diff --git a/.gitattributes b/.gitattributes index a6344aac8c09253b3b630fb776ae94478aa0275b..c837427826e96b969748296d09b8d4f9b05d9742 100644 --- a/.gitattributes +++ b/.gitattributes @@ -33,3 +33,9 @@ saved_model/**/* 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 +lora/sft-captcha/checkpoint-400/tokenizer.json filter=lfs diff=lfs merge=lfs -text +lora/sft-captcha/checkpoint-500/tokenizer.json filter=lfs diff=lfs merge=lfs -text +lora/sft-captcha/checkpoint-600/tokenizer.json filter=lfs diff=lfs merge=lfs -text +lora/sft-captcha/checkpoint-700/tokenizer.json filter=lfs diff=lfs merge=lfs -text +lora/sft-captcha/checkpoint-775/tokenizer.json filter=lfs diff=lfs merge=lfs -text +lora/sft-captcha/tokenizer.json filter=lfs diff=lfs merge=lfs -text diff --git a/lora/sft-captcha/README.md b/lora/sft-captcha/README.md new file mode 100644 index 0000000000000000000000000000000000000000..fbcd86d0dfea0c0f0e8a71715bea35b6a43940ec --- /dev/null +++ b/lora/sft-captcha/README.md @@ -0,0 +1,62 @@ +--- +library_name: peft +license: other +base_model: Qwen/Qwen3-VL-4B-Instruct +tags: +- base_model:adapter:Qwen/Qwen3-VL-4B-Instruct +- llama-factory +- lora +- transformers +pipeline_tag: text-generation +model-index: +- name: sft-captcha + results: [] +--- + + + +# sft-captcha + +This model is a fine-tuned version of [Qwen/Qwen3-VL-4B-Instruct](https://huggingface.co/Qwen/Qwen3-VL-4B-Instruct) on the new_dataset dataset. + +## Model description + +More information needed + +## Intended uses & limitations + +More information needed + +## Training and evaluation data + +More information needed + +## Training procedure + +### Training hyperparameters + +The following hyperparameters were used during training: +- learning_rate: 5e-05 +- train_batch_size: 4 +- eval_batch_size: 8 +- seed: 42 +- gradient_accumulation_steps: 8 +- total_train_batch_size: 32 +- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments +- lr_scheduler_type: cosine +- lr_scheduler_warmup_steps: 0.1 +- num_epochs: 5 +- mixed_precision_training: Native AMP + +### Training results + + + +### Framework versions + +- PEFT 0.18.1 +- Transformers 5.6.0 +- Pytorch 2.5.1+cu121 +- Datasets 4.0.0 +- Tokenizers 0.22.2 \ No newline at end of file diff --git a/lora/sft-captcha/adapter_config.json b/lora/sft-captcha/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..91f8411a80c3d6f80ff5a9f2d6d5c6e0952c8e01 --- /dev/null +++ b/lora/sft-captcha/adapter_config.json @@ -0,0 +1,46 @@ +{ + "alora_invocation_tokens": null, + "alpha_pattern": {}, + "arrow_config": null, + "auto_mapping": null, + "base_model_name_or_path": "Qwen/Qwen3-VL-4B-Instruct", + "bias": "none", + "corda_config": null, + "ensure_weight_tying": false, + "eva_config": null, + "exclude_modules": null, + "fan_in_fan_out": false, + "inference_mode": true, + "init_lora_weights": true, + "layer_replication": null, + "layers_pattern": null, + "layers_to_transform": null, + "loftq_config": {}, + "lora_alpha": 64, + "lora_bias": false, + "lora_dropout": 0.05, + "megatron_config": null, + "megatron_core": "megatron.core", + "modules_to_save": null, + "peft_type": "LORA", + "peft_version": "0.18.1", + "qalora_group_size": 16, + "r": 32, + "rank_pattern": {}, + "revision": null, + "target_modules": [ + "k_proj", + "v_proj", + "gate_proj", + "up_proj", + "o_proj", + "down_proj", + "q_proj" + ], + "target_parameters": null, + "task_type": "CAUSAL_LM", + "trainable_token_indices": null, + "use_dora": false, + "use_qalora": false, + "use_rslora": false +} \ No newline at end of file diff --git a/lora/sft-captcha/adapter_model.safetensors b/lora/sft-captcha/adapter_model.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..5177c7f9d30cf56dbef24383fbfe0da65a7aa807 --- /dev/null +++ b/lora/sft-captcha/adapter_model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:fbbc568b3b10ea6e601db64bce50e3cdd551083e7288e2c8dd9d2d4220605219 +size 264316456 diff --git a/lora/sft-captcha/all_results.json b/lora/sft-captcha/all_results.json new file mode 100644 index 0000000000000000000000000000000000000000..d4dc19bd6c47cfa49c4f3b5e91ac214ed06d61de --- /dev/null +++ b/lora/sft-captcha/all_results.json @@ -0,0 +1,8 @@ +{ + "epoch": 5.0, + "total_flos": 7.509104074275226e+17, + "train_loss": 0.31352297429115544, + "train_runtime": 25009.3076, + "train_samples_per_second": 0.99, + "train_steps_per_second": 0.031 +} \ No newline at end of file diff --git a/lora/sft-captcha/chat_template.jinja b/lora/sft-captcha/chat_template.jinja new file mode 100644 index 0000000000000000000000000000000000000000..124386803f142761528f710e77ae483f5f8c4fc4 --- /dev/null +++ b/lora/sft-captcha/chat_template.jinja @@ -0,0 +1,120 @@ +{%- if tools %} + {{- '<|im_start|>system\n' }} + {%- if messages[0].role == 'system' %} + {%- if messages[0].content is string %} + {{- messages[0].content }} + {%- else %} + {%- for content in messages[0].content %} + {%- if 'text' in content %} + {{- content.text }} + {%- endif %} + {%- endfor %} + {%- endif %} + {{- '\n\n' }} + {%- endif %} + {{- "# 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' }} + {%- if messages[0].content is string %} + {{- messages[0].content }} + {%- else %} + {%- for content in messages[0].content %} + {%- if 'text' in content %} + {{- content.text }} + {%- endif %} + {%- endfor %} + {%- endif %} + {{- '<|im_end|>\n' }} + {%- endif %} +{%- endif %} +{%- set image_count = namespace(value=0) %} +{%- set video_count = namespace(value=0) %} +{%- for message in messages %} + {%- if message.role == "user" %} + {{- '<|im_start|>' + message.role + '\n' }} + {%- if message.content is string %} + {{- message.content }} + {%- else %} + {%- for content in message.content %} + {%- if content.type == 'image' or 'image' in content or 'image_url' in content %} + {%- set image_count.value = image_count.value + 1 %} + {%- if add_vision_id %}Picture {{ image_count.value }}: {% endif -%} + <|vision_start|><|image_pad|><|vision_end|> + {%- elif content.type == 'video' or 'video' in content %} + {%- set video_count.value = video_count.value + 1 %} + {%- if add_vision_id %}Video {{ video_count.value }}: {% endif -%} + <|vision_start|><|video_pad|><|vision_end|> + {%- elif 'text' in content %} + {{- content.text }} + {%- endif %} + {%- endfor %} + {%- endif %} + {{- '<|im_end|>\n' }} + {%- elif message.role == "assistant" %} + {{- '<|im_start|>' + message.role + '\n' }} + {%- if message.content is string %} + {{- message.content }} + {%- else %} + {%- for content_item in message.content %} + {%- if 'text' in content_item %} + {{- content_item.text }} + {%- endif %} + {%- endfor %} + {%- endif %} + {%- if message.tool_calls %} + {%- for tool_call in message.tool_calls %} + {%- if (loop.first and message.content) or (not loop.first) %} + {{- '\n' }} + {%- endif %} + {%- if tool_call.function %} + {%- set tool_call = tool_call.function %} + {%- endif %} + {{- '\n{"name": "' }} + {{- tool_call.name }} + {{- '", "arguments": ' }} + {%- if tool_call.arguments is string %} + {{- tool_call.arguments }} + {%- else %} + {{- tool_call.arguments | tojson }} + {%- endif %} + {{- '}\n' }} + {%- endfor %} + {%- endif %} + {{- '<|im_end|>\n' }} + {%- elif message.role == "tool" %} + {%- if loop.first or (messages[loop.index0 - 1].role != "tool") %} + {{- '<|im_start|>user' }} + {%- endif %} + {{- '\n\n' }} + {%- if message.content is string %} + {{- message.content }} + {%- else %} + {%- for content in message.content %} + {%- if content.type == 'image' or 'image' in content or 'image_url' in content %} + {%- set image_count.value = image_count.value + 1 %} + {%- if add_vision_id %}Picture {{ image_count.value }}: {% endif -%} + <|vision_start|><|image_pad|><|vision_end|> + {%- elif content.type == 'video' or 'video' in content %} + {%- set video_count.value = video_count.value + 1 %} + {%- if add_vision_id %}Video {{ video_count.value }}: {% endif -%} + <|vision_start|><|video_pad|><|vision_end|> + {%- elif 'text' in content %} + {{- content.text }} + {%- endif %} + {%- endfor %} + {%- endif %} + {{- '\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/lora/sft-captcha/checkpoint-400/README.md b/lora/sft-captcha/checkpoint-400/README.md new file mode 100644 index 0000000000000000000000000000000000000000..776a59eddf85aa2890ebb734de124c8a090269b6 --- /dev/null +++ b/lora/sft-captcha/checkpoint-400/README.md @@ -0,0 +1,208 @@ +--- +base_model: Qwen/Qwen3-VL-4B-Instruct +library_name: peft +pipeline_tag: text-generation +tags: +- base_model:adapter:Qwen/Qwen3-VL-4B-Instruct +- llama-factory +- lora +- transformers +--- + +# Model Card for Model ID + + + + + +## Model Details + +### Model Description + + + + + +- **Developed by:** [More Information Needed] +- **Funded by [optional]:** [More Information Needed] +- **Shared by [optional]:** [More Information Needed] +- **Model type:** [More Information Needed] +- **Language(s) (NLP):** [More Information Needed] +- **License:** [More Information Needed] +- **Finetuned from model [optional]:** [More Information Needed] + +### Model Sources [optional] + + + +- **Repository:** [More Information Needed] +- **Paper [optional]:** [More Information Needed] +- **Demo [optional]:** [More Information Needed] + +## Uses + + + +### Direct Use + + + +[More Information Needed] + +### Downstream Use [optional] + + + +[More Information Needed] + +### Out-of-Scope Use + + + +[More Information Needed] + +## Bias, Risks, and Limitations + + + +[More Information Needed] + +### Recommendations + + + +Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations. + +## How to Get Started with the Model + +Use the code below to get started with the model. + +[More Information Needed] + +## Training Details + +### Training Data + + + +[More Information Needed] + +### Training Procedure + + + +#### Preprocessing [optional] + +[More Information Needed] + + +#### Training Hyperparameters + +- **Training regime:** [More Information Needed] + +#### Speeds, Sizes, Times [optional] + + + +[More Information Needed] + +## Evaluation + + + +### Testing Data, Factors & Metrics + +#### Testing Data + + + +[More Information Needed] + +#### Factors + + + +[More Information Needed] + +#### Metrics + + + +[More Information Needed] + +### Results + +[More Information Needed] + +#### Summary + + + +## Model Examination [optional] + + + +[More Information Needed] + +## Environmental Impact + + + +Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700). + +- **Hardware Type:** [More Information Needed] +- **Hours used:** [More Information Needed] +- **Cloud Provider:** [More Information Needed] +- **Compute Region:** [More Information Needed] +- **Carbon Emitted:** [More Information Needed] + +## Technical Specifications [optional] + +### Model Architecture and Objective + +[More Information Needed] + +### Compute Infrastructure + +[More Information Needed] + +#### Hardware + +[More Information Needed] + +#### Software + +[More Information Needed] + +## Citation [optional] + + + +**BibTeX:** + +[More Information Needed] + +**APA:** + +[More Information Needed] + +## Glossary [optional] + + + +[More Information Needed] + +## More Information [optional] + +[More Information Needed] + +## Model Card Authors [optional] + +[More Information Needed] + +## Model Card Contact + +[More Information Needed] +### Framework versions + +- PEFT 0.18.1 \ No newline at end of file diff --git a/lora/sft-captcha/checkpoint-400/adapter_config.json b/lora/sft-captcha/checkpoint-400/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..91f8411a80c3d6f80ff5a9f2d6d5c6e0952c8e01 --- /dev/null +++ b/lora/sft-captcha/checkpoint-400/adapter_config.json @@ -0,0 +1,46 @@ +{ + "alora_invocation_tokens": null, + "alpha_pattern": {}, + "arrow_config": null, + "auto_mapping": null, + "base_model_name_or_path": "Qwen/Qwen3-VL-4B-Instruct", + "bias": "none", + "corda_config": null, + "ensure_weight_tying": false, + "eva_config": null, + "exclude_modules": null, + "fan_in_fan_out": false, + "inference_mode": true, + "init_lora_weights": true, + "layer_replication": null, + "layers_pattern": null, + "layers_to_transform": null, + "loftq_config": {}, + "lora_alpha": 64, + "lora_bias": false, + "lora_dropout": 0.05, + "megatron_config": null, + "megatron_core": "megatron.core", + "modules_to_save": null, + "peft_type": "LORA", + "peft_version": "0.18.1", + "qalora_group_size": 16, + "r": 32, + "rank_pattern": {}, + "revision": null, + "target_modules": [ + "k_proj", + "v_proj", + "gate_proj", + "up_proj", + "o_proj", + "down_proj", + "q_proj" + ], + "target_parameters": null, + "task_type": "CAUSAL_LM", + "trainable_token_indices": null, + "use_dora": false, + "use_qalora": false, + "use_rslora": false +} \ No newline at end of file diff --git a/lora/sft-captcha/checkpoint-400/adapter_model.safetensors b/lora/sft-captcha/checkpoint-400/adapter_model.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..7e51c8945b0e004cd93c3413e209c6a249c87d18 --- /dev/null +++ b/lora/sft-captcha/checkpoint-400/adapter_model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e66eeeee1ebad3f84ed03697a7cea74e5b0223bb59941d542f5bdc61e448335a +size 264316456 diff --git a/lora/sft-captcha/checkpoint-400/chat_template.jinja b/lora/sft-captcha/checkpoint-400/chat_template.jinja new file mode 100644 index 0000000000000000000000000000000000000000..124386803f142761528f710e77ae483f5f8c4fc4 --- /dev/null +++ b/lora/sft-captcha/checkpoint-400/chat_template.jinja @@ -0,0 +1,120 @@ +{%- if tools %} + {{- '<|im_start|>system\n' }} + {%- if messages[0].role == 'system' %} + {%- if messages[0].content is string %} + {{- messages[0].content }} + {%- else %} + {%- for content in messages[0].content %} + {%- if 'text' in content %} + {{- content.text }} + {%- endif %} + {%- endfor %} + {%- endif %} + {{- '\n\n' }} + {%- endif %} + {{- "# 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' }} + {%- if messages[0].content is string %} + {{- messages[0].content }} + {%- else %} + {%- for content in messages[0].content %} + {%- if 'text' in content %} + {{- content.text }} + {%- endif %} + {%- endfor %} + {%- endif %} + {{- '<|im_end|>\n' }} + {%- endif %} +{%- endif %} +{%- set image_count = namespace(value=0) %} +{%- set video_count = namespace(value=0) %} +{%- for message in messages %} + {%- if message.role == "user" %} + {{- '<|im_start|>' + message.role + '\n' }} + {%- if message.content is string %} + {{- message.content }} + {%- else %} + {%- for content in message.content %} + {%- if content.type == 'image' or 'image' in content or 'image_url' in content %} + {%- set image_count.value = image_count.value + 1 %} + {%- if add_vision_id %}Picture {{ image_count.value }}: {% endif -%} + <|vision_start|><|image_pad|><|vision_end|> + {%- elif content.type == 'video' or 'video' in content %} + {%- set video_count.value = video_count.value + 1 %} + {%- if add_vision_id %}Video {{ video_count.value }}: {% endif -%} + <|vision_start|><|video_pad|><|vision_end|> + {%- elif 'text' in content %} + {{- content.text }} + {%- endif %} + {%- endfor %} + {%- endif %} + {{- '<|im_end|>\n' }} + {%- elif message.role == "assistant" %} + {{- '<|im_start|>' + message.role + '\n' }} + {%- if message.content is string %} + {{- message.content }} + {%- else %} + {%- for content_item in message.content %} + {%- if 'text' in content_item %} + {{- content_item.text }} + {%- endif %} + {%- endfor %} + {%- endif %} + {%- if message.tool_calls %} + {%- for tool_call in message.tool_calls %} + {%- if (loop.first and message.content) or (not loop.first) %} + {{- '\n' }} + {%- endif %} + {%- if tool_call.function %} + {%- set tool_call = tool_call.function %} + {%- endif %} + {{- '\n{"name": "' }} + {{- tool_call.name }} + {{- '", "arguments": ' }} + {%- if tool_call.arguments is string %} + {{- tool_call.arguments }} + {%- else %} + {{- tool_call.arguments | tojson }} + {%- endif %} + {{- '}\n' }} + {%- endfor %} + {%- endif %} + {{- '<|im_end|>\n' }} + {%- elif message.role == "tool" %} + {%- if loop.first or (messages[loop.index0 - 1].role != "tool") %} + {{- '<|im_start|>user' }} + {%- endif %} + {{- '\n\n' }} + {%- if message.content is string %} + {{- message.content }} + {%- else %} + {%- for content in message.content %} + {%- if content.type == 'image' or 'image' in content or 'image_url' in content %} + {%- set image_count.value = image_count.value + 1 %} + {%- if add_vision_id %}Picture {{ image_count.value }}: {% endif -%} + <|vision_start|><|image_pad|><|vision_end|> + {%- elif content.type == 'video' or 'video' in content %} + {%- set video_count.value = video_count.value + 1 %} + {%- if add_vision_id %}Video {{ video_count.value }}: {% endif -%} + <|vision_start|><|video_pad|><|vision_end|> + {%- elif 'text' in content %} + {{- content.text }} + {%- endif %} + {%- endfor %} + {%- endif %} + {{- '\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/lora/sft-captcha/checkpoint-400/optimizer.pt b/lora/sft-captcha/checkpoint-400/optimizer.pt new file mode 100644 index 0000000000000000000000000000000000000000..85495e9a2f81f5b8642843114e9164eac97db704 --- /dev/null +++ b/lora/sft-captcha/checkpoint-400/optimizer.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:49df9d74c32cb35aab23ecd0094e6250cd5a11c35a06e201cb5136a52fd041ca +size 528907898 diff --git a/lora/sft-captcha/checkpoint-400/processor_config.json b/lora/sft-captcha/checkpoint-400/processor_config.json new file mode 100644 index 0000000000000000000000000000000000000000..33818c7f9e991ad735fd240209f4fa73e6c28c50 --- /dev/null +++ b/lora/sft-captcha/checkpoint-400/processor_config.json @@ -0,0 +1,60 @@ +{ + "image_processor": { + "do_convert_rgb": true, + "do_normalize": true, + "do_rescale": true, + "do_resize": true, + "image_mean": [ + 0.5, + 0.5, + 0.5 + ], + "image_processor_type": "Qwen2VLImageProcessor", + "image_std": [ + 0.5, + 0.5, + 0.5 + ], + "merge_size": 2, + "patch_size": 16, + "resample": 3, + "rescale_factor": 0.00392156862745098, + "size": { + "longest_edge": 16777216, + "shortest_edge": 65536 + }, + "temporal_patch_size": 2 + }, + "processor_class": "Qwen3VLProcessor", + "video_processor": { + "do_convert_rgb": true, + "do_normalize": true, + "do_rescale": true, + "do_resize": true, + "do_sample_frames": true, + "fps": 2, + "image_mean": [ + 0.5, + 0.5, + 0.5 + ], + "image_std": [ + 0.5, + 0.5, + 0.5 + ], + "max_frames": 768, + "merge_size": 2, + "min_frames": 4, + "patch_size": 16, + "resample": 3, + "rescale_factor": 0.00392156862745098, + "return_metadata": false, + "size": { + "longest_edge": 25165824, + "shortest_edge": 4096 + }, + "temporal_patch_size": 2, + "video_processor_type": "Qwen3VLVideoProcessor" + } +} diff --git a/lora/sft-captcha/checkpoint-400/rng_state.pth b/lora/sft-captcha/checkpoint-400/rng_state.pth new file mode 100644 index 0000000000000000000000000000000000000000..2a7e5810223832eda86d6e29e8127bd5152a44f3 --- /dev/null +++ b/lora/sft-captcha/checkpoint-400/rng_state.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:475fd467f417f0259650e7c80bd36d7eec27e4769fb5d99ba3ebaf101c351054 +size 14244 diff --git a/lora/sft-captcha/checkpoint-400/scaler.pt b/lora/sft-captcha/checkpoint-400/scaler.pt new file mode 100644 index 0000000000000000000000000000000000000000..834fccb9e03a3266c967e2aa2ac0269bf2253fcc --- /dev/null +++ b/lora/sft-captcha/checkpoint-400/scaler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c801982aae9be06d302403c1fff693e53dedf89c1d3b689ee29fedad84a96d23 +size 988 diff --git a/lora/sft-captcha/checkpoint-400/scheduler.pt b/lora/sft-captcha/checkpoint-400/scheduler.pt new file mode 100644 index 0000000000000000000000000000000000000000..18dba97ef8b4e1cb3d1486588b5d0dd31fd148f6 --- /dev/null +++ b/lora/sft-captcha/checkpoint-400/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0f6555b29b300370c41c8532c7d7e8f3e8a569486e0856a98f852e0aafd33af3 +size 1064 diff --git a/lora/sft-captcha/checkpoint-400/tokenizer.json b/lora/sft-captcha/checkpoint-400/tokenizer.json new file mode 100644 index 0000000000000000000000000000000000000000..c7afbed2efcdf019f88ab0572ec29d3bf595dfe2 --- /dev/null +++ b/lora/sft-captcha/checkpoint-400/tokenizer.