{
  "model": "phi2-mental-health",
  "base_model": "microsoft/phi-2",
  "dataset": "ShenLab/MentalChat16K",
  "lora_config": {
    "rank": 16,
    "alpha": 32,
    "target_modules": [
      "q_proj",
      "k_proj",
      "v_proj",
      "dense"
    ],
    "dropout": 0.1
  },
  "training": {
    "final_train_loss": 0.7486542798042297,
    "total_steps": 2500,
    "epochs": 4,
    "learning_rate": 0.0002,
    "per_device_batch_size": 4,
    "gradient_accumulation": 2
  },
  "evaluation": {
    "eval_loss": 0.7297702431678772,
    "eval_runtime": 4064.1661,
    "eval_samples_per_second": 0.116,
    "eval_steps_per_second": 0.029,
    "epoch": 3.7397157816005984
  },
  "test_eval": {
    "eval_loss": 0.7111775875091553,
    "eval_runtime": 39.2705,
    "eval_samples_per_second": 12.019,
    "eval_steps_per_second": 3.005,
    "epoch": 3.7397157816005984
  },
  "dataset_stats": {
    "train_size": 5347,
    "val_size": 472,
    "test_size": 472
  }
}