{
  "best_global_step": 4266,
  "best_metric": 0.38985955715179443,
  "best_model_checkpoint": "bert_base_km_50_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": 1.748183250427246,
      "learning_rate": 4.9e-05,
      "loss": 0.4831,
      "step": 1422
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7856789512738066,
      "eval_combined_score": 0.7248023969416391,
      "eval_f1": 0.6639258426094714,
      "eval_loss": 0.4464315176010132,
      "eval_runtime": 25.6464,
      "eval_samples_per_second": 1576.439,
      "eval_steps_per_second": 6.161,
      "step": 1422
    },
    {
      "epoch": 2.0,
      "grad_norm": 2.887326240539551,
      "learning_rate": 4.8e-05,
      "loss": 0.3789,
      "step": 2844
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8142221122928518,
      "eval_combined_score": 0.7791161002826176,
      "eval_f1": 0.7440100882723834,
      "eval_loss": 0.3968706429004669,
      "eval_runtime": 25.54,
      "eval_samples_per_second": 1583.009,
      "eval_steps_per_second": 6.186,
      "step": 2844
    },
    {
      "epoch": 3.0,
      "grad_norm": 3.3585822582244873,
      "learning_rate": 4.7e-05,
      "loss": 0.3007,
      "step": 4266
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.824387830818699,
      "eval_combined_score": 0.7931782254536033,
      "eval_f1": 0.7619686200885074,
      "eval_loss": 0.38985955715179443,
      "eval_runtime": 25.3302,
      "eval_samples_per_second": 1596.121,
      "eval_steps_per_second": 6.238,
      "step": 4266
    },
    {
      "epoch": 4.0,
      "grad_norm": 4.299377918243408,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.2341,
      "step": 5688
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8296809300024734,
      "eval_combined_score": 0.7937189565230829,
      "eval_f1": 0.7577569830436924,
      "eval_loss": 0.43014436960220337,
      "eval_runtime": 25.3567,
      "eval_samples_per_second": 1594.449,
      "eval_steps_per_second": 6.231,
      "step": 5688
    },
    {
      "epoch": 5.0,
      "grad_norm": 3.312682867050171,
      "learning_rate": 4.5e-05,
      "loss": 0.18,
      "step": 7110
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.829532525352461,
      "eval_combined_score": 0.8036981262745029,
      "eval_f1": 0.7778637271965448,
      "eval_loss": 0.44985222816467285,
      "eval_runtime": 25.3685,
      "eval_samples_per_second": 1593.712,
      "eval_steps_per_second": 6.228,
      "step": 7110
    },
    {
      "epoch": 6.0,
      "grad_norm": 4.098166465759277,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.1381,
      "step": 8532
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8346277516695523,
      "eval_combined_score": 0.8068380000912203,
      "eval_f1": 0.7790482485128883,
      "eval_loss": 0.5405033826828003,
      "eval_runtime": 25.4803,
      "eval_samples_per_second": 1586.715,
      "eval_steps_per_second": 6.201,
      "step": 8532
    },
    {
      "epoch": 7.0,
      "grad_norm": 3.422797679901123,
      "learning_rate": 4.3e-05,
      "loss": 0.1072,
      "step": 9954
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8325500865693791,
      "eval_combined_score": 0.8052186128384953,
      "eval_f1": 0.7778871391076115,
      "eval_loss": 0.5881912708282471,
      "eval_runtime": 25.3579,
      "eval_samples_per_second": 1594.376,
      "eval_steps_per_second": 6.231,
      "step": 9954
    },
    {
      "epoch": 8.0,
      "grad_norm": 2.0576579570770264,
      "learning_rate": 4.2e-05,
      "loss": 0.0866,
      "step": 11376
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8286420974523868,
      "eval_combined_score": 0.804016628907062,
      "eval_f1": 0.7793911603617374,
      "eval_loss": 0.5793479084968567,
      "eval_runtime": 25.3077,
      "eval_samples_per_second": 1597.535,
      "eval_steps_per_second": 6.243,
      "step": 11376
    },
    {
      "epoch": 8.0,
      "step": 11376,
      "total_flos": 3.8292762019405824e+17,
      "train_loss": 0.23859816630346029,
      "train_runtime": 4865.116,
      "train_samples_per_second": 3739.335,
      "train_steps_per_second": 14.614
    }
  ],
  "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
}