{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 20.0,
  "eval_steps": 100,
  "global_step": 1160,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.1724137931034483,
      "grad_norm": 9.273743629455566,
      "learning_rate": 9e-07,
      "loss": 9.21,
      "step": 10
    },
    {
      "epoch": 0.3448275862068966,
      "grad_norm": 10.659689903259277,
      "learning_rate": 1.9e-06,
      "loss": 9.0596,
      "step": 20
    },
    {
      "epoch": 0.5172413793103449,
      "grad_norm": 13.765968322753906,
      "learning_rate": 2.8000000000000003e-06,
      "loss": 8.6308,
      "step": 30
    },
    {
      "epoch": 0.6896551724137931,
      "grad_norm": 13.644205093383789,
      "learning_rate": 3.8e-06,
      "loss": 7.9066,
      "step": 40
    },
    {
      "epoch": 0.8620689655172413,
      "grad_norm": 18.20844078063965,
      "learning_rate": 4.800000000000001e-06,
      "loss": 7.8077,
      "step": 50
    },
    {
      "epoch": 1.0344827586206897,
      "grad_norm": 18.099712371826172,
      "learning_rate": 5.8e-06,
      "loss": 6.6628,
      "step": 60
    },
    {
      "epoch": 1.206896551724138,
      "grad_norm": 10.263163566589355,
      "learning_rate": 6.800000000000001e-06,
      "loss": 4.6321,
      "step": 70
    },
    {
      "epoch": 1.3793103448275863,
      "grad_norm": 4.330440998077393,
      "learning_rate": 7.8e-06,
      "loss": 3.9137,
      "step": 80
    },
    {
      "epoch": 1.5517241379310345,
      "grad_norm": 1.776586651802063,
      "learning_rate": 8.8e-06,
      "loss": 3.3387,
      "step": 90
    },
    {
      "epoch": 1.7241379310344827,
      "grad_norm": 4.104515075683594,
      "learning_rate": 9.800000000000001e-06,
      "loss": 3.158,
      "step": 100
    },
    {
      "epoch": 1.7241379310344827,
      "eval_loss": 3.6802542209625244,
      "eval_runtime": 14.7124,
      "eval_samples_per_second": 91.352,
      "eval_steps_per_second": 2.855,
      "eval_wer": 1.0,
      "step": 100
    },
    {
      "epoch": 1.896551724137931,
      "grad_norm": 0.8036434650421143,
      "learning_rate": 1.08e-05,
      "loss": 3.0952,
      "step": 110
    },
    {
      "epoch": 2.0689655172413794,
      "grad_norm": 0.9299842119216919,
      "learning_rate": 1.18e-05,
      "loss": 3.0977,
      "step": 120
    },
    {
      "epoch": 2.2413793103448274,
      "grad_norm": 2.4941225051879883,
      "learning_rate": 1.2800000000000001e-05,
      "loss": 3.0263,
      "step": 130
    },
    {
      "epoch": 2.413793103448276,
      "grad_norm": 1.423953652381897,
      "learning_rate": 1.3800000000000002e-05,
      "loss": 3.0224,
      "step": 140
    },
    {
      "epoch": 2.586206896551724,
      "grad_norm": 1.3000880479812622,
      "learning_rate": 1.48e-05,
      "loss": 3.0211,
      "step": 150
    },
    {
      "epoch": 2.7586206896551726,
      "grad_norm": 1.0403013229370117,
      "learning_rate": 1.58e-05,
      "loss": 3.0063,
      "step": 160
    },
    {
      "epoch": 2.9310344827586206,
      "grad_norm": 0.8328425884246826,
      "learning_rate": 1.6800000000000002e-05,
      "loss": 2.9927,
      "step": 170
    },
    {
      "epoch": 3.103448275862069,
      "grad_norm": 1.010672926902771,
      "learning_rate": 1.78e-05,
      "loss": 3.0028,
      "step": 180
    },
    {
      "epoch": 3.2758620689655173,
      "grad_norm": 2.069948434829712,
      "learning_rate": 1.88e-05,
      "loss": 2.9785,
      "step": 190
    },
    {
      "epoch": 3.4482758620689653,
      "grad_norm": 1.1152503490447998,
      "learning_rate": 1.9800000000000004e-05,
      "loss": 2.9744,
      "step": 200
    },
    {
      "epoch": 3.4482758620689653,
      "eval_loss": 3.1165249347686768,
