{
  "best_metric": 5.218643517767322,
  "best_model_checkpoint": "OUTCOMESAI/whisper-large-v3-common-n-medical-50-50/checkpoint-4750",
  "epoch": 1.937984496124031,
  "eval_steps": 250,
  "global_step": 5000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.03875968992248062,
      "grad_norm": 4.327393531799316,
      "learning_rate": 4.123797088618779e-07,
      "loss": 6.6181,
      "step": 100
    },
    {
      "epoch": 0.07751937984496124,
      "grad_norm": 7.569775104522705,
      "learning_rate": 4.775003968157492e-07,
      "loss": 5.126,
      "step": 200
    },
    {
      "epoch": 0.09689922480620156,
      "eval_loss": 0.369384765625,
      "eval_runtime": 2703.3784,
      "eval_samples_per_second": 0.83,
      "eval_steps_per_second": 0.026,
      "eval_wer": 5.660122234842339,
      "step": 250
    },
    {
      "epoch": 0.11627906976744186,
      "grad_norm": 3.0063843727111816,
      "learning_rate": 4.953684210526315e-07,
      "loss": 4.8144,
      "step": 300
    },
    {
      "epoch": 0.15503875968992248,
      "grad_norm": 3.107106924057007,
      "learning_rate": 4.848421052631578e-07,
      "loss": 4.5831,
      "step": 400
    },
    {
      "epoch": 0.1937984496124031,
      "grad_norm": 2.855757236480713,
      "learning_rate": 4.7431578947368417e-07,
      "loss": 4.367,
      "step": 500
    },
    {
      "epoch": 0.1937984496124031,
      "eval_loss": 0.358642578125,
      "eval_runtime": 2810.867,
      "eval_samples_per_second": 0.798,
      "eval_steps_per_second": 0.025,
      "eval_wer": 5.815620098436908,
      "step": 500
    },
    {
      "epoch": 0.23255813953488372,
      "grad_norm": 2.5796642303466797,
      "learning_rate": 4.637894736842105e-07,
      "loss": 4.2067,
      "step": 600
    },
    {
      "epoch": 0.2713178294573643,
      "grad_norm": 2.537674903869629,
      "learning_rate": 4.532631578947368e-07,
      "loss": 4.1514,
      "step": 700
    },
    {
      "epoch": 0.29069767441860467,
      "eval_loss": 0.35107421875,
      "eval_runtime": 2945.7156,
      "eval_samples_per_second": 0.761,
      "eval_steps_per_second": 0.024,
      "eval_wer": 5.883903942884959,
      "step": 750
    },
    {
      "epoch": 0.31007751937984496,
      "grad_norm": 2.430060386657715,
      "learning_rate": 4.4273684210526315e-07,
      "loss": 4.0786,
      "step": 800
    },
    {
      "epoch": 0.3488372093023256,
      "grad_norm": 2.526545286178589,
      "learning_rate": 4.322105263157895e-07,
      "loss": 4.0004,
      "step": 900
    },
    {
      "epoch": 0.3875968992248062,
      "grad_norm": 2.5485153198242188,
      "learning_rate": 4.2168421052631575e-07,
      "loss": 3.962,
      "step": 1000
    },
    {
      "epoch": 0.3875968992248062,
      "eval_loss": 0.344970703125,
      "eval_runtime": 2929.618,
      "eval_samples_per_second": 0.766,
      "eval_steps_per_second": 0.024,
      "eval_wer": 5.780464059711179,
      "step": 1000
    },
    {
      "epoch": 0.4263565891472868,
      "grad_norm": 2.46012282371521,
      "learning_rate": 4.111578947368421e-07,
      "loss": 3.9314,
      "step": 1100
    },
    {
      "epoch": 0.46511627906976744,
      "grad_norm": 2.6299571990966797,
      "learning_rate": 4.0063157894736836e-07,
      "loss": 3.9038,
      "step": 1200
    },
    {
      "epoch": 0.4844961240310077,
      "eval_loss": 0.34033203125,
      "eval_runtime": 3171.0392,
      "eval_samples_per_second": 0.707,
      "eval_steps_per_second": 0.022,
      "eval_wer": 6.174617340040023,
      "step": 1250
    },
    {
      "epoch": 0.5038759689922481,
