commit 6bf7a7c528084c4d82b9d692903a416edf1a3260 Author: ModelHub XC Date: Thu Jun 25 22:11:24 2026 +0800 初始化项目，由ModelHub XC社区提供模型 Model: brysgo/gol-grpo-fixed-validation-37156495 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..7c71750 --- /dev/null +++ b/README.md @@ -0,0 +1,67 @@ +--- +base_model: Qwen/Qwen2.5-0.5B-Instruct +library_name: transformers +model_name: gol-grpo-fixed-50 +tags: +- generated_from_trainer +- trl +- grpo +licence: license +--- + +# Model Card for gol-grpo-fixed-50 + +This model is a fine-tuned version of [Qwen/Qwen2.5-0.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-0.5B-Instruct). +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="None", device="cuda") +output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0] +print(output["generated_text"]) +``` + +## Training procedure + + + + + +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: 1.4.0 +- Transformers: 5.9.0 +- Pytorch: 2.11.0+cu130 +- 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/checkpoint-30/chat_template.jinja b/checkpoint-30/chat_template.jinja new file mode 100644 index 0000000..bdf7919 --- /dev/null +++ b/checkpoint-30/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/checkpoint-30/config.json b/checkpoint-30/config.json new file mode 100644 index 0000000..f62c429 --- /dev/null +++ b/checkpoint-30/config.json @@ -0,0 +1,57 @@ +{ + "architectures": [ + "Qwen2ForCausalLM" + ], + "attention_dropout": 0.0, + "bos_token_id": null, + "dtype": "float32", + "eos_token_id": 151645, + "hidden_act": "silu", + "hidden_size": 896, + "initializer_range": 0.02, + "intermediate_size": 4864, + "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" + ], + "max_position_embeddings": 32768, + "max_window_layers": 21, + "model_type": "qwen2", + "num_attention_heads": 14, + "num_hidden_layers": 24, + "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": "5.9.0", + "use_cache": false, + "use_sliding_window": false, + "vocab_size": 151936 +} diff --git a/checkpoint-30/generation_config.json b/checkpoint-30/generation_config.json new file mode 100644 index 0000000..21d3d9d --- /dev/null +++ b/checkpoint-30/generation_config.json @@ -0,0 +1,13 @@ +{ + "do_sample": true, + "eos_token_id": [ + 151645, + 151643 + ], + "pad_token_id": 151643, + "repetition_penalty": 1.1, + "temperature": 0.7, + "top_k": 20, + "top_p": 0.8, + "transformers_version": "5.9.0" +} diff --git a/checkpoint-30/model.safetensors b/checkpoint-30/model.safetensors new file mode 100644 index 0000000..4535412 --- /dev/null +++ b/checkpoint-30/model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:920b4e656f4ea47583dd538101d84eddb6cd13f1ddbdc2b17fd461a14b6a0bd3 +size 1976163472 diff --git a/checkpoint-30/tokenizer.json b/checkpoint-30/tokenizer.json new file mode 100644 index 0000000..34510ff --- /dev/null +++ b/checkpoint-30/tokenizer.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3fd169731d2cbde95e10bf356d66d5997fd885dd8dbb6fb4684da3f23b2585d8 +size 11421892 diff --git a/checkpoint-30/tokenizer_config.json b/checkpoint-30/tokenizer_config.json new file mode 100644 index 0000000..4d8760d --- /dev/null +++ b/checkpoint-30/tokenizer_config.json @@ -0,0 +1,32 @@ +{ + "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|>", + "padding_side": "left", + "split_special_tokens": false, + "tokenizer_class": "Qwen2Tokenizer", + "truncation_side": "left", + "unk_token": null +} diff --git a/checkpoint-30/trainer_state.json b/checkpoint-30/trainer_state.json new file mode 100644 index 0000000..1857b44 --- /dev/null +++ b/checkpoint-30/trainer_state.json @@ -0,0 +1,844 @@ +{ + "best_global_step": null, + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 0.19736842105263158, + "eval_steps": 500, + "global_step": 30, + "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": 0.125, + "completions/max_length": 192.0, + "completions/max_terminated_length": 175.0, + "completions/mean_length": 135.25, + "completions/mean_terminated_length": 127.14286041259766, + "completions/min_length": 