{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.9982748706152962,
  "eval_steps": 500,
  "global_step": 434,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.004600345025876941,
      "grad_norm": 0.005687589757144451,
      "learning_rate": 0.0,
      "loss": 1.2964,
      "mean_token_accuracy": 0.6567770391702652,
      "num_tokens": 104794.0,
      "step": 1
    },
    {
      "epoch": 0.009200690051753882,
      "grad_norm": 0.0055035436525940895,
      "learning_rate": 2.2727272727272729e-07,
      "loss": 1.3236,
      "mean_token_accuracy": 0.6467703059315681,
      "num_tokens": 198477.0,
      "step": 2
    },
    {
      "epoch": 0.013801035077630822,
      "grad_norm": 0.00571180647239089,
      "learning_rate": 4.5454545454545457e-07,
      "loss": 1.2871,
      "mean_token_accuracy": 0.655419260263443,
      "num_tokens": 297137.0,
      "step": 3
    },
    {
      "epoch": 0.018401380103507763,
      "grad_norm": 0.006776052992790937,
      "learning_rate": 6.818181818181818e-07,
      "loss": 1.3236,
      "mean_token_accuracy": 0.6503552347421646,
      "num_tokens": 391965.0,
      "step": 4
    },
    {
      "epoch": 0.023001725129384705,
      "grad_norm": 0.00528043182566762,
      "learning_rate": 9.090909090909091e-07,
      "loss": 1.3344,
      "mean_token_accuracy": 0.6503330767154694,
      "num_tokens": 487852.0,
      "step": 5
    },
    {
      "epoch": 0.027602070155261643,
      "grad_norm": 0.005563394166529179,
      "learning_rate": 1.1363636363636364e-06,
      "loss": 1.3086,
      "mean_token_accuracy": 0.654597781598568,
      "num_tokens": 584002.0,
      "step": 6
    },
    {
      "epoch": 0.03220241518113859,
      "grad_norm": 0.007420171983540058,
      "learning_rate": 1.3636363636363636e-06,
      "loss": 1.3444,
      "mean_token_accuracy": 0.6459237858653069,
      "num_tokens": 675831.0,
      "step": 7
    },
    {
      "epoch": 0.03680276020701553,
      "grad_norm": 0.006749135907739401,
      "learning_rate": 1.590909090909091e-06,
      "loss": 1.313,
      "mean_token_accuracy": 0.6493777111172676,
      "num_tokens": 773911.0,
      "step": 8
    },
    {
      "epoch": 0.041403105232892465,
      "grad_norm": 0.00610280130058527,
      "learning_rate": 1.8181818181818183e-06,
      "loss": 1.2976,
      "mean_token_accuracy": 0.657502181828022,
      "num_tokens": 868063.0,
      "step": 9
    },
    {
      "epoch": 0.04600345025876941,
      "grad_norm": 0.007394636515527964,
      "learning_rate": 2.0454545454545457e-06,
      "loss": 1.291,
      "mean_token_accuracy": 0.6572054326534271,
      "num_tokens": 967303.0,
      "step": 10
    },
    {
      "epoch": 0.05060379528464635,
      "grad_norm": 0.007649592589586973,
      "learning_rate": 2.2727272727272728e-06,
      "loss": 1.3437,
      "mean_token_accuracy": 0.6450314894318581,
      "num_tokens": 1061613.0,
      "step": 11
    },
    {
      "epoch": 0.055204140310523286,
      "grad_norm": 0.007066652178764343,
      "learning_rate": 2.5e-06,
      "loss": 1.3062,
      "mean_token_accuracy": 0.6532740890979767,
      "num_tokens": 1154782.0,
      "step": 12
    },
    {
      "epoch": 0.05980448533640023,
      "grad_norm": 0.006177663803100586,
      "learning_rate": 2.7272727272727272e-06,
      "loss": 1.3027,
      "mean_token_accuracy": 0.6530071422457695,
      "num_tokens": 1254453.0,
      "step": 13
    },
    {
      "epoch": 0.06440483036227718,
      "grad_norm": 0.008032996207475662,
      "learning_rate": 2.954545454545455e-06,
      "loss": 1.2904,
      "mean_token_accuracy": 0.6574910432100296,
      "num_tokens": 1349953.0,
      "step": 14
    },
    {
      "epoch": 0.06900517538815411,
      "grad_norm": 0.010039424523711205,
      "learning_rate": 3.181818181818182e-06,
      "loss": 1.3105,
      "mean_token_accuracy": 0.6531528457999229,
      "num_tokens": 1447847.0,
      "step": 15
    },
    {
      "epoch": 0.07360552041403105,
      "grad_norm": 0.00691134762018919,
      "learning_rate": 3.409090909090909e-06,
      "loss": 1.3162,
      "mean_token_accuracy": 0.6518233045935631,
      "num_tokens": 1539663.0,
      "step": 16
    },
    {
      "epoch": 0.078205865439908,
      "grad_norm": 0.007641863543540239,
      "learning_rate": 3.6363636363636366e-06,
      "loss": 1.2965,
      "mean_token_accuracy": 0.656828798353672,
      "num_tokens": 1634446.0,
      "step": 17
    },
    {
      "epoch": 0.08280621046578493,
      "grad_norm": 0.006876361556351185,
      "learning_rate": 3.863636363636364e-06,
      "loss": 1.353,
      "mean_token_accuracy": 0.6474368572235107,
      "num_tokens": 1729775.0,
      "step": 18
    },
    {
      "epoch": 0.08740655549166187,
      "grad_norm": 0.00726681761443615,
      "learning_rate": 4.0909090909090915e-06,
      "loss": 1.3106,
      "mean_token_accuracy": 0.6513411849737167,
      "num_tokens": 1824828.0,
      "step": 19
    },
    {
      "epoch": 0.09200690051753882,
      "grad_norm": 0.008261698298156261,
      "learning_rate": 4.3181818181818185e-06,
      "loss": 1.3019,
      "mean_token_accuracy": 0.6508027985692024,
      "num_tokens": 1925138.0,
      "step": 20
    },
    {
      "epoch": 0.09660724554341575,
      "grad_norm": 0.008603152818977833,
      "learning_rate": 4.5454545454545455e-06,
      "loss": 1.2888,
      "mean_token_accuracy": 0.6574650928378105,
      "num_tokens": 2024720.0,
      "step": 21
    },
    {
      "epoch": 0.1012075905692927,
      "grad_norm": 0.011507058516144753,
      "learning_rate": 4.772727272727273e-06,
      "loss": 1.3541,
      "mean_token_accuracy": 0.6430061087012291,
      "num_tokens": 2115523.0,
      "step": 22
    },
    {
      "epoch": 0.10580793559516964,
      "grad_norm": 0.011566699482500553,
      "learning_rate": 5e-06,
      "loss": 1.2786,
      "mean_token_accuracy": 0.6559519320726395,
      "num_tokens": 2211187.0,
      "step": 23
    },
    {
      "epoch": 0.11040828062104657,
      "grad_norm": 0.011184126138687134,
      "learning_rate": 5.2272727272727274e-06,
      "loss": 1.2869,
      "mean_token_accuracy": 0.6546742096543312,
      "num_tokens": 2307075.0,
      "step": 24
    },
    {
      "epoch": 0.11500862564692352,
      "grad_norm": 0.009821473620831966,
      "learning_rate": 5.4545454545454545e-06,
      "loss": 1.2486,
      "mean_token_accuracy": 0.6607496812939644,
      "num_tokens": 2408388.0,
      "step": 25
    },
    {
      "epoch": 0.11960897067280046,
      "grad_norm": 0.011240242049098015,
      "learning_rate": 5.681818181818183e-06,
      "loss": 1.3021,
      "mean_token_accuracy": 0.6554695144295692,
      "num_tokens": 2502118.0,
      "step": 26
    },
    {
      "epoch": 0.1242093156986774,
      "grad_norm": 0.009518246166408062,
      "learning_rate": 5.90909090909091e-06,
      "loss": 1.2953,
      "mean_token_accuracy": 0.6546049192547798,
      "num_tokens": 2595264.0,
      "step": 27
    },
    {
      "epoch": 0.12880966072455435,
      "grad_norm": 0.008872764185070992,
      "learning_rate": 6.136363636363637e-06,
      "loss": 1.3026,
      "mean_token_accuracy": 0.6526112258434296,
      "num_tokens": 2692102.0,
      "step": 28
    },
    {
      "epoch": 0.13341000575043127,
      "grad_norm": 0.009111359715461731,
      "learning_rate": 6.363636363636364e-06,
      "loss": 1.2366,
      "mean_token_accuracy": 0.6660361588001251,
      "num_tokens": 2785080.0,
      "step": 29
    },
    {
      "epoch": 0.13801035077630822,
      "grad_norm": 0.008477196097373962,
      "learning_rate": 6.590909090909091e-06,
      "loss": 1.2575,
      "mean_token_accuracy": 0.6607440859079361,
      "num_tokens": 2881020.0,
      "step": 30
    },
    {
      "epoch": 0.14261069580218516,
      "grad_norm": 0.007433631923049688,
      "learning_rate": 6.818181818181818e-06,
      "loss": 1.2407,
      "mean_token_accuracy": 0.663087397813797,
      "num_tokens": 2978761.0,
      "step": 31
    },
    {
      "epoch": 0.1472110408280621,
      "grad_norm": 0.008866243064403534,
      "learning_rate": 7.045454545454546e-06,
      "loss": 1.2441,
      "mean_token_accuracy": 0.6632796227931976,
      "num_tokens": 3076332.0,
      "step": 32
    },
    {
      "epoch": 0.15181138585393905,
      "grad_norm": 0.007918394170701504,
      "learning_rate": 7.272727272727273e-06,
      "loss": 1.2707,
      "mean_token_accuracy": 0.661369614303112,
      "num_tokens": 3174677.0,
      "step": 33
    },
    {
      "epoch": 0.156411730879816,
      "grad_norm": 0.009430650621652603,
      "learning_rate": 7.500000000000001e-06,
      "loss": 1.237,
      "mean_token_accuracy": 0.6620191782712936,
      "num_tokens": 3271182.0,
      "step": 34
    },
    {
      "epoch": 0.16101207590569291,
      "grad_norm": 0.009918849915266037,
      "learning_rate": 7.727272727272727e-06,
      "loss": 1.2784,
      "mean_token_accuracy": 0.6557858139276505,
      "num_tokens": 3364423.0,
      "step": 35
    },
    {
      "epoch": 0.16561242093156986,
      "grad_norm": 0.01214979775249958,
      "learning_rate": 7.954545454545455e-06,
      "loss": 1.2267,
      "mean_token_accuracy": 0.665397971868515,
      "num_tokens": 3458719.0,
      "step": 36
    },
    {
      "epoch": 0.1702127659574468,
      "grad_norm": 0.0094231516122818,
      "learning_rate": 8.181818181818183e-06,
      "loss": 1.2112,
      "mean_token_accuracy": 0.6734237298369408,
      "num_tokens": 3549995.0,
      "step": 37
    },
    {
      "epoch": 0.17481311098332375,
      "grad_norm": 0.008751116693019867,
      "learning_rate": 8.40909090909091e-06,
      "loss": 1.2069,
      "mean_token_accuracy": 0.6694932207465172,
      "num_tokens": 3643349.0,
      "step": 38
    },
    {
      "epoch": 0.1794134560092007,
      "grad_norm": 0.008826850913465023,
      "learning_rate": 8.636363636363637e-06,
      "loss": 1.2127,
      "mean_token_accuracy": 0.669578805565834,
      "num_tokens": 3736816.0,
      "step": 39
    },
    {
      "epoch": 0.18401380103507764,
      "grad_norm": 0.008413583971560001,
      "learning_rate": 8.863636363636365e-06,
      "loss": 1.2028,
      "mean_token_accuracy": 0.672471858561039,
      "num_tokens": 3832791.0,
      "step": 40
    },
    {
      "epoch": 0.18861414606095459,
      "grad_norm": 0.006991323083639145,
      "learning_rate": 9.090909090909091e-06,
      "loss": 1.1791,
      "mean_token_accuracy": 0.6750742718577385,
      "num_tokens": 3932062.0,
      "step": 41
    },
    {
      "epoch": 0.1932144910868315,
      "grad_norm": 0.007404603064060211,
      "learning_rate": 9.318181818181819e-06,
      "loss": 1.1645,
      "mean_token_accuracy": 0.6763711273670197,
      "num_tokens": 4024968.0,
      "step": 42
    },
    {
      "epoch": 0.19781483611270845,
      "grad_norm": 0.007536900695413351,
      "learning_rate": 9.545454545454547e-06,
      "loss": 1.1986,
      "mean_token_accuracy": 0.6699943095445633,
      "num_tokens": 4125477.0,
      "step": 43
    },
    {
      "epoch": 0.2024151811385854,
      "grad_norm": 0.008920799009501934,
      "learning_rate": 9.772727272727273e-06,
      "loss": 1.1808,
      "mean_token_accuracy": 0.6771005317568779,
      "num_tokens": 4220159.0,
      "step": 44
    },
    {
      "epoch": 0.20701552616446234,
      "grad_norm": 0.007982825860381126,
      "learning_rate": 1e-05,
      "loss": 1.1712,
      "mean_token_accuracy": 0.6806080117821693,
      "num_tokens": 4313530.0,
      "step": 45
    },
    {
      "epoch": 0.21161587119033928,
      "grad_norm": 0.007420428097248077,
      "learning_rate": 9.999837778582641e-06,
      "loss": 1.1685,
      "mean_token_accuracy": 0.6754245385527611,
      "num_tokens": 4413260.0,
      "step": 46
    },
    {
      "epoch": 0.21621621621621623,
      "grad_norm": 0.008077534846961498,
      "learning_rate": 9.999351124856873e-06,
      "loss": 1.155,
      "mean_token_accuracy": 0.6799673438072205,
      "num_tokens": 4511873.0,
      "step": 47
    },
    {
      "epoch": 0.22081656124209315,
      "grad_norm": 0.008561821654438972,
      "learning_rate": 9.998540070400966e-06,
      "loss": 1.171,
      "mean_token_accuracy": 0.6745180711150169,
      "num_tokens": 4602762.0,
      "step": 48
    },
    {
      "epoch": 0.2254169062679701,
      "grad_norm": 0.00783666130155325,
      "learning_rate": 9.997404667843076e-06,
      "loss": 1.1694,
      "mean_token_accuracy": 0.6786425858736038,
      "num_tokens": 4703481.0,
      "step": 49
    },
    {
      "epoch": 0.23001725129384704,
      "grad_norm": 0.007960589602589607,
      "learning_rate": 9.995944990857848e-06,
      "loss": 1.1707,
      "mean_token_accuracy": 0.6771489679813385,
      "num_tokens": 4799562.0,
      "step": 50
    },
    {
      "epoch": 0.23461759631972398,
      "grad_norm": 0.007685200311243534,
      "learning_rate": 9.994161134161635e-06,
      "loss": 1.1846,
      "mean_token_accuracy": 0.6739635095000267,
      "num_tokens": 4900840.0,
      "step": 51
    },
    {
      "epoch": 0.23921794134560093,
      "grad_norm": 0.008777720853686333,
      "learning_rate": 9.992053213506333e-06,
      "loss": 1.1876,
      "mean_token_accuracy": 0.6759511306881905,
      "num_tokens": 4997883.0,
      "step": 52
    },
    {
      "epoch": 0.24381828637147787,
      "grad_norm": 0.0073701320216059685,
      "learning_rate": 9.989621365671902e-06,
      "loss": 1.1798,
      "mean_token_accuracy": 0.6745479926466942,
      "num_tokens": 5094702.0,
