{
  "best_metric": 7.122989865404904,
  "best_model_checkpoint": "./whisper-medium-mix-pt/checkpoint-5000",
  "epoch": 1.0622,
  "eval_steps": 1000,
  "global_step": 5000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.005,
      "grad_norm": 5.954268455505371,
      "learning_rate": 4.6000000000000004e-07,
      "loss": 0.7195,
      "step": 25
    },
    {
      "epoch": 0.01,
      "grad_norm": 6.020993232727051,
      "learning_rate": 9.600000000000001e-07,
      "loss": 0.5893,
      "step": 50
    },
    {
      "epoch": 0.015,
      "grad_norm": 4.6267008781433105,
      "learning_rate": 1.46e-06,
      "loss": 0.351,
      "step": 75
    },
    {
      "epoch": 0.02,
      "grad_norm": 5.107844829559326,
      "learning_rate": 1.9600000000000003e-06,
      "loss": 0.27,
      "step": 100
    },
    {
      "epoch": 0.025,
      "grad_norm": 4.793674468994141,
      "learning_rate": 2.46e-06,
      "loss": 0.5947,
      "step": 125
    },
    {
      "epoch": 0.03,
      "grad_norm": 5.168308734893799,
      "learning_rate": 2.96e-06,
      "loss": 0.634,
      "step": 150
    },
    {
      "epoch": 0.035,
      "grad_norm": 5.14207124710083,
      "learning_rate": 3.46e-06,
      "loss": 0.4872,
      "step": 175
    },
    {
      "epoch": 0.04,
      "grad_norm": 4.885250568389893,
      "learning_rate": 3.96e-06,
      "loss": 0.3443,
      "step": 200
    },
    {
      "epoch": 0.045,
      "grad_norm": 4.922640323638916,
      "learning_rate": 4.4600000000000005e-06,
      "loss": 0.3077,
      "step": 225
    },
    {
      "epoch": 0.05,
      "grad_norm": 3.2033262252807617,
      "learning_rate": 4.960000000000001e-06,
      "loss": 0.2798,
      "step": 250
    },
    {
      "epoch": 0.055,
      "grad_norm": 4.390773773193359,
      "learning_rate": 5.460000000000001e-06,
      "loss": 0.237,
      "step": 275
    },
    {
      "epoch": 0.06,
      "grad_norm": 4.187547206878662,
      "learning_rate": 5.9600000000000005e-06,
      "loss": 0.2354,
      "step": 300
    },
    {
      "epoch": 0.065,
      "grad_norm": 4.3109869956970215,
      "learning_rate": 6.460000000000001e-06,
      "loss": 0.2123,
      "step": 325
    },
    {
      "epoch": 0.07,
      "grad_norm": 3.2677078247070312,
      "learning_rate": 6.96e-06,
      "loss": 0.1923,
      "step": 350
    },
    {
      "epoch": 0.075,
      "grad_norm": 4.168946743011475,
      "learning_rate": 7.4600000000000006e-06,
      "loss": 0.204,
      "step": 375
    },
    {
      "epoch": 0.08,
      "grad_norm": 3.5255637168884277,
      "learning_rate": 7.960000000000002e-06,
      "loss": 0.1864,
      "step": 400
    },
    {
      "epoch": 0.085,
      "grad_norm": 2.9705564975738525,
      "learning_rate": 8.46e-06,
      "loss": 0.1935,
      "step": 425
    },
    {
      "epoch": 0.09,
      "grad_norm": 2.4820713996887207,
      "learning_rate": 8.96e-06,
      "loss": 0.1751,
      "step": 450
    },
    {
      "epoch": 0.095,
      "grad_norm": 3.700296401977539,
      "learning_rate": 9.460000000000001e-06,
      "loss": 0.1594,
      "step": 475
    },
    {
      "epoch": 0.1,
      "grad_norm": 3.0839247703552246,
      "learning_rate": 9.960000000000001e-06,
      "loss": 0.1709,
      "step": 500
    },
    {
      "epoch": 0.105,
      "grad_norm": 3.175724983215332,
      "learning_rate": 9.94888888888889e-06,
      "loss": 0.1246,
      "step": 525
    },
    {
      "epoch": 0.11,
      "grad_norm": 3.514580726623535,
      "learning_rate": 9.893333333333334e-06,
      "loss": 0.1323,
      "step": 550
    },
    {
      "epoch": 0.115,
      "grad_norm": 4.5611724853515625,
      "learning_rate": 9.837777777777778e-06,
      "loss": 0.2125,
      "step": 575
    },
    {
      "epoch": 0.12,
      "grad_norm": 3.8415772914886475,
      "learning_rate": 9.782222222222222e-06,
      "loss": 0.3451,
      "step": 600
    },
    {
      "epoch": 0.125,
      "grad_norm": 3.0463509559631348,
      "learning_rate": 9.726666666666668e-06,
      "loss": 0.205,
      "step": 625
    },
    {
      "epoch": 0.13,
      "grad_norm": 2.989981174468994,
      "learning_rate": 9.671111111111112e-06,
      "loss": 0.1433,
      "step": 650
