{
  "best_metric": 15.184536972434753,
  "best_model_checkpoint": "./checkpoint-100",
  "epoch": 1.9821605550049552,
  "eval_steps": 100,
  "global_step": 1000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.05,
      "learning_rate": 1.15e-07,
      "loss": 1.7239,
      "step": 25
    },
    {
      "epoch": 0.1,
      "learning_rate": 2.4e-07,
      "loss": 1.5685,
      "step": 50
    },
    {
      "epoch": 0.15,
      "learning_rate": 3.65e-07,
      "loss": 1.3334,
      "step": 75
    },
    {
      "epoch": 0.2,
      "learning_rate": 4.9e-07,
      "loss": 0.9295,
      "step": 100
    },
    {
      "epoch": 0.2,
      "eval_loss": 0.8013624548912048,
      "eval_runtime": 4914.4036,
      "eval_samples_per_second": 3.284,
      "eval_steps_per_second": 0.103,
      "eval_wer": 15.184536972434753,
      "step": 100
    },
    {
      "epoch": 0.25,
      "learning_rate": 4.872222222222222e-07,
      "loss": 0.5131,
      "step": 125
    },
    {
      "epoch": 0.3,
      "learning_rate": 4.733333333333333e-07,
      "loss": 0.349,
      "step": 150
    },
    {
      "epoch": 0.35,
      "learning_rate": 4.594444444444444e-07,
      "loss": 0.3085,
      "step": 175
    },
    {
      "epoch": 0.4,
      "learning_rate": 4.455555555555555e-07,
      "loss": 0.2976,
      "step": 200
    },
    {
      "epoch": 0.4,
      "eval_loss": 0.4207054078578949,
      "eval_runtime": 4946.2,
      "eval_samples_per_second": 3.263,
      "eval_steps_per_second": 0.102,
      "eval_wer": 16.028877878031764,
      "step": 200
    },
    {
      "epoch": 0.45,
      "learning_rate": 4.3166666666666663e-07,
      "loss": 0.2986,
      "step": 225
    },
    {
      "epoch": 0.5,
      "learning_rate": 4.177777777777778e-07,
      "loss": 0.28,
      "step": 250
    },
    {
      "epoch": 0.55,
      "learning_rate": 4.038888888888889e-07,
      "loss": 0.2705,
      "step": 275
    },
    {
      "epoch": 0.59,
      "learning_rate": 3.8999999999999997e-07,
      "loss": 0.2699,
      "step": 300
    },
    {
      "epoch": 0.59,
      "eval_loss": 0.3999289870262146,
      "eval_runtime": 5024.7316,
      "eval_samples_per_second": 3.212,
      "eval_steps_per_second": 0.101,
      "eval_wer": 15.826661778792142,
      "step": 300
    },
    {
      "epoch": 0.64,
      "learning_rate": 3.761111111111111e-07,
      "loss": 0.2783,
      "step": 325
    },
    {
      "epoch": 0.69,
      "learning_rate": 3.622222222222222e-07,
      "loss": 0.2716,
      "step": 350
    },
    {
      "epoch": 0.74,
      "learning_rate": 3.483333333333333e-07,
      "loss": 0.2705,
      "step": 375
    },
    {
      "epoch": 0.79,
      "learning_rate": 3.3444444444444443e-07,
      "loss": 0.2773,
      "step": 400
    },
    {
      "epoch": 0.79,
      "eval_loss": 0.390967458486557,
      "eval_runtime": 4987.4922,
      "eval_samples_per_second": 3.236,
      "eval_steps_per_second": 0.101,
      "eval_wer": 15.726736279460285,
      "step": 400
    },
    {
      "epoch": 0.84,
      "learning_rate": 3.2055555555555555e-07,
      "loss": 0.2518,
      "step": 425
    },
    {
      "epoch": 0.89,
      "learning_rate": 3.066666666666666e-07,
      "loss": 0.2697,
      "step": 450
    },
    {
      "epoch": 0.94,
      "learning_rate": 2.927777777777778e-07,
      "loss": 0.2623,
      "step": 475
    },
    {
      "epoch": 0.99,
      "learning_rate": 2.788888888888889e-07,
      "loss": 0.2631,
      "step": 500
    },
    {
      "epoch": 0.99,
      "eval_loss": 0.3862614035606384,
      "eval_runtime": 4949.1159,
      "eval_samples_per_second": 3.261,
      "eval_steps_per_second": 0.102,
      "eval_wer": 15.59724702292965,
      "step": 500
    },
    {
      "epoch": 1.04,
      "learning_rate": 2.65e-07,
      "loss": 0.2435,
      "step": 525
    },
    {
      "epoch": 1.09,
