trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.9419354838709677,
  "eval_steps": 500,
  "global_step": 114,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.03,
      "learning_rate": 4.166666666666667e-06,
      "loss": 0.1438,
      "step": 1
    },
    {
      "epoch": 0.31,
      "learning_rate": 5e-05,
      "loss": 0.146,
      "step": 12
    },
    {
      "epoch": 0.62,
      "learning_rate": 4.411764705882353e-05,
      "loss": 0.1822,
      "step": 24
    },
    {
      "epoch": 0.93,
      "learning_rate": 3.8235294117647055e-05,
      "loss": 0.1916,
      "step": 36
    },
    {
      "epoch": 0.98,
      "eval_Macro F1": 0.9408451121862865,
      "eval_Macro Precision": 0.9411664777049882,
      "eval_Macro Recall": 0.9427118741264285,
      "eval_Micro F1": 0.9461054287962234,
      "eval_Micro Precision": 0.9461054287962234,
      "eval_Micro Recall": 0.9461054287962234,
      "eval_Weighted F1": 0.9464960091472664,
      "eval_Weighted Precision": 0.9487006349374144,
      "eval_Weighted Recall": 0.9461054287962234,
      "eval_accuracy": 0.9461054287962234,
      "eval_loss": 0.13961628079414368,
      "eval_runtime": 63.2832,
      "eval_samples_per_second": 40.169,
      "eval_steps_per_second": 1.264,
      "step": 38
    },
    {
      "epoch": 1.24,
      "learning_rate": 3.235294117647059e-05,
      "loss": 0.1565,
      "step": 48
    },
    {
      "epoch": 1.55,
      "learning_rate": 2.647058823529412e-05,
      "loss": 0.1692,
      "step": 60
    },
    {
      "epoch": 1.86,
      "learning_rate": 2.058823529411765e-05,
      "loss": 0.1597,
      "step": 72
    },
    {
      "epoch": 1.99,
      "eval_Macro F1": 0.9484522652918805,
      "eval_Macro Precision": 0.9471910825399321,
      "eval_Macro Recall": 0.95148334257645,
      "eval_Micro F1": 0.9520062942564909,
      "eval_Micro Precision": 0.952006294256491,
      "eval_Micro Recall": 0.952006294256491,
      "eval_Weighted F1": 0.9522977862410118,
      "eval_Weighted Precision": 0.9540997275393984,
      "eval_Weighted Recall": 0.952006294256491,
      "eval_accuracy": 0.952006294256491,
      "eval_loss": 0.12265925854444504,
      "eval_runtime": 64.9371,
      "eval_samples_per_second": 39.146,
      "eval_steps_per_second": 1.232,
      "step": 77
    },
    {
      "epoch": 2.17,
      "learning_rate": 1.4705882352941177e-05,
      "loss": 0.1619,
      "step": 84
    },
    {
      "epoch": 2.48,
      "learning_rate": 8.823529411764707e-06,
      "loss": 0.1611,
      "step": 96
    },
    {
      "epoch": 2.79,
      "learning_rate": 2.9411764705882355e-06,
      "loss": 0.1722,
      "step": 108
    },
    {
      "epoch": 2.94,
      "eval_Macro F1": 0.947301058449996,
      "eval_Macro Precision": 0.9458287584858323,
      "eval_Macro Recall": 0.9498167521420487,
      "eval_Micro F1": 0.9512195121951219,
      "eval_Micro Precision": 0.9512195121951219,
      "eval_Micro Recall": 0.9512195121951219,
      "eval_Weighted F1": 0.9515018097434871,
      "eval_Weighted Precision": 0.9527033488356239,
      "eval_Weighted Recall": 0.9512195121951219,
      "eval_accuracy": 0.9512195121951219,
      "eval_loss": 0.11822786182165146,
      "eval_runtime": 63.9617,
      "eval_samples_per_second": 39.743,
      "eval_steps_per_second": 1.251,
      "step": 114
    },
    {
      "epoch": 2.94,
      "step": 114,
      "total_flos": 1.1265748270391624e+18,
      "train_loss": 0.16734615334293299,
      "train_runtime": 848.5959,
      "train_samples_per_second": 17.436,
      "train_steps_per_second": 0.134
    }
  ],
  "logging_steps": 12,
  "max_steps": 114,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 1.1265748270391624e+18,
  "trial_name": null,
  "trial_params": null
}