      "step": 53
    },
    {
      "epoch": 0.2484186313973548,
      "grad_norm": 0.0080358712002635,
      "learning_rate": 9.986865748457457e-06,
      "loss": 1.1485,
      "mean_token_accuracy": 0.6811746656894684,
      "num_tokens": 5189105.0,
      "step": 54
    },
    {
      "epoch": 0.25301897642323173,
      "grad_norm": 0.008727259002625942,
      "learning_rate": 9.983786540671052e-06,
      "loss": 1.193,
      "mean_token_accuracy": 0.6724725291132927,
      "num_tokens": 5284168.0,
      "step": 55
    },
    {
      "epoch": 0.2576193214491087,
      "grad_norm": 0.008287778124213219,
      "learning_rate": 9.980383942118066e-06,
      "loss": 1.1849,
      "mean_token_accuracy": 0.6733070015907288,
      "num_tokens": 5382877.0,
      "step": 56
    },
    {
      "epoch": 0.2622196664749856,
      "grad_norm": 0.007565150503069162,
      "learning_rate": 9.976658173588244e-06,
      "loss": 1.1741,
      "mean_token_accuracy": 0.6780361458659172,
      "num_tokens": 5476577.0,
      "step": 57
    },
    {
      "epoch": 0.26682001150086254,
      "grad_norm": 0.00806084182113409,
      "learning_rate": 9.972609476841368e-06,
      "loss": 1.161,
      "mean_token_accuracy": 0.6804856136441231,
      "num_tokens": 5577955.0,
      "step": 58
    },
    {
      "epoch": 0.2714203565267395,
      "grad_norm": 0.010196089744567871,
      "learning_rate": 9.968238114591567e-06,
      "loss": 1.1499,
      "mean_token_accuracy": 0.6841338276863098,
      "num_tokens": 5666759.0,
      "step": 59
    },
    {
      "epoch": 0.27602070155261643,
      "grad_norm": 0.007678396068513393,
      "learning_rate": 9.96354437049027e-06,
      "loss": 1.1543,
      "mean_token_accuracy": 0.6782903298735619,
      "num_tokens": 5763702.0,
      "step": 60
    },
    {
      "epoch": 0.2806210465784934,
      "grad_norm": 0.00881399866193533,
      "learning_rate": 9.958528549107812e-06,
      "loss": 1.1556,
      "mean_token_accuracy": 0.6805967539548874,
      "num_tokens": 5858188.0,
      "step": 61
    },
    {
      "epoch": 0.2852213916043703,
      "grad_norm": 0.009992823004722595,
      "learning_rate": 9.953190975913648e-06,
      "loss": 1.2055,
      "mean_token_accuracy": 0.668336994946003,
      "num_tokens": 5953081.0,
      "step": 62
    },
    {
      "epoch": 0.2898217366302473,
      "grad_norm": 0.008360505104064941,
      "learning_rate": 9.947531997255256e-06,
      "loss": 1.1395,
      "mean_token_accuracy": 0.6817946434020996,
      "num_tokens": 6054917.0,
      "step": 63
    },
    {
      "epoch": 0.2944220816561242,
      "grad_norm": 0.00789540633559227,
      "learning_rate": 9.941551980335653e-06,
      "loss": 1.1541,
      "mean_token_accuracy": 0.6804044917225838,
      "num_tokens": 6151706.0,
      "step": 64
    },
    {
      "epoch": 0.29902242668200113,
      "grad_norm": 0.010097729973495007,
      "learning_rate": 9.935251313189564e-06,
      "loss": 1.1614,
      "mean_token_accuracy": 0.6784722730517387,
      "num_tokens": 6248791.0,
      "step": 65
    },
    {
      "epoch": 0.3036227717078781,
      "grad_norm": 0.009608102962374687,
      "learning_rate": 9.928630404658255e-06,
      "loss": 1.1586,
      "mean_token_accuracy": 0.6793174743652344,
      "num_tokens": 6342795.0,
      "step": 66
    },
    {
      "epoch": 0.308223116733755,
      "grad_norm": 0.008577210828661919,
      "learning_rate": 9.921689684362989e-06,
      "loss": 1.1543,
      "mean_token_accuracy": 0.6808740571141243,
      "num_tokens": 6441301.0,
      "step": 67
    },
    {
      "epoch": 0.312823461759632,
      "grad_norm": 0.00851823017001152,
      "learning_rate": 9.914429602677163e-06,
      "loss": 1.1917,
      "mean_token_accuracy": 0.6714753583073616,
      "num_tokens": 6539454.0,
      "step": 68
    },
    {
      "epoch": 0.3174238067855089,
      "grad_norm": 0.009974654763936996,
      "learning_rate": 9.906850630697068e-06,
      "loss": 1.1883,
      "mean_token_accuracy": 0.6764531210064888,
      "num_tokens": 6634535.0,
      "step": 69
    },
    {
      "epoch": 0.32202415181138583,
      "grad_norm": 0.008425155654549599,
      "learning_rate": 9.89895326021134e-06,
      "loss": 1.1289,
      "mean_token_accuracy": 0.6835334897041321,
      "num_tokens": 6733173.0,
      "step": 70
    },
    {
      "epoch": 0.3266244968372628,
      "grad_norm": 0.009581580758094788,
      "learning_rate": 9.890738003669029e-06,
      "loss": 1.1895,
      "mean_token_accuracy": 0.6751143932342529,
      "num_tokens": 6825744.0,
      "step": 71
    },
    {
      "epoch": 0.3312248418631397,
      "grad_norm": 0.00899854488670826,
      "learning_rate": 9.882205394146362e-06,
      "loss": 1.1807,
      "mean_token_accuracy": 0.6726124361157417,
      "num_tokens": 6921349.0,
      "step": 72
    },
    {
      "epoch": 0.3358251868890167,
      "grad_norm": 0.009314511902630329,
      "learning_rate": 9.873355985312141e-06,
      "loss": 1.1511,
      "mean_token_accuracy": 0.6797289550304413,
      "num_tokens": 7016428.0,
      "step": 73
    },
    {
      "epoch": 0.3404255319148936,
      "grad_norm": 0.008227611891925335,
      "learning_rate": 9.864190351391822e-06,
      "loss": 1.1399,
      "mean_token_accuracy": 0.680743120610714,
      "num_tokens": 7116651.0,
      "step": 74
    },
    {
      "epoch": 0.3450258769407706,
      "grad_norm": 0.009333742782473564,
      "learning_rate": 9.854709087130261e-06,
      "loss": 1.1893,
      "mean_token_accuracy": 0.6743270009756088,
      "num_tokens": 7214530.0,
      "step": 75
    },
    {
      "epoch": 0.3496262219666475,
      "grad_norm": 0.008280204609036446,
      "learning_rate": 9.844912807753105e-06,
      "loss": 1.1246,
      "mean_token_accuracy": 0.6881353706121445,
      "num_tokens": 7314471.0,
      "step": 76
    },
    {
      "epoch": 0.3542265669925244,
      "grad_norm": 0.009001548402011395,
      "learning_rate": 9.834802148926883e-06,
      "loss": 1.1389,
      "mean_token_accuracy": 0.6820591241121292,
      "num_tokens": 7411143.0,
      "step": 77
    },
    {
      "epoch": 0.3588269120184014,
      "grad_norm": 0.008818632923066616,
      "learning_rate": 9.824377766717758e-06,
      "loss": 1.1199,
      "mean_token_accuracy": 0.6870076656341553,
      "num_tokens": 7505547.0,
      "step": 78
    },
    {
      "epoch": 0.3634272570442783,
      "grad_norm": 0.009602286852896214,
      "learning_rate": 9.813640337548955e-06,
      "loss": 1.1457,
      "mean_token_accuracy": 0.681760624051094,
      "num_tokens": 7604105.0,
      "step": 79
    },
    {
      "epoch": 0.3680276020701553,
      "grad_norm": 0.009009558707475662,
      "learning_rate": 9.802590558156863e-06,
      "loss": 1.1536,
      "mean_token_accuracy": 0.6847949475049973,
      "num_tokens": 7702689.0,
      "step": 80
    },
    {
      "epoch": 0.3726279470960322,
      "grad_norm": 0.009743778966367245,
      "learning_rate": 9.791229145545832e-06,
      "loss": 1.1386,
      "mean_token_accuracy": 0.6835950613021851,
      "num_tokens": 7797600.0,
      "step": 81
    },
    {
      "epoch": 0.37722829212190917,
      "grad_norm": 0.009100291877985,
      "learning_rate": 9.779556836941646e-06,
      "loss": 1.1349,
      "mean_token_accuracy": 0.6835153922438622,
      "num_tokens": 7896536.0,
      "step": 82
    },
    {
      "epoch": 0.3818286371477861,
      "grad_norm": 0.009762028232216835,
      "learning_rate": 9.767574389743683e-06,
      "loss": 1.1444,
      "mean_token_accuracy": 0.6820423156023026,
      "num_tokens": 7995947.0,
      "step": 83
    },
    {
      "epoch": 0.386428982173663,
      "grad_norm": 0.00889329332858324,
      "learning_rate": 9.755282581475769e-06,
      "loss": 1.1405,
      "mean_token_accuracy": 0.6851356402039528,
      "num_tokens": 8094263.0,
      "step": 84
    },
    {
      "epoch": 0.39102932719954,
      "grad_norm": 0.00981220230460167,
      "learning_rate": 9.742682209735727e-06,
      "loss": 1.1735,
      "mean_token_accuracy": 0.677773728966713,
      "num_tokens": 8189505.0,
      "step": 85
    },
    {
      "epoch": 0.3956296722254169,
      "grad_norm": 0.010011876933276653,
      "learning_rate": 9.729774092143627e-06,
      "loss": 1.129,
      "mean_token_accuracy": 0.6852159947156906,
      "num_tokens": 8293089.0,
      "step": 86
    },
    {
      "epoch": 0.40023001725129387,
      "grad_norm": 0.010612046346068382,
      "learning_rate": 9.716559066288716e-06,
      "loss": 1.1742,
      "mean_token_accuracy": 0.674958735704422,
      "num_tokens": 8387541.0,
      "step": 87
    },
    {
      "epoch": 0.4048303622771708,
      "grad_norm": 0.0103145157918334,
      "learning_rate": 9.703037989675088e-06,
      "loss": 1.153,
      "mean_token_accuracy": 0.6805268451571465,
      "num_tokens": 8484151.0,
      "step": 88
    },
    {
      "epoch": 0.4094307073030477,
      "grad_norm": 0.009694715961813927,
      "learning_rate": 9.689211739666023e-06,
      "loss": 1.1877,
      "mean_token_accuracy": 0.6738117784261703,
      "num_tokens": 8581092.0,
      "step": 89
    },
    {
      "epoch": 0.4140310523289247,
      "grad_norm": 0.01016835868358612,
      "learning_rate": 9.675081213427076e-06,
      "loss": 1.13,
      "mean_token_accuracy": 0.6874581500887871,
      "num_tokens": 8679981.0,
      "step": 90
    },
    {
      "epoch": 0.4186313973548016,
      "grad_norm": 0.008928492665290833,
      "learning_rate": 9.66064732786784e-06,
      "loss": 1.1369,
      "mean_token_accuracy": 0.6865670680999756,
      "num_tokens": 8779582.0,
      "step": 91
    },
    {
      "epoch": 0.42323174238067857,
      "grad_norm": 0.009940912947058678,
      "learning_rate": 9.645911019582467e-06,
      "loss": 1.1602,
      "mean_token_accuracy": 0.6783237531781197,
      "num_tokens": 8875552.0,
      "step": 92
    },
    {
      "epoch": 0.4278320874065555,
      "grad_norm": 0.011580446735024452,
      "learning_rate": 9.630873244788884e-06,
      "loss": 1.139,
      "mean_token_accuracy": 0.6843747869133949,
      "num_tokens": 8966375.0,
      "step": 93
    },
    {
      "epoch": 0.43243243243243246,
      "grad_norm": 0.009115920402109623,
      "learning_rate": 9.615534979266745e-06,
      "loss": 1.1646,
      "mean_token_accuracy": 0.6751029714941978,
      "num_tokens": 9063222.0,
      "step": 94
    },
    {
      "epoch": 0.4370327774583094,
      "grad_norm": 0.009414221160113811,
      "learning_rate": 9.599897218294122e-06,
      "loss": 1.1572,
      "mean_token_accuracy": 0.6786936447024345,
      "num_tokens": 9158240.0,
      "step": 95
    },
    {
      "epoch": 0.4416331224841863,
      "grad_norm": 0.009733911603689194,
      "learning_rate": 9.583960976582914e-06,
      "loss": 1.1106,
      "mean_token_accuracy": 0.6883520036935806,
      "num_tokens": 9257427.0,
      "step": 96
    },
    {
      "epoch": 0.44623346751006326,
      "grad_norm": 0.01011735387146473,
      "learning_rate": 9.567727288213005e-06,
      "loss": 1.1555,
      "mean_token_accuracy": 0.6811134442687035,
      "num_tokens": 9354084.0,
      "step": 97
    },
    {
      "epoch": 0.4508338125359402,
      "grad_norm": 0.011312155053019524,
      "learning_rate": 9.551197206565174e-06,
      "loss": 1.1386,
      "mean_token_accuracy": 0.682069718837738,
      "num_tokens": 9449568.0,
      "step": 98
    },
    {
      "epoch": 0.45543415756181715,
      "grad_norm": 0.010424958541989326,
      "learning_rate": 9.534371804252727e-06,
      "loss": 1.1424,
      "mean_token_accuracy": 0.6819394826889038,
      "num_tokens": 9543754.0,
      "step": 99
    },
    {
      "epoch": 0.46003450258769407,
      "grad_norm": 0.010464823804795742,
      "learning_rate": 9.517252173051912e-06,
      "loss": 1.1736,
      "mean_token_accuracy": 0.6762436330318451,
      "num_tokens": 9640369.0,
      "step": 100
    },
    {
      "epoch": 0.46463484761357104,
      "grad_norm": 0.010858048684895039,
      "learning_rate": 9.499839423831062e-06,
      "loss": 1.1636,
      "mean_token_accuracy": 0.6757931411266327,
      "num_tokens": 9732896.0,
      "step": 101
    },
    {
      "epoch": 0.46923519263944796,
      "grad_norm": 0.01097831316292286,
      "learning_rate": 9.48213468647852e-06,
      "loss": 1.1372,
      "mean_token_accuracy": 0.6813743412494659,
      "num_tokens": 9828655.0,
      "step": 102
    },
    {
      "epoch": 0.4738355376653249,
      "grad_norm": 0.009854613803327084,
      "learning_rate": 9.46413910982932e-06,
      "loss": 1.1422,
      "mean_token_accuracy": 0.6805589944124222,
      "num_tokens": 9925327.0,
      "step": 103
    },
    {
      "epoch": 0.47843588269120185,
      "grad_norm": 0.011986501514911652,
      "learning_rate": 9.445853861590647e-06,
      "loss": 1.1552,
      "mean_token_accuracy": 0.6841816902160645,
      "num_tokens": 10019986.0,
      "step": 104
    },
    {
      "epoch": 0.48303622771707877,
      "grad_norm": 0.009618732146918774,
      "learning_rate": 9.427280128266049e-06,
      "loss": 1.114,
      "mean_token_accuracy": 0.6898596957325935,
      "num_tokens": 10120938.0,
      "step": 105
    },
    {
      "epoch": 0.48763657274295574,
      "grad_norm": 0.011056702584028244,
      "learning_rate": 9.40841911507847e-06,
      "loss": 1.1625,
      "mean_token_accuracy": 0.6775639355182648,
      "num_tokens": 10216437.0,
      "step": 106
    },
    {
      "epoch": 0.49223691776883266,
