{
  "best_metric": 35.60262364321316,
  "best_model_checkpoint": "./checkpoint-1000",
  "epoch": 3.4305317324185247,
  "eval_steps": 500,
  "global_step": 1000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.09,
      "learning_rate": 5e-07,
      "loss": 0.25,
      "step": 25
    },
    {
      "epoch": 0.17,
      "learning_rate": 1e-06,
      "loss": 0.2435,
      "step": 50
    },
    {
      "epoch": 0.26,
      "learning_rate": 9.736842105263158e-07,
      "loss": 0.2513,
      "step": 75
    },
    {
      "epoch": 0.34,
      "learning_rate": 9.473684210526315e-07,
      "loss": 0.2497,
      "step": 100
    },
    {
      "epoch": 0.43,
      "learning_rate": 9.210526315789473e-07,
      "loss": 0.2489,
      "step": 125
    },
    {
      "epoch": 0.51,
      "learning_rate": 8.947368421052631e-07,
      "loss": 0.2486,
      "step": 150
    },
    {
      "epoch": 0.6,
      "learning_rate": 8.684210526315789e-07,
      "loss": 0.2449,
      "step": 175
    },
    {
      "epoch": 0.69,
      "learning_rate": 8.421052631578947e-07,
      "loss": 0.2439,
      "step": 200
    },
    {
      "epoch": 0.77,
      "learning_rate": 8.157894736842105e-07,
      "loss": 0.2467,
      "step": 225
    },
    {
      "epoch": 0.86,
      "learning_rate": 7.894736842105263e-07,
      "loss": 0.2441,
      "step": 250
    },
    {
      "epoch": 0.94,
      "learning_rate": 7.631578947368421e-07,
      "loss": 0.2415,
      "step": 275
    },
    {
      "epoch": 1.03,
      "learning_rate": 7.368421052631578e-07,
      "loss": 0.2432,
      "step": 300
    },
    {
      "epoch": 1.11,
      "learning_rate": 7.105263157894736e-07,
      "loss": 0.2397,
      "step": 325
    },
    {
      "epoch": 1.2,
      "learning_rate": 6.842105263157895e-07,
      "loss": 0.2384,
      "step": 350
    },
    {
      "epoch": 1.29,
      "learning_rate": 6.578947368421053e-07,
      "loss": 0.2367,
      "step": 375
    },
    {
      "epoch": 1.37,
      "learning_rate": 6.31578947368421e-07,
      "loss": 0.2365,
      "step": 400
    },
    {
      "epoch": 1.46,
      "learning_rate": 6.052631578947368e-07,
      "loss": 0.2389,
      "step": 425
    },
    {
      "epoch": 1.54,
      "learning_rate": 5.789473684210526e-07,
      "loss": 0.2394,
      "step": 450
    },
    {
      "epoch": 1.63,
      "learning_rate": 5.526315789473684e-07,
      "loss": 0.2389,
      "step": 475
    },
    {
      "epoch": 1.72,
      "learning_rate": 5.263157894736842e-07,
      "loss": 0.2423,
      "step": 500
    },
    {
      "epoch": 1.72,
      "eval_loss": 0.2709830701351166,
      "eval_runtime": 1895.8676,
      "eval_samples_per_second": 4.913,
      "eval_steps_per_second": 0.154,
      "eval_wer": 35.95697654649092,
      "step": 500
    },
    {
      "epoch": 1.8,
      "learning_rate": 5e-07,
      "loss": 0.2376,
      "step": 525
    },
    {
      "epoch": 1.89,
      "learning_rate": 4.7368421052631574e-07,
      "loss": 0.2327,
      "step": 550
    },
    {
      "epoch": 1.97,
      "learning_rate": 4.4736842105263156e-07,
      "loss": 0.2357,
      "step": 575
    },
    {
      "epoch": 2.06,
      "learning_rate": 4.2105263157894733e-07,
      "loss": 0.234,
      "step": 600
    },
    {
      "epoch": 2.14,
      "learning_rate": 3.9473684210526315e-07,
      "loss": 0.2314,
      "step": 625
    },
    {
      "epoch": 2.23,
      "learning_rate": 3.684210526315789e-07,
      "loss": 0.2357,
      "step": 650
    },
    {
      "epoch": 2.32,
      "learning_rate": 3.4210526315789473e-07,
      "loss": 0.2358,
      "step": 675
    },
    {
      "epoch": 2.4,
      "learning_rate": 3.157894736842105e-07,
      "loss": 0.2383,
      "step": 700
    },
    {
      "epoch": 2.49,
      "learning_rate": 2.894736842105263e-07,
      "loss": 0.2293,
      "step": 725
    },
    {
      "epoch": 2.57,
      "learning_rate": 2.631578947368421e-07,
      "loss": 0.2321,
      "step": 750
    },
    {
      "epoch": 2.66,
      "learning_rate": 2.3684210526315787e-07,
      "loss": 0.2302,
      "step": 775
    },
    {
      "epoch": 2.74,
      "learning_rate": 2.1052631578947366e-07,
      "loss": 0.2254,
      "step": 800
    },
    {
      "epoch": 2.83,
      "learning_rate": 1.8421052631578946e-07,
      "loss": 0.234,
      "step": 825
    },
    {
      "epoch": 2.92,
      "learning_rate": 1.5789473684210525e-07,
      "loss": 0.2358,
      "step": 850
    },
    {
      "epoch": 3.0,
      "learning_rate": 1.3157894736842104e-07,
      "loss": 0.2321,
      "step": 875
    },
    {
      "epoch": 3.09,
      "learning_rate": 1.0526315789473683e-07,
      "loss": 0.2375,
      "step": 900
    },
    {
      "epoch": 3.17,
      "learning_rate": 7.894736842105262e-08,
      "loss": 0.2258,
      "step": 925
    },
    {
      "epoch": 3.26,
      "learning_rate": 5.2631578947368416e-08,
      "loss": 0.2254,
      "step": 950
    },
    {
      "epoch": 3.34,
      "learning_rate": 2.6315789473684208e-08,
      "loss": 0.234,
      "step": 975
    },
    {
      "epoch": 3.43,
      "learning_rate": 0.0,
      "loss": 0.2329,
      "step": 1000
    },
    {
      "epoch": 3.43,
      "eval_loss": 0.267061322927475,
      "eval_runtime": 1899.4594,
      "eval_samples_per_second": 4.904,
      "eval_steps_per_second": 0.154,
      "eval_wer": 35.60262364321316,
      "step": 1000
    },
    {
      "epoch": 3.43,
      "step": 1000,
      "total_flos": 4.14559151456256e+18,
      "train_loss": 0.23807335948944092,
      "train_runtime": 7036.0339,
      "train_samples_per_second": 9.096,
      "train_steps_per_second": 0.142
    }
  ],
  "logging_steps": 25,
  "max_steps": 1000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 500,
  "total_flos": 4.14559151456256e+18,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}