{
  "best_global_step": 4266,
  "best_metric": 0.40034225583076477,
  "best_model_checkpoint": "bert_base_km_100_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": 2.1033525466918945,
      "learning_rate": 4.9e-05,
      "loss": 0.4851,
      "step": 1422
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7841949047736829,
      "eval_combined_score": 0.7207300485293563,
      "eval_f1": 0.6572651922850297,
      "eval_loss": 0.45431217551231384,
      "eval_runtime": 25.7195,
      "eval_samples_per_second": 1571.958,
      "eval_steps_per_second": 6.143,
      "step": 1422
    },
    {
      "epoch": 2.0,
      "grad_norm": 2.518594980239868,
      "learning_rate": 4.8e-05,
      "loss": 0.3784,
      "step": 2844
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8115755627009646,
      "eval_combined_score": 0.7842808271588273,
      "eval_f1": 0.7569860916166901,
      "eval_loss": 0.40324944257736206,
      "eval_runtime": 25.423,
      "eval_samples_per_second": 1590.294,
      "eval_steps_per_second": 6.215,
      "step": 2844
    },
    {
      "epoch": 3.0,
      "grad_norm": 3.3977935314178467,
      "learning_rate": 4.7e-05,
      "loss": 0.2978,
      "step": 4266
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8197378184516448,
      "eval_combined_score": 0.7942450990976335,
      "eval_f1": 0.7687523797436223,
      "eval_loss": 0.40034225583076477,
      "eval_runtime": 25.4506,
      "eval_samples_per_second": 1588.571,
      "eval_steps_per_second": 6.208,
      "step": 4266
    },
    {
      "epoch": 4.0,
      "grad_norm": 4.080958366394043,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.2292,
      "step": 5688
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8305218896858768,
      "eval_combined_score": 0.796796214552482,
      "eval_f1": 0.7630705394190871,
      "eval_loss": 0.43787506222724915,
      "eval_runtime": 25.4377,
      "eval_samples_per_second": 1589.372,
      "eval_steps_per_second": 6.211,
      "step": 5688
    },
    {
      "epoch": 5.0,
      "grad_norm": 2.104090452194214,
      "learning_rate": 4.5e-05,
      "loss": 0.1741,
      "step": 7110
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8309176354192431,
      "eval_combined_score": 0.8029579948539833,
      "eval_f1": 0.7749983542887235,
      "eval_loss": 0.47364217042922974,
      "eval_runtime": 25.4349,
      "eval_samples_per_second": 1589.547,
      "eval_steps_per_second": 6.212,
      "step": 7110
    },
    {
      "epoch": 6.0,
      "grad_norm": 4.073667049407959,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.1338,
      "step": 8532
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8355181795696265,
      "eval_combined_score": 0.8028503256156123,
      "eval_f1": 0.770182471661598,
      "eval_loss": 0.5308232307434082,
      "eval_runtime": 25.4722,
      "eval_samples_per_second": 1587.219,
      "eval_steps_per_second": 6.203,
      "step": 8532
    },
    {
      "epoch": 7.0,
      "grad_norm": 3.648453950881958,
      "learning_rate": 4.3e-05,
      "loss": 0.1055,
      "step": 9954
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8292357160524363,
      "eval_combined_score": 0.8033348148282813,
      "eval_f1": 0.7774339136041264,
      "eval_loss": 0.5860157608985901,
      "eval_runtime": 25.4436,
      "eval_samples_per_second": 1589.002,
      "eval_steps_per_second": 6.21,
      "step": 9954
    },
    {
      "epoch": 8.0,
      "grad_norm": 2.077934741973877,
      "learning_rate": 4.2e-05,
      "loss": 0.0843,
      "step": 11376
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8314617858026219,
      "eval_combined_score": 0.8038425513577969,
      "eval_f1": 0.7762233169129721,
      "eval_loss": 0.6411646604537964,
      "eval_runtime": 25.4895,
      "eval_samples_per_second": 1586.142,
      "eval_steps_per_second": 6.199,
      "step": 11376
    },
    {
      "epoch": 8.0,
      "step": 11376,
      "total_flos": 3.8292762019405824e+17,
      "train_loss": 0.23603766611692104,
      "train_runtime": 4899.3821,
      "train_samples_per_second": 3713.183,
      "train_steps_per_second": 14.512
    }
  ],
  "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": 3.8292762019405824e+17,
  "train_batch_size": 256,
  "trial_name": null,
  "trial_params": null
}