trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 294.0004,
  "global_step": 5000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 29.0,
      "learning_rate": 1e-05,
      "loss": 0.0638,
      "step": 500
    },
    {
      "epoch": 58.0,
      "learning_rate": 1e-05,
      "loss": 0.0004,
      "step": 1000
    },
    {
      "epoch": 58.0,
      "eval_loss": 0.11287793517112732,
      "eval_runtime": 14.5402,
      "eval_samples_per_second": 1.238,
      "eval_steps_per_second": 0.206,
      "eval_wer": 1.8292682926829267,
      "step": 1000
    },
    {
      "epoch": 88.0,
      "learning_rate": 1e-05,
      "loss": 0.0001,
      "step": 1500
    },
    {
      "epoch": 117.0,
      "learning_rate": 1e-05,
      "loss": 0.0,
      "step": 2000
    },
    {
      "epoch": 117.0,
      "eval_loss": 0.12321429699659348,
      "eval_runtime": 12.3922,
      "eval_samples_per_second": 1.453,
      "eval_steps_per_second": 0.242,
      "eval_wer": 1.8292682926829267,
      "step": 2000
    },
    {
      "epoch": 147.0,
      "learning_rate": 1e-05,
      "loss": 0.0,
      "step": 2500
    },
    {
      "epoch": 176.0,
      "learning_rate": 1e-05,
      "loss": 0.0,
      "step": 3000
    },
    {
      "epoch": 176.0,
      "eval_loss": 0.13265825808048248,
      "eval_runtime": 14.0309,
      "eval_samples_per_second": 1.283,
      "eval_steps_per_second": 0.214,
      "eval_wer": 1.8292682926829267,
      "step": 3000
    },
    {
      "epoch": 205.0,
      "learning_rate": 1e-05,
      "loss": 0.0,
      "step": 3500
    },
    {
      "epoch": 235.0,
      "learning_rate": 1e-05,
      "loss": 0.0,
      "step": 4000
    },
    {
      "epoch": 235.0,
      "eval_loss": 0.14010873436927795,
      "eval_runtime": 12.2156,
      "eval_samples_per_second": 1.474,
      "eval_steps_per_second": 0.246,
      "eval_wer": 4.878048780487805,
      "step": 4000
    },
    {
      "epoch": 264.0,
      "learning_rate": 1e-05,
      "loss": 0.0,
      "step": 4500
    },
    {
      "epoch": 294.0,
      "learning_rate": 1e-05,
      "loss": 0.0,
      "step": 5000
    },
    {
      "epoch": 294.0,
      "eval_loss": 0.15023881196975708,
      "eval_runtime": 13.2184,
      "eval_samples_per_second": 1.362,
      "eval_steps_per_second": 0.227,
      "eval_wer": 4.878048780487805,
      "step": 5000
    },
    {
      "epoch": 294.0,
      "step": 5000,
      "total_flos": 2.30868320256e+19,
      "train_loss": 0.006433991515403613,
      "train_runtime": 23651.0479,
      "train_samples_per_second": 3.383,
      "train_steps_per_second": 0.211
    }
  ],
  "max_steps": 5000,
  "num_train_epochs": 9223372036854775807,
  "total_flos": 2.30868320256e+19,
  "trial_name": null,
  "trial_params": null
}