trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.9230769230769234,
  "eval_steps": 500,
  "global_step": 57,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.05,
      "learning_rate": 8.333333333333334e-06,
      "loss": 1.5498,
      "step": 1
    },
    {
      "epoch": 0.62,
      "learning_rate": 4.411764705882353e-05,
      "loss": 1.3111,
      "step": 12
    },
    {
      "epoch": 0.97,
      "eval_Macro F1": 0.38463266827131015,
      "eval_Macro Precision": 0.3513860547447244,
      "eval_Macro Recall": 0.45842986022199095,
      "eval_Micro F1": 0.46153846153846156,
      "eval_Micro Precision": 0.46153846153846156,
      "eval_Micro Recall": 0.46153846153846156,
      "eval_Weighted F1": 0.39257303711149655,
      "eval_Weighted Precision": 0.36331970047517986,
      "eval_Weighted Recall": 0.46153846153846156,
      "eval_accuracy": 0.46153846153846156,
      "eval_loss": 1.004745602607727,
      "eval_runtime": 492.4481,
      "eval_samples_per_second": 2.587,
      "eval_steps_per_second": 0.081,
      "step": 19
    },
    {
      "epoch": 1.23,
      "learning_rate": 3.235294117647059e-05,
      "loss": 1.0359,
      "step": 24
    },
    {
      "epoch": 1.85,
      "learning_rate": 2.058823529411765e-05,
      "loss": 0.923,
      "step": 36
    },
    {
      "epoch": 2.0,
      "eval_Macro F1": 0.5093855651518693,
      "eval_Macro Precision": 0.7237298314233377,
      "eval_Macro Recall": 0.5863700553639922,
      "eval_Micro F1": 0.6004709576138147,
      "eval_Micro Precision": 0.6004709576138147,
      "eval_Micro Recall": 0.6004709576138147,
      "eval_Weighted F1": 0.5303808904978752,
      "eval_Weighted Precision": 0.7224619896227847,
      "eval_Weighted Recall": 0.6004709576138147,
      "eval_accuracy": 0.6004709576138147,
      "eval_loss": 0.7626671195030212,
      "eval_runtime": 31.9216,
      "eval_samples_per_second": 39.91,
      "eval_steps_per_second": 1.253,
      "step": 39
    },
    {
      "epoch": 2.46,
      "learning_rate": 8.823529411764707e-06,
      "loss": 0.7743,
      "step": 48
    },
    {
      "epoch": 2.92,
      "eval_Macro F1": 0.5605146596385748,
      "eval_Macro Precision": 0.7334668145101633,
      "eval_Macro Recall": 0.6217819801854411,
      "eval_Micro F1": 0.640502354788069,
      "eval_Micro Precision": 0.640502354788069,
      "eval_Micro Recall": 0.640502354788069,
      "eval_Weighted F1": 0.5794525732393071,
      "eval_Weighted Precision": 0.727397633667412,
      "eval_Weighted Recall": 0.640502354788069,
      "eval_accuracy": 0.640502354788069,
      "eval_loss": 0.6403084993362427,
      "eval_runtime": 32.0563,
      "eval_samples_per_second": 39.743,
      "eval_steps_per_second": 1.248,
      "step": 57
    },
    {
      "epoch": 2.92,
      "step": 57,
      "total_flos": 5.6158236080942285e+17,
      "train_loss": 0.9780613832306444,
      "train_runtime": 1760.8586,
      "train_samples_per_second": 4.21,
      "train_steps_per_second": 0.032
    }
  ],
  "logging_steps": 12,
  "max_steps": 57,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 5.6158236080942285e+17,
  "trial_name": null,
  "trial_params": null
}