license: apache-2.0
Neural Krishna DPO
Fine-tuning + lnegth(choose)
- Training Args:
´´´python
LoRA configuration
peft_config = LoraConfig( r=16, lora_alpha=16, lora_dropout=0.05, bias="none", task_type="CAUSAL_LM", target_modules=['k_proj', 'gate_proj', 'v_proj', 'up_proj', 'q_proj', 'o_proj', 'down_proj'] )
Model to fine-tune
model = AutoModelForCausalLM.from_pretrained( model_name, torch_dtype=torch.float16, load_in_4bit=True ) model.config.use_cache = False
Training arguments
training_args = TrainingArguments( per_device_train_batch_size=4, gradient_accumulation_steps=4, gradient_checkpointing=True, learning_rate=5e-5, lr_scheduler_type="cosine", max_steps=120, save_strategy="no", logging_steps=1, output_dir=new_model, optim="paged_adamw_32bit", warmup_steps=50, bf16=True, report_to="wandb", )
Create DPO trainer
dpo_trainer = DPOTrainer( model, args=training_args, train_dataset=dataset, tokenizer=tokenizer, peft_config=peft_config, beta=0.1, max_prompt_length=1024, max_length=1536, )
Fine-tune model with DPO
dpo_trainer.train() ´´´