{
  "best_metric": 3.24710634661695,
  "best_model_checkpoint": "./final-whisper-for-initial-publish-v2/checkpoint-2000",
  "epoch": 3.2948929159802307,
  "eval_steps": 500,
  "global_step": 2000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.04118616144975288,
      "grad_norm": 67.08155822753906,
      "learning_rate": 4.2000000000000006e-07,
      "loss": 2.8368,
      "step": 25
    },
    {
      "epoch": 0.08237232289950576,
      "grad_norm": 17.44784164428711,
      "learning_rate": 9.200000000000001e-07,
      "loss": 2.0282,
      "step": 50
    },
    {
      "epoch": 0.12355848434925865,
      "grad_norm": 12.481098175048828,
      "learning_rate": 1.42e-06,
      "loss": 1.3152,
      "step": 75
    },
    {
      "epoch": 0.16474464579901152,
      "grad_norm": 8.900260925292969,
      "learning_rate": 1.9200000000000003e-06,
      "loss": 0.7248,
      "step": 100
    },
    {
      "epoch": 0.20593080724876442,
      "grad_norm": 7.8956146240234375,
      "learning_rate": 2.42e-06,
      "loss": 0.5731,
      "step": 125
    },
    {
      "epoch": 0.2471169686985173,
      "grad_norm": 9.542287826538086,
      "learning_rate": 2.92e-06,
      "loss": 0.474,
      "step": 150
    },
    {
      "epoch": 0.2883031301482702,
      "grad_norm": 6.937229633331299,
      "learning_rate": 3.4200000000000007e-06,
      "loss": 0.4301,
      "step": 175
    },
    {
      "epoch": 0.32948929159802304,
      "grad_norm": 6.919783115386963,
      "learning_rate": 3.920000000000001e-06,
      "loss": 0.351,
      "step": 200
    },
    {
      "epoch": 0.37067545304777594,
      "grad_norm": 6.427859783172607,
      "learning_rate": 4.42e-06,
      "loss": 0.2726,
      "step": 225
    },
    {
      "epoch": 0.41186161449752884,
      "grad_norm": 5.83529806137085,
      "learning_rate": 4.92e-06,
      "loss": 0.1663,
      "step": 250
    },
    {
      "epoch": 0.45304777594728174,
      "grad_norm": 2.4859721660614014,
      "learning_rate": 5.420000000000001e-06,
      "loss": 0.0893,
      "step": 275
    },
    {
      "epoch": 0.4942339373970346,
      "grad_norm": 3.2526772022247314,
      "learning_rate": 5.92e-06,
      "loss": 0.076,
      "step": 300
    },
    {
      "epoch": 0.5354200988467874,
      "grad_norm": 4.417870044708252,
      "learning_rate": 6.42e-06,
      "loss": 0.0764,
      "step": 325
    },
    {
      "epoch": 0.5766062602965404,
      "grad_norm": 3.149733781814575,
      "learning_rate": 6.92e-06,
      "loss": 0.0679,
      "step": 350
    },
    {
      "epoch": 0.6177924217462932,
      "grad_norm": 3.8823468685150146,
      "learning_rate": 7.420000000000001e-06,
      "loss": 0.0525,
      "step": 375
    },
    {
      "epoch": 0.6589785831960461,
      "grad_norm": 6.1009745597839355,
      "learning_rate": 7.92e-06,
      "loss": 0.0552,
      "step": 400
    },
    {
      "epoch": 0.700164744645799,
      "grad_norm": 5.714720726013184,
      "learning_rate": 8.42e-06,
      "loss": 0.0529,
      "step": 425
    },
    {
      "epoch": 0.7413509060955519,
      "grad_norm": 4.403444290161133,
      "learning_rate": 8.920000000000001e-06,
      "loss": 0.0472,
      "step": 450
    },
    {
      "epoch": 0.7825370675453048,
      "grad_norm": 3.7249884605407715,
      "learning_rate": 9.42e-06,
      "loss": 0.0424,
      "step": 475
    },
    {
      "epoch": 0.8237232289950577,
      "grad_norm": 3.8490548133850098,
      "learning_rate": 9.920000000000002e-06,
      "loss": 0.0461,
      "step": 500
    },
    {
      "epoch": 0.8237232289950577,
      "eval_loss": 0.04240331053733826,
      "eval_runtime": 1653.3644,
      "eval_samples_per_second": 1.468,
      "eval_steps_per_second": 0.184,
      "eval_wer": 11.04637613609881,