      "grad_norm": 0.01116709690541029,
      "learning_rate": 9.389272045892023e-06,
      "loss": 1.145,
      "mean_token_accuracy": 0.6837708950042725,
      "num_tokens": 10311173.0,
      "step": 107
    },
    {
      "epoch": 0.4968372627947096,
      "grad_norm": 0.011039802804589272,
      "learning_rate": 9.36984016313259e-06,
      "loss": 1.1417,
      "mean_token_accuracy": 0.6808711588382721,
      "num_tokens": 10407597.0,
      "step": 108
    },
    {
      "epoch": 0.5014376078205866,
      "grad_norm": 0.010891326703131199,
      "learning_rate": 9.350124727707197e-06,
      "loss": 1.168,
      "mean_token_accuracy": 0.6797266602516174,
      "num_tokens": 10508399.0,
      "step": 109
    },
    {
      "epoch": 0.5060379528464635,
      "grad_norm": 0.011290277354419231,
      "learning_rate": 9.330127018922195e-06,
      "loss": 1.1601,
      "mean_token_accuracy": 0.6797808483242989,
      "num_tokens": 10607967.0,
      "step": 110
    },
    {
      "epoch": 0.5106382978723404,
      "grad_norm": 0.01199404802173376,
      "learning_rate": 9.309848334400247e-06,
      "loss": 1.1465,
      "mean_token_accuracy": 0.6825785636901855,
      "num_tokens": 10700032.0,
      "step": 111
    },
    {
      "epoch": 0.5152386428982174,
      "grad_norm": 0.011274462565779686,
      "learning_rate": 9.289289989996133e-06,
      "loss": 1.1476,
      "mean_token_accuracy": 0.6835216283798218,
      "num_tokens": 10793853.0,
      "step": 112
    },
    {
      "epoch": 0.5198389879240943,
      "grad_norm": 0.010716607794165611,
      "learning_rate": 9.268453319711362e-06,
      "loss": 1.137,
      "mean_token_accuracy": 0.6838621497154236,
      "num_tokens": 10893892.0,
      "step": 113
    },
    {
      "epoch": 0.5244393329499712,
      "grad_norm": 0.011201828718185425,
      "learning_rate": 9.247339675607606e-06,
      "loss": 1.1673,
      "mean_token_accuracy": 0.6785961911082268,
      "num_tokens": 10990079.0,
      "step": 114
    },
    {
      "epoch": 0.5290396779758482,
      "grad_norm": 0.012047258205711842,
      "learning_rate": 9.225950427718974e-06,
      "loss": 1.141,
      "mean_token_accuracy": 0.6847905442118645,
      "num_tokens": 11087797.0,
      "step": 115
    },
    {
      "epoch": 0.5336400230017251,
      "grad_norm": 0.009881600737571716,
      "learning_rate": 9.204286963963112e-06,
      "loss": 1.1815,
      "mean_token_accuracy": 0.6752007529139519,
      "num_tokens": 11186382.0,
      "step": 116
    },
    {
      "epoch": 0.5382403680276021,
      "grad_norm": 0.011906330473721027,
      "learning_rate": 9.182350690051134e-06,
      "loss": 1.1841,
      "mean_token_accuracy": 0.6728686764836311,
      "num_tokens": 11280569.0,
      "step": 117
    },
    {
      "epoch": 0.542840713053479,
      "grad_norm": 0.01070008147507906,
      "learning_rate": 9.160143029396422e-06,
      "loss": 1.1368,
      "mean_token_accuracy": 0.6841261833906174,
      "num_tokens": 11375905.0,
      "step": 118
    },
    {
      "epoch": 0.547441058079356,
      "grad_norm": 0.01177423819899559,
      "learning_rate": 9.13766542302225e-06,
      "loss": 1.1204,
      "mean_token_accuracy": 0.6871694549918175,
      "num_tokens": 11469920.0,
      "step": 119
    },
    {
      "epoch": 0.5520414031052329,
      "grad_norm": 0.011396740563213825,
      "learning_rate": 9.114919329468283e-06,
      "loss": 1.1295,
      "mean_token_accuracy": 0.6842972412705421,
      "num_tokens": 11566631.0,
      "step": 120
    },
    {
      "epoch": 0.5566417481311098,
      "grad_norm": 0.011250436305999756,
      "learning_rate": 9.091906224695935e-06,
      "loss": 1.1233,
      "mean_token_accuracy": 0.6875024884939194,
      "num_tokens": 11663849.0,
      "step": 121
    },
    {
      "epoch": 0.5612420931569868,
      "grad_norm": 0.01271882839500904,
      "learning_rate": 9.0686276019926e-06,
      "loss": 1.1367,
      "mean_token_accuracy": 0.6822795867919922,
      "num_tokens": 11758571.0,
      "step": 122
    },
    {
      "epoch": 0.5658424381828637,
      "grad_norm": 0.01079560723155737,
      "learning_rate": 9.045084971874738e-06,
      "loss": 1.1209,
      "mean_token_accuracy": 0.6869323998689651,
      "num_tokens": 11858943.0,
      "step": 123
    },
    {
      "epoch": 0.5704427832087406,
      "grad_norm": 0.010741114616394043,
      "learning_rate": 9.021279861989884e-06,
      "loss": 1.1339,
      "mean_token_accuracy": 0.681599348783493,
      "num_tokens": 11959125.0,
      "step": 124
    },
    {
      "epoch": 0.5750431282346176,
      "grad_norm": 0.011083470657467842,
      "learning_rate": 8.997213817017508e-06,
      "loss": 1.1707,
      "mean_token_accuracy": 0.6807960867881775,
      "num_tokens": 12050647.0,
      "step": 125
    },
    {
      "epoch": 0.5796434732604946,
      "grad_norm": 0.012295430526137352,
      "learning_rate": 8.972888398568772e-06,
      "loss": 1.1365,
      "mean_token_accuracy": 0.682566873729229,
      "num_tokens": 12147275.0,
      "step": 126
    },
    {
      "epoch": 0.5842438182863715,
      "grad_norm": 0.012151171453297138,
      "learning_rate": 8.948305185085226e-06,
      "loss": 1.1393,
      "mean_token_accuracy": 0.6846441850066185,
      "num_tokens": 12245419.0,
      "step": 127
    },
    {
      "epoch": 0.5888441633122484,
      "grad_norm": 0.01221384946256876,
      "learning_rate": 8.923465771736361e-06,
      "loss": 1.1423,
      "mean_token_accuracy": 0.6798125579953194,
      "num_tokens": 12345045.0,
      "step": 128
    },
    {
      "epoch": 0.5934445083381253,
      "grad_norm": 0.010475218296051025,
      "learning_rate": 8.898371770316113e-06,
      "loss": 1.1229,
      "mean_token_accuracy": 0.6866893246769905,
      "num_tokens": 12440926.0,
      "step": 129
    },
    {
      "epoch": 0.5980448533640023,
      "grad_norm": 0.010910892859101295,
      "learning_rate": 8.873024809138272e-06,
      "loss": 1.135,
      "mean_token_accuracy": 0.6809633672237396,
      "num_tokens": 12538689.0,
      "step": 130
    },
    {
      "epoch": 0.6026451983898793,
      "grad_norm": 0.012574547901749611,
      "learning_rate": 8.84742653293083e-06,
      "loss": 1.1386,
      "mean_token_accuracy": 0.6841543838381767,
      "num_tokens": 12633929.0,
      "step": 131
    },
    {
      "epoch": 0.6072455434157562,
      "grad_norm": 0.0117676155641675,
      "learning_rate": 8.821578602729242e-06,
      "loss": 1.1872,
      "mean_token_accuracy": 0.6732319965958595,
      "num_tokens": 12730071.0,
      "step": 132
    },
    {
      "epoch": 0.6118458884416331,
      "grad_norm": 0.01177041232585907,
      "learning_rate": 8.795482695768658e-06,
      "loss": 1.1557,
      "mean_token_accuracy": 0.6811974123120308,
      "num_tokens": 12827268.0,
      "step": 133
    },
    {
      "epoch": 0.61644623346751,
      "grad_norm": 0.012073309160768986,
      "learning_rate": 8.769140505375084e-06,
      "loss": 1.1482,
      "mean_token_accuracy": 0.6824083104729652,
      "num_tokens": 12920648.0,
      "step": 134
    },
    {
      "epoch": 0.621046578493387,
      "grad_norm": 0.013189052231609821,
      "learning_rate": 8.742553740855507e-06,
      "loss": 1.1202,
      "mean_token_accuracy": 0.6862241849303246,
      "num_tokens": 13010217.0,
      "step": 135
    },
    {
      "epoch": 0.625646923519264,
      "grad_norm": 0.017087044194340706,
      "learning_rate": 8.715724127386971e-06,
      "loss": 1.1678,
      "mean_token_accuracy": 0.6766590625047684,
      "num_tokens": 13108063.0,
      "step": 136
    },
    {
      "epoch": 0.6302472685451409,
      "grad_norm": 0.013833911158144474,
      "learning_rate": 8.688653405904653e-06,
      "loss": 1.139,
      "mean_token_accuracy": 0.6834281384944916,
      "num_tokens": 13204017.0,
      "step": 137
    },
    {
      "epoch": 0.6348476135710178,
      "grad_norm": 0.011891160160303116,
      "learning_rate": 8.661343332988869e-06,
      "loss": 1.1432,
      "mean_token_accuracy": 0.6826580688357353,
      "num_tokens": 13298917.0,
      "step": 138
    },
    {
      "epoch": 0.6394479585968947,
      "grad_norm": 0.010969611816108227,
      "learning_rate": 8.633795680751116e-06,
      "loss": 1.1384,
      "mean_token_accuracy": 0.6838229894638062,
      "num_tokens": 13398469.0,
      "step": 139
    },
    {
      "epoch": 0.6440483036227717,
      "grad_norm": 0.011026185005903244,
      "learning_rate": 8.606012236719073e-06,
      "loss": 1.1356,
      "mean_token_accuracy": 0.6841299682855606,
      "num_tokens": 13491680.0,
      "step": 140
    },
    {
      "epoch": 0.6486486486486487,
      "grad_norm": 0.011263255029916763,
      "learning_rate": 8.577994803720605e-06,
      "loss": 1.129,
      "mean_token_accuracy": 0.6875933781266212,
      "num_tokens": 13590200.0,
      "step": 141
    },
    {
      "epoch": 0.6532489936745256,
      "grad_norm": 0.011134397238492966,
      "learning_rate": 8.549745199766792e-06,
      "loss": 1.1408,
      "mean_token_accuracy": 0.682669073343277,
      "num_tokens": 13687397.0,
      "step": 142
    },
    {
      "epoch": 0.6578493387004025,
      "grad_norm": 0.012590516358613968,
      "learning_rate": 8.521265257933948e-06,
      "loss": 1.1685,
      "mean_token_accuracy": 0.677878700196743,
      "num_tokens": 13783952.0,
      "step": 143
    },
    {
      "epoch": 0.6624496837262794,
      "grad_norm": 0.012606673873960972,
      "learning_rate": 8.492556826244687e-06,
      "loss": 1.1415,
      "mean_token_accuracy": 0.6823642030358315,
      "num_tokens": 13878727.0,
      "step": 144
    },
    {
      "epoch": 0.6670500287521565,
      "grad_norm": 0.011452843435108662,
      "learning_rate": 8.463621767547998e-06,
      "loss": 1.115,
      "mean_token_accuracy": 0.6876819431781769,
      "num_tokens": 13973528.0,
      "step": 145
    },
    {
      "epoch": 0.6716503737780334,
      "grad_norm": 0.013077128678560257,
      "learning_rate": 8.434461959398377e-06,
      "loss": 1.1375,
      "mean_token_accuracy": 0.6831812486052513,
      "num_tokens": 14067816.0,
      "step": 146
    },
    {
      "epoch": 0.6762507188039103,
      "grad_norm": 0.012629431672394276,
      "learning_rate": 8.405079293933986e-06,
      "loss": 1.1549,
      "mean_token_accuracy": 0.6779226139187813,
      "num_tokens": 14166980.0,
      "step": 147
    },
    {
      "epoch": 0.6808510638297872,
      "grad_norm": 0.01266096904873848,
      "learning_rate": 8.375475677753882e-06,
      "loss": 1.1391,
      "mean_token_accuracy": 0.6808317601680756,
      "num_tokens": 14262155.0,
      "step": 148
    },
    {
      "epoch": 0.6854514088556641,
      "grad_norm": 0.012114723213016987,
      "learning_rate": 8.345653031794292e-06,
      "loss": 1.1339,
      "mean_token_accuracy": 0.6848656088113785,
      "num_tokens": 14357686.0,
      "step": 149
    },
    {
      "epoch": 0.6900517538815412,
      "grad_norm": 0.01152091845870018,
      "learning_rate": 8.315613291203977e-06,
      "loss": 1.1362,
      "mean_token_accuracy": 0.6844386830925941,
      "num_tokens": 14452969.0,
      "step": 150
    },
    {
      "epoch": 0.6946520989074181,
      "grad_norm": 0.011453399434685707,
      "learning_rate": 8.285358405218655e-06,
      "loss": 1.1354,
      "mean_token_accuracy": 0.6829889118671417,
      "num_tokens": 14555977.0,
      "step": 151
    },
    {
      "epoch": 0.699252443933295,
      "grad_norm": 0.01063123345375061,
      "learning_rate": 8.25489033703452e-06,
      "loss": 1.1511,
      "mean_token_accuracy": 0.6795672103762627,
      "num_tokens": 14656890.0,
      "step": 152
    },
    {
      "epoch": 0.7038527889591719,
      "grad_norm": 0.01153411716222763,
      "learning_rate": 8.224211063680854e-06,
      "loss": 1.1349,
      "mean_token_accuracy": 0.6822441294789314,
      "num_tokens": 14756458.0,
      "step": 153
    },
    {
      "epoch": 0.7084531339850488,
      "grad_norm": 0.012573798187077045,
      "learning_rate": 8.19332257589174e-06,
      "loss": 1.1318,
      "mean_token_accuracy": 0.6857383400201797,
      "num_tokens": 14847811.0,
      "step": 154
    },
    {
      "epoch": 0.7130534790109259,
      "grad_norm": 0.015371086075901985,
      "learning_rate": 8.162226877976886e-06,
      "loss": 1.1367,
      "mean_token_accuracy": 0.6819384023547173,
      "num_tokens": 14944682.0,
      "step": 155
    },
    {
      "epoch": 0.7176538240368028,
      "grad_norm": 0.012037052772939205,
      "learning_rate": 8.13092598769157e-06,
      "loss": 1.0943,
      "mean_token_accuracy": 0.6912998482584953,
      "num_tokens": 15044543.0,
      "step": 156
    },
    {
      "epoch": 0.7222541690626797,
      "grad_norm": 0.012677938677370548,
      "learning_rate": 8.099421936105702e-06,
      "loss": 1.1261,
      "mean_token_accuracy": 0.6894383281469345,
      "num_tokens": 15135392.0,
      "step": 157
    },
    {
      "epoch": 0.7268545140885566,
      "grad_norm": 0.013612372800707817,
      "learning_rate": 8.067716767472045e-06,
      "loss": 1.1509,
      "mean_token_accuracy": 0.6794408410787582,
      "num_tokens": 15230933.0,
      "step": 158
    },
    {
      "epoch": 0.7314548591144335,
      "grad_norm": 0.012262539006769657,
      "learning_rate": 8.035812539093557e-06,
      "loss": 1.1656,
      "mean_token_accuracy": 0.6754115670919418,
      "num_tokens": 15326416.0,
      "step": 159
    },
    {
      "epoch": 0.7360552041403106,