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:be75606093db2094d7cd20f3c2f385c212750648bd6ea4fb2bf507a6a4c55506 +size 11422650 diff --git a/lora/sft-captcha/checkpoint-400/tokenizer_config.json b/lora/sft-captcha/checkpoint-400/tokenizer_config.json new file mode 100644 index 0000000000000000000000000000000000000000..19febf221cc1a71c2b17feb1d29923a875d42d6e --- /dev/null +++ b/lora/sft-captcha/checkpoint-400/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": 262144, + "pad_token": "<|endoftext|>", + "padding_side": "right", + "processor_class": "Qwen3VLProcessor", + "split_special_tokens": false, + "tokenizer_class": "Qwen2Tokenizer", + "unk_token": null +} diff --git a/lora/sft-captcha/checkpoint-400/trainer_state.json b/lora/sft-captcha/checkpoint-400/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..47f98675626c81a4194370898007a41dcee48344 --- /dev/null +++ b/lora/sft-captcha/checkpoint-400/trainer_state.json @@ -0,0 +1,314 @@ +{ + "best_global_step": null, + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 2.581583198707593, + "eval_steps": 500, + "global_step": 400, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.06462035541195477, + "grad_norm": 2.068004846572876, + "learning_rate": 5.76923076923077e-06, + "loss": 0.9769228935241699, + "step": 10 + }, + { + "epoch": 0.12924071082390953, + "grad_norm": 0.9047558307647705, + "learning_rate": 1.217948717948718e-05, + "loss": 0.7594649791717529, + "step": 20 + }, + { + "epoch": 0.1938610662358643, + "grad_norm": 0.4503297805786133, + "learning_rate": 1.858974358974359e-05, + "loss": 0.5557493209838867, + "step": 30 + }, + { + "epoch": 0.25848142164781907, + "grad_norm": 0.37129801511764526, + "learning_rate": 2.5e-05, + "loss": 0.4763153076171875, + "step": 40 + }, + { + "epoch": 0.32310177705977383, + "grad_norm": 0.2953853905200958, + "learning_rate": 3.141025641025641e-05, + "loss": 0.4166147708892822, + "step": 50 + }, + { + "epoch": 0.3877221324717286, + "grad_norm": 0.29160749912261963, + "learning_rate": 3.782051282051282e-05, + "loss": 0.3905663013458252, + "step": 60 + }, + { + "epoch": 0.45234248788368336, + "grad_norm": 0.27939939498901367, + "learning_rate": 4.423076923076923e-05, + "loss": 0.3649837732315063, + "step": 70 + }, + { + "epoch": 0.5169628432956381, + "grad_norm": 0.3130805492401123, + "learning_rate": 4.999974605277764e-05, + "loss": 0.3609442949295044, + "step": 80 + }, + { + "epoch": 0.5815831987075929, + "grad_norm": 0.3720378577709198, + "learning_rate": 4.996927862815294e-05, + "loss": 0.3543627977371216, + "step": 90 + }, + { + "epoch": 0.6462035541195477, + "grad_norm": 0.2810651361942291, + "learning_rate": 4.988809267347141e-05, + "loss": 0.345255446434021, + "step": 100 + }, + { + "epoch": 0.7108239095315024, + "grad_norm": 0.3126155734062195, + "learning_rate": 4.975635309667227e-05, + "loss": 0.33807756900787356, + "step": 110 + }, + { + "epoch": 0.7754442649434572, + "grad_norm": 0.3109033703804016, + "learning_rate": 4.957432749209755e-05, + "loss": 0.3394132614135742, + "step": 120 + }, + { + "epoch": 0.840064620355412, + "grad_norm": 0.3006167411804199, + "learning_rate": 4.934238559694448e-05, + "loss": 0.3244359016418457, + "step": 130 + }, + { + "epoch": 0.9046849757673667, + "grad_norm": 0.3104219138622284, + "learning_rate": 4.90609985402415e-05, + "loss": 0.3221057176589966, + "step": 140 + }, + { + "epoch": 0.9693053311793215, + "grad_norm": 0.330458402633667, + "learning_rate": 4.873073788587338e-05, + "loss": 0.3329624176025391, + "step": 150 + }, + { + "epoch": 1.0323101777059773, + "grad_norm": 0.2666167616844177, + "learning_rate": 4.835227447159939e-05, + "loss": 0.3191133737564087, + "step": 160 + }, + { + "epoch": 1.0969305331179322, + "grad_norm": 0.3217558264732361, + "learning_rate": 4.7926377046422575e-05, + "loss": 0.3055490255355835, + "step": 170 + }, + { + "epoch": 1.1615508885298869, + "grad_norm": 0.31330329179763794, + "learning_rate": 4.7453910709078145e-05, + "loss": 0.31946792602539065, + "step": 180 + }, + { + "epoch": 1.2261712439418417, + "grad_norm": 0.3254242539405823, + "learning_rate": 4.69358351508126e-05, + "loss": 0.3053389549255371, + "step": 190 + }, + { + "epoch": 1.2907915993537964, + "grad_norm": 0.3096122145652771, + "learning_rate": 4.637320270602314e-05, + "loss": 0.30810558795928955, + "step": 200 + }, + { + "epoch": 1.3554119547657513, + "grad_norm": 0.30546557903289795, + "learning_rate": 4.576715621471678e-05, + "loss": 0.31534519195556643, + "step": 210 + }, + { + "epoch": 1.420032310177706, + "grad_norm": 0.32512548565864563, + "learning_rate": 4.511892670113104e-05, + "loss": 0.3046459913253784, + "step": 220 + }, + { + "epoch": 1.4846526655896608, + "grad_norm": 0.32627010345458984, + "learning_rate": 4.442983087323167e-05, + "loss": 0.2983341455459595, + "step": 230 + }, + { + "epoch": 1.5492730210016155, + "grad_norm": 0.3522598147392273, + "learning_rate": 4.370126844816628e-05, + "loss": 0.3133318185806274, + "step": 240 + }, + { + "epoch": 1.6138933764135701, + "grad_norm": 0.30818620324134827, + "learning_rate": 4.293471930910668e-05, + "loss": 0.30533685684204104, + "step": 250 + }, + { + "epoch": 1.678513731825525, + "grad_norm": 0.2774396538734436, + "learning_rate": 4.213174049925498e-05, + "loss": 0.2984048843383789, + "step": 260 + }, + { + "epoch": 1.7431340872374799, + "grad_norm": 0.2943349778652191, + "learning_rate": 4.129396305911934e-05, + "loss": 0.2998037338256836, + "step": 270 + }, + { + "epoch": 1.8077544426494345, + "grad_norm": 0.30737221240997314, + "learning_rate": 4.042308871348371e-05, + "loss": 0.3016969680786133, + "step": 280 + }, + { + "epoch": 1.8723747980613892, + "grad_norm": 0.3105849325656891, + "learning_rate": 3.9520886414800936e-05, + "loss": 0.3014758348464966, + "step": 290 + }, + { + "epoch": 1.936995153473344, + "grad_norm": 0.29112061858177185, + "learning_rate": 3.858918875003053e-05, + "loss": 0.297167444229126, + "step": 300 + }, + { + "epoch": 2.0, + "grad_norm": 0.3953491747379303, + "learning_rate": 3.7629888218219774e-05, + "loss": 0.29504544734954835, + "step": 310 + }, + { + "epoch": 2.0646203554119547, + "grad_norm": 0.29763659834861755, + "learning_rate": 3.664493338638898e-05, + "loss": 0.28955745697021484, + "step": 320 + }, + { + "epoch": 2.1292407108239093, + "grad_norm": 0.3380239009857178, + "learning_rate": 3.5636324931529554e-05, + "loss": 0.28567984104156496, + "step": 330 + }, + { + "epoch": 2.1938610662358644, + "grad_norm": 0.31489691138267517, + "learning_rate": 3.46061115767542e-05, + "loss": 0.28932580947875974, + "step": 340 + }, + { + "epoch": 2.258481421647819, + "grad_norm": 0.3106003403663635, + "learning_rate": 3.3556385929854164e-05, + "loss": 0.2879155158996582, + "step": 350 + }, + { + "epoch": 2.3231017770597737, + "grad_norm": 0.2789410948753357, + "learning_rate": 3.248928023271624e-05, + "loss": 0.2938661813735962, + "step": 360 + }, + { + "epoch": 2.387722132471729, + "grad_norm": 0.2844054102897644, + "learning_rate": 3.140696203023348e-05, + "loss": 0.2907016515731812, + "step": 370 + }, + { + "epoch": 2.4523424878836835, + "grad_norm": 0.3297322392463684, + "learning_rate": 3.0311629767507232e-05, + "loss": 0.28118228912353516, + "step": 380 + }, + { + "epoch": 2.516962843295638, + "grad_norm": 0.30098527669906616, + "learning_rate": 2.920550832428337e-05, + "loss": 0.2846884489059448, + "step": 390 + }, + { + "epoch": 2.581583198707593, + "grad_norm": 0.31197914481163025, + "learning_rate": 2.809084449569365e-05, + "loss": 0.28516063690185545, + "step": 400 + } + ], + "logging_steps": 10, + "max_steps": 775, + "num_input_tokens_seen": 0, + "num_train_epochs": 5, + "save_steps": 100, + "stateful_callbacks": { + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": false + }, + "attributes": {} + } + }, + "total_flos": 3.876922101492941e+17, + "train_batch_size": 4, + "trial_name": null, + "trial_params": null +} diff --git a/lora/sft-captcha/checkpoint-400/training_args.bin b/lora/sft-captcha/checkpoint-400/training_args.bin new file mode 100644 index 0000000000000000000000000000000000000000..fb2a0e0e3f614c6145d25df84879d6af93eeaf5b --- /dev/null +++ b/lora/sft-captcha/checkpoint-400/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:994239c83a21a412f370376b6709e6af83b0c60df0ba98c0dc5d4b0c9b6306e1 +size 5496 diff --git a/lora/sft-captcha/checkpoint-500/README.md b/lora/sft-captcha/checkpoint-500/README.md new file mode 100644 index 0000000000000000000000000000000000000000..776a59eddf85aa2890ebb734de124c8a090269b6 --- /dev/null +++ b/lora/sft-captcha/checkpoint-500/README.md @@ -0,0 +1,208 @@ +--- +base_model: Qwen/Qwen3-VL-4B-Instruct +library_name: peft +pipeline_tag: text-generation +tags: +- base_model:adapter:Qwen/Qwen3-VL-4B-Instruct +- llama-factory +- lora +- transformers +--- + +# Model Card for Model ID + + + + + +## Model Details + +### Model Description + + + + + +- **Developed by:** [More Information Needed] +- **Funded by [optional]:** [More Information Needed] +- **Shared by [optional]:** [More Information Needed] +- **Model type:** [More Information Needed] +- **Language(s) (NLP):** [More Information Needed] +- **License:** [More Information Needed] +- **Finetuned from model [optional]:** [More Information Needed] + +### Model Sources [optional] + + + +- **Repository:** [More Information Needed] +- **Paper [optional]:** [More Information Needed] +- **Demo [optional]:** [More Information Needed] + +## Uses + + + +### Direct Use + + + +[More Information Needed] + +### Downstream Use [optional] + + + +[More Information Needed] + +### Out-of-Scope Use + + + +[More Information Needed] + +## Bias, Risks, and Limitations + + + +[More Information Needed] + +### Recommendations + + + +Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations. + +## How to Get Started with the Model + +Use the code below to get started with the model. + +[More Information Needed] + +## Training Details + +### Training Data + + + +[More Information Needed] + +### Training Procedure + + + +#### Preprocessing [optional] + +[More Information Needed] + + +#### Training Hyperparameters + +- **Training regime:** [More Information Needed] + +#### Speeds, Sizes, Times [optional] + + + +[More Information Needed] + +## Evaluation + + + +### Testing Data, Factors & Metrics + +#### Testing Data + + + +[More Information Needed] + +#### Factors + + + +[More Information Needed] + +#### Metrics + + + +[More Information Needed] + +### Results + +[More Information Needed] + +#### Summary + + + +## Model Examination [optional] + + + +[More Information Needed] + +## Environmental Impact + + + +Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700). + +- **Hardware Type:** [More Information Needed] +- **Hours used:** [More Information Needed] +- **Cloud Provider:** [More Information Needed] +- **Compute Region:** [More Information Needed] +- **Carbon Emitted:** [More Information Needed] + +## Technical Specifications [optional] + +### Model Architecture and Objective + +[More Information Needed] + +### Compute Infrastructure + +[More Information Needed] + +#### Hardware + +[More Information Needed] + +#### Software + +[More Information Needed] + +## Citation [optional] + + + +**BibTeX:** + +[More Information Needed] + +**APA:** + +[More Information Needed] + +## Glossary [optional] + + + +[More Information Needed] + +## More Information [optional] + +[More Information Needed] + +## Model Card Authors [optional] + +[More Information Needed] + +## Model Card Contact + +[More Information Needed] +### Framework versions + +- PEFT 0.18.1 \ No newline at end of file diff --git a/lora/sft-captcha/checkpoint-500/adapter_config.json b/lora/sft-captcha/checkpoint-500/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..91f8411a80c3d6f80ff5a9f2d6d5c6e0952c8e01 --- /dev/null +++ b/lora/sft-captcha/checkpoint-500/adapter_config.json @@ -0,0 +1,46 @@ +{ + "alora_invocation_tokens": null, + "alpha_pattern": {}, + "arrow_config": null, + "auto_mapping": null, + "base_model_name_or_path": "Qwen/Qwen3-VL-4B-Instruct", + "bias": "none", + "corda_config": null, + "ensure_weight_tying": false, + "eva_config": null, + "exclude_modules": null, + "fan_in_fan_out": false, + "inference_mode": true, + "init_lora_weights": true, + "layer_replication": null, + "layers_pattern": null, + "layers_to_transform": null, + "loftq_config": {}, + "lora_alpha": 64, + "lora_bias": false, + "lora_dropout": 0.05, + "megatron_config": null, + "megatron_core": "megatron.core", + "modules_to_save": null, + "peft_type": "LORA", + "peft_version": "0.18.1", + "qalora_group_size": 16, + "r": 32, + "rank_pattern": {}, + "revision": null, + "target_modules": [ + "k_proj", + "v_proj", + "gate_proj", + "up_proj", + "o_proj", + "down_proj", + "q_proj" + ], + "target_parameters": null, + "task_type": "CAUSAL_LM", + "trainable_token_indices": null, + "use_dora": false, + "use_qalora": false, + "use_rslora": false +} \ No newline at end of file diff --git a/lora/sft-captcha/checkpoint-500/adapter_model.safetensors b/lora/sft-captcha/checkpoint-500/adapter_model.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..20bfb7bb9d52f6684e0a86b2f21e9864e0ffdfbe --- /dev/null +++ b/lora/sft-captcha/checkpoint-500/adapter_model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4f24bc5bebdaf93e8b3a102055eef9253fc374041639654ee6cc5cb8caa6961b +size 264316456 diff --git a/lora/sft-captcha/checkpoint-500/chat_template.jinja b/lora/sft-captcha/checkpoint-500/chat_template.jinja new file mode 100644 index 0000000000000000000000000000000000000000..124386803f142761528f710e77ae483f5f8c4fc4 --- /dev/null +++ b/lora/sft-captcha/checkpoint-500/chat_template.jinja @@ -0,0 +1,120 @@ +{%- if tools %} + {{- '<|im_start|>system\n' }} + {%- if messages[0].role == 'system' %} + {%- if messages[0].content is string %} + {{- messages[0].content }} + {%- else %} + {%- for content in messages[0].content %} + {%- if 'text' in content %} + {{- content.text }} + {%- endif %} + {%- endfor %} + {%- endif %} + {{- '\n\n' }} + {%- endif %} + {{- "# 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' }} + {%- if messages[0].content is string %} + {{- messages[0].content }} + {%- else %} + {%- for content in messages[0].content %} + {%- if 'text' in content %} + {{- content.text }} + {%- endif %} + {%- endfor %} + {%- endif %} + {{- '<|im_end|>\n' }} + {%- endif %} +{%- endif %} +{%- set image_count = namespace(value=0) %} +{%- set video_count = namespace(value=0) %} +{%- for message in messages %} + {%- if message.role == "user" %} + {{- '<|im_start|>' + message.role + '\n' }} + {%- if message.content is string %} + {{- message.content }} + {%- else %} + {%- for content in message.content %} + {%- if content.type == 'image' or 'image' in content or 'image_url' in content %} + {%- set image_count.value = image_count.value + 1 %} + {%- if add_vision_id %}Picture {{ image_count.value }}: {% endif -%} + <|vision_start|><|image_pad|><|vision_end|> + {%- elif content.type == 'video' or 'video' in content %} + {%- set video_count.value = video_count.value + 1 %} + {%- if add_vision_id %}Video {{ video_count.value }}: {% endif -%} + <|vision_start|><|video_pad|><|vision_end|> + {%- elif 'text' in content %} + {{- content.text }} + {%- endif %} + {%- endfor %} + {%- endif %} + {{- '<|im_end|>\n' }} + {%- elif message.role == "assistant" %} + {{- '<|im_start|>' + message.role + '\n' }} + {%- if message.content is string %} + {{- message.content }} + {%- else %} + {%- for content_item in message.content %} + {%- if 'text' in content_item %} + {{- content_item.text }} + {%- endif %} + {%- endfor %} + {%- endif %} + {%- if message.tool_calls %} + {%- for tool_call in message.tool_calls %} + {%- if (loop.first and message.content) or (not loop.first) %} + {{- '\n' }} + {%- endif %} + {%- if tool_call.function %} + {%- set tool_call = tool_call.function %} + {%- endif %} + {{- '\n{"name": "' }} + {{- tool_call.name }} + {{- '", "arguments": ' }} + {%- if tool_call.arguments is string %} + {{- tool_call.arguments }} + {%- else %} + {{- tool_call.arguments | tojson }} + {%- endif %} + {{- '}\n' }} + {%- endfor %} + {%- endif %} + {{- '<|im_end|>\n' }} + {%- elif message.role == "tool" %} + {%- if loop.first or (messages[loop.index0 - 1].role != "tool") %} + {{- '<|im_start|>user' }} + {%- endif %} + {{- '\n\n' }} + {%- if message.content is string %} + {{- message.content }} + {%- else %} + {%- for content in message.content %} + {%- if content.type == 'image' or 'image' in content or 'image_url' in content %} + {%- set image_count.value = image_count.value + 1 %} + {%- if add_vision_id %}Picture {{ image_count.value }}: {% endif -%} + <|vision_start|><|image_pad|><|vision_end|> + {%- elif content.type == 'video' or 'video' in content %} + {%- set video_count.value = video_count.value + 1 %} + {%- if add_vision_id %}Video {{ video_count.value }}: {% endif -%} + <|vision_start|><|video_pad|><|vision_end|> + {%- elif 'text' in content %} + {{- content.text }} + {%- endif %} + {%- endfor %} + {%- endif %} + {{- '\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/lora/sft-captcha/checkpoint-500/optimizer.pt b/lora/sft-captcha/checkpoint-500/optimizer.pt new file mode 100644 index 0000000000000000000000000000000000000000..afc5aad4924b9e862c93cff2266c320767af9977 --- /dev/null +++ b/lora/sft-captcha/checkpoint-500/optimizer.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:959218718c23668aca16b7c6e08b83540dc806dc9b120b61a8f4d51851d4b54c +size 528907898 diff --git a/lora/sft-captcha/checkpoint-500/processor_config.json b/lora/sft-captcha/checkpoint-500/processor_config.json new file mode 100644 index 0000000000000000000000000000000000000000..33818c7f9e991ad735fd240209f4fa73e6c28c50 --- /dev/null +++ b/lora/sft-captcha/checkpoint-500/processor_config.json @@ -0,0 +1,60 @@ +{ + "image_processor": { + "do_convert_rgb": true, + "do_normalize": true, + "do_rescale": true, + "do_resize": true, + "image_mean": [ + 0.5, + 0.5, + 0.5 + ], + "image_processor_type": "Qwen2VLImageProcessor", + "image_std": [ + 0.5, + 0.5, + 0.5 + ], + "merge_size": 2, + "patch_size": 16, + "resample": 3, + "rescale_factor": 0.00392156862745098, + "size": { + "longest_edge": 16777216, + "shortest_edge": 65536 + }, + "temporal_patch_size": 2 + }, + "processor_class": "Qwen3VLProcessor", + "video_processor": { + "do_convert_rgb": true, + "do_normalize": true, + "do_rescale": true, + "do_resize": true, + "do_sample_frames": true, + "fps": 2, + "image_mean": [ + 0.5, + 0.5, + 0.5 + ], + "image_std": [ + 0.5, + 0.5, + 0.5 + ], + "max_frames": 768, + "merge_size": 2, + "min_frames": 4, + "patch_size": 16, + "resample": 3, + "rescale_factor": 0.00392156862745098, + "return_metadata": false, + "size": { + "longest_edge": 25165824, + "shortest_edge": 4096 + }, + "temporal_patch_size": 2, + "video_processor_type": "Qwen3VLVideoProcessor" + } +} diff --git a/lora/sft-captcha/checkpoint-500/rng_state.pth b/lora/sft-captcha/checkpoint-500/rng_state.pth new file mode 100644 index 0000000000000000000000000000000000000000..215ff460b6e53e59a9364ee6f20208d8a5eca888 --- /dev/null +++ b/lora/sft-captcha/checkpoint-500/rng_state.