{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0080645161290323,
  "eval_steps": 200,
  "global_step": 3000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.06720430107526881,
      "eval_loss": 3.1445324420928955,
      "eval_runtime": 187.2734,
      "eval_samples_per_second": 37.496,
      "eval_steps_per_second": 2.344,
      "eval_wer": 1.0,
      "step": 200
    },
    {
      "epoch": 0.13440860215053763,
      "eval_loss": 2.7407212257385254,
      "eval_runtime": 185.5858,
      "eval_samples_per_second": 37.837,
      "eval_steps_per_second": 2.365,
      "eval_wer": 0.9999867125526515,
      "step": 400
    },
    {
      "epoch": 0.16801075268817203,
      "grad_norm": 8.384466171264648,
      "learning_rate": 0.00027866666666666665,
      "loss": 4.0188,
      "step": 500
    },
    {
      "epoch": 0.20161290322580644,
      "eval_loss": 1.2700377702713013,
      "eval_runtime": 187.6926,
      "eval_samples_per_second": 37.412,
      "eval_steps_per_second": 2.339,
      "eval_wer": 0.8484433755431244,
      "step": 600
    },
    {
      "epoch": 0.26881720430107525,
      "eval_loss": 0.9953192472457886,
      "eval_runtime": 187.642,
      "eval_samples_per_second": 37.422,
      "eval_steps_per_second": 2.34,
      "eval_wer": 0.7435389787267968,
      "step": 800
    },
    {
      "epoch": 0.33602150537634407,
      "grad_norm": 5.077725410461426,
      "learning_rate": 0.0002232222222222222,
      "loss": 1.0707,
      "step": 1000
    },
    {
      "epoch": 0.33602150537634407,
      "eval_loss": 0.8646696209907532,
      "eval_runtime": 187.682,
      "eval_samples_per_second": 37.414,
      "eval_steps_per_second": 2.339,
      "eval_wer": 0.6541277455188084,
      "step": 1000
    },
    {
      "epoch": 0.4032258064516129,
      "eval_loss": 0.7888585925102234,
      "eval_runtime": 187.1558,
      "eval_samples_per_second": 37.52,
      "eval_steps_per_second": 2.346,
      "eval_wer": 0.5784025830797646,
      "step": 1200
    },
    {
      "epoch": 0.47043010752688175,
      "eval_loss": 0.7465152740478516,
      "eval_runtime": 185.9206,
      "eval_samples_per_second": 37.769,
      "eval_steps_per_second": 2.361,
      "eval_wer": 0.5440412442365697,
      "step": 1400
    },
    {
      "epoch": 0.5040322580645161,
      "grad_norm": 7.621553897857666,
      "learning_rate": 0.00016777777777777776,
      "loss": 0.8175,
      "step": 1500
    },
    {
      "epoch": 0.5376344086021505,
      "eval_loss": 0.68277907371521,
      "eval_runtime": 187.6331,
      "eval_samples_per_second": 37.424,
      "eval_steps_per_second": 2.34,
      "eval_wer": 0.5042453394278426,
      "step": 1600
    },
    {
      "epoch": 0.6048387096774194,
      "eval_loss": 0.6549283862113953,
      "eval_runtime": 186.6705,
      "eval_samples_per_second": 37.617,
      "eval_steps_per_second": 2.352,
      "eval_wer": 0.4952098752308694,
      "step": 1800
    },
    {
      "epoch": 0.6720430107526881,
      "grad_norm": 6.983826637268066,
      "learning_rate": 0.00011233333333333333,
      "loss": 0.7148,
      "step": 2000
    },
    {
      "epoch": 0.6720430107526881,
      "eval_loss": 0.6289859414100647,
      "eval_runtime": 188.3084,
      "eval_samples_per_second": 37.29,
      "eval_steps_per_second": 2.331,
      "eval_wer": 0.4905858435535949,
      "step": 2000
    },
    {
      "epoch": 0.739247311827957,
      "eval_loss": 0.6112708449363708,
      "eval_runtime": 188.0636,
      "eval_samples_per_second": 37.338,
      "eval_steps_per_second": 2.334,
      "eval_wer": 0.45763297412934,
      "step": 2200
    },
    {
      "epoch": 0.8064516129032258,
      "eval_loss": 0.5718730688095093,
      "eval_runtime": 187.8424,
      "eval_samples_per_second": 37.382,
      "eval_steps_per_second": 2.337,
      "eval_wer": 0.4404788796024396,
      "step": 2400
    },
    {
      "epoch": 0.8400537634408602,
      "grad_norm": 6.760587215423584,
      "learning_rate": 5.6999999999999996e-05,
      "loss": 0.6374,
      "step": 2500
    },
    {
      "epoch": 0.8736559139784946,
      "eval_loss": 0.5643858313560486,
      "eval_runtime": 188.0058,
      "eval_samples_per_second": 37.35,
      "eval_steps_per_second": 2.335,
      "eval_wer": 0.431376978168724,
      "step": 2600
    },
    {
      "epoch": 0.9408602150537635,
      "eval_loss": 0.5482733249664307,
      "eval_runtime": 188.1104,
      "eval_samples_per_second": 37.329,
      "eval_steps_per_second": 2.334,
      "eval_wer": 0.41896650234523447,
      "step": 2800
    },
    {
      "epoch": 1.0080645161290323,
      "grad_norm": 1.7233390808105469,
      "learning_rate": 1.4444444444444445e-06,
      "loss": 0.6013,
      "step": 3000
    },
    {
      "epoch": 1.0080645161290323,
      "eval_loss": 0.5417820811271667,
      "eval_runtime": 189.023,
      "eval_samples_per_second": 37.149,
      "eval_steps_per_second": 2.322,
      "eval_wer": 0.4178370693206128,
      "step": 3000
    },
    {
      "epoch": 1.0080645161290323,
      "step": 3000,
      "total_flos": 6.709869296482936e+18,
      "train_loss": 1.3100875549316406,
      "train_runtime": 4950.9591,
      "train_samples_per_second": 9.695,
      "train_steps_per_second": 0.606
    }
  ],
  "logging_steps": 500,
  "max_steps": 3000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 6.709869296482936e+18,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}