commit 72640e3c3c2d65c639f268000b7c2d16502b7407 Author: ModelHub XC Date: Tue Jun 16 06:58:17 2026 +0800 初始化项目，由ModelHub XC社区提供模型 Model: harsha070/expfinal-qwen-mbpp-s42-lambda-0p75 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..497c7fb --- /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-lambda-0p75 +tags: +- generated_from_trainer +- grpo +- trl +licence: license +--- + +# Model Card for expfinal-qwen-mbpp-s42-lambda-0p75 + +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-lambda-0p75", 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/ygj72uo9) + + + +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..d35f913 --- /dev/null +++ b/completions/completions_00010.parquet @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4759825f34ab1a4e2b3436d0e0c513856587e242d04bbcaee84f641f6f7db680 +size 14278 diff --git a/completions/completions_00020.parquet b/completions/completions_00020.parquet new file mode 100644 index 0000000..06866d5 --- /dev/null +++ b/completions/completions_00020.parquet @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9a1f1f51bdc71b4c076b70511255f730521face27b392168d5e7bcec64c2dd09 +size 14435 diff --git a/completions/completions_00030.parquet b/completions/completions_00030.parquet new file mode 100644 index 0000000..998e5c1 --- /dev/null +++ b/completions/completions_00030.parquet @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:58a26e1179f8e08b57c51f03344f21b695c85b4ce2d17ff148919d3a5e6f2904 +size 14987 diff --git a/completions/completions_00040.parquet b/completions/completions_00040.parquet new file mode 100644 index 0000000..56bfb87 --- /dev/null +++ b/completions/completions_00040.parquet @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:800593a66d57f53c12faff4e1f426b11f846d1d8a809090ed918e0f9d64e96ab +size 14276 diff --git a/completions/completions_00050.parquet b/completions/completions_00050.parquet new file mode 100644 index 0000000..0fb5c79 --- /dev/null +++ b/completions/completions_00050.parquet @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5b7ebf3cc40b42a20629d5207a831e6540c96fdf7200636c9f311be925d55fbd +size 16951 diff --git a/completions/completions_00060.parquet b/completions/completions_00060.parquet new file mode 100644 index 0000000..0ee0d3a --- /dev/null +++ b/completions/completions_00060.parquet @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9ab839d5cd2fbb31028790207b60e13c9a41d0fc315d3ee52fd59914a46b57a1 +size 17181 diff --git a/completions/completions_00070.parquet b/completions/completions_00070.parquet new file mode 100644 index 0000000..f39dd92 --- /dev/null +++ b/completions/completions_00070.parquet @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a09134033cacd8cb6f64573317575cc1df987bb9091b134dd57d440d015c10c3 +size 15649 diff --git a/completions/completions_00080.parquet b/completions/completions_00080.parquet new file mode 100644 index 0000000..77d6b57 --- /dev/null +++ b/completions/completions_00080.parquet @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:aaec41b4a4bae7fd30722d5614a141884b348085686861df21831771161086df +size 17667 diff --git a/completions/completions_00090.parquet b/completions/completions_00090.parquet new file mode 100644 index 0000000..0cc77fe --- /dev/null +++ b/completions/completions_00090.parquet @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:515ba7091175f986b1a02166baba9170033faa51a7bd8b9801fc80547e351cd1 +size 16809 diff --git a/completions/completions_00100.parquet b/completions/completions_00100.parquet new file mode 100644 index 0000000..911c1b6 --- /dev/null +++ b/completions/completions_00100.parquet @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:28477cfbb7b51953c24cb577e9d0c7fb8519d255a79964524a90af665516a2aa +size 15182 diff --git a/completions/completions_00110.parquet b/completions/completions_00110.parquet new file mode 100644 index 0000000..09542d1 --- /dev/null +++ b/completions/completions_00110.parquet @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:85940ea412bb805d2f2f34defea92f36dda1fc608d4535fb7618f7e076082d85 +size 16789 diff --git a/completions/completions_00120.parquet b/completions/completions_00120.parquet new file mode 100644 index 0000000..634154e --- /dev/null +++ b/completions/completions_00120.parquet @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5e38b75427052a43a4268081a0052a30a5c80b66ac4158ff109284bdb81e9438 +size 18474 diff --git a/completions/completions_00130.parquet b/completions/completions_00130.parquet new file mode 100644 index 0000000..319569c --- /dev/null +++ b/completions/completions_00130.parquet @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:cff4469e753ffa11b38a6e2153091f527ae743d9a9a4924d4992ec7a61ff322b +size 12896 diff --git a/completions/completions_00140.parquet b/completions/completions_00140.parquet new file mode 100644 index 0000000..e533376 --- /dev/null +++ b/completions/completions_00140.parquet @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5957b6ed35fe7a10bf49400f90f31ea2609cb1d3bde8d3198dd13df849d9d79b +size 14117 diff --git a/completions/completions_00150.parquet b/completions/completions_00150.parquet new file mode 100644 index 0000000..0ce533d --- /dev/null +++ b/completions/completions_00150.parquet @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f9e7c50fa7aea973d00c0c85500fae58d0d80ac7cde60b4b657b1f0462239385 +size 15649 diff --git a/completions/completions_00160.parquet b/completions/completions_00160.parquet new file mode 100644 index 0000000..1194850 --- /dev/null +++ b/completions/completions_00160.parquet @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:03a407d307866993e202b1e9801d5798223dbf42fe263f45b708725c27118c9b +size 13980 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..7048198 --- /dev/null +++ b/last-checkpoint/model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6753ac3d39840dcbcb8da23e74296953ddc011bb9abcf19b87e7c7e7bfb56447 +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..fbb1e2d --- /dev/null +++ b/last-checkpoint/trainer_state.json @@ -0,0 +1,482 @@ +{ + "best_global_step": null, + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 1.0666666666666667, + "eval_steps": 500, + "global_step": 160, + "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": 0.0625, + "completions/max_length": 419.8, + "completions/max_terminated_length": 406.8, + "completions/mean_length": 314.8625, + "completions/mean_terminated_length": 307.5414337158203, + "completions/min_length": 206.3, + "completions/min_terminated_length": 206.3, + "entropy": 0.44715926721692084, + "epoch": 0.06666666666666667, + "frac_reward_zero_std": 0.15, + "grad_norm": 4.625, + "kl": 0.09871633112052222, + "learning_rate": 9.4375e-06, + "loss": -0.010261553525924682, + "num_tokens": 38057.0, + "reward": 0.9905517578125, + "reward_std": 0.6565703094005585, + "rewards/JointRewardFunction/mean": 0.9905517578125, + "rewards/JointRewardFunction/std": 0.6565703094005585, + "step": 10, + "step_time": 19.035294922399043 + }, + { + "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": 0.2, + "completions/max_length": 458.4, + "completions/max_terminated_length": 426.1, + "completions/mean_length": 343.8, + "completions/mean_terminated_length": 313.08583984375, + "completions/min_length": 211.7, + "completions/min_terminated_length": 211.7, + "entropy": 0.31359912948682905, + "epoch": 0.13333333333333333, + "frac_reward_zero_std": 0.1, + "grad_norm": 3.125, + "kl": 0.14868967132642866, + "learning_rate": 8.8125e-06, + "loss": -0.023439544439315795, + "num_tokens": 79321.0, + "reward": 1.072357177734375, + "reward_std": 0.5760923266410828, + "rewards/JointRewardFunction/mean": 1.072357177734375, + "rewards/JointRewardFunction/std": 0.5760923385620117, + "step": 20, + "step_time": 21.0788949906022 + }, + { + "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": 0.0, + "completions/max_length": 