73.0, + "completions/min_terminated_length": 73.0, + "entropy": 0.7073924690485001, + "epoch": 0.006578947368421052, + "frac_reward_zero_std": 0.0, + "grad_norm": 11.996140480041504, + "learning_rate": 1e-06, + "loss": -0.16150084137916565, + "num_tokens": 2426.0, + "reward": -2.15625, + "reward_std": 0.5827506184577942, + "rewards/gol_tool_reward/mean": -2.15625, + "rewards/gol_tool_reward/std": 0.5827506184577942, + "step": 1, + "step_time": 5.788707780069672 + }, + { + "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": 172.0, + "completions/max_terminated_length": 172.0, + "completions/mean_length": 126.75, + "completions/mean_terminated_length": 126.75, + "completions/min_length": 101.0, + "completions/min_terminated_length": 101.0, + "entropy": 0.6544792205095291, + "epoch": 0.013157894736842105, + "frac_reward_zero_std": 0.0, + "grad_norm": 12.064927101135254, + "learning_rate": 9.666666666666666e-07, + "loss": -0.05266939848661423, + "num_tokens": 4784.0, + "reward": -2.28125, + "reward_std": 0.7454326152801514, + "rewards/gol_tool_reward/mean": -2.28125, + "rewards/gol_tool_reward/std": 0.7454325556755066, + "step": 2, + "step_time": 4.945590557064861 + }, + { + "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": 192.0, + "completions/max_terminated_length": 150.0, + "completions/mean_length": 141.125, + "completions/mean_terminated_length": 133.85714721679688, + "completions/min_length": 113.0, + "completions/min_terminated_length": 113.0, + "entropy": 0.7222911566495895, + "epoch": 0.019736842105263157, + "frac_reward_zero_std": 0.5, + "grad_norm": 8.669416427612305, + "learning_rate": 9.333333333333333e-07, + "loss": 0.061358727514743805, + "num_tokens": 7257.0, + "reward": -1.7750000953674316, + "reward_std": 0.6943651437759399, + "rewards/gol_tool_reward/mean": -1.7750000953674316, + "rewards/gol_tool_reward/std": 0.6943651437759399, + "step": 3, + "step_time": 5.346994869993068 + }, + { + "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": 192.0, + "completions/max_terminated_length": 178.0, + "completions/mean_length": 137.75, + "completions/mean_terminated_length": 130.0, + "completions/min_length": 101.0, + "completions/min_terminated_length": 101.0, + "entropy": 0.584687352180481, + "epoch": 0.02631578947368421, + "frac_reward_zero_std": 0.0, + "grad_norm": 11.648664474487305, + "learning_rate": 9e-07, + "loss": 0.0426417738199234, + "num_tokens": 9703.0, + "reward": -1.6624999046325684, + "reward_std": 0.4860555827617645, + "rewards/gol_tool_reward/mean": -1.6624999046325684, + "rewards/gol_tool_reward/std": 0.4860555827617645, + "step": 4, + "step_time": 5.309761939919554 + }, + { + "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": 192.0, + "completions/max_terminated_length": 164.0, + "completions/mean_length": 131.75, + "completions/mean_terminated_length": 123.14286041259766, + "completions/min_length": 83.0, + "completions/min_terminated_length": 83.0, + "entropy": 0.6980996131896973, + "epoch": 0.03289473684210526, + "frac_reward_zero_std": 0.0, + "grad_norm": 10.72757625579834, + "learning_rate": 8.666666666666667e-07, + "loss": 0.014450110495090485, + "num_tokens": 12101.0, + "reward": -1.8328126668930054, + "reward_std": 1.0132369995117188, + "rewards/gol_tool_reward/mean": -1.8328126668930054, + "rewards/gol_tool_reward/std": 1.0132369995117188, + "step": 5, + "step_time": 5.347554020117968 + }, + { + "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": 0.25, + "completions/max_length": 192.0, + "completions/max_terminated_length": 136.0, + "completions/mean_length": 123.75, + "completions/mean_terminated_length": 101.0, + "completions/min_length": 69.0, + "completions/min_terminated_length": 69.0, + "entropy": 0.5499220415949821, + "epoch": 0.039473684210526314, + "frac_reward_zero_std": 0.5, + "grad_norm": 9.949792861938477, + "learning_rate": 8.333333333333333e-07, + "loss": 0.14735683798789978, + "num_tokens": 14435.0, + "reward": -1.431249976158142, + "reward_std": 