{
  "best_global_step": 7670,
  "best_metric": 0.7941984534263611,
  "best_model_checkpoint": "tiny_bert_rand_10_v2_mnli/checkpoint-7670",
  "epoch": 10.0,
  "eval_steps": 500,
  "global_step": 15340,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 1.068098545074463,
      "learning_rate": 4.9e-05,
      "loss": 0.9938,
      "step": 1534
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.547427407030056,
      "eval_loss": 0.9396762847900391,
      "eval_runtime": 2.8729,
      "eval_samples_per_second": 3416.367,
      "eval_steps_per_second": 13.575,
      "step": 1534
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.5339981317520142,
      "learning_rate": 4.8e-05,
      "loss": 0.9048,
      "step": 3068
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.5955170657157413,
      "eval_loss": 0.8722593784332275,
      "eval_runtime": 2.8012,
      "eval_samples_per_second": 3503.828,
      "eval_steps_per_second": 13.922,
      "step": 3068
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.7663615942001343,
      "learning_rate": 4.7e-05,
      "loss": 0.8336,
      "step": 4602
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6360672440142638,
      "eval_loss": 0.82261723279953,
      "eval_runtime": 2.7892,
      "eval_samples_per_second": 3518.932,
      "eval_steps_per_second": 13.983,
      "step": 4602
    },
    {
      "epoch": 4.0,
      "grad_norm": 1.8656636476516724,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.7696,
      "step": 6136
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6478858889454916,
      "eval_loss": 0.8077008128166199,
      "eval_runtime": 2.8007,
      "eval_samples_per_second": 3504.463,
      "eval_steps_per_second": 13.925,
      "step": 6136
    },
    {
      "epoch": 5.0,
      "grad_norm": 2.0056347846984863,
      "learning_rate": 4.5e-05,
      "loss": 0.7202,
      "step": 7670
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.6551197147223637,
      "eval_loss": 0.7941984534263611,
      "eval_runtime": 2.8251,
      "eval_samples_per_second": 3474.242,
      "eval_steps_per_second": 13.805,
      "step": 7670
    },
    {
      "epoch": 6.0,
      "grad_norm": 2.4566726684570312,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.6775,
      "step": 9204
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.6630667345899134,
      "eval_loss": 0.822736382484436,
      "eval_runtime": 2.9338,
      "eval_samples_per_second": 3345.474,
      "eval_steps_per_second": 13.293,
      "step": 9204
    },
    {
      "epoch": 7.0,
      "grad_norm": 2.1102540493011475,
      "learning_rate": 4.3e-05,
      "loss": 0.6399,
      "step": 10738
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.6600101884870097,
      "eval_loss": 0.8302655816078186,
      "eval_runtime": 2.822,
      "eval_samples_per_second": 3477.987,
      "eval_steps_per_second": 13.82,
      "step": 10738
    },
    {
      "epoch": 8.0,
      "grad_norm": 2.309288740158081,
      "learning_rate": 4.2e-05,
      "loss": 0.6018,
      "step": 12272
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.6581762608252675,
      "eval_loss": 0.8506349921226501,
      "eval_runtime": 2.9439,
      "eval_samples_per_second": 3334.039,
      "eval_steps_per_second": 13.248,
      "step": 12272
    },
    {
      "epoch": 9.0,
      "grad_norm": 3.0031216144561768,
      "learning_rate": 4.1e-05,
      "loss": 0.5655,
      "step": 13806
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.6542027508914926,
      "eval_loss": 0.9014685750007629,
      "eval_runtime": 2.7942,
      "eval_samples_per_second": 3512.676,
      "eval_steps_per_second": 13.958,
      "step": 13806
    },
    {
      "epoch": 10.0,
      "grad_norm": 3.36259126663208,
      "learning_rate": 4e-05,
      "loss": 0.5303,
      "step": 15340
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.656444218033622,
      "eval_loss": 0.925642728805542,
      "eval_runtime": 2.7952,
      "eval_samples_per_second": 3511.315,
      "eval_steps_per_second": 13.952,
      "step": 15340
    },
    {
      "epoch": 10.0,
      "step": 15340,
      "total_flos": 1.029836100834816e+17,
      "train_loss": 0.7237090746147226,
      "train_runtime": 2080.8841,
      "train_samples_per_second": 9435.941,
      "train_steps_per_second": 36.859
    }
  ],
  "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": 1.029836100834816e+17,
  "train_batch_size": 256,
  "trial_name": null,
  "trial_params": null
}