{
  "best_global_step": 4266,
  "best_metric": 0.41218966245651245,
  "best_model_checkpoint": "tiny_bert_km_20_v1_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": 1.566367506980896,
      "learning_rate": 4.9e-05,
      "loss": 0.4973,
      "step": 1422
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7837991590403166,
      "eval_combined_score": 0.7351328375279778,
      "eval_f1": 0.686466516015639,
      "eval_loss": 0.45083168148994446,
      "eval_runtime": 12.0361,
      "eval_samples_per_second": 3359.063,
      "eval_steps_per_second": 13.127,
      "step": 1422
    },
    {
      "epoch": 2.0,
      "grad_norm": 2.3436715602874756,
      "learning_rate": 4.8e-05,
      "loss": 0.409,
      "step": 2844
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.802003462775167,
      "eval_combined_score": 0.7610590720400082,
      "eval_f1": 0.7201146813048495,
      "eval_loss": 0.4178438186645508,
      "eval_runtime": 11.8445,
      "eval_samples_per_second": 3413.4,
      "eval_steps_per_second": 13.34,
      "step": 2844
    },
    {
      "epoch": 3.0,
      "grad_norm": 3.1545252799987793,
      "learning_rate": 4.7e-05,
      "loss": 0.35,
      "step": 4266
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8135790254761316,
      "eval_combined_score": 0.7745289575139757,
      "eval_f1": 0.7354788895518197,
      "eval_loss": 0.41218966245651245,
      "eval_runtime": 11.8297,
      "eval_samples_per_second": 3417.671,
      "eval_steps_per_second": 13.356,
      "step": 4266
    },
    {
      "epoch": 4.0,
      "grad_norm": 2.3504576683044434,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.3027,
      "step": 5688
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8184516448182043,
      "eval_combined_score": 0.7798455784550726,
      "eval_f1": 0.741239512091941,
      "eval_loss": 0.42339518666267395,
      "eval_runtime": 11.7949,
      "eval_samples_per_second": 3427.766,
      "eval_steps_per_second": 13.396,
      "step": 5688
    },
    {
      "epoch": 5.0,
      "grad_norm": 2.9670097827911377,
      "learning_rate": 4.5e-05,
      "loss": 0.2638,
      "step": 7110
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8230274548602523,
      "eval_combined_score": 0.7899003746069286,
      "eval_f1": 0.7567732943536051,
      "eval_loss": 0.43965277075767517,
      "eval_runtime": 11.8862,
      "eval_samples_per_second": 3401.425,
      "eval_steps_per_second": 13.293,
      "step": 7110
    },
    {
      "epoch": 6.0,
      "grad_norm": 2.1944124698638916,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.2301,
      "step": 8532
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8204798417017066,
      "eval_combined_score": 0.7908570667157129,
      "eval_f1": 0.761234291729719,
      "eval_loss": 0.4461761713027954,
      "eval_runtime": 11.8322,
      "eval_samples_per_second": 3416.952,
      "eval_steps_per_second": 13.353,
      "step": 8532
    },
    {
      "epoch": 7.0,
      "grad_norm": 4.315927028656006,
      "learning_rate": 4.3e-05,
      "loss": 0.2039,
      "step": 9954
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8193668068266139,
      "eval_combined_score": 0.7913855448588637,
      "eval_f1": 0.7634042828911135,
      "eval_loss": 0.4793422818183899,
      "eval_runtime": 11.8419,
      "eval_samples_per_second": 3414.147,
      "eval_steps_per_second": 13.342,
      "step": 9954
    },
    {
      "epoch": 8.0,
      "grad_norm": 2.269212484359741,
      "learning_rate": 4.2e-05,
      "loss": 0.1794,
      "step": 11376
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8208755874350729,
      "eval_combined_score": 0.7926663762973842,
      "eval_f1": 0.7644571651596955,
      "eval_loss": 0.5060178637504578,
      "eval_runtime": 11.815,
      "eval_samples_per_second": 3421.917,
      "eval_steps_per_second": 13.373,
      "step": 11376
    },
    {
      "epoch": 8.0,
      "step": 11376,
      "total_flos": 7.633075201391002e+16,
      "train_loss": 0.30452447534445803,
      "train_runtime": 1634.6208,
      "train_samples_per_second": 11129.37,
      "train_steps_per_second": 43.496
    }
  ],
  "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
}