0.0883883461356163, + "rewards/gol_tool_reward/mean": -1.431249976158142, + "rewards/gol_tool_reward/std": 0.0883883461356163, + "step": 6, + "step_time": 5.282774350955151 + }, + { + "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": 0.25, + "completions/max_length": 192.0, + "completions/max_terminated_length": 141.0, + "completions/mean_length": 134.625, + "completions/mean_terminated_length": 115.5, + "completions/min_length": 86.0, + "completions/min_terminated_length": 86.0, + "entropy": 0.5586348995566368, + "epoch": 0.046052631578947366, + "frac_reward_zero_std": 0.5, + "grad_norm": 7.489627361297607, + "learning_rate": 8e-07, + "loss": 0.10210999846458435, + "num_tokens": 16856.0, + "reward": -1.7750000953674316, + "reward_std": 0.6943651437759399, + "rewards/gol_tool_reward/mean": -1.7750000953674316, + "rewards/gol_tool_reward/std": 0.6943651437759399, + "step": 7, + "step_time": 5.278500100132078 + }, + { + "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": 192.0, + "completions/max_terminated_length": 171.0, + "completions/mean_length": 146.375, + "completions/mean_terminated_length": 139.85714721679688, + "completions/min_length": 92.0, + "completions/min_terminated_length": 92.0, + "entropy": 0.6310850158333778, + "epoch": 0.05263157894736842, + "frac_reward_zero_std": 0.0, + "grad_norm": 10.776825904846191, + "learning_rate": 7.666666666666667e-07, + "loss": 0.019737035036087036, + "num_tokens": 19371.0, + "reward": -1.568750023841858, + "reward_std": 0.9996205568313599, + "rewards/gol_tool_reward/mean": -1.568750023841858, + "rewards/gol_tool_reward/std": 0.9996204972267151, + "step": 8, + "step_time": 5.483975553070195 + }, + { + "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": 192.0, + "completions/max_terminated_length": 175.0, + "completions/mean_length": 136.75, + "completions/mean_terminated_length": 128.85714721679688, + "completions/min_length": 102.0, + "completions/min_terminated_length": 102.0, + "entropy": 0.5831628292798996, + "epoch": 0.05921052631578947, + "frac_reward_zero_std": 0.5, + "grad_norm": 8.288241386413574, + "learning_rate": 7.333333333333332e-07, + "loss": 0.06289202719926834, + "num_tokens": 21809.0, + "reward": -1.537500023841858, + "reward_std": 0.5680480599403381, + "rewards/gol_tool_reward/mean": -1.537500023841858, + "rewards/gol_tool_reward/std": 0.5680480599403381, + "step": 9, + "step_time": 5.505063942982815 + }, + { + "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": 192.0, + "completions/max_terminated_length": 163.0, + "completions/mean_length": 126.25, + "completions/mean_terminated_length": 116.85714721679688, + "completions/min_length": 76.0, + "completions/min_terminated_length": 76.0, + "entropy": 0.4830511435866356, + "epoch": 0.06578947368421052, + "frac_reward_zero_std": 0.0, + "grad_norm": 11.556458473205566, + "learning_rate": 7e-07, + "loss": 0.0357169508934021, + "num_tokens": 24163.0, + "reward": -1.7000000476837158, + "reward_std": 0.769044041633606, + "rewards/gol_tool_reward/mean": -1.7000000476837158, + "rewards/gol_tool_reward/std": 0.7690439820289612, + "step": 10, + "step_time": 5.544649249990471 + }, + { + "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": 0.25, + "completions/max_length": 192.0, + "completions/max_terminated_length": 186.0, + "completions/mean_length": 165.0, + "completions/mean_terminated_length": 156.0, + "completions/min_length": 125.0, + "completions/min_terminated_length": 125.0, + "entropy": 0.5843875631690025, + "epoch": 0.07236842105263158, + "frac_reward_zero_std": 0.0, + "grad_norm": 12.836280822753906, + "learning_rate": 6.666666666666666e-07, + "loss": 0.047385718673467636, + "num_tokens": 26827.0, + "reward": -1.1484375, + "reward_std": 1.135318636894226, + "rewards/gol_tool_reward/mean": -1.1484375, + "rewards/gol_tool_reward/std": 1.135318636894226, + "step": 11, + "step_time": 5.533237581024878 + }, + { + "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": 192.0, + "completions/max_terminated_length": 164.0, + "completions/mean_length": 