{
  "best_global_step": 19000,
  "best_metric": 0.1748135060855909,
  "best_model_checkpoint": "/home/cluster-dgxa100/slp01/bagas-fine-tune-whisper/whisper-tiny-javanese-openslr-v8/checkpoint-19000",
  "epoch": 32.0,
  "eval_steps": 1000,
  "global_step": 20000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.32,
      "grad_norm": 19.961477279663086,
      "learning_rate": 1.97e-06,
      "loss": 3.217,
      "step": 200
    },
    {
      "epoch": 0.64,
      "grad_norm": 16.765470504760742,
      "learning_rate": 3.97e-06,
      "loss": 1.6255,
      "step": 400
    },
    {
      "epoch": 0.96,
      "grad_norm": 16.053678512573242,
      "learning_rate": 5.9700000000000004e-06,
      "loss": 1.1789,
      "step": 600
    },
    {
      "epoch": 1.28,
      "grad_norm": 14.488689422607422,
      "learning_rate": 7.970000000000002e-06,
      "loss": 0.9348,
      "step": 800
    },
    {
      "epoch": 1.6,
      "grad_norm": 12.85074234008789,
      "learning_rate": 9.970000000000001e-06,
      "loss": 0.8116,
      "step": 1000
    },
    {
      "epoch": 1.6,
      "eval_loss": 0.6225642561912537,
      "eval_runtime": 205.3808,
      "eval_samples_per_second": 6.086,
      "eval_steps_per_second": 1.524,
      "eval_wer": 0.47516686297605026,
      "step": 1000
    },
    {
      "epoch": 1.92,
      "grad_norm": 14.117347717285156,
      "learning_rate": 1.1970000000000002e-05,
      "loss": 0.757,
      "step": 1200
    },
    {
      "epoch": 2.24,
      "grad_norm": 11.961180686950684,
      "learning_rate": 1.3970000000000002e-05,
      "loss": 0.618,
      "step": 1400
    },
    {
      "epoch": 2.56,
      "grad_norm": 11.46937370300293,
      "learning_rate": 1.597e-05,
      "loss": 0.5826,
      "step": 1600
    },
    {
      "epoch": 2.88,
      "grad_norm": 10.750535011291504,
      "learning_rate": 1.7970000000000002e-05,
      "loss": 0.5615,
      "step": 1800
    },
    {
      "epoch": 3.2,
      "grad_norm": 12.116422653198242,
      "learning_rate": 1.9970000000000004e-05,
      "loss": 0.4683,
      "step": 2000
    },
    {
      "epoch": 3.2,
      "eval_loss": 0.41659122705459595,
      "eval_runtime": 206.2639,
      "eval_samples_per_second": 6.06,
      "eval_steps_per_second": 1.517,
      "eval_wer": 0.41647035728307813,
      "step": 2000
    },
    {
      "epoch": 3.52,
      "grad_norm": 9.442646980285645,
      "learning_rate": 1.978111111111111e-05,
      "loss": 0.4231,
      "step": 2200
    },
    {
      "epoch": 3.84,
      "grad_norm": 13.404372215270996,
      "learning_rate": 1.955888888888889e-05,
      "loss": 0.4156,
      "step": 2400
    },
    {
      "epoch": 4.16,
      "grad_norm": 9.753326416015625,
      "learning_rate": 1.9336666666666667e-05,
      "loss": 0.3613,
      "step": 2600
    },
    {
      "epoch": 4.48,
      "grad_norm": 10.980928421020508,
      "learning_rate": 1.9114444444444447e-05,
      "loss": 0.3039,
      "step": 2800
    },
    {
      "epoch": 4.8,
      "grad_norm": 11.34424114227295,
      "learning_rate": 1.8892222222222223e-05,
      "loss": 0.3047,
      "step": 3000
    },
    {
      "epoch": 4.8,
      "eval_loss": 0.34299781918525696,
      "eval_runtime": 205.0374,
      "eval_samples_per_second": 6.096,
      "eval_steps_per_second": 1.527,
      "eval_wer": 0.3496270121711818,
      "step": 3000
    },
    {
      "epoch": 5.12,
      "grad_norm": 7.82305383682251,
      "learning_rate": 1.8670000000000003e-05,
      "loss": 0.2757,
      "step": 3200
    },
    {
      "epoch": 5.44,
      "grad_norm": 7.7784576416015625,