      "eval_runtime": 14.3806,
      "eval_samples_per_second": 93.459,
      "eval_steps_per_second": 2.921,
      "eval_wer": 1.0,
      "step": 200
    },
    {
      "epoch": 3.6206896551724137,
      "grad_norm": 1.179228663444519,
      "learning_rate": 2.08e-05,
      "loss": 2.9759,
      "step": 210
    },
    {
      "epoch": 3.793103448275862,
      "grad_norm": 0.9685543179512024,
      "learning_rate": 2.18e-05,
      "loss": 2.9684,
      "step": 220
    },
    {
      "epoch": 3.9655172413793105,
      "grad_norm": 0.710486114025116,
      "learning_rate": 2.2800000000000002e-05,
      "loss": 2.9692,
      "step": 230
    },
    {
      "epoch": 4.137931034482759,
      "grad_norm": 0.4123358726501465,
      "learning_rate": 2.38e-05,
      "loss": 2.9609,
      "step": 240
    },
    {
      "epoch": 4.310344827586207,
      "grad_norm": 1.9703031778335571,
      "learning_rate": 2.48e-05,
      "loss": 2.9573,
      "step": 250
    },
    {
      "epoch": 4.482758620689655,
      "grad_norm": 1.264237403869629,
      "learning_rate": 2.58e-05,
      "loss": 2.952,
      "step": 260
    },
    {
      "epoch": 4.655172413793103,
      "grad_norm": 0.4067547619342804,
      "learning_rate": 2.6800000000000004e-05,
      "loss": 2.9432,
      "step": 270
    },
    {
      "epoch": 4.827586206896552,
      "grad_norm": 0.5488032698631287,
      "learning_rate": 2.7800000000000005e-05,
      "loss": 2.9385,
      "step": 280
    },
    {
      "epoch": 5.0,
      "grad_norm": 1.3628854751586914,
      "learning_rate": 2.88e-05,
      "loss": 2.9384,
      "step": 290
    },
    {
      "epoch": 5.172413793103448,
      "grad_norm": 0.5179504752159119,
      "learning_rate": 2.98e-05,
      "loss": 2.9266,
      "step": 300
    },
    {
      "epoch": 5.172413793103448,
      "eval_loss": 3.017453908920288,
      "eval_runtime": 14.4537,
      "eval_samples_per_second": 92.987,
      "eval_steps_per_second": 2.906,
      "eval_wer": 1.0,
      "step": 300
    },
    {
      "epoch": 5.344827586206897,
      "grad_norm": 1.5788885354995728,
      "learning_rate": 3.08e-05,
      "loss": 2.9229,
      "step": 310
    },
    {
      "epoch": 5.517241379310345,
      "grad_norm": 1.777976155281067,
      "learning_rate": 3.18e-05,
      "loss": 2.8996,
      "step": 320
    },
    {
      "epoch": 5.689655172413794,
      "grad_norm": 0.7765872478485107,
      "learning_rate": 3.2800000000000004e-05,
      "loss": 2.8698,
      "step": 330
    },
    {
      "epoch": 5.862068965517241,
      "grad_norm": 1.5203367471694946,
      "learning_rate": 3.38e-05,
      "loss": 2.8412,
      "step": 340
    },
    {
      "epoch": 6.0344827586206895,
      "grad_norm": 3.4122729301452637,
      "learning_rate": 3.48e-05,
      "loss": 2.7877,
      "step": 350
    },
    {
      "epoch": 6.206896551724138,
      "grad_norm": 1.2957384586334229,
      "learning_rate": 3.58e-05,
      "loss": 2.6514,
      "step": 360
    },
    {
      "epoch": 6.379310344827586,
      "grad_norm": 1.9932421445846558,
      "learning_rate": 3.68e-05,
      "loss": 2.5492,
      "step": 370
    },
    {
      "epoch": 6.551724137931035,
      "grad_norm": 2.7136945724487305,
      "learning_rate": 3.7800000000000004e-05,
      "loss": 2.3894,
      "step": 380
    },
    {
      "epoch": 6.724137931034483,
      "grad_norm": 3.479182720184326,
      "learning_rate": 3.88e-05,
      "loss": 2.2604,
      "step": 390
    },
    {
      "epoch": 6.896551724137931,
      "grad_norm": 1.5610332489013672,
      "learning_rate": 3.9800000000000005e-05,
      "loss": 2.1336,
      "step": 400
    },
    {
      "epoch": 6.896551724137931,
      "eval_loss": 2.2134621143341064,
