trainer_state.json

{
  "best_metric": 40.88698140200286,
  "best_model_checkpoint": "./DICT_LUKE_ASR/checkpoint-500",
  "epoch": 0.9487666034155597,
  "global_step": 500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.05,
      "learning_rate": 8.000000000000001e-07,
      "loss": 3.0133,
      "step": 25
    },
    {
      "epoch": 0.09,
      "learning_rate": 1.8000000000000001e-06,
      "loss": 1.4076,
      "step": 50
    },
    {
      "epoch": 0.14,
      "learning_rate": 2.8000000000000003e-06,
      "loss": 0.6474,
      "step": 75
    },
    {
      "epoch": 0.19,
      "learning_rate": 3.8000000000000005e-06,
      "loss": 0.1948,
      "step": 100
    },
    {
      "epoch": 0.24,
      "learning_rate": 4.800000000000001e-06,
      "loss": 0.1533,
      "step": 125
    },
    {
      "epoch": 0.28,
      "learning_rate": 5.8e-06,
      "loss": 0.1456,
      "step": 150
    },
    {
      "epoch": 0.33,
      "learning_rate": 6.760000000000001e-06,
      "loss": 0.1275,
      "step": 175
    },
    {
      "epoch": 0.38,
      "learning_rate": 7.76e-06,
      "loss": 0.1275,
      "step": 200
    },
    {
      "epoch": 0.43,
      "learning_rate": 8.76e-06,
      "loss": 0.1157,
      "step": 225
    },
    {
      "epoch": 0.47,
      "learning_rate": 9.760000000000001e-06,
      "loss": 0.1344,
      "step": 250
    },
    {
      "epoch": 0.52,
      "learning_rate": 1.0760000000000002e-05,
      "loss": 0.1237,
      "step": 275
    },
    {
      "epoch": 0.57,
      "learning_rate": 1.1760000000000001e-05,
      "loss": 0.1505,
      "step": 300
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.2760000000000001e-05,
      "loss": 0.1157,
      "step": 325
    },
    {
      "epoch": 0.66,
      "learning_rate": 1.376e-05,
      "loss": 0.1372,
      "step": 350
    },
    {
      "epoch": 0.71,
      "learning_rate": 1.4760000000000001e-05,
      "loss": 0.1399,
      "step": 375
    },
    {
      "epoch": 0.76,
      "learning_rate": 1.576e-05,
      "loss": 0.1282,
      "step": 400
    },
    {
      "epoch": 0.81,
      "learning_rate": 1.6760000000000002e-05,
      "loss": 0.1313,
      "step": 425
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.7760000000000003e-05,
      "loss": 0.1289,
      "step": 450
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.876e-05,
      "loss": 0.1077,
      "step": 475
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.976e-05,
      "loss": 0.1025,
      "step": 500
    },
    {
      "epoch": 0.95,
      "eval_loss": 0.10049215704202652,
      "eval_runtime": 1187.251,
      "eval_samples_per_second": 1.578,
      "eval_steps_per_second": 0.099,
      "eval_wer": 40.88698140200286,
      "step": 500
    }
  ],
  "max_steps": 5000,
  "num_train_epochs": 10,
  "total_flos": 3.39707363328e+19,
  "trial_name": null,
  "trial_params": null
}