last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0015677052714089751,
  "eval_steps": 5,
  "global_step": 40,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 3.919263178522438e-05,
      "eval_loss": NaN,
      "eval_runtime": 1085.419,
      "eval_samples_per_second": 9.898,
      "eval_steps_per_second": 4.949,
      "step": 1
    },
    {
      "epoch": 0.00011757789535567314,
      "grad_norm": NaN,
      "learning_rate": 6e-05,
      "loss": 0.0,
      "step": 3
    },
    {
      "epoch": 0.0001959631589261219,
      "eval_loss": NaN,
      "eval_runtime": 1087.9639,
      "eval_samples_per_second": 9.875,
      "eval_steps_per_second": 4.938,
      "step": 5
    },
    {
      "epoch": 0.00023515579071134627,
      "grad_norm": NaN,
      "learning_rate": 0.00012,
      "loss": 0.0,
      "step": 6
    },
    {
      "epoch": 0.0003527336860670194,
      "grad_norm": NaN,
      "learning_rate": 0.00018,
      "loss": 0.0,
      "step": 9
    },
    {
      "epoch": 0.0003919263178522438,
      "eval_loss": NaN,
      "eval_runtime": 1796.0138,
      "eval_samples_per_second": 5.982,
      "eval_steps_per_second": 2.991,
      "step": 10
    },
    {
      "epoch": 0.00047031158142269255,
      "grad_norm": NaN,
      "learning_rate": 0.00019781476007338058,
      "loss": 0.0,
      "step": 12
    },
    {
      "epoch": 0.0005878894767783657,
      "grad_norm": NaN,
      "learning_rate": 0.00018660254037844388,
      "loss": 0.0,
      "step": 15
    },
    {
      "epoch": 0.0005878894767783657,
      "eval_loss": NaN,
      "eval_runtime": 1798.0063,
      "eval_samples_per_second": 5.976,
      "eval_steps_per_second": 2.988,
      "step": 15
    },
    {
      "epoch": 0.0007054673721340388,
      "grad_norm": NaN,
      "learning_rate": 0.00016691306063588583,
      "loss": 0.0,
      "step": 18
    },
    {
      "epoch": 0.0007838526357044876,
      "eval_loss": NaN,
      "eval_runtime": 1797.8378,
      "eval_samples_per_second": 5.976,
      "eval_steps_per_second": 2.988,
      "step": 20
    },
    {
      "epoch": 0.0008230452674897119,
      "grad_norm": NaN,
      "learning_rate": 0.00014067366430758004,
      "loss": 0.0,
      "step": 21
    },
    {
      "epoch": 0.0009406231628453851,
      "grad_norm": NaN,
      "learning_rate": 0.00011045284632676536,
      "loss": 0.0,
      "step": 24
    },
    {
      "epoch": 0.0009798157946306093,
      "eval_loss": NaN,
      "eval_runtime": 1798.0158,
      "eval_samples_per_second": 5.975,
      "eval_steps_per_second": 2.988,
      "step": 25
    },
    {
      "epoch": 0.0010582010582010583,
      "grad_norm": NaN,
      "learning_rate": 7.920883091822408e-05,
      "loss": 0.0,
      "step": 27
    },
    {
      "epoch": 0.0011757789535567313,
      "grad_norm": NaN,
      "learning_rate": 5.000000000000002e-05,
      "loss": 0.0,
      "step": 30
    },
    {
      "epoch": 0.0011757789535567313,
      "eval_loss": NaN,
      "eval_runtime": 1798.0739,
      "eval_samples_per_second": 5.975,
      "eval_steps_per_second": 2.988,
      "step": 30
    },
    {
      "epoch": 0.0012933568489124044,
      "grad_norm": NaN,
      "learning_rate": 2.5685517452260567e-05,
      "loss": 0.0,
      "step": 33
    },
    {
      "epoch": 0.0013717421124828531,
      "eval_loss": NaN,
      "eval_runtime": 1798.4968,
      "eval_samples_per_second": 5.974,
      "eval_steps_per_second": 2.987,
      "step": 35
    },
    {
      "epoch": 0.0014109347442680777,
      "grad_norm": NaN,
      "learning_rate": 8.645454235739903e-06,
      "loss": 0.0,
      "step": 36
    },
    {
      "epoch": 0.0015285126396237508,
      "grad_norm": NaN,
      "learning_rate": 5.478104631726711e-07,
      "loss": 0.0,
      "step": 39
    },
    {
      "epoch": 0.0015677052714089751,
      "eval_loss": NaN,
      "eval_runtime": 1805.5926,
      "eval_samples_per_second": 5.95,
      "eval_steps_per_second": 2.975,
      "step": 40
    }
  ],
  "logging_steps": 3,
  "max_steps": 40,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 10,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.406258997362688e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}