{
    "model": "unsloth/gemma-2-9b",
    "training_type": "SFT and DPO",
    "max_seq_length": 1024,
    "dtype": "float16",
    "training_args": {
        "sft": {
            "learning_rate": 2e-06,
            "batch_size": 4,
            "gradient_accumulation_steps": 4,
            "num_train_epochs": 3,
            "scheduler": "cosine",
            "optim": "adamw_8bit",
            "seed": 2802
        }
    }
}