      "grad_norm": 2.334170341491699,
      "learning_rate": 3.9010526315789474e-07,
      "loss": 3.8762,
      "step": 1300
    },
    {
      "epoch": 0.5426356589147286,
      "grad_norm": 2.6218087673187256,
      "learning_rate": 3.7957894736842106e-07,
      "loss": 3.8491,
      "step": 1400
    },
    {
      "epoch": 0.5813953488372093,
      "grad_norm": 2.390178680419922,
      "learning_rate": 3.6905263157894734e-07,
      "loss": 3.8313,
      "step": 1500
    },
    {
      "epoch": 0.5813953488372093,
      "eval_loss": 0.3359375,
      "eval_runtime": 3066.5551,
      "eval_samples_per_second": 0.731,
      "eval_steps_per_second": 0.023,
      "eval_wer": 5.973822272702688,
      "step": 1500
    },
    {
      "epoch": 0.6201550387596899,
      "grad_norm": 2.461087465286255,
      "learning_rate": 3.5852631578947367e-07,
      "loss": 3.804,
      "step": 1600
    },
    {
      "epoch": 0.6589147286821705,
      "grad_norm": 2.4371871948242188,
      "learning_rate": 3.4799999999999994e-07,
      "loss": 3.7778,
      "step": 1700
    },
    {
      "epoch": 0.6782945736434108,
      "eval_loss": 0.333251953125,
      "eval_runtime": 3069.4158,
      "eval_samples_per_second": 0.731,
      "eval_steps_per_second": 0.023,
      "eval_wer": 5.921764292281898,
      "step": 1750
    },
    {
      "epoch": 0.6976744186046512,
      "grad_norm": 2.413174867630005,
      "learning_rate": 3.374736842105263e-07,
      "loss": 3.7737,
      "step": 1800
    },
    {
      "epoch": 0.7364341085271318,
      "grad_norm": 2.4858949184417725,
      "learning_rate": 3.2694736842105265e-07,
      "loss": 3.7538,
      "step": 1900
    },
    {
      "epoch": 0.7751937984496124,
      "grad_norm": 2.6870527267456055,
      "learning_rate": 3.164210526315789e-07,
      "loss": 3.7421,
      "step": 2000
    },
    {
      "epoch": 0.7751937984496124,
      "eval_loss": 0.33056640625,
      "eval_runtime": 3182.1093,
      "eval_samples_per_second": 0.705,
      "eval_steps_per_second": 0.022,
      "eval_wer": 6.13270052463627,
      "step": 2000
    },
    {
      "epoch": 0.813953488372093,
      "grad_norm": 2.504830837249756,
      "learning_rate": 3.0589473684210525e-07,
      "loss": 3.7371,
      "step": 2100
    },
    {
      "epoch": 0.8527131782945736,
      "grad_norm": 2.3637797832489014,
      "learning_rate": 2.953684210526315e-07,
      "loss": 3.7367,
      "step": 2200
    },
    {
      "epoch": 0.872093023255814,
      "eval_loss": 0.328125,
      "eval_runtime": 2953.0944,
      "eval_samples_per_second": 0.76,
      "eval_steps_per_second": 0.024,
      "eval_wer": 5.656065768835524,
      "step": 2250
    },
    {
      "epoch": 0.8914728682170543,
      "grad_norm": 2.3057029247283936,
      "learning_rate": 2.848421052631579e-07,
      "loss": 3.7176,
      "step": 2300
    },
    {
      "epoch": 0.9302325581395349,
      "grad_norm": 2.277510643005371,
      "learning_rate": 2.7431578947368423e-07,
      "loss": 3.7003,
      "step": 2400
    },
    {
      "epoch": 0.9689922480620154,
      "grad_norm": 2.3119587898254395,
      "learning_rate": 2.637894736842105e-07,
      "loss": 3.6878,
      "step": 2500
    },
    {
      "epoch": 0.9689922480620154,
      "eval_loss": 0.32568359375,
      "eval_runtime": 2931.2759,
      "eval_samples_per_second": 0.765,
      "eval_steps_per_second": 0.024,
      "eval_wer": 5.515441613932608,
      "step": 2500
    },
    {
      "epoch": 1.0077519379844961,
      "grad_norm": 2.32456636428833,
      "learning_rate": 2.5326315789473683e-07,
