.08631132342852652, + "learning_rate": 6.1e-06, + "loss": 0.0021702755242586137, + "num_tokens": 210533.0, + "reward": 0.7625, + "reward_std": 0.3475900113582611, + "rewards/JointRewardFunction/mean": 0.7625, + "rewards/JointRewardFunction/std": 0.34759002923965454, + "step": 40, + "step_time": 22.401593531500158 + }, + { + "clip_ratio/high_max": 0.0, + "clip_ratio/high_mean": 0.0, + "clip_ratio/low_mean": 0.0, + "clip_ratio/low_min": 0.0, + "clip_ratio/region_mean": 0.0, + "completions/clipped_ratio": 0.925, + "completions/max_length": 512.0, + "completions/max_terminated_length": 235.5, + "completions/mean_length": 509.125, + "completions/mean_terminated_length": 234.8, + "completions/min_length": 490.1, + "completions/min_terminated_length": 234.1, + "entropy": 0.15993599854409696, + "epoch": 0.3333333333333333, + "frac_reward_zero_std": 0.6, + "grad_norm": 1.3125, + "kl": 0.10001510812435299, + "learning_rate": 5.1e-06, + "loss": -0.001868080161511898, + "num_tokens": 263979.0, + "reward": 0.8375, + "reward_std": 0.29291952252388, + "rewards/JointRewardFunction/mean": 0.8375, + "rewards/JointRewardFunction/std": 0.2929195284843445, + "step": 50, + "step_time": 22.454355105500007 + }, + { + "clip_ratio/high_max": 0.0, + "clip_ratio/high_mean": 0.0, + "clip_ratio/low_mean": 0.0, + "clip_ratio/low_min": 0.0, + "clip_ratio/region_mean": 0.0, + "completions/clipped_ratio": 0.9125, + "completions/max_length": 512.0, + "completions/max_terminated_length": 222.0, + "completions/mean_length": 505.65, + "completions/mean_terminated_length": 219.93333435058594, + "completions/min_length": 474.0, + "completions/min_terminated_length": 218.0, + "entropy": 0.15155332698486745, + "epoch": 0.4, + "frac_reward_zero_std": 0.65, + "grad_norm": 1.265625, + "kl": 0.0995997670572251, + "learning_rate": 4.1e-06, + "loss": 0.0030021272599697114, + "num_tokens": 317831.0, + "reward": 0.8125, + "reward_std": 0.32402919232845306, + "rewards/JointRewardFunction/mean": 0.8125, + "rewards/JointRewardFunction/std": 0.32402920424938203, + "step": 60, + "step_time": 22.315176106599846 + }, + { + "clip_ratio/high_max": 0.0, + "clip_ratio/high_mean": 0.0, + "clip_ratio/low_mean": 0.0, + "clip_ratio/low_min": 0.0, + "clip_ratio/region_mean": 0.0, + "completions/clipped_ratio": 0.9125, + "completions/max_length": 512.0, + "completions/max_terminated_length": 182.4, + "completions/mean_length": 505.3875, + "completions/mean_terminated_length": 174.23333435058595, + "completions/min_length": 475.2, + "completions/min_terminated_length": 168.0, + "entropy": 0.13972499519586562, + "epoch": 0.4666666666666667, + "frac_reward_zero_std": 0.65, + "grad_norm": 0.0235595703125, + "kl": 0.11641144650056959, + "learning_rate": 3.1000000000000004e-06, + "loss": 0.002840310521423817, + "num_tokens": 372058.0, + "reward": 0.8, + "reward_std": 0.3038551896810532, + "rewards/JointRewardFunction/mean": 0.8, + "rewards/JointRewardFunction/std": 0.30385519564151764, + "step": 70, + "step_time": 22.520917878499766 + }, + { + "clip_ratio/high_max": 0.0, + "clip_ratio/high_mean": 0.0, + "clip_ratio/low_mean": 0.0, + "clip_ratio/low_min": 0.0, + "clip_ratio/region_mean": 0.0, + "completions/clipped_ratio": 0.8875, + "completions/max_length": 512.0, + "completions/max_terminated_length": 238.4, + "completions/mean_length": 504.55, + "completions/mean_terminated_length": 225.6, + "completions/min_length": 466.8, + "completions/min_terminated_length": 210.8, + "entropy": 0.14867580346763135, + "epoch": 0.5333333333333333, + "frac_reward_zero_std": 0.7, + "grad_norm": 1.109375, + "kl": 0.09885827025864273, + "learning_rate": 2.1000000000000002e-06, + "loss": 9.710401645861566e-05, + "num_tokens": 426094.0, + "reward": 0.8625, + "reward_std": 0.2558668524026871, + "rewards/JointRewardFunction/mean": 0.8625, + "rewards/JointRewardFunction/std": 0.25586686432361605, + "step": 80, + "step_time": 22.384258029100057 + }, + { + "clip_ratio/high_max": 