    },
    {
      "epoch": 0.135,
      "grad_norm": 2.3929381370544434,
      "learning_rate": 9.615555555555558e-06,
      "loss": 0.1383,
      "step": 675
    },
    {
      "epoch": 0.14,
      "grad_norm": 2.7740297317504883,
      "learning_rate": 9.56e-06,
      "loss": 0.1288,
      "step": 700
    },
    {
      "epoch": 0.145,
      "grad_norm": 3.0236458778381348,
      "learning_rate": 9.504444444444446e-06,
      "loss": 0.1278,
      "step": 725
    },
    {
      "epoch": 0.15,
      "grad_norm": 4.738789081573486,
      "learning_rate": 9.44888888888889e-06,
      "loss": 0.1316,
      "step": 750
    },
    {
      "epoch": 0.155,
      "grad_norm": 2.820563316345215,
      "learning_rate": 9.393333333333334e-06,
      "loss": 0.1298,
      "step": 775
    },
    {
      "epoch": 0.16,
      "grad_norm": 3.1976284980773926,
      "learning_rate": 9.33777777777778e-06,
      "loss": 0.125,
      "step": 800
    },
    {
      "epoch": 0.165,
      "grad_norm": 3.4737789630889893,
      "learning_rate": 9.282222222222222e-06,
      "loss": 0.1437,
      "step": 825
    },
    {
      "epoch": 0.17,
      "grad_norm": 2.761700391769409,
      "learning_rate": 9.226666666666668e-06,
      "loss": 0.1285,
      "step": 850
    },
    {
      "epoch": 0.175,
      "grad_norm": 2.7969601154327393,
      "learning_rate": 9.171111111111112e-06,
      "loss": 0.1091,
      "step": 875
    },
    {
      "epoch": 0.18,
      "grad_norm": 3.8701090812683105,
      "learning_rate": 9.115555555555556e-06,
      "loss": 0.1383,
      "step": 900
    },
    {
      "epoch": 0.185,
      "grad_norm": 2.473555088043213,
      "learning_rate": 9.060000000000001e-06,
      "loss": 0.1067,
      "step": 925
    },
    {
      "epoch": 0.19,
      "grad_norm": 3.0442628860473633,
      "learning_rate": 9.004444444444445e-06,
      "loss": 0.1265,
      "step": 950
    },
    {
      "epoch": 0.195,
      "grad_norm": 3.2474780082702637,
      "learning_rate": 8.94888888888889e-06,
      "loss": 0.1254,
      "step": 975
    },
    {
      "epoch": 0.2,
      "grad_norm": 1.9548333883285522,
      "learning_rate": 8.893333333333333e-06,
      "loss": 0.1116,
      "step": 1000
    },
    {
      "epoch": 0.2,
      "eval_loss": 0.15698818862438202,
      "eval_runtime": 947.4383,
      "eval_samples_per_second": 9.992,
      "eval_steps_per_second": 1.25,
      "eval_wer": 8.582435499856134,
      "step": 1000
    },
    {
      "epoch": 0.205,
      "grad_norm": 2.7346816062927246,
      "learning_rate": 8.83777777777778e-06,
      "loss": 0.1005,
      "step": 1025
    },
    {
      "epoch": 0.21,
      "grad_norm": 3.2607483863830566,
      "learning_rate": 8.782222222222223e-06,
      "loss": 0.0931,
      "step": 1050
    },
    {
      "epoch": 0.215,
      "grad_norm": 2.8954336643218994,
      "learning_rate": 8.726666666666667e-06,
      "loss": 0.1111,
      "step": 1075
    },
    {
      "epoch": 0.22,
      "grad_norm": 3.1846327781677246,
      "learning_rate": 8.671111111111113e-06,
      "loss": 0.1003,
      "step": 1100
    },
    {
      "epoch": 0.225,
      "grad_norm": 2.284499168395996,
      "learning_rate": 8.615555555555555e-06,
      "loss": 0.1018,
      "step": 1125
    },
    {
      "epoch": 0.23,
      "grad_norm": 2.8060786724090576,
      "learning_rate": 8.560000000000001e-06,
      "loss": 0.1145,
      "step": 1150
    },
    {
      "epoch": 0.235,
      "grad_norm": 2.9943974018096924,
      "learning_rate": 8.504444444444445e-06,
      "loss": 0.1247,
      "step": 1175
    },
    {
      "epoch": 0.24,
      "grad_norm": 2.2265496253967285,
      "learning_rate": 8.448888888888889e-06,
      "loss": 0.1193,
      "step": 1200
    },
    {
      "epoch": 0.245,
      "grad_norm": 2.4557271003723145,
      "learning_rate": 8.393333333333335e-06,
      "loss": 0.1209,
      "step": 1225
    },
    {
      "epoch": 0.25,
      "grad_norm": 2.332017183303833,
      "learning_rate": 8.337777777777777e-06,
      "loss": 0.114,
      "step": 1250
    },
    {
      "epoch": 0.255,
      "grad_norm": 3.056497812271118,
      "learning_rate": 8.282222222222223e-06,
      "loss": 0.1043,
      "step": 1275