      "loss": 3.6732,
      "step": 2600
    },
    {
      "epoch": 1.0465116279069768,
      "grad_norm": 2.489800453186035,
      "learning_rate": 2.4273684210526316e-07,
      "loss": 3.6769,
      "step": 2700
    },
    {
      "epoch": 1.0658914728682172,
      "eval_loss": 0.32421875,
      "eval_runtime": 2928.9686,
      "eval_samples_per_second": 0.766,
      "eval_steps_per_second": 0.024,
      "eval_wer": 5.48028557520688,
      "step": 2750
    },
    {
      "epoch": 1.0852713178294573,
      "grad_norm": 2.307305097579956,
      "learning_rate": 2.3221052631578946e-07,
      "loss": 3.6646,
      "step": 2800
    },
    {
      "epoch": 1.124031007751938,
      "grad_norm": 2.3552660942077637,
      "learning_rate": 2.216842105263158e-07,
      "loss": 3.6604,
      "step": 2900
    },
    {
      "epoch": 1.1627906976744187,
      "grad_norm": 2.3546299934387207,
      "learning_rate": 2.111578947368421e-07,
      "loss": 3.6508,
      "step": 3000
    },
    {
      "epoch": 1.1627906976744187,
      "eval_loss": 0.323486328125,
      "eval_runtime": 2929.9225,
      "eval_samples_per_second": 0.766,
      "eval_steps_per_second": 0.024,
      "eval_wer": 5.463383633511818,
      "step": 3000
    },
    {
      "epoch": 1.2015503875968991,
      "grad_norm": 2.2800722122192383,
      "learning_rate": 2.006315789473684e-07,
      "loss": 3.6412,
      "step": 3100
    },
    {
      "epoch": 1.2403100775193798,
      "grad_norm": 2.368227005004883,
      "learning_rate": 1.9010526315789475e-07,
      "loss": 3.6292,
      "step": 3200
    },
    {
      "epoch": 1.2596899224806202,
      "eval_loss": 0.322021484375,
      "eval_runtime": 2816.4937,
      "eval_samples_per_second": 0.796,
      "eval_steps_per_second": 0.025,
      "eval_wer": 5.3511547406566065,
      "step": 3250
    },
    {
      "epoch": 1.2790697674418605,
      "grad_norm": 2.3508598804473877,
      "learning_rate": 1.7957894736842105e-07,
      "loss": 3.6196,
      "step": 3300
    },
    {
      "epoch": 1.3178294573643412,
      "grad_norm": 2.4686381816864014,
      "learning_rate": 1.6905263157894737e-07,
      "loss": 3.6282,
      "step": 3400
    },
    {
      "epoch": 1.3565891472868217,
      "grad_norm": 2.3237240314483643,
      "learning_rate": 1.5852631578947367e-07,
      "loss": 3.6179,
      "step": 3500
    },
    {
      "epoch": 1.3565891472868217,
      "eval_loss": 0.321044921875,
      "eval_runtime": 2685.856,
      "eval_samples_per_second": 0.835,
      "eval_steps_per_second": 0.026,
      "eval_wer": 5.225404294445346,
      "step": 3500
    },
    {
      "epoch": 1.3953488372093024,
      "grad_norm": 2.2395377159118652,
      "learning_rate": 1.4799999999999998e-07,
      "loss": 3.6078,
      "step": 3600
    },
    {
      "epoch": 1.4341085271317828,
      "grad_norm": 2.3762331008911133,
      "learning_rate": 1.374736842105263e-07,
      "loss": 3.6032,
      "step": 3700
    },
    {
      "epoch": 1.4534883720930232,
      "eval_loss": 0.320556640625,
      "eval_runtime": 2693.0163,
      "eval_samples_per_second": 0.833,
      "eval_steps_per_second": 0.026,
      "eval_wer": 5.2206717507707285,
      "step": 3750
    },
    {
      "epoch": 1.4728682170542635,
      "grad_norm": 2.3068864345550537,
      "learning_rate": 1.2694736842105263e-07,
      "loss": 3.5997,
      "step": 3800
    },
    {
      "epoch": 1.5116279069767442,
      "grad_norm": 2.40097713470459,
      "learning_rate": 1.1642105263157894e-07,
      "loss": 3.588,
      "step": 3900
    },
    {
      "epoch": 1.550387596899225,
