{
  "best_global_step": 2752,
  "best_metric": 0.0057959225960075855,
  "best_model_checkpoint": "projects/PetBERT_annonymisation/data/case_sensitive/model/checkpoint-2752",
  "epoch": 7.0,
  "eval_steps": 500,
  "global_step": 4816,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0,
      "eval_f1": 0.011880094095301812,
      "eval_loss": 2.4431910514831543,
      "eval_precision": 0.08875808758622007,
      "eval_recall": 0.05952071418656867,
      "eval_runtime": 23.7929,
      "eval_samples_per_second": 139.285,
      "eval_steps_per_second": 4.371,
      "step": 0
    },
    {
      "epoch": 0.7267441860465116,
      "grad_norm": 0.14246754348278046,
      "learning_rate": 4.996373546511628e-05,
      "loss": 0.0494,
      "step": 500
    },
    {
      "epoch": 1.0,
      "eval_f1": 0.8935989317265615,
      "eval_loss": 0.006424472201615572,
      "eval_precision": 0.8856397165335348,
      "eval_recall": 0.9052291580842897,
      "eval_runtime": 24.8439,
      "eval_samples_per_second": 133.393,
      "eval_steps_per_second": 4.186,
      "step": 688
    },
    {
      "epoch": 1.4534883720930232,
      "grad_norm": 0.013319989666342735,
      "learning_rate": 4.992739825581396e-05,
      "loss": 0.0044,
      "step": 1000
    },
    {
      "epoch": 2.0,
      "eval_f1": 0.921251638253682,
      "eval_loss": 0.006238908972591162,
      "eval_precision": 0.935008628082375,
      "eval_recall": 0.9144731612976222,
      "eval_runtime": 24.802,
      "eval_samples_per_second": 133.618,
      "eval_steps_per_second": 4.193,
      "step": 1376
    },
    {
      "epoch": 2.1802325581395348,
      "grad_norm": 0.04109887406229973,
      "learning_rate": 4.989106104651163e-05,
      "loss": 0.0032,
      "step": 1500
    },
    {
      "epoch": 2.9069767441860463,
      "grad_norm": 0.3754558861255646,
      "learning_rate": 4.985472383720931e-05,
      "loss": 0.0023,
      "step": 2000
    },
    {
      "epoch": 3.0,
      "eval_f1": 0.931311286064836,
      "eval_loss": 0.006664152257144451,
      "eval_precision": 0.9347111088313973,
      "eval_recall": 0.9310447645995317,
      "eval_runtime": 24.7968,
      "eval_samples_per_second": 133.646,
      "eval_steps_per_second": 4.194,
      "step": 2064
    },
    {
      "epoch": 3.633720930232558,
      "grad_norm": 0.002286644419655204,
      "learning_rate": 4.981838662790698e-05,
      "loss": 0.002,
      "step": 2500
    },
    {
      "epoch": 4.0,
      "eval_f1": 0.9188838249757393,
      "eval_loss": 0.0057959225960075855,
      "eval_precision": 0.9199713934746218,
      "eval_recall": 0.9194240810494807,
      "eval_runtime": 24.791,
      "eval_samples_per_second": 133.677,
      "eval_steps_per_second": 4.195,
      "step": 2752
    },
    {
      "epoch": 4.3604651162790695,
      "grad_norm": 0.0007801814354024827,
      "learning_rate": 4.978204941860465e-05,
      "loss": 0.0016,
      "step": 3000
    },
    {
      "epoch": 5.0,
      "eval_f1": 0.8780986671708252,
      "eval_loss": 0.008505718782544136,
      "eval_precision": 0.868721341594642,
      "eval_recall": 0.9097564277642886,
      "eval_runtime": 24.8121,
      "eval_samples_per_second": 133.564,
      "eval_steps_per_second": 4.192,
      "step": 3440
    },
    {
      "epoch": 5.087209302325581,
      "grad_norm": 0.09036080539226532,
      "learning_rate": 4.974571220930232e-05,
      "loss": 0.0011,
      "step": 3500
    },
    {
      "epoch": 5.813953488372093,
      "grad_norm": 0.08693202584981918,
      "learning_rate": 4.9709375e-05,
      "loss": 0.0012,
      "step": 4000
    },
    {
      "epoch": 6.0,
      "eval_f1": 0.9238025924793921,
      "eval_loss": 0.008721762336790562,
      "eval_precision": 0.9375153926584012,
      "eval_recall": 0.917450432310493,
      "eval_runtime": 24.7949,
      "eval_samples_per_second": 133.656,
      "eval_steps_per_second": 4.194,
      "step": 4128
    },
    {
      "epoch": 6.540697674418604,
      "grad_norm": 0.03196759149432182,
      "learning_rate": 4.967303779069767e-05,
      "loss": 0.0013,
      "step": 4500
    },
    {
      "epoch": 7.0,
      "eval_f1": 0.9309400750123076,
      "eval_loss": 0.009504728950560093,
      "eval_precision": 0.9228721238354496,
      "eval_recall": 0.9401722615087472,
      "eval_runtime": 24.8542,
      "eval_samples_per_second": 133.338,
      "eval_steps_per_second": 4.184,
      "step": 4816
    }
  ],
  "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": 4.023931632713933e+16,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}