    },
    {
      "epoch": 0.26,
      "grad_norm": 3.024203062057495,
      "learning_rate": 8.226666666666667e-06,
      "loss": 0.1247,
      "step": 1300
    },
    {
      "epoch": 0.265,
      "grad_norm": 2.543893814086914,
      "learning_rate": 8.171111111111113e-06,
      "loss": 0.1347,
      "step": 1325
    },
    {
      "epoch": 0.27,
      "grad_norm": 2.2330307960510254,
      "learning_rate": 8.115555555555557e-06,
      "loss": 0.119,
      "step": 1350
    },
    {
      "epoch": 0.275,
      "grad_norm": 2.819495677947998,
      "learning_rate": 8.06e-06,
      "loss": 0.0998,
      "step": 1375
    },
    {
      "epoch": 0.28,
      "grad_norm": 2.6235690116882324,
      "learning_rate": 8.004444444444445e-06,
      "loss": 0.1037,
      "step": 1400
    },
    {
      "epoch": 0.285,
      "grad_norm": 3.0939559936523438,
      "learning_rate": 7.948888888888889e-06,
      "loss": 0.1027,
      "step": 1425
    },
    {
      "epoch": 0.29,
      "grad_norm": 2.9563097953796387,
      "learning_rate": 7.893333333333335e-06,
      "loss": 0.106,
      "step": 1450
    },
    {
      "epoch": 0.295,
      "grad_norm": 2.6161818504333496,
      "learning_rate": 7.837777777777779e-06,
      "loss": 0.1112,
      "step": 1475
    },
    {
      "epoch": 0.3,
      "grad_norm": 2.455970525741577,
      "learning_rate": 7.782222222222223e-06,
      "loss": 0.1138,
      "step": 1500
    },
    {
      "epoch": 0.305,
      "grad_norm": 2.8548126220703125,
      "learning_rate": 7.726666666666667e-06,
      "loss": 0.0962,
      "step": 1525
    },
    {
      "epoch": 0.31,
      "grad_norm": 2.414487838745117,
      "learning_rate": 7.67111111111111e-06,
      "loss": 0.105,
      "step": 1550
    },
    {
      "epoch": 0.315,
      "grad_norm": 2.747642993927002,
      "learning_rate": 7.6155555555555564e-06,
      "loss": 0.1111,
      "step": 1575
    },
    {
      "epoch": 0.32,
      "grad_norm": 2.5105972290039062,
      "learning_rate": 7.5600000000000005e-06,
      "loss": 0.1066,
      "step": 1600
    },
    {
      "epoch": 0.325,
      "grad_norm": 2.765288829803467,
      "learning_rate": 7.504444444444445e-06,
      "loss": 0.1423,
      "step": 1625
    },
    {
      "epoch": 0.33,
      "grad_norm": 3.616039991378784,
      "learning_rate": 7.44888888888889e-06,
      "loss": 0.1527,
      "step": 1650
    },
    {
      "epoch": 0.335,
      "grad_norm": 3.961867094039917,
      "learning_rate": 7.393333333333333e-06,
      "loss": 0.1409,
      "step": 1675
    },
    {
      "epoch": 0.34,
      "grad_norm": 3.5502121448516846,
      "learning_rate": 7.337777777777778e-06,
      "loss": 0.1375,
      "step": 1700
    },
    {
      "epoch": 0.345,
      "grad_norm": 2.9982211589813232,
      "learning_rate": 7.282222222222222e-06,
      "loss": 0.1208,
      "step": 1725
    },
    {
      "epoch": 0.35,
      "grad_norm": 3.1248340606689453,
      "learning_rate": 7.226666666666667e-06,
      "loss": 0.1375,
      "step": 1750
    },
    {
      "epoch": 0.355,
      "grad_norm": 2.1703038215637207,
      "learning_rate": 7.171111111111112e-06,
      "loss": 0.1513,
      "step": 1775
    },
    {
      "epoch": 0.36,
      "grad_norm": 2.672173500061035,
      "learning_rate": 7.115555555555557e-06,
      "loss": 0.1169,
      "step": 1800
    },
    {
      "epoch": 0.365,
      "grad_norm": 2.1961538791656494,
      "learning_rate": 7.06e-06,
      "loss": 0.1072,
      "step": 1825
    },
    {
      "epoch": 0.37,
      "grad_norm": 2.129289388656616,
      "learning_rate": 7.004444444444445e-06,
      "loss": 0.1022,
      "step": 1850
    },
    {
      "epoch": 0.375,
      "grad_norm": 2.4798481464385986,
      "learning_rate": 6.948888888888889e-06,
      "loss": 0.1129,
      "step": 1875
    },
    {
      "epoch": 0.38,
      "grad_norm": 2.9369380474090576,
      "learning_rate": 6.893333333333334e-06,
      "loss": 0.1149,
      "step": 1900
    },
    {
      "epoch": 0.385,
      "grad_norm": 2.7497618198394775,
      "learning_rate": 6.837777777777779e-06,
      "loss": 0.1181,
      "step": 1925
    },
    {
      "epoch": 0.39,
      "grad_norm": 2.4043569564819336,
