{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.4989587671803415,
  "eval_steps": 240,
  "global_step": 3600,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00041649312786339027,
      "grad_norm": 7.0,
      "learning_rate": 2e-06,
      "loss": 0.7314,
      "step": 1
    },
    {
      "epoch": 0.04164931278633902,
      "grad_norm": 0.0712890625,
      "learning_rate": 0.0002,
      "loss": 0.377,
      "step": 100
    },
    {
      "epoch": 0.08329862557267805,
      "grad_norm": 0.12109375,
      "learning_rate": 0.0004,
      "loss": 0.2401,
      "step": 200
    },
    {
      "epoch": 0.09995835068721366,
      "eval_peoplespeech-clean-transcription_loss": 3.7064812183380127,
      "eval_peoplespeech-clean-transcription_model_preparation_time": 0.0062,
      "eval_peoplespeech-clean-transcription_runtime": 14.4755,
      "eval_peoplespeech-clean-transcription_samples_per_second": 4.421,
      "eval_peoplespeech-clean-transcription_steps_per_second": 0.069,
      "step": 240
    },
    {
      "epoch": 0.12494793835901707,
      "grad_norm": 0.10791015625,
      "learning_rate": 0.0006,
      "loss": 0.209,
      "step": 300
    },
    {
      "epoch": 0.1665972511453561,
      "grad_norm": 0.08740234375,
      "learning_rate": 0.0008,
      "loss": 0.1586,
      "step": 400
    },
    {
      "epoch": 0.19991670137442732,
      "eval_peoplespeech-clean-transcription_loss": 2.0534801483154297,
      "eval_peoplespeech-clean-transcription_model_preparation_time": 0.0062,
      "eval_peoplespeech-clean-transcription_runtime": 13.9793,
      "eval_peoplespeech-clean-transcription_samples_per_second": 4.578,
      "eval_peoplespeech-clean-transcription_steps_per_second": 0.072,
      "step": 480
    },
    {
      "epoch": 0.20824656393169513,
      "grad_norm": 0.059814453125,
      "learning_rate": 0.001,
      "loss": 0.1178,
      "step": 500
    },
    {
      "epoch": 0.24989587671803415,
      "grad_norm": 0.05126953125,
      "learning_rate": 0.0012,
      "loss": 0.1031,
      "step": 600
    },
    {
      "epoch": 0.2915451895043732,
      "grad_norm": 0.04443359375,
      "learning_rate": 0.0014,
      "loss": 0.0942,
      "step": 700
    },
    {
      "epoch": 0.299875052061641,
      "eval_peoplespeech-clean-transcription_loss": 1.931348443031311,
      "eval_peoplespeech-clean-transcription_model_preparation_time": 0.0062,
      "eval_peoplespeech-clean-transcription_runtime": 14.1549,
      "eval_peoplespeech-clean-transcription_samples_per_second": 4.521,
      "eval_peoplespeech-clean-transcription_steps_per_second": 0.071,
      "step": 720
    },
    {
      "epoch": 0.3331945022907122,
      "grad_norm": 0.040283203125,
      "learning_rate": 0.0016,
      "loss": 0.0885,
      "step": 800
    },
    {
      "epoch": 0.3748438150770512,
      "grad_norm": 0.035888671875,
      "learning_rate": 0.0018000000000000002,
      "loss": 0.0853,
      "step": 900
    },
    {
      "epoch": 0.39983340274885465,
      "eval_peoplespeech-clean-transcription_loss": 1.8942928314208984,
      "eval_peoplespeech-clean-transcription_model_preparation_time": 0.0062,
      "eval_peoplespeech-clean-transcription_runtime": 13.7985,
      "eval_peoplespeech-clean-transcription_samples_per_second": 4.638,
      "eval_peoplespeech-clean-transcription_steps_per_second": 0.072,
      "step": 960
    },
    {
      "epoch": 0.41649312786339027,
      "grad_norm": 0.033203125,
      "learning_rate": 0.002,
      "loss": 0.0817,
      "step": 1000
    },
    {
      "epoch": 0.45814244064972925,
      "grad_norm": 0.03173828125,
      "learning_rate": 0.001996926043706003,
      "loss": 0.0798,
      "step": 1100
    },
    {
      "epoch": 0.4997917534360683,
      "grad_norm": 0.029296875,
      "learning_rate": 0.0019877251730624503,
      "loss": 0.0777,
      "step": 1200
    },
    {
      "epoch": 0.4997917534360683,
      "eval_peoplespeech-clean-transcription_loss": 1.863105297088623,
      "eval_peoplespeech-clean-transcription_model_preparation_time": 0.0062,
      "eval_peoplespeech-clean-transcription_runtime": 13.1079,
      "eval_peoplespeech-clean-transcription_samples_per_second": 4.883,
      "eval_peoplespeech-clean-transcription_steps_per_second": 0.076,
      "step": 1200
