train_llama3.json

{
  "stage": "sft",
  "do_train": true,
  "model_name_or_path": "unsloth/llama-3-8b-Instruct-bnb-4bit",
  "dataset": "sql_create_context_v4",
  "template": "llama3",
  "finetuning_type": "lora",
  "lora_target": "all",
  "output_dir": "llama3_lora",
  "per_device_train_batch_size": 2,
  "gradient_accumulation_steps": 4,
  "lr_scheduler_type": "cosine",
  "logging_steps": 10,
  "warmup_ratio": 0.1,
  "save_steps": 1000,
  "learning_rate": 5e-05,
  "num_train_epochs": 100,
  "max_samples": 1000,
  "max_grad_norm": 1.0,
  "quantization_bit": 4,
  "loraplus_lr_ratio": 16.0,
  "fp16": true
}