{
  "best_global_step": 6136,
  "best_metric": 0.773952066898346,
  "best_model_checkpoint": "bert_base_rand_50_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": 1.3422194719314575,
      "learning_rate": 4.9e-05,
      "loss": 0.9773,
      "step": 1534
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5768721344880285,
      "eval_loss": 0.9061076045036316,
      "eval_runtime": 6.2161,
      "eval_samples_per_second": 1578.974,
      "eval_steps_per_second": 6.274,
      "step": 1534
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.5185693502426147,
      "learning_rate": 4.8e-05,
      "loss": 0.868,
      "step": 3068
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6297503820682628,
      "eval_loss": 0.8349848985671997,
      "eval_runtime": 6.1145,
      "eval_samples_per_second": 1605.197,
      "eval_steps_per_second": 6.378,
      "step": 3068
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.8507657051086426,
      "learning_rate": 4.7e-05,
      "loss": 0.7722,
      "step": 4602
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6542027508914926,
      "eval_loss": 0.7925336956977844,
      "eval_runtime": 6.1568,
      "eval_samples_per_second": 1594.167,
      "eval_steps_per_second": 6.334,
      "step": 4602
    },
    {
      "epoch": 4.0,
      "grad_norm": 2.0266313552856445,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.6949,
      "step": 6136
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6617422312786552,
      "eval_loss": 0.773952066898346,
      "eval_runtime": 6.1302,
      "eval_samples_per_second": 1601.078,
      "eval_steps_per_second": 6.362,
      "step": 6136
    },
    {
      "epoch": 5.0,
      "grad_norm": 2.0750410556793213,
      "learning_rate": 4.5e-05,
      "loss": 0.626,
      "step": 7670
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.6711156393275599,
      "eval_loss": 0.8049560189247131,
      "eval_runtime": 6.1252,
      "eval_samples_per_second": 1602.403,
      "eval_steps_per_second": 6.367,
      "step": 7670
    },
    {
      "epoch": 6.0,
      "grad_norm": 2.611304998397827,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.5557,
      "step": 9204
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.677330616403464,
      "eval_loss": 0.831805408000946,
      "eval_runtime": 6.1373,
      "eval_samples_per_second": 1599.237,
      "eval_steps_per_second": 6.355,
      "step": 9204
    },
    {
      "epoch": 7.0,
      "grad_norm": 2.5233824253082275,
      "learning_rate": 4.3e-05,
      "loss": 0.484,
      "step": 10738
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.6729495669893021,
      "eval_loss": 0.8868318796157837,
      "eval_runtime": 6.1258,
      "eval_samples_per_second": 1602.248,
      "eval_steps_per_second": 6.367,
      "step": 10738
    },
    {
      "epoch": 8.0,
      "grad_norm": 3.048797130584717,
      "learning_rate": 4.2e-05,
      "loss": 0.4158,
      "step": 12272
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.6696892511462048,
      "eval_loss": 0.9602928757667542,
      "eval_runtime": 6.1243,
      "eval_samples_per_second": 1602.624,
      "eval_steps_per_second": 6.368,
      "step": 12272
    },
    {
      "epoch": 9.0,
      "grad_norm": 3.61452317237854,
      "learning_rate": 4.1e-05,
      "loss": 0.3534,
      "step": 13806
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.6623535404992359,
      "eval_loss": 1.0435149669647217,
      "eval_runtime": 6.1354,
      "eval_samples_per_second": 1599.734,
      "eval_steps_per_second": 6.357,
      "step": 13806
    },
    {
      "epoch": 9.0,
      "step": 13806,
      "total_flos": 4.6496324415879475e+17,
      "train_loss": 0.6385982048679016,
      "train_runtime": 5725.78,
      "train_samples_per_second": 3429.245,
      "train_steps_per_second": 13.396
    }
  ],
  "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
}