      "grad_norm": 0.014869904145598412,
      "learning_rate": 8.003711321189895e-06,
      "loss": 1.1469,
      "mean_token_accuracy": 0.6832912117242813,
      "num_tokens": 15418479.0,
      "step": 160
    },
    {
      "epoch": 0.7406555491661875,
      "grad_norm": 0.011351753026247025,
      "learning_rate": 7.971415196763088e-06,
      "loss": 1.1279,
      "mean_token_accuracy": 0.6886353641748428,
      "num_tokens": 15518079.0,
      "step": 161
    },
    {
      "epoch": 0.7452558941920644,
      "grad_norm": 0.011147459968924522,
      "learning_rate": 7.938926261462366e-06,
      "loss": 1.0905,
      "mean_token_accuracy": 0.6901291012763977,
      "num_tokens": 15611875.0,
      "step": 162
    },
    {
      "epoch": 0.7498562392179413,
      "grad_norm": 0.013785228133201599,
      "learning_rate": 7.906246623448184e-06,
      "loss": 1.1525,
      "mean_token_accuracy": 0.6813407689332962,
      "num_tokens": 15707848.0,
      "step": 163
    },
    {
      "epoch": 0.7544565842438183,
      "grad_norm": 0.01464665774255991,
      "learning_rate": 7.87337840325542e-06,
      "loss": 1.1265,
      "mean_token_accuracy": 0.683470606803894,
      "num_tokens": 15796371.0,
      "step": 164
    },
    {
      "epoch": 0.7590569292696953,
      "grad_norm": 0.011676521971821785,
      "learning_rate": 7.84032373365578e-06,
      "loss": 1.1639,
      "mean_token_accuracy": 0.6822836399078369,
      "num_tokens": 15892440.0,
      "step": 165
    },
    {
      "epoch": 0.7636572742955722,
      "grad_norm": 0.013809471391141415,
      "learning_rate": 7.807084759519405e-06,
      "loss": 1.1393,
      "mean_token_accuracy": 0.6840439140796661,
      "num_tokens": 15989888.0,
      "step": 166
    },
    {
      "epoch": 0.7682576193214491,
      "grad_norm": 0.013256164267659187,
      "learning_rate": 7.773663637675695e-06,
      "loss": 1.1323,
      "mean_token_accuracy": 0.6872632876038551,
      "num_tokens": 16088051.0,
      "step": 167
    },
    {
      "epoch": 0.772857964347326,
      "grad_norm": 0.013578861020505428,
      "learning_rate": 7.740062536773352e-06,
      "loss": 1.1514,
      "mean_token_accuracy": 0.6812737733125687,
      "num_tokens": 16185711.0,
      "step": 168
    },
    {
      "epoch": 0.777458309373203,
      "grad_norm": 0.012444855645298958,
      "learning_rate": 7.706283637139658e-06,
      "loss": 1.1495,
      "mean_token_accuracy": 0.6834460496902466,
      "num_tokens": 16280860.0,
      "step": 169
    },
    {
      "epoch": 0.78205865439908,
      "grad_norm": 0.012594765983521938,
      "learning_rate": 7.672329130639007e-06,
      "loss": 1.1296,
      "mean_token_accuracy": 0.6830948293209076,
      "num_tokens": 16374245.0,
      "step": 170
    },
    {
      "epoch": 0.7866589994249569,
      "grad_norm": 0.013151990249752998,
      "learning_rate": 7.638201220530664e-06,
      "loss": 1.1253,
      "mean_token_accuracy": 0.688128337264061,
      "num_tokens": 16468182.0,
      "step": 171
    },
    {
      "epoch": 0.7912593444508338,
      "grad_norm": 0.013742980547249317,
      "learning_rate": 7.603902121325813e-06,
      "loss": 1.1271,
      "mean_token_accuracy": 0.6843288838863373,
      "num_tokens": 16561819.0,
      "step": 172
    },
    {
      "epoch": 0.7958596894767107,
      "grad_norm": 0.014118033461272717,
      "learning_rate": 7.5694340586438446e-06,
      "loss": 1.1218,
      "mean_token_accuracy": 0.6890603974461555,
      "num_tokens": 16656202.0,
      "step": 173
    },
    {
      "epoch": 0.8004600345025877,
      "grad_norm": 0.012105434201657772,
      "learning_rate": 7.534799269067952e-06,
      "loss": 1.1091,
      "mean_token_accuracy": 0.6900731474161148,
      "num_tokens": 16749603.0,
      "step": 174
    },
    {
      "epoch": 0.8050603795284647,
      "grad_norm": 0.012348281219601631,
      "learning_rate": 7.500000000000001e-06,
      "loss": 1.1282,
      "mean_token_accuracy": 0.6861552968621254,
      "num_tokens": 16845665.0,
      "step": 175
    },
    {
      "epoch": 0.8096607245543416,
      "grad_norm": 0.013485051691532135,
      "learning_rate": 7.465038509514688e-06,
      "loss": 1.1042,
      "mean_token_accuracy": 0.691558264195919,
      "num_tokens": 16938887.0,
      "step": 176
    },
    {
      "epoch": 0.8142610695802185,
      "grad_norm": 0.013376619666814804,
      "learning_rate": 7.42991706621303e-06,
      "loss": 1.1247,
      "mean_token_accuracy": 0.6862233057618141,
      "num_tokens": 17034004.0,
      "step": 177
    },
    {
      "epoch": 0.8188614146060954,
      "grad_norm": 0.013932105153799057,
      "learning_rate": 7.3946379490751545e-06,
      "loss": 1.1544,
      "mean_token_accuracy": 0.6811259537935257,
      "num_tokens": 17134895.0,
      "step": 178
    },
    {
      "epoch": 0.8234617596319724,
      "grad_norm": 0.013256735168397427,
      "learning_rate": 7.35920344731241e-06,
      "loss": 1.1575,
      "mean_token_accuracy": 0.679460808634758,
      "num_tokens": 17235475.0,
      "step": 179
    },
    {
      "epoch": 0.8280621046578494,
      "grad_norm": 0.011622334830462933,
      "learning_rate": 7.323615860218844e-06,
      "loss": 1.1184,
      "mean_token_accuracy": 0.6867079436779022,
      "num_tokens": 17335072.0,
      "step": 180
    },
    {
      "epoch": 0.8326624496837263,
      "grad_norm": 0.01442211028188467,
      "learning_rate": 7.287877497021978e-06,
      "loss": 1.1415,
      "mean_token_accuracy": 0.6843341290950775,
      "num_tokens": 17433348.0,
      "step": 181
    },
    {
      "epoch": 0.8372627947096032,
      "grad_norm": 0.012319355271756649,
      "learning_rate": 7.251990676732985e-06,
      "loss": 1.1413,
      "mean_token_accuracy": 0.6828998699784279,
      "num_tokens": 17534045.0,
      "step": 182
    },
    {
      "epoch": 0.8418631397354802,
      "grad_norm": 0.014468186534941196,
      "learning_rate": 7.215957727996208e-06,
      "loss": 1.1267,
      "mean_token_accuracy": 0.6874820590019226,
      "num_tokens": 17630770.0,
      "step": 183
    },
    {
      "epoch": 0.8464634847613571,
      "grad_norm": 0.013780116103589535,
      "learning_rate": 7.179780988938051e-06,
      "loss": 1.1264,
      "mean_token_accuracy": 0.6865449622273445,
      "num_tokens": 17721005.0,
      "step": 184
    },
    {
      "epoch": 0.851063829787234,
      "grad_norm": 0.013034569099545479,
      "learning_rate": 7.143462807015271e-06,
      "loss": 1.1428,
      "mean_token_accuracy": 0.6805073171854019,
      "num_tokens": 17818303.0,
      "step": 185
    },
    {
      "epoch": 0.855664174813111,
      "grad_norm": 0.014521626755595207,
      "learning_rate": 7.107005538862647e-06,
      "loss": 1.1435,
      "mean_token_accuracy": 0.6784381717443466,
      "num_tokens": 17911334.0,
      "step": 186
    },
    {
      "epoch": 0.8602645198389879,
      "grad_norm": 0.012255755253136158,
      "learning_rate": 7.07041155014006e-06,
      "loss": 1.119,
      "mean_token_accuracy": 0.6881618723273277,
      "num_tokens": 18003358.0,
      "step": 187
    },
    {
      "epoch": 0.8648648648648649,
      "grad_norm": 0.011319384910166264,
      "learning_rate": 7.033683215379002e-06,
      "loss": 1.1473,
      "mean_token_accuracy": 0.6812754571437836,
      "num_tokens": 18099567.0,
      "step": 188
    },
    {
      "epoch": 0.8694652098907418,
      "grad_norm": 0.012943817302584648,
      "learning_rate": 6.9968229178284775e-06,
      "loss": 1.124,
      "mean_token_accuracy": 0.6847608834505081,
      "num_tokens": 18193605.0,
      "step": 189
    },
    {
      "epoch": 0.8740655549166187,
      "grad_norm": 0.014849803410470486,
      "learning_rate": 6.959833049300376e-06,
      "loss": 1.1638,
      "mean_token_accuracy": 0.6793822422623634,
      "num_tokens": 18291528.0,
      "step": 190
    },
    {
      "epoch": 0.8786658999424957,
      "grad_norm": 0.013070415705442429,
      "learning_rate": 6.922716010014256e-06,
      "loss": 1.1657,
      "mean_token_accuracy": 0.6762072518467903,
      "num_tokens": 18387717.0,
      "step": 191
    },
    {
      "epoch": 0.8832662449683726,
      "grad_norm": 0.015733331441879272,
      "learning_rate": 6.885474208441602e-06,
      "loss": 1.1427,
      "mean_token_accuracy": 0.6810364499688148,
      "num_tokens": 18482986.0,
      "step": 192
    },
    {
      "epoch": 0.8878665899942496,
      "grad_norm": 0.014921659603714943,
      "learning_rate": 6.848110061149555e-06,
      "loss": 1.1138,
      "mean_token_accuracy": 0.686975933611393,
      "num_tokens": 18578439.0,
      "step": 193
    },
    {
      "epoch": 0.8924669350201265,
      "grad_norm": 0.012976438738405704,
      "learning_rate": 6.810625992644085e-06,
      "loss": 1.136,
      "mean_token_accuracy": 0.6828045099973679,
      "num_tokens": 18673147.0,
      "step": 194
    },
    {
      "epoch": 0.8970672800460034,
      "grad_norm": 0.013467282988131046,
      "learning_rate": 6.773024435212678e-06,
      "loss": 1.1548,
      "mean_token_accuracy": 0.6820717006921768,
      "num_tokens": 18769013.0,
      "step": 195
    },
    {
      "epoch": 0.9016676250718804,
      "grad_norm": 0.013113681226968765,
      "learning_rate": 6.735307828766515e-06,
      "loss": 1.162,
      "mean_token_accuracy": 0.6819472461938858,
      "num_tokens": 18866436.0,
      "step": 196
    },
    {
      "epoch": 0.9062679700977573,
      "grad_norm": 0.012697749771177769,
      "learning_rate": 6.697478620682137e-06,
      "loss": 1.1221,
      "mean_token_accuracy": 0.6850037425756454,
      "num_tokens": 18961175.0,
      "step": 197
    },
    {
      "epoch": 0.9108683151236343,
      "grad_norm": 0.013651424087584019,
      "learning_rate": 6.659539265642643e-06,
      "loss": 1.1327,
      "mean_token_accuracy": 0.683879666030407,
      "num_tokens": 19053487.0,
      "step": 198
    },
    {
      "epoch": 0.9154686601495112,
      "grad_norm": 0.01471368595957756,
      "learning_rate": 6.6214922254784145e-06,
      "loss": 1.1362,
      "mean_token_accuracy": 0.6838243156671524,
      "num_tokens": 19154536.0,
      "step": 199
    },
    {
      "epoch": 0.9200690051753881,
      "grad_norm": 0.013274794444441795,
      "learning_rate": 6.583339969007364e-06,
      "loss": 1.1419,
      "mean_token_accuracy": 0.6834047809243202,
      "num_tokens": 19242500.0,
      "step": 200
    },
    {
      "epoch": 0.9246693502012651,
      "grad_norm": 0.013419239781796932,
      "learning_rate": 6.545084971874738e-06,
      "loss": 1.1361,
      "mean_token_accuracy": 0.6821422278881073,
      "num_tokens": 19338362.0,
      "step": 201
    },
    {
      "epoch": 0.9292696952271421,
      "grad_norm": 0.01378414686769247,
      "learning_rate": 6.50672971639248e-06,
      "loss": 1.1408,
      "mean_token_accuracy": 0.680683322250843,
      "num_tokens": 19430444.0,
      "step": 202
    },
    {
      "epoch": 0.933870040253019,
      "grad_norm": 0.013185984455049038,
      "learning_rate": 6.468276691378155e-06,
      "loss": 1.1471,
      "mean_token_accuracy": 0.6821170374751091,
      "num_tokens": 19527300.0,
      "step": 203
    },
    {
      "epoch": 0.9384703852788959,
      "grad_norm": 0.014027146622538567,
      "learning_rate": 6.429728391993446e-06,
      "loss": 1.1402,
      "mean_token_accuracy": 0.6847390085458755,
      "num_tokens": 19623076.0,
      "step": 204
    },
    {
      "epoch": 0.9430707303047728,
      "grad_norm": 0.014988254755735397,
      "learning_rate": 6.391087319582264e-06,
      "loss": 1.1237,
      "mean_token_accuracy": 0.688391737639904,
      "num_tokens": 19717760.0,
      "step": 205
    },
    {
      "epoch": 0.9476710753306498,
      "grad_norm": 0.013494355604052544,
      "learning_rate": 6.35235598150842e-06,
      "loss": 1.1441,
      "mean_token_accuracy": 0.6846969500184059,
      "num_tokens": 19820465.0,
      "step": 206
    },
    {
      "epoch": 0.9522714203565268,
      "grad_norm": 0.012427960522472858,
      "learning_rate": 6.313536890992935e-06,
      "loss": 1.1321,
      "mean_token_accuracy": 0.6843677386641502,
      "num_tokens": 19919080.0,
      "step": 207
    },
    {
      "epoch": 0.9568717653824037,
      "grad_norm": 0.013420927338302135,
      "learning_rate": 6.274632566950967e-06,
      "loss": 1.1541,
      "mean_token_accuracy": 0.6806612834334373,
      "num_tokens": 20012947.0,
      "step": 208
    },
    {
      "epoch": 0.9614721104082806,
      "grad_norm": 0.013078980147838593,
      "learning_rate": 6.235645533828348e-06,
      "loss": 1.1041,
      "mean_token_accuracy": 0.6901779100298882,
      "num_tokens": 20112769.0,
      "step": 209
    },
    {
      "epoch": 0.9660724554341575,
      "grad_norm": 0.013666641898453236,
      "learning_rate": 6.1965783214377895e-06,
      "loss": 1.1299,
      "mean_token_accuracy": 0.6861997470259666,
      "num_tokens": 20210131.0,
      "step": 210
    },
    {
      "epoch": 0.9706728004600345,
      "grad_norm": 0.014031516388058662,
      "learning_rate": 6.157433464794717e-06,
      "loss": 1.1481,
      "mean_token_accuracy": 0.6819183602929115,
      "num_tokens": 20307517.0,
      "step": 211
    },
    {
      "epoch": 0.9752731454859115,
      "grad_norm": 0.014556187205016613,
      "learning_rate": 6.118213503952779e-06,
      "loss": 1.1727,
      "mean_token_accuracy": 0.6784057542681694,
      "num_tokens": 20407086.0,
      "step": 212
    },
    {