      "learning_rate": 6.782222222222222e-06,
      "loss": 0.0998,
      "step": 1950
    },
    {
      "epoch": 0.395,
      "grad_norm": 2.9279298782348633,
      "learning_rate": 6.726666666666667e-06,
      "loss": 0.1035,
      "step": 1975
    },
    {
      "epoch": 0.4,
      "grad_norm": 2.7362992763519287,
      "learning_rate": 6.671111111111112e-06,
      "loss": 0.105,
      "step": 2000
    },
    {
      "epoch": 0.4,
      "eval_loss": 0.14844202995300293,
      "eval_runtime": 940.7954,
      "eval_samples_per_second": 10.063,
      "eval_steps_per_second": 1.259,
      "eval_wer": 7.93983183605614,
      "step": 2000
    },
    {
      "epoch": 0.405,
      "grad_norm": 2.0498945713043213,
      "learning_rate": 6.615555555555556e-06,
      "loss": 0.1086,
      "step": 2025
    },
    {
      "epoch": 0.41,
      "grad_norm": 2.7259154319763184,
      "learning_rate": 6.560000000000001e-06,
      "loss": 0.1176,
      "step": 2050
    },
    {
      "epoch": 0.415,
      "grad_norm": 2.4379186630249023,
      "learning_rate": 6.504444444444446e-06,
      "loss": 0.1157,
      "step": 2075
    },
    {
      "epoch": 0.42,
      "grad_norm": 2.397427558898926,
      "learning_rate": 6.448888888888889e-06,
      "loss": 0.1229,
      "step": 2100
    },
    {
      "epoch": 0.425,
      "grad_norm": 2.2779042720794678,
      "learning_rate": 6.393333333333334e-06,
      "loss": 0.0934,
      "step": 2125
    },
    {
      "epoch": 0.43,
      "grad_norm": 1.9669123888015747,
      "learning_rate": 6.3377777777777786e-06,
      "loss": 0.0968,
      "step": 2150
    },
    {
      "epoch": 0.435,
      "grad_norm": 2.051048755645752,
      "learning_rate": 6.282222222222223e-06,
      "loss": 0.0876,
      "step": 2175
    },
    {
      "epoch": 0.44,
      "grad_norm": 2.082129716873169,
      "learning_rate": 6.2266666666666675e-06,
      "loss": 0.0832,
      "step": 2200
    },
    {
      "epoch": 0.445,
      "grad_norm": 2.0666728019714355,
      "learning_rate": 6.171111111111112e-06,
      "loss": 0.081,
      "step": 2225
    },
    {
      "epoch": 0.45,
      "grad_norm": 2.1512069702148438,
      "learning_rate": 6.1155555555555555e-06,
      "loss": 0.0795,
      "step": 2250
    },
    {
      "epoch": 0.455,
      "grad_norm": 2.393537998199463,
      "learning_rate": 6.0600000000000004e-06,
      "loss": 0.0797,
      "step": 2275
    },
    {
      "epoch": 0.46,
      "grad_norm": 2.4059464931488037,
      "learning_rate": 6.004444444444445e-06,
      "loss": 0.086,
      "step": 2300
    },
    {
      "epoch": 0.465,
      "grad_norm": 2.3822548389434814,
      "learning_rate": 5.948888888888889e-06,
      "loss": 0.0883,
      "step": 2325
    },
    {
      "epoch": 0.47,
      "grad_norm": 2.063124656677246,
      "learning_rate": 5.893333333333334e-06,
      "loss": 0.082,
      "step": 2350
    },
    {
      "epoch": 0.475,
      "grad_norm": 2.988023281097412,
      "learning_rate": 5.837777777777777e-06,
      "loss": 0.194,
      "step": 2375
    },
    {
      "epoch": 0.48,
      "grad_norm": 3.959836959838867,
      "learning_rate": 5.782222222222222e-06,
      "loss": 0.2993,
      "step": 2400
    },
    {
      "epoch": 0.485,
      "grad_norm": 2.79730486869812,
      "learning_rate": 5.726666666666667e-06,
      "loss": 0.2067,
      "step": 2425
    },
    {
      "epoch": 0.49,
      "grad_norm": 1.7770684957504272,
      "learning_rate": 5.671111111111112e-06,
      "loss": 0.1769,
      "step": 2450
    },
    {
      "epoch": 0.495,
      "grad_norm": 2.2204787731170654,
      "learning_rate": 5.615555555555556e-06,
      "loss": 0.0911,
      "step": 2475
    },
    {
      "epoch": 0.5,
      "grad_norm": 3.053565263748169,
      "learning_rate": 5.560000000000001e-06,
      "loss": 0.1141,
      "step": 2500
    },
    {
      "epoch": 0.505,
      "grad_norm": 5.307312488555908,
      "learning_rate": 5.504444444444444e-06,
      "loss": 0.4682,
      "step": 2525
    },
    {
      "epoch": 0.51,
      "grad_norm": 4.99273157119751,
      "learning_rate": 5.448888888888889e-06,
      "loss": 0.4198,
      "step": 2550
    },
    {
      "epoch": 0.515,
      "grad_norm": 4.5542707443237305,
