last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.034255599472990776,
  "eval_steps": 10,
  "global_step": 26,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0013175230566534915,
      "eval_loss": 2.429551601409912,
      "eval_runtime": 34.1229,
      "eval_samples_per_second": 9.378,
      "eval_steps_per_second": 4.689,
      "step": 1
    },
    {
      "epoch": 0.006587615283267457,
      "grad_norm": 1.0831778049468994,
      "learning_rate": 5e-05,
      "loss": 2.1113,
      "step": 5
    },
    {
      "epoch": 0.013175230566534914,
      "grad_norm": 0.8124720454216003,
      "learning_rate": 0.0001,
      "loss": 1.9417,
      "step": 10
    },
    {
      "epoch": 0.013175230566534914,
      "eval_loss": 1.9890544414520264,
      "eval_runtime": 34.7293,
      "eval_samples_per_second": 9.214,
      "eval_steps_per_second": 4.607,
      "step": 10
    },
    {
      "epoch": 0.019762845849802372,
      "grad_norm": 0.7260387539863586,
      "learning_rate": 9.619397662556435e-05,
      "loss": 1.5863,
      "step": 15
    },
    {
      "epoch": 0.026350461133069828,
      "grad_norm": 0.6288080811500549,
      "learning_rate": 8.535533905932738e-05,
      "loss": 1.4501,
      "step": 20
    },
    {
      "epoch": 0.026350461133069828,
      "eval_loss": 1.3593571186065674,
      "eval_runtime": 34.6864,
      "eval_samples_per_second": 9.226,
      "eval_steps_per_second": 4.613,
      "step": 20
    },
    {
      "epoch": 0.03293807641633729,
      "grad_norm": 0.6772316694259644,
      "learning_rate": 6.91341716182545e-05,
      "loss": 1.2137,
      "step": 25
    }
  ],
  "logging_steps": 5,
  "max_steps": 50,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 13,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 8544453975343104.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}