{
  "best_global_step": 7670,
  "best_metric": 0.836931049823761,
  "best_model_checkpoint": "tiny_bert_km_10_v1_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": 1.0128730535507202,
      "learning_rate": 4.9e-05,
      "loss": 0.9971,
      "step": 1534
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5416199694345389,
      "eval_loss": 0.9425157308578491,
      "eval_runtime": 2.8647,
      "eval_samples_per_second": 3426.211,
      "eval_steps_per_second": 13.614,
      "step": 1534
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.3565973043441772,
      "learning_rate": 4.8e-05,
      "loss": 0.9096,
      "step": 3068
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.5772796739684157,
      "eval_loss": 0.886116623878479,
      "eval_runtime": 2.7743,
      "eval_samples_per_second": 3537.791,
      "eval_steps_per_second": 14.057,
      "step": 3068
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.5406967401504517,
      "learning_rate": 4.7e-05,
      "loss": 0.8541,
      "step": 4602
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6075394803871625,
      "eval_loss": 0.8557488322257996,
      "eval_runtime": 2.7714,
      "eval_samples_per_second": 3541.538,
      "eval_steps_per_second": 14.072,
      "step": 4602
    },
    {
      "epoch": 4.0,
      "grad_norm": 1.644335389137268,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.8053,
      "step": 6136
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6218033622007132,
      "eval_loss": 0.8488360643386841,
      "eval_runtime": 2.792,
      "eval_samples_per_second": 3515.397,
      "eval_steps_per_second": 13.968,
      "step": 6136
    },
    {
      "epoch": 5.0,
      "grad_norm": 1.686502456665039,
      "learning_rate": 4.5e-05,
      "loss": 0.7589,
      "step": 7670
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.6271013754457463,
      "eval_loss": 0.836931049823761,
      "eval_runtime": 2.8834,
      "eval_samples_per_second": 3403.986,
      "eval_steps_per_second": 13.526,
      "step": 7670
    },
    {
      "epoch": 6.0,
      "grad_norm": 1.943951964378357,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.7141,
      "step": 9204
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.6303616912888436,
      "eval_loss": 0.861319363117218,
      "eval_runtime": 2.8779,
      "eval_samples_per_second": 3410.461,
      "eval_steps_per_second": 13.551,
      "step": 9204
    },
    {
      "epoch": 7.0,
      "grad_norm": 1.9541428089141846,
      "learning_rate": 4.3e-05,
      "loss": 0.6695,
      "step": 10738
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.6335201222618441,
      "eval_loss": 0.8625016808509827,
      "eval_runtime": 2.8785,
      "eval_samples_per_second": 3409.718,
      "eval_steps_per_second": 13.549,
      "step": 10738
    },
    {
      "epoch": 8.0,
      "grad_norm": 2.083730459213257,
      "learning_rate": 4.2e-05,
      "loss": 0.6257,
      "step": 12272
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.6352521650534896,
      "eval_loss": 0.8963066935539246,
      "eval_runtime": 2.8807,
      "eval_samples_per_second": 3407.176,
      "eval_steps_per_second": 13.538,
      "step": 12272
    },
    {
      "epoch": 9.0,
      "grad_norm": 2.7612860202789307,
      "learning_rate": 4.1e-05,
      "loss": 0.5833,
      "step": 13806
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.6297503820682628,
      "eval_loss": 0.9533789157867432,
      "eval_runtime": 2.86,
      "eval_samples_per_second": 3431.842,
      "eval_steps_per_second": 13.636,
      "step": 13806
    },
    {
      "epoch": 10.0,
      "grad_norm": 3.026907205581665,
      "learning_rate": 4e-05,
      "loss": 0.5404,
      "step": 15340
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.6322975038206826,
      "eval_loss": 0.9586830735206604,
      "eval_runtime": 2.8614,
      "eval_samples_per_second": 3430.185,
      "eval_steps_per_second": 13.63,
      "step": 15340
    },
    {
      "epoch": 10.0,
      "step": 15340,
      "total_flos": 1.029836100834816e+17,
      "train_loss": 0.7457945424543677,
      "train_runtime": 2096.8973,
      "train_samples_per_second": 9363.882,
      "train_steps_per_second": 36.578
    }
  ],
  "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
}