      "eval_runtime": 14.4845,
      "eval_samples_per_second": 92.789,
      "eval_steps_per_second": 2.9,
      "eval_wer": 1.011654374943094,
      "step": 400
    },
    {
      "epoch": 7.068965517241379,
      "grad_norm": 1.7063915729522705,
      "learning_rate": 4.08e-05,
      "loss": 2.0117,
      "step": 410
    },
    {
      "epoch": 7.241379310344827,
      "grad_norm": 2.4562268257141113,
      "learning_rate": 4.18e-05,
      "loss": 1.827,
      "step": 420
    },
    {
      "epoch": 7.413793103448276,
      "grad_norm": 2.533010482788086,
      "learning_rate": 4.2800000000000004e-05,
      "loss": 1.6851,
      "step": 430
    },
    {
      "epoch": 7.586206896551724,
      "grad_norm": 3.3842196464538574,
      "learning_rate": 4.38e-05,
      "loss": 1.5765,
      "step": 440
    },
    {
      "epoch": 7.758620689655173,
      "grad_norm": 4.379862308502197,
      "learning_rate": 4.4800000000000005e-05,
      "loss": 1.4748,
      "step": 450
    },
    {
      "epoch": 7.931034482758621,
      "grad_norm": 2.6355247497558594,
      "learning_rate": 4.58e-05,
      "loss": 1.3422,
      "step": 460
    },
    {
      "epoch": 8.10344827586207,
      "grad_norm": 3.2931857109069824,
      "learning_rate": 4.6800000000000006e-05,
      "loss": 1.2681,
      "step": 470
    },
    {
      "epoch": 8.275862068965518,
      "grad_norm": 1.8078395128250122,
      "learning_rate": 4.78e-05,
      "loss": 1.1503,
      "step": 480
    },
    {
      "epoch": 8.448275862068966,
      "grad_norm": 2.258090019226074,
      "learning_rate": 4.88e-05,
      "loss": 1.0564,
      "step": 490
    },
    {
      "epoch": 8.620689655172415,
      "grad_norm": 2.1892685890197754,
      "learning_rate": 4.9800000000000004e-05,
      "loss": 1.0119,
      "step": 500
    },
    {
      "epoch": 8.620689655172415,
      "eval_loss": 1.022745966911316,
      "eval_runtime": 14.558,
      "eval_samples_per_second": 92.321,
      "eval_steps_per_second": 2.885,
      "eval_wer": 0.825093326049349,
      "step": 500
    },
    {
      "epoch": 8.793103448275861,
      "grad_norm": 1.7447644472122192,
      "learning_rate": 5.08e-05,
      "loss": 0.9426,
      "step": 510
    },
    {
      "epoch": 8.96551724137931,
      "grad_norm": 2.7097115516662598,
      "learning_rate": 5.1800000000000005e-05,
      "loss": 0.8773,
      "step": 520
    },
    {
      "epoch": 9.137931034482758,
      "grad_norm": 1.6821448802947998,
      "learning_rate": 5.28e-05,
      "loss": 0.7822,
      "step": 530
    },
    {
      "epoch": 9.310344827586206,
      "grad_norm": 1.5076041221618652,
      "learning_rate": 5.380000000000001e-05,
      "loss": 0.7334,
      "step": 540
    },
    {
      "epoch": 9.482758620689655,
      "grad_norm": 2.951249361038208,
      "learning_rate": 5.4800000000000004e-05,
      "loss": 0.6852,
      "step": 550
    },
    {
      "epoch": 9.655172413793103,
      "grad_norm": 1.5876814126968384,
      "learning_rate": 5.580000000000001e-05,
      "loss": 0.653,
      "step": 560
    },
    {
      "epoch": 9.827586206896552,
      "grad_norm": 2.462275743484497,
      "learning_rate": 5.68e-05,
      "loss": 0.6415,
      "step": 570
    },
    {
      "epoch": 10.0,
      "grad_norm": 1.9461971521377563,
      "learning_rate": 5.7799999999999995e-05,
      "loss": 0.6092,
      "step": 580
    },
    {
      "epoch": 10.172413793103448,
      "grad_norm": 1.474002480506897,
      "learning_rate": 5.88e-05,
      "loss": 0.5086,
      "step": 590
    },
    {
      "epoch": 10.344827586206897,
      "grad_norm": 1.8599964380264282,
      "learning_rate": 5.9800000000000003e-05,
      "loss": 0.4995,