143.0, + "completions/mean_terminated_length": 136.0, + "completions/min_length": 104.0, + "completions/min_terminated_length": 104.0, + "entropy": 0.510637991130352, + "epoch": 0.07894736842105263, + "frac_reward_zero_std": 0.0, + "grad_norm": 10.66059684753418, + "learning_rate": 6.333333333333332e-07, + "loss": 0.08006405830383301, + "num_tokens": 29315.0, + "reward": -0.04531252384185791, + "reward_std": 1.7028460502624512, + "rewards/gol_tool_reward/mean": -0.04531252384185791, + "rewards/gol_tool_reward/std": 1.7028461694717407, + "step": 12, + "step_time": 5.5445326251210645 + }, + { + "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": 180.0, + "completions/max_terminated_length": 180.0, + "completions/mean_length": 134.0, + "completions/mean_terminated_length": 134.0, + "completions/min_length": 108.0, + "completions/min_terminated_length": 108.0, + "entropy": 0.548699863255024, + "epoch": 0.08552631578947369, + "frac_reward_zero_std": 0.0, + "grad_norm": 11.632511138916016, + "learning_rate": 6e-07, + "loss": -0.1205173134803772, + "num_tokens": 31731.0, + "reward": -0.0703125149011612, + "reward_std": 1.1454672813415527, + "rewards/gol_tool_reward/mean": -0.0703125149011612, + "rewards/gol_tool_reward/std": 1.1454674005508423, + "step": 13, + "step_time": 5.114862979855388 + }, + { + "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": 192.0, + "completions/max_terminated_length": 170.0, + "completions/mean_length": 130.25, + "completions/mean_terminated_length": 121.42857360839844, + "completions/min_length": 79.0, + "completions/min_terminated_length": 79.0, + "entropy": 0.595326341688633, + "epoch": 0.09210526315789473, + "frac_reward_zero_std": 0.0, + "grad_norm": 12.228357315063477, + "learning_rate": 5.666666666666666e-07, + "loss": 0.16490532457828522, + "num_tokens": 34117.0, + "reward": -0.22187499701976776, + "reward_std": 1.3627170324325562, + "rewards/gol_tool_reward/mean": -0.22187499701976776, + "rewards/gol_tool_reward/std": 1.3627170324325562, + "step": 14, + "step_time": 5.458060751087032 + }, + { + "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": 181.0, + "completions/max_terminated_length": 181.0, + "completions/mean_length": 123.125, + "completions/mean_terminated_length": 123.125, + "completions/min_length": 94.0, + "completions/min_terminated_length": 94.0, + "entropy": 0.5212945640087128, + "epoch": 0.09868421052631579, + "frac_reward_zero_std": 0.0, + "grad_norm": 11.923970222473145, + "learning_rate": 5.333333333333333e-07, + "loss": -0.03577335178852081, + "num_tokens": 36446.0, + "reward": -0.27812498807907104, + "reward_std": 1.2783524990081787, + "rewards/gol_tool_reward/mean": -0.27812498807907104, + "rewards/gol_tool_reward/std": 1.2783526182174683, + "step": 15, + "step_time": 5.0575181999010965 + }, + { + "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": 160.0, + "completions/max_terminated_length": 160.0, + "completions/mean_length": 118.375, + "completions/mean_terminated_length": 118.375, + "completions/min_length": 51.0, + "completions/min_terminated_length": 51.0, + "entropy": 0.5411187782883644, + "epoch": 0.10526315789473684, + "frac_reward_zero_std": 0.0, + "grad_norm": 12.17962646484375, + "learning_rate": 5e-07, + "loss": -0.07631053775548935, + "num_tokens": 38737.0, + "reward": -0.42812496423721313, + "reward_std": 0.8981565237045288, + "rewards/gol_tool_reward/mean": -0.42812496423721313, + "rewards/gol_tool_reward/std": 0.8981566429138184, + "step": 16, + "step_time": 4.665566484094597 + }, + { + "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": 192.0, + "completions/max_terminated_length": 166.0, + "completions/mean_length": 133.375, + "completions/mean_terminated_length": 125.00000762939453, + "completions/min_length": 91.0, + "completions/min_terminated_length": 91.0, + "entropy": 0.5414882004261017, + "epoch": 0.1118421052631579, + "frac_reward_zero_std": 0.0, + "grad_norm": 10.49337387084961, + "learning_rate": 4.6666666666666666e-07, + "loss": 