      "grad_norm": 2.320568561553955,
      "learning_rate": 1.0589473684210526e-07,
      "loss": 3.5922,
      "step": 4000
    },
    {
      "epoch": 1.550387596899225,
      "eval_loss": 0.320068359375,
      "eval_runtime": 2803.374,
      "eval_samples_per_second": 0.8,
      "eval_steps_per_second": 0.025,
      "eval_wer": 5.303829303910433,
      "step": 4000
    },
    {
      "epoch": 1.5891472868217056,
      "grad_norm": 2.4094741344451904,
      "learning_rate": 9.557894736842105e-08,
      "loss": 3.5832,
      "step": 4100
    },
    {
      "epoch": 1.627906976744186,
      "grad_norm": 2.448002815246582,
      "learning_rate": 8.505263157894736e-08,
      "loss": 3.5743,
      "step": 4200
    },
    {
      "epoch": 1.6472868217054264,
      "eval_loss": 0.31982421875,
      "eval_runtime": 2778.7179,
      "eval_samples_per_second": 0.807,
      "eval_steps_per_second": 0.026,
      "eval_wer": 5.263264643842285,
      "step": 4250
    },
    {
      "epoch": 1.6666666666666665,
      "grad_norm": 2.362182855606079,
      "learning_rate": 7.452631578947369e-08,
      "loss": 3.5831,
      "step": 4300
    },
    {
      "epoch": 1.7054263565891472,
      "grad_norm": 2.3121325969696045,
      "learning_rate": 6.4e-08,
      "loss": 3.5738,
      "step": 4400
    },
    {
      "epoch": 1.744186046511628,
      "grad_norm": 2.3000595569610596,
      "learning_rate": 5.347368421052632e-08,
      "loss": 3.5882,
      "step": 4500
    },
    {
      "epoch": 1.744186046511628,
      "eval_loss": 0.31982421875,
      "eval_runtime": 2718.6235,
      "eval_samples_per_second": 0.825,
      "eval_steps_per_second": 0.026,
      "eval_wer": 5.225404294445346,
      "step": 4500
    },
    {
      "epoch": 1.7829457364341086,
      "grad_norm": 2.3451480865478516,
      "learning_rate": 4.2947368421052626e-08,
      "loss": 3.5873,
      "step": 4600
    },
    {
      "epoch": 1.8217054263565893,
      "grad_norm": 2.301790237426758,
      "learning_rate": 3.2421052631578947e-08,
      "loss": 3.6021,
      "step": 4700
    },
    {
      "epoch": 1.8410852713178296,
      "eval_loss": 0.319580078125,
      "eval_runtime": 2680.8066,
      "eval_samples_per_second": 0.837,
      "eval_steps_per_second": 0.026,
      "eval_wer": 5.218643517767322,
      "step": 4750
    },
    {
      "epoch": 1.8604651162790697,
      "grad_norm": 2.2558062076568604,
      "learning_rate": 2.189473684210526e-08,
      "loss": 3.5833,
      "step": 4800
    },
    {
      "epoch": 1.8992248062015504,
      "grad_norm": 2.2542638778686523,
      "learning_rate": 1.136842105263158e-08,
      "loss": 3.5732,
      "step": 4900
    },
    {
      "epoch": 1.937984496124031,
      "grad_norm": 2.378117322921753,
      "learning_rate": 8.421052631578947e-10,
      "loss": 3.5865,
      "step": 5000
    },
    {
      "epoch": 1.937984496124031,
      "eval_loss": 0.3193359375,
      "eval_runtime": 2696.1288,
      "eval_samples_per_second": 0.832,
      "eval_steps_per_second": 0.026,
      "eval_wer": 5.221347828438531,
      "step": 5000
    },
    {
      "epoch": 1.937984496124031,
      "step": 5000,
      "total_flos": 1.087083891827713e+21,
      "train_loss": 3.86036796875,
      "train_runtime": 85811.7084,
      "train_samples_per_second": 3.729,
      "train_steps_per_second": 0.058
    }
  ],
  "logging_steps": 100,
  "max_steps": 5000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 250,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.087083891827713e+21,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}