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:35d1aab3d1e368c6691900c295360091c6e7e65e662b8d111e42b019a91efaf2 +size 14244 diff --git a/lora/sft-captcha/checkpoint-500/scaler.pt b/lora/sft-captcha/checkpoint-500/scaler.pt new file mode 100644 index 0000000000000000000000000000000000000000..b1640696e8c648247d0af4fea4d97302082e3749 --- /dev/null +++ b/lora/sft-captcha/checkpoint-500/scaler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:18b984273ea2d45b7ffb1d047bb359d93111e41fcad70d16a1b453fd38f72636 +size 988 diff --git a/lora/sft-captcha/checkpoint-500/scheduler.pt b/lora/sft-captcha/checkpoint-500/scheduler.pt new file mode 100644 index 0000000000000000000000000000000000000000..65c7b317d9b8700dcdfe7abcaf9d4961f806c6e4 --- /dev/null +++ b/lora/sft-captcha/checkpoint-500/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4e9a07f637a01e092d5d5626103dd0ebc2ecea11a83d0124a7099d655e090b6b +size 1064 diff --git a/lora/sft-captcha/checkpoint-500/tokenizer.json b/lora/sft-captcha/checkpoint-500/tokenizer.json new file mode 100644 index 0000000000000000000000000000000000000000..c7afbed2efcdf019f88ab0572ec29d3bf595dfe2 --- /dev/null +++ b/lora/sft-captcha/checkpoint-500/tokenizer.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:be75606093db2094d7cd20f3c2f385c212750648bd6ea4fb2bf507a6a4c55506 +size 11422650 diff --git a/lora/sft-captcha/checkpoint-500/tokenizer_config.json b/lora/sft-captcha/checkpoint-500/tokenizer_config.json new file mode 100644 index 0000000000000000000000000000000000000000..19febf221cc1a71c2b17feb1d29923a875d42d6e --- /dev/null +++ b/lora/sft-captcha/checkpoint-500/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": 262144, + "pad_token": "<|endoftext|>", + "padding_side": "right", + "processor_class": "Qwen3VLProcessor", + "split_special_tokens": false, + "tokenizer_class": "Qwen2Tokenizer", + "unk_token": null +} diff --git a/lora/sft-captcha/checkpoint-500/trainer_state.json b/lora/sft-captcha/checkpoint-500/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..3885063f5f6783cf8639ea4d4902b59f593cc8f4 --- /dev/null +++ b/lora/sft-captcha/checkpoint-500/trainer_state.json @@ -0,0 +1,384 @@ +{ + "best_global_step": null, + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 3.2261712439418417, + "eval_steps": 500, + "global_step": 500, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.06462035541195477, + "grad_norm": 2.068004846572876, + "learning_rate": 5.76923076923077e-06, + "loss": 0.9769228935241699, + "step": 10 + }, + { + "epoch": 0.12924071082390953, + "grad_norm": 0.9047558307647705, + "learning_rate": 1.217948717948718e-05, + "loss": 0.7594649791717529, + "step": 20 + }, + { + "epoch": 0.1938610662358643, + "grad_norm": 0.4503297805786133, + "learning_rate": 1.858974358974359e-05, + "loss": 0.5557493209838867, + "step": 30 + }, + { + "epoch": 0.25848142164781907, + "grad_norm": 0.37129801511764526, + "learning_rate": 2.5e-05, + "loss": 0.4763153076171875, + "step": 40 + }, + { + "epoch": 0.32310177705977383, + "grad_norm": 0.2953853905200958, + "learning_rate": 3.141025641025641e-05, + "loss": 0.4166147708892822, + "step": 50 + }, + { + "epoch": 0.3877221324717286, + "grad_norm": 0.29160749912261963, + "learning_rate": 3.782051282051282e-05, + "loss": 0.3905663013458252, + "step": 60 + }, + { + "epoch": 0.45234248788368336, + "grad_norm": 0.27939939498901367, + "learning_rate": 4.423076923076923e-05, + "loss": 0.3649837732315063, + "step": 70 + }, + { + "epoch": 0.5169628432956381, + "grad_norm": 0.3130805492401123, + "learning_rate": 4.999974605277764e-05, + "loss": 0.3609442949295044, + "step": 80 + }, + { + "epoch": 0.5815831987075929, + "grad_norm": 0.3720378577709198, + "learning_rate": 4.996927862815294e-05, + "loss": 0.3543627977371216, + "step": 90 + }, + { + "epoch": 0.6462035541195477, + "grad_norm": 0.2810651361942291, + "learning_rate": 4.988809267347141e-05, + "loss": 0.345255446434021, + "step": 100 + }, + { + "epoch": 0.7108239095315024, + "grad_norm": 0.3126155734062195, + "learning_rate": 4.975635309667227e-05, + "loss": 0.33807756900787356, + "step": 110 + }, + { + "epoch": 0.7754442649434572, + "grad_norm": 0.3109033703804016, + "learning_rate": 4.957432749209755e-05, + "loss": 0.3394132614135742, + "step": 120 + }, + { + "epoch": 0.840064620355412, + "grad_norm": 0.3006167411804199, + "learning_rate": 4.934238559694448e-05, + "loss": 0.3244359016418457, + "step": 130 + }, + { + "epoch": 0.9046849757673667, + "grad_norm": 0.3104219138622284, + "learning_rate": 4.90609985402415e-05, + "loss": 0.3221057176589966, + "step": 140 + }, + { + "epoch": 0.9693053311793215, + "grad_norm": 0.330458402633667, + "learning_rate": 4.873073788587338e-05, + "loss": 0.3329624176025391, + "step": 150 + }, + { + "epoch": 1.0323101777059773, + "grad_norm": 0.2666167616844177, + "learning_rate": 4.835227447159939e-05, + "loss": 0.3191133737564087, + "step": 160 + }, + { + "epoch": 1.0969305331179322, + "grad_norm": 0.3217558264732361, + "learning_rate": 4.7926377046422575e-05, + "loss": 0.3055490255355835, + "step": 170 + }, + { + "epoch": 1.1615508885298869, + "grad_norm": 0.31330329179763794, + "learning_rate": 4.7453910709078145e-05, + "loss": 0.31946792602539065, + "step": 180 + }, + { + "epoch": 1.2261712439418417, + "grad_norm": 0.3254242539405823, + "learning_rate": 4.69358351508126e-05, + "loss": 0.3053389549255371, + "step": 190 + }, + { + "epoch": 1.2907915993537964, + "grad_norm": 0.3096122145652771, + "learning_rate": 4.637320270602314e-05, + "loss": 0.30810558795928955, + "step": 200 + }, + { + "epoch": 1.3554119547657513, + "grad_norm": 0.30546557903289795, + "learning_rate": 4.576715621471678e-05, + "loss": 0.31534519195556643, + "step": 210 + }, + { + "epoch": 1.420032310177706, + "grad_norm": 0.32512548565864563, + "learning_rate": 4.511892670113104e-05, + "loss": 0.3046459913253784, + "step": 220 + }, + { + "epoch": 1.4846526655896608, + "grad_norm": 0.32627010345458984, + "learning_rate": 4.442983087323167e-05, + "loss": 0.2983341455459595, + "step": 230 + }, + { + "epoch": 1.5492730210016155, + "grad_norm": 0.3522598147392273, + "learning_rate": 4.370126844816628e-05, + "loss": 0.3133318185806274, + "step": 240 + }, + { + "epoch": 1.6138933764135701, + "grad_norm": 0.30818620324134827, + "learning_rate": 4.293471930910668e-05, + "loss": 0.30533685684204104, + "step": 250 + }, + { + "epoch": 1.678513731825525, + "grad_norm": 0.2774396538734436, + "learning_rate": 4.213174049925498e-05, + "loss": 0.2984048843383789, + "step": 260 + }, + { + "epoch": 1.7431340872374799, + "grad_norm": 0.2943349778652191, + "learning_rate": 4.129396305911934e-05, + "loss": 0.2998037338256836, + "step": 270 + }, + { + "epoch": 1.8077544426494345, + "grad_norm": 0.30737221240997314, + "learning_rate": 4.042308871348371e-05, + "loss": 0.3016969680786133, + "step": 280 + }, + { + "epoch": 1.8723747980613892, + "grad_norm": 0.3105849325656891, + "learning_rate": 3.9520886414800936e-05, + "loss": 0.3014758348464966, + "step": 290 + }, + { + "epoch": 1.936995153473344, + "grad_norm": 0.29112061858177185, + "learning_rate": 3.858918875003053e-05, + "loss": 0.297167444229126, + "step": 300 + }, + { + "epoch": 2.0, + "grad_norm": 0.3953491747379303, + "learning_rate": 3.7629888218219774e-05, + "loss": 0.29504544734954835, + "step": 310 + }, + { + "epoch": 2.0646203554119547, + "grad_norm": 0.29763659834861755, + "learning_rate": 3.664493338638898e-05, + "loss": 0.28955745697021484, + "step": 320 + }, + { + "epoch": 2.1292407108239093, + "grad_norm": 0.3380239009857178, + "learning_rate": 3.5636324931529554e-05, + "loss": 0.28567984104156496, + "step": 330 + }, + { + "epoch": 2.1938610662358644, + "grad_norm": 0.31489691138267517, + "learning_rate": 3.46061115767542e-05, + "loss": 0.28932580947875974, + "step": 340 + }, + { + "epoch": 2.258481421647819, + "grad_norm": 0.3106003403663635, + "learning_rate": 3.3556385929854164e-05, + "loss": 0.2879155158996582, + "step": 350 + }, + { + "epoch": 2.3231017770597737, + "grad_norm": 0.2789410948753357, + "learning_rate": 3.248928023271624e-05, + "loss": 0.2938661813735962, + "step": 360 + }, + { + "epoch": 2.387722132471729, + "grad_norm": 0.2844054102897644, + "learning_rate": 3.140696203023348e-05, + "loss": 0.2907016515731812, + "step": 370 + }, + { + "epoch": 2.4523424878836835, + "grad_norm": 0.3297322392463684, + "learning_rate": 3.0311629767507232e-05, + "loss": 0.28118228912353516, + "step": 380 + }, + { + "epoch": 2.516962843295638, + "grad_norm": 0.30098527669906616, + "learning_rate": 2.920550832428337e-05, + "loss": 0.2846884489059448, + "step": 390 + }, + { + "epoch": 2.581583198707593, + "grad_norm": 0.31197914481163025, + "learning_rate": 2.809084449569365e-05, + "loss": 0.28516063690185545, + "step": 400 + }, + { + "epoch": 2.6462035541195474, + "grad_norm": 0.30557820200920105, + "learning_rate": 2.6969902428481557e-05, + "loss": 0.2813657283782959, + "step": 410 + }, + { + "epoch": 2.7108239095315025, + "grad_norm": 0.3514881730079651, + "learning_rate": 2.584495902198307e-05, + "loss": 0.2873976469039917, + "step": 420 + }, + { + "epoch": 2.775444264943457, + "grad_norm": 0.3010100722312927, + "learning_rate": 2.4718299303203742e-05, + "loss": 0.2792309045791626, + "step": 430 + }, + { + "epoch": 2.840064620355412, + "grad_norm": 0.33192241191864014, + "learning_rate": 2.3592211785386642e-05, + "loss": 0.2784327745437622, + "step": 440 + }, + { + "epoch": 2.904684975767367, + "grad_norm": 0.3093332350254059, + "learning_rate": 2.246898381949894e-05, + "loss": 0.28588788509368895, + "step": 450 + }, + { + "epoch": 2.9693053311793216, + "grad_norm": 0.33140891790390015, + "learning_rate": 2.1350896948079226e-05, + "loss": 0.2833216428756714, + "step": 460 + }, + { + "epoch": 3.0323101777059773, + "grad_norm": 0.31609705090522766, + "learning_rate": 2.0240222270883288e-05, + "loss": 0.28030478954315186, + "step": 470 + }, + { + "epoch": 3.096930533117932, + "grad_norm": 0.2810538113117218, + "learning_rate": 1.913921583174154e-05, + "loss": 0.27578275203704833, + "step": 480 + }, + { + "epoch": 3.161550888529887, + "grad_norm": 0.3165380358695984, + "learning_rate": 1.805011403599862e-05, + "loss": 0.27012245655059813, + "step": 490 + }, + { + "epoch": 3.2261712439418417, + "grad_norm": 0.3220997452735901, + "learning_rate": 1.697512910784352e-05, + "loss": 0.2735466957092285, + "step": 500 + } + ], + "logging_steps": 10, + "max_steps": 775, + "num_input_tokens_seen": 0, + "num_train_epochs": 5, + "save_steps": 100, + "stateful_callbacks": { + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": false + }, + "attributes": {} + } + }, + "total_flos": 4.8452189110984704e+17, + "train_batch_size": 4, + "trial_name": null, + "trial_params": null +} diff --git a/lora/sft-captcha/checkpoint-500/training_args.bin b/lora/sft-captcha/checkpoint-500/training_args.bin new file mode 100644 index 0000000000000000000000000000000000000000..fb2a0e0e3f614c6145d25df84879d6af93eeaf5b --- /dev/null +++ b/lora/sft-captcha/checkpoint-500/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:994239c83a21a412f370376b6709e6af83b0c60df0ba98c0dc5d4b0c9b6306e1 +size 5496 diff --git a/lora/sft-captcha/checkpoint-600/README.md b/lora/sft-captcha/checkpoint-600/README.md new file mode 100644 index 0000000000000000000000000000000000000000..776a59eddf85aa2890ebb734de124c8a090269b6 --- /dev/null +++ b/lora/sft-captcha/checkpoint-600/README.md @@ -0,0 +1,208 @@ +--- +base_model: Qwen/Qwen3-VL-4B-Instruct +library_name: peft +pipeline_tag: text-generation +tags: +- base_model:adapter:Qwen/Qwen3-VL-4B-Instruct +- llama-factory +- lora +- transformers +--- + +# Model Card for Model ID + + + + + +## Model Details + +### Model Description + + + + + +- **Developed by:** [More Information Needed] +- **Funded by [optional]:** [More Information Needed] +- **Shared by [optional]:** [More Information Needed] +- **Model type:** [More Information Needed] +- **Language(s) (NLP):** [More Information Needed] +- **License:** [More Information Needed] +- **Finetuned from model [optional]:** [More Information Needed] + +### Model Sources [optional] + + + +- **Repository:** [More Information Needed] +- **Paper [optional]:** [More Information Needed] +- **Demo [optional]:** [More Information Needed] + +## Uses + + + +### Direct Use + + + +[More Information Needed] + +### Downstream Use [optional] + + + +[More Information Needed] + +### Out-of-Scope Use + + + +[More Information Needed] + +## Bias, Risks, and Limitations + + + +[More Information Needed] + +### Recommendations + + + +Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations. + +## How to Get Started with the Model + +Use the code below to get started with the model. + +[More Information Needed] + +## Training Details + +### Training Data + + + +[More Information Needed] + +### Training Procedure + + + +#### Preprocessing [optional] + +[More Information Needed] + + +#### Training Hyperparameters + +- **Training regime:** [More Information Needed] + +#### Speeds, Sizes, Times [optional] + + + +[More Information Needed] + +## Evaluation + + + +### Testing Data, Factors & Metrics + +#### Testing Data + + + +[More Information Needed] + +#### Factors + + + +[More Information Needed] + +#### Metrics + + + +[More Information Needed] + +### Results + +[More Information Needed] + +#### Summary + + + +## Model Examination [optional] + + + +[More Information Needed] + +## Environmental Impact + + + +Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700). + +- **Hardware Type:** [More Information Needed] +- **Hours used:** [More Information Needed] +- **Cloud Provider:** [More Information Needed] +- **Compute Region:** [More Information Needed] +- **Carbon Emitted:** [More Information Needed] + +## Technical Specifications [optional] + +### Model Architecture and Objective + +[More Information Needed] + +### Compute Infrastructure + +[More Information Needed] + +#### Hardware + +[More Information Needed] + +#### Software + +[More Information Needed] + +## Citation [optional] + + + +**BibTeX:** + +[More Information Needed] + +**APA:** + +[More Information Needed] + +## Glossary [optional] + + + +[More Information Needed] + +## More Information [optional] + +[More Information Needed] + +## Model Card Authors [optional] + +[More Information Needed] + +## Model Card Contact + +[More Information Needed] +### Framework versions + +- PEFT 0.18.1 \ No newline at end of file diff --git a/lora/sft-captcha/checkpoint-600/adapter_config.json b/lora/sft-captcha/checkpoint-600/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..91f8411a80c3d6f80ff5a9f2d6d5c6e0952c8e01 --- /dev/null +++ b/lora/sft-captcha/checkpoint-600/adapter_config.json @@ -0,0 +1,46 @@ +{ + "alora_invocation_tokens": null, + "alpha_pattern": {}, + "arrow_config": null, + "auto_mapping": null, + "base_model_name_or_path": "Qwen/Qwen3-VL-4B-Instruct", + "bias": "none", + "corda_config": null, + "ensure_weight_tying": false, + "eva_config": null, + "exclude_modules": null, + "fan_in_fan_out": false, + "inference_mode": true, + "init_lora_weights": true, + "layer_replication": null, + "layers_pattern": null, + "layers_to_transform": null, + "loftq_config": {}, + "lora_alpha": 64, + "lora_bias": false, + "lora_dropout": 0.05, + "megatron_config": null, + "megatron_core": "megatron.core", + "modules_to_save": null, + "peft_type": "LORA", + "peft_version": "0.18.1", + "qalora_group_size": 16, + "r": 32, + "rank_pattern": {}, + "revision": null, + "target_modules": [ + "k_proj", + "v_proj", + "gate_proj", + "up_proj", + "o_proj", + "down_proj", + "q_proj" + ], + "target_parameters": null, + "task_type": "CAUSAL_LM", + "trainable_token_indices": null, + "use_dora": false, + "use_qalora": false, + "use_rslora": false +} \ No newline at end of file diff --git a/lora/sft-captcha/checkpoint-600/adapter_model.safetensors b/lora/sft-captcha/checkpoint-600/adapter_model.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..513bd9b3da4ea06acdb8726aa3efc8281bf31869 --- /dev/null +++ b/lora/sft-captcha/checkpoint-600/adapter_model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6a5721f058d1fe20172472aa0fe233c7960f2fedeed5f13f328b9ad96c0d00b7 +size 264316456 diff --git a/lora/sft-captcha/checkpoint-600/chat_template.jinja b/lora/sft-captcha/checkpoint-600/chat_template.jinja new file mode 100644 index 0000000000000000000000000000000000000000..124386803f142761528f710e77ae483f5f8c4fc4 --- /dev/null +++ b/lora/sft-captcha/checkpoint-600/chat_template.jinja @@ -0,0 +1,120 @@ +{%- if tools %} + {{- '<|im_start|>system\n' }} + {%- if messages[0].role == 'system' %} + {%- if messages[0].content is string %} + {{- messages[0].content }} + {%- else %} + {%- for content in messages[0].content %} + {%- if 'text' in content %} + {{- content.text }} + {%- endif %} + {%- endfor %} + {%- endif %} + {{- '\n\n' }} + {%- endif %} + {{- "# 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' }} + {%- if messages[0].content is string %} + {{- messages[0].content }} + {%- else %} + {%- for content in messages[0].content %} + {%- if 'text' in content %} + {{- content.text }} + {%- endif %} + {%- endfor %} + {%- endif %} + {{- '<|im_end|>\n' }} + {%- endif %} +{%- endif %} +{%- set image_count = namespace(value=0) %} +{%- set video_count = namespace(value=0) %} +{%- for message in messages %} + {%- if message.role == "user" %} + {{- '<|im_start|>' + message.role + '\n' }} + {%- if message.content is string %} + {{- message.content }} + {%- else %} + {%- for content in message.content %} + {%- if content.type == 'image' or 'image' in content or 'image_url' in content %} + {%- set image_count.value = image_count.value + 1 %} + {%- if add_vision_id %}Picture {{ image_count.value }}: {% endif -%} + <|vision_start|><|image_pad|><|vision_end|> + {%- elif content.type == 'video' or 'video' in content %} + {%- set video_count.value = video_count.value + 1 %} + {%- if add_vision_id %}Video {{ video_count.value }}: {% endif -%} + <|vision_start|><|video_pad|><|vision_end|> + {%- elif 'text' in content %} + {{- content.text }} + {%- endif %} + {%- endfor %} + {%- endif %} + {{- '<|im_end|>\n' }} + {%- elif message.role == "assistant" %} + {{- '<|im_start|>' + message.role + '\n' }} + {%- if message.content is string %} + {{- message.content }} + {%- else %} + {%- for content_item in message.content %} + {%- if 'text' in content_item %} + {{- content_item.text }} + {%- endif %} + {%- endfor %} + {%- endif %} + {%- if message.tool_calls %} + {%- for tool_call in message.tool_calls %} + {%- if (loop.first and message.content) or (not loop.first) %} + {{- '\n' }} + {%- endif %} + {%- if tool_call.function %} + {%- set tool_call = tool_call.function %} + {%- endif %} + {{- '\n{"name": "' }} + {{- tool_call.name }} + {{- '", "arguments": ' }} + {%- if tool_call.arguments is string %} + {{- tool_call.arguments }} + {%- else %} + {{- tool_call.arguments | tojson }} + {%- endif %} + {{- '}\n' }} + {%- endfor %} + {%- endif %} + {{- '<|im_end|>\n' }} + {%- elif message.role == "tool" %} + {%- if loop.first or (messages[loop.index0 - 1].role != "tool") %} + {{- '<|im_start|>user' }} + {%- endif %} + {{- '\n\n' }} + {%- if message.content is string %} + {{- message.content }} + {%- else %} + {%- for content in message.content %} + {%- if content.type == 'image' or 'image' in content or 'image_url' in content %} + {%- set image_count.value = image_count.value + 1 %} + {%- if add_vision_id %}Picture {{ image_count.value }}: {% endif -%} + <|vision_start|><|image_pad|><|vision_end|> + {%- elif content.type == 'video' or 'video' in content %} + {%- set video_count.value = video_count.value + 1 %} + {%- if add_vision_id %}Video {{ video_count.value }}: {% endif -%} + <|vision_start|><|video_pad|><|vision_end|> + {%- elif 'text' in content %} + {{- content.text }} + {%- endif %} + {%- endfor %} + {%- endif %} + {{- '\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/lora/sft-captcha/checkpoint-600/optimizer.pt b/lora/sft-captcha/checkpoint-600/optimizer.pt new file mode 100644 index 0000000000000000000000000000000000000000..eeac073b14215dffe3eb3a0b4dbc2f26a67ae33f --- /dev/null +++ b/lora/sft-captcha/checkpoint-600/optimizer.