{
  "best_global_step": 6136,
  "best_metric": 0.6076479554176331,
  "best_model_checkpoint": "tinybert_base_train_kd_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": 2.6455118656158447,
      "learning_rate": 4.90006518904824e-05,
      "loss": 0.8098,
      "step": 1534
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6904737646459501,
      "eval_loss": 0.7136619091033936,
      "eval_runtime": 4.8793,
      "eval_samples_per_second": 2011.543,
      "eval_steps_per_second": 7.993,
      "step": 1534
    },
    {
      "epoch": 2.0,
      "grad_norm": 2.2584753036499023,
      "learning_rate": 4.80006518904824e-05,
      "loss": 0.6666,
      "step": 3068
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7276617422312787,
      "eval_loss": 0.6575940847396851,
      "eval_runtime": 4.9057,
      "eval_samples_per_second": 2000.715,
      "eval_steps_per_second": 7.95,
      "step": 3068
    },
    {
      "epoch": 3.0,
      "grad_norm": 2.2152528762817383,
      "learning_rate": 4.70006518904824e-05,
      "loss": 0.5873,
      "step": 4602
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7377483443708609,
      "eval_loss": 0.627395749092102,
      "eval_runtime": 4.9208,
      "eval_samples_per_second": 1994.577,
      "eval_steps_per_second": 7.925,
      "step": 4602
    },
    {
      "epoch": 4.0,
      "grad_norm": 2.7104854583740234,
      "learning_rate": 4.60006518904824e-05,
      "loss": 0.522,
      "step": 6136
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.7492613346917982,
      "eval_loss": 0.6076479554176331,
      "eval_runtime": 4.8283,
      "eval_samples_per_second": 2032.81,
      "eval_steps_per_second": 8.077,
      "step": 6136
    },
    {
      "epoch": 5.0,
      "grad_norm": 2.4938905239105225,
      "learning_rate": 4.50006518904824e-05,
      "loss": 0.4623,
      "step": 7670
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.7570045848191543,
      "eval_loss": 0.6132704615592957,
      "eval_runtime": 4.9556,
      "eval_samples_per_second": 1980.583,
      "eval_steps_per_second": 7.87,
      "step": 7670
    },
    {
      "epoch": 6.0,
      "grad_norm": 2.4697744846343994,
      "learning_rate": 4.40006518904824e-05,
      "loss": 0.4069,
      "step": 9204
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.7575140091696383,
      "eval_loss": 0.6447588801383972,
      "eval_runtime": 5.0012,
      "eval_samples_per_second": 1962.533,
      "eval_steps_per_second": 7.798,
      "step": 9204
    },
    {
      "epoch": 7.0,
      "grad_norm": 3.08911395072937,
      "learning_rate": 4.30006518904824e-05,
      "loss": 0.3547,
      "step": 10738
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.760570555272542,
      "eval_loss": 0.6818413734436035,
      "eval_runtime": 4.9938,
      "eval_samples_per_second": 1965.428,
      "eval_steps_per_second": 7.81,
      "step": 10738
    },
    {
      "epoch": 8.0,
      "grad_norm": 3.197273015975952,
      "learning_rate": 4.20006518904824e-05,
      "loss": 0.3073,
      "step": 12272
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.7602649006622516,
      "eval_loss": 0.703414261341095,
      "eval_runtime": 5.5899,
      "eval_samples_per_second": 1755.83,
      "eval_steps_per_second": 6.977,
      "step": 12272
    },
    {
      "epoch": 9.0,
      "grad_norm": 2.992473602294922,
      "learning_rate": 4.10006518904824e-05,
      "loss": 0.2658,
      "step": 13806
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.7489556800815079,
      "eval_loss": 0.8077275156974792,
      "eval_runtime": 5.4298,
      "eval_samples_per_second": 1807.619,
      "eval_steps_per_second": 7.183,
      "step": 13806
    },
    {
      "epoch": 9.0,
      "step": 13806,
      "total_flos": 2.340951304214108e+17,
      "train_loss": 0.4869576942811889,
      "train_runtime": 3465.4123,
      "train_samples_per_second": 5666.021,
      "train_steps_per_second": 22.133
    }
  ],
  "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": 2.340951304214108e+17,
  "train_batch_size": 256,
  "trial_name": null,
  "trial_params": null
}