      "learning_rate": 5.393333333333334e-06,
      "loss": 0.3805,
      "step": 2575
    },
    {
      "epoch": 0.52,
      "grad_norm": 3.3522355556488037,
      "learning_rate": 5.337777777777779e-06,
      "loss": 0.3213,
      "step": 2600
    },
    {
      "epoch": 0.525,
      "grad_norm": 3.8808069229125977,
      "learning_rate": 5.282222222222223e-06,
      "loss": 0.2504,
      "step": 2625
    },
    {
      "epoch": 0.53,
      "grad_norm": 3.6517879962921143,
      "learning_rate": 5.226666666666667e-06,
      "loss": 0.2508,
      "step": 2650
    },
    {
      "epoch": 0.535,
      "grad_norm": 5.0575151443481445,
      "learning_rate": 5.171111111111111e-06,
      "loss": 0.288,
      "step": 2675
    },
    {
      "epoch": 0.54,
      "grad_norm": 4.049450874328613,
      "learning_rate": 5.115555555555556e-06,
      "loss": 0.2526,
      "step": 2700
    },
    {
      "epoch": 0.545,
      "grad_norm": 3.075061798095703,
      "learning_rate": 5.060000000000001e-06,
      "loss": 0.25,
      "step": 2725
    },
    {
      "epoch": 0.55,
      "grad_norm": 3.80968976020813,
      "learning_rate": 5.004444444444445e-06,
      "loss": 0.2388,
      "step": 2750
    },
    {
      "epoch": 0.555,
      "grad_norm": 3.044834852218628,
      "learning_rate": 4.94888888888889e-06,
      "loss": 0.1718,
      "step": 2775
    },
    {
      "epoch": 0.56,
      "grad_norm": 2.3228962421417236,
      "learning_rate": 4.893333333333334e-06,
      "loss": 0.148,
      "step": 2800
    },
    {
      "epoch": 0.565,
      "grad_norm": 2.9311017990112305,
      "learning_rate": 4.837777777777778e-06,
      "loss": 0.1308,
      "step": 2825
    },
    {
      "epoch": 0.57,
      "grad_norm": 2.485396146774292,
      "learning_rate": 4.7822222222222226e-06,
      "loss": 0.1197,
      "step": 2850
    },
    {
      "epoch": 0.575,
      "grad_norm": 3.5410890579223633,
      "learning_rate": 4.7266666666666674e-06,
      "loss": 0.0878,
      "step": 2875
    },
    {
      "epoch": 0.58,
      "grad_norm": 2.246903657913208,
      "learning_rate": 4.6711111111111115e-06,
      "loss": 0.0793,
      "step": 2900
    },
    {
      "epoch": 0.585,
      "grad_norm": 2.251746416091919,
      "learning_rate": 4.6155555555555555e-06,
      "loss": 0.0795,
      "step": 2925
    },
    {
      "epoch": 0.59,
      "grad_norm": 1.791506052017212,
      "learning_rate": 4.56e-06,
      "loss": 0.0918,
      "step": 2950
    },
    {
      "epoch": 0.595,
      "grad_norm": 1.3495018482208252,
      "learning_rate": 4.504444444444444e-06,
      "loss": 0.077,
      "step": 2975
    },
    {
      "epoch": 0.6,
      "grad_norm": 1.8840142488479614,
      "learning_rate": 4.448888888888889e-06,
      "loss": 0.0783,
      "step": 3000
    },
    {
      "epoch": 0.6,
      "eval_loss": 0.13742496073246002,
      "eval_runtime": 944.2672,
      "eval_samples_per_second": 10.026,
      "eval_steps_per_second": 1.254,
      "eval_wer": 7.447488730458135,
      "step": 3000
    },
    {
      "epoch": 0.605,
      "grad_norm": 1.7588751316070557,
      "learning_rate": 4.393333333333334e-06,
      "loss": 0.0747,
      "step": 3025
    },
    {
      "epoch": 0.61,
      "grad_norm": 1.9886692762374878,
      "learning_rate": 4.337777777777778e-06,
      "loss": 0.076,
      "step": 3050
    },
    {
      "epoch": 0.615,
      "grad_norm": 2.5392279624938965,
      "learning_rate": 4.282222222222222e-06,
      "loss": 0.0796,
      "step": 3075
    },
    {
      "epoch": 0.62,
      "grad_norm": 2.3111472129821777,
      "learning_rate": 4.226666666666667e-06,
      "loss": 0.0898,
      "step": 3100
    },
    {
      "epoch": 0.625,
      "grad_norm": 1.8798396587371826,
      "learning_rate": 4.171111111111111e-06,
      "loss": 0.0742,
      "step": 3125
    },
    {
      "epoch": 0.63,
      "grad_norm": 1.8901393413543701,
      "learning_rate": 4.115555555555556e-06,
      "loss": 0.0815,
      "step": 3150
    },
    {
      "epoch": 0.635,
      "grad_norm": 3.506927490234375,
      "learning_rate": 4.060000000000001e-06,
      "loss": 0.1055,
      "step": 3175
    },
    {
      "epoch": 0.64,
