{
  "best_global_step": 1750,
  "best_metric": 0.5023255944252014,
  "best_model_checkpoint": "./results/checkpoint-1500",
  "epoch": 2.6041666666666665,
  "eval_steps": 250,
  "global_step": 2500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.2604166666666667,
      "eval_f1": 0.749100249517516,
      "eval_loss": 0.7024011015892029,
      "eval_runtime": 43.4725,
      "eval_samples_per_second": 170.729,
      "eval_steps_per_second": 5.337,
      "step": 250
    },
    {
      "epoch": 0.5208333333333334,
      "grad_norm": 8.554716110229492,
      "learning_rate": 1.7920833333333333e-05,
      "loss": 1.4443548583984376,
      "step": 500
    },
    {
      "epoch": 0.5208333333333334,
      "eval_f1": 0.7762738847259911,
      "eval_loss": 0.6240520477294922,
      "eval_runtime": 42.884,
      "eval_samples_per_second": 173.071,
      "eval_steps_per_second": 5.41,
      "step": 500
    },
    {
      "epoch": 0.78125,
      "eval_f1": 0.7936541579212073,
      "eval_loss": 0.5607157945632935,
      "eval_runtime": 42.7919,
      "eval_samples_per_second": 173.444,
      "eval_steps_per_second": 5.422,
      "step": 750
    },
    {
      "epoch": 1.0416666666666667,
      "grad_norm": 7.3601179122924805,
      "learning_rate": 1.5837500000000002e-05,
      "loss": 1.206718505859375,
      "step": 1000
    },
    {
      "epoch": 1.0416666666666667,
      "eval_f1": 0.8017692154501209,
      "eval_loss": 0.5573240518569946,
      "eval_runtime": 42.8238,
      "eval_samples_per_second": 173.315,
      "eval_steps_per_second": 5.418,
      "step": 1000
    },
    {
      "epoch": 1.3020833333333333,
      "eval_f1": 0.8070317417405216,
      "eval_loss": 0.5319511294364929,
      "eval_runtime": 42.9254,
      "eval_samples_per_second": 172.905,
      "eval_steps_per_second": 5.405,
      "step": 1250
    },
    {
      "epoch": 1.5625,
      "grad_norm": 7.68901252746582,
      "learning_rate": 1.3754166666666668e-05,
      "loss": 1.0192529296875,
      "step": 1500
    },
    {
      "epoch": 1.5625,
      "eval_f1": 0.8202881539259634,
      "eval_loss": 0.516129195690155,
      "eval_runtime": 42.8694,
      "eval_samples_per_second": 173.131,
      "eval_steps_per_second": 5.412,
      "step": 1500
    },
    {
      "epoch": 1.8229166666666665,
      "eval_f1": 0.8237475395887287,
      "eval_loss": 0.5023255944252014,
      "eval_runtime": 42.9086,
      "eval_samples_per_second": 172.972,
      "eval_steps_per_second": 5.407,
      "step": 1750
    },
    {
      "epoch": 2.0833333333333335,
      "grad_norm": 11.637955665588379,
      "learning_rate": 1.1670833333333334e-05,
      "loss": 0.9360776977539063,
      "step": 2000
    },
    {
      "epoch": 2.0833333333333335,
      "eval_f1": 0.8224263127844592,
      "eval_loss": 0.5213221311569214,
      "eval_runtime": 42.9603,
      "eval_samples_per_second": 172.764,
      "eval_steps_per_second": 5.4,
      "step": 2000
    },
    {
      "epoch": 2.34375,
      "eval_f1": 0.8301988049494216,
      "eval_loss": 0.5083206295967102,
      "eval_runtime": 42.7754,
      "eval_samples_per_second": 173.511,
      "eval_steps_per_second": 5.424,
      "step": 2250
    },
    {
      "epoch": 2.6041666666666665,
      "grad_norm": 9.42564582824707,
      "learning_rate": 9.587500000000001e-06,
      "loss": 0.801677734375,
      "step": 2500
    },
    {
      "epoch": 2.6041666666666665,
      "eval_f1": 0.8208076756586397,
      "eval_loss": 0.522527813911438,
      "eval_runtime": 42.9058,
      "eval_samples_per_second": 172.983,
      "eval_steps_per_second": 5.407,
      "step": 2500
    }
  ],
  "logging_steps": 500,
  "max_steps": 4800,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 3,
        "early_stopping_threshold": 0.0
      },
      "attributes": {
        "early_stopping_patience_counter": 3
      }
    },
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 4.209852481536e+16,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}