{
  "best_metric": 0.09445525705814362,
  "best_model_checkpoint": "/kaggle/working/checkpoint-48000",
  "epoch": 1.9991670137442732,
  "global_step": 48000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.25,
      "learning_rate": 1.750104123281966e-05,
      "loss": 0.0319,
      "step": 6000
    },
    {
      "epoch": 0.25,
      "eval_accuracy": 0.9821939782823297,
      "eval_f1": 0.9897475612268823,
      "eval_loss": 0.16341181099414825,
      "eval_precision": 0.9992109831150386,
      "eval_recall": 0.9804617117117117,
      "eval_runtime": 789.0185,
      "eval_samples_per_second": 102.71,
      "eval_steps_per_second": 12.839,
      "step": 6000
    },
    {
      "epoch": 0.5,
      "learning_rate": 1.5002082465639318e-05,
      "loss": 0.0147,
      "step": 12000
    },
    {
      "epoch": 0.5,
      "eval_accuracy": 0.9766658440276407,
      "eval_f1": 0.9865158764680296,
      "eval_loss": 0.10613188147544861,
      "eval_precision": 0.999638723102935,
      "eval_recall": 0.9737331081081081,
      "eval_runtime": 793.5025,
      "eval_samples_per_second": 102.129,
      "eval_steps_per_second": 12.766,
      "step": 12000
    },
    {
      "epoch": 0.75,
      "learning_rate": 1.2503123698458976e-05,
      "loss": 0.0107,
      "step": 18000
    },
    {
      "epoch": 0.75,
      "eval_accuracy": 0.9957428430404739,
      "eval_f1": 0.9975677676336846,
      "eval_loss": 0.043012455105781555,
      "eval_precision": 0.9992232186992444,
      "eval_recall": 0.9959177927927928,
      "eval_runtime": 792.2966,
      "eval_samples_per_second": 102.285,
      "eval_steps_per_second": 12.786,
      "step": 18000
    },
    {
      "epoch": 1.0,
      "learning_rate": 1.0004164931278636e-05,
      "loss": 0.0074,
      "step": 24000
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9847235932872656,
      "eval_f1": 0.9912112564069798,
      "eval_loss": 0.09504717588424683,
      "eval_precision": 0.9998567786657501,
      "eval_recall": 0.9827139639639639,
      "eval_runtime": 790.115,
      "eval_samples_per_second": 102.567,
      "eval_steps_per_second": 12.821,
      "step": 24000
    },
    {
      "epoch": 1.25,
      "learning_rate": 7.505206164098294e-06,
      "loss": 0.0041,
      "step": 30000
    },
    {
      "epoch": 1.25,
      "eval_accuracy": 0.9903998025666337,
      "eval_f1": 0.9944949973819397,
      "eval_loss": 0.08727405965328217,
      "eval_precision": 0.9998292689867114,
      "eval_recall": 0.9892173423423424,
      "eval_runtime": 788.7991,
      "eval_samples_per_second": 102.738,
      "eval_steps_per_second": 12.842,
      "step": 30000
    },
    {
      "epoch": 1.5,
      "learning_rate": 5.006247396917952e-06,
      "loss": 0.0031,
      "step": 36000
    },
    {
      "epoch": 1.5,
      "eval_accuracy": 0.9819718657453109,
      "eval_f1": 0.9896102233695304,
      "eval_loss": 0.1629093438386917,
      "eval_precision": 1.0,
      "eval_recall": 0.9794341216216216,
      "eval_runtime": 788.9221,
      "eval_samples_per_second": 102.722,
      "eval_steps_per_second": 12.84,
      "step": 36000
    },
    {
      "epoch": 1.75,
      "learning_rate": 2.5072886297376097e-06,
      "loss": 0.0022,
      "step": 42000
    },
    {
      "epoch": 1.75,
      "eval_accuracy": 0.9804540967423495,
      "eval_f1": 0.9887258181611126,
      "eval_loss": 0.1487206518650055,
      "eval_precision": 0.9999856028103314,
      "eval_recall": 0.9777167792792792,
      "eval_runtime": 791.8683,
      "eval_samples_per_second": 102.34,
      "eval_steps_per_second": 12.793,
      "step": 42000
    },
    {
      "epoch": 2.0,
      "learning_rate": 8.329862557267807e-09,
      "loss": 0.001,
      "step": 48000
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9904368213228035,
      "eval_f1": 0.9945157204220418,
      "eval_loss": 0.09445525705814362,
      "eval_precision": 0.9999430791342336,
      "eval_recall": 0.9891469594594594,
      "eval_runtime": 792.2508,
      "eval_samples_per_second": 102.291,
      "eval_steps_per_second": 12.786,
      "step": 48000
    }
  ],
  "max_steps": 48020,
  "num_train_epochs": 2,
  "total_flos": 7.687934370520642e+16,
  "trial_name": null,
  "trial_params": null
}