{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.8933417481855475,
  "eval_steps": 200,
  "global_step": 6000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.06311139160618491,
      "eval_loss": 3.0427980422973633,
      "eval_runtime": 195.72,
      "eval_samples_per_second": 35.878,
      "eval_steps_per_second": 0.562,
      "eval_wer": 1.0,
      "step": 200
    },
    {
      "epoch": 0.12622278321236982,
      "eval_loss": 3.1055634021759033,
      "eval_runtime": 193.3912,
      "eval_samples_per_second": 36.31,
      "eval_steps_per_second": 0.569,
      "eval_wer": 1.0,
      "step": 400
    },
    {
      "epoch": 0.1577784790154623,
      "grad_norm": 6.020185947418213,
      "learning_rate": 0.00024599999999999996,
      "loss": 4.232,
      "step": 500
    },
    {
      "epoch": 0.18933417481855475,
      "eval_loss": 1.2093147039413452,
      "eval_runtime": 194.3659,
      "eval_samples_per_second": 36.128,
      "eval_steps_per_second": 0.566,
      "eval_wer": 0.8286185040991775,
      "step": 600
    },
    {
      "epoch": 0.25244556642473964,
      "eval_loss": 1.0886054039001465,
      "eval_runtime": 194.9365,
      "eval_samples_per_second": 36.022,
      "eval_steps_per_second": 0.564,
      "eval_wer": 0.7324838225328533,
      "step": 800
    },
    {
      "epoch": 0.3155569580309246,
      "grad_norm": 4.922729969024658,
      "learning_rate": 0.00027822222222222224,
      "loss": 1.0379,
      "step": 1000
    },
    {
      "epoch": 0.3155569580309246,
      "eval_loss": 0.9017586708068848,
      "eval_runtime": 193.3111,
      "eval_samples_per_second": 36.325,
      "eval_steps_per_second": 0.569,
      "eval_wer": 0.673049070543058,
      "step": 1000
    },
    {
      "epoch": 0.3786683496371095,
      "eval_loss": 0.8131064772605896,
      "eval_runtime": 192.8235,
      "eval_samples_per_second": 36.417,
      "eval_steps_per_second": 0.57,
      "eval_wer": 0.5803292629452956,
      "step": 1200
    },
    {
      "epoch": 0.4417797412432944,
      "eval_loss": 0.7567442655563354,
      "eval_runtime": 192.8955,
      "eval_samples_per_second": 36.403,
      "eval_steps_per_second": 0.57,
      "eval_wer": 0.5543788782736948,
      "step": 1400
    },
    {
      "epoch": 0.47333543704638686,
      "grad_norm": 25.353076934814453,
      "learning_rate": 0.00025049999999999996,
      "loss": 0.8008,
      "step": 1500
    },
    {
      "epoch": 0.5048911328494793,
      "eval_loss": 0.704021155834198,
      "eval_runtime": 193.2182,
      "eval_samples_per_second": 36.342,
      "eval_steps_per_second": 0.569,
      "eval_wer": 0.5137857266240583,
      "step": 1600
    },
    {
      "epoch": 0.5680025244556642,
      "eval_loss": 0.6949470043182373,
      "eval_runtime": 192.7198,
      "eval_samples_per_second": 36.436,
      "eval_steps_per_second": 0.571,
      "eval_wer": 0.5236450125566378,
      "step": 1800
    },
    {
      "epoch": 0.6311139160618492,
      "grad_norm": NaN,
      "learning_rate": 0.00022277777777777774,
      "loss": 0.7212,
      "step": 2000
    },
    {
      "epoch": 0.6311139160618492,
      "eval_loss": 0.672233521938324,
      "eval_runtime": 194.6494,
      "eval_samples_per_second": 36.075,
      "eval_steps_per_second": 0.565,
      "eval_wer": 0.4992758341195073,
      "step": 2000
    },
    {
      "epoch": 0.694225307668034,
      "eval_loss": 0.6403974890708923,