      "learning_rate": 1.844777777777778e-05,
      "loss": 0.2217,
      "step": 3400
    },
    {
      "epoch": 5.76,
      "grad_norm": 12.481864929199219,
      "learning_rate": 1.8225555555555555e-05,
      "loss": 0.2279,
      "step": 3600
    },
    {
      "epoch": 6.08,
      "grad_norm": 6.796186447143555,
      "learning_rate": 1.8003333333333334e-05,
      "loss": 0.2183,
      "step": 3800
    },
    {
      "epoch": 6.4,
      "grad_norm": 5.8489460945129395,
      "learning_rate": 1.7781111111111114e-05,
      "loss": 0.1679,
      "step": 4000
    },
    {
      "epoch": 6.4,
      "eval_loss": 0.31804612278938293,
      "eval_runtime": 201.9885,
      "eval_samples_per_second": 6.188,
      "eval_steps_per_second": 1.55,
      "eval_wer": 0.24676089517078917,
      "step": 4000
    },
    {
      "epoch": 6.72,
      "grad_norm": 8.320049285888672,
      "learning_rate": 1.755888888888889e-05,
      "loss": 0.1735,
      "step": 4200
    },
    {
      "epoch": 7.04,
      "grad_norm": 6.053928375244141,
      "learning_rate": 1.733777777777778e-05,
      "loss": 0.1708,
      "step": 4400
    },
    {
      "epoch": 7.36,
      "grad_norm": 13.38781452178955,
      "learning_rate": 1.7115555555555557e-05,
      "loss": 0.1268,
      "step": 4600
    },
    {
      "epoch": 7.68,
      "grad_norm": 7.869670391082764,
      "learning_rate": 1.6893333333333336e-05,
      "loss": 0.1355,
      "step": 4800
    },
    {
      "epoch": 8.0,
      "grad_norm": 6.039541721343994,
      "learning_rate": 1.6671111111111113e-05,
      "loss": 0.1405,
      "step": 5000
    },
    {
      "epoch": 8.0,
      "eval_loss": 0.28736841678619385,
      "eval_runtime": 201.5675,
      "eval_samples_per_second": 6.201,
      "eval_steps_per_second": 1.553,
      "eval_wer": 0.21181782489202983,
      "step": 5000
    },
    {
      "epoch": 8.32,
      "grad_norm": 4.737486362457275,
      "learning_rate": 1.644888888888889e-05,
      "loss": 0.1042,
      "step": 5200
    },
    {
      "epoch": 8.64,
      "grad_norm": 8.452003479003906,
      "learning_rate": 1.6226666666666668e-05,
      "loss": 0.1113,
      "step": 5400
    },
    {
      "epoch": 8.96,
      "grad_norm": 4.064589500427246,
      "learning_rate": 1.6004444444444444e-05,
      "loss": 0.1094,
      "step": 5600
    },
    {
      "epoch": 9.28,
      "grad_norm": 7.312466621398926,
      "learning_rate": 1.5782222222222224e-05,
      "loss": 0.0854,
      "step": 5800
    },
    {
      "epoch": 9.6,
      "grad_norm": 5.71759033203125,
      "learning_rate": 1.556e-05,
      "loss": 0.0856,
      "step": 6000
    },
    {
      "epoch": 9.6,
      "eval_loss": 0.2818545699119568,
      "eval_runtime": 205.5208,
      "eval_samples_per_second": 6.082,
      "eval_steps_per_second": 1.523,
      "eval_wer": 0.22781703965449548,
      "step": 6000
    },
    {
      "epoch": 9.92,
      "grad_norm": 5.981679439544678,
      "learning_rate": 1.533777777777778e-05,
      "loss": 0.0895,
      "step": 6200
    },
    {
      "epoch": 10.24,
      "grad_norm": 4.288158893585205,
      "learning_rate": 1.5115555555555557e-05,
      "loss": 0.0744,
      "step": 6400
    },
    {
      "epoch": 10.56,
      "grad_norm": 4.479767799377441,
      "learning_rate": 1.4893333333333335e-05,
      "loss": 0.0721,
      "step": 6600
    },
    {
      "epoch": 10.88,
      "grad_norm": 3.508542776107788,
      "learning_rate": 1.4671111111111111e-05,
      "loss": 0.0732,
      "step": 6800
    },
    {
      "epoch": 11.2,
      "grad_norm": 5.090618133544922,