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d534d1b2f938995e792afddf132a3a16a82d8dc8544391a789fcd4f471f78053 +size 528907898 diff --git a/lora/sft-captcha/checkpoint-600/processor_config.json b/lora/sft-captcha/checkpoint-600/processor_config.json new file mode 100644 index 0000000000000000000000000000000000000000..33818c7f9e991ad735fd240209f4fa73e6c28c50 --- /dev/null +++ b/lora/sft-captcha/checkpoint-600/processor_config.json @@ -0,0 +1,60 @@ +{ + "image_processor": { + "do_convert_rgb": true, + "do_normalize": true, + "do_rescale": true, + "do_resize": true, + "image_mean": [ + 0.5, + 0.5, + 0.5 + ], + "image_processor_type": "Qwen2VLImageProcessor", + "image_std": [ + 0.5, + 0.5, + 0.5 + ], + "merge_size": 2, + "patch_size": 16, + "resample": 3, + "rescale_factor": 0.00392156862745098, + "size": { + "longest_edge": 16777216, + "shortest_edge": 65536 + }, + "temporal_patch_size": 2 + }, + "processor_class": "Qwen3VLProcessor", + "video_processor": { + "do_convert_rgb": true, + "do_normalize": true, + "do_rescale": true, + "do_resize": true, + "do_sample_frames": true, + "fps": 2, + "image_mean": [ + 0.5, + 0.5, + 0.5 + ], + "image_std": [ + 0.5, + 0.5, + 0.5 + ], + "max_frames": 768, + "merge_size": 2, + "min_frames": 4, + "patch_size": 16, + "resample": 3, + "rescale_factor": 0.00392156862745098, + "return_metadata": false, + "size": { + "longest_edge": 25165824, + "shortest_edge": 4096 + }, + "temporal_patch_size": 2, + "video_processor_type": "Qwen3VLVideoProcessor" + } +} diff --git a/lora/sft-captcha/checkpoint-600/rng_state.pth b/lora/sft-captcha/checkpoint-600/rng_state.pth new file mode 100644 index 0000000000000000000000000000000000000000..6de6d413e6ace775ef244fc54067313c9b593b2d --- /dev/null +++ b/lora/sft-captcha/checkpoint-600/rng_state.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0f541f824f68fa5ee039423d4bc10b573113d8ccb48fb258d6b1fda9dace15c4 +size 14244 diff --git a/lora/sft-captcha/checkpoint-600/scaler.pt b/lora/sft-captcha/checkpoint-600/scaler.pt new file mode 100644 index 0000000000000000000000000000000000000000..22cf649db85f1ace860ed1049e0bceb509519d57 --- /dev/null +++ b/lora/sft-captcha/checkpoint-600/scaler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e32b8db4f4cba2b7b0157ab60a064591c3802a9067529d5fddbb1c6c1e99660b +size 988 diff --git a/lora/sft-captcha/checkpoint-600/scheduler.pt b/lora/sft-captcha/checkpoint-600/scheduler.pt new file mode 100644 index 0000000000000000000000000000000000000000..1ada30734976ed6bcb4f0948acd9db80c21f8ad6 --- /dev/null +++ b/lora/sft-captcha/checkpoint-600/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b74e2a2ae06cc083da5b0fbf7843ecfcacdb4293c8a4f47b648e760b6dddaad8 +size 1064 diff --git a/lora/sft-captcha/checkpoint-600/tokenizer.json b/lora/sft-captcha/checkpoint-600/tokenizer.json new file mode 100644 index 0000000000000000000000000000000000000000..c7afbed2efcdf019f88ab0572ec29d3bf595dfe2 --- /dev/null +++ b/lora/sft-captcha/checkpoint-600/tokenizer.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:be75606093db2094d7cd20f3c2f385c212750648bd6ea4fb2bf507a6a4c55506 +size 11422650 diff --git a/lora/sft-captcha/checkpoint-600/tokenizer_config.json b/lora/sft-captcha/checkpoint-600/tokenizer_config.json new file mode 100644 index 0000000000000000000000000000000000000000..19febf221cc1a71c2b17feb1d29923a875d42d6e --- /dev/null +++ b/lora/sft-captcha/checkpoint-600/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": 262144, + "pad_token": "<|endoftext|>", + "padding_side": "right", + "processor_class": "Qwen3VLProcessor", + "split_special_tokens": false, + "tokenizer_class": "Qwen2Tokenizer", + "unk_token": null +} diff --git a/lora/sft-captcha/checkpoint-600/trainer_state.json b/lora/sft-captcha/checkpoint-600/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..89a0c5fbdffbd9eef546f8311e721cf97d59f02a --- /dev/null +++ b/lora/sft-captcha/checkpoint-600/trainer_state.json @@ -0,0 +1,454 @@ +{ + "best_global_step": null, + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 3.872374798061389, + "eval_steps": 500, + "global_step": 600, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.06462035541195477, + "grad_norm": 2.068004846572876, + "learning_rate": 5.76923076923077e-06, + "loss": 0.9769228935241699, + "step": 10 + }, + { + "epoch": 0.12924071082390953, + "grad_norm": 0.9047558307647705, + "learning_rate": 1.217948717948718e-05, + "loss": 0.7594649791717529, + "step": 20 + }, + { + "epoch": 0.1938610662358643, + "grad_norm": 0.4503297805786133, + "learning_rate": 1.858974358974359e-05, + "loss": 0.5557493209838867, + "step": 30 + }, + { + "epoch": 0.25848142164781907, + "grad_norm": 0.37129801511764526, + "learning_rate": 2.5e-05, + "loss": 0.4763153076171875, + "step": 40 + }, + { + "epoch": 0.32310177705977383, + "grad_norm": 0.2953853905200958, + "learning_rate": 3.141025641025641e-05, + "loss": 0.4166147708892822, + "step": 50 + }, + { + "epoch": 0.3877221324717286, + "grad_norm": 0.29160749912261963, + "learning_rate": 3.782051282051282e-05, + "loss": 0.3905663013458252, + "step": 60 + }, + { + "epoch": 0.45234248788368336, + "grad_norm": 0.27939939498901367, + "learning_rate": 4.423076923076923e-05, + "loss": 0.3649837732315063, + "step": 70 + }, + { + "epoch": 0.5169628432956381, + "grad_norm": 0.3130805492401123, + "learning_rate": 4.999974605277764e-05, + "loss": 0.3609442949295044, + "step": 80 + }, + { + "epoch": 0.5815831987075929, + "grad_norm": 0.3720378577709198, + "learning_rate": 4.996927862815294e-05, + "loss": 0.3543627977371216, + "step": 90 + }, + { + "epoch": 0.6462035541195477, + "grad_norm": 0.2810651361942291, + "learning_rate": 4.988809267347141e-05, + "loss": 0.345255446434021, + "step": 100 + }, + { + "epoch": 0.7108239095315024, + "grad_norm": 0.3126155734062195, + "learning_rate": 4.975635309667227e-05, + "loss": 0.33807756900787356, + "step": 110 + }, + { + "epoch": 0.7754442649434572, + "grad_norm": 0.3109033703804016, + "learning_rate": 4.957432749209755e-05, + "loss": 0.3394132614135742, + "step": 120 + }, + { + "epoch": 0.840064620355412, + "grad_norm": 0.3006167411804199, + "learning_rate": 4.934238559694448e-05, + "loss": 0.3244359016418457, + "step": 130 + }, + { + "epoch": 0.9046849757673667, + "grad_norm": 0.3104219138622284, + "learning_rate": 4.90609985402415e-05, + "loss": 0.3221057176589966, + "step": 140 + }, + { + "epoch": 0.9693053311793215, + "grad_norm": 0.330458402633667, + "learning_rate": 4.873073788587338e-05, + "loss": 0.3329624176025391, + "step": 150 + }, + { + "epoch": 1.0323101777059773, + "grad_norm": 0.2666167616844177, + "learning_rate": 4.835227447159939e-05, + "loss": 0.3191133737564087, + "step": 160 + }, + { + "epoch": 1.0969305331179322, + "grad_norm": 0.3217558264732361, + "learning_rate": 4.7926377046422575e-05, + "loss": 0.3055490255355835, + "step": 170 + }, + { + "epoch": 1.1615508885298869, + "grad_norm": 0.31330329179763794, + "learning_rate": 4.7453910709078145e-05, + "loss": 0.31946792602539065, + "step": 180 + }, + { + "epoch": 1.2261712439418417, + "grad_norm": 0.3254242539405823, + "learning_rate": 4.69358351508126e-05, + "loss": 0.3053389549255371, + "step": 190 + }, + { + "epoch": 1.2907915993537964, + "grad_norm": 0.3096122145652771, + "learning_rate": 4.637320270602314e-05, + "loss": 0.30810558795928955, + "step": 200 + }, + { + "epoch": 1.3554119547657513, + "grad_norm": 0.30546557903289795, + "learning_rate": 4.576715621471678e-05, + "loss": 0.31534519195556643, + "step": 210 + }, + { + "epoch": 1.420032310177706, + "grad_norm": 0.32512548565864563, + "learning_rate": 4.511892670113104e-05, + "loss": 0.3046459913253784, + "step": 220 + }, + { + "epoch": 1.4846526655896608, + "grad_norm": 0.32627010345458984, + "learning_rate": 4.442983087323167e-05, + "loss": 0.2983341455459595, + "step": 230 + }, + { + "epoch": 1.5492730210016155, + "grad_norm": 0.3522598147392273, + "learning_rate": 4.370126844816628e-05, + "loss": 0.3133318185806274, + "step": 240 + }, + { + "epoch": 1.6138933764135701, + "grad_norm": 0.30818620324134827, + "learning_rate": 4.293471930910668e-05, + "loss": 0.30533685684204104, + "step": 250 + }, + { + "epoch": 1.678513731825525, + "grad_norm": 0.2774396538734436, + "learning_rate": 4.213174049925498e-05, + "loss": 0.2984048843383789, + "step": 260 + }, + { + "epoch": 1.7431340872374799, + "grad_norm": 0.2943349778652191, + "learning_rate": 4.129396305911934e-05, + "loss": 0.2998037338256836, + "step": 270 + }, + { + "epoch": 1.8077544426494345, + "grad_norm": 0.30737221240997314, + "learning_rate": 4.042308871348371e-05, + "loss": 0.3016969680786133, + "step": 280 + }, + { + "epoch": 1.8723747980613892, + "grad_norm": 0.3105849325656891, + "learning_rate": 3.9520886414800936e-05, + "loss": 0.3014758348464966, + "step": 290 + }, + { + "epoch": 1.936995153473344, + "grad_norm": 0.29112061858177185, + "learning_rate": 3.858918875003053e-05, + "loss": 0.297167444229126, + "step": 300 + }, + { + "epoch": 2.0, + "grad_norm": 0.3953491747379303, + "learning_rate": 3.7629888218219774e-05, + "loss": 0.29504544734954835, + "step": 310 + }, + { + "epoch": 2.0646203554119547, + "grad_norm": 0.29763659834861755, + "learning_rate": 3.664493338638898e-05, + "loss": 0.28955745697021484, + "step": 320 + }, + { + "epoch": 2.1292407108239093, + "grad_norm": 0.3380239009857178, + "learning_rate": 3.5636324931529554e-05, + "loss": 0.28567984104156496, + "step": 330 + }, + { + "epoch": 2.1938610662358644, + "grad_norm": 0.31489691138267517, + "learning_rate": 3.46061115767542e-05, + "loss": 0.28932580947875974, + "step": 340 + }, + { + "epoch": 2.258481421647819, + "grad_norm": 0.3106003403663635, + "learning_rate": 3.3556385929854164e-05, + "loss": 0.2879155158996582, + "step": 350 + }, + { + "epoch": 2.3231017770597737, + "grad_norm": 0.2789410948753357, + "learning_rate": 3.248928023271624e-05, + "loss": 0.2938661813735962, + "step": 360 + }, + { + "epoch": 2.387722132471729, + "grad_norm": 0.2844054102897644, + "learning_rate": 3.140696203023348e-05, + "loss": 0.2907016515731812, + "step": 370 + }, + { + "epoch": 2.4523424878836835, + "grad_norm": 0.3297322392463684, + "learning_rate": 3.0311629767507232e-05, + "loss": 0.28118228912353516, + "step": 380 + }, + { + "epoch": 2.516962843295638, + "grad_norm": 0.30098527669906616, + "learning_rate": 2.920550832428337e-05, + "loss": 0.2846884489059448, + "step": 390 + }, + { + "epoch": 2.581583198707593, + "grad_norm": 0.31197914481163025, + "learning_rate": 2.809084449569365e-05, + "loss": 0.28516063690185545, + "step": 400 + }, + { + "epoch": 2.6462035541195474, + "grad_norm": 0.30557820200920105, + "learning_rate": 2.6969902428481557e-05, + "loss": 0.2813657283782959, + "step": 410 + }, + { + "epoch": 2.7108239095315025, + "grad_norm": 0.3514881730079651, + "learning_rate": 2.584495902198307e-05, + "loss": 0.2873976469039917, + "step": 420 + }, + { + "epoch": 2.775444264943457, + "grad_norm": 0.3010100722312927, + "learning_rate": 2.4718299303203742e-05, + "loss": 0.2792309045791626, + "step": 430 + }, + { + "epoch": 2.840064620355412, + "grad_norm": 0.33192241191864014, + "learning_rate": 2.3592211785386642e-05, + "loss": 0.2784327745437622, + "step": 440 + }, + { + "epoch": 2.904684975767367, + "grad_norm": 0.3093332350254059, + "learning_rate": 2.246898381949894e-05, + "loss": 0.28588788509368895, + "step": 450 + }, + { + "epoch": 2.9693053311793216, + "grad_norm": 0.33140891790390015, + "learning_rate": 2.1350896948079226e-05, + "loss": 0.2833216428756714, + "step": 460 + }, + { + "epoch": 3.0323101777059773, + "grad_norm": 0.31609705090522766, + "learning_rate": 2.0240222270883288e-05, + "loss": 0.28030478954315186, + "step": 470 + }, + { + "epoch": 3.096930533117932, + "grad_norm": 0.2810538113117218, + "learning_rate": 1.913921583174154e-05, + "loss": 0.27578275203704833, + "step": 480 + }, + { + "epoch": 3.161550888529887, + "grad_norm": 0.3165380358695984, + "learning_rate": 1.805011403599862e-05, + "loss": 0.27012245655059813, + "step": 490 + }, + { + "epoch": 3.2261712439418417, + "grad_norm": 0.3220997452735901, + "learning_rate": 1.697512910784352e-05, + "loss": 0.2735466957092285, + "step": 500 + }, + { + "epoch": 3.2907915993537964, + "grad_norm": 0.3499109447002411, + "learning_rate": 1.5916444596757218e-05, + "loss": 0.27160613536834716, + "step": 510 + }, + { + "epoch": 3.355411954765751, + "grad_norm": 0.3451310694217682, + "learning_rate": 1.4876210942205604e-05, + "loss": 0.27280707359313966, + "step": 520 + }, + { + "epoch": 3.420032310177706, + "grad_norm": 0.33574041724205017, + "learning_rate": 1.3856541105586545e-05, + "loss": 0.27128801345825193, + "step": 530 + }, + { + "epoch": 3.484652665589661, + "grad_norm": 0.3565731346607208, + "learning_rate": 1.2859506278303979e-05, + "loss": 0.2736222743988037, + "step": 540 + }, + { + "epoch": 3.5492730210016155, + "grad_norm": 0.32914265990257263, + "learning_rate": 1.1887131674686596e-05, + "loss": 0.27256240844726565, + "step": 550 + }, + { + "epoch": 3.61389337641357, + "grad_norm": 0.31808099150657654, + "learning_rate": 1.0941392418297026e-05, + "loss": 0.27215673923492434, + "step": 560 + }, + { + "epoch": 3.678513731825525, + "grad_norm": 0.3377521336078644, + "learning_rate": 1.0024209529987149e-05, + "loss": 0.2722803592681885, + "step": 570 + }, + { + "epoch": 3.74313408723748, + "grad_norm": 0.33442333340644836, + "learning_rate": 9.137446025848914e-06, + "loss": 0.27033517360687254, + "step": 580 + }, + { + "epoch": 3.8077544426494345, + "grad_norm": 0.32362648844718933, + "learning_rate": 8.282903132986526e-06, + "loss": 0.268932580947876, + "step": 590 + }, + { + "epoch": 3.872374798061389, + "grad_norm": 0.33127161860466003, + "learning_rate": 7.462316630796834e-06, + "loss": 0.274311900138855, + "step": 600 + } + ], + "logging_steps": 10, + "max_steps": 775, + "num_input_tokens_seen": 0, + "num_train_epochs": 5, + "save_steps": 100, + "stateful_callbacks": { + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": false + }, + "attributes": {} + } + }, + "total_flos": 5.816662323100877e+17, + "train_batch_size": 4, + "trial_name": null, + "trial_params": null +} diff --git a/lora/sft-captcha/checkpoint-600/training_args.bin b/lora/sft-captcha/checkpoint-600/training_args.bin new file mode 100644 index 0000000000000000000000000000000000000000..fb2a0e0e3f614c6145d25df84879d6af93eeaf5b --- /dev/null +++ b/lora/sft-captcha/checkpoint-600/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:994239c83a21a412f370376b6709e6af83b0c60df0ba98c0dc5d4b0c9b6306e1 +size 5496 diff --git a/lora/sft-captcha/checkpoint-700/README.md b/lora/sft-captcha/checkpoint-700/README.md new file mode 100644 index 0000000000000000000000000000000000000000..776a59eddf85aa2890ebb734de124c8a090269b6 --- /dev/null +++ b/lora/sft-captcha/checkpoint-700/README.md @@ -0,0 +1,208 @@ +--- +base_model: Qwen/Qwen3-VL-4B-Instruct +library_name: peft +pipeline_tag: text-generation +tags: +- base_model:adapter:Qwen/Qwen3-VL-4B-Instruct +- llama-factory +- lora +- transformers +--- + +# Model Card for Model ID + + + + + +## Model Details + +### Model Description + + + + + +- **Developed by:** [More Information Needed] +- **Funded by [optional]:** [More Information Needed] +- **Shared by [optional]:** [More Information Needed] +- **Model type:** [More Information Needed] +- **Language(s) (NLP):** [More Information Needed] +- **License:** [More Information Needed] +- **Finetuned from model [optional]:** [More Information Needed] + +### Model Sources [optional] + + + +- **Repository:** [More Information Needed] +- **Paper [optional]:** [More Information Needed] +- **Demo [optional]:** [More Information Needed] + +## Uses + + + +### Direct Use + + + +[More Information Needed] + +### Downstream Use [optional] + + + +[More Information Needed] + +### Out-of-Scope Use + + + +[More Information Needed] + +## Bias, Risks, and Limitations + + + +[More Information Needed] + +### Recommendations + + + +Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations. + +## How to Get Started with the Model + +Use the code below to get started with the model. + +[More Information Needed] + +## Training Details + +### Training Data + + + +[More Information Needed] + +### Training Procedure + + + +#### Preprocessing [optional] + +[More Information Needed] + + +#### Training Hyperparameters + +- **Training regime:** [More Information Needed] + +#### Speeds, Sizes, Times [optional] + + + +[More Information Needed] + +## Evaluation + + + +### Testing Data, Factors & Metrics + +#### Testing Data + + + +[More Information Needed] + +#### Factors + + + +[More Information Needed] + +#### Metrics + + + +[More Information Needed] + +### Results + +[More Information Needed] + +#### Summary + + + +## Model Examination [optional] + + + +[More Information Needed] + +## Environmental Impact + + + +Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700). + +- **Hardware Type:** [More Information Needed] +- **Hours used:** [More Information Needed] +- **Cloud Provider:** [More Information Needed] +- **Compute Region:** [More Information Needed] +- **Carbon Emitted:** [More Information Needed] + +## Technical Specifications [optional] + +### Model Architecture and Objective + +[More Information Needed] + +### Compute Infrastructure + +[More Information Needed] + +#### Hardware + +[More Information Needed] + +#### Software + +[More Information Needed] + +## Citation [optional] + + + +**BibTeX:** + +[More Information Needed] + +**APA:** + +[More Information Needed] + +## Glossary [optional] + + + +[More Information Needed] + +## More Information [optional] + +[More Information Needed] + +## Model Card Authors [optional] + +[More Information Needed] + +## Model Card Contact + +[More Information Needed] +### Framework versions + +- PEFT 0.18.1 \ No newline at end of file diff --git a/lora/sft-captcha/checkpoint-700/adapter_config.json b/lora/sft-captcha/checkpoint-700/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..91f8411a80c3d6f80ff5a9f2d6d5c6e0952c8e01 --- /dev/null +++ b/lora/sft-captcha/checkpoint-700/adapter_config.json @@ -0,0 +1,46 @@ +{ + "alora_invocation_tokens": null, + "alpha_pattern": {}, + "arrow_config": null, + "auto_mapping": null, + "base_model_name_or_path": "Qwen/Qwen3-VL-4B-Instruct", + "bias": "none", + "corda_config": null, + "ensure_weight_tying": false, + "eva_config": null, + "exclude_modules": null, + "fan_in_fan_out": false, + "inference_mode": true, + "init_lora_weights": true, + "layer_replication": null, + "layers_pattern": null, + "layers_to_transform": null, + "loftq_config": {}, + "lora_alpha": 64, + "lora_bias": false, + "lora_dropout": 0.05, + "megatron_config": null, + "megatron_core": "megatron.core", + "modules_to_save": null, + "peft_type": "LORA", + "peft_version": "0.18.1", + "qalora_group_size": 16, + "r": 32, + "rank_pattern": {}, + "revision": null, + "target_modules": [ + "k_proj", + "v_proj", + "gate_proj", + "up_proj", + "o_proj", + "down_proj", + "q_proj" + ], + "target_parameters": null, + "task_type": "CAUSAL_LM", + "trainable_token_indices": null, + "use_dora": false, + "use_qalora": false, + "use_rslora": false +} \ No newline at end of file diff --git a/lora/sft-captcha/checkpoint-700/adapter_model.safetensors b/lora/sft-captcha/checkpoint-700/adapter_model.