{
  "best_global_step": 2844,
  "best_metric": 0.38952478766441345,
  "best_model_checkpoint": "bert_base_rand_5_v1_qqp/checkpoint-2844",
  "epoch": 7.0,
  "eval_steps": 500,
  "global_step": 9954,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 1.669679045677185,
      "learning_rate": 4.9e-05,
      "loss": 0.4751,
      "step": 1422
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7920108830076675,
      "eval_combined_score": 0.7358767139264258,
      "eval_f1": 0.6797425448451841,
      "eval_loss": 0.4341031610965729,
      "eval_runtime": 25.4809,
      "eval_samples_per_second": 1586.68,
      "eval_steps_per_second": 6.201,
      "step": 1422
    },
    {
      "epoch": 2.0,
      "grad_norm": 2.3344945907592773,
      "learning_rate": 4.8e-05,
      "loss": 0.3657,
      "step": 2844
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8175364828097947,
      "eval_combined_score": 0.791802964728189,
      "eval_f1": 0.7660694466465832,
      "eval_loss": 0.38952478766441345,
      "eval_runtime": 25.2135,
      "eval_samples_per_second": 1603.509,
      "eval_steps_per_second": 6.266,
      "step": 2844
    },
    {
      "epoch": 3.0,
      "grad_norm": 3.8463003635406494,
      "learning_rate": 4.7e-05,
      "loss": 0.2876,
      "step": 4266
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8242146920603513,
      "eval_combined_score": 0.7997650236249403,
      "eval_f1": 0.7753153551895292,
      "eval_loss": 0.40631604194641113,
      "eval_runtime": 25.2174,
      "eval_samples_per_second": 1603.258,
      "eval_steps_per_second": 6.266,
      "step": 4266
    },
    {
      "epoch": 4.0,
      "grad_norm": 4.032034397125244,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.2268,
      "step": 5688
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8353697749196142,
      "eval_combined_score": 0.8064028704732492,
      "eval_f1": 0.7774359660268843,
      "eval_loss": 0.4113540053367615,
      "eval_runtime": 25.2081,
      "eval_samples_per_second": 1603.852,
      "eval_steps_per_second": 6.268,
      "step": 5688
    },
    {
      "epoch": 5.0,
      "grad_norm": 2.0207948684692383,
      "learning_rate": 4.5e-05,
      "loss": 0.1778,
      "step": 7110
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8396982438783082,
      "eval_combined_score": 0.812394799559607,
      "eval_f1": 0.7850913552409059,
      "eval_loss": 0.4330194592475891,
      "eval_runtime": 25.2478,
      "eval_samples_per_second": 1601.329,
      "eval_steps_per_second": 6.258,
      "step": 7110
    },
    {
      "epoch": 6.0,
      "grad_norm": 4.257767677307129,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.1429,
      "step": 8532
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8441256492703438,
      "eval_combined_score": 0.8127517859895048,
      "eval_f1": 0.7813779227086658,
      "eval_loss": 0.5096306204795837,
      "eval_runtime": 25.1961,
      "eval_samples_per_second": 1604.617,
      "eval_steps_per_second": 6.271,
      "step": 8532
    },
    {
      "epoch": 7.0,
      "grad_norm": 5.623451232910156,
      "learning_rate": 4.3e-05,
      "loss": 0.1145,
      "step": 9954
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8407865446450655,
      "eval_combined_score": 0.8129200696645036,
      "eval_f1": 0.7850535946839416,
      "eval_loss": 0.554709792137146,
      "eval_runtime": 25.1988,
      "eval_samples_per_second": 1604.443,
      "eval_steps_per_second": 6.27,
      "step": 9954
    },
    {
      "epoch": 7.0,
      "step": 9954,
      "total_flos": 3.3506166766980096e+17,
      "train_loss": 0.2557692494238144,
      "train_runtime": 4257.8695,
      "train_samples_per_second": 4272.63,
      "train_steps_per_second": 16.698
    }
  ],
  "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": 3.3506166766980096e+17,
  "train_batch_size": 256,
  "trial_name": null,
  "trial_params": null
}