    },
    {
      "epoch": 0.5414410662224073,
      "grad_norm": 0.028076171875,
      "learning_rate": 0.0019724602393453973,
      "loss": 0.0761,
      "step": 1300
    },
    {
      "epoch": 0.5830903790087464,
      "grad_norm": 0.0277099609375,
      "learning_rate": 0.001951235517530571,
      "loss": 0.0744,
      "step": 1400
    },
    {
      "epoch": 0.599750104123282,
      "eval_peoplespeech-clean-transcription_loss": 1.8575688600540161,
      "eval_peoplespeech-clean-transcription_model_preparation_time": 0.0062,
      "eval_peoplespeech-clean-transcription_runtime": 14.2046,
      "eval_peoplespeech-clean-transcription_samples_per_second": 4.506,
      "eval_peoplespeech-clean-transcription_steps_per_second": 0.07,
      "step": 1440
    },
    {
      "epoch": 0.6247396917950854,
      "grad_norm": 0.02490234375,
      "learning_rate": 0.0019241959939895518,
      "loss": 0.073,
      "step": 1500
    },
    {
      "epoch": 0.6663890045814244,
      "grad_norm": 0.0263671875,
      "learning_rate": 0.0018915263760858401,
      "loss": 0.071,
      "step": 1600
    },
    {
      "epoch": 0.6997084548104956,
      "eval_peoplespeech-clean-transcription_loss": 1.83821439743042,
      "eval_peoplespeech-clean-transcription_model_preparation_time": 0.0062,
      "eval_peoplespeech-clean-transcription_runtime": 13.7329,
      "eval_peoplespeech-clean-transcription_samples_per_second": 4.66,
      "eval_peoplespeech-clean-transcription_steps_per_second": 0.073,
      "step": 1680
    },
    {
      "epoch": 0.7080383173677635,
      "grad_norm": 0.0252685546875,
      "learning_rate": 0.0018534498304362756,
      "loss": 0.0704,
      "step": 1700
    },
    {
      "epoch": 0.7496876301541024,
      "grad_norm": 0.027587890625,
      "learning_rate": 0.0018102264584567542,
      "loss": 0.0693,
      "step": 1800
    },
    {
      "epoch": 0.7913369429404414,
      "grad_norm": 0.0264892578125,
      "learning_rate": 0.0017621515196058187,
      "loss": 0.0679,
      "step": 1900
    },
    {
      "epoch": 0.7996668054977093,
      "eval_peoplespeech-clean-transcription_loss": 1.835157036781311,
      "eval_peoplespeech-clean-transcription_model_preparation_time": 0.0062,
      "eval_peoplespeech-clean-transcription_runtime": 13.3808,
      "eval_peoplespeech-clean-transcription_samples_per_second": 4.783,
      "eval_peoplespeech-clean-transcription_steps_per_second": 0.075,
      "step": 1920
    },
    {
      "epoch": 0.8329862557267805,
      "grad_norm": 0.02294921875,
      "learning_rate": 0.001709553414463167,
      "loss": 0.0678,
      "step": 2000
    },
    {
      "epoch": 0.8746355685131195,
      "grad_norm": 0.0220947265625,
      "learning_rate": 0.0016527914414207012,
      "loss": 0.0669,
      "step": 2100
    },
    {
      "epoch": 0.899625156184923,
      "eval_peoplespeech-clean-transcription_loss": 1.822905421257019,
      "eval_peoplespeech-clean-transcription_model_preparation_time": 0.0062,
      "eval_peoplespeech-clean-transcription_runtime": 13.1096,
      "eval_peoplespeech-clean-transcription_samples_per_second": 4.882,
      "eval_peoplespeech-clean-transcription_steps_per_second": 0.076,
      "step": 2160
    },
    {
      "epoch": 0.9162848812994585,
      "grad_norm": 0.0244140625,
      "learning_rate": 0.0015922533423101844,
      "loss": 0.0659,
      "step": 2200
    },
    {
      "epoch": 0.9579341940857976,
      "grad_norm": 0.0255126953125,
      "learning_rate": 0.0015283526537333662,
      "loss": 0.0658,
      "step": 2300
    },
    {
      "epoch": 0.9995835068721366,
      "grad_norm": 0.0225830078125,
      "learning_rate": 0.0014615258821876727,
      "loss": 0.0653,
      "step": 2400
    },
    {
      "epoch": 0.9995835068721366,
      "eval_peoplespeech-clean-transcription_loss": 1.8253135681152344,
      "eval_peoplespeech-clean-transcription_model_preparation_time": 0.0062,
      "eval_peoplespeech-clean-transcription_runtime": 13.4347,
      "eval_peoplespeech-clean-transcription_samples_per_second": 4.764,
      "eval_peoplespeech-clean-transcription_steps_per_second": 0.074,
      "step": 2400
    },
    {
      "epoch": 1.0408163265306123,
      "grad_norm": 0.0264892578125,
      "learning_rate": 0.0013922295222842153,
