{
  "best_global_step": 4266,
  "best_metric": 0.3982488512992859,
  "best_model_checkpoint": "tiny_bert_km_5_v1_qqp/checkpoint-4266",
  "epoch": 8.0,
  "eval_steps": 500,
  "global_step": 11376,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 1.9771767854690552,
      "learning_rate": 4.9e-05,
      "loss": 0.4854,
      "step": 1422
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7934454612911205,
      "eval_combined_score": 0.7517149167889879,
      "eval_f1": 0.7099843722868554,
      "eval_loss": 0.4320138096809387,
      "eval_runtime": 11.5465,
      "eval_samples_per_second": 3501.498,
      "eval_steps_per_second": 13.684,
      "step": 1422
    },
    {
      "epoch": 2.0,
      "grad_norm": 2.2720088958740234,
      "learning_rate": 4.8e-05,
      "loss": 0.3893,
      "step": 2844
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8129606727677468,
      "eval_combined_score": 0.7775239925912404,
      "eval_f1": 0.742087312414734,
      "eval_loss": 0.40190523862838745,
      "eval_runtime": 11.3551,
      "eval_samples_per_second": 3560.508,
      "eval_steps_per_second": 13.914,
      "step": 2844
    },
    {
      "epoch": 3.0,
      "grad_norm": 3.3009727001190186,
      "learning_rate": 4.7e-05,
      "loss": 0.3257,
      "step": 4266
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8229037843185754,
      "eval_combined_score": 0.7895255928608748,
      "eval_f1": 0.7561474014031742,
      "eval_loss": 0.3982488512992859,
      "eval_runtime": 11.3549,
      "eval_samples_per_second": 3560.578,
      "eval_steps_per_second": 13.915,
      "step": 4266
    },
    {
      "epoch": 4.0,
      "grad_norm": 4.036802291870117,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.2734,
      "step": 5688
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8248330447687361,
      "eval_combined_score": 0.7847485096932807,
      "eval_f1": 0.7446639746178252,
      "eval_loss": 0.43422242999076843,
      "eval_runtime": 11.3579,
      "eval_samples_per_second": 3559.649,
      "eval_steps_per_second": 13.911,
      "step": 5688
    },
    {
      "epoch": 5.0,
      "grad_norm": 3.4999232292175293,
      "learning_rate": 4.5e-05,
      "loss": 0.2309,
      "step": 7110
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8301756121691813,
      "eval_combined_score": 0.7970016750994504,
      "eval_f1": 0.7638277380297194,
      "eval_loss": 0.4551413953304291,
      "eval_runtime": 11.3406,
      "eval_samples_per_second": 3565.062,
      "eval_steps_per_second": 13.932,
      "step": 7110
    },
    {
      "epoch": 6.0,
      "grad_norm": 3.7123641967773438,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.1964,
      "step": 8532
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8268859757605739,
      "eval_combined_score": 0.7995084282188838,
      "eval_f1": 0.7721308806771936,
      "eval_loss": 0.4515395164489746,
      "eval_runtime": 11.3722,
      "eval_samples_per_second": 3555.17,
      "eval_steps_per_second": 13.894,
      "step": 8532
    },
    {
      "epoch": 7.0,
      "grad_norm": 4.537855625152588,
      "learning_rate": 4.3e-05,
      "loss": 0.1681,
      "step": 9954
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8293346524857779,
      "eval_combined_score": 0.8022382054712015,
      "eval_f1": 0.7751417584566251,
      "eval_loss": 0.49950671195983887,
      "eval_runtime": 11.3648,
      "eval_samples_per_second": 3557.485,
      "eval_steps_per_second": 13.903,
      "step": 9954
    },
    {
      "epoch": 8.0,
      "grad_norm": 3.828883647918701,
      "learning_rate": 4.2e-05,
      "loss": 0.146,
      "step": 11376
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8303734850358645,
      "eval_combined_score": 0.8031132688612451,
      "eval_f1": 0.7758530526866257,
      "eval_loss": 0.5505490899085999,
      "eval_runtime": 11.3526,
      "eval_samples_per_second": 3561.298,
      "eval_steps_per_second": 13.918,
      "step": 11376
    },
    {
      "epoch": 8.0,
      "step": 11376,
      "total_flos": 7.633075201391002e+16,
      "train_loss": 0.27687875127993555,
      "train_runtime": 1613.3516,
      "train_samples_per_second": 11276.091,
      "train_steps_per_second": 44.07
    }
  ],
  "logging_steps": 1,
  "max_steps": 71100,
  "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": 7.633075201391002e+16,
  "train_batch_size": 256,
  "trial_name": null,
  "trial_params": null
}