trainer_state.json

{
  "best_metric": 0.07267534732818604,
  "best_model_checkpoint": "bert-large-uncased-hate-offensive-normal-speech-lr-1e-05/checkpoint-612",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 765,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "learning_rate": 9.986928104575165e-06,
      "loss": 1.3111,
      "step": 1
    },
    {
      "epoch": 1.0,
      "learning_rate": 8.000000000000001e-06,
      "loss": 0.9139,
      "step": 153
    },
    {
      "epoch": 1.0,
      "eval_Macro F1": 0.7946891596376133,
      "eval_Macro Precision": 0.8030025975570529,
      "eval_Macro Recall": 0.793248276332462,
      "eval_Micro F1": 0.8110749185667753,
      "eval_Micro Precision": 0.8110749185667753,
      "eval_Micro Recall": 0.8110749185667753,
      "eval_Weighted F1": 0.808597679674613,
      "eval_Weighted Precision": 0.812490327736885,
      "eval_Weighted Recall": 0.8110749185667753,
      "eval_accuracy": 0.8110749185667753,
      "eval_loss": 0.485382616519928,
      "eval_runtime": 1.6985,
      "eval_samples_per_second": 180.753,
      "eval_steps_per_second": 11.775,
      "step": 153
    },
    {
      "epoch": 2.0,
      "learning_rate": 6e-06,
      "loss": 0.2278,
      "step": 306
    },
    {
      "epoch": 2.0,
      "eval_Macro F1": 0.965223571946261,
      "eval_Macro Precision": 0.9654715673031662,
      "eval_Macro Recall": 0.9657485621293646,
      "eval_Micro F1": 0.9674267100977199,
      "eval_Micro Precision": 0.9674267100977199,
      "eval_Micro Recall": 0.9674267100977199,
      "eval_Weighted F1": 0.9675123332857436,
      "eval_Weighted Precision": 0.9682947849236886,
      "eval_Weighted Recall": 0.9674267100977199,
      "eval_accuracy": 0.9674267100977199,
      "eval_loss": 0.0775093287229538,
      "eval_runtime": 1.6928,
      "eval_samples_per_second": 181.358,
      "eval_steps_per_second": 11.815,
      "step": 306
    },
    {
      "epoch": 3.0,
      "learning_rate": 4.000000000000001e-06,
      "loss": 0.0582,
      "step": 459
    },
    {
      "epoch": 3.0,
      "eval_Macro F1": 0.9756573721562932,
      "eval_Macro Precision": 0.9768818096344228,
      "eval_Macro Recall": 0.9753907251940297,
      "eval_Micro F1": 0.9771986970684039,
      "eval_Micro Precision": 0.9771986970684039,
      "eval_Micro Recall": 0.9771986970684039,
      "eval_Weighted F1": 0.9772671060837873,
      "eval_Weighted Precision": 0.9781958385960247,
      "eval_Weighted Recall": 0.9771986970684039,
      "eval_accuracy": 0.9771986970684039,
      "eval_loss": 0.10066915303468704,
      "eval_runtime": 1.6955,
      "eval_samples_per_second": 181.07,
      "eval_steps_per_second": 11.796,
      "step": 459
    },
    {
      "epoch": 4.0,
      "learning_rate": 2.0000000000000003e-06,
      "loss": 0.0228,
      "step": 612
    },
    {
      "epoch": 4.0,
      "eval_Macro F1": 0.9756041514089026,
      "eval_Macro Precision": 0.9756981683292011,
      "eval_Macro Recall": 0.975548273738675,
      "eval_Micro F1": 0.9771986970684039,
      "eval_Micro Precision": 0.9771986970684039,
      "eval_Micro Recall": 0.9771986970684039,
      "eval_Weighted F1": 0.977214672387114,
      "eval_Weighted Precision": 0.9772649217239973,
      "eval_Weighted Recall": 0.9771986970684039,
      "eval_accuracy": 0.9771986970684039,
      "eval_loss": 0.07267534732818604,
      "eval_runtime": 1.6931,
      "eval_samples_per_second": 181.322,
      "eval_steps_per_second": 11.812,
      "step": 612
    },
    {
      "epoch": 5.0,
      "learning_rate": 0.0,
      "loss": 0.0159,
      "step": 765
    },
    {
      "epoch": 5.0,
      "eval_Macro F1": 0.9719944646254973,
      "eval_Macro Precision": 0.9722554765464863,
      "eval_Macro Recall": 0.9718852700756714,
      "eval_Micro F1": 0.9739413680781759,
      "eval_Micro Precision": 0.9739413680781759,
      "eval_Micro Recall": 0.9739413680781759,
      "eval_Weighted F1": 0.9739714001894333,
      "eval_Weighted Precision": 0.9741378480054397,
      "eval_Weighted Recall": 0.9739413680781759,
      "eval_accuracy": 0.9739413680781759,
      "eval_loss": 0.10134366899728775,
      "eval_runtime": 1.7036,
      "eval_samples_per_second": 180.21,
      "eval_steps_per_second": 11.74,
      "step": 765
    },
    {
      "epoch": 5.0,
      "step": 765,
      "total_flos": 2539812775083840.0,
      "train_loss": 0.24821887405869228,
      "train_runtime": 280.5122,
      "train_samples_per_second": 43.634,
      "train_steps_per_second": 2.727
    }
  ],
  "logging_steps": 500,
  "max_steps": 765,
  "num_train_epochs": 5,
  "save_steps": 500,
  "total_flos": 2539812775083840.0,
  "trial_name": null,
  "trial_params": null
}