trainer_state.json

{
  "best_global_step": 1376,
  "best_metric": 0.008029412478208542,
  "best_model_checkpoint": "projects/PetBERT_annonymisation/data/case_sensitive/model/checkpoint-1376",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 3440,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0,
      "eval_f1": 0.010973861448813302,
      "eval_loss": 2.4063162803649902,
      "eval_precision": 0.09108980806958722,
      "eval_recall": 0.10473207318294019,
      "eval_runtime": 23.7722,
      "eval_samples_per_second": 139.407,
      "eval_steps_per_second": 4.375,
      "step": 0
    },
    {
      "epoch": 0.7267441860465116,
      "grad_norm": 0.2666139304637909,
      "learning_rate": 4.996373546511628e-05,
      "loss": 0.0518,
      "step": 500
    },
    {
      "epoch": 1.0,
      "eval_f1": 0.9198586422350306,
      "eval_loss": 0.010010140016674995,
      "eval_precision": 0.9136306910527949,
      "eval_recall": 0.9281047546302535,
      "eval_runtime": 24.8489,
      "eval_samples_per_second": 133.366,
      "eval_steps_per_second": 4.185,
      "step": 688
    },
    {
      "epoch": 1.4534883720930232,
      "grad_norm": 0.11741874366998672,
      "learning_rate": 4.992739825581396e-05,
      "loss": 0.0036,
      "step": 1000
    },
    {
      "epoch": 2.0,
      "eval_f1": 0.9300254346144702,
      "eval_loss": 0.008029412478208542,
      "eval_precision": 0.9130114164496201,
      "eval_recall": 0.9489397415434179,
      "eval_runtime": 25.1412,
      "eval_samples_per_second": 131.815,
      "eval_steps_per_second": 4.137,
      "step": 1376
    },
    {
      "epoch": 2.1802325581395348,
      "grad_norm": 0.3377048373222351,
      "learning_rate": 4.989106104651163e-05,
      "loss": 0.0026,
      "step": 1500
    },
    {
      "epoch": 2.9069767441860463,
      "grad_norm": 0.2929118573665619,
      "learning_rate": 4.985472383720931e-05,
      "loss": 0.0011,
      "step": 2000
    },
    {
      "epoch": 3.0,
      "eval_f1": 0.9402795283929368,
      "eval_loss": 0.01124291867017746,
      "eval_precision": 0.9422754585534239,
      "eval_recall": 0.9386206401984958,
      "eval_runtime": 24.8815,
      "eval_samples_per_second": 133.191,
      "eval_steps_per_second": 4.18,
      "step": 2064
    },
    {
      "epoch": 3.633720930232558,
      "grad_norm": 0.002858501160517335,
      "learning_rate": 4.981838662790698e-05,
      "loss": 0.0012,
      "step": 2500
    },
    {
      "epoch": 4.0,
      "eval_f1": 0.9321076846329125,
      "eval_loss": 0.011350538581609726,
      "eval_precision": 0.9585579423470411,
      "eval_recall": 0.9092514508781232,
      "eval_runtime": 24.8844,
      "eval_samples_per_second": 133.176,
      "eval_steps_per_second": 4.179,
      "step": 2752
    },
    {
      "epoch": 4.3604651162790695,
      "grad_norm": 0.3254820704460144,
      "learning_rate": 4.978204941860465e-05,
      "loss": 0.001,
      "step": 3000
    },
    {
      "epoch": 5.0,
      "eval_f1": 0.9375959937426187,
      "eval_loss": 0.012251886539161205,
      "eval_precision": 0.9347887957568951,
      "eval_recall": 0.9407576824824448,
      "eval_runtime": 24.848,
      "eval_samples_per_second": 133.371,
      "eval_steps_per_second": 4.185,
      "step": 3440
    }
  ],
  "logging_steps": 500,
  "max_steps": 688000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1000,
  "save_steps": 500,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 3,
        "early_stopping_threshold": 0.0
      },
      "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": 2.874236880509952e+16,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}