      "learning_rate": 1.444888888888889e-05,
      "loss": 0.066,
      "step": 7000
    },
    {
      "epoch": 11.2,
      "eval_loss": 0.27581551671028137,
      "eval_runtime": 200.6769,
      "eval_samples_per_second": 6.229,
      "eval_steps_per_second": 1.56,
      "eval_wer": 0.2033765213977228,
      "step": 7000
    },
    {
      "epoch": 11.52,
      "grad_norm": 4.929802417755127,
      "learning_rate": 1.4227777777777778e-05,
      "loss": 0.0617,
      "step": 7200
    },
    {
      "epoch": 11.84,
      "grad_norm": 4.811327934265137,
      "learning_rate": 1.4005555555555556e-05,
      "loss": 0.0618,
      "step": 7400
    },
    {
      "epoch": 12.16,
      "grad_norm": 3.7202506065368652,
      "learning_rate": 1.3783333333333336e-05,
      "loss": 0.0582,
      "step": 7600
    },
    {
      "epoch": 12.48,
      "grad_norm": 3.58890962600708,
      "learning_rate": 1.3561111111111112e-05,
      "loss": 0.0524,
      "step": 7800
    },
    {
      "epoch": 12.8,
      "grad_norm": 2.8380696773529053,
      "learning_rate": 1.333888888888889e-05,
      "loss": 0.0534,
      "step": 8000
    },
    {
      "epoch": 12.8,
      "eval_loss": 0.2711934447288513,
      "eval_runtime": 205.6866,
      "eval_samples_per_second": 6.077,
      "eval_steps_per_second": 1.522,
      "eval_wer": 0.22340007852375343,
      "step": 8000
    },
    {
      "epoch": 13.12,
      "grad_norm": 5.2906670570373535,
      "learning_rate": 1.311666666666667e-05,
      "loss": 0.0547,
      "step": 8200
    },
    {
      "epoch": 13.44,
      "grad_norm": 4.201717853546143,
      "learning_rate": 1.2894444444444445e-05,
      "loss": 0.0482,
      "step": 8400
    },
    {
      "epoch": 13.76,
      "grad_norm": 6.0169477462768555,
      "learning_rate": 1.2672222222222223e-05,
      "loss": 0.0513,
      "step": 8600
    },
    {
      "epoch": 14.08,
      "grad_norm": 2.1733462810516357,
      "learning_rate": 1.2450000000000003e-05,
      "loss": 0.0474,
      "step": 8800
    },
    {
      "epoch": 14.4,
      "grad_norm": 4.900807857513428,
      "learning_rate": 1.2227777777777779e-05,
      "loss": 0.044,
      "step": 9000
    },
    {
      "epoch": 14.4,
      "eval_loss": 0.2689875662326813,
      "eval_runtime": 205.0191,
      "eval_samples_per_second": 6.097,
      "eval_steps_per_second": 1.527,
      "eval_wer": 0.2009226541028661,
      "step": 9000
    },
    {
      "epoch": 14.72,
      "grad_norm": 2.0183374881744385,
      "learning_rate": 1.2005555555555557e-05,
      "loss": 0.0439,
      "step": 9200
    },
    {
      "epoch": 15.04,
      "grad_norm": 4.9282450675964355,
      "learning_rate": 1.1784444444444444e-05,
      "loss": 0.0436,
      "step": 9400
    },
    {
      "epoch": 15.36,
      "grad_norm": 3.2207448482513428,
      "learning_rate": 1.1562222222222224e-05,
      "loss": 0.0372,
      "step": 9600
    },
    {
      "epoch": 15.68,
      "grad_norm": 5.770218849182129,
      "learning_rate": 1.134e-05,
      "loss": 0.0388,
      "step": 9800
    },
    {
      "epoch": 16.0,
      "grad_norm": 9.450571060180664,
      "learning_rate": 1.1117777777777778e-05,
      "loss": 0.0393,
      "step": 10000
    },
    {
      "epoch": 16.0,
      "eval_loss": 0.2668628692626953,
      "eval_runtime": 205.3305,
      "eval_samples_per_second": 6.088,
      "eval_steps_per_second": 1.524,
      "eval_wer": 0.1930702787593247,
      "step": 10000
    },
    {
      "epoch": 16.32,
      "grad_norm": 6.021556854248047,
      "learning_rate": 1.0895555555555557e-05,
      "loss": 0.0328,
      "step": 10200
    },
