trainer_state.json

{
  "best_metric": 0.01635025069117546,
  "best_model_checkpoint": "bert-base-uncased-hate-offensive-normal-speech-lr-2e-05/checkpoint-765",
  "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": 1.997385620915033e-05,
      "loss": 1.1996,
      "step": 1
    },
    {
      "epoch": 1.0,
      "learning_rate": 1.6000000000000003e-05,
      "loss": 0.6115,
      "step": 153
    },
    {
      "epoch": 1.0,
      "eval_Macro F1": 0.9729341469616068,
      "eval_Macro Precision": 0.9718924972004479,
      "eval_Macro Recall": 0.9749090532326395,
      "eval_Micro F1": 0.9739413680781759,
      "eval_Micro Precision": 0.9739413680781759,
      "eval_Micro Recall": 0.9739413680781759,
      "eval_Weighted F1": 0.974189084096731,
      "eval_Weighted Precision": 0.97531214549646,
      "eval_Weighted Recall": 0.9739413680781759,
      "eval_accuracy": 0.9739413680781759,
      "eval_loss": 0.08827492594718933,
      "eval_runtime": 0.6114,
      "eval_samples_per_second": 502.164,
      "eval_steps_per_second": 32.714,
      "step": 153
    },
    {
      "epoch": 2.0,
      "learning_rate": 1.2e-05,
      "loss": 0.0759,
      "step": 306
    },
    {
      "epoch": 2.0,
      "eval_Macro F1": 0.993179353485733,
      "eval_Macro Precision": 0.9926739926739927,
      "eval_Macro Recall": 0.9937955373406194,
      "eval_Micro F1": 0.993485342019544,
      "eval_Micro Precision": 0.993485342019544,
      "eval_Micro Recall": 0.993485342019544,
      "eval_Weighted F1": 0.9935063052016132,
      "eval_Weighted Precision": 0.9936285213158177,
      "eval_Weighted Recall": 0.993485342019544,
      "eval_accuracy": 0.993485342019544,
      "eval_loss": 0.029480930417776108,
      "eval_runtime": 0.5986,
      "eval_samples_per_second": 512.887,
      "eval_steps_per_second": 33.413,
      "step": 306
    },
    {
      "epoch": 3.0,
      "learning_rate": 8.000000000000001e-06,
      "loss": 0.0279,
      "step": 459
    },
    {
      "epoch": 3.0,
      "eval_Macro F1": 0.9860112452705044,
      "eval_Macro Precision": 0.985663082437276,
      "eval_Macro Recall": 0.9868510928961749,
      "eval_Micro F1": 0.9869706840390879,
      "eval_Micro Precision": 0.9869706840390879,
      "eval_Micro Recall": 0.9869706840390879,
      "eval_Weighted F1": 0.9870296055601882,
      "eval_Weighted Precision": 0.9875310847255788,
      "eval_Weighted Recall": 0.9869706840390879,
      "eval_accuracy": 0.9869706840390879,
      "eval_loss": 0.027695246040821075,
      "eval_runtime": 0.6054,
      "eval_samples_per_second": 507.13,
      "eval_steps_per_second": 33.038,
      "step": 459
    },
    {
      "epoch": 4.0,
      "learning_rate": 4.000000000000001e-06,
      "loss": 0.0207,
      "step": 612
    },
    {
      "epoch": 4.0,
      "eval_Macro F1": 0.9932684864595648,
      "eval_Macro Precision": 0.9935862691960252,
      "eval_Macro Recall": 0.9930555555555555,
      "eval_Micro F1": 0.993485342019544,
      "eval_Micro Precision": 0.993485342019544,
      "eval_Micro Recall": 0.993485342019544,
      "eval_Weighted F1": 0.9934668006404771,
      "eval_Weighted Precision": 0.9935480169133939,
      "eval_Weighted Recall": 0.993485342019544,
      "eval_accuracy": 0.993485342019544,
      "eval_loss": 0.026556001976132393,
      "eval_runtime": 0.6018,
      "eval_samples_per_second": 510.162,
      "eval_steps_per_second": 33.235,
      "step": 612
    },
    {
      "epoch": 5.0,
      "learning_rate": 0.0,
      "loss": 0.0099,
      "step": 765
    },
    {
      "epoch": 5.0,
      "eval_Macro F1": 0.9932684864595648,
      "eval_Macro Precision": 0.9935862691960252,
      "eval_Macro Recall": 0.9930555555555555,
      "eval_Micro F1": 0.993485342019544,
      "eval_Micro Precision": 0.993485342019544,
      "eval_Micro Recall": 0.993485342019544,
      "eval_Weighted F1": 0.9934668006404771,
      "eval_Weighted Precision": 0.9935480169133939,
      "eval_Weighted Recall": 0.993485342019544,
      "eval_accuracy": 0.993485342019544,
      "eval_loss": 0.01635025069117546,
      "eval_runtime": 0.5984,
      "eval_samples_per_second": 513.075,
      "eval_steps_per_second": 33.425,
      "step": 765
    },
    {
      "epoch": 5.0,
      "step": 765,
      "total_flos": 717066286251840.0,
      "train_loss": 0.14996177059373045,
      "train_runtime": 83.8708,
      "train_samples_per_second": 145.939,
      "train_steps_per_second": 9.121
    }
  ],
  "logging_steps": 500,
  "max_steps": 765,
  "num_train_epochs": 5,
  "save_steps": 500,
  "total_flos": 717066286251840.0,
  "trial_name": null,
  "trial_params": null
}