{
  "best_metric": 0.9323447636700648,
  "best_model_checkpoint": "/content/dissertation/scripts/ner/output/checkpoint-4333",
  "epoch": 9.989615784008308,
  "eval_steps": 500,
  "global_step": 4810,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.9989615784008308,
      "eval_accuracy": 0.9987441743644127,
      "eval_f1": 0.9291628334866606,
      "eval_loss": 0.004182814620435238,
      "eval_precision": 0.9173478655767484,
      "eval_recall": 0.9412861136999068,
      "eval_runtime": 15.1684,
      "eval_samples_per_second": 457.925,
      "eval_steps_per_second": 57.29,
      "step": 481
    },
    {
      "epoch": 1.0384215991692627,
      "grad_norm": 0.07515838742256165,
      "learning_rate": 4.48024948024948e-05,
      "loss": 0.0156,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9986464990372004,
      "eval_f1": 0.9189439555349699,
      "eval_loss": 0.004935940261930227,
      "eval_precision": 0.9134438305709024,
      "eval_recall": 0.9245107176141659,
      "eval_runtime": 15.18,
      "eval_samples_per_second": 457.575,
      "eval_steps_per_second": 57.246,
      "step": 963
    },
    {
      "epoch": 2.0768431983385254,
      "grad_norm": 0.005299085285514593,
      "learning_rate": 3.9604989604989604e-05,
      "loss": 0.0039,
      "step": 1000
    },
    {
      "epoch": 2.9989615784008308,
      "eval_accuracy": 0.9985627773281612,
      "eval_f1": 0.9191873589164785,
      "eval_loss": 0.005292736925184727,
      "eval_precision": 0.8914185639229422,
      "eval_recall": 0.9487418452935694,
      "eval_runtime": 15.1155,
      "eval_samples_per_second": 459.529,
      "eval_steps_per_second": 57.491,
      "step": 1444
    },
    {
      "epoch": 3.115264797507788,
      "grad_norm": 0.06427361071109772,
      "learning_rate": 3.4407484407484405e-05,
      "loss": 0.0024,
      "step": 1500
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9984720788100354,
      "eval_f1": 0.9167412712623096,
      "eval_loss": 0.0061088865622878075,
      "eval_precision": 0.8819982773471146,
      "eval_recall": 0.9543336439888164,
      "eval_runtime": 15.4198,
      "eval_samples_per_second": 450.46,
      "eval_steps_per_second": 56.356,
      "step": 1926
    },
    {
      "epoch": 4.153686396677051,
      "grad_norm": 0.04428843781352043,
      "learning_rate": 2.920997920997921e-05,
      "loss": 0.0017,
      "step": 2000
    },
    {
      "epoch": 4.998961578400831,
      "eval_accuracy": 0.9986046381826807,
      "eval_f1": 0.9254284390921722,
      "eval_loss": 0.007444376591593027,
      "eval_precision": 0.919889502762431,
      "eval_recall": 0.9310344827586207,
      "eval_runtime": 15.1205,
      "eval_samples_per_second": 459.377,
      "eval_steps_per_second": 57.472,
      "step": 2407
    },
    {
      "epoch": 5.192107995846314,
      "grad_norm": 0.004208261147141457,
      "learning_rate": 2.4012474012474013e-05,
      "loss": 0.0011,
      "step": 2500
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9985767309463344,
      "eval_f1": 0.9266943291839557,
      "eval_loss": 0.007942954078316689,
      "eval_precision": 0.916970802919708,
      "eval_recall": 0.9366262814538676,
      "eval_runtime": 15.1887,
      "eval_samples_per_second": 457.314,
      "eval_steps_per_second": 57.214,
      "step": 2889
    },
    {
      "epoch": 6.230529595015576,
      "grad_norm": 0.004876282997429371,
      "learning_rate": 1.8814968814968818e-05,
      "loss": 0.0007,
      "step": 3000
    },
    {
      "epoch": 6.998961578400831,
      "eval_accuracy": 0.9986744062735468,
      "eval_f1": 0.9254004576659038,
      "eval_loss": 0.006659395061433315,
      "eval_precision": 0.9091726618705036,
      "eval_recall": 0.9422180801491147,
      "eval_runtime": 15.1599,
      "eval_samples_per_second": 458.182,
      "eval_steps_per_second": 57.322,
      "step": 3370
    },
    {
      "epoch": 7.268951194184839,
      "grad_norm": 0.0032651671208441257,
      "learning_rate": 1.3617463617463619e-05,
      "loss": 0.0005,
      "step": 3500
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9986674294644602,
      "eval_f1": 0.9275092936802974,
      "eval_loss": 0.007326104678213596,
      "eval_precision": 0.9249304911955515,
      "eval_recall": 0.9301025163094129,
      "eval_runtime": 15.1332,
      "eval_samples_per_second": 458.991,
      "eval_steps_per_second": 57.423,
      "step": 3852
    },
    {
      "epoch": 8.307372793354102,
      "grad_norm": 0.0013715826207771897,
      "learning_rate": 8.419958419958421e-06,
      "loss": 0.0004,
      "step": 4000
    },
    {
      "epoch": 8.99896157840083,
      "eval_accuracy": 0.9987162671280663,
      "eval_f1": 0.9323447636700648,
      "eval_loss": 0.008018395863473415,
      "eval_precision": 0.9271889400921659,
      "eval_recall": 0.9375582479030755,
      "eval_runtime": 15.4151,
      "eval_samples_per_second": 450.598,
      "eval_steps_per_second": 56.373,
      "step": 4333
    },
    {
      "epoch": 9.345794392523365,
      "grad_norm": 0.005300257820636034,
      "learning_rate": 3.2224532224532228e-06,
      "loss": 0.0002,
      "step": 4500
    },
    {
      "epoch": 9.989615784008308,
      "eval_accuracy": 0.9987092903189797,
      "eval_f1": 0.9315448658649399,
      "eval_loss": 0.007857992313802242,
      "eval_precision": 0.9247015610651974,
      "eval_recall": 0.9384902143522833,
      "eval_runtime": 15.1843,
      "eval_samples_per_second": 457.447,
      "eval_steps_per_second": 57.23,
      "step": 4810
    },
    {
      "epoch": 9.989615784008308,
      "step": 4810,
      "total_flos": 1.3132946802550608e+16,
      "train_loss": 0.0027612092573652642,
      "train_runtime": 2084.9871,
      "train_samples_per_second": 147.78,
      "train_steps_per_second": 2.307
    }
  ],
  "logging_steps": 500,
  "max_steps": 4810,
  "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": 1.3132946802550608e+16,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}