{
  "best_global_step": 6136,
  "best_metric": 0.7868309020996094,
  "best_model_checkpoint": "bert_base_km_100_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": 1.0563355684280396,
      "learning_rate": 4.9e-05,
      "loss": 0.9919,
      "step": 1534
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.567396841569027,
      "eval_loss": 0.9109046459197998,
      "eval_runtime": 6.2368,
      "eval_samples_per_second": 1573.736,
      "eval_steps_per_second": 6.253,
      "step": 1534
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.4128915071487427,
      "learning_rate": 4.8e-05,
      "loss": 0.8775,
      "step": 3068
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6129393785022924,
      "eval_loss": 0.8532446026802063,
      "eval_runtime": 6.1845,
      "eval_samples_per_second": 1587.027,
      "eval_steps_per_second": 6.306,
      "step": 3068
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.7205497026443481,
      "learning_rate": 4.7e-05,
      "loss": 0.7872,
      "step": 4602
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6459500764136525,
      "eval_loss": 0.7985056638717651,
      "eval_runtime": 6.1934,
      "eval_samples_per_second": 1584.752,
      "eval_steps_per_second": 6.297,
      "step": 4602
    },
    {
      "epoch": 4.0,
      "grad_norm": 1.723157286643982,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.7081,
      "step": 6136
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6558329088130412,
      "eval_loss": 0.7868309020996094,
      "eval_runtime": 6.1961,
      "eval_samples_per_second": 1584.072,
      "eval_steps_per_second": 6.294,
      "step": 6136
    },
    {
      "epoch": 5.0,
      "grad_norm": 2.4626529216766357,
      "learning_rate": 4.5e-05,
      "loss": 0.6356,
      "step": 7670
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.6589913397860417,
      "eval_loss": 0.7993543744087219,
      "eval_runtime": 6.2127,
      "eval_samples_per_second": 1579.834,
      "eval_steps_per_second": 6.277,
      "step": 7670
    },
    {
      "epoch": 6.0,
      "grad_norm": 2.7789957523345947,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.5597,
      "step": 9204
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.6656138563423332,
      "eval_loss": 0.8493095636367798,
      "eval_runtime": 6.2169,
      "eval_samples_per_second": 1578.771,
      "eval_steps_per_second": 6.273,
      "step": 9204
    },
    {
      "epoch": 7.0,
      "grad_norm": 2.8432204723358154,
      "learning_rate": 4.3e-05,
      "loss": 0.481,
      "step": 10738
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.6660213958227204,
      "eval_loss": 0.9018713235855103,
      "eval_runtime": 6.2212,
      "eval_samples_per_second": 1577.674,
      "eval_steps_per_second": 6.269,
      "step": 10738
    },
    {
      "epoch": 8.0,
      "grad_norm": 2.848738193511963,
      "learning_rate": 4.2e-05,
      "loss": 0.4057,
      "step": 12272
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.6586856851757514,
      "eval_loss": 0.996565043926239,
      "eval_runtime": 6.2354,
      "eval_samples_per_second": 1574.078,
      "eval_steps_per_second": 6.255,
      "step": 12272
    },
    {
      "epoch": 9.0,
      "grad_norm": 3.7572710514068604,
      "learning_rate": 4.1e-05,
      "loss": 0.3381,
      "step": 13806
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.6597045338767193,
      "eval_loss": 1.1030168533325195,
      "eval_runtime": 6.218,
      "eval_samples_per_second": 1578.476,
      "eval_steps_per_second": 6.272,
      "step": 13806
    },
    {
      "epoch": 9.0,
      "step": 13806,
      "total_flos": 4.6496324415879475e+17,
      "train_loss": 0.6427538210147535,
      "train_runtime": 5761.4438,
      "train_samples_per_second": 3408.017,
      "train_steps_per_second": 13.313
    }
  ],
  "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
}