{
  "best_global_step": 6136,
  "best_metric": 0.8063306212425232,
  "best_model_checkpoint": "bert_base_rand_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": 1.1981600522994995,
      "learning_rate": 4.9e-05,
      "loss": 0.98,
      "step": 1534
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5750382068262863,
      "eval_loss": 0.905325174331665,
      "eval_runtime": 6.3505,
      "eval_samples_per_second": 1545.554,
      "eval_steps_per_second": 6.141,
      "step": 1534
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.4905537366867065,
      "learning_rate": 4.8e-05,
      "loss": 0.8735,
      "step": 3068
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6041772796739684,
      "eval_loss": 0.860352635383606,
      "eval_runtime": 6.2421,
      "eval_samples_per_second": 1572.4,
      "eval_steps_per_second": 6.248,
      "step": 3068
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.687028169631958,
      "learning_rate": 4.7e-05,
      "loss": 0.7969,
      "step": 4602
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6297503820682628,
      "eval_loss": 0.8222507238388062,
      "eval_runtime": 6.2368,
      "eval_samples_per_second": 1573.713,
      "eval_steps_per_second": 6.253,
      "step": 4602
    },
    {
      "epoch": 4.0,
      "grad_norm": 1.9236862659454346,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.7229,
      "step": 6136
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6495160468670402,
      "eval_loss": 0.8063306212425232,
      "eval_runtime": 6.3118,
      "eval_samples_per_second": 1555.032,
      "eval_steps_per_second": 6.179,
      "step": 6136
    },
    {
      "epoch": 5.0,
      "grad_norm": 2.2754101753234863,
      "learning_rate": 4.5e-05,
      "loss": 0.6538,
      "step": 7670
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.6612328069281712,
      "eval_loss": 0.8120583295822144,
      "eval_runtime": 6.2567,
      "eval_samples_per_second": 1568.728,
      "eval_steps_per_second": 6.233,
      "step": 7670
    },
    {
      "epoch": 6.0,
      "grad_norm": 2.9715635776519775,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.5838,
      "step": 9204
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.6595007641365257,
      "eval_loss": 0.8697757124900818,
      "eval_runtime": 6.2422,
      "eval_samples_per_second": 1572.365,
      "eval_steps_per_second": 6.248,
      "step": 9204
    },
    {
      "epoch": 7.0,
      "grad_norm": 2.3772523403167725,
      "learning_rate": 4.3e-05,
      "loss": 0.5103,
      "step": 10738
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.6576668364747835,
      "eval_loss": 0.9112951755523682,
      "eval_runtime": 6.2387,
      "eval_samples_per_second": 1573.242,
      "eval_steps_per_second": 6.251,
      "step": 10738
    },
    {
      "epoch": 8.0,
      "grad_norm": 3.106327772140503,
      "learning_rate": 4.2e-05,
      "loss": 0.4426,
      "step": 12272
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.6551197147223637,
      "eval_loss": 1.000422716140747,
      "eval_runtime": 6.3494,
      "eval_samples_per_second": 1545.813,
      "eval_steps_per_second": 6.142,
      "step": 12272
    },
    {
      "epoch": 9.0,
      "grad_norm": 4.314876556396484,
      "learning_rate": 4.1e-05,
      "loss": 0.3765,
      "step": 13806
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.655425369332654,
      "eval_loss": 1.1618634462356567,
      "eval_runtime": 6.244,
      "eval_samples_per_second": 1571.911,
      "eval_steps_per_second": 6.246,
      "step": 13806
    },
    {
      "epoch": 9.0,
      "step": 13806,
      "total_flos": 4.6496324415879475e+17,
      "train_loss": 0.6600460896332437,
      "train_runtime": 5801.0591,
      "train_samples_per_second": 3384.744,
      "train_steps_per_second": 13.222
    }
  ],
  "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
}