trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 5.0,
  "global_step": 665,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "learning_rate": 1.9969924812030075e-05,
      "loss": 1.3801,
      "step": 1
    },
    {
      "epoch": 1.0,
      "learning_rate": 1.6000000000000003e-05,
      "loss": 0.7426,
      "step": 133
    },
    {
      "epoch": 1.0,
      "eval_Macro F1": 0.6689648126125525,
      "eval_Macro Precision": 0.8992027427922928,
      "eval_Macro Recall": 0.680893472825423,
      "eval_Micro F1": 0.880321665089877,
      "eval_Micro Precision": 0.880321665089877,
      "eval_Micro Recall": 0.880321665089877,
      "eval_Weighted F1": 0.8635636544503065,
      "eval_Weighted Precision": 0.8861584487449019,
      "eval_Weighted Recall": 0.880321665089877,
      "eval_accuracy": 0.880321665089877,
      "eval_loss": 0.3819724917411804,
      "eval_runtime": 89.2566,
      "eval_samples_per_second": 23.685,
      "eval_steps_per_second": 0.381,
      "step": 133
    },
    {
      "epoch": 2.0,
      "learning_rate": 1.2e-05,
      "loss": 0.332,
      "step": 266
    },
    {
      "epoch": 2.0,
      "eval_Macro F1": 0.852520274558422,
      "eval_Macro Precision": 0.8705013592781246,
      "eval_Macro Recall": 0.8402059008388016,
      "eval_Micro F1": 0.9006622516556292,
      "eval_Micro Precision": 0.9006622516556292,
      "eval_Micro Recall": 0.9006622516556292,
      "eval_Weighted F1": 0.8987197931272273,
      "eval_Weighted Precision": 0.9014826393048196,
      "eval_Weighted Recall": 0.9006622516556292,
      "eval_accuracy": 0.9006622516556292,
      "eval_loss": 0.3082645535469055,
      "eval_runtime": 89.7702,
      "eval_samples_per_second": 23.549,
      "eval_steps_per_second": 0.379,
      "step": 266
    },
    {
      "epoch": 3.0,
      "learning_rate": 8.000000000000001e-06,
      "loss": 0.2381,
      "step": 399
    },
    {
      "epoch": 3.0,
      "eval_Macro F1": 0.8685813310685866,
      "eval_Macro Precision": 0.8862440804192933,
      "eval_Macro Recall": 0.8539033677339591,
      "eval_Micro F1": 0.9105960264900662,
      "eval_Micro Precision": 0.9105960264900662,
      "eval_Micro Recall": 0.9105960264900662,
      "eval_Weighted F1": 0.9097386578247573,
      "eval_Weighted Precision": 0.9095946797682414,
      "eval_Weighted Recall": 0.9105960264900662,
      "eval_accuracy": 0.9105960264900662,
      "eval_loss": 0.2869592308998108,
      "eval_runtime": 92.5923,
      "eval_samples_per_second": 22.831,
      "eval_steps_per_second": 0.367,
      "step": 399
    },
    {
      "epoch": 4.0,
      "learning_rate": 4.000000000000001e-06,
      "loss": 0.1911,
      "step": 532
    },
    {
      "epoch": 4.0,
      "eval_Macro F1": 0.8843230817497421,
      "eval_Macro Precision": 0.8873071926241887,
      "eval_Macro Recall": 0.8819268015939119,
      "eval_Micro F1": 0.9162724692526018,
      "eval_Micro Precision": 0.9162724692526018,
      "eval_Micro Recall": 0.9162724692526018,
      "eval_Weighted F1": 0.9158446415019297,
      "eval_Weighted Precision": 0.9159184568651126,
      "eval_Weighted Recall": 0.9162724692526018,
      "eval_accuracy": 0.9162724692526018,
      "eval_loss": 0.27965518832206726,
      "eval_runtime": 94.0341,
      "eval_samples_per_second": 22.481,
      "eval_steps_per_second": 0.362,
      "step": 532
    },
    {
      "epoch": 5.0,
      "learning_rate": 0.0,
      "loss": 0.1584,
      "step": 665
    },
    {
      "epoch": 5.0,
      "eval_Macro F1": 0.8854074422801754,
      "eval_Macro Precision": 0.8853385079570577,
      "eval_Macro Recall": 0.8859207019693811,
      "eval_Micro F1": 0.9172185430463576,
      "eval_Micro Precision": 0.9172185430463576,
      "eval_Micro Recall": 0.9172185430463576,
      "eval_Weighted F1": 0.9169847452219524,
      "eval_Weighted Precision": 0.917103894576438,
      "eval_Weighted Recall": 0.9172185430463576,
      "eval_accuracy": 0.9172185430463576,
      "eval_loss": 0.2838347256183624,
      "eval_runtime": 90.9928,
      "eval_samples_per_second": 23.233,
      "eval_steps_per_second": 0.374,
      "step": 665
    },
    {
      "epoch": 5.0,
      "step": 665,
      "total_flos": 964831600689600.0,
      "train_loss": 0.3334039892469134,
      "train_runtime": 6660.313,
      "train_samples_per_second": 6.348,
      "train_steps_per_second": 0.1
    }
  ],
  "max_steps": 665,
  "num_train_epochs": 5,
  "total_flos": 964831600689600.0,
  "trial_name": null,
  "trial_params": null
}