{
  "best_global_step": 2844,
  "best_metric": 0.3935573101043701,
  "best_model_checkpoint": "bert_base_km_50_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": 2.09523868560791,
      "learning_rate": 4.9e-05,
      "loss": 0.4867,
      "step": 1422
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7796685629483057,
      "eval_combined_score": 0.7094811878557531,
      "eval_f1": 0.6392938127632005,
      "eval_loss": 0.46669575572013855,
      "eval_runtime": 25.9605,
      "eval_samples_per_second": 1557.366,
      "eval_steps_per_second": 6.086,
      "step": 1422
    },
    {
      "epoch": 2.0,
      "grad_norm": 2.261963367462158,
      "learning_rate": 4.8e-05,
      "loss": 0.3802,
      "step": 2844
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8152114766262676,
      "eval_combined_score": 0.7866100462845004,
      "eval_f1": 0.7580086159427332,
      "eval_loss": 0.3935573101043701,
      "eval_runtime": 25.37,
      "eval_samples_per_second": 1593.615,
      "eval_steps_per_second": 6.228,
      "step": 2844
    },
    {
      "epoch": 3.0,
      "grad_norm": 4.046843528747559,
      "learning_rate": 4.7e-05,
      "loss": 0.3026,
      "step": 4266
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8212218649517685,
      "eval_combined_score": 0.7966261977253639,
      "eval_f1": 0.7720305304989592,
      "eval_loss": 0.3943940997123718,
      "eval_runtime": 25.2606,
      "eval_samples_per_second": 1600.516,
      "eval_steps_per_second": 6.255,
      "step": 4266
    },
    {
      "epoch": 4.0,
      "grad_norm": 4.8188042640686035,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.2343,
      "step": 5688
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8318575315359882,
      "eval_combined_score": 0.8016997027218857,
      "eval_f1": 0.7715418739077833,
      "eval_loss": 0.4298071265220642,
      "eval_runtime": 25.6427,
      "eval_samples_per_second": 1576.669,
      "eval_steps_per_second": 6.162,
      "step": 5688
    },
    {
      "epoch": 5.0,
      "grad_norm": 4.673189640045166,
      "learning_rate": 4.5e-05,
      "loss": 0.181,
      "step": 7110
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8309918377442493,
      "eval_combined_score": 0.8042346429415551,
      "eval_f1": 0.7774774481388609,
      "eval_loss": 0.46103334426879883,
      "eval_runtime": 25.5377,
      "eval_samples_per_second": 1583.148,
      "eval_steps_per_second": 6.187,
      "step": 7110
    },
    {
      "epoch": 6.0,
      "grad_norm": 4.1060051918029785,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.1403,
      "step": 8532
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.836804353203067,
      "eval_combined_score": 0.8069118082338418,
      "eval_f1": 0.7770192632646165,
      "eval_loss": 0.5261883735656738,
      "eval_runtime": 25.2801,
      "eval_samples_per_second": 1599.282,
      "eval_steps_per_second": 6.25,
      "step": 8532
    },
    {
      "epoch": 7.0,
      "grad_norm": 4.5016326904296875,
      "learning_rate": 4.3e-05,
      "loss": 0.1126,
      "step": 9954
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8348503586445709,
      "eval_combined_score": 0.8079482158856566,
      "eval_f1": 0.7810460731267421,
      "eval_loss": 0.5777826905250549,
      "eval_runtime": 25.242,
      "eval_samples_per_second": 1601.696,
      "eval_steps_per_second": 6.259,
      "step": 9954
    },
    {
      "epoch": 7.0,
      "step": 9954,
      "total_flos": 3.3506166766980096e+17,
      "train_loss": 0.2625292777440288,
      "train_runtime": 4273.1656,
      "train_samples_per_second": 4257.336,
      "train_steps_per_second": 16.639
    }
  ],
  "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
}