{
  "best_global_step": 4266,
  "best_metric": 0.4161389470100403,
  "best_model_checkpoint": "tiny_bert_rand_5_v2_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": 1.7300523519515991,
      "learning_rate": 4.9e-05,
      "loss": 0.4948,
      "step": 1422
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7834776156319565,
      "eval_combined_score": 0.7347445466968965,
      "eval_f1": 0.6860114777618365,
      "eval_loss": 0.44807204604148865,
      "eval_runtime": 11.998,
      "eval_samples_per_second": 3369.732,
      "eval_steps_per_second": 13.169,
      "step": 1422
    },
    {
      "epoch": 2.0,
      "grad_norm": 2.451735734939575,
      "learning_rate": 4.8e-05,
      "loss": 0.4052,
      "step": 2844
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8023497402918625,
      "eval_combined_score": 0.7576325672898148,
      "eval_f1": 0.7129153942877672,
      "eval_loss": 0.42114099860191345,
      "eval_runtime": 11.4595,
      "eval_samples_per_second": 3528.079,
      "eval_steps_per_second": 13.788,
      "step": 2844
    },
    {
      "epoch": 3.0,
      "grad_norm": 3.203301429748535,
      "learning_rate": 4.7e-05,
      "loss": 0.3464,
      "step": 4266
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8146178580262181,
      "eval_combined_score": 0.7758868704474419,
      "eval_f1": 0.7371558828686656,
      "eval_loss": 0.4161389470100403,
      "eval_runtime": 11.4612,
      "eval_samples_per_second": 3527.559,
      "eval_steps_per_second": 13.786,
      "step": 4266
    },
    {
      "epoch": 4.0,
      "grad_norm": 3.619699239730835,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.2996,
      "step": 5688
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8190699975265892,
      "eval_combined_score": 0.7806998865719529,
      "eval_f1": 0.7423297756173166,
      "eval_loss": 0.4293212294578552,
      "eval_runtime": 12.0834,
      "eval_samples_per_second": 3345.917,
      "eval_steps_per_second": 13.076,
      "step": 5688
    },
    {
      "epoch": 5.0,
      "grad_norm": 2.874742269515991,
      "learning_rate": 4.5e-05,
      "loss": 0.2623,
      "step": 7110
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8209992579767499,
      "eval_combined_score": 0.789118259695161,
      "eval_f1": 0.7572372614135722,
      "eval_loss": 0.43629372119903564,
      "eval_runtime": 11.5402,
      "eval_samples_per_second": 3503.407,
      "eval_steps_per_second": 13.691,
      "step": 7110
    },
    {
      "epoch": 6.0,
      "grad_norm": 2.806262731552124,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.2317,
      "step": 8532
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8217907494434825,
      "eval_combined_score": 0.7909998776671268,
      "eval_f1": 0.7602090058907711,
      "eval_loss": 0.45417508482933044,
      "eval_runtime": 11.5347,
      "eval_samples_per_second": 3505.072,
      "eval_steps_per_second": 13.698,
      "step": 8532
    },
    {
      "epoch": 7.0,
      "grad_norm": 3.6159353256225586,
      "learning_rate": 4.3e-05,
      "loss": 0.207,
      "step": 9954
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8216423447934702,
      "eval_combined_score": 0.7933131060407725,
      "eval_f1": 0.7649838672880749,
      "eval_loss": 0.4872119724750519,
      "eval_runtime": 11.5245,
      "eval_samples_per_second": 3508.174,
      "eval_steps_per_second": 13.71,
      "step": 9954
    },
    {
      "epoch": 8.0,
      "grad_norm": 3.678138494491577,
      "learning_rate": 4.2e-05,
      "loss": 0.184,
      "step": 11376
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8273311897106109,
      "eval_combined_score": 0.7958507637890484,
      "eval_f1": 0.7643703378674858,
      "eval_loss": 0.5372677445411682,
      "eval_runtime": 11.4885,
      "eval_samples_per_second": 3519.164,
      "eval_steps_per_second": 13.753,
      "step": 11376
    },
    {
      "epoch": 8.0,
      "step": 11376,
      "total_flos": 7.633075201391002e+16,
      "train_loss": 0.30389826006024195,
      "train_runtime": 1605.4025,
      "train_samples_per_second": 11331.925,
      "train_steps_per_second": 44.288
    }
  ],
  "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
}