checkpoint-200/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.09204924634679554,
  "global_step": 200,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "learning_rate": 1.0305343511450381e-05,
      "loss": 13.0539,
      "step": 20
    },
    {
      "epoch": 0.02,
      "learning_rate": 2.118320610687023e-05,
      "loss": 20.9649,
      "step": 40
    },
    {
      "epoch": 0.03,
      "learning_rate": 3.206106870229007e-05,
      "loss": 21.1001,
      "step": 60
    },
    {
      "epoch": 0.04,
      "learning_rate": 4.3511450381679383e-05,
      "loss": 5.5852,
      "step": 80
    },
    {
      "epoch": 0.05,
      "learning_rate": 5.438931297709923e-05,
      "loss": 5.2002,
      "step": 100
    },
    {
      "epoch": 0.05,
      "eval_loss": 3.5950469970703125,
      "eval_runtime": 194.8685,
      "eval_samples_per_second": 24.853,
      "eval_steps_per_second": 0.78,
      "eval_wer": 1.0,
      "step": 100
    },
    {
      "epoch": 0.06,
      "learning_rate": 6.583969465648854e-05,
      "loss": 3.281,
      "step": 120
    },
    {
      "epoch": 0.06,
      "learning_rate": 7.495301487862176e-05,
      "loss": 3.3442,
      "step": 140
    },
    {
      "epoch": 0.07,
      "learning_rate": 7.472983555207517e-05,
      "loss": 3.5229,
      "step": 160
    },
    {
      "epoch": 0.08,
      "learning_rate": 7.449490994518402e-05,
      "loss": 3.089,
      "step": 180
    },
    {
      "epoch": 0.09,
      "learning_rate": 7.427173061863743e-05,
      "loss": 3.482,
      "step": 200
    },
    {
      "epoch": 0.09,
      "eval_loss": 3.274627923965454,
      "eval_runtime": 194.5649,
      "eval_samples_per_second": 24.891,
      "eval_steps_per_second": 0.781,
      "eval_wer": 1.0,
      "step": 200
    }
  ],
  "max_steps": 6516,
  "num_train_epochs": 3,
  "total_flos": 3.810397715849134e+18,
  "trial_name": null,
  "trial_params": null
}