351.8, + "completions/max_terminated_length": 351.8, + "completions/mean_length": 260.625, + "completions/mean_terminated_length": 260.625, + "completions/min_length": 174.2, + "completions/min_terminated_length": 174.2, + "entropy": 0.3843053654767573, + "epoch": 0.2, + "frac_reward_zero_std": 0.1, + "grad_norm": 4.5625, + "kl": 0.18629266200587152, + "learning_rate": 8.1875e-06, + "loss": 0.004870641976594925, + "num_tokens": 111587.0, + "reward": 1.3501953125, + "reward_std": 0.6562108278274537, + "rewards/JointRewardFunction/mean": 1.3501953125, + "rewards/JointRewardFunction/std": 0.6562108367681503, + "step": 30, + "step_time": 16.898221987598664 + }, + { + "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": 0.0375, + "completions/max_length": 429.1, + "completions/max_terminated_length": 425.2, + "completions/mean_length": 327.5125, + "completions/mean_terminated_length": 321.53809814453126, + "completions/min_length": 227.9, + "completions/min_terminated_length": 227.9, + "entropy": 0.44414809048175813, + "epoch": 0.26666666666666666, + "frac_reward_zero_std": 0.35, + "grad_norm": 6.34375, + "kl": 0.14110819818452, + "learning_rate": 7.5625e-06, + "loss": -0.03268973827362061, + "num_tokens": 151100.0, + "reward": 1.493701171875, + "reward_std": 0.3899420065339655, + "rewards/JointRewardFunction/mean": 1.493701171875, + "rewards/JointRewardFunction/std": 0.38994201249442995, + "step": 40, + "step_time": 20.12026042639991 + }, + { + "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": 0.225, + "completions/max_length": 472.2, + "completions/max_terminated_length": 427.6, + "completions/mean_length": 352.75, + "completions/mean_terminated_length": 313.9489349365234, + "completions/min_length": 201.0, + "completions/min_terminated_length": 201.0, + "entropy": 0.5324544316157699, + "epoch": 0.3333333333333333, + "frac_reward_zero_std": 0.25, + "grad_norm": 5.3125, + "kl": 0.13007921809330583, + "learning_rate": 6.9375e-06, + "loss": -0.03016577959060669, + "num_tokens": 192036.0, + "reward": 1.0732177734375, + "reward_std": 0.5164182722568512, + "rewards/JointRewardFunction/mean": 1.0732177734375, + "rewards/JointRewardFunction/std": 0.5164182841777801, + "step": 50, + "step_time": 21.588160099001833 + }, + { + "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": 0.225, + "completions/max_length": 495.4, + "completions/max_terminated_length": 461.4, + "completions/mean_length": 396.95, + "completions/mean_terminated_length": 363.4666687011719, + "completions/min_length": 224.4, + "completions/min_terminated_length": 224.4, + "entropy": 0.47121726758778093, + "epoch": 0.4, + "frac_reward_zero_std": 0.05, + "grad_norm": 2.46875, + "kl": 0.10462474231608213, + "learning_rate": 6.3125e-06, + "loss": -0.031146174669265746, + "num_tokens": 237192.0, + "reward": 1.193585205078125, + "reward_std": 0.5929911494255066, + "rewards/JointRewardFunction/mean": 1.193585205078125, + "rewards/JointRewardFunction/std": 0.592991155385971, + "step": 60, + "step_time": 22.437090325800817 + }, + { + "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": 0.275, + "completions/max_length": 497.6, + "completions/max_terminated_length": 455.6, + "completions/mean_length": 392.0375, + "completions/mean_terminated_length": 351.8452484130859, + "completions/min_length": 243.3, + "completions/min_terminated_length": 243.3, + "entropy": 0.4913482774049044, + "epoch": 0.4666666666666667, + "frac_reward_zero_std": 0.2, + "grad_norm": 4.21875, + "kl": 0.11023280541412532, + "learning_rate": 5.6875e-06, + "loss": -0.02786441147327423, + "num_tokens": 282351.0, + "reward": 1.166510009765625, + "reward_std": 0.5283917605876922, + "rewards/JointRewardFunction/mean": 1.166510009765625, + "rewards/JointRewardFunction/std": 0.528391769528389, + "step": 70, + "step_time": 22.72093682029881 + }, + { + "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": 0.3125, + "completions/max_length": 