      "step": 500
    },
    {
      "epoch": 0.8649093904448105,
      "grad_norm": 2.538515329360962,
      "learning_rate": 9.961818181818183e-06,
      "loss": 0.0384,
      "step": 525
    },
    {
      "epoch": 0.9060955518945635,
      "grad_norm": 1.6598188877105713,
      "learning_rate": 9.916363636363637e-06,
      "loss": 0.0375,
      "step": 550
    },
    {
      "epoch": 0.9472817133443163,
      "grad_norm": 3.256746292114258,
      "learning_rate": 9.870909090909092e-06,
      "loss": 0.0346,
      "step": 575
    },
    {
      "epoch": 0.9884678747940692,
      "grad_norm": 3.04457426071167,
      "learning_rate": 9.825454545454546e-06,
      "loss": 0.034,
      "step": 600
    },
    {
      "epoch": 1.029654036243822,
      "grad_norm": 0.6082770228385925,
      "learning_rate": 9.780000000000001e-06,
      "loss": 0.0272,
      "step": 625
    },
    {
      "epoch": 1.0708401976935749,
      "grad_norm": 1.6969329118728638,
      "learning_rate": 9.734545454545455e-06,
      "loss": 0.0206,
      "step": 650
    },
    {
      "epoch": 1.112026359143328,
      "grad_norm": 1.9652377367019653,
      "learning_rate": 9.68909090909091e-06,
      "loss": 0.0197,
      "step": 675
    },
    {
      "epoch": 1.1532125205930808,
      "grad_norm": 2.402329206466675,
      "learning_rate": 9.643636363636364e-06,
      "loss": 0.0221,
      "step": 700
    },
    {
      "epoch": 1.1943986820428336,
      "grad_norm": 3.167133092880249,
      "learning_rate": 9.598181818181818e-06,
      "loss": 0.0176,
      "step": 725
    },
    {
      "epoch": 1.2355848434925865,
      "grad_norm": 0.4075870215892792,
      "learning_rate": 9.552727272727273e-06,
      "loss": 0.0165,
      "step": 750
    },
    {
      "epoch": 1.2767710049423393,
      "grad_norm": 2.739176034927368,
      "learning_rate": 9.507272727272729e-06,
      "loss": 0.0156,
      "step": 775
    },
    {
      "epoch": 1.3179571663920924,
      "grad_norm": 2.4662702083587646,
      "learning_rate": 9.461818181818183e-06,
      "loss": 0.0133,
      "step": 800
    },
    {
      "epoch": 1.3591433278418452,
      "grad_norm": 2.955909252166748,
      "learning_rate": 9.416363636363636e-06,
      "loss": 0.0184,
      "step": 825
    },
    {
      "epoch": 1.400329489291598,
      "grad_norm": 2.0838463306427,
      "learning_rate": 9.370909090909092e-06,
      "loss": 0.0138,
      "step": 850
    },
    {
      "epoch": 1.441515650741351,
      "grad_norm": 2.2245779037475586,
      "learning_rate": 9.325454545454547e-06,
      "loss": 0.0132,
      "step": 875
    },
    {
      "epoch": 1.4827018121911038,
      "grad_norm": 2.146961212158203,
      "learning_rate": 9.280000000000001e-06,
      "loss": 0.0126,
      "step": 900
    },
    {
      "epoch": 1.5238879736408566,
      "grad_norm": 1.955930471420288,
      "learning_rate": 9.234545454545455e-06,
      "loss": 0.0111,
      "step": 925
    },
    {
      "epoch": 1.5650741350906094,
      "grad_norm": 2.04664945602417,
      "learning_rate": 9.18909090909091e-06,
      "loss": 0.0118,
      "step": 950
    },
    {
      "epoch": 1.6062602965403623,
      "grad_norm": 2.1349849700927734,
      "learning_rate": 9.143636363636364e-06,
      "loss": 0.0116,
      "step": 975
    },
    {
      "epoch": 1.6474464579901154,
      "grad_norm": 2.336024761199951,
      "learning_rate": 9.09818181818182e-06,
      "loss": 0.0124,
      "step": 1000
    },
    {
      "epoch": 1.6474464579901154,
      "eval_loss": 0.02211674302816391,
      "eval_runtime": 1633.7086,
      "eval_samples_per_second": 1.486,
      "eval_steps_per_second": 0.186,
      "eval_wer": 5.11147362697118,
      "step": 1000
    },
    {
      "epoch": 1.6886326194398682,
      "grad_norm": 0.4564249813556671,