      "step": 600
    },
    {
      "epoch": 10.344827586206897,
      "eval_loss": 0.7699851989746094,
      "eval_runtime": 14.4827,
      "eval_samples_per_second": 92.8,
      "eval_steps_per_second": 2.9,
      "eval_wer": 0.6573795866338887,
      "step": 600
    },
    {
      "epoch": 10.517241379310345,
      "grad_norm": 1.4216625690460205,
      "learning_rate": 6.08e-05,
      "loss": 0.4884,
      "step": 610
    },
    {
      "epoch": 10.689655172413794,
      "grad_norm": 2.1162352561950684,
      "learning_rate": 6.18e-05,
      "loss": 0.4398,
      "step": 620
    },
    {
      "epoch": 10.862068965517242,
      "grad_norm": 1.4033368825912476,
      "learning_rate": 6.280000000000001e-05,
      "loss": 0.4458,
      "step": 630
    },
    {
      "epoch": 11.03448275862069,
      "grad_norm": 1.6064072847366333,
      "learning_rate": 6.38e-05,
      "loss": 0.4414,
      "step": 640
    },
    {
      "epoch": 11.206896551724139,
      "grad_norm": 2.22825288772583,
      "learning_rate": 6.48e-05,
      "loss": 0.3528,
      "step": 650
    },
    {
      "epoch": 11.379310344827585,
      "grad_norm": 1.4636731147766113,
      "learning_rate": 6.58e-05,
      "loss": 0.3699,
      "step": 660
    },
    {
      "epoch": 11.551724137931034,
      "grad_norm": 1.3389075994491577,
      "learning_rate": 6.680000000000001e-05,
      "loss": 0.3621,
      "step": 670
    },
    {
      "epoch": 11.724137931034482,
      "grad_norm": 2.6080307960510254,
      "learning_rate": 6.780000000000001e-05,
      "loss": 0.338,
      "step": 680
    },
    {
      "epoch": 11.89655172413793,
      "grad_norm": 1.3124964237213135,
      "learning_rate": 6.879999999999999e-05,
      "loss": 0.3143,
      "step": 690
    },
    {
      "epoch": 12.068965517241379,
      "grad_norm": 1.1012139320373535,
      "learning_rate": 6.98e-05,
      "loss": 0.3233,
      "step": 700
    },
    {
      "epoch": 12.068965517241379,
      "eval_loss": 0.49695342779159546,
      "eval_runtime": 14.4504,
      "eval_samples_per_second": 93.008,
      "eval_steps_per_second": 2.906,
      "eval_wer": 0.5240826732222525,
      "step": 700
    },
    {
      "epoch": 12.241379310344827,
      "grad_norm": 2.3211851119995117,
      "learning_rate": 7.08e-05,
      "loss": 0.2827,
      "step": 710
    },
    {
      "epoch": 12.413793103448276,
      "grad_norm": 1.2446343898773193,
      "learning_rate": 7.18e-05,
      "loss": 0.2478,
      "step": 720
    },
    {
      "epoch": 12.586206896551724,
      "grad_norm": 1.1632747650146484,
      "learning_rate": 7.280000000000001e-05,
      "loss": 0.2761,
      "step": 730
    },
    {
      "epoch": 12.758620689655173,
      "grad_norm": 1.39242684841156,
      "learning_rate": 7.38e-05,
      "loss": 0.2893,
      "step": 740
    },
    {
      "epoch": 12.931034482758621,
      "grad_norm": 1.546246886253357,
      "learning_rate": 7.48e-05,
      "loss": 0.2382,
      "step": 750
    },
    {
      "epoch": 13.10344827586207,
      "grad_norm": 1.10295832157135,
      "learning_rate": 7.58e-05,
      "loss": 0.2563,
      "step": 760
    },
    {
      "epoch": 13.275862068965518,
      "grad_norm": 1.3670897483825684,
      "learning_rate": 7.680000000000001e-05,
      "loss": 0.2438,
      "step": 770
    },
    {
      "epoch": 13.448275862068966,
      "grad_norm": 1.2440053224563599,
      "learning_rate": 7.780000000000001e-05,
      "loss": 0.2149,
      "step": 780
    },
    {
      "epoch": 13.620689655172415,
      "grad_norm": 1.1088565587997437,
      "learning_rate": 7.88e-05,
      "loss": 0.2341,
      "step": 790
    },
    {
      "epoch": 13.793103448275861,
      "grad_norm": 1.2869517803192139,