507.2, + "completions/max_terminated_length": 447.3, + "completions/mean_length": 409.9125, + "completions/mean_terminated_length": 365.60191040039064, + "completions/min_length": 248.5, + "completions/min_terminated_length": 248.5, + "entropy": 0.43941801562905314, + "epoch": 0.5333333333333333, + "frac_reward_zero_std": 0.2, + "grad_norm": 1.640625, + "kl": 0.12287256089039147, + "learning_rate": 5.0625e-06, + "loss": -0.024439637362957, + "num_tokens": 328816.0, + "reward": 1.467431640625, + "reward_std": 0.4256438536103815, + "rewards/JointRewardFunction/mean": 1.467431640625, + "rewards/JointRewardFunction/std": 0.42564386553131045, + "step": 80, + "step_time": 23.137156244499057 + }, + { + "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": 0.1875, + "completions/max_length": 486.2, + "completions/max_terminated_length": 463.3, + "completions/mean_length": 379.0625, + "completions/mean_terminated_length": 360.1238128662109, + "completions/min_length": 241.6, + "completions/min_terminated_length": 241.6, + "entropy": 0.3873957570642233, + "epoch": 0.6, + "frac_reward_zero_std": 0.35, + "grad_norm": 2.15625, + "kl": 0.12395602646283806, + "learning_rate": 4.4375e-06, + "loss": 0.0433960497379303, + "num_tokens": 373289.0, + "reward": 1.3514892578125, + "reward_std": 0.43082923418842256, + "rewards/JointRewardFunction/mean": 1.3514892578125, + "rewards/JointRewardFunction/std": 0.4308292371686548, + "step": 90, + "step_time": 22.175798371599377 + }, + { + "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": 0.1375, + "completions/max_length": 475.4, + "completions/max_terminated_length": 439.4, + "completions/mean_length": 344.5625, + "completions/mean_terminated_length": 319.0309600830078, + "completions/min_length": 209.0, + "completions/min_terminated_length": 209.0, + "entropy": 0.3818320112302899, + "epoch": 0.6666666666666666, + "frac_reward_zero_std": 0.15, + "grad_norm": 1.7265625, + "kl": 0.1494429545942694, + "learning_rate": 3.8125e-06, + "loss": -0.00486765056848526, + "num_tokens": 415738.0, + "reward": 1.4651611328125, + "reward_std": 0.453005512803793, + "rewards/JointRewardFunction/mean": 1.4651611328125, + "rewards/JointRewardFunction/std": 0.45300552546977996, + "step": 100, + "step_time": 21.99789129000128 + }, + { + "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": 0.125, + "completions/max_length": 500.8, + "completions/max_terminated_length": 454.0, + "completions/mean_length": 348.225, + "completions/mean_terminated_length": 326.1460815429688, + "completions/min_length": 184.0, + "completions/min_terminated_length": 184.0, + "entropy": 0.38654828127473595, + "epoch": 0.7333333333333333, + "frac_reward_zero_std": 0.25, + "grad_norm": 4.25, + "kl": 0.19588984637521206, + "learning_rate": 3.1875e-06, + "loss": -0.04184774756431579, + "num_tokens": 456544.0, + "reward": 1.449822998046875, + "reward_std": 0.40167671740055083, + "rewards/JointRewardFunction/mean": 1.449822998046875, + "rewards/JointRewardFunction/std": 0.40167671740055083, + "step": 110, + "step_time": 22.696052773901464 + }, + { + "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": 0.1875, + "completions/max_length": 474.8, + "completions/max_terminated_length": 426.7, + "completions/mean_length": 338.5375, + "completions/mean_terminated_length": 308.2898864746094, + "completions/min_length": 179.5, + "completions/min_terminated_length": 179.5, + "entropy": 0.42813673401251434, + "epoch": 0.8, + "frac_reward_zero_std": 0.15, + "grad_norm": 3.625, + "kl": 0.188834874285385, + "learning_rate": 2.5625e-06, + "loss": -0.022572511434555055, + "num_tokens": 495891.0, + "reward": 1.37578125, + "reward_std": 0.485464009642601, + "rewards/JointRewardFunction/mean": 1.37578125, + "rewards/JointRewardFunction/std": 0.4854640245437622, + "step": 120, + "step_time": 21.932242763400062 + }, + { + "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": 0.225, + "completions/max_length": 