trainer_state.json

{
  "best_metric": 0.029342563822865486,
  "best_model_checkpoint": "roberta-large-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": 1.1222,
      "step": 1
    },
    {
      "epoch": 1.0,
      "learning_rate": 1.6000000000000003e-05,
      "loss": 0.5253,
      "step": 153
    },
    {
      "epoch": 1.0,
      "eval_Macro F1": 0.9632836299502966,
      "eval_Macro Precision": 0.9633333333333334,
      "eval_Macro Recall": 0.9662454462659381,
      "eval_Micro F1": 0.9641693811074918,
      "eval_Micro Precision": 0.9641693811074918,
      "eval_Micro Recall": 0.9641693811074918,
      "eval_Weighted F1": 0.9647479825674883,
      "eval_Weighted Precision": 0.9681107491856679,
      "eval_Weighted Recall": 0.9641693811074918,
      "eval_accuracy": 0.9641693811074918,
      "eval_loss": 0.12700782716274261,
      "eval_runtime": 1.6798,
      "eval_samples_per_second": 182.759,
      "eval_steps_per_second": 11.906,
      "step": 153
    },
    {
      "epoch": 2.0,
      "learning_rate": 1.2e-05,
      "loss": 0.0921,
      "step": 306
    },
    {
      "epoch": 2.0,
      "eval_Macro F1": 0.980327393886716,
      "eval_Macro Precision": 0.9817575757575758,
      "eval_Macro Recall": 0.9790873600622173,
      "eval_Micro F1": 0.9804560260586319,
      "eval_Micro Precision": 0.9804560260586319,
      "eval_Micro Recall": 0.9804560260586319,
      "eval_Weighted F1": 0.9804746376641527,
      "eval_Weighted Precision": 0.9806946994373705,
      "eval_Weighted Recall": 0.9804560260586319,
      "eval_accuracy": 0.9804560260586319,
      "eval_loss": 0.08777838945388794,
      "eval_runtime": 1.6847,
      "eval_samples_per_second": 182.225,
      "eval_steps_per_second": 11.871,
      "step": 306
    },
    {
      "epoch": 3.0,
      "learning_rate": 8.000000000000001e-06,
      "loss": 0.0413,
      "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": 0.05901196971535683,
      "eval_runtime": 1.7013,
      "eval_samples_per_second": 180.445,
      "eval_steps_per_second": 11.755,
      "step": 459
    },
    {
      "epoch": 4.0,
      "learning_rate": 4.000000000000001e-06,
      "loss": 0.0261,
      "step": 612
    },
    {
      "epoch": 4.0,
      "eval_Macro F1": 0.989644746787604,
      "eval_Macro Precision": 0.9899639655737217,
      "eval_Macro Recall": 0.9895833333333334,
      "eval_Micro F1": 0.990228013029316,
      "eval_Micro Precision": 0.990228013029316,
      "eval_Micro Recall": 0.990228013029316,
      "eval_Weighted F1": 0.9901932977316877,
      "eval_Weighted Precision": 0.9903976746751036,
      "eval_Weighted Recall": 0.990228013029316,
      "eval_accuracy": 0.990228013029316,
      "eval_loss": 0.05230216309428215,
      "eval_runtime": 1.6848,
      "eval_samples_per_second": 182.218,
      "eval_steps_per_second": 11.871,
      "step": 612
    },
    {
      "epoch": 5.0,
      "learning_rate": 0.0,
      "loss": 0.012,
      "step": 765
    },
    {
      "epoch": 5.0,
      "eval_Macro F1": 0.9831697750089234,
      "eval_Macro Precision": 0.9845093632958801,
      "eval_Macro Recall": 0.9820926966292135,
      "eval_Micro F1": 0.9837133550488599,
      "eval_Micro Precision": 0.9837133550488599,
      "eval_Micro Recall": 0.9837133550488599,
      "eval_Weighted F1": 0.983695143647141,
      "eval_Weighted Precision": 0.9839478827361564,
      "eval_Weighted Recall": 0.9837133550488599,
      "eval_accuracy": 0.9837133550488599,
      "eval_loss": 0.029342563822865486,
      "eval_runtime": 1.6892,
      "eval_samples_per_second": 181.741,
      "eval_steps_per_second": 11.84,
      "step": 765
    },
    {
      "epoch": 5.0,
      "step": 765,
      "total_flos": 2539812775083840.0,
      "train_loss": 0.14014302266189477,
      "train_runtime": 281.4865,
      "train_samples_per_second": 43.483,
      "train_steps_per_second": 2.718
    }
  ],
  "logging_steps": 500,
  "max_steps": 765,
  "num_train_epochs": 5,
  "save_steps": 500,
  "total_flos": 2539812775083840.0,
  "trial_name": null,
  "trial_params": null
}