last-checkpoint/trainer_state.json

{
  "best_metric": 70.1955074875208,
  "best_model_checkpoint": "./whisper-it/checkpoint-300",
  "epoch": 0.7633587786259542,
  "eval_steps": 100,
  "global_step": 400,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.05,
      "learning_rate": 4.4e-07,
      "loss": 3.4673,
      "step": 25
    },
    {
      "epoch": 0.1,
      "learning_rate": 9.400000000000001e-07,
      "loss": 2.7417,
      "step": 50
    },
    {
      "epoch": 0.14,
      "learning_rate": 1.44e-06,
      "loss": 2.0307,
      "step": 75
    },
    {
      "epoch": 0.19,
      "learning_rate": 1.94e-06,
      "loss": 1.2496,
      "step": 100
    },
    {
      "epoch": 0.19,
      "eval_loss": 1.233077049255371,
      "eval_runtime": 1969.1907,
      "eval_samples_per_second": 0.762,
      "eval_steps_per_second": 0.095,
      "eval_wer": 176.42124237382143,
      "step": 100
    },
    {
      "epoch": 0.24,
      "learning_rate": 2.4000000000000003e-06,
      "loss": 0.9794,
      "step": 125
    },
    {
      "epoch": 0.29,
      "learning_rate": 2.9e-06,
      "loss": 0.8642,
      "step": 150
    },
    {
      "epoch": 0.33,
      "learning_rate": 3.4000000000000005e-06,
      "loss": 0.8044,
      "step": 175
    },
    {
      "epoch": 0.38,
      "learning_rate": 3.88e-06,
      "loss": 0.7389,
      "step": 200
    },
    {
      "epoch": 0.38,
      "eval_loss": 0.8330782055854797,
      "eval_runtime": 1755.11,
      "eval_samples_per_second": 0.855,
      "eval_steps_per_second": 0.107,
      "eval_wer": 80.49084858569051,
      "step": 200
    },
    {
      "epoch": 0.43,
      "learning_rate": 4.38e-06,
      "loss": 0.6293,
      "step": 225
    },
    {
      "epoch": 0.48,
      "learning_rate": 4.880000000000001e-06,
      "loss": 0.5066,
      "step": 250
    },
    {
      "epoch": 0.52,
      "learning_rate": 5.380000000000001e-06,
      "loss": 0.3526,
      "step": 275
    },
    {
      "epoch": 0.57,
      "learning_rate": 5.8800000000000005e-06,
      "loss": 0.2951,
      "step": 300
    },
    {
      "epoch": 0.57,
      "eval_loss": 0.4260523319244385,
      "eval_runtime": 1708.1953,
      "eval_samples_per_second": 0.878,
      "eval_steps_per_second": 0.11,
      "eval_wer": 70.1955074875208,
      "step": 300
    },
    {
      "epoch": 0.62,
      "learning_rate": 6.380000000000001e-06,
      "loss": 0.2553,
      "step": 325
    },
    {
      "epoch": 0.67,
      "learning_rate": 6.88e-06,
      "loss": 0.2744,
      "step": 350
    },
    {
      "epoch": 0.72,
      "learning_rate": 7.3800000000000005e-06,
      "loss": 0.279,
      "step": 375
    },
    {
      "epoch": 0.76,
      "learning_rate": 7.88e-06,
      "loss": 0.2703,
      "step": 400
    },
    {
      "epoch": 0.76,
      "eval_loss": 0.40512609481811523,
      "eval_runtime": 1849.1681,
      "eval_samples_per_second": 0.811,
      "eval_steps_per_second": 0.102,
      "eval_wer": 101.62922906267333,
      "step": 400
    }
  ],
  "logging_steps": 25,
  "max_steps": 4000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 8,
  "save_steps": 100,
  "total_flos": 3.693893124096e+18,
  "trial_name": null,
  "trial_params": null
}