      "learning_rate": 9.052727272727273e-06,
      "loss": 0.0119,
      "step": 1025
    },
    {
      "epoch": 1.729818780889621,
      "grad_norm": 1.9494842290878296,
      "learning_rate": 9.007272727272729e-06,
      "loss": 0.0114,
      "step": 1050
    },
    {
      "epoch": 1.771004942339374,
      "grad_norm": 1.7129623889923096,
      "learning_rate": 8.961818181818182e-06,
      "loss": 0.0116,
      "step": 1075
    },
    {
      "epoch": 1.812191103789127,
      "grad_norm": 0.8837174773216248,
      "learning_rate": 8.916363636363638e-06,
      "loss": 0.01,
      "step": 1100
    },
    {
      "epoch": 1.8533772652388798,
      "grad_norm": 3.1779348850250244,
      "learning_rate": 8.870909090909091e-06,
      "loss": 0.0087,
      "step": 1125
    },
    {
      "epoch": 1.8945634266886326,
      "grad_norm": 1.1833815574645996,
      "learning_rate": 8.825454545454545e-06,
      "loss": 0.0122,
      "step": 1150
    },
    {
      "epoch": 1.9357495881383855,
      "grad_norm": 2.3456814289093018,
      "learning_rate": 8.78e-06,
      "loss": 0.0089,
      "step": 1175
    },
    {
      "epoch": 1.9769357495881383,
      "grad_norm": 1.5834710597991943,
      "learning_rate": 8.734545454545456e-06,
      "loss": 0.0111,
      "step": 1200
    },
    {
      "epoch": 2.018121911037891,
      "grad_norm": 1.1926366090774536,
      "learning_rate": 8.68909090909091e-06,
      "loss": 0.0069,
      "step": 1225
    },
    {
      "epoch": 2.059308072487644,
      "grad_norm": 0.5671884417533875,
      "learning_rate": 8.643636363636364e-06,
      "loss": 0.0043,
      "step": 1250
    },
    {
      "epoch": 2.100494233937397,
      "grad_norm": 0.8897690176963806,
      "learning_rate": 8.598181818181819e-06,
      "loss": 0.0034,
      "step": 1275
    },
    {
      "epoch": 2.1416803953871497,
      "grad_norm": 1.6989620923995972,
      "learning_rate": 8.552727272727274e-06,
      "loss": 0.003,
      "step": 1300
    },
    {
      "epoch": 2.182866556836903,
      "grad_norm": 2.0327303409576416,
      "learning_rate": 8.507272727272728e-06,
      "loss": 0.0037,
      "step": 1325
    },
    {
      "epoch": 2.224052718286656,
      "grad_norm": 0.6554267406463623,
      "learning_rate": 8.461818181818182e-06,
      "loss": 0.0034,
      "step": 1350
    },
    {
      "epoch": 2.2652388797364087,
      "grad_norm": 0.21176663041114807,
      "learning_rate": 8.416363636363637e-06,
      "loss": 0.0046,
      "step": 1375
    },
    {
      "epoch": 2.3064250411861615,
      "grad_norm": 1.2915725708007812,
      "learning_rate": 8.370909090909091e-06,
      "loss": 0.0026,
      "step": 1400
    },
    {
      "epoch": 2.3476112026359144,
      "grad_norm": 0.15533868968486786,
      "learning_rate": 8.325454545454547e-06,
      "loss": 0.0043,
      "step": 1425
    },
    {
      "epoch": 2.3887973640856672,
      "grad_norm": 0.19825848937034607,
      "learning_rate": 8.28e-06,
      "loss": 0.0042,
      "step": 1450
    },
    {
      "epoch": 2.42998352553542,
      "grad_norm": 0.14249640703201294,
      "learning_rate": 8.234545454545456e-06,
      "loss": 0.0023,
      "step": 1475
    },
    {
      "epoch": 2.471169686985173,
      "grad_norm": 0.12349322438240051,
      "learning_rate": 8.18909090909091e-06,
      "loss": 0.0026,
      "step": 1500
    },
    {
      "epoch": 2.471169686985173,
      "eval_loss": 0.015178721398115158,
      "eval_runtime": 1647.8725,
      "eval_samples_per_second": 1.473,
      "eval_steps_per_second": 0.184,
      "eval_wer": 3.5966752116833685,
      "step": 1500
    },
    {
      "epoch": 2.5123558484349258,
      "grad_norm": 0.24572087824344635,
      "learning_rate": 8.143636363636365e-06,
      "loss": 0.0032,