      "learning_rate": 2.511111111111111e-07,
      "loss": 0.2395,
      "step": 550
    },
    {
      "epoch": 1.14,
      "learning_rate": 2.372222222222222e-07,
      "loss": 0.2667,
      "step": 575
    },
    {
      "epoch": 1.19,
      "learning_rate": 2.2333333333333332e-07,
      "loss": 0.2487,
      "step": 600
    },
    {
      "epoch": 1.19,
      "eval_loss": 0.3833913505077362,
      "eval_runtime": 4993.2782,
      "eval_samples_per_second": 3.233,
      "eval_steps_per_second": 0.101,
      "eval_wer": 15.590742996345918,
      "step": 600
    },
    {
      "epoch": 1.24,
      "learning_rate": 2.0944444444444444e-07,
      "loss": 0.2445,
      "step": 625
    },
    {
      "epoch": 1.29,
      "learning_rate": 1.9555555555555555e-07,
      "loss": 0.2435,
      "step": 650
    },
    {
      "epoch": 1.34,
      "learning_rate": 1.8166666666666667e-07,
      "loss": 0.2442,
      "step": 675
    },
    {
      "epoch": 1.39,
      "learning_rate": 1.6777777777777778e-07,
      "loss": 0.2477,
      "step": 700
    },
    {
      "epoch": 1.39,
      "eval_loss": 0.38139107823371887,
      "eval_runtime": 5034.4769,
      "eval_samples_per_second": 3.206,
      "eval_steps_per_second": 0.1,
      "eval_wer": 15.61557655239289,
      "step": 700
    },
    {
      "epoch": 1.44,
      "learning_rate": 1.5388888888888887e-07,
      "loss": 0.2467,
      "step": 725
    },
    {
      "epoch": 1.49,
      "learning_rate": 1.4e-07,
      "loss": 0.2457,
      "step": 750
    },
    {
      "epoch": 1.54,
      "learning_rate": 1.2611111111111112e-07,
      "loss": 0.251,
      "step": 775
    },
    {
      "epoch": 1.59,
      "learning_rate": 1.1222222222222221e-07,
      "loss": 0.2428,
      "step": 800
    },
    {
      "epoch": 1.59,
      "eval_loss": 0.3801017701625824,
      "eval_runtime": 4971.5469,
      "eval_samples_per_second": 3.247,
      "eval_steps_per_second": 0.102,
      "eval_wer": 15.490226221870085,
      "step": 800
    },
    {
      "epoch": 1.64,
      "learning_rate": 9.833333333333333e-08,
      "loss": 0.2418,
      "step": 825
    },
    {
      "epoch": 1.68,
      "learning_rate": 8.444444444444444e-08,
      "loss": 0.2399,
      "step": 850
    },
    {
      "epoch": 1.73,
      "learning_rate": 7.055555555555554e-08,
      "loss": 0.2607,
      "step": 875
    },
    {
      "epoch": 1.78,
      "learning_rate": 5.666666666666666e-08,
      "loss": 0.2492,
      "step": 900
    },
    {
      "epoch": 1.78,
      "eval_loss": 0.3794139325618744,
      "eval_runtime": 4986.6737,
      "eval_samples_per_second": 3.237,
      "eval_steps_per_second": 0.101,
      "eval_wer": 15.46716649125504,
      "step": 900
    },
    {
      "epoch": 1.83,
      "learning_rate": 4.277777777777777e-08,
      "loss": 0.2423,
      "step": 925
    },
    {
      "epoch": 1.88,
      "learning_rate": 2.8888888888888887e-08,
      "loss": 0.2343,
      "step": 950
    },
    {
      "epoch": 1.93,
      "learning_rate": 1.5e-08,
      "loss": 0.2429,
      "step": 975
    },
    {
      "epoch": 1.98,
      "learning_rate": 1.111111111111111e-09,
      "loss": 0.2471,
      "step": 1000
    },
    {
      "epoch": 1.98,
      "eval_loss": 0.37914004921913147,
      "eval_runtime": 4987.5849,
      "eval_samples_per_second": 3.236,
      "eval_steps_per_second": 0.101,
      "eval_wer": 15.470714142118894,
      "step": 1000
    },
    {
      "epoch": 1.98,
      "step": 1000,
      "total_flos": 1.846773410807808e+19,
      "train_loss": 0.3802435531616211,
      "train_runtime": 58130.333,
      "train_samples_per_second": 1.101,
      "train_steps_per_second": 0.017
    }
  ],
  "logging_steps": 25,
  "max_steps": 1000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 100,
  "total_flos": 1.846773410807808e+19,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}