      "epoch": 0.9798734905117884,
      "grad_norm": 0.014544697478413582,
      "learning_rate": 6.078920983839032e-06,
      "loss": 1.1414,
      "mean_token_accuracy": 0.6819159612059593,
      "num_tokens": 20502784.0,
      "step": 213
    },
    {
      "epoch": 0.9844738355376653,
      "grad_norm": 0.014732340350747108,
      "learning_rate": 6.039558454088796e-06,
      "loss": 1.156,
      "mean_token_accuracy": 0.6811837628483772,
      "num_tokens": 20599039.0,
      "step": 214
    },
    {
      "epoch": 0.9890741805635422,
      "grad_norm": 0.014581098221242428,
      "learning_rate": 6.000128468880223e-06,
      "loss": 1.1515,
      "mean_token_accuracy": 0.6804311126470566,
      "num_tokens": 20692487.0,
      "step": 215
    },
    {
      "epoch": 0.9936745255894192,
      "grad_norm": 0.012514597736299038,
      "learning_rate": 5.9606335867685424e-06,
      "loss": 1.1308,
      "mean_token_accuracy": 0.685503862798214,
      "num_tokens": 20793938.0,
      "step": 216
    },
    {
      "epoch": 0.9982748706152962,
      "grad_norm": 0.014343786984682083,
      "learning_rate": 5.921076370520058e-06,
      "loss": 1.1707,
      "mean_token_accuracy": 0.6801939308643341,
      "num_tokens": 20887727.0,
      "step": 217
    },
    {
      "epoch": 1.004600345025877,
      "grad_norm": 0.016213279217481613,
      "learning_rate": 5.8814593869458455e-06,
      "loss": 2.2427,
      "mean_token_accuracy": 0.6866950869560242,
      "num_tokens": 21004456.0,
      "step": 218
    },
    {
      "epoch": 1.0092006900517538,
      "grad_norm": 0.013519300147891045,
      "learning_rate": 5.841785206735192e-06,
      "loss": 1.1294,
      "mean_token_accuracy": 0.6871752962470055,
      "num_tokens": 21099553.0,
      "step": 219
    },
    {
      "epoch": 1.0138010350776308,
      "grad_norm": 0.01289405021816492,
      "learning_rate": 5.8020564042888015e-06,
      "loss": 1.174,
      "mean_token_accuracy": 0.6760912984609604,
      "num_tokens": 21193220.0,
      "step": 220
    },
    {
      "epoch": 1.0184013801035077,
      "grad_norm": 0.012278413400053978,
      "learning_rate": 5.762275557551728e-06,
      "loss": 1.1335,
      "mean_token_accuracy": 0.6865715309977531,
      "num_tokens": 21292513.0,
      "step": 221
    },
    {
      "epoch": 1.0230017251293848,
      "grad_norm": 0.01394630316644907,
      "learning_rate": 5.722445247846107e-06,
      "loss": 1.1277,
      "mean_token_accuracy": 0.6879585757851601,
      "num_tokens": 21386328.0,
      "step": 222
    },
    {
      "epoch": 1.0276020701552617,
      "grad_norm": 0.01193606574088335,
      "learning_rate": 5.682568059703659e-06,
      "loss": 1.1225,
      "mean_token_accuracy": 0.6864568889141083,
      "num_tokens": 21484561.0,
      "step": 223
    },
    {
      "epoch": 1.0322024151811386,
      "grad_norm": 0.014031891711056232,
      "learning_rate": 5.642646580697974e-06,
      "loss": 1.1372,
      "mean_token_accuracy": 0.6855478435754776,
      "num_tokens": 21580042.0,
      "step": 224
    },
    {
      "epoch": 1.0368027602070156,
      "grad_norm": 0.01325292233377695,
      "learning_rate": 5.6026834012766155e-06,
      "loss": 1.143,
      "mean_token_accuracy": 0.6811042875051498,
      "num_tokens": 21677310.0,
      "step": 225
    },
    {
      "epoch": 1.0414031052328925,
      "grad_norm": 0.012639901600778103,
      "learning_rate": 5.562681114593028e-06,
      "loss": 1.1522,
      "mean_token_accuracy": 0.6808114796876907,
      "num_tokens": 21774453.0,
      "step": 226
    },
    {
      "epoch": 1.0460034502587694,
      "grad_norm": 0.014555993489921093,
      "learning_rate": 5.522642316338268e-06,
      "loss": 1.1434,
      "mean_token_accuracy": 0.6824417561292648,
      "num_tokens": 21868348.0,
      "step": 227
    },
    {
      "epoch": 1.0506037952846463,
      "grad_norm": 0.01433889102190733,
      "learning_rate": 5.482569604572577e-06,
      "loss": 1.1537,
      "mean_token_accuracy": 0.6790372729301453,
      "num_tokens": 21966335.0,
      "step": 228
    },
    {
      "epoch": 1.0552041403105232,
      "grad_norm": 0.016252966597676277,
      "learning_rate": 5.442465579556793e-06,
      "loss": 1.1674,
      "mean_token_accuracy": 0.6783912926912308,
      "num_tokens": 22058574.0,
      "step": 229
    },
    {
      "epoch": 1.0598044853364001,
      "grad_norm": 0.014194665476679802,
      "learning_rate": 5.402332843583631e-06,
      "loss": 1.1083,
      "mean_token_accuracy": 0.6893611028790474,
      "num_tokens": 22155600.0,
      "step": 230
    },
    {
      "epoch": 1.0644048303622773,
      "grad_norm": 0.014426385052502155,
      "learning_rate": 5.362174000808813e-06,
      "loss": 1.1122,
      "mean_token_accuracy": 0.6882274597883224,
      "num_tokens": 22245005.0,
      "step": 231
    },
    {
      "epoch": 1.0690051753881542,
      "grad_norm": 0.012664509005844593,
      "learning_rate": 5.3219916570820976e-06,
      "loss": 1.1259,
      "mean_token_accuracy": 0.6857394576072693,
      "num_tokens": 22346244.0,
      "step": 232
    },
    {
      "epoch": 1.0736055204140311,
      "grad_norm": 0.01445621158927679,
      "learning_rate": 5.281788419778187e-06,
      "loss": 1.1699,
      "mean_token_accuracy": 0.6760598793625832,
      "num_tokens": 22442082.0,
      "step": 233
    },
    {
      "epoch": 1.078205865439908,
      "grad_norm": 0.014202049933373928,
      "learning_rate": 5.241566897627536e-06,
      "loss": 1.15,
      "mean_token_accuracy": 0.6811029836535454,
      "num_tokens": 22535337.0,
      "step": 234
    },
    {
      "epoch": 1.082806210465785,
      "grad_norm": 0.01301807351410389,
      "learning_rate": 5.201329700547077e-06,
      "loss": 1.096,
      "mean_token_accuracy": 0.6926997303962708,
      "num_tokens": 22631641.0,
      "step": 235
    },
    {
      "epoch": 1.0874065554916619,
      "grad_norm": 0.015671228989958763,
      "learning_rate": 5.1610794394708665e-06,
      "loss": 1.1017,
      "mean_token_accuracy": 0.6929657682776451,
      "num_tokens": 22725983.0,
      "step": 236
    },
    {
      "epoch": 1.0920069005175388,
      "grad_norm": 0.012980216182768345,
      "learning_rate": 5.120818726180662e-06,
      "loss": 1.1607,
      "mean_token_accuracy": 0.6774906814098358,
      "num_tokens": 22827099.0,
      "step": 237
    },
    {
      "epoch": 1.0966072455434157,
      "grad_norm": 0.01648387499153614,
      "learning_rate": 5.080550173136457e-06,
      "loss": 1.1342,
      "mean_token_accuracy": 0.6835969239473343,
      "num_tokens": 22917405.0,
      "step": 238
    },
    {
      "epoch": 1.1012075905692926,
      "grad_norm": 0.014721757732331753,
      "learning_rate": 5.0402763933069496e-06,
      "loss": 1.1352,
      "mean_token_accuracy": 0.6829187944531441,
      "num_tokens": 23015436.0,
      "step": 239
    },
    {
      "epoch": 1.1058079355951695,
      "grad_norm": 0.014279097318649292,
      "learning_rate": 5e-06,
      "loss": 1.1392,
      "mean_token_accuracy": 0.6804762333631516,
      "num_tokens": 23111568.0,
      "step": 240
    },
    {
      "epoch": 1.1104082806210465,
      "grad_norm": 0.014015192165970802,
      "learning_rate": 4.959723606693051e-06,
      "loss": 1.1415,
      "mean_token_accuracy": 0.6824319586157799,
      "num_tokens": 23206977.0,
      "step": 241
    },
    {
      "epoch": 1.1150086256469236,
      "grad_norm": 0.014435522258281708,
      "learning_rate": 4.919449826863544e-06,
      "loss": 1.1419,
      "mean_token_accuracy": 0.6825414225459099,
      "num_tokens": 23303985.0,
      "step": 242
    },
    {
      "epoch": 1.1196089706728005,
      "grad_norm": 0.014449840411543846,
      "learning_rate": 4.87918127381934e-06,
      "loss": 1.1272,
      "mean_token_accuracy": 0.6874236464500427,
      "num_tokens": 23402646.0,
      "step": 243
    },
    {
      "epoch": 1.1242093156986774,
      "grad_norm": 0.012992781586945057,
      "learning_rate": 4.838920560529137e-06,
      "loss": 1.1556,
      "mean_token_accuracy": 0.6821000725030899,
      "num_tokens": 23501515.0,
      "step": 244
    },
    {
      "epoch": 1.1288096607245544,
      "grad_norm": 0.01294787134975195,
      "learning_rate": 4.798670299452926e-06,
      "loss": 1.1236,
      "mean_token_accuracy": 0.6848902404308319,
      "num_tokens": 23599557.0,
      "step": 245
    },
    {
      "epoch": 1.1334100057504313,
      "grad_norm": 0.014253921806812286,
      "learning_rate": 4.758433102372466e-06,
      "loss": 1.1115,
      "mean_token_accuracy": 0.6874924078583717,
      "num_tokens": 23692128.0,
      "step": 246
    },
    {
      "epoch": 1.1380103507763082,
      "grad_norm": 0.014446437358856201,
      "learning_rate": 4.718211580221813e-06,
      "loss": 1.0993,
      "mean_token_accuracy": 0.6906405836343765,
      "num_tokens": 23783395.0,
      "step": 247
    },
    {
      "epoch": 1.142610695802185,
      "grad_norm": 0.014871901832520962,
      "learning_rate": 4.678008342917903e-06,
      "loss": 1.1617,
      "mean_token_accuracy": 0.6774732172489166,
      "num_tokens": 23870895.0,
      "step": 248
    },
    {
      "epoch": 1.147211040828062,
      "grad_norm": 0.012906315736472607,
      "learning_rate": 4.637825999191189e-06,
      "loss": 1.1493,
      "mean_token_accuracy": 0.6792214959859848,
      "num_tokens": 23967222.0,
      "step": 249
    },
    {
      "epoch": 1.151811385853939,
      "grad_norm": 0.01403460931032896,
      "learning_rate": 4.597667156416371e-06,
      "loss": 1.1165,
      "mean_token_accuracy": 0.6873978450894356,
      "num_tokens": 24065684.0,
      "step": 250
    },
    {
      "epoch": 1.156411730879816,
      "grad_norm": 0.015156279318034649,
      "learning_rate": 4.557534420443209e-06,
      "loss": 1.1243,
      "mean_token_accuracy": 0.6890147104859352,
      "num_tokens": 24162258.0,
      "step": 251
    },
    {
      "epoch": 1.161012075905693,
      "grad_norm": 0.013031619600951672,
      "learning_rate": 4.517430395427424e-06,
      "loss": 1.1404,
      "mean_token_accuracy": 0.6832455694675446,
      "num_tokens": 24260436.0,
      "step": 252
    },
    {
      "epoch": 1.16561242093157,
      "grad_norm": 0.012547873891890049,
      "learning_rate": 4.477357683661734e-06,
      "loss": 1.1615,
      "mean_token_accuracy": 0.6788115128874779,
      "num_tokens": 24359066.0,
      "step": 253
    },
    {
      "epoch": 1.1702127659574468,
      "grad_norm": 0.012914301827549934,
      "learning_rate": 4.4373188854069736e-06,
      "loss": 1.1265,
      "mean_token_accuracy": 0.685889832675457,
      "num_tokens": 24455995.0,
      "step": 254
    },
    {
      "epoch": 1.1748131109833237,
      "grad_norm": 0.013518140651285648,
      "learning_rate": 4.397316598723385e-06,
      "loss": 1.1116,
      "mean_token_accuracy": 0.6878758445382118,
      "num_tokens": 24552992.0,
      "step": 255
    },
    {
      "epoch": 1.1794134560092007,
      "grad_norm": 0.014318661764264107,
      "learning_rate": 4.357353419302028e-06,
      "loss": 1.1223,
      "mean_token_accuracy": 0.6877824738621712,
      "num_tokens": 24650651.0,
      "step": 256
    },
    {
      "epoch": 1.1840138010350776,
      "grad_norm": 0.01345856860280037,
      "learning_rate": 4.3174319402963436e-06,
      "loss": 1.0991,
      "mean_token_accuracy": 0.6909112930297852,
      "num_tokens": 24746354.0,
      "step": 257
    },
    {
      "epoch": 1.1886141460609545,
      "grad_norm": 0.014836768619716167,
      "learning_rate": 4.277554752153895e-06,
      "loss": 1.1036,
      "mean_token_accuracy": 0.688332587480545,
      "num_tokens": 24839448.0,
      "step": 258
    },
    {
      "epoch": 1.1932144910868314,
      "grad_norm": 0.016939718276262283,
      "learning_rate": 4.237724442448273e-06,
      "loss": 1.1138,
      "mean_token_accuracy": 0.6888449415564537,
      "num_tokens": 24940763.0,
      "step": 259
    },
    {
      "epoch": 1.1978148361127086,
      "grad_norm": 0.014445330947637558,
      "learning_rate": 4.1979435957111984e-06,
      "loss": 1.1269,
      "mean_token_accuracy": 0.6845501437783241,
      "num_tokens": 25039359.0,
      "step": 260
    },
    {
      "epoch": 1.2024151811385855,
      "grad_norm": 0.014106119982898235,
      "learning_rate": 4.158214793264808e-06,
      "loss": 1.1223,
      "mean_token_accuracy": 0.6866239085793495,
      "num_tokens": 25135799.0,
      "step": 261
    },
    {
      "epoch": 1.2070155261644624,
      "grad_norm": 0.013844680972397327,
      "learning_rate": 4.118540613054155e-06,
      "loss": 1.1395,
      "mean_token_accuracy": 0.6820477023720741,
      "num_tokens": 25232039.0,
      "step": 262
    },
    {
      "epoch": 1.2116158711903393,
      "grad_norm": 0.014573228545486927,
      "learning_rate": 4.0789236294799425e-06,
      "loss": 1.1059,
      "mean_token_accuracy": 0.6888193562626839,
      "num_tokens": 25326856.0,
      "step": 263
    },
    {
      "epoch": 1.2162162162162162,
      "grad_norm": 0.014026801101863384,
      "learning_rate": 4.039366413231458e-06,
      "loss": 1.134,
      "mean_token_accuracy": 0.686940111219883,
      "num_tokens": 25422559.0,
      "step": 264
    },
    {
      "epoch": 1.2208165612420931,
      "grad_norm": 0.014971627853810787,
      "learning_rate": 3.999871531119779e-06,
      "loss": 1.111,
      "mean_token_accuracy": 0.6868948489427567,
      "num_tokens": 25515580.0,
      "step": 265
    },
    {