0.016712799668312073, + "num_tokens": 41148.0, + "reward": -0.5531250238418579, + "reward_std": 0.7563643455505371, + "rewards/gol_tool_reward/mean": -0.5531250238418579, + "rewards/gol_tool_reward/std": 0.7563644051551819, + "step": 17, + "step_time": 5.364008491043933 + }, + { + "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": 146.0, + "completions/max_terminated_length": 146.0, + "completions/mean_length": 117.25, + "completions/mean_terminated_length": 117.25, + "completions/min_length": 86.0, + "completions/min_terminated_length": 86.0, + "entropy": 0.4803786799311638, + "epoch": 0.11842105263157894, + "frac_reward_zero_std": 0.0, + "grad_norm": 8.492094993591309, + "learning_rate": 4.3333333333333335e-07, + "loss": 0.002299964427947998, + "num_tokens": 43430.0, + "reward": 0.09843747317790985, + "reward_std": 0.9447796940803528, + "rewards/gol_tool_reward/mean": 0.09843747317790985, + "rewards/gol_tool_reward/std": 0.9447796940803528, + "step": 18, + "step_time": 4.548883846960962 + }, + { + "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": 144.0, + "completions/max_terminated_length": 144.0, + "completions/mean_length": 127.75, + "completions/mean_terminated_length": 127.75, + "completions/min_length": 102.0, + "completions/min_terminated_length": 102.0, + "entropy": 0.5627511814236641, + "epoch": 0.125, + "frac_reward_zero_std": 0.0, + "grad_norm": 9.863720893859863, + "learning_rate": 4e-07, + "loss": -0.03161659836769104, + "num_tokens": 45796.0, + "reward": -0.21875, + "reward_std": 0.3750595152378082, + "rewards/gol_tool_reward/mean": -0.21875, + "rewards/gol_tool_reward/std": 0.3750595152378082, + "step": 19, + "step_time": 4.288126179017127 + }, + { + "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": 192.0, + "completions/max_terminated_length": 147.0, + "completions/mean_length": 126.5, + "completions/mean_terminated_length": 117.14286041259766, + "completions/min_length": 92.0, + "completions/min_terminated_length": 92.0, + "entropy": 0.6194332987070084, + "epoch": 0.13157894736842105, + "frac_reward_zero_std": 0.0, + "grad_norm": 13.363960266113281, + "learning_rate": 3.666666666666666e-07, + "loss": 0.08973874151706696, + "num_tokens": 48152.0, + "reward": -0.7171875238418579, + "reward_std": 0.9436095952987671, + "rewards/gol_tool_reward/mean": -0.7171875238418579, + "rewards/gol_tool_reward/std": 0.9436095356941223, + "step": 20, + "step_time": 5.38997302914504 + }, + { + "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": 168.0, + "completions/max_terminated_length": 168.0, + "completions/mean_length": 128.375, + "completions/mean_terminated_length": 128.375, + "completions/min_length": 95.0, + "completions/min_terminated_length": 95.0, + "entropy": 0.5529855117201805, + "epoch": 0.13815789473684212, + "frac_reward_zero_std": 0.0, + "grad_norm": 11.634380340576172, + "learning_rate": 3.333333333333333e-07, + "loss": 0.003947541117668152, + "num_tokens": 50523.0, + "reward": 0.3453124761581421, + "reward_std": 1.174352765083313, + "rewards/gol_tool_reward/mean": 0.3453124761581421, + "rewards/gol_tool_reward/std": 1.174352765083313, + "step": 21, + "step_time": 4.907234247075394 + }, + { + "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": 192.0, + "completions/max_terminated_length": 138.0, + "completions/mean_length": 126.0, + "completions/mean_terminated_length": 116.5714340209961, + "completions/min_length": 97.0, + "completions/min_terminated_length": 97.0, + "entropy": 0.554837092757225, + "epoch": 0.14473684210526316, + "frac_reward_zero_std": 0.0, + "grad_norm": 11.619768142700195, + "learning_rate": 3e-07, + "loss": 0.07526145875453949, + "num_tokens": 52875.0, + "reward": -0.6468750238418579, + "reward_std": 0.6682115793228149, + "rewards/gol_tool_reward/mean": -0.6468750238418579, + "rewards/gol_tool_reward/std": 0.6682115197181702, + "step": 22, + "step_time": 5.336878409842029 + }, + { + "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": 160.0, + "completions/max_terminated_length": 