{
  "best_global_step": 3528,
  "best_metric": 0.8244905994684517,
  "best_model_checkpoint": "/projects/gyorilab/variants_ner_coarser/checkpoint-3528",
  "epoch": 8.0,
  "eval_steps": 500,
  "global_step": 3528,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9403182734357377,
      "eval_f1": 0.47378233512001866,
      "eval_loss": 0.01702982187271118,
      "eval_precision": 0.5675600223338917,
      "eval_recall": 0.4066,
      "eval_runtime": 8.8591,
      "eval_samples_per_second": 316.058,
      "eval_steps_per_second": 39.507,
      "step": 441
    },
    {
      "epoch": 1.1338627339761769,
      "grad_norm": 0.19258293509483337,
      "learning_rate": 1.7736961451247167e-05,
      "loss": 0.1472896270751953,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9491082668141367,
      "eval_f1": 0.5545414069456812,
      "eval_loss": 0.014995112083852291,
      "eval_precision": 0.6252510040160643,
      "eval_recall": 0.4982,
      "eval_runtime": 8.4202,
      "eval_samples_per_second": 332.534,
      "eval_steps_per_second": 41.567,
      "step": 882
    },
    {
      "epoch": 2.2677254679523537,
      "grad_norm": 0.7697210907936096,
      "learning_rate": 1.546938775510204e-05,
      "loss": 0.052211280822753905,
      "step": 1000
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9703556018016576,
      "eval_f1": 0.7212371134020619,
      "eval_loss": 0.017216848209500313,
      "eval_precision": 0.7442553191489362,
      "eval_recall": 0.6996,
      "eval_runtime": 8.3974,
      "eval_samples_per_second": 333.438,
      "eval_steps_per_second": 41.68,
      "step": 1323
    },
    {
      "epoch": 3.401588201928531,
      "grad_norm": 0.342649906873703,
      "learning_rate": 1.3201814058956916e-05,
      "loss": 0.03277603530883789,
      "step": 1500
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9722474877936971,
      "eval_f1": 0.7500510516642842,
      "eval_loss": 0.01973508670926094,
      "eval_precision": 0.7661660408844388,
      "eval_recall": 0.7346,
      "eval_runtime": 8.5095,
      "eval_samples_per_second": 329.042,
      "eval_steps_per_second": 41.13,
      "step": 1764
    },
    {
      "epoch": 4.5354509359047075,
      "grad_norm": 0.06497496366500854,
      "learning_rate": 1.0934240362811793e-05,
      "loss": 0.019792444229125978,
      "step": 2000
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9771954973113389,
      "eval_f1": 0.7891332470892627,
      "eval_loss": 0.03587024286389351,
      "eval_precision": 0.7853040205981382,
      "eval_recall": 0.793,
      "eval_runtime": 8.4468,
      "eval_samples_per_second": 331.488,
      "eval_steps_per_second": 41.436,
      "step": 2205
    },
    {
      "epoch": 5.669313669880885,
      "grad_norm": 0.02606065385043621,
      "learning_rate": 8.666666666666668e-06,
      "loss": 0.013434083938598632,
      "step": 2500
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9781268873381892,
      "eval_f1": 0.8013205282112845,
      "eval_loss": 0.03571802005171776,
      "eval_precision": 0.8016413130504404,
      "eval_recall": 0.801,
      "eval_runtime": 8.4407,
      "eval_samples_per_second": 331.725,
      "eval_steps_per_second": 41.466,
      "step": 2646
    },
    {
      "epoch": 6.803176403857062,
      "grad_norm": 0.06468810141086578,
      "learning_rate": 6.399092970521542e-06,
      "loss": 0.009369298934936524,
      "step": 3000
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9796622255855751,
      "eval_f1": 0.8202601497832086,
      "eval_loss": 0.06079654023051262,
      "eval_precision": 0.8084693084693084,
      "eval_recall": 0.8324,
      "eval_runtime": 8.4309,
      "eval_samples_per_second": 332.113,
      "eval_steps_per_second": 41.514,
      "step": 3087
    },
    {
      "epoch": 7.9370391378332386,
      "grad_norm": 0.018788253888487816,
      "learning_rate": 4.131519274376418e-06,
      "loss": 0.005781527042388916,
      "step": 3500
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9798732436385333,
      "eval_f1": 0.8244905994684517,
      "eval_loss": 0.06755472719669342,
      "eval_precision": 0.8117852296956775,
      "eval_recall": 0.8376,
      "eval_runtime": 8.4796,
      "eval_samples_per_second": 330.204,
      "eval_steps_per_second": 41.276,
      "step": 3528
    }
  ],
  "logging_steps": 500,
  "max_steps": 4410,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.254969894362148e+16,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}