checkpoint-500/trainer_state.json

{
  "best_metric": 54.54987091580379,
  "best_model_checkpoint": "./whisper-small-Persian/checkpoint-500",
  "epoch": 1.5567651632970452,
  "global_step": 500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.08,
      "learning_rate": 4.4e-08,
      "loss": 1.4232,
      "step": 25
    },
    {
      "epoch": 0.16,
      "learning_rate": 9.4e-08,
      "loss": 1.4027,
      "step": 50
    },
    {
      "epoch": 0.23,
      "learning_rate": 1.44e-07,
      "loss": 1.3731,
      "step": 75
    },
    {
      "epoch": 0.31,
      "learning_rate": 1.94e-07,
      "loss": 1.2969,
      "step": 100
    },
    {
      "epoch": 0.39,
      "learning_rate": 2.4399999999999996e-07,
      "loss": 1.1964,
      "step": 125
    },
    {
      "epoch": 0.47,
      "learning_rate": 2.9399999999999996e-07,
      "loss": 1.0995,
      "step": 150
    },
    {
      "epoch": 0.54,
      "learning_rate": 3.4399999999999996e-07,
      "loss": 1.0068,
      "step": 175
    },
    {
      "epoch": 0.62,
      "learning_rate": 3.94e-07,
      "loss": 0.8949,
      "step": 200
    },
    {
      "epoch": 0.7,
      "learning_rate": 4.44e-07,
      "loss": 0.8521,
      "step": 225
    },
    {
      "epoch": 0.78,
      "learning_rate": 4.94e-07,
      "loss": 0.7694,
      "step": 250
    },
    {
      "epoch": 0.86,
      "learning_rate": 5.44e-07,
      "loss": 0.7196,
      "step": 275
    },
    {
      "epoch": 0.93,
      "learning_rate": 5.939999999999999e-07,
      "loss": 0.6902,
      "step": 300
    },
    {
      "epoch": 1.01,
      "learning_rate": 6.44e-07,
      "loss": 0.6748,
      "step": 325
    },
    {
      "epoch": 1.09,
      "learning_rate": 6.939999999999999e-07,
      "loss": 0.6247,
      "step": 350
    },
    {
      "epoch": 1.17,
      "learning_rate": 7.44e-07,
      "loss": 0.63,
      "step": 375
    },
    {
      "epoch": 1.25,
      "learning_rate": 7.94e-07,
      "loss": 0.5899,
      "step": 400
    },
    {
      "epoch": 1.32,
      "learning_rate": 8.439999999999999e-07,
      "loss": 0.5628,
      "step": 425
    },
    {
      "epoch": 1.4,
      "learning_rate": 8.939999999999999e-07,
      "loss": 0.5634,
      "step": 450
    },
    {
      "epoch": 1.48,
      "learning_rate": 9.439999999999999e-07,
      "loss": 0.5549,
      "step": 475
    },
    {
      "epoch": 1.56,
      "learning_rate": 9.94e-07,
      "loss": 0.5533,
      "step": 500
    },
    {
      "epoch": 1.56,
      "eval_loss": 0.7044046521186829,
      "eval_runtime": 1265.1903,
      "eval_samples_per_second": 8.132,
      "eval_steps_per_second": 0.127,
      "eval_wer": 54.54987091580379,
      "step": 500
    }
  ],
  "max_steps": 5000,
  "num_train_epochs": 16,
  "total_flos": 9.243967607629939e+18,
  "trial_name": null,
  "trial_params": null
}