      "eval_runtime": 192.2183,
      "eval_samples_per_second": 36.531,
      "eval_steps_per_second": 0.572,
      "eval_wer": 0.4761689631804834,
      "step": 2200
    },
    {
      "epoch": 0.757336699274219,
      "eval_loss": 0.6335896849632263,
      "eval_runtime": 192.0564,
      "eval_samples_per_second": 36.562,
      "eval_steps_per_second": 0.573,
      "eval_wer": 0.4685552558497987,
      "step": 2400
    },
    {
      "epoch": 0.7888923950773115,
      "grad_norm": 5.390285968780518,
      "learning_rate": 0.00019505555555555555,
      "loss": 0.6639,
      "step": 2500
    },
    {
      "epoch": 0.8204480908804039,
      "eval_loss": 0.593280553817749,
      "eval_runtime": 191.8655,
      "eval_samples_per_second": 36.599,
      "eval_steps_per_second": 0.573,
      "eval_wer": 0.45866939502252224,
      "step": 2600
    },
    {
      "epoch": 0.8835594824865888,
      "eval_loss": 0.5996308326721191,
      "eval_runtime": 192.1279,
      "eval_samples_per_second": 36.549,
      "eval_steps_per_second": 0.573,
      "eval_wer": 0.45444398676570247,
      "step": 2800
    },
    {
      "epoch": 0.9466708740927737,
      "grad_norm": 7.231433391571045,
      "learning_rate": 0.00016733333333333333,
      "loss": 0.6278,
      "step": 3000
    },
    {
      "epoch": 0.9466708740927737,
      "eval_loss": 0.5639352202415466,
      "eval_runtime": 195.2391,
      "eval_samples_per_second": 35.966,
      "eval_steps_per_second": 0.563,
      "eval_wer": 0.42424161894258494,
      "step": 3000
    },
    {
      "epoch": 1.0097822656989586,
      "eval_loss": 0.566460371017456,
      "eval_runtime": 191.9621,
      "eval_samples_per_second": 36.58,
      "eval_steps_per_second": 0.573,
      "eval_wer": 0.4227534248395541,
      "step": 3200
    },
    {
      "epoch": 1.0728936573051435,
      "eval_loss": 0.5476272106170654,
      "eval_runtime": 193.6396,
      "eval_samples_per_second": 36.263,
      "eval_steps_per_second": 0.568,
      "eval_wer": 0.41909937681871934,
      "step": 3400
    },
    {
      "epoch": 1.104449353108236,
      "grad_norm": 4.519629955291748,
      "learning_rate": 0.00013955555555555555,
      "loss": 0.5528,
      "step": 3500
    },
    {
      "epoch": 1.1360050489113285,
      "eval_loss": 0.5440065860748291,
      "eval_runtime": 191.7996,
      "eval_samples_per_second": 36.611,
      "eval_steps_per_second": 0.574,
      "eval_wer": 0.41887349021379505,
      "step": 3600
    },
    {
      "epoch": 1.1991164405175134,
      "eval_loss": 0.5297770500183105,
      "eval_runtime": 191.4083,
      "eval_samples_per_second": 36.686,
      "eval_steps_per_second": 0.575,
      "eval_wer": 0.4070875244156845,
      "step": 3800
    },
    {
      "epoch": 1.2622278321236984,
      "grad_norm": 1.4278947114944458,
      "learning_rate": 0.00011183333333333332,
      "loss": 0.5103,
      "step": 4000
    },
    {
      "epoch": 1.2622278321236984,
      "eval_loss": 0.5384453535079956,
      "eval_runtime": 195.1408,
      "eval_samples_per_second": 35.984,
      "eval_steps_per_second": 0.564,
      "eval_wer": 0.4025033550804555,
      "step": 4000
    },
    {
      "epoch": 1.325339223729883,
      "eval_loss": 0.531086266040802,
      "eval_runtime": 192.0473,
      "eval_samples_per_second": 36.564,
      "eval_steps_per_second": 0.573,
      "eval_wer": 0.39976614092666657,
      "step": 4200