      "grad_norm": 3.018944025039673,
      "learning_rate": 4.004444444444445e-06,
      "loss": 0.1541,
      "step": 3200
    },
    {
      "epoch": 0.645,
      "grad_norm": 2.304964303970337,
      "learning_rate": 3.948888888888889e-06,
      "loss": 0.1213,
      "step": 3225
    },
    {
      "epoch": 0.65,
      "grad_norm": 2.425555944442749,
      "learning_rate": 3.893333333333333e-06,
      "loss": 0.0935,
      "step": 3250
    },
    {
      "epoch": 0.655,
      "grad_norm": 1.5498875379562378,
      "learning_rate": 3.837777777777778e-06,
      "loss": 0.0838,
      "step": 3275
    },
    {
      "epoch": 0.66,
      "grad_norm": 2.360494613647461,
      "learning_rate": 3.782222222222223e-06,
      "loss": 0.0777,
      "step": 3300
    },
    {
      "epoch": 0.665,
      "grad_norm": 1.5479514598846436,
      "learning_rate": 3.726666666666667e-06,
      "loss": 0.0778,
      "step": 3325
    },
    {
      "epoch": 0.67,
      "grad_norm": 2.920868158340454,
      "learning_rate": 3.6711111111111113e-06,
      "loss": 0.0837,
      "step": 3350
    },
    {
      "epoch": 0.675,
      "grad_norm": 1.8797603845596313,
      "learning_rate": 3.615555555555556e-06,
      "loss": 0.0727,
      "step": 3375
    },
    {
      "epoch": 0.68,
      "grad_norm": 2.5125744342803955,
      "learning_rate": 3.5600000000000002e-06,
      "loss": 0.0798,
      "step": 3400
    },
    {
      "epoch": 0.685,
      "grad_norm": 2.6805381774902344,
      "learning_rate": 3.5044444444444447e-06,
      "loss": 0.0825,
      "step": 3425
    },
    {
      "epoch": 0.69,
      "grad_norm": 2.2335715293884277,
      "learning_rate": 3.4488888888888896e-06,
      "loss": 0.1145,
      "step": 3450
    },
    {
      "epoch": 0.695,
      "grad_norm": 3.3753836154937744,
      "learning_rate": 3.3933333333333336e-06,
      "loss": 0.1083,
      "step": 3475
    },
    {
      "epoch": 0.7,
      "grad_norm": 3.470496654510498,
      "learning_rate": 3.337777777777778e-06,
      "loss": 0.3921,
      "step": 3500
    },
    {
      "epoch": 0.705,
      "grad_norm": 1.5432311296463013,
      "learning_rate": 3.282222222222223e-06,
      "loss": 0.161,
      "step": 3525
    },
    {
      "epoch": 0.71,
      "grad_norm": 2.0430617332458496,
      "learning_rate": 3.226666666666667e-06,
      "loss": 0.1171,
      "step": 3550
    },
    {
      "epoch": 0.715,
      "grad_norm": 1.5512393712997437,
      "learning_rate": 3.1711111111111114e-06,
      "loss": 0.08,
      "step": 3575
    },
    {
      "epoch": 0.72,
      "grad_norm": 1.4717549085617065,
      "learning_rate": 3.1155555555555555e-06,
      "loss": 0.0726,
      "step": 3600
    },
    {
      "epoch": 0.725,
      "grad_norm": 1.5445371866226196,
      "learning_rate": 3.0600000000000003e-06,
      "loss": 0.0589,
      "step": 3625
    },
    {
      "epoch": 0.73,
      "grad_norm": 1.693588376045227,
      "learning_rate": 3.004444444444445e-06,
      "loss": 0.0606,
      "step": 3650
    },
    {
      "epoch": 0.735,
      "grad_norm": 1.875542163848877,
      "learning_rate": 2.948888888888889e-06,
      "loss": 0.0548,
      "step": 3675
    },
    {
      "epoch": 0.74,
      "grad_norm": 1.2488154172897339,
      "learning_rate": 2.8933333333333337e-06,
      "loss": 0.0517,
      "step": 3700
    },
    {
      "epoch": 0.745,
      "grad_norm": 1.5682026147842407,
      "learning_rate": 2.837777777777778e-06,
      "loss": 0.0571,
      "step": 3725
    },
    {
      "epoch": 0.75,
      "grad_norm": 1.1433475017547607,
      "learning_rate": 2.7822222222222222e-06,
      "loss": 0.0513,
      "step": 3750
    },
    {
      "epoch": 0.755,
      "grad_norm": 1.7756175994873047,
      "learning_rate": 2.726666666666667e-06,
      "loss": 0.047,
      "step": 3775
    },
    {
      "epoch": 0.76,
      "grad_norm": 1.6270946264266968,
      "learning_rate": 2.6711111111111116e-06,
      "loss": 0.0528,
      "step": 3800
    },
    {
      "epoch": 0.765,
      "grad_norm": 1.3709731101989746,
      "learning_rate": 2.6155555555555556e-06,
      "loss": 0.0507,
      "step": 3825
    },
    {
      "epoch": 0.77,
      "grad_norm": 1.4345024824142456,
      "learning_rate": 2.56e-06,
