last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.003937395412934344,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0003937395412934344,
      "grad_norm": 2.7983758449554443,
      "learning_rate": 2e-05,
      "loss": 2.1978,
      "step": 1
    },
    {
      "epoch": 0.0003937395412934344,
      "eval_loss": 3.2129366397857666,
      "eval_runtime": 117.1778,
      "eval_samples_per_second": 4.566,
      "eval_steps_per_second": 4.566,
      "step": 1
    },
    {
      "epoch": 0.0007874790825868688,
      "grad_norm": 1.699691653251648,
      "learning_rate": 4e-05,
      "loss": 1.1053,
      "step": 2
    },
    {
      "epoch": 0.0011812186238803033,
      "grad_norm": 2.7148642539978027,
      "learning_rate": 6e-05,
      "loss": 2.0023,
      "step": 3
    },
    {
      "epoch": 0.0011812186238803033,
      "eval_loss": 3.08441424369812,
      "eval_runtime": 116.5547,
      "eval_samples_per_second": 4.59,
      "eval_steps_per_second": 4.59,
      "step": 3
    },
    {
      "epoch": 0.0015749581651737375,
      "grad_norm": 4.442303657531738,
      "learning_rate": 8e-05,
      "loss": 3.3335,
      "step": 4
    },
    {
      "epoch": 0.001968697706467172,
      "grad_norm": 2.875202178955078,
      "learning_rate": 0.0001,
      "loss": 2.1078,
      "step": 5
    },
    {
      "epoch": 0.0023624372477606065,
      "grad_norm": 2.5976645946502686,
      "learning_rate": 0.00012,
      "loss": 1.8696,
      "step": 6
    },
    {
      "epoch": 0.0023624372477606065,
      "eval_loss": 2.2479214668273926,
      "eval_runtime": 116.5681,
      "eval_samples_per_second": 4.59,
      "eval_steps_per_second": 4.59,
      "step": 6
    },
    {
      "epoch": 0.002756176789054041,
      "grad_norm": NaN,
      "learning_rate": 0.00012,
      "loss": 1.9327,
      "step": 7
    },
    {
      "epoch": 0.003149916330347475,
      "grad_norm": 5.197524070739746,
      "learning_rate": 0.00014,
      "loss": 2.3621,
      "step": 8
    },
    {
      "epoch": 0.0035436558716409093,
      "grad_norm": 3.2192602157592773,
      "learning_rate": 0.00016,
      "loss": 1.3753,
      "step": 9
    },
    {
      "epoch": 0.0035436558716409093,
      "eval_loss": 1.492257833480835,
      "eval_runtime": 116.5291,
      "eval_samples_per_second": 4.591,
      "eval_steps_per_second": 4.591,
      "step": 9
    },
    {
      "epoch": 0.003937395412934344,
      "grad_norm": 3.4407730102539062,
      "learning_rate": 0.00018,
      "loss": 1.9109,
      "step": 10
    }
  ],
  "logging_steps": 1,
  "max_steps": 10,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 3,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 6990386933268480.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}