{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.7399813606710158,
  "eval_steps": 318,
  "global_step": 3176,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00023299161230195712,
      "grad_norm": 6.78125,
      "learning_rate": 2e-06,
      "loss": 0.7236,
      "step": 1
    },
    {
      "epoch": 0.023299161230195712,
      "grad_norm": 0.09033203125,
      "learning_rate": 0.0002,
      "loss": 0.3799,
      "step": 100
    },
    {
      "epoch": 0.046598322460391424,
      "grad_norm": 0.11865234375,
      "learning_rate": 0.0004,
      "loss": 0.2452,
      "step": 200
    },
    {
      "epoch": 0.06989748369058714,
      "grad_norm": 0.1396484375,
      "learning_rate": 0.0006,
      "loss": 0.2131,
      "step": 300
    },
    {
      "epoch": 0.07409133271202237,
      "eval_peoplespeech-clean-transcription_loss": 3.0843491554260254,
      "eval_peoplespeech-clean-transcription_model_preparation_time": 0.0063,
      "eval_peoplespeech-clean-transcription_runtime": 14.6247,
      "eval_peoplespeech-clean-transcription_samples_per_second": 4.376,
      "eval_peoplespeech-clean-transcription_steps_per_second": 0.068,
      "step": 318
    },
    {
      "epoch": 0.09319664492078285,
      "grad_norm": 0.07421875,
      "learning_rate": 0.0008,
      "loss": 0.155,
      "step": 400
    },
    {
      "epoch": 0.11649580615097857,
      "grad_norm": 0.0576171875,
      "learning_rate": 0.001,
      "loss": 0.1083,
      "step": 500
    },
    {
      "epoch": 0.13979496738117428,
      "grad_norm": 0.04736328125,
      "learning_rate": 0.0012,
      "loss": 0.0926,
      "step": 600
    },
    {
      "epoch": 0.14818266542404473,
      "eval_peoplespeech-clean-transcription_loss": 1.7947263717651367,
      "eval_peoplespeech-clean-transcription_model_preparation_time": 0.0063,
      "eval_peoplespeech-clean-transcription_runtime": 13.9513,
      "eval_peoplespeech-clean-transcription_samples_per_second": 4.587,
      "eval_peoplespeech-clean-transcription_steps_per_second": 0.072,
      "step": 636
    },
    {
      "epoch": 0.16309412861137,
      "grad_norm": 0.04150390625,
      "learning_rate": 0.0014,
      "loss": 0.0844,
      "step": 700
    },
    {
      "epoch": 0.1863932898415657,
      "grad_norm": 0.034912109375,
      "learning_rate": 0.0016,
      "loss": 0.0793,
      "step": 800
    },
    {
      "epoch": 0.2096924510717614,
      "grad_norm": 0.032958984375,
      "learning_rate": 0.0018000000000000002,
      "loss": 0.0758,
      "step": 900
    },
    {
      "epoch": 0.2222739981360671,
      "eval_peoplespeech-clean-transcription_loss": 1.7458150386810303,
      "eval_peoplespeech-clean-transcription_model_preparation_time": 0.0063,
      "eval_peoplespeech-clean-transcription_runtime": 14.2202,
      "eval_peoplespeech-clean-transcription_samples_per_second": 4.501,
      "eval_peoplespeech-clean-transcription_steps_per_second": 0.07,
      "step": 954
    },
    {
      "epoch": 0.23299161230195714,
      "grad_norm": 0.03125,
      "learning_rate": 0.002,
      "loss": 0.0732,
      "step": 1000
    },
    {
      "epoch": 0.25629077353215285,
      "grad_norm": 0.02783203125,
      "learning_rate": 0.0019984487567773325,
      "loss": 0.0714,
      "step": 1100
    },
    {
      "epoch": 0.27958993476234856,
      "grad_norm": 0.0289306640625,
      "learning_rate": 0.0019938003745660765,
      "loss": 0.0691,
      "step": 1200
    },
    {
      "epoch": 0.29636533084808947,
      "eval_peoplespeech-clean-transcription_loss": 1.7118018865585327,
      "eval_peoplespeech-clean-transcription_model_preparation_time": 0.0063,
      "eval_peoplespeech-clean-transcription_runtime": 14.7608,
      "eval_peoplespeech-clean-transcription_samples_per_second": 4.336,
      "eval_peoplespeech-clean-transcription_steps_per_second": 0.068,
      "step": 1272
    },
    {
      "epoch": 0.30288909599254427,
      "grad_norm": 0.027587890625,
      "learning_rate": 0.0019860708773026797,
      "loss": 0.0676,
      "step": 1300
    },
    {
      "epoch": 0.32618825722274,
      "grad_norm": 0.0257568359375,
      "learning_rate": 0.001975286910165463,
      "loss": 0.0657,
      "step": 1400
    },
    {
      "epoch": 0.3494874184529357,
      "grad_norm": 0.0244140625,
      "learning_rate": 0.0019614856477231713,
      "loss": 0.0648,
      "step": 1500
    },
    {
      "epoch": 0.3704566635601118,
      "eval_peoplespeech-clean-transcription_loss": 1.6898815631866455,
      "eval_peoplespeech-clean-transcription_model_preparation_time": 0.0063,
      "eval_peoplespeech-clean-transcription_runtime": 14.5539,
      "eval_peoplespeech-clean-transcription_samples_per_second": 4.397,
