trainer_state.json

{
  "best_metric": 11.271395917499468,
  "best_model_checkpoint": "whisper-it-small/checkpoint-932",
  "epoch": 2.0,
  "global_step": 932,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.11,
      "learning_rate": 9.600000000000001e-07,
      "loss": 0.9816,
      "step": 50
    },
    {
      "epoch": 0.21,
      "learning_rate": 1.9600000000000003e-06,
      "loss": 0.519,
      "step": 100
    },
    {
      "epoch": 0.32,
      "learning_rate": 2.96e-06,
      "loss": 0.3418,
      "step": 150
    },
    {
      "epoch": 0.43,
      "learning_rate": 3.96e-06,
      "loss": 0.3032,
      "step": 200
    },
    {
      "epoch": 0.54,
      "learning_rate": 4.960000000000001e-06,
      "loss": 0.2901,
      "step": 250
    },
    {
      "epoch": 0.64,
      "learning_rate": 5.9600000000000005e-06,
      "loss": 0.2783,
      "step": 300
    },
    {
      "epoch": 0.75,
      "learning_rate": 6.96e-06,
      "loss": 0.2771,
      "step": 350
    },
    {
      "epoch": 0.86,
      "learning_rate": 7.960000000000002e-06,
      "loss": 0.2659,
      "step": 400
    },
    {
      "epoch": 0.97,
      "learning_rate": 8.96e-06,
      "loss": 0.2529,
      "step": 450
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.2758234441280365,
      "eval_runtime": 5945.4592,
      "eval_samples_per_second": 2.523,
      "eval_steps_per_second": 0.079,
      "eval_wer": 12.490697427174144,
      "step": 466
    },
    {
      "epoch": 1.07,
      "learning_rate": 9.960000000000001e-06,
      "loss": 0.2122,
      "step": 500
    },
    {
      "epoch": 1.18,
      "learning_rate": 8.888888888888888e-06,
      "loss": 0.1893,
      "step": 550
    },
    {
      "epoch": 1.29,
      "learning_rate": 7.731481481481483e-06,
      "loss": 0.1872,
      "step": 600
    },
    {
      "epoch": 1.39,
      "learning_rate": 6.574074074074075e-06,
      "loss": 0.1803,
      "step": 650
    },
    {
      "epoch": 1.5,
      "learning_rate": 5.416666666666667e-06,
      "loss": 0.1843,
      "step": 700
    },
    {
      "epoch": 1.61,
      "learning_rate": 4.2592592592592596e-06,
      "loss": 0.1806,
      "step": 750
    },
    {
      "epoch": 1.72,
      "learning_rate": 3.101851851851852e-06,
      "loss": 0.1787,
      "step": 800
    },
    {
      "epoch": 1.82,
      "learning_rate": 1.944444444444445e-06,
      "loss": 0.1761,
      "step": 850
    },
    {
      "epoch": 1.93,
      "learning_rate": 7.870370370370371e-07,
      "loss": 0.1711,
      "step": 900
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.25167033076286316,
      "eval_runtime": 5711.8942,
      "eval_samples_per_second": 2.627,
      "eval_steps_per_second": 0.082,
      "eval_wer": 11.271395917499468,
      "step": 932
    },
    {
      "epoch": 2.0,
      "step": 932,
      "total_flos": 1.718295190585344e+19,
      "train_loss": 0.2830442016216818,
      "train_runtime": 31129.4467,
      "train_samples_per_second": 1.913,
      "train_steps_per_second": 0.03
    }
  ],
  "max_steps": 932,
  "num_train_epochs": 2,
  "total_flos": 1.718295190585344e+19,
  "trial_name": null,
  "trial_params": null
}