      "learning_rate": 7.98e-05,
      "loss": 0.2452,
      "step": 800
    },
    {
      "epoch": 13.793103448275861,
      "eval_loss": 0.4584949016571045,
      "eval_runtime": 14.4542,
      "eval_samples_per_second": 92.983,
      "eval_steps_per_second": 2.906,
      "eval_wer": 0.49084949467358646,
      "step": 800
    },
    {
      "epoch": 13.96551724137931,
      "grad_norm": 1.8122053146362305,
      "learning_rate": 8.080000000000001e-05,
      "loss": 0.2153,
      "step": 810
    },
    {
      "epoch": 14.137931034482758,
      "grad_norm": 1.5954018831253052,
      "learning_rate": 8.18e-05,
      "loss": 0.1834,
      "step": 820
    },
    {
      "epoch": 14.310344827586206,
      "grad_norm": 1.2168644666671753,
      "learning_rate": 8.28e-05,
      "loss": 0.2001,
      "step": 830
    },
    {
      "epoch": 14.482758620689655,
      "grad_norm": 2.418687343597412,
      "learning_rate": 8.38e-05,
      "loss": 0.2027,
      "step": 840
    },
    {
      "epoch": 14.655172413793103,
      "grad_norm": 4.489363670349121,
      "learning_rate": 8.48e-05,
      "loss": 0.1739,
      "step": 850
    },
    {
      "epoch": 14.827586206896552,
      "grad_norm": 1.143477201461792,
      "learning_rate": 8.58e-05,
      "loss": 0.2029,
      "step": 860
    },
    {
      "epoch": 15.0,
      "grad_norm": 1.1375572681427002,
      "learning_rate": 8.680000000000001e-05,
      "loss": 0.2061,
      "step": 870
    },
    {
      "epoch": 15.172413793103448,
      "grad_norm": 1.596352458000183,
      "learning_rate": 8.78e-05,
      "loss": 0.1517,
      "step": 880
    },
    {
      "epoch": 15.344827586206897,
      "grad_norm": 1.0544416904449463,
      "learning_rate": 8.88e-05,
      "loss": 0.1687,
      "step": 890
    },
    {
      "epoch": 15.517241379310345,
      "grad_norm": 0.9486349821090698,
      "learning_rate": 8.98e-05,
      "loss": 0.181,
      "step": 900
    },
    {
      "epoch": 15.517241379310345,
      "eval_loss": 0.4625524878501892,
      "eval_runtime": 14.6126,
      "eval_samples_per_second": 91.975,
      "eval_steps_per_second": 2.874,
      "eval_wer": 0.4813803150323227,
      "step": 900
    },
    {
      "epoch": 15.689655172413794,
      "grad_norm": 2.3056046962738037,
      "learning_rate": 9.080000000000001e-05,
      "loss": 0.1855,
      "step": 910
    },
    {
      "epoch": 15.862068965517242,
      "grad_norm": 1.6048717498779297,
      "learning_rate": 9.180000000000001e-05,
      "loss": 0.1783,
      "step": 920
    },
    {
      "epoch": 16.03448275862069,
      "grad_norm": 0.8028109073638916,
      "learning_rate": 9.28e-05,
      "loss": 0.1877,
      "step": 930
    },
    {
      "epoch": 16.20689655172414,
      "grad_norm": 1.15639066696167,
      "learning_rate": 9.38e-05,
      "loss": 0.1278,
      "step": 940
    },
    {
      "epoch": 16.379310344827587,
      "grad_norm": 1.2636216878890991,
      "learning_rate": 9.48e-05,
      "loss": 0.1544,
      "step": 950
    },
    {
      "epoch": 16.551724137931036,
      "grad_norm": 1.0739763975143433,
      "learning_rate": 9.58e-05,
      "loss": 0.1659,
      "step": 960
    },
    {
      "epoch": 16.724137931034484,
      "grad_norm": 1.9558533430099487,
      "learning_rate": 9.680000000000001e-05,
      "loss": 0.1622,
      "step": 970
    },
    {
      "epoch": 16.896551724137932,
      "grad_norm": 1.494171142578125,
      "learning_rate": 9.78e-05,
      "loss": 0.1448,
      "step": 980
    },
    {
      "epoch": 17.06896551724138,
      "grad_norm": 0.8110372424125671,
      "learning_rate": 9.88e-05,
      "loss": 0.1522,
      "step": 990
    },
    {
      "epoch": 17.24137931034483,
