{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 9.0,
  "global_step": 31725,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "learning_rate": 1.8e-05,
      "loss": 0.0026,
      "step": 3525
    },
    {
      "epoch": 1.0,
      "eval_EG_f1": 0.9988826815642459,
      "eval_ET_f1": 0.9997551420176297,
      "eval_TE_f1": 0.9980463980463979,
      "eval_loss": 0.000998740317299962,
      "eval_overall_accuracy": 0.9997666459240822,
      "eval_overall_f1": 0.9988922736084187,
      "eval_overall_precision": 0.9988231221876082,
      "eval_overall_recall": 0.998961434604999,
      "eval_runtime": 142.8836,
      "eval_samples_per_second": 43.854,
      "eval_steps_per_second": 2.743,
      "step": 3525
    },
    {
      "epoch": 2.0,
      "learning_rate": 1.6000000000000003e-05,
      "loss": 0.0012,
      "step": 7050
    },
    {
      "epoch": 2.0,
      "eval_EG_f1": 0.998084902649218,
      "eval_ET_f1": 0.9993876301285977,
      "eval_TE_f1": 0.9978038067349927,
      "eval_loss": 0.001965524861589074,
      "eval_overall_accuracy": 0.9996499688861232,
      "eval_overall_f1": 0.9983733084138026,
      "eval_overall_precision": 0.9981314878892733,
      "eval_overall_recall": 0.9986152461399986,
      "eval_runtime": 140.6221,
      "eval_samples_per_second": 44.559,
      "eval_steps_per_second": 2.788,
      "step": 7050
    },
    {
      "epoch": 3.0,
      "learning_rate": 1.4e-05,
      "loss": 0.0007,
      "step": 10575
    },
    {
      "epoch": 3.0,
      "eval_EG_f1": 0.9988031596585016,
      "eval_ET_f1": 0.998775710088149,
      "eval_TE_f1": 0.9981682745145928,
      "eval_loss": 0.0016128044808283448,
      "eval_overall_accuracy": 0.9997666459240822,
      "eval_overall_f1": 0.9986154378677742,
      "eval_overall_precision": 0.9984771924967121,
      "eval_overall_recall": 0.9987537215259987,
      "eval_runtime": 141.5414,
      "eval_samples_per_second": 44.27,
      "eval_steps_per_second": 2.77,
      "step": 10575
    },
    {
      "epoch": 4.0,
      "learning_rate": 1.2e-05,
      "loss": 0.0008,
      "step": 14100
    },
    {
      "epoch": 4.0,
      "eval_EG_f1": 0.9992020427705075,
      "eval_ET_f1": 0.9997550820475141,
      "eval_TE_f1": 0.9992673992673993,
      "eval_loss": 0.0006879049469716847,
      "eval_overall_accuracy": 0.9999092511926986,
      "eval_overall_f1": 0.9993769039047354,
      "eval_overall_precision": 0.9993077189338871,
      "eval_overall_recall": 0.9994460984559994,
      "eval_runtime": 141.9363,
      "eval_samples_per_second": 44.147,
      "eval_steps_per_second": 2.762,
      "step": 14100
    },
    {
      "epoch": 5.0,
      "learning_rate": 1e-05,
      "loss": 0.0004,
      "step": 17625
    },
    {
      "epoch": 5.0,
      "eval_EG_f1": 0.9988828598787105,
      "eval_ET_f1": 0.9996326680543651,
      "eval_TE_f1": 0.9989012330606765,
      "eval_loss": 0.0006999427569098771,
      "eval_overall_accuracy": 0.9998055382700685,
      "eval_overall_f1": 0.9991000346140533,
      "eval_overall_precision": 0.9989617221568492,
      "eval_overall_recall": 0.9992383853769993,
      "eval_runtime": 233.5972,
      "eval_samples_per_second": 26.824,
      "eval_steps_per_second": 1.678,
      "step": 17625
    },
    {
      "epoch": 6.0,
      "learning_rate": 8.000000000000001e-06,
      "loss": 0.0004,
      "step": 21150
    },
    {
      "epoch": 6.0,
      "eval_EG_f1": 0.999680817108203,
      "eval_ET_f1": 0.9995101640950281,
      "eval_TE_f1": 0.9996336549029186,
      "eval_loss": 0.0005218852893449366,
      "eval_overall_accuracy": 0.9998962870773699,
      "eval_overall_f1": 0.9996192058711532,
      "eval_overall_precision": 0.9995846026031571,
      "eval_overall_recall": 0.9996538115349997,
      "eval_runtime": 141.8762,
      "eval_samples_per_second": 44.165,
      "eval_steps_per_second": 2.763,
      "step": 21150
    },
    {
      "epoch": 7.0,
      "learning_rate": 6e-06,
      "loss": 0.0,
      "step": 24675
    },
    {
      "epoch": 7.0,
      "eval_EG_f1": 0.999361634216406,
      "eval_ET_f1": 0.9997550820475141,
      "eval_TE_f1": 0.9993892756809576,
      "eval_loss": 0.0003737392835319042,
      "eval_overall_accuracy": 0.9999351794233562,
      "eval_overall_f1": 0.9994806993249091,
      "eval_overall_precision": 0.999515302589669,
      "eval_overall_recall": 0.9994460984559994,
      "eval_runtime": 244.7883,
      "eval_samples_per_second": 25.598,
      "eval_steps_per_second": 1.601,
      "step": 24675
    },
    {
      "epoch": 8.0,
      "learning_rate": 4.000000000000001e-06,
      "loss": 0.0001,
      "step": 28200
    },
    {
      "epoch": 8.0,
      "eval_EG_f1": 0.9995212256623045,
      "eval_ET_f1": 0.9997550820475141,
      "eval_TE_f1": 0.9997557997557998,
      "eval_loss": 0.00023818030604161322,
      "eval_overall_accuracy": 0.9999481435386849,
      "eval_overall_f1": 0.9996538355026309,
      "eval_overall_precision": 0.9995846313603323,
      "eval_overall_recall": 0.9997230492279997,
      "eval_runtime": 365.8307,
      "eval_samples_per_second": 17.128,
      "eval_steps_per_second": 1.072,
      "step": 28200
    },
    {
      "epoch": 9.0,
      "learning_rate": 2.0000000000000003e-06,
      "loss": 0.0001,
      "step": 31725
    },
    {
      "epoch": 9.0,
      "eval_EG_f1": 0.9988828598787105,
      "eval_ET_f1": 0.9997550820475141,
      "eval_TE_f1": 0.9992677568952892,
      "eval_loss": 0.001098370412364602,
      "eval_overall_accuracy": 0.9997925741547397,
      "eval_overall_f1": 0.9992385435414648,
      "eval_overall_precision": 0.9990310748148661,
      "eval_overall_recall": 0.9994460984559994,
      "eval_runtime": 243.6202,
      "eval_samples_per_second": 25.72,
      "eval_steps_per_second": 1.609,
      "step": 31725
    }
  ],
  "max_steps": 35250,
  "num_train_epochs": 10,
  "total_flos": 7174982911185900.0,
  "trial_name": null,
  "trial_params": null
}