      "epoch": 1.22541690626797,
      "grad_norm": 0.014933711849153042,
      "learning_rate": 3.960441545911205e-06,
      "loss": 1.1362,
      "mean_token_accuracy": 0.6853025332093239,
      "num_tokens": 25611398.0,
      "step": 266
    },
    {
      "epoch": 1.230017251293847,
      "grad_norm": 0.014098069630563259,
      "learning_rate": 3.92107901616097e-06,
      "loss": 1.1326,
      "mean_token_accuracy": 0.6831789240241051,
      "num_tokens": 25703520.0,
      "step": 267
    },
    {
      "epoch": 1.234617596319724,
      "grad_norm": 0.013797523453831673,
      "learning_rate": 3.881786496047224e-06,
      "loss": 1.1414,
      "mean_token_accuracy": 0.6835273951292038,
      "num_tokens": 25802295.0,
      "step": 268
    },
    {
      "epoch": 1.239217941345601,
      "grad_norm": 0.013530906289815903,
      "learning_rate": 3.842566535205286e-06,
      "loss": 1.1197,
      "mean_token_accuracy": 0.68748340010643,
      "num_tokens": 25898232.0,
      "step": 269
    },
    {
      "epoch": 1.243818286371478,
      "grad_norm": 0.01218091044574976,
      "learning_rate": 3.803421678562213e-06,
      "loss": 1.1028,
      "mean_token_accuracy": 0.6864416226744652,
      "num_tokens": 25997971.0,
      "step": 270
    },
    {
      "epoch": 1.2484186313973549,
      "grad_norm": 0.014679036103188992,
      "learning_rate": 3.7643544661716518e-06,
      "loss": 1.1251,
      "mean_token_accuracy": 0.6866933032870293,
      "num_tokens": 26097320.0,
      "step": 271
    },
    {
      "epoch": 1.2530189764232318,
      "grad_norm": 0.013114512898027897,
      "learning_rate": 3.725367433049033e-06,
      "loss": 1.1231,
      "mean_token_accuracy": 0.6846908032894135,
      "num_tokens": 26193610.0,
      "step": 272
    },
    {
      "epoch": 1.2576193214491087,
      "grad_norm": 0.014116368256509304,
      "learning_rate": 3.6864631090070656e-06,
      "loss": 1.1065,
      "mean_token_accuracy": 0.6905641108751297,
      "num_tokens": 26290383.0,
      "step": 273
    },
    {
      "epoch": 1.2622196664749856,
      "grad_norm": 0.013729719445109367,
      "learning_rate": 3.6476440184915817e-06,
      "loss": 1.1274,
      "mean_token_accuracy": 0.6840422749519348,
      "num_tokens": 26386071.0,
      "step": 274
    },
    {
      "epoch": 1.2668200115008625,
      "grad_norm": 0.014113099314272404,
      "learning_rate": 3.6089126804177373e-06,
      "loss": 1.1206,
      "mean_token_accuracy": 0.6884650811553001,
      "num_tokens": 26480843.0,
      "step": 275
    },
    {
      "epoch": 1.2714203565267395,
      "grad_norm": 0.014033479616045952,
      "learning_rate": 3.5702716080065546e-06,
      "loss": 1.1163,
      "mean_token_accuracy": 0.6893876194953918,
      "num_tokens": 26581540.0,
      "step": 276
    },
    {
      "epoch": 1.2760207015526164,
      "grad_norm": 0.01345459371805191,
      "learning_rate": 3.5317233086218474e-06,
      "loss": 1.1634,
      "mean_token_accuracy": 0.6806627959012985,
      "num_tokens": 26677681.0,
      "step": 277
    },
    {
      "epoch": 1.2806210465784935,
      "grad_norm": 0.013448994606733322,
      "learning_rate": 3.4932702836075216e-06,
      "loss": 1.1117,
      "mean_token_accuracy": 0.6878006383776665,
      "num_tokens": 26776798.0,
      "step": 278
    },
    {
      "epoch": 1.2852213916043702,
      "grad_norm": 0.013521680608391762,
      "learning_rate": 3.4549150281252635e-06,
      "loss": 1.1758,
      "mean_token_accuracy": 0.6749053597450256,
      "num_tokens": 26871483.0,
      "step": 279
    },
    {
      "epoch": 1.2898217366302474,
      "grad_norm": 0.012783129699528217,
      "learning_rate": 3.416660030992639e-06,
      "loss": 1.1154,
      "mean_token_accuracy": 0.6897058337926865,
      "num_tokens": 26971553.0,
      "step": 280
    },
    {
      "epoch": 1.2944220816561243,
      "grad_norm": 0.013442179188132286,
      "learning_rate": 3.378507774521587e-06,
      "loss": 1.0993,
      "mean_token_accuracy": 0.6932654827833176,
      "num_tokens": 27072607.0,
      "step": 281
    },
    {
      "epoch": 1.2990224266820012,
      "grad_norm": 0.015558043494820595,
      "learning_rate": 3.340460734357359e-06,
      "loss": 1.136,
      "mean_token_accuracy": 0.6847696825861931,
      "num_tokens": 27166911.0,
      "step": 282
    },
    {
      "epoch": 1.303622771707878,
      "grad_norm": 0.01596686989068985,
      "learning_rate": 3.3025213793178647e-06,
      "loss": 1.1228,
      "mean_token_accuracy": 0.6864028871059418,
      "num_tokens": 27259171.0,
      "step": 283
    },
    {
      "epoch": 1.308223116733755,
      "grad_norm": 0.01548865344375372,
      "learning_rate": 3.2646921712334854e-06,
      "loss": 1.1259,
      "mean_token_accuracy": 0.6838768199086189,
      "num_tokens": 27355256.0,
      "step": 284
    },
    {
      "epoch": 1.312823461759632,
      "grad_norm": 0.014444452710449696,
      "learning_rate": 3.226975564787322e-06,
      "loss": 1.1387,
      "mean_token_accuracy": 0.6838934645056725,
      "num_tokens": 27456222.0,
      "step": 285
    },
    {
      "epoch": 1.3174238067855089,
      "grad_norm": 0.014634891413152218,
      "learning_rate": 3.189374007355917e-06,
      "loss": 1.1166,
      "mean_token_accuracy": 0.6879652291536331,
      "num_tokens": 27550974.0,
      "step": 286
    },
    {
      "epoch": 1.3220241518113858,
      "grad_norm": 0.01424104068428278,
      "learning_rate": 3.151889938850445e-06,
      "loss": 1.1484,
      "mean_token_accuracy": 0.6820650920271873,
      "num_tokens": 27647575.0,
      "step": 287
    },
    {
      "epoch": 1.3266244968372627,
      "grad_norm": 0.013483865186572075,
      "learning_rate": 3.114525791558398e-06,
      "loss": 1.1251,
      "mean_token_accuracy": 0.6878606304526329,
      "num_tokens": 27747320.0,
      "step": 288
    },
    {
      "epoch": 1.3312248418631398,
      "grad_norm": 0.014904686249792576,
      "learning_rate": 3.0772839899857465e-06,
      "loss": 1.1132,
      "mean_token_accuracy": 0.6881635338068008,
      "num_tokens": 27842613.0,
      "step": 289
    },
    {
      "epoch": 1.3358251868890167,
      "grad_norm": 0.013660099357366562,
      "learning_rate": 3.040166950699626e-06,
      "loss": 1.1404,
      "mean_token_accuracy": 0.6817925050854683,
      "num_tokens": 27940439.0,
      "step": 290
    },
    {
      "epoch": 1.3404255319148937,
      "grad_norm": 0.01384556945413351,
      "learning_rate": 3.0031770821715233e-06,
      "loss": 1.1322,
      "mean_token_accuracy": 0.681700274348259,
      "num_tokens": 28033868.0,
      "step": 291
    },
    {
      "epoch": 1.3450258769407706,
      "grad_norm": 0.015682250261306763,
      "learning_rate": 2.966316784621e-06,
      "loss": 1.1612,
      "mean_token_accuracy": 0.6785846799612045,
      "num_tokens": 28131730.0,
      "step": 292
    },
    {
      "epoch": 1.3496262219666475,
      "grad_norm": 0.01346894446760416,
      "learning_rate": 2.9295884498599415e-06,
      "loss": 1.0988,
      "mean_token_accuracy": 0.6885087415575981,
      "num_tokens": 28225656.0,
      "step": 293
    },
    {
      "epoch": 1.3542265669925244,
      "grad_norm": 0.013444703072309494,
      "learning_rate": 2.8929944611373555e-06,
      "loss": 1.1266,
      "mean_token_accuracy": 0.688482291996479,
      "num_tokens": 28317210.0,
      "step": 294
    },
    {
      "epoch": 1.3588269120184013,
      "grad_norm": 0.015197189524769783,
      "learning_rate": 2.8565371929847286e-06,
      "loss": 1.1024,
      "mean_token_accuracy": 0.6932826340198517,
      "num_tokens": 28411161.0,
      "step": 295
    },
    {
      "epoch": 1.3634272570442783,
      "grad_norm": 0.016413524746894836,
      "learning_rate": 2.820219011061949e-06,
      "loss": 1.1173,
      "mean_token_accuracy": 0.6866224035620689,
      "num_tokens": 28505045.0,
      "step": 296
    },
    {
      "epoch": 1.3680276020701552,
      "grad_norm": 0.013771053403615952,
      "learning_rate": 2.7840422720037943e-06,
      "loss": 1.1429,
      "mean_token_accuracy": 0.6815031543374062,
      "num_tokens": 28601874.0,
      "step": 297
    },
    {
      "epoch": 1.3726279470960323,
      "grad_norm": 0.015113587491214275,
      "learning_rate": 2.748009323267016e-06,
      "loss": 1.1238,
      "mean_token_accuracy": 0.6860849261283875,
      "num_tokens": 28696440.0,
      "step": 298
    },
    {
      "epoch": 1.3772282921219092,
      "grad_norm": 0.01364414393901825,
      "learning_rate": 2.712122502978024e-06,
      "loss": 1.1429,
      "mean_token_accuracy": 0.6800694912672043,
      "num_tokens": 28795256.0,
      "step": 299
    },
    {
      "epoch": 1.3818286371477861,
      "grad_norm": 0.013928713276982307,
      "learning_rate": 2.6763841397811576e-06,
      "loss": 1.1323,
      "mean_token_accuracy": 0.6844060495495796,
      "num_tokens": 28891471.0,
      "step": 300
    },
    {
      "epoch": 1.386428982173663,
      "grad_norm": 0.016008542850613594,
      "learning_rate": 2.64079655268759e-06,
      "loss": 1.1386,
      "mean_token_accuracy": 0.6828343719244003,
      "num_tokens": 28985187.0,
      "step": 301
    },
    {
      "epoch": 1.39102932719954,
      "grad_norm": 0.015067597851157188,
      "learning_rate": 2.605362050924848e-06,
      "loss": 1.138,
      "mean_token_accuracy": 0.6833794862031937,
      "num_tokens": 29077725.0,
      "step": 302
    },
    {
      "epoch": 1.395629672225417,
      "grad_norm": 0.016080757603049278,
      "learning_rate": 2.57008293378697e-06,
      "loss": 1.1304,
      "mean_token_accuracy": 0.6871781721711159,
      "num_tokens": 29173318.0,
      "step": 303
    },
    {
      "epoch": 1.4002300172512938,
      "grad_norm": 0.014263632707297802,
      "learning_rate": 2.534961490485313e-06,
      "loss": 1.1578,
      "mean_token_accuracy": 0.6787764355540276,
      "num_tokens": 29268879.0,
      "step": 304
    },
    {
      "epoch": 1.4048303622771707,
      "grad_norm": 0.01258003804832697,
      "learning_rate": 2.5000000000000015e-06,
      "loss": 1.1418,
      "mean_token_accuracy": 0.6824080869555473,
      "num_tokens": 29365800.0,
      "step": 305
    },
    {
      "epoch": 1.4094307073030476,
      "grad_norm": 0.015632053837180138,
      "learning_rate": 2.4652007309320497e-06,
      "loss": 1.1815,
      "mean_token_accuracy": 0.673783928155899,
      "num_tokens": 29460673.0,
      "step": 306
    },
    {
      "epoch": 1.4140310523289248,
      "grad_norm": 0.014565376564860344,
      "learning_rate": 2.430565941356157e-06,
      "loss": 1.1325,
      "mean_token_accuracy": 0.6849493682384491,
      "num_tokens": 29561748.0,
      "step": 307
    },
    {
      "epoch": 1.4186313973548015,
      "grad_norm": 0.013184985145926476,
      "learning_rate": 2.3960978786741878e-06,
      "loss": 1.1266,
      "mean_token_accuracy": 0.6847327277064323,
      "num_tokens": 29659815.0,
      "step": 308
    },
    {
      "epoch": 1.4232317423806786,
      "grad_norm": 0.01590883731842041,
      "learning_rate": 2.3617987794693358e-06,
      "loss": 1.1284,
      "mean_token_accuracy": 0.6860458254814148,
      "num_tokens": 29757215.0,
      "step": 309
    },
    {
      "epoch": 1.4278320874065555,
      "grad_norm": 0.014524009078741074,
      "learning_rate": 2.3276708693609947e-06,
      "loss": 1.1282,
      "mean_token_accuracy": 0.6843872591853142,
      "num_tokens": 29852193.0,
      "step": 310
    },
    {
      "epoch": 1.4324324324324325,
      "grad_norm": 0.014431421644985676,
      "learning_rate": 2.2937163628603437e-06,
      "loss": 1.1361,
      "mean_token_accuracy": 0.6860853359103203,
      "num_tokens": 29946647.0,
      "step": 311
    },
    {
      "epoch": 1.4370327774583094,
      "grad_norm": 0.01360079925507307,
      "learning_rate": 2.2599374632266514e-06,
      "loss": 1.1366,
      "mean_token_accuracy": 0.6804769262671471,
      "num_tokens": 30041785.0,
      "step": 312
    },
    {
      "epoch": 1.4416331224841863,
      "grad_norm": 0.014270540326833725,
      "learning_rate": 2.2263363623243058e-06,
      "loss": 1.1229,
      "mean_token_accuracy": 0.6883322149515152,
      "num_tokens": 30141703.0,
      "step": 313
    },
    {
      "epoch": 1.4462334675100632,
      "grad_norm": 0.01577930524945259,
      "learning_rate": 2.192915240480596e-06,
      "loss": 1.1315,
      "mean_token_accuracy": 0.6846339255571365,
      "num_tokens": 30236086.0,
      "step": 314
    },
    {
      "epoch": 1.4508338125359401,
      "grad_norm": 0.01702301762998104,
      "learning_rate": 2.159676266344222e-06,
      "loss": 1.1335,
      "mean_token_accuracy": 0.684619590640068,
      "num_tokens": 30338166.0,
      "step": 315
    },
    {
      "epoch": 1.4554341575618173,
      "grad_norm": 0.015012836083769798,
      "learning_rate": 2.1266215967445823e-06,
      "loss": 1.1356,
      "mean_token_accuracy": 0.6858243867754936,
      "num_tokens": 30435257.0,
      "step": 316
    },
    {
      "epoch": 1.460034502587694,
      "grad_norm": 0.014962634071707726,
      "learning_rate": 2.0937533765518187e-06,
      "loss": 1.1463,
      "mean_token_accuracy": 0.6821167767047882,
      "num_tokens": 30532225.0,
      "step": 317
    },
    {
      "epoch": 1.464634847613571,
      "grad_norm": 0.012999648228287697,
      "learning_rate": 2.061073738537635e-06,
      "loss": 1.0992,
      "mean_token_accuracy": 0.6904957070946693,
      "num_tokens": 30632883.0,
      "step": 318