      "grad_norm": 1.800034761428833,
      "learning_rate": 9.98e-05,
      "loss": 0.1419,
      "step": 1000
    },
    {
      "epoch": 17.24137931034483,
      "eval_loss": 0.4916817247867584,
      "eval_runtime": 14.4094,
      "eval_samples_per_second": 93.273,
      "eval_steps_per_second": 2.915,
      "eval_wer": 0.4774651734498771,
      "step": 1000
    },
    {
      "epoch": 17.413793103448278,
      "grad_norm": 1.428348422050476,
      "learning_rate": 9.5e-05,
      "loss": 0.1284,
      "step": 1010
    },
    {
      "epoch": 17.586206896551722,
      "grad_norm": 1.0604817867279053,
      "learning_rate": 8.875e-05,
      "loss": 0.1525,
      "step": 1020
    },
    {
      "epoch": 17.75862068965517,
      "grad_norm": 0.9464378952980042,
      "learning_rate": 8.25e-05,
      "loss": 0.1326,
      "step": 1030
    },
    {
      "epoch": 17.93103448275862,
      "grad_norm": 1.1999770402908325,
      "learning_rate": 7.625e-05,
      "loss": 0.1266,
      "step": 1040
    },
    {
      "epoch": 18.103448275862068,
      "grad_norm": 0.815773606300354,
      "learning_rate": 7e-05,
      "loss": 0.118,
      "step": 1050
    },
    {
      "epoch": 18.275862068965516,
      "grad_norm": 0.8608018755912781,
      "learning_rate": 6.375e-05,
      "loss": 0.1203,
      "step": 1060
    },
    {
      "epoch": 18.448275862068964,
      "grad_norm": 1.4489638805389404,
      "learning_rate": 5.7499999999999995e-05,
      "loss": 0.1175,
      "step": 1070
    },
    {
      "epoch": 18.620689655172413,
      "grad_norm": 0.9969916939735413,
      "learning_rate": 5.125e-05,
      "loss": 0.1075,
      "step": 1080
    },
    {
      "epoch": 18.79310344827586,
      "grad_norm": 1.1026653051376343,
      "learning_rate": 4.5e-05,
      "loss": 0.109,
      "step": 1090
    },
    {
      "epoch": 18.96551724137931,
      "grad_norm": 1.557403802871704,
      "learning_rate": 3.875e-05,
      "loss": 0.1175,
      "step": 1100
    },
    {
      "epoch": 18.96551724137931,
      "eval_loss": 0.42788705229759216,
      "eval_runtime": 14.5391,
      "eval_samples_per_second": 92.441,
      "eval_steps_per_second": 2.889,
      "eval_wer": 0.4359464627151052,
      "step": 1100
    },
    {
      "epoch": 19.137931034482758,
      "grad_norm": 0.6065557599067688,
      "learning_rate": 3.2500000000000004e-05,
      "loss": 0.0727,
      "step": 1110
    },
    {
      "epoch": 19.310344827586206,
      "grad_norm": 0.7714155912399292,
      "learning_rate": 2.625e-05,
      "loss": 0.0907,
      "step": 1120
    },
    {
      "epoch": 19.482758620689655,
      "grad_norm": 1.672736644744873,
      "learning_rate": 2e-05,
      "loss": 0.0953,
      "step": 1130
    },
    {
      "epoch": 19.655172413793103,
      "grad_norm": 0.798320472240448,
      "learning_rate": 1.3750000000000002e-05,
      "loss": 0.084,
      "step": 1140
    },
    {
      "epoch": 19.82758620689655,
      "grad_norm": 0.8147380352020264,
      "learning_rate": 7.5e-06,
      "loss": 0.0907,
      "step": 1150
    },
    {
      "epoch": 20.0,
      "grad_norm": 1.1494929790496826,
      "learning_rate": 1.25e-06,
      "loss": 0.0897,
      "step": 1160
    },
    {
      "epoch": 20.0,
      "step": 1160,
      "total_flos": 2.2493941126178012e+18,
      "train_loss": 1.5806312146885642,
      "train_runtime": 1021.9437,
      "train_samples_per_second": 72.333,
      "train_steps_per_second": 1.135
    }
  ],
  "logging_steps": 10,
  "max_steps": 1160,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 20,
  "save_steps": 400,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 2.2493941126178012e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}