{
  "best_metric": 0.922854387656702,
  "best_model_checkpoint": "/content/dissertation/scripts/ner/output/checkpoint-2395",
  "epoch": 10.0,
  "eval_steps": 500,
  "global_step": 4790,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.998338094001558,
      "eval_f1": 0.8886836027713626,
      "eval_loss": 0.005215654149651527,
      "eval_precision": 0.8498233215547704,
      "eval_recall": 0.9312681510164569,
      "eval_runtime": 14.0241,
      "eval_samples_per_second": 484.736,
      "eval_steps_per_second": 60.61,
      "step": 479
    },
    {
      "epoch": 1.0438413361169103,
      "grad_norm": 0.063787080347538,
      "learning_rate": 4.478079331941545e-05,
      "loss": 0.0127,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9983965203843158,
      "eval_f1": 0.90715667311412,
      "eval_loss": 0.0056168013252317905,
      "eval_precision": 0.9062801932367149,
      "eval_recall": 0.9080348499515973,
      "eval_runtime": 13.9845,
      "eval_samples_per_second": 486.11,
      "eval_steps_per_second": 60.782,
      "step": 958
    },
    {
      "epoch": 2.0876826722338206,
      "grad_norm": 0.30002298951148987,
      "learning_rate": 3.95615866388309e-05,
      "loss": 0.0035,
      "step": 1000
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9985393404310569,
      "eval_f1": 0.9184466019417475,
      "eval_loss": 0.0046853721141815186,
      "eval_precision": 0.9211295034079844,
      "eval_recall": 0.9157792836398838,
      "eval_runtime": 14.0593,
      "eval_samples_per_second": 483.522,
      "eval_steps_per_second": 60.458,
      "step": 1437
    },
    {
      "epoch": 3.1315240083507305,
      "grad_norm": 0.03608839586377144,
      "learning_rate": 3.434237995824635e-05,
      "loss": 0.002,
      "step": 1500
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9984030122046221,
      "eval_f1": 0.9054054054054055,
      "eval_loss": 0.006538054905831814,
      "eval_precision": 0.9027911453320501,
      "eval_recall": 0.9080348499515973,
      "eval_runtime": 14.21,
      "eval_samples_per_second": 478.395,
      "eval_steps_per_second": 59.817,
      "step": 1916
    },
    {
      "epoch": 4.175365344467641,
      "grad_norm": 0.002257826505228877,
      "learning_rate": 2.9123173277661797e-05,
      "loss": 0.0014,
      "step": 2000
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9985847831732018,
      "eval_f1": 0.922854387656702,
      "eval_loss": 0.006093208212405443,
      "eval_precision": 0.9193083573487032,
      "eval_recall": 0.9264278799612778,
      "eval_runtime": 13.9969,
      "eval_samples_per_second": 485.68,
      "eval_steps_per_second": 60.728,
      "step": 2395
    },
    {
      "epoch": 5.219206680584551,
      "grad_norm": 0.0006600466440431774,
      "learning_rate": 2.3903966597077245e-05,
      "loss": 0.0007,
      "step": 2500
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9984289794858479,
      "eval_f1": 0.9072978303747534,
      "eval_loss": 0.006890583783388138,
      "eval_precision": 0.9246231155778895,
      "eval_recall": 0.8906098741529526,
      "eval_runtime": 14.017,
      "eval_samples_per_second": 484.982,
      "eval_steps_per_second": 60.641,
      "step": 2874
    },
    {
      "epoch": 6.263048016701461,
      "grad_norm": 0.06641975790262222,
      "learning_rate": 1.8684759916492694e-05,
      "loss": 0.0004,
      "step": 3000
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9985003895092184,
      "eval_f1": 0.910133843212237,
      "eval_loss": 0.007136390544474125,
      "eval_precision": 0.898961284230406,
      "eval_recall": 0.9215876089060987,
      "eval_runtime": 14.0087,
      "eval_samples_per_second": 485.271,
      "eval_steps_per_second": 60.677,
      "step": 3353
    },
    {
      "epoch": 7.306889352818372,
      "grad_norm": 0.0022020984906703234,
      "learning_rate": 1.3465553235908144e-05,
      "loss": 0.0003,
      "step": 3500
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9986172422747338,
      "eval_f1": 0.9218225419664269,
      "eval_loss": 0.007608836982399225,
      "eval_precision": 0.9134980988593155,
      "eval_recall": 0.9303000968054211,
      "eval_runtime": 13.9459,
      "eval_samples_per_second": 487.454,
      "eval_steps_per_second": 60.95,
      "step": 3832
    },
    {
      "epoch": 8.350730688935283,
      "grad_norm": 0.0032535246573388577,
      "learning_rate": 8.246346555323591e-06,
      "loss": 0.0001,
      "step": 4000
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.9985653077122826,
      "eval_f1": 0.9187109187109187,
      "eval_loss": 0.007967105135321617,
      "eval_precision": 0.9130019120458891,
      "eval_recall": 0.9244917715392061,
      "eval_runtime": 14.2735,
      "eval_samples_per_second": 476.269,
      "eval_steps_per_second": 59.551,
      "step": 4311
    },
    {
      "epoch": 9.394572025052192,
      "grad_norm": 0.00014514605572912842,
      "learning_rate": 3.02713987473904e-06,
      "loss": 0.0001,
      "step": 4500
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.9986107504544274,
      "eval_f1": 0.919463087248322,
      "eval_loss": 0.008040270768105984,
      "eval_precision": 0.9107312440645774,
      "eval_recall": 0.9283639883833494,
      "eval_runtime": 14.9407,
      "eval_samples_per_second": 455.0,
      "eval_steps_per_second": 56.892,
      "step": 4790
    },
    {
      "epoch": 10.0,
      "step": 4790,
      "total_flos": 1.7383912956387396e+16,
      "train_loss": 0.002211083533992847,
      "train_runtime": 1477.0977,
      "train_samples_per_second": 207.447,
      "train_steps_per_second": 3.243
    }
  ],
  "logging_steps": 500,
  "max_steps": 4790,
  "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": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.7383912956387396e+16,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}