160.0, + "completions/mean_length": 131.25, + "completions/mean_terminated_length": 131.25, + "completions/min_length": 118.0, + "completions/min_terminated_length": 118.0, + "entropy": 0.5712674260139465, + "epoch": 0.1513157894736842, + "frac_reward_zero_std": 0.0, + "grad_norm": 10.13196849822998, + "learning_rate": 2.6666666666666667e-07, + "loss": -0.0519290491938591, + "num_tokens": 55269.0, + "reward": -0.23125000298023224, + "reward_std": 0.3797908425331116, + "rewards/gol_tool_reward/mean": -0.23125000298023224, + "rewards/gol_tool_reward/std": 0.3797908127307892, + "step": 23, + "step_time": 4.688064404879697 + }, + { + "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": 185.0, + "completions/max_terminated_length": 185.0, + "completions/mean_length": 119.125, + "completions/mean_terminated_length": 119.125, + "completions/min_length": 95.0, + "completions/min_terminated_length": 95.0, + "entropy": 0.5097240433096886, + "epoch": 0.15789473684210525, + "frac_reward_zero_std": 0.0, + "grad_norm": 9.0028076171875, + "learning_rate": 2.3333333333333333e-07, + "loss": -0.05878718942403793, + "num_tokens": 57566.0, + "reward": -0.02656250074505806, + "reward_std": 0.5322810411453247, + "rewards/gol_tool_reward/mean": -0.02656250074505806, + "rewards/gol_tool_reward/std": 0.5322810411453247, + "step": 24, + "step_time": 5.215061811963096 + }, + { + "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": 182.0, + "completions/max_terminated_length": 182.0, + "completions/mean_length": 120.5, + "completions/mean_terminated_length": 120.5, + "completions/min_length": 95.0, + "completions/min_terminated_length": 95.0, + "entropy": 0.5010545030236244, + "epoch": 0.16447368421052633, + "frac_reward_zero_std": 0.0, + "grad_norm": 10.190753936767578, + "learning_rate": 2e-07, + "loss": -0.00694732740521431, + "num_tokens": 59874.0, + "reward": 0.265625, + "reward_std": 1.212799072265625, + "rewards/gol_tool_reward/mean": 0.265625, + "rewards/gol_tool_reward/std": 1.2127991914749146, + "step": 25, + "step_time": 5.231085888110101 + }, + { + "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": 176.0, + "completions/max_terminated_length": 176.0, + "completions/mean_length": 126.375, + "completions/mean_terminated_length": 126.375, + "completions/min_length": 88.0, + "completions/min_terminated_length": 88.0, + "entropy": 0.5022949725389481, + "epoch": 0.17105263157894737, + "frac_reward_zero_std": 0.0, + "grad_norm": 11.172710418701172, + "learning_rate": 1.6666666666666665e-07, + "loss": -0.02744058519601822, + "num_tokens": 62229.0, + "reward": -0.359375, + "reward_std": 0.05966797471046448, + "rewards/gol_tool_reward/mean": -0.359375, + "rewards/gol_tool_reward/std": 0.05966798588633537, + "step": 26, + "step_time": 4.989802666124888 + }, + { + "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": 168.0, + "completions/max_terminated_length": 168.0, + "completions/mean_length": 123.75, + "completions/mean_terminated_length": 123.75, + "completions/min_length": 94.0, + "completions/min_terminated_length": 94.0, + "entropy": 0.5283998399972916, + "epoch": 0.17763157894736842, + "frac_reward_zero_std": 0.0, + "grad_norm": 9.951343536376953, + "learning_rate": 1.3333333333333334e-07, + "loss": 0.0610831081867218, + "num_tokens": 64563.0, + "reward": -0.0078125, + "reward_std": 1.0492969751358032, + "rewards/gol_tool_reward/mean": -0.0078125, + "rewards/gol_tool_reward/std": 1.0492969751358032, + "step": 27, + "step_time": 4.680415777955204 + }, + { + "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": 173.0, + "completions/max_terminated_length": 173.0, + "completions/mean_length": 118.875, + "completions/mean_terminated_length": 118.875, + "completions/min_length": 86.0, + "completions/min_terminated_length": 86.0, + "entropy": 0.4742962121963501, + "epoch": 0.18421052631578946, + "frac_reward_zero_std": 0.0, + "grad_norm": 12.151405334472656, + "learning_rate": 1e-07, + "loss": 0.058089837431907654, + "num_tokens": 66858.0, + "reward": -0.42031246423721313, + "reward_std": 