      "loss": 0.0455,
      "step": 3850
    },
    {
      "epoch": 0.775,
      "grad_norm": 2.159369707107544,
      "learning_rate": 2.504444444444445e-06,
      "loss": 0.0548,
      "step": 3875
    },
    {
      "epoch": 0.78,
      "grad_norm": 2.050881862640381,
      "learning_rate": 2.448888888888889e-06,
      "loss": 0.0833,
      "step": 3900
    },
    {
      "epoch": 0.785,
      "grad_norm": 2.428863763809204,
      "learning_rate": 2.3933333333333334e-06,
      "loss": 0.1491,
      "step": 3925
    },
    {
      "epoch": 0.79,
      "grad_norm": 2.655684471130371,
      "learning_rate": 2.337777777777778e-06,
      "loss": 0.1254,
      "step": 3950
    },
    {
      "epoch": 0.795,
      "grad_norm": 1.752144455909729,
      "learning_rate": 2.2822222222222223e-06,
      "loss": 0.1352,
      "step": 3975
    },
    {
      "epoch": 0.8,
      "grad_norm": 2.89431095123291,
      "learning_rate": 2.226666666666667e-06,
      "loss": 0.1703,
      "step": 4000
    },
    {
      "epoch": 0.8,
      "eval_loss": 0.13704591989517212,
      "eval_runtime": 942.2184,
      "eval_samples_per_second": 10.048,
      "eval_steps_per_second": 1.257,
      "eval_wer": 7.241280092074555,
      "step": 4000
    },
    {
      "epoch": 0.805,
      "grad_norm": 2.1112782955169678,
      "learning_rate": 2.1711111111111113e-06,
      "loss": 0.137,
      "step": 4025
    },
    {
      "epoch": 0.81,
      "grad_norm": 2.241307258605957,
      "learning_rate": 2.1155555555555557e-06,
      "loss": 0.0809,
      "step": 4050
    },
    {
      "epoch": 0.815,
      "grad_norm": 1.5932738780975342,
      "learning_rate": 2.06e-06,
      "loss": 0.0605,
      "step": 4075
    },
    {
      "epoch": 0.82,
      "grad_norm": 1.965742588043213,
      "learning_rate": 2.006666666666667e-06,
      "loss": 0.0549,
      "step": 4100
    },
    {
      "epoch": 0.825,
      "grad_norm": 3.738175868988037,
      "learning_rate": 1.9511111111111113e-06,
      "loss": 0.0557,
      "step": 4125
    },
    {
      "epoch": 0.83,
      "grad_norm": 1.9460804462432861,
      "learning_rate": 1.8955555555555557e-06,
      "loss": 0.0433,
      "step": 4150
    },
    {
      "epoch": 0.835,
      "grad_norm": 1.6488566398620605,
      "learning_rate": 1.8400000000000002e-06,
      "loss": 0.0655,
      "step": 4175
    },
    {
      "epoch": 0.84,
      "grad_norm": 1.511797308921814,
      "learning_rate": 1.7844444444444444e-06,
      "loss": 0.0787,
      "step": 4200
    },
    {
      "epoch": 0.845,
      "grad_norm": 2.361323833465576,
      "learning_rate": 1.728888888888889e-06,
      "loss": 0.0685,
      "step": 4225
    },
    {
      "epoch": 0.85,
      "grad_norm": 1.1696654558181763,
      "learning_rate": 1.6733333333333335e-06,
      "loss": 0.0518,
      "step": 4250
    },
    {
      "epoch": 0.855,
      "grad_norm": 1.6606045961380005,
      "learning_rate": 1.6177777777777778e-06,
      "loss": 0.0509,
      "step": 4275
    },
    {
      "epoch": 0.86,
      "grad_norm": 1.3564594984054565,
      "learning_rate": 1.5622222222222225e-06,
      "loss": 0.0545,
      "step": 4300
    },
    {
      "epoch": 0.865,
      "grad_norm": 2.545974016189575,
      "learning_rate": 1.506666666666667e-06,
      "loss": 0.0953,
      "step": 4325
    },
    {
      "epoch": 0.87,
      "grad_norm": 5.036618232727051,
      "learning_rate": 1.4511111111111112e-06,
      "loss": 0.1465,
      "step": 4350
    },
    {
      "epoch": 0.875,
      "grad_norm": 3.5834786891937256,
      "learning_rate": 1.3955555555555556e-06,
      "loss": 0.4963,
      "step": 4375
    },
    {
      "epoch": 0.88,
      "grad_norm": 4.646224498748779,
      "learning_rate": 1.34e-06,
      "loss": 0.5522,
      "step": 4400
    },
    {
      "epoch": 0.885,
      "grad_norm": 4.311796188354492,
      "learning_rate": 1.2844444444444445e-06,
      "loss": 0.585,
      "step": 4425
    },
    {
      "epoch": 0.89,
      "grad_norm": 2.739590883255005,
      "learning_rate": 1.228888888888889e-06,
      "loss": 0.5151,
      "step": 4450
    },
    {
      "epoch": 0.895,
      "grad_norm": 2.0715537071228027,
