{
  "best_global_step": 6136,
  "best_metric": 0.7743538618087769,
  "best_model_checkpoint": "bert_base_rand_100_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": 1.185520887374878,
      "learning_rate": 4.9e-05,
      "loss": 0.9696,
      "step": 1534
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5816607233825777,
      "eval_loss": 0.9075585603713989,
      "eval_runtime": 6.1744,
      "eval_samples_per_second": 1589.626,
      "eval_steps_per_second": 6.316,
      "step": 1534
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.6829042434692383,
      "learning_rate": 4.8e-05,
      "loss": 0.8564,
      "step": 3068
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.634029546612328,
      "eval_loss": 0.8225669264793396,
      "eval_runtime": 6.1397,
      "eval_samples_per_second": 1598.602,
      "eval_steps_per_second": 6.352,
      "step": 3068
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.659785509109497,
      "learning_rate": 4.7e-05,
      "loss": 0.7602,
      "step": 4602
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6570555272542028,
      "eval_loss": 0.7792320847511292,
      "eval_runtime": 6.1378,
      "eval_samples_per_second": 1599.096,
      "eval_steps_per_second": 6.354,
      "step": 4602
    },
    {
      "epoch": 4.0,
      "grad_norm": 1.9616304636001587,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.6808,
      "step": 6136
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6659195109526236,
      "eval_loss": 0.7743538618087769,
      "eval_runtime": 6.1387,
      "eval_samples_per_second": 1598.878,
      "eval_steps_per_second": 6.353,
      "step": 6136
    },
    {
      "epoch": 5.0,
      "grad_norm": 2.228065013885498,
      "learning_rate": 4.5e-05,
      "loss": 0.6102,
      "step": 7670
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.6746816097809475,
      "eval_loss": 0.7905533909797668,
      "eval_runtime": 6.1368,
      "eval_samples_per_second": 1599.365,
      "eval_steps_per_second": 6.355,
      "step": 7670
    },
    {
      "epoch": 6.0,
      "grad_norm": 2.5042777061462402,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.537,
      "step": 9204
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.6722363728986246,
      "eval_loss": 0.8281874060630798,
      "eval_runtime": 6.1516,
      "eval_samples_per_second": 1595.509,
      "eval_steps_per_second": 6.34,
      "step": 9204
    },
    {
      "epoch": 7.0,
      "grad_norm": 2.568113088607788,
      "learning_rate": 4.3e-05,
      "loss": 0.4654,
      "step": 10738
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.6683647478349465,
      "eval_loss": 0.895623505115509,
      "eval_runtime": 6.141,
      "eval_samples_per_second": 1598.283,
      "eval_steps_per_second": 6.351,
      "step": 10738
    },
    {
      "epoch": 8.0,
      "grad_norm": 2.8277199268341064,
      "learning_rate": 4.2e-05,
      "loss": 0.3956,
      "step": 12272
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.6696892511462048,
      "eval_loss": 0.9840532541275024,
      "eval_runtime": 6.1432,
      "eval_samples_per_second": 1597.698,
      "eval_steps_per_second": 6.348,
      "step": 12272
    },
    {
      "epoch": 9.0,
      "grad_norm": 3.975034475326538,
      "learning_rate": 4.1e-05,
      "loss": 0.3336,
      "step": 13806
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.6606214977075904,
      "eval_loss": 1.1014842987060547,
      "eval_runtime": 6.1556,
      "eval_samples_per_second": 1594.476,
      "eval_steps_per_second": 6.336,
      "step": 13806
    },
    {
      "epoch": 9.0,
      "step": 13806,
      "total_flos": 4.6496324415879475e+17,
      "train_loss": 0.6232022944868569,
      "train_runtime": 5727.1087,
      "train_samples_per_second": 3428.449,
      "train_steps_per_second": 13.392
    }
  ],
  "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
}