{
  "best_global_step": 6136,
  "best_metric": 0.8068473935127258,
  "best_model_checkpoint": "bert_base_km_50_v2_mnli/checkpoint-6136",
  "epoch": 9.0,
  "eval_steps": 500,
  "global_step": 13806,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 1.1364778280258179,
      "learning_rate": 4.9e-05,
      "loss": 0.9905,
      "step": 1534
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5657666836474784,
      "eval_loss": 0.920609176158905,
      "eval_runtime": 6.2534,
      "eval_samples_per_second": 1569.535,
      "eval_steps_per_second": 6.237,
      "step": 1534
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.483047366142273,
      "learning_rate": 4.8e-05,
      "loss": 0.8854,
      "step": 3068
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6031584309730005,
      "eval_loss": 0.8692390322685242,
      "eval_runtime": 6.1696,
      "eval_samples_per_second": 1590.871,
      "eval_steps_per_second": 6.321,
      "step": 3068
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.7138577699661255,
      "learning_rate": 4.7e-05,
      "loss": 0.8065,
      "step": 4602
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6321956189505858,
      "eval_loss": 0.8283551335334778,
      "eval_runtime": 6.1748,
      "eval_samples_per_second": 1589.534,
      "eval_steps_per_second": 6.316,
      "step": 4602
    },
    {
      "epoch": 4.0,
      "grad_norm": 2.1886351108551025,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.7303,
      "step": 6136
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6476821192052981,
      "eval_loss": 0.8068473935127258,
      "eval_runtime": 6.1611,
      "eval_samples_per_second": 1593.072,
      "eval_steps_per_second": 6.33,
      "step": 6136
    },
    {
      "epoch": 5.0,
      "grad_norm": 2.2214322090148926,
      "learning_rate": 4.5e-05,
      "loss": 0.6572,
      "step": 7670
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.6522669383596535,
      "eval_loss": 0.840409517288208,
      "eval_runtime": 6.2208,
      "eval_samples_per_second": 1577.763,
      "eval_steps_per_second": 6.269,
      "step": 7670
    },
    {
      "epoch": 6.0,
      "grad_norm": 3.14277982711792,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.5827,
      "step": 9204
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.6575649516046868,
      "eval_loss": 0.8798835277557373,
      "eval_runtime": 6.1442,
      "eval_samples_per_second": 1597.442,
      "eval_steps_per_second": 6.347,
      "step": 9204
    },
    {
      "epoch": 7.0,
      "grad_norm": 2.993121862411499,
      "learning_rate": 4.3e-05,
      "loss": 0.5062,
      "step": 10738
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.6541008660213958,
      "eval_loss": 0.9258556962013245,
      "eval_runtime": 6.1438,
      "eval_samples_per_second": 1597.542,
      "eval_steps_per_second": 6.348,
      "step": 10738
    },
    {
      "epoch": 8.0,
      "grad_norm": 3.233996629714966,
      "learning_rate": 4.2e-05,
      "loss": 0.4354,
      "step": 12272
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.6509424350483953,
      "eval_loss": 0.9967018961906433,
      "eval_runtime": 6.1462,
      "eval_samples_per_second": 1596.913,
      "eval_steps_per_second": 6.345,
      "step": 12272
    },
    {
      "epoch": 9.0,
      "grad_norm": 3.693145513534546,
      "learning_rate": 4.1e-05,
      "loss": 0.3682,
      "step": 13806
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.649617931737137,
      "eval_loss": 1.1101100444793701,
      "eval_runtime": 6.1447,
      "eval_samples_per_second": 1597.305,
      "eval_steps_per_second": 6.347,
      "step": 13806
    },
    {
      "epoch": 9.0,
      "step": 13806,
      "total_flos": 4.6496324415879475e+17,
      "train_loss": 0.6624809042985726,
      "train_runtime": 5711.7823,
      "train_samples_per_second": 3437.649,
      "train_steps_per_second": 13.428
    }
  ],
  "logging_steps": 1,
  "max_steps": 76700,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 50,
  "save_steps": 500,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 5,
        "early_stopping_threshold": 0.0
      },
      "attributes": {
        "early_stopping_patience_counter": 5
      }
    },
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 4.6496324415879475e+17,
  "train_batch_size": 256,
  "trial_name": null,
  "trial_params": null
}