    {
      "epoch": 16.64,
      "grad_norm": 2.8971662521362305,
      "learning_rate": 1.0673333333333333e-05,
      "loss": 0.0342,
      "step": 10400
    },
    {
      "epoch": 16.96,
      "grad_norm": 3.1022391319274902,
      "learning_rate": 1.0451111111111111e-05,
      "loss": 0.0377,
      "step": 10600
    },
    {
      "epoch": 17.28,
      "grad_norm": 1.2703841924667358,
      "learning_rate": 1.022888888888889e-05,
      "loss": 0.0348,
      "step": 10800
    },
    {
      "epoch": 17.6,
      "grad_norm": 4.3228373527526855,
      "learning_rate": 1.0006666666666667e-05,
      "loss": 0.0344,
      "step": 11000
    },
    {
      "epoch": 17.6,
      "eval_loss": 0.26008835434913635,
      "eval_runtime": 202.8929,
      "eval_samples_per_second": 6.161,
      "eval_steps_per_second": 1.543,
      "eval_wer": 0.18865331762858265,
      "step": 11000
    },
    {
      "epoch": 17.92,
      "grad_norm": 5.954049587249756,
      "learning_rate": 9.784444444444445e-06,
      "loss": 0.034,
      "step": 11200
    },
    {
      "epoch": 18.24,
      "grad_norm": 2.9373772144317627,
      "learning_rate": 9.562222222222223e-06,
      "loss": 0.0321,
      "step": 11400
    },
    {
      "epoch": 18.56,
      "grad_norm": 3.066429615020752,
      "learning_rate": 9.340000000000002e-06,
      "loss": 0.0299,
      "step": 11600
    },
    {
      "epoch": 18.88,
      "grad_norm": 2.1500790119171143,
      "learning_rate": 9.117777777777778e-06,
      "loss": 0.0294,
      "step": 11800
    },
    {
      "epoch": 19.2,
      "grad_norm": 2.651730537414551,
      "learning_rate": 8.895555555555556e-06,
      "loss": 0.0292,
      "step": 12000
    },
    {
      "epoch": 19.2,
      "eval_loss": 0.2627313733100891,
      "eval_runtime": 202.5129,
      "eval_samples_per_second": 6.172,
      "eval_steps_per_second": 1.546,
      "eval_wer": 0.18089909697683548,
      "step": 12000
    },
    {
      "epoch": 19.52,
      "grad_norm": 2.2529680728912354,
      "learning_rate": 8.673333333333334e-06,
      "loss": 0.0284,
      "step": 12200
    },
    {
      "epoch": 19.84,
      "grad_norm": 4.965338230133057,
      "learning_rate": 8.451111111111112e-06,
      "loss": 0.0275,
      "step": 12400
    },
    {
      "epoch": 20.16,
      "grad_norm": 4.967452526092529,
      "learning_rate": 8.22888888888889e-06,
      "loss": 0.0287,
      "step": 12600
    },
    {
      "epoch": 20.48,
      "grad_norm": 4.872837543487549,
      "learning_rate": 8.006666666666667e-06,
      "loss": 0.0264,
      "step": 12800
    },
    {
      "epoch": 20.8,
      "grad_norm": 5.586236000061035,
      "learning_rate": 7.784444444444445e-06,
      "loss": 0.0272,
      "step": 13000
    },
    {
      "epoch": 20.8,
      "eval_loss": 0.25965991616249084,
      "eval_runtime": 203.2854,
      "eval_samples_per_second": 6.149,
      "eval_steps_per_second": 1.54,
      "eval_wer": 0.18315665488810365,
      "step": 13000
    },
    {
      "epoch": 21.12,
      "grad_norm": 2.6139557361602783,
      "learning_rate": 7.562222222222223e-06,
      "loss": 0.0251,
      "step": 13200
    },
    {
      "epoch": 21.44,
      "grad_norm": 6.1768341064453125,
      "learning_rate": 7.340000000000001e-06,
      "loss": 0.025,
      "step": 13400
    },
    {
      "epoch": 21.76,
      "grad_norm": 3.9618000984191895,
      "learning_rate": 7.117777777777778e-06,
      "loss": 0.0257,
      "step": 13600
    },
    {
      "epoch": 22.08,
      "grad_norm": 2.3093647956848145,
      "learning_rate": 6.8955555555555565e-06,
      "loss": 0.0241,