      "learning_rate": 1.1733333333333335e-06,
      "loss": 0.2399,
      "step": 4475
    },
    {
      "epoch": 0.9,
      "grad_norm": 1.7919174432754517,
      "learning_rate": 1.117777777777778e-06,
      "loss": 0.1013,
      "step": 4500
    },
    {
      "epoch": 0.905,
      "grad_norm": 1.8934131860733032,
      "learning_rate": 1.0622222222222222e-06,
      "loss": 0.0792,
      "step": 4525
    },
    {
      "epoch": 0.91,
      "grad_norm": 1.463447093963623,
      "learning_rate": 1.0066666666666668e-06,
      "loss": 0.0594,
      "step": 4550
    },
    {
      "epoch": 0.915,
      "grad_norm": 2.0955190658569336,
      "learning_rate": 9.511111111111111e-07,
      "loss": 0.0641,
      "step": 4575
    },
    {
      "epoch": 0.92,
      "grad_norm": 1.8596012592315674,
      "learning_rate": 8.955555555555557e-07,
      "loss": 0.065,
      "step": 4600
    },
    {
      "epoch": 0.925,
      "grad_norm": 2.5963196754455566,
      "learning_rate": 8.400000000000001e-07,
      "loss": 0.0604,
      "step": 4625
    },
    {
      "epoch": 0.93,
      "grad_norm": 1.361854076385498,
      "learning_rate": 7.844444444444445e-07,
      "loss": 0.0695,
      "step": 4650
    },
    {
      "epoch": 0.935,
      "grad_norm": 1.8713892698287964,
      "learning_rate": 7.28888888888889e-07,
      "loss": 0.0501,
      "step": 4675
    },
    {
      "epoch": 1.0022,
      "grad_norm": 1.7594069242477417,
      "learning_rate": 6.755555555555555e-07,
      "loss": 0.0811,
      "step": 4700
    },
    {
      "epoch": 1.0072,
      "grad_norm": 2.2666430473327637,
      "learning_rate": 6.200000000000001e-07,
      "loss": 0.0794,
      "step": 4725
    },
    {
      "epoch": 1.0122,
      "grad_norm": 1.4609644412994385,
      "learning_rate": 5.644444444444445e-07,
      "loss": 0.0913,
      "step": 4750
    },
    {
      "epoch": 1.0172,
      "grad_norm": 2.5821077823638916,
      "learning_rate": 5.088888888888889e-07,
      "loss": 0.0743,
      "step": 4775
    },
    {
      "epoch": 1.0222,
      "grad_norm": 3.746551036834717,
      "learning_rate": 4.533333333333334e-07,
      "loss": 0.2048,
      "step": 4800
    },
    {
      "epoch": 1.0272,
      "grad_norm": 4.535083293914795,
      "learning_rate": 3.9777777777777783e-07,
      "loss": 0.4846,
      "step": 4825
    },
    {
      "epoch": 1.0322,
      "grad_norm": 4.492420196533203,
      "learning_rate": 3.422222222222223e-07,
      "loss": 0.3757,
      "step": 4850
    },
    {
      "epoch": 1.0372,
      "grad_norm": 3.43796706199646,
      "learning_rate": 2.866666666666667e-07,
      "loss": 0.2405,
      "step": 4875
    },
    {
      "epoch": 1.0422,
      "grad_norm": 3.234219551086426,
      "learning_rate": 2.3111111111111112e-07,
      "loss": 0.1906,
      "step": 4900
    },
    {
      "epoch": 1.0472,
      "grad_norm": 2.204688549041748,
      "learning_rate": 1.7555555555555558e-07,
      "loss": 0.1335,
      "step": 4925
    },
    {
      "epoch": 1.0522,
      "grad_norm": 3.1717331409454346,
      "learning_rate": 1.2000000000000002e-07,
      "loss": 0.1198,
      "step": 4950
    },
    {
      "epoch": 1.0572,
      "grad_norm": 2.4640631675720215,
      "learning_rate": 6.444444444444445e-08,
      "loss": 0.106,
      "step": 4975
    },
    {
      "epoch": 1.0622,
      "grad_norm": 2.141594648361206,
      "learning_rate": 8.88888888888889e-09,
      "loss": 0.0977,
      "step": 5000
    },
    {
      "epoch": 1.0622,
      "eval_loss": 0.13527874648571014,
      "eval_runtime": 940.5129,
      "eval_samples_per_second": 10.066,
      "eval_steps_per_second": 1.259,
      "eval_wer": 7.122989865404904,
      "step": 5000
    },
    {
      "epoch": 1.0622,
      "step": 5000,
      "total_flos": 1.63271270080512e+20,
      "train_loss": 0.15365941531658173,
      "train_runtime": 18539.6479,
      "train_samples_per_second": 8.63,
      "train_steps_per_second": 0.27
    }
  ],
  "logging_steps": 25,
  "max_steps": 5000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9223372036854775807,
  "save_steps": 1000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.63271270080512e+20,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}