| |
| """ |
| Free Fine-Tuning Script for DualBlind Arena Reasoning Dataset |
| Runs on free Google Colab (T4 / L4 GPU) or local Linux/Mac with CUDA/MPS. |
| """ |
|
|
| import os |
| from datasets import load_dataset |
| from unsloth import FastLanguageModel |
| from trl import SFTTrainer |
| from transformers import TrainingArguments |
|
|
| DATASET_NAME = "GlimmaryKarl/DualBlind" |
| BASE_MODEL = "unsloth/Qwen2.5-7B-Instruct" |
|
|
| print(f"Loading base model: {BASE_MODEL}") |
| model, tokenizer = FastLanguageModel.from_pretrained( |
| model_name=BASE_MODEL, |
| max_seq_length=4096, |
| load_in_4bit=True, |
| ) |
|
|
| model = FastLanguageModel.get_peft_model( |
| model, |
| r=16, |
| target_modules=["q_proj", "k_proj", "v_proj", "o_proj", "gate_proj", "up_proj", "down_proj"], |
| lora_alpha=16, |
| lora_dropout=0, |
| bias="none", |
| use_gradient_checkpointing="unsloth", |
| ) |
|
|
| print(f"Loading reasoning SFT dataset from {DATASET_NAME}...") |
| try: |
| dataset = load_dataset(DATASET_NAME, "sft_reasoning", split="train") |
| except Exception: |
| print("Loading from local data/sft_reasoning_train.jsonl...") |
| dataset = load_dataset("json", data_files="data/sft_reasoning_train.jsonl", split="train") |
|
|
| def format_prompts(batch): |
| texts = [] |
| for msgs in batch["messages"]: |
| texts.append(tokenizer.apply_chat_template(msgs, tokenize=False, add_generation_prompt=False)) |
| return {"text": texts} |
|
|
| dataset = dataset.map(format_prompts, batched=True) |
|
|
| trainer = SFTTrainer( |
| model=model, |
| tokenizer=tokenizer, |
| train_dataset=dataset, |
| dataset_text_field="text", |
| max_seq_length=4096, |
| dataset_num_proc=2, |
| packing=False, |
| args=TrainingArguments( |
| per_device_train_batch_size=2, |
| gradient_accumulation_steps=4, |
| warmup_ratio=0.05, |
| num_train_epochs=3, |
| learning_rate=2e-4, |
| fp16=True, |
| logging_steps=5, |
| optim="adamw_8bit", |
| output_dir="reasoning_model_lora", |
| save_strategy="epoch", |
| ), |
| ) |
|
|
| print("Starting training...") |
| trainer.train() |
|
|
| print("Saving fine-tuned LoRA weights...") |
| model.save_pretrained_merged("dualblind_reasoning_model", tokenizer, save_method="merged_16bit") |
| print("Complete! Model saved to ./dualblind_reasoning_model") |
|
|