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..4e799e199fd7263ec507a786416bdd9d53164733 --- /dev/null +++ b/lora/sft-captcha/checkpoint-700/adapter_model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:307c2b4e2255090e13237967266449a5f71c8de87aa3403c0e0a4db1b76359cf +size 264316456 diff --git a/lora/sft-captcha/checkpoint-700/chat_template.jinja b/lora/sft-captcha/checkpoint-700/chat_template.jinja new file mode 100644 index 0000000000000000000000000000000000000000..124386803f142761528f710e77ae483f5f8c4fc4 --- /dev/null +++ b/lora/sft-captcha/checkpoint-700/chat_template.jinja @@ -0,0 +1,120 @@ +{%- if tools %} + {{- '<|im_start|>system\n' }} + {%- if messages[0].role == 'system' %} + {%- if messages[0].content is string %} + {{- messages[0].content }} + {%- else %} + {%- for content in messages[0].content %} + {%- if 'text' in content %} + {{- content.text }} + {%- endif %} + {%- endfor %} + {%- endif %} + {{- '\n\n' }} + {%- endif %} + {{- "# 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' }} + {%- if messages[0].content is string %} + {{- messages[0].content }} + {%- else %} + {%- for content in messages[0].content %} + {%- if 'text' in content %} + {{- content.text }} + {%- endif %} + {%- endfor %} + {%- endif %} + {{- '<|im_end|>\n' }} + {%- endif %} +{%- endif %} +{%- set image_count = namespace(value=0) %} +{%- set video_count = namespace(value=0) %} +{%- for message in messages %} + {%- if message.role == "user" %} + {{- '<|im_start|>' + message.role + '\n' }} + {%- if message.content is string %} + {{- message.content }} + {%- else %} + {%- for content in message.content %} + {%- if content.type == 'image' or 'image' in content or 'image_url' in content %} + {%- set image_count.value = image_count.value + 1 %} + {%- if add_vision_id %}Picture {{ image_count.value }}: {% endif -%} + <|vision_start|><|image_pad|><|vision_end|> + {%- elif content.type == 'video' or 'video' in content %} + {%- set video_count.value = video_count.value + 1 %} + {%- if add_vision_id %}Video {{ video_count.value }}: {% endif -%} + <|vision_start|><|video_pad|><|vision_end|> + {%- elif 'text' in content %} + {{- content.text }} + {%- endif %} + {%- endfor %} + {%- endif %} + {{- '<|im_end|>\n' }} + {%- elif message.role == "assistant" %} + {{- '<|im_start|>' + message.role + '\n' }} + {%- if message.content is string %} + {{- message.content }} + {%- else %} + {%- for content_item in message.content %} + {%- if 'text' in content_item %} + {{- content_item.text }} + {%- endif %} + {%- endfor %} + {%- endif %} + {%- if message.tool_calls %} + {%- for tool_call in message.tool_calls %} + {%- if (loop.first and message.content) or (not loop.first) %} + {{- '\n' }} + {%- endif %} + {%- if tool_call.function %} + {%- set tool_call = tool_call.function %} + {%- endif %} + {{- '\n{"name": "' }} + {{- tool_call.name }} + {{- '", "arguments": ' }} + {%- if tool_call.arguments is string %} + {{- tool_call.arguments }} + {%- else %} + {{- tool_call.arguments | tojson }} + {%- endif %} + {{- '}\n' }} + {%- endfor %} + {%- endif %} + {{- '<|im_end|>\n' }} + {%- elif message.role == "tool" %} + {%- if loop.first or (messages[loop.index0 - 1].role != "tool") %} + {{- '<|im_start|>user' }} + {%- endif %} + {{- '\n\n' }} + {%- if message.content is string %} + {{- message.content }} + {%- else %} + {%- for content in message.content %} + {%- if content.type == 'image' or 'image' in content or 'image_url' in content %} + {%- set image_count.value = image_count.value + 1 %} + {%- if add_vision_id %}Picture {{ image_count.value }}: {% endif -%} + <|vision_start|><|image_pad|><|vision_end|> + {%- elif content.type == 'video' or 'video' in content %} + {%- set video_count.value = video_count.value + 1 %} + {%- if add_vision_id %}Video {{ video_count.value }}: {% endif -%} + <|vision_start|><|video_pad|><|vision_end|> + {%- elif 'text' in content %} + {{- content.text }} + {%- endif %} + {%- endfor %} + {%- endif %} + {{- '\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/lora/sft-captcha/checkpoint-700/optimizer.pt b/lora/sft-captcha/checkpoint-700/optimizer.pt new file mode 100644 index 0000000000000000000000000000000000000000..94b8a8f8bde5a7ef848661b70a33c7d3f235e21a --- /dev/null +++ b/lora/sft-captcha/checkpoint-700/optimizer.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ddb527f07a922693764c931a34b0ea01f4366b3002d99e4c23e2b018833cf4c5 +size 528907898 diff --git a/lora/sft-captcha/checkpoint-700/processor_config.json b/lora/sft-captcha/checkpoint-700/processor_config.json new file mode 100644 index 0000000000000000000000000000000000000000..33818c7f9e991ad735fd240209f4fa73e6c28c50 --- /dev/null +++ b/lora/sft-captcha/checkpoint-700/processor_config.json @@ -0,0 +1,60 @@ +{ + "image_processor": { + "do_convert_rgb": true, + "do_normalize": true, + "do_rescale": true, + "do_resize": true, + "image_mean": [ + 0.5, + 0.5, + 0.5 + ], + "image_processor_type": "Qwen2VLImageProcessor", + "image_std": [ + 0.5, + 0.5, + 0.5 + ], + "merge_size": 2, + "patch_size": 16, + "resample": 3, + "rescale_factor": 0.00392156862745098, + "size": { + "longest_edge": 16777216, + "shortest_edge": 65536 + }, + "temporal_patch_size": 2 + }, + "processor_class": "Qwen3VLProcessor", + "video_processor": { + "do_convert_rgb": true, + "do_normalize": true, + "do_rescale": true, + "do_resize": true, + "do_sample_frames": true, + "fps": 2, + "image_mean": [ + 0.5, + 0.5, + 0.5 + ], + "image_std": [ + 0.5, + 0.5, + 0.5 + ], + "max_frames": 768, + "merge_size": 2, + "min_frames": 4, + "patch_size": 16, + "resample": 3, + "rescale_factor": 0.00392156862745098, + "return_metadata": false, + "size": { + "longest_edge": 25165824, + "shortest_edge": 4096 + }, + "temporal_patch_size": 2, + "video_processor_type": "Qwen3VLVideoProcessor" + } +} diff --git a/lora/sft-captcha/checkpoint-700/rng_state.pth b/lora/sft-captcha/checkpoint-700/rng_state.pth new file mode 100644 index 0000000000000000000000000000000000000000..da3f80e9733e799eb63d5be540536c060ca27550 --- /dev/null +++ b/lora/sft-captcha/checkpoint-700/rng_state.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ae18d25634bcbc519cbb38556624bfd0959cd3af155b6441a3b1f1a3fea3a28f +size 14244 diff --git a/lora/sft-captcha/checkpoint-700/scaler.pt b/lora/sft-captcha/checkpoint-700/scaler.pt new file mode 100644 index 0000000000000000000000000000000000000000..c6135025e213a869a87e81da7ad5fa351022c707 --- /dev/null +++ b/lora/sft-captcha/checkpoint-700/scaler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:02a6b6b8cf99b0c2174cd2e01d3f497f96d31938cf35230876ea28c69c51c665 +size 988 diff --git a/lora/sft-captcha/checkpoint-700/scheduler.pt b/lora/sft-captcha/checkpoint-700/scheduler.pt new file mode 100644 index 0000000000000000000000000000000000000000..c5243dedf879fc2810116c5dc693dbe0c2e91c9d --- /dev/null +++ b/lora/sft-captcha/checkpoint-700/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f3c32d392a2222f0dba5abdcf8cd42ea247418c13cc8b220a8f908b72c04cbf2 +size 1064 diff --git a/lora/sft-captcha/checkpoint-700/tokenizer.json b/lora/sft-captcha/checkpoint-700/tokenizer.json new file mode 100644 index 0000000000000000000000000000000000000000..c7afbed2efcdf019f88ab0572ec29d3bf595dfe2 --- /dev/null +++ b/lora/sft-captcha/checkpoint-700/tokenizer.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:be75606093db2094d7cd20f3c2f385c212750648bd6ea4fb2bf507a6a4c55506 +size 11422650 diff --git a/lora/sft-captcha/checkpoint-700/tokenizer_config.json b/lora/sft-captcha/checkpoint-700/tokenizer_config.json new file mode 100644 index 0000000000000000000000000000000000000000..19febf221cc1a71c2b17feb1d29923a875d42d6e --- /dev/null +++ b/lora/sft-captcha/checkpoint-700/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": 262144, + "pad_token": "<|endoftext|>", + "padding_side": "right", + "processor_class": "Qwen3VLProcessor", + "split_special_tokens": false, + "tokenizer_class": "Qwen2Tokenizer", + "unk_token": null +} diff --git a/lora/sft-captcha/checkpoint-700/trainer_state.json b/lora/sft-captcha/checkpoint-700/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..b4f94ed45d83a9c884394c60816c9cc3b496db07 --- /dev/null +++ b/lora/sft-captcha/checkpoint-700/trainer_state.json @@ -0,0 +1,524 @@ +{ + "best_global_step": null, + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 4.516962843295638, + "eval_steps": 500, + "global_step": 700, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.06462035541195477, + "grad_norm": 2.068004846572876, + "learning_rate": 5.76923076923077e-06, + "loss": 0.9769228935241699, + "step": 10 + }, + { + "epoch": 0.12924071082390953, + "grad_norm": 0.9047558307647705, + "learning_rate": 1.217948717948718e-05, + "loss": 0.7594649791717529, + "step": 20 + }, + { + "epoch": 0.1938610662358643, + "grad_norm": 0.4503297805786133, + "learning_rate": 1.858974358974359e-05, + "loss": 0.5557493209838867, + "step": 30 + }, + { + "epoch": 0.25848142164781907, + "grad_norm": 0.37129801511764526, + "learning_rate": 2.5e-05, + "loss": 0.4763153076171875, + "step": 40 + }, + { + "epoch": 0.32310177705977383, + "grad_norm": 0.2953853905200958, + "learning_rate": 3.141025641025641e-05, + "loss": 0.4166147708892822, + "step": 50 + }, + { + "epoch": 0.3877221324717286, + "grad_norm": 0.29160749912261963, + "learning_rate": 3.782051282051282e-05, + "loss": 0.3905663013458252, + "step": 60 + }, + { + "epoch": 0.45234248788368336, + "grad_norm": 0.27939939498901367, + "learning_rate": 4.423076923076923e-05, + "loss": 0.3649837732315063, + "step": 70 + }, + { + "epoch": 0.5169628432956381, + "grad_norm": 0.3130805492401123, + "learning_rate": 4.999974605277764e-05, + "loss": 0.3609442949295044, + "step": 80 + }, + { + "epoch": 0.5815831987075929, + "grad_norm": 0.3720378577709198, + "learning_rate": 4.996927862815294e-05, + "loss": 0.3543627977371216, + "step": 90 + }, + { + "epoch": 0.6462035541195477, + "grad_norm": 0.2810651361942291, + "learning_rate": 4.988809267347141e-05, + "loss": 0.345255446434021, + "step": 100 + }, + { + "epoch": 0.7108239095315024, + "grad_norm": 0.3126155734062195, + "learning_rate": 4.975635309667227e-05, + "loss": 0.33807756900787356, + "step": 110 + }, + { + "epoch": 0.7754442649434572, + "grad_norm": 0.3109033703804016, + "learning_rate": 4.957432749209755e-05, + "loss": 0.3394132614135742, + "step": 120 + }, + { + "epoch": 0.840064620355412, + "grad_norm": 0.3006167411804199, + "learning_rate": 4.934238559694448e-05, + "loss": 0.3244359016418457, + "step": 130 + }, + { + "epoch": 0.9046849757673667, + "grad_norm": 0.3104219138622284, + "learning_rate": 4.90609985402415e-05, + "loss": 0.3221057176589966, + "step": 140 + }, + { + "epoch": 0.9693053311793215, + "grad_norm": 0.330458402633667, + "learning_rate": 4.873073788587338e-05, + "loss": 0.3329624176025391, + "step": 150 + }, + { + "epoch": 1.0323101777059773, + "grad_norm": 0.2666167616844177, + "learning_rate": 4.835227447159939e-05, + "loss": 0.3191133737564087, + "step": 160 + }, + { + "epoch": 1.0969305331179322, + "grad_norm": 0.3217558264732361, + "learning_rate": 4.7926377046422575e-05, + "loss": 0.3055490255355835, + "step": 170 + }, + { + "epoch": 1.1615508885298869, + "grad_norm": 0.31330329179763794, + "learning_rate": 4.7453910709078145e-05, + "loss": 0.31946792602539065, + "step": 180 + }, + { + "epoch": 1.2261712439418417, + "grad_norm": 0.3254242539405823, + "learning_rate": 4.69358351508126e-05, + "loss": 0.3053389549255371, + "step": 190 + }, + { + "epoch": 1.2907915993537964, + "grad_norm": 0.3096122145652771, + "learning_rate": 4.637320270602314e-05, + "loss": 0.30810558795928955, + "step": 200 + }, + { + "epoch": 1.3554119547657513, + "grad_norm": 0.30546557903289795, + "learning_rate": 4.576715621471678e-05, + "loss": 0.31534519195556643, + "step": 210 + }, + { + "epoch": 1.420032310177706, + "grad_norm": 0.32512548565864563, + "learning_rate": 4.511892670113104e-05, + "loss": 0.3046459913253784, + "step": 220 + }, + { + "epoch": 1.4846526655896608, + "grad_norm": 0.32627010345458984, + "learning_rate": 4.442983087323167e-05, + "loss": 0.2983341455459595, + "step": 230 + }, + { + "epoch": 1.5492730210016155, + "grad_norm": 0.3522598147392273, + "learning_rate": 4.370126844816628e-05, + "loss": 0.3133318185806274, + "step": 240 + }, + { + "epoch": 1.6138933764135701, + "grad_norm": 0.30818620324134827, + "learning_rate": 4.293471930910668e-05, + "loss": 0.30533685684204104, + "step": 250 + }, + { + "epoch": 1.678513731825525, + "grad_norm": 0.2774396538734436, + "learning_rate": 4.213174049925498e-05, + "loss": 0.2984048843383789, + "step": 260 + }, + { + "epoch": 1.7431340872374799, + "grad_norm": 0.2943349778652191, + "learning_rate": 4.129396305911934e-05, + "loss": 0.2998037338256836, + "step": 270 + }, + { + "epoch": 1.8077544426494345, + "grad_norm": 0.30737221240997314, + "learning_rate": 4.042308871348371e-05, + "loss": 0.3016969680786133, + "step": 280 + }, + { + "epoch": 1.8723747980613892, + "grad_norm": 0.3105849325656891, + "learning_rate": 3.9520886414800936e-05, + "loss": 0.3014758348464966, + "step": 290 + }, + { + "epoch": 1.936995153473344, + "grad_norm": 0.29112061858177185, + "learning_rate": 3.858918875003053e-05, + "loss": 0.297167444229126, + "step": 300 + }, + { + "epoch": 2.0, + "grad_norm": 0.3953491747379303, + "learning_rate": 3.7629888218219774e-05, + "loss": 0.29504544734954835, + "step": 310 + }, + { + "epoch": 2.0646203554119547, + "grad_norm": 0.29763659834861755, + "learning_rate": 3.664493338638898e-05, + "loss": 0.28955745697021484, + "step": 320 + }, + { + "epoch": 2.1292407108239093, + "grad_norm": 0.3380239009857178, + "learning_rate": 3.5636324931529554e-05, + "loss": 0.28567984104156496, + "step": 330 + }, + { + "epoch": 2.1938610662358644, + "grad_norm": 0.31489691138267517, + "learning_rate": 3.46061115767542e-05, + "loss": 0.28932580947875974, + "step": 340 + }, + { + "epoch": 2.258481421647819, + "grad_norm": 0.3106003403663635, + "learning_rate": 3.3556385929854164e-05, + "loss": 0.2879155158996582, + "step": 350 + }, + { + "epoch": 2.3231017770597737, + "grad_norm": 0.2789410948753357, + "learning_rate": 3.248928023271624e-05, + "loss": 0.2938661813735962, + "step": 360 + }, + { + "epoch": 2.387722132471729, + "grad_norm": 0.2844054102897644, + "learning_rate": 3.140696203023348e-05, + "loss": 0.2907016515731812, + "step": 370 + }, + { + "epoch": 2.4523424878836835, + "grad_norm": 0.3297322392463684, + "learning_rate": 3.0311629767507232e-05, + "loss": 0.28118228912353516, + "step": 380 + }, + { + "epoch": 2.516962843295638, + "grad_norm": 0.30098527669906616, + "learning_rate": 2.920550832428337e-05, + "loss": 0.2846884489059448, + "step": 390 + }, + { + "epoch": 2.581583198707593, + "grad_norm": 0.31197914481163025, + "learning_rate": 2.809084449569365e-05, + "loss": 0.28516063690185545, + "step": 400 + }, + { + "epoch": 2.6462035541195474, + "grad_norm": 0.30557820200920105, + "learning_rate": 2.6969902428481557e-05, + "loss": 0.2813657283782959, + "step": 410 + }, + { + "epoch": 2.7108239095315025, + "grad_norm": 0.3514881730079651, + "learning_rate": 2.584495902198307e-05, + "loss": 0.2873976469039917, + "step": 420 + }, + { + "epoch": 2.775444264943457, + "grad_norm": 0.3010100722312927, + "learning_rate": 2.4718299303203742e-05, + "loss": 0.2792309045791626, + "step": 430 + }, + { + "epoch": 2.840064620355412, + "grad_norm": 0.33192241191864014, + "learning_rate": 2.3592211785386642e-05, + "loss": 0.2784327745437622, + "step": 440 + }, + { + "epoch": 2.904684975767367, + "grad_norm": 0.3093332350254059, + "learning_rate": 2.246898381949894e-05, + "loss": 0.28588788509368895, + "step": 450 + }, + { + "epoch": 2.9693053311793216, + "grad_norm": 0.33140891790390015, + "learning_rate": 2.1350896948079226e-05, + "loss": 0.2833216428756714, + "step": 460 + }, + { + "epoch": 3.0323101777059773, + "grad_norm": 0.31609705090522766, + "learning_rate": 2.0240222270883288e-05, + "loss": 0.28030478954315186, + "step": 470 + }, + { + "epoch": 3.096930533117932, + "grad_norm": 0.2810538113117218, + "learning_rate": 1.913921583174154e-05, + "loss": 0.27578275203704833, + "step": 480 + }, + { + "epoch": 3.161550888529887, + "grad_norm": 0.3165380358695984, + "learning_rate": 1.805011403599862e-05, + "loss": 0.27012245655059813, + "step": 490 + }, + { + "epoch": 3.2261712439418417, + "grad_norm": 0.3220997452735901, + "learning_rate": 1.697512910784352e-05, + "loss": 0.2735466957092285, + "step": 500 + }, + { + "epoch": 3.2907915993537964, + "grad_norm": 0.3499109447002411, + "learning_rate": 1.5916444596757218e-05, + "loss": 0.27160613536834716, + "step": 510 + }, + { + "epoch": 3.355411954765751, + "grad_norm": 0.3451310694217682, + "learning_rate": 1.4876210942205604e-05, + "loss": 0.27280707359313966, + "step": 520 + }, + { + "epoch": 3.420032310177706, + "grad_norm": 0.33574041724205017, + "learning_rate": 1.3856541105586545e-05, + "loss": 0.27128801345825193, + "step": 530 + }, + { + "epoch": 3.484652665589661, + "grad_norm": 0.3565731346607208, + "learning_rate": 1.2859506278303979e-05, + "loss": 0.2736222743988037, + "step": 540 + }, + { + "epoch": 3.5492730210016155, + "grad_norm": 0.32914265990257263, + "learning_rate": 1.1887131674686596e-05, + "loss": 0.27256240844726565, + "step": 550 + }, + { + "epoch": 3.61389337641357, + "grad_norm": 0.31808099150657654, + "learning_rate": 1.0941392418297026e-05, + "loss": 0.27215673923492434, + "step": 560 + }, + { + "epoch": 3.678513731825525, + "grad_norm": 0.3377521336078644, + "learning_rate": 1.0024209529987149e-05, + "loss": 0.2722803592681885, + "step": 570 + }, + { + "epoch": 3.74313408723748, + "grad_norm": 0.33442333340644836, + "learning_rate": 9.137446025848914e-06, + "loss": 0.27033517360687254, + "step": 580 + }, + { + "epoch": 3.8077544426494345, + "grad_norm": 0.32362648844718933, + "learning_rate": 8.282903132986526e-06, + "loss": 0.268932580947876, + "step": 590 + }, + { + "epoch": 3.872374798061389, + "grad_norm": 0.33127161860466003, + "learning_rate": 7.462316630796834e-06, + "loss": 0.274311900138855, + "step": 600 + }, + { + "epoch": 3.936995153473344, + "grad_norm": 0.35842862725257874, + "learning_rate": 6.677353325189409e-06, + "loss": 0.2703368902206421, + "step": 610 + }, + { + "epoch": 4.0, + "grad_norm": 0.4013536870479584, + "learning_rate": 5.929607662908282e-06, + "loss": 0.27242910861968994, + "step": 620 + }, + { + "epoch": 4.064620355411955, + "grad_norm": 0.3234418034553528, + "learning_rate": 5.220598492832171e-06, + "loss": 0.2654104232788086, + "step": 630 + }, + { + "epoch": 4.129240710823909, + "grad_norm": 0.3155248761177063, + "learning_rate": 4.551765980832059e-06, + "loss": 0.26429879665374756, + "step": 640 + }, + { + "epoch": 4.193861066235864, + "grad_norm": 0.3309212327003479, + "learning_rate": 3.924468684452523e-06, + "loss": 0.2637316226959229, + "step": 650 + }, + { + "epoch": 4.258481421647819, + "grad_norm": 0.3667273223400116, + "learning_rate": 3.339980793358982e-06, + "loss": 0.2668980360031128, + "step": 660 + }, + { + "epoch": 4.323101777059774, + "grad_norm": 0.3333350718021393, + "learning_rate": 2.7994895411561333e-06, + "loss": 0.2676853179931641, + "step": 670 + }, + { + "epoch": 4.387722132471729, + "grad_norm": 0.32394319772720337, + "learning_rate": 2.3040927938347528e-06, + "loss": 0.26267526149749754, + "step": 680 + }, + { + "epoch": 4.4523424878836835, + "grad_norm": 0.3705294132232666, + "learning_rate": 1.854796819745383e-06, + "loss": 0.26965837478637694, + "step": 690 + }, + { + "epoch": 4.516962843295638, + "grad_norm": 0.3425300121307373, + "learning_rate": 1.4525142456284779e-06, + "loss": 0.2649491310119629, + "step": 700 + } + ], + "logging_steps": 10, + "max_steps": 775, + "num_input_tokens_seen": 0, + "num_train_epochs": 5, + "save_steps": 100, + "stateful_callbacks": { + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": false + }, + "attributes": {} + } + }, + "total_flos": 6.784722249213542e+17, + "train_batch_size": 4, + "trial_name": null, + "trial_params": null +} diff --git a/lora/sft-captcha/checkpoint-700/training_args.bin b/lora/sft-captcha/checkpoint-700/training_args.bin new file mode 100644 index 0000000000000000000000000000000000000000..fb2a0e0e3f614c6145d25df84879d6af93eeaf5b --- /dev/null +++ b/lora/sft-captcha/checkpoint-700/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:994239c83a21a412f370376b6709e6af83b0c60df0ba98c0dc5d4b0c9b6306e1 +size 5496 diff --git a/lora/sft-captcha/checkpoint-775/README.md b/lora/sft-captcha/checkpoint-775/README.md new file mode 100644 index 0000000000000000000000000000000000000000..776a59eddf85aa2890ebb734de124c8a090269b6 --- /dev/null +++ b/lora/sft-captcha/checkpoint-775/README.md @@ -0,0 +1,208 @@ +--- +base_model: Qwen/Qwen3-VL-4B-Instruct +library_name: peft +pipeline_tag: text-generation +tags: +- base_model:adapter:Qwen/Qwen3-VL-4B-Instruct +- llama-factory +- lora +- transformers +--- + +# Model Card for Model ID + + + + + +## Model Details + +### Model Description + + + + + +- **Developed by:** [More Information Needed] +- **Funded by [optional]:** [More Information Needed] +- **Shared by [optional]:** [More Information Needed] +- **Model type:** [More Information Needed] +- **Language(s) (NLP):** [More Information Needed] +- **License:** [More Information Needed] +- **Finetuned from model [optional]:** [More Information Needed] + +### Model Sources [optional] + + + +- **Repository:** [More Information Needed] +- **Paper [optional]:** [More Information Needed] +- **Demo [optional]:** [More Information Needed] + +## Uses + + + +### Direct Use + + + +[More Information Needed] + +### Downstream Use [optional] + + + +[More Information Needed] + +### Out-of-Scope Use + + + +[More Information Needed] + +## Bias, Risks, and Limitations + + + +[More Information Needed] + +### Recommendations + + + +Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations. + +## How to Get Started with the Model + +Use the code below to get started with the model. + +[More Information Needed] + +## Training Details + +### Training Data + + + +[More Information Needed] + +### Training Procedure + + + +#### Preprocessing [optional] + +[More Information Needed] + + +#### Training Hyperparameters + +- **Training regime:** [More Information Needed] + +#### Speeds, Sizes, Times [optional] + + + +[More Information Needed] + +## Evaluation + + + +### Testing Data, Factors & Metrics + +#### Testing Data + + + +[More Information Needed] + +#### Factors + + + +[More Information Needed] + +#### Metrics + + + +[More Information Needed] + +### Results + +[More Information Needed] + +#### Summary + + + +## Model Examination [optional] + + + +[More Information Needed] + +## Environmental Impact + + + +Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700). + +- **Hardware Type:** [More Information Needed] +- **Hours used:** [More Information Needed] +- **Cloud Provider:** [More Information Needed] +- **Compute Region:** [More Information Needed] +- **Carbon Emitted:** [More Information Needed] + +## Technical Specifications [optional] + +### Model Architecture and Objective + +[More Information Needed] + +### Compute Infrastructure + +[More Information Needed] + +#### Hardware + +[More Information Needed] + +#### Software + +[More Information Needed] + +## Citation [optional] + + + +**BibTeX:** + +[More Information Needed] + +**APA:** + +[More Information Needed] + +## Glossary [optional] + + + +[More Information Needed] + +## More Information [optional] + +[More Information Needed] + +## Model Card Authors [optional] + +[More Information Needed] + +## Model Card Contact + +[More Information Needed] +### Framework versions + +- PEFT 0.18.1 \ No newline at end of file diff --git a/lora/sft-captcha/checkpoint-775/adapter_config.json b/lora/sft-captcha/checkpoint-775/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..91f8411a80c3d6f80ff5a9f2d6d5c6e0952c8e01 --- /dev/null +++ b/lora/sft-captcha/checkpoint-775/adapter_config.json @@ -0,0 +1,46 @@ +{ + "alora_invocation_tokens": null, + "alpha_pattern": {}, + "arrow_config": null, + "auto_mapping": null, + "base_model_name_or_path": "Qwen/Qwen3-VL-4B-Instruct", + "bias": "none", + "corda_config": null, + "ensure_weight_tying": false, + "eva_config": null, + "exclude_modules": null, + "fan_in_fan_out": false, + "inference_mode": true, + "init_lora_weights": true, + "layer_replication": null, + "layers_pattern": null, + "layers_to_transform": null, + "loftq_config": {}, + "lora_alpha": 64, + "lora_bias": false, + "lora_dropout": 0.05, + "megatron_config": null, + "megatron_core": "megatron.core", + "modules_to_save": null, + "peft_type": "LORA", + "peft_version": "0.18.1", + "qalora_group_size": 16, + "r": 32, + "rank_pattern": {}, + "revision": null, + "target_modules": [ + "k_proj", + "v_proj", + "gate_proj", + "up_proj", + "o_proj", + "down_proj", + "q_proj" + ], + "target_parameters": null, + "task_type": "CAUSAL_LM", + "trainable_token_indices": null, + "use_dora": false, + "use_qalora": false, + "use_rslora": false +} \ No newline at end of file diff --git a/lora/sft-captcha/checkpoint-775/adapter_model.safetensors b/lora/sft-captcha/checkpoint-775/adapter_model.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..5177c7f9d30cf56dbef24383fbfe0da65a7aa807 --- /dev/null +++ b/lora/sft-captcha/checkpoint-775/adapter_model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:fbbc568b3b10ea6e601db64bce50e3cdd551083e7288e2c8dd9d2d4220605219 +size 264316456 diff --git a/lora/sft-captcha/checkpoint-775/chat_template.jinja b/lora/sft-captcha/checkpoint-775/chat_template.jinja new file mode 100644 index 0000000000000000000000000000000000000000..124386803f142761528f710e77ae483f5f8c4fc4 --- /dev/null +++ b/lora/sft-captcha/checkpoint-775/chat_template.jinja @@ -0,0 +1,120 @@ +{%- if tools %} + {{- '<|im_start|>system\n' }} + {%- if messages[0].role == 'system' %} + {%- if messages[0].content is string %} + {{- messages[0].content }} + {%- else %} + {%- for content in messages[0].content %} + {%- if 'text' in content %} + {{- content.text }} + {%- endif %} + {%- endfor %} + {%- endif %} + {{- '\n\n' }} + {%- endif %} + {{- "# 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' }} + {%- if messages[0].content is string %} + {{- messages[0].content }} + {%- else %} + {%- for content in messages[0].content %} + {%- if 'text' in content %} + {{- content.text }} + {%- endif %} + {%- endfor %} + {%- endif %} + {{- '<|im_end|>\n' }} + {%- endif %} +{%- endif %} +{%- set image_count = namespace(value=0) %} +{%- set video_count = namespace(value=0) %} +{%- for message in messages %} + {%- if message.role == "user" %} + {{- '<|im_start|>' + message.role + '\n' }} + {%- if message.content is string %} + {{- message.content }} + {%- else %} + {%- for content in message.content %} + {%- if content.type == 'image' or 'image' in content or 'image_url' in content %} + {%- set image_count.value = image_count.value + 1 %} + {%- if add_vision_id %}Picture {{ image_count.value }}: {% endif -%} + <|vision_start|><|image_pad|><|vision_end|> + {%- elif content.type == 'video' or 'video' in content %} + {%- set video_count.value = video_count.value + 1 %} + {%- if add_vision_id %}Video {{ video_count.value }}: {% endif -%} + <|vision_start|><|video_pad|><|vision_end|> + {%- elif 'text' in content %} + {{- content.text }} + {%- endif %} + {%- endfor %} + {%- endif %} + {{- '<|im_end|>\n' }} + {%- elif message.role == "assistant" %} + {{- '<|im_start|>' + message.role + '\n' }} + {%- if message.content is string %} + {{- message.content }} + {%- else %} + {%- for content_item in message.content %} + {%- if 'text' in content_item %} + {{- content_item.text }} + {%- endif %} + {%- endfor %} + {%- endif %} + {%- if message.tool_calls %} + {%- for tool_call in message.tool_calls %} + {%- if (loop.first and message.content) or (not loop.first) %} + {{- '\n' }} + {%- endif %} + {%- if tool_call.function %} + {%- set tool_call = tool_call.function %} + {%- endif %} + {{- '\n{"name": "' }} + {{- tool_call.name }} + {{- '", "arguments": ' }} + {%- if tool_call.arguments is string %} + {{- tool_call.arguments }} + {%- else %} + {{- tool_call.arguments | tojson }} + {%- endif %} + {{- '}\n' }} + {%- endfor %} + {%- endif %} + {{- '<|im_end|>\n' }} + {%- elif message.role == "tool" %} + {%- if loop.first or (messages[loop.index0 - 1].role != "tool") %} + {{- '<|im_start|>user' }} + {%- endif %} + {{- '\n\n' }} + {%- if message.content is string %} + {{- message.content }} + {%- else %} + {%- for content in message.content %} + {%- if content.type == 'image' or 'image' in content or 'image_url' in content %} + {%- set image_count.value = image_count.value + 1 %} + {%- if add_vision_id %}Picture {{ image_count.value }}: {% endif -%} + <|vision_start|><|image_pad|><|vision_end|> + {%- elif content.type == 'video' or 'video' in content %} + {%- set video_count.value = video_count.value + 1 %} + {%- if add_vision_id %}Video {{ video_count.value }}: {% endif -%} + <|vision_start|><|video_pad|><|vision_end|> + {%- elif 'text' in content %} + {{- content.text }} + {%- endif %} + {%- endfor %} + {%- endif %} + {{- '\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/lora/sft-captcha/checkpoint-775/optimizer.pt b/lora/sft-captcha/checkpoint-775/optimizer.pt new file mode 100644 index 0000000000000000000000000000000000000000..b93710ed8c8a7311dc68249219730494631bdeef --- /dev/null +++ b/lora/sft-captcha/checkpoint-775/optimizer.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:725a4fb0cea549358e4d7fbfe9585238cd30a551606347de98f2791dba3001f8 +size 528907898 diff --git a/lora/sft-captcha/checkpoint-775/processor_config.json b/lora/sft-captcha/checkpoint-775/processor_config.json new file mode 100644 index 0000000000000000000000000000000000000000..33818c7f9e991ad735fd240209f4fa73e6c28c50 --- /dev/null +++ b/lora/sft-captcha/checkpoint-775/processor_config.json @@ -0,0 +1,60 @@ +{ + "image_processor": { + "do_convert_rgb": true, + "do_normalize": true, + "do_rescale": true, + "do_resize": true, + "image_mean": [ + 0.5, + 0.5, + 0.5 + ], + "image_processor_type": "Qwen2VLImageProcessor", + "image_std": [ + 0.5, + 0.5, + 0.5 + ], + "merge_size": 2, + "patch_size": 16, + "resample": 3, + "rescale_factor": 0.00392156862745098, + "size": { + "longest_edge": 16777216, + "shortest_edge": 65536 + }, + "temporal_patch_size": 2 + }, + "processor_class": "Qwen3VLProcessor", + "video_processor": { + "do_convert_rgb": true, + "do_normalize": true, + "do_rescale": true, + "do_resize": true, + "do_sample_frames": true, + "fps": 2, + "image_mean": [ + 0.5, + 0.5, + 0.5 + ], + "image_std": [ + 0.5, + 0.5, + 0.5 + ], + "max_frames": 768, + "merge_size": 2, + "min_frames": 4, + "patch_size": 16, + "resample": 3, + "rescale_factor": 0.00392156862745098, + "return_metadata": false, + "size": { + "longest_edge": 25165824, + "shortest_edge": 4096 + }, + "temporal_patch_size": 2, + "video_processor_type": "Qwen3VLVideoProcessor" + } +} diff --git a/lora/sft-captcha/checkpoint-775/rng_state.pth b/lora/sft-captcha/checkpoint-775/rng_state.pth new file mode 100644 index 0000000000000000000000000000000000000000..ac70eb12c20d9001a98d3940f2bbf573287040fd --- /dev/null +++ b/lora/sft-captcha/checkpoint-775/rng_state.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:98c547d08cf23980589afcbc0dbec67b390fc4a6c68656d890ecdf390c8a8cf0 +size 14244 diff --git a/lora/sft-captcha/checkpoint-775/scaler.pt b/lora/sft-captcha/checkpoint-775/scaler.pt new file mode 100644 index 0000000000000000000000000000000000000000..abadb41c59069728adc5e1d04f6504f61364c8a7 --- /dev/null +++ b/lora/sft-captcha/checkpoint-775/scaler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f924b41b4cf872a4b09a3b7b8a16b7dbade463089970effa53a1f8006a4b35c9 +size 988 diff --git a/lora/sft-captcha/checkpoint-775/scheduler.pt b/lora/sft-captcha/checkpoint-775/scheduler.pt new file mode 100644 index 0000000000000000000000000000000000000000..1e6af574369d709aecdcbcdb422b0bc821bf3c3a --- /dev/null +++ b/lora/sft-captcha/checkpoint-775/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:35dec5869836f5b9c808cc50a84891010a12ba0e4ea93da98628d77d5444d78a +size 1064 diff --git a/lora/sft-captcha/checkpoint-775/tokenizer.json b/lora/sft-captcha/checkpoint-775/tokenizer.json new file mode 100644 index 0000000000000000000000000000000000000000..c7afbed2efcdf019f88ab0572ec29d3bf595dfe2 --- /dev/null +++ b/lora/sft-captcha/checkpoint-775/tokenizer.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:be75606093db2094d7cd20f3c2f385c212750648bd6ea4fb2bf507a6a4c55506 +size 11422650 diff --git a/lora/sft-captcha/checkpoint-775/tokenizer_config.json b/lora/sft-captcha/checkpoint-775/tokenizer_config.json new file mode 100644 index 0000000000000000000000000000000000000000..19febf221cc1a71c2b17feb1d29923a875d42d6e --- /dev/null +++ b/lora/sft-captcha/checkpoint-775/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": 262144, + "pad_token": "<|endoftext|>", + "padding_side": "right", + "processor_class": "Qwen3VLProcessor", + "split_special_tokens": false, + "tokenizer_class": "Qwen2Tokenizer", + "unk_token": null +} diff --git a/lora/sft-captcha/checkpoint-775/trainer_state.json b/lora/sft-captcha/checkpoint-775/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..461c90dde228012c986d9fcbf583d065ea7450a5 --- /dev/null +++ b/lora/sft-captcha/checkpoint-775/trainer_state.json @@ -0,0 +1,573 @@ +{ + "best_global_step": null, + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 5.0, + "eval_steps": 500, + "global_step": 775, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.06462035541195477, + "grad_norm": 2.068004846572876, + "learning_rate": 5.76923076923077e-06, + "loss": 0.9769228935241699, + "step": 10 + }, + { + "epoch": 0.12924071082390953, + "grad_norm": 0.9047558307647705, + "learning_rate": 1.217948717948718e-05, + "loss": 0.7594649791717529, + "step": 20 + }, + { + "epoch": 0.1938610662358643, + "grad_norm": 0.4503297805786133, + "learning_rate": 1.858974358974359e-05, + "loss": 0.5557493209838867, + "step": 30 + }, + { + "epoch": 0.25848142164781907, + "grad_norm": 0.37129801511764526, + "learning_rate": 2.5e-05, + "loss": 0.4763153076171875, + "step": 40 + }, + { + "epoch": 0.32310177705977383, + "grad_norm": 0.2953853905200958, + "learning_rate": 3.141025641025641e-05, + "loss": 0.4166147708892822, + "step": 50 + }, + { + "epoch": 0.3877221324717286, + "grad_norm": 0.29160749912261963, + "learning_rate": 3.782051282051282e-05, + "loss": 0.3905663013458252, + "step": 60 + }, + { + "epoch": 0.45234248788368336, + "grad_norm": 0.27939939498901367, + "learning_rate": 4.423076923076923e-05, + "loss": 0.3649837732315063, + "step": 70 + }, + { + "epoch": 0.5169628432956381, + "grad_norm": 0.3130805492401123, + "learning_rate": 4.999974605277764e-05, + "loss": 0.3609442949295044, + "step": 80 + }, + { + "epoch": 0.5815831987075929, + "grad_norm": 0.3720378577709198, + "learning_rate": 4.996927862815294e-05, + "loss": 0.3543627977371216, + "step": 90 + }, + { + "epoch": 0.6462035541195477, + "grad_norm": 0.2810651361942291, + "learning_rate": 4.988809267347141e-05, + "loss": 0.345255446434021, + "step": 100 + }, + { + "epoch": 0.7108239095315024, + "grad_norm": 0.3126155734062195, + "learning_rate": 4.975635309667227e-05, + "loss": 0.33807756900787356, + "step": 110 + }, + { + "epoch": 0.7754442649434572, + "grad_norm": 0.3109033703804016, + "learning_rate": 4.957432749209755e-05, + "loss": 0.3394132614135742, + "step": 120 + }, + { + "epoch": 0.840064620355412, + "grad_norm": 0.3006167411804199, + "learning_rate": 4.934238559694448e-05, + "loss": 0.3244359016418457, + "step": 130 + }, + { + "epoch": 0.9046849757673667, + "grad_norm": 0.3104219138622284, + "learning_rate": 4.90609985402415e-05, + "loss": 0.3221057176589966, + "step": 140 + }, + { + "epoch": 0.9693053311793215, + "grad_norm": 0.330458402633667, + "learning_rate": 4.873073788587338e-05, + "loss": 0.3329624176025391, + "step": 150 + }, + { + "epoch": 1.0323101777059773, + "grad_norm": 0.2666167616844177, + "learning_rate": 4.835227447159939e-05, + "loss": 0.3191133737564087, + "step": 160 + }, + { + "epoch": 1.0969305331179322, + "grad_norm": 0.3217558264732361, + "learning_rate": 4.7926377046422575e-05, + "loss": 0.3055490255355835, + "step": 170 + }, + { + "epoch": 1.1615508885298869, + "grad_norm": 0.31330329179763794, + "learning_rate": 4.7453910709078145e-05, + "loss": 0.31946792602539065, + "step": 180 + }, + { + "epoch": 1.2261712439418417, + "grad_norm": 0.3254242539405823, + "learning_rate": 4.69358351508126e-05, + "loss": 0.3053389549255371, + "step": 190 + }, + { + "epoch": 1.2907915993537964, + "grad_norm": 0.3096122145652771, + "learning_rate": 4.637320270602314e-05, + "loss": 0.30810558795928955, + "step": 200 + }, + { + "epoch": 1.3554119547657513, + "grad_norm": 0.30546557903289795, + "learning_rate": 4.576715621471678e-05, + "loss": 0.31534519195556643, + "step": 210 + }, + { + "epoch": 1.420032310177706, + "grad_norm": 0.32512548565864563, + "learning_rate": 4.511892670113104e-05, + "loss": 0.3046459913253784, + "step": 220 + }, + { + "epoch": 1.4846526655896608, + "grad_norm": 0.32627010345458984, + "learning_rate": 4.442983087323167e-05, + "loss": 0.2983341455459595, + "step": 230 + }, + { + "epoch": 1.5492730210016155, + "grad_norm": 0.3522598147392273, + "learning_rate": 4.370126844816628e-05, + "loss": 0.3133318185806274, + "step": 240 + }, + { + "epoch": 1.6138933764135701, + "grad_norm": 0.30818620324134827, + "learning_rate": 4.293471930910668e-05, + "loss": 0.30533685684204104, + "step": 250 + }, + { + "epoch": 1.678513731825525, + "grad_norm": 0.2774396538734436, + "learning_rate": 4.213174049925498e-05, + "loss": 0.2984048843383789, + "step": 260 + }, + { + "epoch": 1.7431340872374799, + "grad_norm": 0.2943349778652191, + "learning_rate": 4.129396305911934e-05, + "loss": 0.2998037338256836, + "step": 270 + }, + { + "epoch": 1.8077544426494345, + "grad_norm": 0.30737221240997314, + "learning_rate": 4.042308871348371e-05, + "loss": 0.3016969680786133, + "step": 280 + }, + { + "epoch": 1.8723747980613892, + "grad_norm": 0.3105849325656891, + "learning_rate": 3.9520886414800936e-05, + "loss": 0.3014758348464966, + "step": 290 + }, + { + "epoch": 1.936995153473344, + "grad_norm": 0.29112061858177185, + "learning_rate": 3.858918875003053e-05, + "loss": 0.297167444229126, + "step": 300 + }, + { + "epoch": 2.0, + "grad_norm": 0.3953491747379303, + "learning_rate": 3.7629888218219774e-05, + "loss": 0.29504544734954835, + "step": 310 + }, + { + "epoch": 2.0646203554119547, + "grad_norm": 0.29763659834861755, + "learning_rate": 3.664493338638898e-05, + "loss": 0.28955745697021484, + "step": 320 + }, + { + "epoch": 2.1292407108239093, + "grad_norm": 0.3380239009857178, + "learning_rate": 3.5636324931529554e-05, + "loss": 0.28567984104156496, + "step": 330 + }, + { + "epoch": 2.1938610662358644, + "grad_norm": 0.31489691138267517, + "learning_rate": 3.46061115767542e-05, + "loss": 0.28932580947875974, + "step": 340 + }, + { + "epoch": 2.258481421647819, + "grad_norm": 0.3106003403663635, + "learning_rate": 3.3556385929854164e-05, + "loss": 0.2879155158996582, + "step": 350 + }, + { + "epoch": 2.3231017770597737, + "grad_norm": 0.2789410948753357, + "learning_rate": 3.248928023271624e-05, + "loss": 0.2938661813735962, + "step": 360 + }, + { + "epoch": 2.387722132471729, + "grad_norm": 0.2844054102897644, + "learning_rate": 3.140696203023348e-05, + "loss": 0.2907016515731812, + "step": 370 + }, + { + "epoch": 2.4523424878836835, + "grad_norm": 0.3297322392463684, + "learning_rate": 3.0311629767507232e-05, + "loss": 0.28118228912353516, + "step": 380 + }, + { + "epoch": 2.516962843295638, + "grad_norm": 0.30098527669906616, + "learning_rate": 2.920550832428337e-05, + "loss": 0.2846884489059448, + "step": 390 + }, + { + "epoch": 2.581583198707593, + "grad_norm": 0.31197914481163025, + "learning_rate": 2.809084449569365e-05, + "loss": 0.28516063690185545, + "step": 400 + }, + { + "epoch": 2.6462035541195474, + "grad_norm": 