    },
    {
      "epoch": 1.469235192639448,
      "grad_norm": 0.013315939344465733,
      "learning_rate": 2.028584803236914e-06,
      "loss": 1.1199,
      "mean_token_accuracy": 0.689661055803299,
      "num_tokens": 30728533.0,
      "step": 319
    },
    {
      "epoch": 1.473835537665325,
      "grad_norm": 0.015063401311635971,
      "learning_rate": 1.996288678810105e-06,
      "loss": 1.1339,
      "mean_token_accuracy": 0.6830965057015419,
      "num_tokens": 30823487.0,
      "step": 320
    },
    {
      "epoch": 1.4784358826912019,
      "grad_norm": 0.016178306192159653,
      "learning_rate": 1.9641874609064443e-06,
      "loss": 1.1068,
      "mean_token_accuracy": 0.6887371689081192,
      "num_tokens": 30916751.0,
      "step": 321
    },
    {
      "epoch": 1.4830362277170788,
      "grad_norm": 0.015641840174794197,
      "learning_rate": 1.9322832325279563e-06,
      "loss": 1.1464,
      "mean_token_accuracy": 0.6822967156767845,
      "num_tokens": 31010220.0,
      "step": 322
    },
    {
      "epoch": 1.4876365727429557,
      "grad_norm": 0.01568695902824402,
      "learning_rate": 1.9005780638942982e-06,
      "loss": 1.1169,
      "mean_token_accuracy": 0.6888584122061729,
      "num_tokens": 31102774.0,
      "step": 323
    },
    {
      "epoch": 1.4922369177688326,
      "grad_norm": 0.015229462645947933,
      "learning_rate": 1.8690740123084316e-06,
      "loss": 1.1639,
      "mean_token_accuracy": 0.6777781769633293,
      "num_tokens": 31197179.0,
      "step": 324
    },
    {
      "epoch": 1.4968372627947095,
      "grad_norm": 0.014780288562178612,
      "learning_rate": 1.8377731220231144e-06,
      "loss": 1.1229,
      "mean_token_accuracy": 0.6873270496726036,
      "num_tokens": 31289592.0,
      "step": 325
    },
    {
      "epoch": 1.5014376078205864,
      "grad_norm": 0.015464117750525475,
      "learning_rate": 1.8066774241082612e-06,
      "loss": 1.0989,
      "mean_token_accuracy": 0.6906232610344887,
      "num_tokens": 31388081.0,
      "step": 326
    },
    {
      "epoch": 1.5060379528464636,
      "grad_norm": 0.014275867491960526,
      "learning_rate": 1.7757889363191484e-06,
      "loss": 1.1497,
      "mean_token_accuracy": 0.6790970116853714,
      "num_tokens": 31482452.0,
      "step": 327
    },
    {
      "epoch": 1.5106382978723403,
      "grad_norm": 0.014304906129837036,
      "learning_rate": 1.7451096629654813e-06,
      "loss": 1.111,
      "mean_token_accuracy": 0.6866638660430908,
      "num_tokens": 31581779.0,
      "step": 328
    },
    {
      "epoch": 1.5152386428982174,
      "grad_norm": 0.014677311293780804,
      "learning_rate": 1.7146415947813472e-06,
      "loss": 1.1048,
      "mean_token_accuracy": 0.6891909688711166,
      "num_tokens": 31678876.0,
      "step": 329
    },
    {
      "epoch": 1.5198389879240943,
      "grad_norm": 0.013847602531313896,
      "learning_rate": 1.6843867087960252e-06,
      "loss": 1.1727,
      "mean_token_accuracy": 0.6768988743424416,
      "num_tokens": 31777929.0,
      "step": 330
    },
    {
      "epoch": 1.5244393329499712,
      "grad_norm": 0.015412612818181515,
      "learning_rate": 1.6543469682057105e-06,
      "loss": 1.1145,
      "mean_token_accuracy": 0.6917125135660172,
      "num_tokens": 31876162.0,
      "step": 331
    },
    {
      "epoch": 1.5290396779758482,
      "grad_norm": 0.012880155816674232,
      "learning_rate": 1.6245243222461198e-06,
      "loss": 1.1185,
      "mean_token_accuracy": 0.6893917471170425,
      "num_tokens": 31977070.0,
      "step": 332
    },
    {
      "epoch": 1.533640023001725,
      "grad_norm": 0.014607089571654797,
      "learning_rate": 1.5949207060660138e-06,
      "loss": 1.1254,
      "mean_token_accuracy": 0.6842816695570946,
      "num_tokens": 32070721.0,
      "step": 333
    },
    {
      "epoch": 1.5382403680276022,
      "grad_norm": 0.014117407612502575,
      "learning_rate": 1.5655380406016236e-06,
      "loss": 1.1336,
      "mean_token_accuracy": 0.6836338341236115,
      "num_tokens": 32163042.0,
      "step": 334
    },
    {
      "epoch": 1.542840713053479,
      "grad_norm": 0.013967139646410942,
      "learning_rate": 1.5363782324520033e-06,
      "loss": 1.1209,
      "mean_token_accuracy": 0.6864241659641266,
      "num_tokens": 32258782.0,
      "step": 335
    },
    {
      "epoch": 1.547441058079356,
      "grad_norm": 0.01426258496940136,
      "learning_rate": 1.5074431737553158e-06,
      "loss": 1.1433,
      "mean_token_accuracy": 0.6818135678768158,
      "num_tokens": 32351941.0,
      "step": 336
    },
    {
      "epoch": 1.5520414031052328,
      "grad_norm": 0.014212261885404587,
      "learning_rate": 1.4787347420660541e-06,
      "loss": 1.1589,
      "mean_token_accuracy": 0.6766619980335236,
      "num_tokens": 32447081.0,
      "step": 337
    },
    {
      "epoch": 1.55664174813111,
      "grad_norm": 0.017143219709396362,
      "learning_rate": 1.450254800233209e-06,
      "loss": 1.1048,
      "mean_token_accuracy": 0.6900773867964745,
      "num_tokens": 32540969.0,
      "step": 338
    },
    {
      "epoch": 1.5612420931569868,
      "grad_norm": 0.014866476878523827,
      "learning_rate": 1.4220051962793952e-06,
      "loss": 1.1518,
      "mean_token_accuracy": 0.682394802570343,
      "num_tokens": 32637119.0,
      "step": 339
    },
    {
      "epoch": 1.5658424381828637,
      "grad_norm": 0.013940082862973213,
      "learning_rate": 1.3939877632809279e-06,
      "loss": 1.1103,
      "mean_token_accuracy": 0.6884757280349731,
      "num_tokens": 32733583.0,
      "step": 340
    },
    {
      "epoch": 1.5704427832087406,
      "grad_norm": 0.01513680163770914,
      "learning_rate": 1.366204319248885e-06,
      "loss": 1.1401,
      "mean_token_accuracy": 0.6841761991381645,
      "num_tokens": 32831547.0,
      "step": 341
    },
    {
      "epoch": 1.5750431282346176,
      "grad_norm": 0.01321039255708456,
      "learning_rate": 1.3386566670111339e-06,
      "loss": 1.1051,
      "mean_token_accuracy": 0.6915635168552399,
      "num_tokens": 32932973.0,
      "step": 342
    },
    {
      "epoch": 1.5796434732604947,
      "grad_norm": 0.013876788318157196,
      "learning_rate": 1.3113465940953495e-06,
      "loss": 1.1363,
      "mean_token_accuracy": 0.6843502447009087,
      "num_tokens": 33030925.0,
      "step": 343
    },
    {
      "epoch": 1.5842438182863714,
      "grad_norm": 0.014943170361220837,
      "learning_rate": 1.2842758726130283e-06,
      "loss": 1.1179,
      "mean_token_accuracy": 0.6865890249609947,
      "num_tokens": 33125536.0,
      "step": 344
    },
    {
      "epoch": 1.5888441633122485,
      "grad_norm": 0.013300514779984951,
      "learning_rate": 1.257446259144494e-06,
      "loss": 1.1575,
      "mean_token_accuracy": 0.6767172366380692,
      "num_tokens": 33225737.0,
      "step": 345
    },
    {
      "epoch": 1.5934445083381252,
      "grad_norm": 0.01461979653686285,
      "learning_rate": 1.2308594946249163e-06,
      "loss": 1.1086,
      "mean_token_accuracy": 0.6917338967323303,
      "num_tokens": 33317585.0,
      "step": 346
    },
    {
      "epoch": 1.5980448533640024,
      "grad_norm": 0.015281404368579388,
      "learning_rate": 1.2045173042313429e-06,
      "loss": 1.1211,
      "mean_token_accuracy": 0.6883839294314384,
      "num_tokens": 33415173.0,
      "step": 347
    },
    {
      "epoch": 1.6026451983898793,
      "grad_norm": 0.01422287430614233,
      "learning_rate": 1.1784213972707581e-06,
      "loss": 1.125,
      "mean_token_accuracy": 0.6859614327549934,
      "num_tokens": 33509461.0,
      "step": 348
    },
    {
      "epoch": 1.6072455434157562,
      "grad_norm": 0.015178248286247253,
      "learning_rate": 1.1525734670691702e-06,
      "loss": 1.1305,
      "mean_token_accuracy": 0.6844439208507538,
      "num_tokens": 33603830.0,
      "step": 349
    },
    {
      "epoch": 1.6118458884416331,
      "grad_norm": 0.015206449665129185,
      "learning_rate": 1.1269751908617277e-06,
      "loss": 1.1145,
      "mean_token_accuracy": 0.6888925060629845,
      "num_tokens": 33696633.0,
      "step": 350
    },
    {
      "epoch": 1.61644623346751,
      "grad_norm": 0.014618100598454475,
      "learning_rate": 1.1016282296838887e-06,
      "loss": 1.0988,
      "mean_token_accuracy": 0.6908671483397484,
      "num_tokens": 33791601.0,
      "step": 351
    },
    {
      "epoch": 1.621046578493387,
      "grad_norm": 0.01533578522503376,
      "learning_rate": 1.0765342282636416e-06,
      "loss": 1.1244,
      "mean_token_accuracy": 0.6876219883561134,
      "num_tokens": 33888255.0,
      "step": 352
    },
    {
      "epoch": 1.6256469235192639,
      "grad_norm": 0.014407423324882984,
      "learning_rate": 1.0516948149147755e-06,
      "loss": 1.1467,
      "mean_token_accuracy": 0.6829477101564407,
      "num_tokens": 33982641.0,
      "step": 353
    },
    {
      "epoch": 1.630247268545141,
      "grad_norm": 0.015139218419790268,
      "learning_rate": 1.0271116014312293e-06,
      "loss": 1.1447,
      "mean_token_accuracy": 0.6811694651842117,
      "num_tokens": 34080372.0,
      "step": 354
    },
    {
      "epoch": 1.6348476135710177,
      "grad_norm": 0.014464878477156162,
      "learning_rate": 1.0027861829824953e-06,
      "loss": 1.1289,
      "mean_token_accuracy": 0.6872419267892838,
      "num_tokens": 34177100.0,
      "step": 355
    },
    {
      "epoch": 1.6394479585968948,
      "grad_norm": 0.013368290849030018,
      "learning_rate": 9.787201380101157e-07,
      "loss": 1.1615,
      "mean_token_accuracy": 0.6785909533500671,
      "num_tokens": 34277683.0,
      "step": 356
    },
    {
      "epoch": 1.6440483036227715,
      "grad_norm": 0.014261895790696144,
      "learning_rate": 9.549150281252633e-07,
      "loss": 1.13,
      "mean_token_accuracy": 0.6863186731934547,
      "num_tokens": 34374641.0,
      "step": 357
    },
    {
      "epoch": 1.6486486486486487,
      "grad_norm": 0.013263655826449394,
      "learning_rate": 9.313723980074018e-07,
      "loss": 1.1045,
      "mean_token_accuracy": 0.6919630914926529,
      "num_tokens": 34473009.0,
      "step": 358
    },
    {
      "epoch": 1.6532489936745256,
      "grad_norm": 0.01498090848326683,
      "learning_rate": 9.080937753040647e-07,
      "loss": 1.1229,
      "mean_token_accuracy": 0.6858481094241142,
      "num_tokens": 34569063.0,
      "step": 359
    },
    {
      "epoch": 1.6578493387004025,
      "grad_norm": 0.014527600258588791,
      "learning_rate": 8.850806705317183e-07,
      "loss": 1.145,
      "mean_token_accuracy": 0.6824858039617538,
      "num_tokens": 34663301.0,
      "step": 360
    },
    {
      "epoch": 1.6624496837262794,
      "grad_norm": 0.014273539185523987,
      "learning_rate": 8.623345769777514e-07,
      "loss": 1.0855,
      "mean_token_accuracy": 0.6919998005032539,
      "num_tokens": 34755484.0,
      "step": 361
    },
    {
      "epoch": 1.6670500287521564,
      "grad_norm": 0.014079663902521133,
      "learning_rate": 8.398569706035791e-07,
      "loss": 1.1314,
      "mean_token_accuracy": 0.685786284506321,
      "num_tokens": 34850202.0,
      "step": 362
    },
    {
      "epoch": 1.6716503737780335,
      "grad_norm": 0.015356908552348614,
      "learning_rate": 8.176493099488664e-07,
      "loss": 1.1421,
      "mean_token_accuracy": 0.6809604987502098,
      "num_tokens": 34946332.0,
      "step": 363
    },
    {
      "epoch": 1.6762507188039102,
      "grad_norm": 0.015410685911774635,
      "learning_rate": 7.957130360368898e-07,
      "loss": 1.1175,
      "mean_token_accuracy": 0.6878371834754944,
      "num_tokens": 35041587.0,
      "step": 364
    },
    {
      "epoch": 1.6808510638297873,
      "grad_norm": 0.016270428895950317,
      "learning_rate": 7.740495722810271e-07,
      "loss": 1.1735,
      "mean_token_accuracy": 0.6750294417142868,
      "num_tokens": 35139785.0,
      "step": 365
    },
    {
      "epoch": 1.685451408855664,
      "grad_norm": 0.015215057879686356,
      "learning_rate": 7.526603243923958e-07,
      "loss": 1.1524,
      "mean_token_accuracy": 0.6767367720603943,
      "num_tokens": 35235093.0,
      "step": 366
    },
    {
      "epoch": 1.6900517538815412,
      "grad_norm": 0.014159487560391426,
      "learning_rate": 7.315466802886401e-07,
      "loss": 1.1088,
      "mean_token_accuracy": 0.6886250227689743,
      "num_tokens": 35332416.0,
      "step": 367
    },
    {
      "epoch": 1.694652098907418,
      "grad_norm": 0.014062878675758839,
      "learning_rate": 7.107100100038672e-07,
      "loss": 1.1578,
      "mean_token_accuracy": 0.6797412261366844,
      "num_tokens": 35431910.0,
      "step": 368
    },
    {
      "epoch": 1.699252443933295,
      "grad_norm": 0.014110508374869823,
      "learning_rate": 6.901516655997536e-07,
      "loss": 1.1387,
      "mean_token_accuracy": 0.6830648630857468,
      "num_tokens": 35529803.0,
      "step": 369
    },
    {
      "epoch": 1.703852788959172,
      "grad_norm": 0.015163487754762173,
      "learning_rate": 6.698729810778065e-07,
      "loss": 1.1303,
      "mean_token_accuracy": 0.6845860555768013,
      "num_tokens": 35627791.0,
      "step": 370
    },
    {
      "epoch": 1.7084531339850488,
      "grad_norm": 0.012958538718521595,
      "learning_rate": 6.498752722928042e-07,
      "loss": 1.1231,
      "mean_token_accuracy": 0.6880553737282753,