      "step": 13800
    },
    {
      "epoch": 22.4,
      "grad_norm": 1.8838666677474976,
      "learning_rate": 6.674444444444445e-06,
      "loss": 0.0231,
      "step": 14000
    },
    {
      "epoch": 22.4,
      "eval_loss": 0.25562289357185364,
      "eval_runtime": 205.1594,
      "eval_samples_per_second": 6.093,
      "eval_steps_per_second": 1.526,
      "eval_wer": 0.18138987043580684,
      "step": 14000
    },
    {
      "epoch": 22.72,
      "grad_norm": 6.57139253616333,
      "learning_rate": 6.452222222222223e-06,
      "loss": 0.0237,
      "step": 14200
    },
    {
      "epoch": 23.04,
      "grad_norm": 6.151878833770752,
      "learning_rate": 6.2300000000000005e-06,
      "loss": 0.0231,
      "step": 14400
    },
    {
      "epoch": 23.36,
      "grad_norm": 3.6745595932006836,
      "learning_rate": 6.007777777777778e-06,
      "loss": 0.0226,
      "step": 14600
    },
    {
      "epoch": 23.68,
      "grad_norm": 1.7786234617233276,
      "learning_rate": 5.785555555555556e-06,
      "loss": 0.0217,
      "step": 14800
    },
    {
      "epoch": 24.0,
      "grad_norm": 4.625872611999512,
      "learning_rate": 5.563333333333334e-06,
      "loss": 0.0223,
      "step": 15000
    },
    {
      "epoch": 24.0,
      "eval_loss": 0.2562410533428192,
      "eval_runtime": 202.4663,
      "eval_samples_per_second": 6.174,
      "eval_steps_per_second": 1.546,
      "eval_wer": 0.1827640361209266,
      "step": 15000
    },
    {
      "epoch": 24.32,
      "grad_norm": 2.113445520401001,
      "learning_rate": 5.341111111111111e-06,
      "loss": 0.0205,
      "step": 15200
    },
    {
      "epoch": 24.64,
      "grad_norm": 3.8710196018218994,
      "learning_rate": 5.118888888888889e-06,
      "loss": 0.0216,
      "step": 15400
    },
    {
      "epoch": 24.96,
      "grad_norm": 1.624074935913086,
      "learning_rate": 4.896666666666667e-06,
      "loss": 0.0221,
      "step": 15600
    },
    {
      "epoch": 25.28,
      "grad_norm": 11.679327964782715,
      "learning_rate": 4.6744444444444445e-06,
      "loss": 0.0192,
      "step": 15800
    },
    {
      "epoch": 25.6,
      "grad_norm": 5.923542499542236,
      "learning_rate": 4.452222222222223e-06,
      "loss": 0.0192,
      "step": 16000
    },
    {
      "epoch": 25.6,
      "eval_loss": 0.25337618589401245,
      "eval_runtime": 203.6663,
      "eval_samples_per_second": 6.137,
      "eval_steps_per_second": 1.537,
      "eval_wer": 0.18021201413427562,
      "step": 16000
    },
    {
      "epoch": 25.92,
      "grad_norm": 1.646501064300537,
      "learning_rate": 4.23e-06,
      "loss": 0.0198,
      "step": 16200
    },
    {
      "epoch": 26.24,
      "grad_norm": 3.09792160987854,
      "learning_rate": 4.008888888888889e-06,
      "loss": 0.0188,
      "step": 16400
    },
    {
      "epoch": 26.56,
      "grad_norm": 1.3494817018508911,
      "learning_rate": 3.7866666666666667e-06,
      "loss": 0.0158,
      "step": 16600
    },
    {
      "epoch": 26.88,
      "grad_norm": 2.5502867698669434,
      "learning_rate": 3.565555555555556e-06,
      "loss": 0.02,
      "step": 16800
    },
    {
      "epoch": 27.2,
      "grad_norm": 2.5397415161132812,
      "learning_rate": 3.3433333333333332e-06,
      "loss": 0.0167,
      "step": 17000
    },
    {
      "epoch": 27.2,
      "eval_loss": 0.2511875331401825,
      "eval_runtime": 201.8937,
      "eval_samples_per_second": 6.191,
      "eval_steps_per_second": 1.55,
      "eval_wer": 0.1762858264625049,
      "step": 17000
    },
    {
      "epoch": 27.52,