0.30557820200920105, + "learning_rate": 2.6969902428481557e-05, + "loss": 0.2813657283782959, + "step": 410 + }, + { + "epoch": 2.7108239095315025, + "grad_norm": 0.3514881730079651, + "learning_rate": 2.584495902198307e-05, + "loss": 0.2873976469039917, + "step": 420 + }, + { + "epoch": 2.775444264943457, + "grad_norm": 0.3010100722312927, + "learning_rate": 2.4718299303203742e-05, + "loss": 0.2792309045791626, + "step": 430 + }, + { + "epoch": 2.840064620355412, + "grad_norm": 0.33192241191864014, + "learning_rate": 2.3592211785386642e-05, + "loss": 0.2784327745437622, + "step": 440 + }, + { + "epoch": 2.904684975767367, + "grad_norm": 0.3093332350254059, + "learning_rate": 2.246898381949894e-05, + "loss": 0.28588788509368895, + "step": 450 + }, + { + "epoch": 2.9693053311793216, + "grad_norm": 0.33140891790390015, + "learning_rate": 2.1350896948079226e-05, + "loss": 0.2833216428756714, + "step": 460 + }, + { + "epoch": 3.0323101777059773, + "grad_norm": 0.31609705090522766, + "learning_rate": 2.0240222270883288e-05, + "loss": 0.28030478954315186, + "step": 470 + }, + { + "epoch": 3.096930533117932, + "grad_norm": 0.2810538113117218, + "learning_rate": 1.913921583174154e-05, + "loss": 0.27578275203704833, + "step": 480 + }, + { + "epoch": 3.161550888529887, + "grad_norm": 0.3165380358695984, + "learning_rate": 1.805011403599862e-05, + "loss": 0.27012245655059813, + "step": 490 + }, + { + "epoch": 3.2261712439418417, + "grad_norm": 0.3220997452735901, + "learning_rate": 1.697512910784352e-05, + "loss": 0.2735466957092285, + "step": 500 + }, + { + "epoch": 3.2907915993537964, + "grad_norm": 0.3499109447002411, + "learning_rate": 1.5916444596757218e-05, + "loss": 0.27160613536834716, + "step": 510 + }, + { + "epoch": 3.355411954765751, + "grad_norm": 0.3451310694217682, + "learning_rate": 1.4876210942205604e-05, + "loss": 0.27280707359313966, + "step": 520 + }, + { + "epoch": 3.420032310177706, + "grad_norm": 0.33574041724205017, + "learning_rate": 1.3856541105586545e-05, + "loss": 0.27128801345825193, + "step": 530 + }, + { + "epoch": 3.484652665589661, + "grad_norm": 0.3565731346607208, + "learning_rate": 1.2859506278303979e-05, + "loss": 0.2736222743988037, + "step": 540 + }, + { + "epoch": 3.5492730210016155, + "grad_norm": 0.32914265990257263, + "learning_rate": 1.1887131674686596e-05, + "loss": 0.27256240844726565, + "step": 550 + }, + { + "epoch": 3.61389337641357, + "grad_norm": 0.31808099150657654, + "learning_rate": 1.0941392418297026e-05, + "loss": 0.27215673923492434, + "step": 560 + }, + { + "epoch": 3.678513731825525, + "grad_norm": 0.3377521336078644, + "learning_rate": 1.0024209529987149e-05, + "loss": 0.2722803592681885, + "step": 570 + }, + { + "epoch": 3.74313408723748, + "grad_norm": 0.33442333340644836, + "learning_rate": 9.137446025848914e-06, + "loss": 0.27033517360687254, + "step": 580 + }, + { + "epoch": 3.8077544426494345, + "grad_norm": 0.32362648844718933, + "learning_rate": 8.282903132986526e-06, + "loss": 0.268932580947876, + "step": 590 + }, + { + "epoch": 3.872374798061389, + "grad_norm": 0.33127161860466003, + "learning_rate": 7.462316630796834e-06, + "loss": 0.274311900138855, + "step": 600 + }, + { + "epoch": 3.936995153473344, + "grad_norm": 0.35842862725257874, + "learning_rate": 6.677353325189409e-06, + "loss": 0.2703368902206421, + "step": 610 + }, + { + "epoch": 4.0, + "grad_norm": 0.4013536870479584, + "learning_rate": 5.929607662908282e-06, + "loss": 0.27242910861968994, + "step": 620 + }, + { + "epoch": 4.064620355411955, + "grad_norm": 0.3234418034553528, + "learning_rate": 5.220598492832171e-06, + "loss": 0.2654104232788086, + "step": 630 + }, + { + "epoch": 4.129240710823909, + "grad_norm": 0.3155248761177063, + "learning_rate": 4.551765980832059e-06, + "loss": 0.26429879665374756, + "step": 640 + }, + { + "epoch": 4.193861066235864, + "grad_norm": 0.3309212327003479, + "learning_rate": 3.924468684452523e-06, + "loss": 0.2637316226959229, + "step": 650 + }, + { + "epoch": 4.258481421647819, + "grad_norm": 0.3667273223400116, + "learning_rate": 3.339980793358982e-06, + "loss": 0.2668980360031128, + "step": 660 + }, + { + "epoch": 4.323101777059774, + "grad_norm": 0.3333350718021393, + "learning_rate": 2.7994895411561333e-06, + "loss": 0.2676853179931641, + "step": 670 + }, + { + "epoch": 4.387722132471729, + "grad_norm": 0.32394319772720337, + "learning_rate": 2.3040927938347528e-06, + "loss": 0.26267526149749754, + "step": 680 + }, + { + "epoch": 4.4523424878836835, + "grad_norm": 0.3705294132232666, + "learning_rate": 1.854796819745383e-06, + "loss": 0.26965837478637694, + "step": 690 + }, + { + "epoch": 4.516962843295638, + "grad_norm": 0.3425300121307373, + "learning_rate": 1.4525142456284779e-06, + "loss": 0.2649491310119629, + "step": 700 + }, + { + "epoch": 4.581583198707593, + "grad_norm": 0.32138586044311523, + "learning_rate": 1.098062202852984e-06, + "loss": 0.264300274848938, + "step": 710 + }, + { + "epoch": 4.646203554119547, + "grad_norm": 0.34855136275291443, + "learning_rate": 7.921606676285776e-07, + "loss": 0.2624504804611206, + "step": 720 + }, + { + "epoch": 4.710823909531502, + "grad_norm": 0.33398354053497314, + "learning_rate": 5.354309985631605e-07, + "loss": 0.2657909870147705, + "step": 730 + }, + { + "epoch": 4.775444264943458, + "grad_norm": 0.32164713740348816, + "learning_rate": 3.283946745360755e-07, + "loss": 0.2667935609817505, + "step": 740 + }, + { + "epoch": 4.840064620355412, + "grad_norm": 0.3328765332698822, + "learning_rate": 1.714722354507592e-07, + "loss": 0.26322848796844484, + "step": 750 + }, + { + "epoch": 4.904684975767367, + "grad_norm": 0.3057484030723572, + "learning_rate": 6.498242801838694e-08, + "loss": 0.26327192783355713, + "step": 760 + }, + { + "epoch": 4.969305331179322, + "grad_norm": 0.3382052183151245, + "learning_rate": 9.141558307687592e-09, + "loss": 0.2605318546295166, + "step": 770 + } + ], + "logging_steps": 10, + "max_steps": 775, + "num_input_tokens_seen": 0, + "num_train_epochs": 5, + "save_steps": 100, + "stateful_callbacks": { + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": true + }, + "attributes": {} + } + }, + "total_flos": 7.509104074275226e+17, + "train_batch_size": 4, + "trial_name": null, + "trial_params": null +} diff --git a/lora/sft-captcha/checkpoint-775/training_args.bin b/lora/sft-captcha/checkpoint-775/training_args.bin new file mode 100644 index 0000000000000000000000000000000000000000..fb2a0e0e3f614c6145d25df84879d6af93eeaf5b --- /dev/null +++ b/lora/sft-captcha/checkpoint-775/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:994239c83a21a412f370376b6709e6af83b0c60df0ba98c0dc5d4b0c9b6306e1 +size 5496 diff --git a/lora/sft-captcha/processor_config.json b/lora/sft-captcha/processor_config.json new file mode 100644 index 0000000000000000000000000000000000000000..33818c7f9e991ad735fd240209f4fa73e6c28c50 --- /dev/null +++ b/lora/sft-captcha/processor_config.json @@ -0,0 +1,60 @@ +{ + "image_processor": { + "do_convert_rgb": true, + "do_normalize": true, + "do_rescale": true, + "do_resize": true, + "image_mean": [ + 0.5, + 0.5, + 0.5 + ], + "image_processor_type": "Qwen2VLImageProcessor", + "image_std": [ + 0.5, + 0.5, + 0.5 + ], + "merge_size": 2, + "patch_size": 16, + "resample": 3, + "rescale_factor": 0.00392156862745098, + "size": { + "longest_edge": 16777216, + "shortest_edge": 65536 + }, + "temporal_patch_size": 2 + }, + "processor_class": "Qwen3VLProcessor", + "video_processor": { + "do_convert_rgb": true, + "do_normalize": true, + "do_rescale": true, + "do_resize": true, + "do_sample_frames": true, + "fps": 2, + "image_mean": [ + 0.5, + 0.5, + 0.5 + ], + "image_std": [ + 0.5, + 0.5, + 0.5 + ], + "max_frames": 768, + "merge_size": 2, + "min_frames": 4, + "patch_size": 16, + "resample": 3, + "rescale_factor": 0.00392156862745098, + "return_metadata": false, + "size": { + "longest_edge": 25165824, + "shortest_edge": 4096 + }, + "temporal_patch_size": 2, + "video_processor_type": "Qwen3VLVideoProcessor" + } +} diff --git a/lora/sft-captcha/tokenizer.json b/lora/sft-captcha/tokenizer.json new file mode 100644 index 0000000000000000000000000000000000000000..c7afbed2efcdf019f88ab0572ec29d3bf595dfe2 --- /dev/null +++ b/lora/sft-captcha/tokenizer.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:be75606093db2094d7cd20f3c2f385c212750648bd6ea4fb2bf507a6a4c55506 +size 11422650 diff --git a/lora/sft-captcha/tokenizer_config.json b/lora/sft-captcha/tokenizer_config.json new file mode 100644 index 0000000000000000000000000000000000000000..19febf221cc1a71c2b17feb1d29923a875d42d6e --- /dev/null +++ b/lora/sft-captcha/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": 262144, + "pad_token": "<|endoftext|>", + "padding_side": "right", + "processor_class": "Qwen3VLProcessor", + "split_special_tokens": false, + "tokenizer_class": "Qwen2Tokenizer", + "unk_token": null +} diff --git a/lora/sft-captcha/train_results.json b/lora/sft-captcha/train_results.json new file mode 100644 index 0000000000000000000000000000000000000000..d4dc19bd6c47cfa49c4f3b5e91ac214ed06d61de --- /dev/null +++ b/lora/sft-captcha/train_results.json @@ -0,0 +1,8 @@ +{ + "epoch": 5.0, + "total_flos": 7.509104074275226e+17, + "train_loss": 0.31352297429115544, + "train_runtime": 25009.3076, + "train_samples_per_second": 0.99, + "train_steps_per_second": 0.031 +} \ No newline at end of file diff --git a/lora/sft-captcha/trainer_log.jsonl b/lora/sft-captcha/trainer_log.jsonl new file mode 100644 index 0000000000000000000000000000000000000000..ae1d7bb71250806147aeaffc8435e5eb5a1deee7 --- /dev/null +++ b/lora/sft-captcha/trainer_log.jsonl @@ -0,0 +1,78 @@ +{"current_steps": 10, "total_steps": 775, "loss": 0.9769228935241699, "lr": 5.76923076923077e-06, "epoch": 0.06462035541195477, "percentage": 1.29, "elapsed_time": "0:05:23", "remaining_time": "6:51:51"} +{"current_steps": 20, "total_steps": 775, "loss": 0.7594649791717529, "lr": 1.217948717948718e-05, "epoch": 0.12924071082390953, "percentage": 2.58, "elapsed_time": "0:10:49", "remaining_time": "6:48:36"} +{"current_steps": 30, "total_steps": 775, "loss": 0.5557493209838867, "lr": 1.858974358974359e-05, "epoch": 0.1938610662358643, "percentage": 3.87, "elapsed_time": "0:16:12", "remaining_time": "6:42:25"} +{"current_steps": 40, "total_steps": 775, "loss": 0.4763153076171875, "lr": 2.5e-05, "epoch": 0.25848142164781907, "percentage": 5.16, "elapsed_time": "0:21:37", "remaining_time": "6:37:17"} +{"current_steps": 50, "total_steps": 775, "loss": 0.4166147708892822, "lr": 3.141025641025641e-05, "epoch": 0.32310177705977383, "percentage": 6.45, "elapsed_time": "0:26:57", "remaining_time": "6:30:53"} +{"current_steps": 60, "total_steps": 775, "loss": 0.3905663013458252, "lr": 3.782051282051282e-05, "epoch": 0.3877221324717286, "percentage": 7.74, "elapsed_time": "0:32:18", "remaining_time": "6:25:04"} +{"current_steps": 70, "total_steps": 775, "loss": 0.3649837732315063, "lr": 4.423076923076923e-05, "epoch": 0.45234248788368336, "percentage": 9.03, "elapsed_time": "0:37:43", "remaining_time": "6:19:52"} +{"current_steps": 80, "total_steps": 775, "loss": 0.3609442949295044, "lr": 4.999974605277764e-05, "epoch": 0.5169628432956381, "percentage": 10.32, "elapsed_time": "0:43:06", "remaining_time": "6:14:34"} +{"current_steps": 90, "total_steps": 775, "loss": 0.3543627977371216, "lr": 4.996927862815294e-05, "epoch": 0.5815831987075929, "percentage": 11.61, "elapsed_time": "0:48:29", "remaining_time": "6:09:08"} +{"current_steps": 100, "total_steps": 775, "loss": 0.345255446434021, "lr": 4.988809267347141e-05, "epoch": 0.6462035541195477, "percentage": 12.9, "elapsed_time": "0:53:54", "remaining_time": "6:03:51"} +{"current_steps": 110, "total_steps": 775, "loss": 0.33807756900787356, "lr": 4.975635309667227e-05, "epoch": 0.7108239095315024, "percentage": 14.19, "elapsed_time": "0:59:15", "remaining_time": "5:58:11"} +{"current_steps": 120, "total_steps": 775, "loss": 0.3394132614135742, "lr": 4.957432749209755e-05, "epoch": 0.7754442649434572, "percentage": 15.48, "elapsed_time": "1:04:37", "remaining_time": "5:52:44"} +{"current_steps": 130, "total_steps": 775, "loss": 0.3244359016418457, "lr": 4.934238559694448e-05, "epoch": 0.840064620355412, "percentage": 16.77, "elapsed_time": "1:10:00", "remaining_time": "5:47:20"} +{"current_steps": 140, "total_steps": 775, "loss": 0.3221057176589966, "lr": 4.90609985402415e-05, "epoch": 0.9046849757673667, "percentage": 18.06, "elapsed_time": "1:15:22", "remaining_time": "5:41:50"} +{"current_steps": 150, "total_steps": 775, "loss": 0.3329624176025391, "lr": 4.873073788587338e-05, "epoch": 0.9693053311793215, "percentage": 19.35, "elapsed_time": "1:20:48", "remaining_time": "5:36:41"} +{"current_steps": 160, "total_steps": 775, "loss": 0.3191133737564087, "lr": 4.835227447159939e-05, "epoch": 1.0323101777059773, "percentage": 20.65, "elapsed_time": "1:26:01", "remaining_time": "5:30:39"} +{"current_steps": 170, "total_steps": 775, "loss": 0.3055490255355835, "lr": 4.7926377046422575e-05, "epoch": 1.0969305331179322, "percentage": 21.94, "elapsed_time": "1:31:25", "remaining_time": "5:25:20"} +{"current_steps": 180, "total_steps": 775, "loss": 0.31946792602539065, "lr": 4.7453910709078145e-05, "epoch": 1.1615508885298869, "percentage": 23.23, "elapsed_time": "1:36:49", "remaining_time": "5:20:03"} +{"current_steps": 190, "total_steps": 775, "loss": 0.3053389549255371, "lr": 4.69358351508126e-05, "epoch": 1.2261712439418417, "percentage": 24.52, "elapsed_time": "1:42:14", "remaining_time": "5:14:46"} +{"current_steps": 200, "total_steps": 775, "loss": 0.30810558795928955, "lr": 4.637320270602314e-05, "epoch": 1.2907915993537964, "percentage": 25.81, "elapsed_time": "1:47:37", "remaining_time": "5:09:23"} +{"current_steps": 210, "total_steps": 775, "loss": 0.31534519195556643, "lr": 4.576715621471678e-05, "epoch": 1.3554119547657513, "percentage": 27.1, "elapsed_time": "1:52:58", "remaining_time": "5:03:58"} +{"current_steps": 220, "total_steps": 775, "loss": 0.3046459913253784, "lr": 4.511892670113104e-05, "epoch": 1.420032310177706, "percentage": 28.39, "elapsed_time": "1:58:17", "remaining_time": "4:58:26"} +{"current_steps": 230, "total_steps": 775, "loss": 0.2983341455459595, "lr": 4.442983087323167e-05, "epoch": 1.4846526655896608, "percentage": 29.68, "elapsed_time": "2:03:40", "remaining_time": "4:53:02"} +{"current_steps": 240, "total_steps": 775, "loss": 0.3133318185806274, "lr": 4.370126844816628e-05, "epoch": 1.5492730210016155, "percentage": 30.97, "elapsed_time": "2:09:02", "remaining_time": "4:47:39"} +{"current_steps": 250, "total_steps": 775, "loss": 0.30533685684204104, "lr": 4.293471930910668e-05, "epoch": 1.6138933764135701, "percentage": 32.26, "elapsed_time": "2:14:24", "remaining_time": "4:42:15"} +{"current_steps": 260, "total_steps": 775, "loss": 0.2984048843383789, "lr": 4.213174049925498e-05, "epoch": 1.678513731825525, "percentage": 33.55, "elapsed_time": "2:19:43", "remaining_time": "4:36:46"} +{"current_steps": 270, "total_steps": 775, "loss": 0.2998037338256836, "lr": 4.129396305911934e-05, "epoch": 1.7431340872374799, "percentage": 34.84, "elapsed_time": "2:25:10", "remaining_time": "4:31:31"} +{"current_steps": 280, "total_steps": 775, "loss": 0.3016969680786133, "lr": 4.042308871348371e-05, "epoch": 1.8077544426494345, "percentage": 36.13, "elapsed_time": "2:30:32", "remaining_time": "4:26:08"} +{"current_steps": 290, "total_steps": 775, "loss": 0.3014758348464966, "lr": 3.9520886414800936e-05, "epoch": 1.8723747980613892, "percentage": 37.42, "elapsed_time": "2:35:57", "remaining_time": "4:20:50"} +{"current_steps": 300, "total_steps": 775, "loss": 0.297167444229126, "lr": 3.858918875003053e-05, "epoch": 1.936995153473344, "percentage": 38.71, "elapsed_time": "2:41:22", "remaining_time": "4:15:31"} +{"current_steps": 310, "total_steps": 775, "loss": 0.29504544734954835, "lr": 3.7629888218219774e-05, "epoch": 2.0, "percentage": 40.0, "elapsed_time": "2:46:36", "remaining_time": "4:09:54"} +{"current_steps": 320, "total_steps": 775, "loss": 0.28955745697021484, "lr": 3.664493338638898e-05, "epoch": 2.0646203554119547, "percentage": 41.29, "elapsed_time": "2:51:57", "remaining_time": "4:04:30"} +{"current_steps": 330, "total_steps": 775, "loss": 0.28567984104156496, "lr": 3.5636324931529554e-05, "epoch": 2.1292407108239093, "percentage": 42.58, "elapsed_time": "2:57:20", "remaining_time": "3:59:08"} +{"current_steps": 340, "total_steps": 775, "loss": 0.28932580947875974, "lr": 3.46061115767542e-05, "epoch": 2.1938610662358644, "percentage": 43.87, "elapsed_time": "3:02:41", "remaining_time": "3:53:44"} +{"current_steps": 350, "total_steps": 775, "loss": 0.2879155158996582, "lr": 3.3556385929854164e-05, "epoch": 2.258481421647819, "percentage": 45.16, "elapsed_time": "3:08:02", "remaining_time": "3:48:20"} +{"current_steps": 360, "total_steps": 775, "loss": 0.2938661813735962, "lr": 3.248928023271624e-05, "epoch": 2.3231017770597737, "percentage": 46.45, "elapsed_time": "3:13:30", "remaining_time": "3:43:04"} +{"current_steps": 370, "total_steps": 775, "loss": 0.2907016515731812, "lr": 3.140696203023348e-05, "epoch": 2.387722132471729, "percentage": 47.74, "elapsed_time": "3:18:53", "remaining_time": "3:37:41"} +{"current_steps": 380, "total_steps": 775, "loss": 0.28118228912353516, "lr": 3.0311629767507232e-05, "epoch": 2.4523424878836835, "percentage": 49.03, "elapsed_time": "3:24:18", "remaining_time": "3:32:22"} +{"current_steps": 390, "total_steps": 775, "loss": 0.2846884489059448, "lr": 2.920550832428337e-05, "epoch": 2.516962843295638, "percentage": 50.32, "elapsed_time": "3:29:43", "remaining_time": "3:27:02"} +{"current_steps": 400, "total_steps": 775, "loss": 0.28516063690185545, "lr": 2.809084449569365e-05, "epoch": 2.581583198707593, "percentage": 51.61, "elapsed_time": "3:35:05", "remaining_time": "3:21:38"} +{"current_steps": 410, "total_steps": 775, "loss": 0.2813657283782959, "lr": 2.6969902428481557e-05, "epoch": 2.6462035541195474, "percentage": 52.9, "elapsed_time": "3:40:33", "remaining_time": "3:16:20"} +{"current_steps": 420, "total_steps": 775, "loss": 0.2873976469039917, "lr": 2.584495902198307e-05, "epoch": 2.7108239095315025, "percentage": 54.19, "elapsed_time": "3:45:56", "remaining_time": "3:10:58"} +{"current_steps": 430, "total_steps": 775, "loss": 0.2792309045791626, "lr": 2.4718299303203742e-05, "epoch": 2.775444264943457, "percentage": 55.48, "elapsed_time": "3:51:20", "remaining_time": "3:05:36"} +{"current_steps": 440, "total_steps": 775, "loss": 0.2784327745437622, "lr": 2.3592211785386642e-05, "epoch": 2.840064620355412, "percentage": 56.77, "elapsed_time": "3:56:44", "remaining_time": "3:00:14"} +{"current_steps": 450, "total_steps": 775, "loss": 0.28588788509368895, "lr": 2.246898381949894e-05, "epoch": 2.904684975767367, "percentage": 58.06, "elapsed_time": "4:02:06", "remaining_time": "2:54:51"} +{"current_steps": 460, "total_steps": 775, "loss": 0.2833216428756714, "lr": 2.1350896948079226e-05, "epoch": 2.9693053311793216, "percentage": 59.35, "elapsed_time": "4:07:28", "remaining_time": "2:49:28"} +{"current_steps": 470, "total_steps": 775, "loss": 0.28030478954315186, "lr": 2.0240222270883288e-05, "epoch": 3.0323101777059773, "percentage": 60.65, "elapsed_time": "4:12:40", "remaining_time": "2:43:58"} +{"current_steps": 480, "total_steps": 775, "loss": 0.27578275203704833, "lr": 1.913921583174154e-05, "epoch": 3.096930533117932, "percentage": 61.94, "elapsed_time": "4:18:05", "remaining_time": "2:38:37"} +{"current_steps": 490, "total_steps": 775, "loss": 0.27012245655059813, "lr": 1.805011403599862e-05, "epoch": 3.161550888529887, "percentage": 63.23, "elapsed_time": "4:23:29", "remaining_time": "2:33:15"} +{"current_steps": 500, "total_steps": 775, "loss": 0.2735466957092285, "lr": 