480.4, + "completions/max_terminated_length": 457.0, + "completions/mean_length": 353.0375, + "completions/mean_terminated_length": 309.4391723632813, + "completions/min_length": 93.9, + "completions/min_terminated_length": 93.9, + "entropy": 0.4207899322733283, + "epoch": 0.8666666666666667, + "frac_reward_zero_std": 0.15, + "grad_norm": 3.140625, + "kl": 0.19314469541423024, + "learning_rate": 1.9375e-06, + "loss": -0.10344405174255371, + "num_tokens": 538386.0, + "reward": 1.3167236328125, + "reward_std": 0.5021238774061203, + "rewards/JointRewardFunction/mean": 1.3167236328125, + "rewards/JointRewardFunction/std": 0.5021239161491394, + "step": 130, + "step_time": 22.154275490202053 + }, + { + "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": 0.1375, + "completions/max_length": 468.1, + "completions/max_terminated_length": 443.4, + "completions/mean_length": 351.3625, + "completions/mean_terminated_length": 333.25274353027345, + "completions/min_length": 215.1, + "completions/min_terminated_length": 215.1, + "entropy": 0.3897366087883711, + "epoch": 0.9333333333333333, + "frac_reward_zero_std": 0.4, + "grad_norm": 1.8828125, + "kl": 0.1783840640448034, + "learning_rate": 1.3125000000000001e-06, + "loss": -0.012397536635398864, + "num_tokens": 579127.0, + "reward": 1.5228759765625, + "reward_std": 0.3425671649631113, + "rewards/JointRewardFunction/mean": 1.5228759765625, + "rewards/JointRewardFunction/std": 0.3425671679433435, + "step": 140, + "step_time": 21.521125249801845 + }, + { + "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": 0.15, + "completions/max_length": 482.1, + "completions/max_terminated_length": 444.4, + "completions/mean_length": 354.5375, + "completions/mean_terminated_length": 328.6072692871094, + "completions/min_length": 207.4, + "completions/min_terminated_length": 207.4, + "entropy": 0.4106031972914934, + "epoch": 1.0, + "frac_reward_zero_std": 0.2, + "grad_norm": 1.703125, + "kl": 0.16290537370368838, + "learning_rate": 6.875000000000001e-07, + "loss": 0.014532069861888885, + "num_tokens": 619802.0, + "reward": 1.502001953125, + "reward_std": 0.44335485696792604, + "rewards/JointRewardFunction/mean": 1.502001953125, + "rewards/JointRewardFunction/std": 0.443354868888855, + "step": 150, + "step_time": 22.203639667498646 + }, + { + "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": 0.125, + "completions/max_length": 432.7, + "completions/max_terminated_length": 389.6, + "completions/mean_length": 310.0125, + "completions/mean_terminated_length": 283.19702911376953, + "completions/min_length": 129.3, + "completions/min_terminated_length": 129.3, + "entropy": 0.43815543316304684, + "epoch": 1.0666666666666667, + "frac_reward_zero_std": 0.2, + "grad_norm": 3.484375, + "kl": 0.21010836036875843, + "learning_rate": 6.250000000000001e-08, + "loss": -0.05822192430496216, + "num_tokens": 656519.0, + "reward": 1.34354248046875, + "reward_std": 0.5034710764884949, + "rewards/JointRewardFunction/mean": 1.34354248046875, + "rewards/JointRewardFunction/std": 0.503471103310585, + "step": 160, + "step_time": 20.17411774720167 + } + ], + "logging_steps": 10, + "max_steps": 160, + "num_input_tokens_seen": 656519, + "num_train_epochs": 2, + "save_steps": 10, + "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..229d444 --- /dev/null +++ b/last-checkpoint/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:425595137c947d190b8d7b621fa2afaf7a55b15db5aef2061b280d7894998eec +size 7249 diff --git a/model.safetensors b/model.safetensors new file mode 100644 index 0000000..7048198 --- /dev/null +++ b/model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6753ac3d39840dcbcb8da23e74296953ddc011bb9abcf19b87e7c7e7bfb56447 +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..229d444 --- /dev/null +++ b/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:425595137c947d190b8d7b621fa2afaf7a55b15db5aef2061b280d7894998eec +size 7249