      "grad_norm": 2.679438352584839,
      "learning_rate": 3.1211111111111115e-06,
      "loss": 0.0181,
      "step": 17200
    },
    {
      "epoch": 27.84,
      "grad_norm": 3.130686044692993,
      "learning_rate": 2.898888888888889e-06,
      "loss": 0.018,
      "step": 17400
    },
    {
      "epoch": 28.16,
      "grad_norm": 5.3460774421691895,
      "learning_rate": 2.6766666666666667e-06,
      "loss": 0.0178,
      "step": 17600
    },
    {
      "epoch": 28.48,
      "grad_norm": 3.5381155014038086,
      "learning_rate": 2.4544444444444446e-06,
      "loss": 0.0174,
      "step": 17800
    },
    {
      "epoch": 28.8,
      "grad_norm": 2.955854892730713,
      "learning_rate": 2.2322222222222224e-06,
      "loss": 0.0178,
      "step": 18000
    },
    {
      "epoch": 28.8,
      "eval_loss": 0.24957236647605896,
      "eval_runtime": 205.824,
      "eval_samples_per_second": 6.073,
      "eval_steps_per_second": 1.521,
      "eval_wer": 0.17942677659992148,
      "step": 18000
    },
    {
      "epoch": 29.12,
      "grad_norm": 1.8030221462249756,
      "learning_rate": 2.0100000000000002e-06,
      "loss": 0.017,
      "step": 18200
    },
    {
      "epoch": 29.44,
      "grad_norm": 2.740708112716675,
      "learning_rate": 1.787777777777778e-06,
      "loss": 0.0161,
      "step": 18400
    },
    {
      "epoch": 29.76,
      "grad_norm": 2.2259738445281982,
      "learning_rate": 1.5655555555555557e-06,
      "loss": 0.0153,
      "step": 18600
    },
    {
      "epoch": 30.08,
      "grad_norm": 3.472790479660034,
      "learning_rate": 1.3433333333333335e-06,
      "loss": 0.0156,
      "step": 18800
    },
    {
      "epoch": 30.4,
      "grad_norm": 3.7026050090789795,
      "learning_rate": 1.1211111111111112e-06,
      "loss": 0.0143,
      "step": 19000
    },
    {
      "epoch": 30.4,
      "eval_loss": 0.24610121548175812,
      "eval_runtime": 205.7197,
      "eval_samples_per_second": 6.076,
      "eval_steps_per_second": 1.521,
      "eval_wer": 0.1748135060855909,
      "step": 19000
    },
    {
      "epoch": 30.72,
      "grad_norm": 0.6519302129745483,
      "learning_rate": 8.98888888888889e-07,
      "loss": 0.0145,
      "step": 19200
    },
    {
      "epoch": 31.04,
      "grad_norm": 2.7453384399414062,
      "learning_rate": 6.766666666666667e-07,
      "loss": 0.0159,
      "step": 19400
    },
    {
      "epoch": 31.36,
      "grad_norm": 5.072993278503418,
      "learning_rate": 4.544444444444445e-07,
      "loss": 0.016,
      "step": 19600
    },
    {
      "epoch": 31.68,
      "grad_norm": 2.2992451190948486,
      "learning_rate": 2.3222222222222223e-07,
      "loss": 0.0165,
      "step": 19800
    },
    {
      "epoch": 32.0,
      "grad_norm": 0.8698179721832275,
      "learning_rate": 1e-08,
      "loss": 0.0147,
      "step": 20000
    },
    {
      "epoch": 32.0,
      "eval_loss": 0.24500565230846405,
      "eval_runtime": 201.9819,
      "eval_samples_per_second": 6.189,
      "eval_steps_per_second": 1.55,
      "eval_wer": 0.1762858264625049,
      "step": 20000
    },
    {
      "epoch": 32.0,
      "step": 20000,
      "total_flos": 7.87725543407616e+18,
      "train_loss": 0.17171463202238083,
      "train_runtime": 42247.0614,
      "train_samples_per_second": 7.574,
      "train_steps_per_second": 0.473
    }
  ],
  "logging_steps": 200,
  "max_steps": 20000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 32,
  "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": 7.87725543407616e+18,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}