1.697512910784352e-05, "epoch": 3.2261712439418417, "percentage": 64.52, "elapsed_time": "4:28:53", "remaining_time": "2:27:53"} +{"current_steps": 510, "total_steps": 775, "loss": 0.27160613536834716, "lr": 1.5916444596757218e-05, "epoch": 3.2907915993537964, "percentage": 65.81, "elapsed_time": "4:34:15", "remaining_time": "2:22:30"} +{"current_steps": 520, "total_steps": 775, "loss": 0.27280707359313966, "lr": 1.4876210942205604e-05, "epoch": 3.355411954765751, "percentage": 67.1, "elapsed_time": "4:39:42", "remaining_time": "2:17:10"} +{"current_steps": 530, "total_steps": 775, "loss": 0.27128801345825193, "lr": 1.3856541105586545e-05, "epoch": 3.420032310177706, "percentage": 68.39, "elapsed_time": "4:45:04", "remaining_time": "2:11:46"} +{"current_steps": 540, "total_steps": 775, "loss": 0.2736222743988037, "lr": 1.2859506278303979e-05, "epoch": 3.484652665589661, "percentage": 69.68, "elapsed_time": "4:50:24", "remaining_time": "2:06:22"} +{"current_steps": 550, "total_steps": 775, "loss": 0.27256240844726565, "lr": 1.1887131674686596e-05, "epoch": 3.5492730210016155, "percentage": 70.97, "elapsed_time": "4:55:47", "remaining_time": "2:01:00"} +{"current_steps": 560, "total_steps": 775, "loss": 0.27215673923492434, "lr": 1.0941392418297026e-05, "epoch": 3.61389337641357, "percentage": 72.26, "elapsed_time": "5:01:10", "remaining_time": "1:55:37"} +{"current_steps": 570, "total_steps": 775, "loss": 0.2722803592681885, "lr": 1.0024209529987149e-05, "epoch": 3.678513731825525, "percentage": 73.55, "elapsed_time": "5:06:34", "remaining_time": "1:50:15"} +{"current_steps": 580, "total_steps": 775, "loss": 0.27033517360687254, "lr": 9.137446025848914e-06, "epoch": 3.74313408723748, "percentage": 74.84, "elapsed_time": "5:11:57", "remaining_time": "1:44:53"} +{"current_steps": 590, "total_steps": 775, "loss": 0.268932580947876, "lr": 8.282903132986526e-06, "epoch": 3.8077544426494345, "percentage": 76.13, "elapsed_time": "5:17:24", "remaining_time": "1:39:31"} +{"current_steps": 600, "total_steps": 775, "loss": 0.274311900138855, "lr": 7.462316630796834e-06, "epoch": 3.872374798061389, "percentage": 77.42, "elapsed_time": "5:22:50", "remaining_time": "1:34:09"} +{"current_steps": 610, "total_steps": 775, "loss": 0.2703368902206421, "lr": 6.677353325189409e-06, "epoch": 3.936995153473344, "percentage": 78.71, "elapsed_time": "5:28:13", "remaining_time": "1:28:46"} +{"current_steps": 620, "total_steps": 775, "loss": 0.27242910861968994, "lr": 5.929607662908282e-06, "epoch": 4.0, "percentage": 80.0, "elapsed_time": "5:33:27", "remaining_time": "1:23:21"} +{"current_steps": 630, "total_steps": 775, "loss": 0.2654104232788086, "lr": 5.220598492832171e-06, "epoch": 4.064620355411955, "percentage": 81.29, "elapsed_time": "5:38:51", "remaining_time": "1:17:59"} +{"current_steps": 640, "total_steps": 775, "loss": 0.26429879665374756, "lr": 4.551765980832059e-06, "epoch": 4.129240710823909, "percentage": 82.58, "elapsed_time": "5:44:11", "remaining_time": "1:12:36"} +{"current_steps": 650, "total_steps": 775, "loss": 0.2637316226959229, "lr": 3.924468684452523e-06, "epoch": 4.193861066235864, "percentage": 83.87, "elapsed_time": "5:49:33", "remaining_time": "1:07:13"} +{"current_steps": 660, "total_steps": 775, "loss": 0.2668980360031128, "lr": 3.339980793358982e-06, "epoch": 4.258481421647819, "percentage": 85.16, "elapsed_time": "5:54:59", "remaining_time": "1:01:51"} +{"current_steps": 670, "total_steps": 775, "loss": 0.2676853179931641, "lr": 2.7994895411561333e-06, "epoch": 4.323101777059774, "percentage": 86.45, "elapsed_time": "6:00:24", "remaining_time": "0:56:28"} +{"current_steps": 680, "total_steps": 775, "loss": 0.26267526149749754, "lr": 2.3040927938347528e-06, "epoch": 4.387722132471729, "percentage": 87.74, "elapsed_time": "6:05:46", "remaining_time": "0:51:06"} +{"current_steps": 690, "total_steps": 775, "loss": 0.26965837478637694, "lr": 1.854796819745383e-06, "epoch": 4.4523424878836835, "percentage": 89.03, "elapsed_time": "6:11:10", "remaining_time": "0:45:43"} +{"current_steps": 700, "total_steps": 775, "loss": 0.2649491310119629, "lr": 1.4525142456284779e-06, "epoch": 4.516962843295638, "percentage": 90.32, "elapsed_time": "6:16:35", "remaining_time": "0:40:20"} +{"current_steps": 710, "total_steps": 775, "loss": 0.264300274848938, "lr": 1.098062202852984e-06, "epoch": 4.581583198707593, "percentage": 91.61, "elapsed_time": "6:21:57", "remaining_time": "0:34:58"} +{"current_steps": 720, "total_steps": 775, "loss": 0.2624504804611206, "lr": 7.921606676285776e-07, "epoch": 4.646203554119547, "percentage": 92.9, "elapsed_time": "6:27:22", "remaining_time": "0:29:35"} +{"current_steps": 730, "total_steps": 775, "loss": 0.2657909870147705, "lr": 5.354309985631605e-07, "epoch": 4.710823909531502, "percentage": 94.19, "elapsed_time": "6:32:46", "remaining_time": "0:24:12"} +{"current_steps": 740, "total_steps": 775, "loss": 0.2667935609817505, "lr": 3.283946745360755e-07, "epoch": 4.775444264943458, "percentage": 95.48, "elapsed_time": "6:38:09", "remaining_time": "0:18:49"} +{"current_steps": 750, "total_steps": 775, "loss": 0.26322848796844484, "lr": 1.714722354507592e-07, "epoch": 4.840064620355412, "percentage": 96.77, "elapsed_time": "6:43:29", "remaining_time": "0:13:26"} +{"current_steps": 760, "total_steps": 775, "loss": 0.26327192783355713, "lr": 6.498242801838694e-08, "epoch": 4.904684975767367, "percentage": 98.06, "elapsed_time": "6:48:52", "remaining_time": "0:08:04"} +{"current_steps": 770, "total_steps": 775, "loss": 0.2605318546295166, "lr": 9.141558307687592e-09, "epoch": 4.969305331179322, "percentage": 99.35, "elapsed_time": "6:54:14", "remaining_time": "0:02:41"} +{"current_steps": 775, "total_steps": 775, "epoch": 5.0, "percentage": 100.0, "elapsed_time": "6:56:49", "remaining_time": "0:00:00"} diff --git a/lora/sft-captcha/trainer_state.json b/lora/sft-captcha/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..10a836dd720ddd0bcf985f3561e56fa8eb3e1d00 --- /dev/null +++ b/lora/sft-captcha/trainer_state.json @@ -0,0 +1,582 @@ +{ + "best_global_step": null, + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 5.0, + "eval_steps": 500, + "global_step": 775, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.06462035541195477, + "grad_norm": 2.068004846572876, + "learning_rate": 5.76923076923077e-06, + "loss": 0.9769228935241699, + "step": 10 + }, + { + "epoch": 0.12924071082390953, + "grad_norm": 0.9047558307647705, + "learning_rate": 1.217948717948718e-05, + "loss": 0.7594649791717529, + "step": 20 + }, + { + "epoch": 0.1938610662358643, + "grad_norm": 0.4503297805786133, + "learning_rate": 1.858974358974359e-05, + "loss": 0.5557493209838867, + "step": 30 + }, + { + "epoch": 0.25848142164781907, + "grad_norm": 0.37129801511764526, + "learning_rate": 2.5e-05, + "loss": 0.4763153076171875, + "step": 40 + }, + { + "epoch": 0.32310177705977383, + "grad_norm": 0.2953853905200958, + "learning_rate": 3.141025641025641e-05, + "loss": 0.4166147708892822, + "step": 50 + }, + { + "epoch": 0.3877221324717286, + "grad_norm": 0.29160749912261963, + "learning_rate": 3.782051282051282e-05, + "loss": 0.3905663013458252, + "step": 60 + }, + { + "epoch": 0.45234248788368336, + "grad_norm": 0.27939939498901367, + "learning_rate": 4.423076923076923e-05, + "loss": 0.3649837732315063, + "step": 70 + }, + { + "epoch": 0.5169628432956381, + "grad_norm": 0.3130805492401123, + "learning_rate": 4.999974605277764e-05, + "loss": 0.3609442949295044, + "step": 80 + }, + { + "epoch": 0.5815831987075929, + "grad_norm": 0.3720378577709198, + "learning_rate": 4.996927862815294e-05, + "loss": 0.3543627977371216, + "step": 90 + }, + { + "epoch": 0.6462035541195477, + "grad_norm": 0.2810651361942291, + "learning_rate": 4.988809267347141e-05, + "loss": 0.345255446434021, + "step": 100 + }, + { + "epoch": 0.7108239095315024, + "grad_norm": 0.3126155734062195, + "learning_rate": 4.975635309667227e-05, + "loss": 0.33807756900787356, + "step": 110 + }, + { + "epoch": 0.7754442649434572, + "grad_norm": 0.3109033703804016, + "learning_rate": 4.957432749209755e-05, + "loss": 0.3394132614135742, + "step": 120 + }, + { + "epoch": 0.840064620355412, + "grad_norm": 0.3006167411804199, + "learning_rate": 4.934238559694448e-05, + "loss": 0.3244359016418457, + "step": 130 + }, + { + "epoch": 0.9046849757673667, + "grad_norm": 0.3104219138622284, + "learning_rate": 4.90609985402415e-05, + "loss": 0.3221057176589966, + "step": 140 + }, + { + "epoch": 0.9693053311793215, + "grad_norm": 0.330458402633667, + "learning_rate": 4.873073788587338e-05, + "loss": 0.3329624176025391, + "step": 150 + }, + { + "epoch": 1.0323101777059773, + "grad_norm": 0.2666167616844177, + "learning_rate": 4.835227447159939e-05, + "loss": 0.3191133737564087, + "step": 160 + }, + { + "epoch": 1.0969305331179322, + "grad_norm": 0.3217558264732361, + "learning_rate": 4.7926377046422575e-05, + "loss": 0.3055490255355835, + "step": 170 + }, + { + "epoch": 1.1615508885298869, + "grad_norm": 0.31330329179763794, + "learning_rate": 4.7453910709078145e-05, + "loss": 0.31946792602539065, + "step": 180 + }, + { + "epoch": 1.2261712439418417, + "grad_norm": 0.3254242539405823, + "learning_rate": 4.69358351508126e-05, + "loss": 0.3053389549255371, + "step": 190 + }, + { + "epoch": 1.2907915993537964, + "grad_norm": 0.3096122145652771, + "learning_rate": 4.637320270602314e-05, + "loss": 0.30810558795928955, + "step": 200 + }, + { + "epoch": 1.3554119547657513, + "grad_norm": 0.30546557903289795, + "learning_rate": 4.576715621471678e-05, + "loss": 0.31534519195556643, + "step": 210 + }, + { + "epoch": 1.420032310177706, + "grad_norm": 0.32512548565864563, + "learning_rate": 4.511892670113104e-05, + "loss": 0.3046459913253784, + "step": 220 + }, + { + "epoch": 1.4846526655896608, + "grad_norm": 0.32627010345458984, + "learning_rate": 4.442983087323167e-05, + "loss": 0.2983341455459595, + "step": 230 + }, + { + "epoch": 1.5492730210016155, + "grad_norm": 0.3522598147392273, + "learning_rate": 4.370126844816628e-05, + "loss": 0.3133318185806274, + "step": 240 + }, + { + "epoch": 1.6138933764135701, + "grad_norm": 0.30818620324134827, + "learning_rate": 4.293471930910668e-05, + "loss": 0.30533685684204104, + "step": 250 + }, + { + "epoch": 1.678513731825525, + "grad_norm": 0.2774396538734436, + "learning_rate": 4.213174049925498e-05, + "loss": 0.2984048843383789, + "step": 260 + }, + { + "epoch": 1.7431340872374799, + "grad_norm": 0.2943349778652191, + "learning_rate": 4.129396305911934e-05, + "loss": 0.2998037338256836, + "step": 270 + }, + { + "epoch": 1.8077544426494345, + "grad_norm": 0.30737221240997314, + "learning_rate": 4.042308871348371e-05, + "loss": 0.3016969680786133, + "step": 280 + }, + { + "epoch": 1.8723747980613892, + "grad_norm": 0.3105849325656891, + "learning_rate": 3.9520886414800936e-05, + "loss": 0.3014758348464966, + "step": 290 + }, + { + "epoch": 1.936995153473344, + "grad_norm": 0.29112061858177185, + "learning_rate": 3.858918875003053e-05, + "loss": 0.297167444229126, + "step": 300 + }, + { + "epoch": 2.0, + "grad_norm": 0.3953491747379303, + "learning_rate": 3.7629888218219774e-05, + "loss": 0.29504544734954835, + "step": 310 + }, + { + "epoch": 2.0646203554119547, + "grad_norm": 0.29763659834861755, + "learning_rate": 3.664493338638898e-05, + "loss": 0.28955745697021484, + "step": 320 + }, + { + "epoch": 2.1292407108239093, + "grad_norm": 0.3380239009857178, + "learning_rate": 3.5636324931529554e-05, + "loss": 0.28567984104156496, + "step": 330 + }, + { + "epoch": 2.1938610662358644, + "grad_norm": 0.31489691138267517, + "learning_rate": 3.46061115767542e-05, + "loss": 0.28932580947875974, + "step": 340 + }, + { + "epoch": 2.258481421647819, + "grad_norm": 0.3106003403663635, + "learning_rate": 3.3556385929854164e-05, + "loss": 0.2879155158996582, + "step": 350 + }, + { + "epoch": 2.3231017770597737, + "grad_norm": 0.2789410948753357, + "learning_rate": 3.248928023271624e-05, + "loss": 0.2938661813735962, + "step": 360 + }, + { + "epoch": 2.387722132471729, + "grad_norm": 0.2844054102897644, + "learning_rate": 3.140696203023348e-05, + "loss": 0.2907016515731812, + "step": 370 + }, + { + "epoch": 2.4523424878836835, + "grad_norm": 0.3297322392463684, + "learning_rate": 3.0311629767507232e-05, + "loss": 0.28118228912353516, + "step": 380 + }, + { + "epoch": 2.516962843295638, + "grad_norm": 0.30098527669906616, + "learning_rate": 2.920550832428337e-05, + "loss": 0.2846884489059448, + "step": 390 + }, + { + "epoch": 2.581583198707593, + "grad_norm": 0.31197914481163025, + "learning_rate": 2.809084449569365e-05, + "loss": 0.28516063690185545, + "step": 400 + }, + { + "epoch": 2.6462035541195474, + "grad_norm": 0.30557820200920105, + "learning_rate": 2.6969902428481557e-05, + "loss": 0.2813657283782959, + "step": 410 + }, + { + "epoch": 2.7108239095315025, + "grad_norm": 0.3514881730079651, + "learning_rate": 2.584495902198307e-05, + "loss": 0.2873976469039917, + "step": 420 + }, + { + "epoch": 2.775444264943457, + "grad_norm": 0.3010100722312927, + "learning_rate": 2.4718299303203742e-05, + "loss": 0.2792309045791626, + "step": 430 + }, + { + "epoch": 2.840064620355412, + "grad_norm": 0.33192241191864014, + "learning_rate": 2.3592211785386642e-05, + "loss": 0.2784327745437622, + "step": 440 + }, + { + "epoch": 2.904684975767367, + "grad_norm": 0.3093332350254059, + "learning_rate": 2.246898381949894e-05, + "loss": 0.28588788509368895, + "step": 450 + }, + { + "epoch": 2.9693053311793216, + "grad_norm": 0.33140891790390015, + "learning_rate": 2.1350896948079226e-05, + "loss": 0.2833216428756714, + "step": 460 + }, + { + "epoch": 3.0323101777059773, + "grad_norm": 0.31609705090522766, + "learning_rate": 2.0240222270883288e-05, + "loss": 0.28030478954315186, + "step": 470 + }, + { + "epoch": 3.096930533117932, + "grad_norm": 0.2810538113117218, + "learning_rate": 1.913921583174154e-05, + "loss": 0.27578275203704833, + "step": 480 + }, + { + "epoch": 3.161550888529887, + "grad_norm": 0.3165380358695984, + "learning_rate": 1.805011403599862e-05, + "loss": 0.27012245655059813, + "step": 490 + }, + { + "epoch": 3.2261712439418417, + "grad_norm": 0.3220997452735901, + "learning_rate": 1.697512910784352e-05, + "loss": 0.2735466957092285, + "step": 500 + }, + { + "epoch": 3.2907915993537964, + "grad_norm": 0.3499109447002411, + "learning_rate": 1.5916444596757218e-05, + "loss": 0.27160613536834716, + "step": 510 + }, + { + "epoch": 3.355411954765751, + "grad_norm": 0.3451310694217682, + "learning_rate": 1.4876210942205604e-05, + "loss": 0.27280707359313966, + "step": 520 + }, + { + "epoch": 3.420032310177706, + "grad_norm": 0.33574041724205017, + "learning_rate": 1.3856541105586545e-05, + "loss": 0.27128801345825193, + "step": 530 + }, + { + "epoch": 3.484652665589661, + "grad_norm": 0.3565731346607208, + "learning_rate": 1.2859506278303979e-05, + "loss": 0.2736222743988037, + "step": 540 + }, + { + "epoch": 3.5492730210016155, + "grad_norm": 0.32914265990257263, + "learning_rate": 1.1887131674686596e-05, + "loss": 0.27256240844726565, + "step": 550 + }, + { + "epoch": 3.61389337641357, + "grad_norm": 0.31808099150657654, + "learning_rate": 1.0941392418297026e-05, + "loss": 0.27215673923492434, + "step": 560 + }, + { + "epoch": 3.678513731825525, + "grad_norm": 0.3377521336078644, + "learning_rate": 1.0024209529987149e-05, + "loss": 0.2722803592681885, + "step": 570 + }, + { + "epoch": 3.74313408723748, + "grad_norm": 0.33442333340644836, + "learning_rate": 9.137446025848914e-06, + "loss": 0.27033517360687254, + "step": 580 + }, + { + "epoch": 3.8077544426494345, + "grad_norm": 0.32362648844718933, + "learning_rate": 8.282903132986526e-06, + "loss": 0.268932580947876, + "step": 590 + }, + { + "epoch": 3.872374798061389, + "grad_norm": 0.33127161860466003, + "learning_rate": 7.462316630796834e-06, + "loss": 0.274311900138855, + "step": 600 + }, + { + "epoch": 3.936995153473344, + "grad_norm": 0.35842862725257874, + "learning_rate": 6.677353325189409e-06, + "loss": 0.2703368902206421, + "step": 610 + }, + { + "epoch": 4.0, + "grad_norm": 0.4013536870479584, + "learning_rate": 5.929607662908282e-06, + "loss": 0.27242910861968994, + "step": 620 + }, + { + "epoch": 4.064620355411955, + "grad_norm": 0.3234418034553528, + "learning_rate": 5.220598492832171e-06, + "loss": 0.2654104232788086, + "step": 630 + }, + { + "epoch": 4.129240710823909, + "grad_norm": 0.3155248761177063, + "learning_rate": 4.551765980832059e-06, + "loss": 0.26429879665374756, + "step": 640 + }, + { + "epoch": 4.193861066235864, + "grad_norm": 0.3309212327003479, + "learning_rate": 3.924468684452523e-06, + "loss": 0.2637316226959229, + "step": 650 + }, + { + "epoch": 4.258481421647819, + "grad_norm": 0.3667273223400116, + "learning_rate": 3.339980793358982e-06, + "loss": 0.2668980360031128, + "step": 660 + }, + { + "epoch": 4.323101777059774, + "grad_norm": 0.3333350718021393, + "learning_rate": 2.7994895411561333e-06, + "loss": 0.2676853179931641, + "step": 670 + }, + { + "epoch": 4.387722132471729, + "grad_norm": 0.32394319772720337, + "learning_rate": 2.3040927938347528e-06, + "loss": 0.26267526149749754, + "step": 680 + }, + { + "epoch": 4.4523424878836835, + "grad_norm": 0.3705294132232666, + "learning_rate": 1.854796819745383e-06, + "loss": 0.26965837478637694, + "step": 690 + }, + { + "epoch": 4.516962843295638, + "grad_norm": 0.3425300121307373, + "learning_rate": 1.4525142456284779e-06, + "loss": 0.2649491310119629, + "step": 700 + }, + { + "epoch": 4.581583198707593, + "grad_norm": 0.32138586044311523, + "learning_rate": 1.098062202852984e-06, + "loss": 0.264300274848938, + "step": 710 + }, + { + "epoch": 4.646203554119547, + "grad_norm": 0.34855136275291443, + "learning_rate": 7.921606676285776e-07, + "loss": 0.2624504804611206, + "step": 720 + }, + { + "epoch": 4.710823909531502, + "grad_norm": 0.33398354053497314, + "learning_rate": 5.354309985631605e-07, + "loss": 0.2657909870147705, + "step": 730 + }, + { + "epoch": 4.775444264943458, + "grad_norm": 0.32164713740348816, + "learning_rate": 3.283946745360755e-07, + "loss": 0.2667935609817505, + "step": 740 + }, + { + "epoch": 4.840064620355412, + "grad_norm": 0.3328765332698822, + "learning_rate": 1.714722354507592e-07, + "loss": 0.26322848796844484, + "step": 750 + }, + { + "epoch": 4.904684975767367, + "grad_norm": 0.3057484030723572, + "learning_rate": 6.498242801838694e-08, + "loss": 0.26327192783355713, + "step": 760 + }, + { + "epoch": 4.969305331179322, + "grad_norm": 0.3382052183151245, + "learning_rate": 9.141558307687592e-09, + "loss": 0.2605318546295166, + "step": 770 + }, + { + "epoch": 5.0, + "step": 775, + "total_flos": 7.509104074275226e+17, + "train_loss": 0.31352297429115544, + "train_runtime": 25009.3076, + "train_samples_per_second": 0.99, + "train_steps_per_second": 0.031 + } + ], + "logging_steps": 10, + "max_steps": 775, + "num_input_tokens_seen": 0, + "num_train_epochs": 5, + "save_steps": 100, + "stateful_callbacks": { + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": true + }, + "attributes": {} + } + }, + "total_flos": 7.509104074275226e+17, + "train_batch_size": 4, + "trial_name": null, + "trial_params": null +} diff --git a/lora/sft-captcha/training_args.bin b/lora/sft-captcha/training_args.bin new file mode 100644 index 0000000000000000000000000000000000000000..fb2a0e0e3f614c6145d25df84879d6af93eeaf5b --- /dev/null +++ b/lora/sft-captcha/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:994239c83a21a412f370376b6709e6af83b0c60df0ba98c0dc5d4b0c9b6306e1 +size 5496 diff --git a/lora/sft-captcha/training_loss.png b/lora/sft-captcha/training_loss.png new file mode 100644 index 0000000000000000000000000000000000000000..f33a710187a31dd153a22bffce0d408a9394f29b Binary files /dev/null and b/lora/sft-captcha/training_loss.png differ