0.11199040710926056, + "rewards/gol_tool_reward/mean": -0.42031246423721313, + "rewards/gol_tool_reward/std": 0.11199040710926056, + "step": 28, + "step_time": 4.9447458980139345 + }, + { + "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": 160.0, + "completions/max_terminated_length": 160.0, + "completions/mean_length": 134.125, + "completions/mean_terminated_length": 134.125, + "completions/min_length": 98.0, + "completions/min_terminated_length": 98.0, + "entropy": 0.5360276401042938, + "epoch": 0.19078947368421054, + "frac_reward_zero_std": 0.0, + "grad_norm": 10.772101402282715, + "learning_rate": 6.666666666666667e-08, + "loss": -0.08468331396579742, + "num_tokens": 69275.0, + "reward": 0.04843750596046448, + "reward_std": 0.7822938561439514, + "rewards/gol_tool_reward/mean": 0.04843750596046448, + "rewards/gol_tool_reward/std": 0.7822939157485962, + "step": 29, + "step_time": 4.704459578031674 + }, + { + "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": 146.0, + "completions/max_terminated_length": 146.0, + "completions/mean_length": 124.5, + "completions/mean_terminated_length": 124.5, + "completions/min_length": 107.0, + "completions/min_terminated_length": 107.0, + "entropy": 0.5363342761993408, + "epoch": 0.19736842105263158, + "frac_reward_zero_std": 0.0, + "grad_norm": 10.216547966003418, + "learning_rate": 3.3333333333333334e-08, + "loss": -0.06437680870294571, + "num_tokens": 71615.0, + "reward": -0.2890625, + "reward_std": 0.22573663294315338, + "rewards/gol_tool_reward/mean": -0.2890625, + "rewards/gol_tool_reward/std": 0.22573664784431458, + "step": 30, + "step_time": 4.333932535140775 + } + ], + "logging_steps": 1, + "max_steps": 30, + "num_input_tokens_seen": 71615, + "num_train_epochs": 1, + "save_steps": 30, + "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": 2, + "trial_name": null, + "trial_params": null +} diff --git a/checkpoint-30/training_args.bin b/checkpoint-30/training_args.bin new file mode 100644 index 0000000..9ada855 --- /dev/null +++ b/checkpoint-30/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c3c6fa49e7b8309ff6c0b5302fb4d29d3f1a49ce05cd69ad53d549e0885b0897 +size 7185 diff --git a/config.json b/config.json new file mode 100644 index 0000000..f62c429 --- /dev/null +++ b/config.json @@ -0,0 +1,57 @@ +{ + "architectures": [ + "Qwen2ForCausalLM" + ], + "attention_dropout": 0.0, + "bos_token_id": null, + "dtype": "float32", + "eos_token_id": 151645, + "hidden_act": "silu", + "hidden_size": 896, + "initializer_range": 0.02, + "intermediate_size": 4864, + "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" + ], + "max_position_embeddings": 32768, + "max_window_layers": 21, + "model_type": "qwen2", + "num_attention_heads": 14, + "num_hidden_layers": 24, + "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": "5.9.0", + "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..21d3d9d --- /dev/null +++ b/generation_config.json @@ -0,0 +1,13 @@ +{ + "do_sample": true, + "eos_token_id": [ + 151645, + 151643 + ], + "pad_token_id": 151643, + "repetition_penalty": 1.1, + "temperature": 0.7, + "top_k": 20, + "top_p": 0.8, + "transformers_version": "5.9.0" +} diff --git a/model.safetensors b/model.safetensors new file mode 100644 index 0000000..4535412 --- /dev/null +++ b/model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:920b4e656f4ea47583dd538101d84eddb6cd13f1ddbdc2b17fd461a14b6a0bd3 +size 1976163472 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..4d8760d --- /dev/null +++ b/tokenizer_config.json @@ -0,0 +1,32 @@ +{ + "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|>", + "padding_side": "left", + "split_special_tokens": false, + "tokenizer_class": "Qwen2Tokenizer", + "truncation_side": "left", + "unk_token": null +} diff --git a/training_args.bin b/training_args.bin new file mode 100644 index 0000000..9ada855 --- /dev/null +++ b/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c3c6fa49e7b8309ff6c0b5302fb4d29d3f1a49ce05cd69ad53d549e0885b0897 +size 7185