      "step": 1525
    },
    {
      "epoch": 2.5535420098846786,
      "grad_norm": 0.2075294703245163,
      "learning_rate": 8.098181818181819e-06,
      "loss": 0.0023,
      "step": 1550
    },
    {
      "epoch": 2.594728171334432,
      "grad_norm": 0.264713317155838,
      "learning_rate": 8.052727272727272e-06,
      "loss": 0.0044,
      "step": 1575
    },
    {
      "epoch": 2.6359143327841847,
      "grad_norm": 1.9720451831817627,
      "learning_rate": 8.007272727272728e-06,
      "loss": 0.0048,
      "step": 1600
    },
    {
      "epoch": 2.6771004942339376,
      "grad_norm": 0.7824451327323914,
      "learning_rate": 7.961818181818183e-06,
      "loss": 0.0048,
      "step": 1625
    },
    {
      "epoch": 2.7182866556836904,
      "grad_norm": 0.681707501411438,
      "learning_rate": 7.916363636363637e-06,
      "loss": 0.0036,
      "step": 1650
    },
    {
      "epoch": 2.7594728171334433,
      "grad_norm": 0.39950481057167053,
      "learning_rate": 7.870909090909091e-06,
      "loss": 0.0047,
      "step": 1675
    },
    {
      "epoch": 2.800658978583196,
      "grad_norm": 0.7777596712112427,
      "learning_rate": 7.825454545454546e-06,
      "loss": 0.0029,
      "step": 1700
    },
    {
      "epoch": 2.841845140032949,
      "grad_norm": 0.06510231643915176,
      "learning_rate": 7.78e-06,
      "loss": 0.0067,
      "step": 1725
    },
    {
      "epoch": 2.883031301482702,
      "grad_norm": 2.2729947566986084,
      "learning_rate": 7.734545454545455e-06,
      "loss": 0.0024,
      "step": 1750
    },
    {
      "epoch": 2.9242174629324547,
      "grad_norm": 1.6635749340057373,
      "learning_rate": 7.68909090909091e-06,
      "loss": 0.0035,
      "step": 1775
    },
    {
      "epoch": 2.9654036243822075,
      "grad_norm": 0.06815456598997116,
      "learning_rate": 7.643636363636365e-06,
      "loss": 0.004,
      "step": 1800
    },
    {
      "epoch": 3.0065897858319603,
      "grad_norm": 1.5607892274856567,
      "learning_rate": 7.598181818181819e-06,
      "loss": 0.0046,
      "step": 1825
    },
    {
      "epoch": 3.047775947281713,
      "grad_norm": 0.8395469784736633,
      "learning_rate": 7.552727272727274e-06,
      "loss": 0.0018,
      "step": 1850
    },
    {
      "epoch": 3.088962108731466,
      "grad_norm": 0.04032081738114357,
      "learning_rate": 7.507272727272728e-06,
      "loss": 0.0011,
      "step": 1875
    },
    {
      "epoch": 3.130148270181219,
      "grad_norm": 0.3644678294658661,
      "learning_rate": 7.461818181818182e-06,
      "loss": 0.0014,
      "step": 1900
    },
    {
      "epoch": 3.171334431630972,
      "grad_norm": 0.35729554295539856,
      "learning_rate": 7.416363636363637e-06,
      "loss": 0.0009,
      "step": 1925
    },
    {
      "epoch": 3.212520593080725,
      "grad_norm": 0.05964287370443344,
      "learning_rate": 7.370909090909092e-06,
      "loss": 0.0014,
      "step": 1950
    },
    {
      "epoch": 3.253706754530478,
      "grad_norm": 0.799541711807251,
      "learning_rate": 7.325454545454546e-06,
      "loss": 0.0015,
      "step": 1975
    },
    {
      "epoch": 3.2948929159802307,
      "grad_norm": 0.058581918478012085,
      "learning_rate": 7.280000000000001e-06,
      "loss": 0.0008,
      "step": 2000
    },
    {
      "epoch": 3.2948929159802307,
      "eval_loss": 0.015332825481891632,
      "eval_runtime": 1629.67,
      "eval_samples_per_second": 1.489,
      "eval_steps_per_second": 0.187,
      "eval_wer": 3.24710634661695,
      "step": 2000
    }
  ],
  "logging_steps": 25,
  "max_steps": 6000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 1000,
  "total_flos": 9.23126978543616e+18,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}