0.0, + "clip_ratio/high_mean": 0.0, + "clip_ratio/low_mean": 0.0, + "clip_ratio/low_min": 0.0, + "clip_ratio/region_mean": 0.0, + "completions/clipped_ratio": 0.9375, + "completions/max_length": 512.0, + "completions/max_terminated_length": 144.1, + "completions/mean_length": 508.4875, + "completions/mean_terminated_length": 139.3, + "completions/min_length": 492.9, + "completions/min_terminated_length": 134.5, + "entropy": 0.1391168820671737, + "epoch": 0.6, + "frac_reward_zero_std": 0.65, + "grad_norm": 1.1953125, + "kl": 0.11537287631072105, + "learning_rate": 1.1e-06, + "loss": -0.004437017068266868, + "num_tokens": 480921.0, + "reward": 0.7875, + "reward_std": 0.29291952252388, + "rewards/JointRewardFunction/mean": 0.7875, + "rewards/JointRewardFunction/std": 0.2929195284843445, + "step": 90, + "step_time": 22.70512801439986 + }, + { + "clip_ratio/high_max": 0.0, + "clip_ratio/high_mean": 0.0, + "clip_ratio/low_mean": 0.0, + "clip_ratio/low_min": 0.0, + "clip_ratio/region_mean": 0.0, + "completions/clipped_ratio": 0.9125, + "completions/max_length": 512.0, + "completions/max_terminated_length": 230.5, + "completions/mean_length": 504.3625, + "completions/mean_terminated_length": 220.3, + "completions/min_length": 466.1, + "completions/min_terminated_length": 210.1, + "entropy": 0.1445184069685638, + "epoch": 0.6666666666666666, + "frac_reward_zero_std": 0.75, + "grad_norm": 0.0107421875, + "kl": 0.09200373792555183, + "learning_rate": 1.0000000000000001e-07, + "loss": 0.002389365620911121, + "num_tokens": 536154.0, + "reward": 0.8625, + "reward_std": 0.22220884561538695, + "rewards/JointRewardFunction/mean": 0.8625, + "rewards/JointRewardFunction/std": 0.22220885157585143, + "step": 100, + "step_time": 22.48313269170003 + } + ], + "logging_steps": 10, + "max_steps": 100, + "num_input_tokens_seen": 536154, + "num_train_epochs": 1, + "save_steps": 20, + "stateful_callbacks": { + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": true + }, + "attributes": {} + } + }, + "total_flos": 0.0, + "train_batch_size": 1, + "trial_name": null, + "trial_params": null +} diff --git a/last-checkpoint/training_args.bin b/last-checkpoint/training_args.bin new file mode 100644 index 0000000..c7da1ed --- /dev/null +++ b/last-checkpoint/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5496dd1567f20e3054dbbb5283ffc95a09e5c6ca9e0c146986d319e67f4bdd33 +size 7249 diff --git a/model.safetensors b/model.safetensors new file mode 100644 index 0000000..43ca29d --- /dev/null +++ b/model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:141a7a20f546a1ac816a72981736a5a085805cd0eb72586ee102a1557cf9f193 +size 6171927112 diff --git a/tokenizer.json b/tokenizer.json new file mode 100644 index 0000000..34510ff --- /dev/null +++ b/tokenizer.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3fd169731d2cbde95e10bf356d66d5997fd885dd8dbb6fb4684da3f23b2585d8 +size 11421892 diff --git a/tokenizer_config.json b/tokenizer_config.json new file mode 100644 index 0000000..770e41d --- /dev/null +++ b/tokenizer_config.json @@ -0,0 +1,30 @@ +{ + "add_prefix_space": false, + "backend": "tokenizers", + "bos_token": null, + "clean_up_tokenization_spaces": false, + "eos_token": "<|im_end|>", + "errors": "replace", + "extra_special_tokens": [ + "<|im_start|>", + "<|im_end|>", + "<|object_ref_start|>", + "<|object_ref_end|>", + "<|box_start|>", + "<|box_end|>", + "<|quad_start|>", + "<|quad_end|>", + "<|vision_start|>", + "<|vision_end|>", + "<|vision_pad|>", + "<|image_pad|>", + "<|video_pad|>" + ], + "is_local": false, + "local_files_only": false, + "model_max_length": 131072, + "pad_token": "<|endoftext|>", + "split_special_tokens": false, + "tokenizer_class": "Qwen2Tokenizer", + "unk_token": null +} diff --git a/training_args.bin b/training_args.bin new file mode 100644 index 0000000..c7da1ed --- /dev/null +++ b/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5496dd1567f20e3054dbbb5283ffc95a09e5c6ca9e0c146986d319e67f4bdd33 +size 7249