trainer_state.json

{
  "best_metric": 0.9321585507595793,
  "best_model_checkpoint": "tmp_ner_fantastic-bale-10_45/run-33/checkpoint-875",
  "epoch": 7.0,
  "eval_steps": 500,
  "global_step": 875,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9730935342517575,
      "eval_loss": 0.08587262034416199,
      "eval_macro_f1": 0.9170047043200686,
      "eval_macro_precision": 0.9278276425309375,
      "eval_macro_recall": 0.90752738549263,
      "eval_runtime": 5.7832,
      "eval_samples_per_second": 172.914,
      "eval_steps_per_second": 21.614,
      "step": 125
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9750480754074587,
      "eval_loss": 0.08141306042671204,
      "eval_macro_f1": 0.9264657139223695,
      "eval_macro_precision": 0.9157927892781152,
      "eval_macro_recall": 0.9377952332788448,
      "eval_runtime": 6.0324,
      "eval_samples_per_second": 165.773,
      "eval_steps_per_second": 20.722,
      "step": 250
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9743781721887709,
      "eval_loss": 0.08608711510896683,
      "eval_macro_f1": 0.9247086477797719,
      "eval_macro_precision": 0.9173611315562689,
      "eval_macro_recall": 0.933305032585042,
      "eval_runtime": 6.0132,
      "eval_samples_per_second": 166.3,
      "eval_steps_per_second": 20.787,
      "step": 375
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.6703500151634216,
      "learning_rate": 5.4624258845588244e-05,
      "loss": 0.1282,
      "step": 500
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9770420226348476,
      "eval_loss": 0.09856158494949341,
      "eval_macro_f1": 0.9314150814858593,
      "eval_macro_precision": 0.9211465798788551,
      "eval_macro_recall": 0.9421883284262093,
      "eval_runtime": 5.932,
      "eval_samples_per_second": 168.577,
      "eval_steps_per_second": 21.072,
      "step": 500
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9750086693357712,
      "eval_loss": 0.11021321266889572,
      "eval_macro_f1": 0.9267488028708639,
      "eval_macro_precision": 0.9153942226603791,
      "eval_macro_recall": 0.9386080517042859,
      "eval_runtime": 5.9689,
      "eval_samples_per_second": 167.536,
      "eval_steps_per_second": 20.942,
      "step": 625
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9760253459853094,
      "eval_loss": 0.11098870635032654,
      "eval_macro_f1": 0.9287954097228024,
      "eval_macro_precision": 0.9205523043756338,
      "eval_macro_recall": 0.9374208211553338,
      "eval_runtime": 5.9208,
      "eval_samples_per_second": 168.896,
      "eval_steps_per_second": 21.112,
      "step": 750
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9773021027079852,
      "eval_loss": 0.11490638554096222,
      "eval_macro_f1": 0.9321585507595793,
      "eval_macro_precision": 0.9266098427277691,
      "eval_macro_recall": 0.9378807443139501,
      "eval_runtime": 6.075,
      "eval_samples_per_second": 164.61,
      "eval_steps_per_second": 20.576,
      "step": 875
    }
  ],
  "logging_steps": 500,
  "max_steps": 4000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 32,
  "save_steps": 500,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 3,
        "early_stopping_threshold": 0.001
      },
      "attributes": {
        "early_stopping_patience_counter": 3
      }
    },
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 723885139142880.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": {
    "learning_rate": 6.150691546013236e-05,
    "per_device_train_batch_size": 8,
    "warmup_ratio": 0.014555287490025246,
    "weight_decay": 0.02919246813677108
  }
}