    },
    {
      "epoch": 1.388450615336068,
      "eval_loss": 0.5395579934120178,
      "eval_runtime": 191.3816,
      "eval_samples_per_second": 36.691,
      "eval_steps_per_second": 0.575,
      "eval_wer": 0.4039383993940924,
      "step": 4400
    },
    {
      "epoch": 1.4200063111391605,
      "grad_norm": 2.4821906089782715,
      "learning_rate": 8.411111111111111e-05,
      "loss": 0.5194,
      "step": 4500
    },
    {
      "epoch": 1.451562006942253,
      "eval_loss": 0.5501742959022522,
      "eval_runtime": 190.6511,
      "eval_samples_per_second": 36.832,
      "eval_steps_per_second": 0.577,
      "eval_wer": 0.4048818081558352,
      "step": 4600
    },
    {
      "epoch": 1.514673398548438,
      "eval_loss": 0.6632032990455627,
      "eval_runtime": 193.4271,
      "eval_samples_per_second": 36.303,
      "eval_steps_per_second": 0.569,
      "eval_wer": 0.4365059328452411,
      "step": 4800
    },
    {
      "epoch": 1.577784790154623,
      "grad_norm": 11.192009925842285,
      "learning_rate": 5.649999999999999e-05,
      "loss": 0.6034,
      "step": 5000
    },
    {
      "epoch": 1.577784790154623,
      "eval_loss": 0.7074605226516724,
      "eval_runtime": 191.8555,
      "eval_samples_per_second": 36.6,
      "eval_steps_per_second": 0.573,
      "eval_wer": 0.4368115441342564,
      "step": 5000
    },
    {
      "epoch": 1.6408961817608079,
      "eval_loss": 0.7465850710868835,
      "eval_runtime": 191.8575,
      "eval_samples_per_second": 36.6,
      "eval_steps_per_second": 0.573,
      "eval_wer": 0.44187406157403103,
      "step": 5200
    },
    {
      "epoch": 1.7040075733669928,
      "eval_loss": 0.7624653577804565,
      "eval_runtime": 189.4967,
      "eval_samples_per_second": 37.056,
      "eval_steps_per_second": 0.58,
      "eval_wer": 0.449819955088428,
      "step": 5400
    },
    {
      "epoch": 1.7355632691700853,
      "grad_norm": 17.51552963256836,
      "learning_rate": 2.8777777777777776e-05,
      "loss": 0.74,
      "step": 5500
    },
    {
      "epoch": 1.7671189649731778,
      "eval_loss": 0.7502115368843079,
      "eval_runtime": 192.6564,
      "eval_samples_per_second": 36.448,
      "eval_steps_per_second": 0.571,
      "eval_wer": 0.4474813643550937,
      "step": 5600
    },
    {
      "epoch": 1.8302303565793627,
      "eval_loss": 0.7739897966384888,
      "eval_runtime": 189.3036,
      "eval_samples_per_second": 37.094,
      "eval_steps_per_second": 0.581,
      "eval_wer": 0.45959951633691654,
      "step": 5800
    },
    {
      "epoch": 1.8933417481855475,
      "grad_norm": 10.696316719055176,
      "learning_rate": 1e-06,
      "loss": 0.7844,
      "step": 6000
    },
    {
      "epoch": 1.8933417481855475,
      "eval_loss": 0.7821305990219116,
      "eval_runtime": 189.7595,
      "eval_samples_per_second": 37.005,
      "eval_steps_per_second": 0.58,
      "eval_wer": 0.45755324944524906,
      "step": 6000
    },
    {
      "epoch": 1.8933417481855475,
      "step": 6000,
      "total_flos": 1.3297699695003722e+19,
      "train_loss": 0.9828314208984374,
      "train_runtime": 9865.427,
      "train_samples_per_second": 9.731,
      "train_steps_per_second": 0.608
    }
  ],
  "logging_steps": 500,
  "max_steps": 6000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.3297699695003722e+19,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}