      "loss": 0.0606,
      "step": 2500
    },
    {
      "epoch": 1.0824656393169512,
      "grad_norm": 0.0205078125,
      "learning_rate": 0.0013209369384267193,
      "loss": 0.0639,
      "step": 2600
    },
    {
      "epoch": 1.099125364431487,
      "eval_peoplespeech-clean-transcription_loss": 1.7945704460144043,
      "eval_peoplespeech-clean-transcription_model_preparation_time": 0.0062,
      "eval_peoplespeech-clean-transcription_runtime": 14.4627,
      "eval_peoplespeech-clean-transcription_samples_per_second": 4.425,
      "eval_peoplespeech-clean-transcription_steps_per_second": 0.069,
      "step": 2640
    },
    {
      "epoch": 1.1241149521032903,
      "grad_norm": 0.0218505859375,
      "learning_rate": 0.0012481351312526605,
      "loss": 0.0637,
      "step": 2700
    },
    {
      "epoch": 1.1657642648896294,
      "grad_norm": 0.0225830078125,
      "learning_rate": 0.0011743214109250992,
      "loss": 0.0636,
      "step": 2800
    },
    {
      "epoch": 1.1990837151187006,
      "eval_peoplespeech-clean-transcription_loss": 1.799329400062561,
      "eval_peoplespeech-clean-transcription_model_preparation_time": 0.0062,
      "eval_peoplespeech-clean-transcription_runtime": 13.4877,
      "eval_peoplespeech-clean-transcription_samples_per_second": 4.745,
      "eval_peoplespeech-clean-transcription_steps_per_second": 0.074,
      "step": 2880
    },
    {
      "epoch": 1.2074135776759682,
      "grad_norm": 0.0216064453125,
      "learning_rate": 0.0011,
      "loss": 0.063,
      "step": 2900
    },
    {
      "epoch": 1.2490628904623073,
      "grad_norm": 0.0230712890625,
      "learning_rate": 0.001025678589074901,
      "loss": 0.0625,
      "step": 3000
    },
    {
      "epoch": 1.2907122032486464,
      "grad_norm": 0.025146484375,
      "learning_rate": 0.0009518648687473394,
      "loss": 0.0615,
      "step": 3100
    },
    {
      "epoch": 1.299042065805914,
      "eval_peoplespeech-clean-transcription_loss": 1.777367353439331,
      "eval_peoplespeech-clean-transcription_model_preparation_time": 0.0062,
      "eval_peoplespeech-clean-transcription_runtime": 13.9312,
      "eval_peoplespeech-clean-transcription_samples_per_second": 4.594,
      "eval_peoplespeech-clean-transcription_steps_per_second": 0.072,
      "step": 3120
    },
    {
      "epoch": 1.3323615160349855,
      "grad_norm": 0.0233154296875,
      "learning_rate": 0.0008790630615732809,
      "loss": 0.0608,
      "step": 3200
    },
    {
      "epoch": 1.3740108288213244,
      "grad_norm": 0.0235595703125,
      "learning_rate": 0.0008077704777157851,
      "loss": 0.0608,
      "step": 3300
    },
    {
      "epoch": 1.3990004164931278,
      "eval_peoplespeech-clean-transcription_loss": 1.7695162296295166,
      "eval_peoplespeech-clean-transcription_model_preparation_time": 0.0062,
      "eval_peoplespeech-clean-transcription_runtime": 13.6148,
      "eval_peoplespeech-clean-transcription_samples_per_second": 4.701,
      "eval_peoplespeech-clean-transcription_steps_per_second": 0.073,
      "step": 3360
    },
    {
      "epoch": 1.4156601416076635,
      "grad_norm": 0.02099609375,
      "learning_rate": 0.0007384741178123277,
      "loss": 0.0596,
      "step": 3400
    },
    {
      "epoch": 1.4573094543940024,
      "grad_norm": 0.021240234375,
      "learning_rate": 0.0006716473462666338,
      "loss": 0.0595,
      "step": 3500
    },
    {
      "epoch": 1.4989587671803415,
      "grad_norm": 0.0216064453125,
      "learning_rate": 0.0006077466576898161,
      "loss": 0.0591,
      "step": 3600
    },
    {
      "epoch": 1.4989587671803415,
      "eval_peoplespeech-clean-transcription_loss": 1.7758369445800781,
      "eval_peoplespeech-clean-transcription_model_preparation_time": 0.0062,
      "eval_peoplespeech-clean-transcription_runtime": 13.8514,
      "eval_peoplespeech-clean-transcription_samples_per_second": 4.62,
      "eval_peoplespeech-clean-transcription_steps_per_second": 0.072,
      "step": 3600
    }
  ],
  "logging_steps": 100,
  "max_steps": 4800,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 1200,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 5.3862866198064e+18,
  "train_batch_size": 576,
  "trial_name": null,
  "trial_params": null
}