{
  "best_global_step": 3726,
  "best_metric": 0.18097351491451263,
  "best_model_checkpoint": "./model\\checkpoint-3726",
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 5589,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.2683843263553409,
      "grad_norm": 4.979691505432129,
      "learning_rate": 4.557165861513688e-05,
      "loss": 0.618,
      "step": 500
    },
    {
      "epoch": 0.5367686527106817,
      "grad_norm": 10.513946533203125,
      "learning_rate": 4.109858650921453e-05,
      "loss": 0.3634,
      "step": 1000
    },
    {
      "epoch": 0.8051529790660226,
      "grad_norm": 0.9428963661193848,
      "learning_rate": 3.662551440329218e-05,
      "loss": 0.2803,
      "step": 1500
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9363758389261745,
      "eval_f1": 0.936185778659979,
      "eval_loss": 0.21076564490795135,
      "eval_runtime": 37.2544,
      "eval_samples_per_second": 199.977,
      "eval_steps_per_second": 12.509,
      "step": 1863
    },
    {
      "epoch": 1.0735373054213633,
      "grad_norm": 4.273161888122559,
      "learning_rate": 3.215244229736983e-05,
      "loss": 0.2266,
      "step": 2000
    },
    {
      "epoch": 1.3419216317767042,
      "grad_norm": 9.790454864501953,
      "learning_rate": 2.767937019144749e-05,
      "loss": 0.1773,
      "step": 2500
    },
    {
      "epoch": 1.6103059581320451,
      "grad_norm": 0.9836813807487488,
      "learning_rate": 2.3215244229736983e-05,
      "loss": 0.1549,
      "step": 3000
    },
    {
      "epoch": 1.8786902844873858,
      "grad_norm": 5.523079872131348,
      "learning_rate": 1.8751118268026483e-05,
      "loss": 0.1524,
      "step": 3500
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9499328859060403,
      "eval_f1": 0.9499109066150679,
      "eval_loss": 0.18097351491451263,
      "eval_runtime": 37.2921,
      "eval_samples_per_second": 199.774,
      "eval_steps_per_second": 12.496,
      "step": 3726
    },
    {
      "epoch": 2.1470746108427265,
      "grad_norm": 0.0565815269947052,
      "learning_rate": 1.4278046162104134e-05,
      "loss": 0.1086,
      "step": 4000
    },
    {
      "epoch": 2.4154589371980677,
      "grad_norm": 0.1514054834842682,
      "learning_rate": 9.804974056181785e-06,
      "loss": 0.0918,
      "step": 4500
    },
    {
      "epoch": 2.6838432635534084,
      "grad_norm": 0.029303351417183876,
      "learning_rate": 5.331901950259439e-06,
      "loss": 0.084,
      "step": 5000
    },
    {
      "epoch": 2.952227589908749,
      "grad_norm": 0.0346352718770504,
      "learning_rate": 8.588298443370907e-07,
      "loss": 0.0709,
      "step": 5500
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9575838926174497,
      "eval_f1": 0.9576189144243515,
      "eval_loss": 0.19314545392990112,
      "eval_runtime": 37.4152,
      "eval_samples_per_second": 199.117,
      "eval_steps_per_second": 12.455,
      "step": 5589
    }
  ],
  "logging_steps": 500,
  "max_steps": 5589,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 2.3521550204630016e+16,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}