trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 12.903225806451612,
  "global_step": 1600,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 3.23,
      "learning_rate": 0.00025788819875776394,
      "loss": 0.2987,
      "step": 400
    },
    {
      "epoch": 3.23,
      "eval_loss": 0.367538183927536,
      "eval_runtime": 226.1299,
      "eval_samples_per_second": 8.155,
      "eval_steps_per_second": 1.022,
      "eval_wer": 0.4119895580831624,
      "step": 400
    },
    {
      "epoch": 6.45,
      "learning_rate": 0.00022062111801242234,
      "loss": 0.1718,
      "step": 800
    },
    {
      "epoch": 6.45,
      "eval_loss": 0.3576325476169586,
      "eval_runtime": 227.1637,
      "eval_samples_per_second": 8.117,
      "eval_steps_per_second": 1.017,
      "eval_wer": 0.3210889427559202,
      "step": 800
    },
    {
      "epoch": 9.68,
      "learning_rate": 0.00018335403726708072,
      "loss": 0.1294,
      "step": 1200
    },
    {
      "epoch": 9.68,
      "eval_loss": 0.36709052324295044,
      "eval_runtime": 227.6934,
      "eval_samples_per_second": 8.099,
      "eval_steps_per_second": 1.015,
      "eval_wer": 0.32090247995524895,
      "step": 1200
    },
    {
      "epoch": 12.9,
      "learning_rate": 0.00014608695652173912,
      "loss": 0.1032,
      "step": 1600
    },
    {
      "epoch": 12.9,
      "eval_loss": 0.35343116521835327,
      "eval_runtime": 227.5075,
      "eval_samples_per_second": 8.105,
      "eval_steps_per_second": 1.015,
      "eval_wer": 0.2946112250606004,
      "step": 1600
    }
  ],
  "max_steps": 3720,
  "num_train_epochs": 30,
  "total_flos": 6.469831852034365e+18,
  "trial_name": null,
  "trial_params": null
}