{
  "best_global_step": 6136,
  "best_metric": 0.8197717070579529,
  "best_model_checkpoint": "bert_base_rand_5_v1_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": 0.9701244235038757,
      "learning_rate": 4.9e-05,
      "loss": 0.9872,
      "step": 1534
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5554763117677025,
      "eval_loss": 0.9291633367538452,
      "eval_runtime": 6.1604,
      "eval_samples_per_second": 1593.246,
      "eval_steps_per_second": 6.331,
      "step": 1534
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.4962955713272095,
      "learning_rate": 4.8e-05,
      "loss": 0.891,
      "step": 3068
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.5991849210392257,
      "eval_loss": 0.874858021736145,
      "eval_runtime": 6.1661,
      "eval_samples_per_second": 1591.758,
      "eval_steps_per_second": 6.325,
      "step": 3068
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.5300061702728271,
      "learning_rate": 4.7e-05,
      "loss": 0.8134,
      "step": 4602
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6161996943453897,
      "eval_loss": 0.8477750420570374,
      "eval_runtime": 6.1433,
      "eval_samples_per_second": 1597.671,
      "eval_steps_per_second": 6.348,
      "step": 4602
    },
    {
      "epoch": 4.0,
      "grad_norm": 2.027043104171753,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.7398,
      "step": 6136
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6413652572592969,
      "eval_loss": 0.8197717070579529,
      "eval_runtime": 6.165,
      "eval_samples_per_second": 1592.04,
      "eval_steps_per_second": 6.326,
      "step": 6136
    },
    {
      "epoch": 5.0,
      "grad_norm": 2.2500391006469727,
      "learning_rate": 4.5e-05,
      "loss": 0.6672,
      "step": 7670
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.6495160468670402,
      "eval_loss": 0.8426722288131714,
      "eval_runtime": 6.1491,
      "eval_samples_per_second": 1596.169,
      "eval_steps_per_second": 6.342,
      "step": 7670
    },
    {
      "epoch": 6.0,
      "grad_norm": 2.5848565101623535,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.5962,
      "step": 9204
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.6585838003056547,
      "eval_loss": 0.8499076962471008,
      "eval_runtime": 6.1433,
      "eval_samples_per_second": 1597.678,
      "eval_steps_per_second": 6.348,
      "step": 9204
    },
    {
      "epoch": 7.0,
      "grad_norm": 2.6250078678131104,
      "learning_rate": 4.3e-05,
      "loss": 0.5244,
      "step": 10738
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.6546102903718798,
      "eval_loss": 0.8883566856384277,
      "eval_runtime": 6.163,
      "eval_samples_per_second": 1592.569,
      "eval_steps_per_second": 6.328,
      "step": 10738
    },
    {
      "epoch": 8.0,
      "grad_norm": 3.2551000118255615,
      "learning_rate": 4.2e-05,
      "loss": 0.456,
      "step": 12272
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.6505348955680081,
      "eval_loss": 0.9953705072402954,
      "eval_runtime": 6.2961,
      "eval_samples_per_second": 1558.902,
      "eval_steps_per_second": 6.194,
      "step": 12272
    },
    {
      "epoch": 9.0,
      "grad_norm": 3.5964415073394775,
      "learning_rate": 4.1e-05,
      "loss": 0.3923,
      "step": 13806
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.650127356087621,
      "eval_loss": 1.0533967018127441,
      "eval_runtime": 6.1447,
      "eval_samples_per_second": 1597.314,
      "eval_steps_per_second": 6.347,
      "step": 13806
    },
    {
      "epoch": 9.0,
      "step": 13806,
      "total_flos": 4.6496324415879475e+17,
      "train_loss": 0.6741655204117548,
      "train_runtime": 5736.554,
      "train_samples_per_second": 3422.804,
      "train_steps_per_second": 13.37
    }
  ],
  "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
}