train_qwen2_7b.json

{
  "model_name_or_path": "Qwen/Qwen2-7B",
  "stage": "sft",
  "do_train": true,
  "finetuning_type": "lora",
  "lora_target": "all",
  "loraplus_lr_ratio": 16.0,
  "dataset": "mgtv_train",
  "template": "qwen",
  "cutoff_len": 1024,
  "max_samples": 2000,
  "overwrite_cache": "true",
  "preprocessing_num_workers": 16,
  "output_dir": "/content/qwen2-7b",
  "logging_steps": 10,
  "save_steps": 10,
  "plot_loss": "true",
  "overwrite_output_dir": "true",
  "per_device_train_batch_size": 1,
  "gradient_accumulation_steps": 8,
  "learning_rate": 0.0001,
  "num_train_epochs": 6.0,
  "lr_scheduler_type": "cosine",
  "warmup_ratio": 0.1,
  "bf16": true,
  "ddp_timeout": 180000000,
  "val_size": 0.1,
  "per_device_eval_batch_size": 1,
  "eval_strategy": "steps",
  "eval_steps": 10,
  "report_to": "wandb"
}