      "eval_peoplespeech-clean-transcription_steps_per_second": 0.069,
      "step": 1590
    },
    {
      "epoch": 0.3727865796831314,
      "grad_norm": 0.0208740234375,
      "learning_rate": 0.0019447146657865335,
      "loss": 0.0641,
      "step": 1600
    },
    {
      "epoch": 0.3960857409133271,
      "grad_norm": 0.02587890625,
      "learning_rate": 0.001925031777404586,
      "loss": 0.0629,
      "step": 1700
    },
    {
      "epoch": 0.4193849021435228,
      "grad_norm": 0.0269775390625,
      "learning_rate": 0.0019025048335711181,
      "loss": 0.0623,
      "step": 1800
    },
    {
      "epoch": 0.4426840633737186,
      "grad_norm": 0.0208740234375,
      "learning_rate": 0.0018772114893282391,
      "loss": 0.0616,
      "step": 1900
    },
    {
      "epoch": 0.4445479962721342,
      "eval_peoplespeech-clean-transcription_loss": 1.6862739324569702,
      "eval_peoplespeech-clean-transcription_model_preparation_time": 0.0063,
      "eval_peoplespeech-clean-transcription_runtime": 14.4561,
      "eval_peoplespeech-clean-transcription_samples_per_second": 4.427,
      "eval_peoplespeech-clean-transcription_steps_per_second": 0.069,
      "step": 1908
    },
    {
      "epoch": 0.4659832246039143,
      "grad_norm": 0.02099609375,
      "learning_rate": 0.0018492389360733603,
      "loss": 0.0612,
      "step": 2000
    },
    {
      "epoch": 0.48928238583411,
      "grad_norm": 0.0245361328125,
      "learning_rate": 0.0018186836009923876,
      "loss": 0.0605,
      "step": 2100
    },
    {
      "epoch": 0.5125815470643057,
      "grad_norm": 0.0206298828125,
      "learning_rate": 0.001785650814655233,
      "loss": 0.0603,
      "step": 2200
    },
    {
      "epoch": 0.5186393289841565,
      "eval_peoplespeech-clean-transcription_loss": 1.6496751308441162,
      "eval_peoplespeech-clean-transcription_model_preparation_time": 0.0063,
      "eval_peoplespeech-clean-transcription_runtime": 14.7041,
      "eval_peoplespeech-clean-transcription_samples_per_second": 4.353,
      "eval_peoplespeech-clean-transcription_steps_per_second": 0.068,
      "step": 2226
    },
    {
      "epoch": 0.5358807082945014,
      "grad_norm": 0.022705078125,
      "learning_rate": 0.0017502544479195221,
      "loss": 0.0599,
      "step": 2300
    },
    {
      "epoch": 0.5591798695246971,
      "grad_norm": 0.0189208984375,
      "learning_rate": 0.0017126165193941567,
      "loss": 0.0591,
      "step": 2400
    },
    {
      "epoch": 0.5824790307548928,
      "grad_norm": 0.0208740234375,
      "learning_rate": 0.0016728667748158923,
      "loss": 0.0591,
      "step": 2500
    },
    {
      "epoch": 0.5927306616961789,
      "eval_peoplespeech-clean-transcription_loss": 1.6663990020751953,
      "eval_peoplespeech-clean-transcription_model_preparation_time": 0.0063,
      "eval_peoplespeech-clean-transcription_runtime": 14.7316,
      "eval_peoplespeech-clean-transcription_samples_per_second": 4.344,
      "eval_peoplespeech-clean-transcription_steps_per_second": 0.068,
      "step": 2544
    },
    {
      "epoch": 0.6057781919850885,
      "grad_norm": 0.021728515625,
      "learning_rate": 0.0016311422397889072,
      "loss": 0.0585,
      "step": 2600
    },
    {
      "epoch": 0.6290773532152842,
      "grad_norm": 0.022216796875,
      "learning_rate": 0.0015875867474291552,
      "loss": 0.0582,
      "step": 2700
    },
    {
      "epoch": 0.65237651444548,
      "grad_norm": 0.0189208984375,
      "learning_rate": 0.0015423504425418212,
      "loss": 0.0574,
      "step": 2800
    },
    {
      "epoch": 0.6668219944082013,
      "eval_peoplespeech-clean-transcription_loss": 1.619414210319519,
      "eval_peoplespeech-clean-transcription_model_preparation_time": 0.0063,
      "eval_peoplespeech-clean-transcription_runtime": 14.4777,
      "eval_peoplespeech-clean-transcription_samples_per_second": 4.421,
      "eval_peoplespeech-clean-transcription_steps_per_second": 0.069,
      "step": 2862
    },
    {
      "epoch": 0.6756756756756757,
      "grad_norm": 0.0191650390625,
      "learning_rate": 0.0014955892640410716,
      "loss": 0.057,
      "step": 2900
    },
    {
      "epoch": 0.6989748369058714,
      "grad_norm": 0.017822265625,
      "learning_rate": 0.0014474644073963134,
      "loss": 0.0573,
      "step": 3000
    },
    {
      "epoch": 0.7222739981360671,
      "grad_norm": 0.02001953125,
      "learning_rate": 0.0013981417689580185,
      "loss": 0.0572,
      "step": 3100
    }
  ],
  "logging_steps": 100,
  "max_steps": 6350,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 1588,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 5.537292414638621e+18,
  "train_batch_size": 672,
  "trial_name": null,
  "trial_params": null
}