      "num_tokens": 35722436.0,
      "step": 371
    },
    {
      "epoch": 1.713053479010926,
      "grad_norm": 0.014430136419832706,
      "learning_rate": 6.301598368674106e-07,
      "loss": 1.1245,
      "mean_token_accuracy": 0.6855921223759651,
      "num_tokens": 35819723.0,
      "step": 372
    },
    {
      "epoch": 1.7176538240368027,
      "grad_norm": 0.014730778522789478,
      "learning_rate": 6.107279541079769e-07,
      "loss": 1.1046,
      "mean_token_accuracy": 0.6905967816710472,
      "num_tokens": 35915276.0,
      "step": 373
    },
    {
      "epoch": 1.7222541690626798,
      "grad_norm": 0.014451798982918262,
      "learning_rate": 5.915808849215304e-07,
      "loss": 1.1047,
      "mean_token_accuracy": 0.6911017447710037,
      "num_tokens": 36008996.0,
      "step": 374
    },
    {
      "epoch": 1.7268545140885565,
      "grad_norm": 0.01583203487098217,
      "learning_rate": 5.727198717339511e-07,
      "loss": 1.107,
      "mean_token_accuracy": 0.6893082559108734,
      "num_tokens": 36103140.0,
      "step": 375
    },
    {
      "epoch": 1.7314548591144336,
      "grad_norm": 0.013766398653388023,
      "learning_rate": 5.541461384093549e-07,
      "loss": 1.1337,
      "mean_token_accuracy": 0.6845080852508545,
      "num_tokens": 36201738.0,
      "step": 376
    },
    {
      "epoch": 1.7360552041403106,
      "grad_norm": 0.0149979954585433,
      "learning_rate": 5.358608901706802e-07,
      "loss": 1.1454,
      "mean_token_accuracy": 0.6823095753788948,
      "num_tokens": 36300425.0,
      "step": 377
    },
    {
      "epoch": 1.7406555491661875,
      "grad_norm": 0.014548883773386478,
      "learning_rate": 5.178653135214811e-07,
      "loss": 1.1457,
      "mean_token_accuracy": 0.6806545779109001,
      "num_tokens": 36391644.0,
      "step": 378
    },
    {
      "epoch": 1.7452558941920644,
      "grad_norm": 0.013143756426870823,
      "learning_rate": 5.001605761689399e-07,
      "loss": 1.1443,
      "mean_token_accuracy": 0.6823921278119087,
      "num_tokens": 36492760.0,
      "step": 379
    },
    {
      "epoch": 1.7498562392179413,
      "grad_norm": 0.014462238177657127,
      "learning_rate": 4.827478269480895e-07,
      "loss": 1.1372,
      "mean_token_accuracy": 0.681210495531559,
      "num_tokens": 36588188.0,
      "step": 380
    },
    {
      "epoch": 1.7544565842438185,
      "grad_norm": 0.014685888774693012,
      "learning_rate": 4.6562819574727304e-07,
      "loss": 1.138,
      "mean_token_accuracy": 0.6838786154985428,
      "num_tokens": 36688704.0,
      "step": 381
    },
    {
      "epoch": 1.7590569292696951,
      "grad_norm": 0.01780293695628643,
      "learning_rate": 4.4880279343482713e-07,
      "loss": 1.1058,
      "mean_token_accuracy": 0.6894217655062675,
      "num_tokens": 36783354.0,
      "step": 382
    },
    {
      "epoch": 1.7636572742955723,
      "grad_norm": 0.014380701817572117,
      "learning_rate": 4.322727117869951e-07,
      "loss": 1.1325,
      "mean_token_accuracy": 0.6812404841184616,
      "num_tokens": 36883898.0,
      "step": 383
    },
    {
      "epoch": 1.768257619321449,
      "grad_norm": 0.01344380248337984,
      "learning_rate": 4.1603902341708804e-07,
      "loss": 1.1054,
      "mean_token_accuracy": 0.6925413459539413,
      "num_tokens": 36981882.0,
      "step": 384
    },
    {
      "epoch": 1.7728579643473261,
      "grad_norm": 0.0131448432803154,
      "learning_rate": 4.001027817058789e-07,
      "loss": 1.1081,
      "mean_token_accuracy": 0.6875624433159828,
      "num_tokens": 37079406.0,
      "step": 385
    },
    {
      "epoch": 1.777458309373203,
      "grad_norm": 0.014320706948637962,
      "learning_rate": 3.844650207332562e-07,
      "loss": 1.1185,
      "mean_token_accuracy": 0.6889297738671303,
      "num_tokens": 37174074.0,
      "step": 386
    },
    {
      "epoch": 1.78205865439908,
      "grad_norm": 0.015010056085884571,
      "learning_rate": 3.691267552111183e-07,
      "loss": 1.1738,
      "mean_token_accuracy": 0.6784838512539864,
      "num_tokens": 37274254.0,
      "step": 387
    },
    {
      "epoch": 1.7866589994249569,
      "grad_norm": 0.013940027914941311,
      "learning_rate": 3.54088980417534e-07,
      "loss": 1.1029,
      "mean_token_accuracy": 0.6877925917506218,
      "num_tokens": 37373486.0,
      "step": 388
    },
    {
      "epoch": 1.7912593444508338,
      "grad_norm": 0.016674095764756203,
      "learning_rate": 3.3935267213216163e-07,
      "loss": 1.1333,
      "mean_token_accuracy": 0.68411485850811,
      "num_tokens": 37470185.0,
      "step": 389
    },
    {
      "epoch": 1.7958596894767107,
      "grad_norm": 0.015059294179081917,
      "learning_rate": 3.2491878657292643e-07,
      "loss": 1.1164,
      "mean_token_accuracy": 0.6874109357595444,
      "num_tokens": 37563960.0,
      "step": 390
    },
    {
      "epoch": 1.8004600345025876,
      "grad_norm": 0.015169203281402588,
      "learning_rate": 3.1078826033397845e-07,
      "loss": 1.1202,
      "mean_token_accuracy": 0.6886393278837204,
      "num_tokens": 37662917.0,
      "step": 391
    },
    {
      "epoch": 1.8050603795284648,
      "grad_norm": 0.013790096156299114,
      "learning_rate": 2.9696201032491434e-07,
      "loss": 1.1368,
      "mean_token_accuracy": 0.6837680712342262,
      "num_tokens": 37753862.0,
      "step": 392
    },
    {
      "epoch": 1.8096607245543415,
      "grad_norm": 0.013124031946063042,
      "learning_rate": 2.834409337112842e-07,
      "loss": 1.1389,
      "mean_token_accuracy": 0.6866304948925972,
      "num_tokens": 37854261.0,
      "step": 393
    },
    {
      "epoch": 1.8142610695802186,
      "grad_norm": 0.015313307754695415,
      "learning_rate": 2.7022590785637406e-07,
      "loss": 1.1293,
      "mean_token_accuracy": 0.6882248967885971,
      "num_tokens": 37949798.0,
      "step": 394
    },
    {
      "epoch": 1.8188614146060953,
      "grad_norm": 0.016208309680223465,
      "learning_rate": 2.573177902642726e-07,
      "loss": 1.1332,
      "mean_token_accuracy": 0.6835260838270187,
      "num_tokens": 38046730.0,
      "step": 395
    },
    {
      "epoch": 1.8234617596319724,
      "grad_norm": 0.013482874259352684,
      "learning_rate": 2.447174185242324e-07,
      "loss": 1.088,
      "mean_token_accuracy": 0.6929525211453438,
      "num_tokens": 38142774.0,
      "step": 396
    },
    {
      "epoch": 1.8280621046578494,
      "grad_norm": 0.015274173580110073,
      "learning_rate": 2.3242561025631882e-07,
      "loss": 1.1542,
      "mean_token_accuracy": 0.6808317825198174,
      "num_tokens": 38239269.0,
      "step": 397
    },
    {
      "epoch": 1.8326624496837263,
      "grad_norm": 0.016665548086166382,
      "learning_rate": 2.204431630583548e-07,
      "loss": 1.1181,
      "mean_token_accuracy": 0.686672255396843,
      "num_tokens": 38335638.0,
      "step": 398
    },
    {
      "epoch": 1.8372627947096032,
      "grad_norm": 0.014923077076673508,
      "learning_rate": 2.0877085445416889e-07,
      "loss": 1.1016,
      "mean_token_accuracy": 0.690706379711628,
      "num_tokens": 38431268.0,
      "step": 399
    },
    {
      "epoch": 1.84186313973548,
      "grad_norm": 0.014740503393113613,
      "learning_rate": 1.9740944184313882e-07,
      "loss": 1.1428,
      "mean_token_accuracy": 0.6803831607103348,
      "num_tokens": 38524271.0,
      "step": 400
    },
    {
      "epoch": 1.8464634847613572,
      "grad_norm": 0.01503705233335495,
      "learning_rate": 1.8635966245104663e-07,
      "loss": 1.1378,
      "mean_token_accuracy": 0.6864303201436996,
      "num_tokens": 38617153.0,
      "step": 401
    },
    {
      "epoch": 1.851063829787234,
      "grad_norm": 0.014517681673169136,
      "learning_rate": 1.7562223328224327e-07,
      "loss": 1.1266,
      "mean_token_accuracy": 0.6875582709908485,
      "num_tokens": 38716205.0,
      "step": 402
    },
    {
      "epoch": 1.855664174813111,
      "grad_norm": 0.013916529715061188,
      "learning_rate": 1.651978510731189e-07,
      "loss": 1.1618,
      "mean_token_accuracy": 0.680492527782917,
      "num_tokens": 38816231.0,
      "step": 403
    },
    {
      "epoch": 1.8602645198389878,
      "grad_norm": 0.01476625818759203,
      "learning_rate": 1.5508719224689716e-07,
      "loss": 1.1313,
      "mean_token_accuracy": 0.6825379431247711,
      "num_tokens": 38915191.0,
      "step": 404
    },
    {
      "epoch": 1.864864864864865,
      "grad_norm": 0.01506135892122984,
      "learning_rate": 1.4529091286973994e-07,
      "loss": 1.1528,
      "mean_token_accuracy": 0.6801683902740479,
      "num_tokens": 39012971.0,
      "step": 405
    },
    {
      "epoch": 1.8694652098907418,
      "grad_norm": 0.01537392009049654,
      "learning_rate": 1.358096486081778e-07,
      "loss": 1.1495,
      "mean_token_accuracy": 0.6829325184226036,
      "num_tokens": 39109365.0,
      "step": 406
    },
    {
      "epoch": 1.8740655549166187,
      "grad_norm": 0.014476312324404716,
      "learning_rate": 1.2664401468786114e-07,
      "loss": 1.1271,
      "mean_token_accuracy": 0.6879825815558434,
      "num_tokens": 39202764.0,
      "step": 407
    },
    {
      "epoch": 1.8786658999424957,
      "grad_norm": 0.01422926690429449,
      "learning_rate": 1.1779460585363945e-07,
      "loss": 1.1463,
      "mean_token_accuracy": 0.6799934059381485,
      "num_tokens": 39297379.0,
      "step": 408
    },
    {
      "epoch": 1.8832662449683726,
      "grad_norm": 0.014644317328929901,
      "learning_rate": 1.0926199633097156e-07,
      "loss": 1.1222,
      "mean_token_accuracy": 0.6866084560751915,
      "num_tokens": 39389572.0,
      "step": 409
    },
    {
      "epoch": 1.8878665899942497,
      "grad_norm": 0.014122752472758293,
      "learning_rate": 1.0104673978866164e-07,
      "loss": 1.1297,
      "mean_token_accuracy": 0.687385693192482,
      "num_tokens": 39487297.0,
      "step": 410
    },
    {
      "epoch": 1.8924669350201264,
      "grad_norm": 0.014691289514303207,
      "learning_rate": 9.314936930293283e-08,
      "loss": 1.1637,
      "mean_token_accuracy": 0.6754296645522118,
      "num_tokens": 39581833.0,
      "step": 411
    },
    {
      "epoch": 1.8970672800460036,
      "grad_norm": 0.016847725957632065,
      "learning_rate": 8.557039732283945e-08,
      "loss": 1.09,
      "mean_token_accuracy": 0.6946672350168228,
      "num_tokens": 39678219.0,
      "step": 412
    },
    {
      "epoch": 1.9016676250718803,
      "grad_norm": 0.014185824431478977,
      "learning_rate": 7.831031563701131e-08,
      "loss": 1.1357,
      "mean_token_accuracy": 0.6832516267895699,
      "num_tokens": 39774070.0,
      "step": 413
    },
    {
      "epoch": 1.9062679700977574,
      "grad_norm": 0.01294031459838152,
      "learning_rate": 7.136959534174592e-08,
      "loss": 1.115,
      "mean_token_accuracy": 0.6877038702368736,
      "num_tokens": 39872679.0,
      "step": 414
    },
    {
      "epoch": 1.9108683151236343,
      "grad_norm": 0.01454690657556057,
      "learning_rate": 6.474868681043578e-08,
      "loss": 1.124,
      "mean_token_accuracy": 0.6823562011122704,
      "num_tokens": 39968778.0,
      "step": 415
    },
    {
      "epoch": 1.9154686601495112,
      "grad_norm": 0.014392836950719357,
      "learning_rate": 5.844801966434832e-08,
      "loss": 1.1087,
      "mean_token_accuracy": 0.6888928934931755,
      "num_tokens": 40063775.0,
      "step": 416
    },
    {
      "epoch": 1.9200690051753881,
      "grad_norm": 0.014324750751256943,
      "learning_rate": 5.246800274474439e-08,
      "loss": 1.0951,
      "mean_token_accuracy": 0.6892815381288528,
      "num_tokens": 40160271.0,
      "step": 417
    },
    {
      "epoch": 1.924669350201265,
      "grad_norm": 0.014842728152871132,
      "learning_rate": 4.680902408635335e-08,
      "loss": 1.1024,
      "mean_token_accuracy": 0.6918300613760948,
      "num_tokens": 40261319.0,
      "step": 418
    },
    {
      "epoch": 1.9292696952271422,
      "grad_norm": 0.014851558022201061,
      "learning_rate": 4.147145089218985e-08,
      "loss": 1.1159,
      "mean_token_accuracy": 0.6875586584210396,
      "num_tokens": 40358248.0,
      "step": 419
    },
    {
      "epoch": 1.933870040253019,
      "grad_norm": 0.01473570428788662,
      "learning_rate": 3.645562950973014e-08,
      "loss": 1.1382,
      "mean_token_accuracy": 0.6858562529087067,
      "num_tokens": 40457606.0,
      "step": 420
    },
    {
      "epoch": 1.938470385278896,
      "grad_norm": 0.013431058265268803,
      "learning_rate": 3.1761885408435055e-08,
      "loss": 1.0909,
      "mean_token_accuracy": 0.695126011967659,
      "num_tokens": 40556693.0,
      "step": 421
    },
    {
      "epoch": 1.9430707303047727,
      "grad_norm": 0.014296426437795162,
      "learning_rate": 2.7390523158633552e-08,
      "loss": 1.1435,
      "mean_token_accuracy": 0.684424415230751,
      "num_tokens": 40648892.0,
      "step": 422
    },
    {
      "epoch": 1.9476710753306499,
      "grad_norm": 0.015391189604997635,
      "learning_rate": 2.3341826411756863e-08,
      "loss": 1.1361,
      "mean_token_accuracy": 0.6864252239465714,
      "num_tokens": 40745057.0,
      "step": 423
    },
    {
      "epoch": 1.9522714203565268,
      "grad_norm": 0.014759673736989498,
      "learning_rate": 1.9616057881935436e-08,
      "loss": 1.1287,
      "mean_token_accuracy": 0.6864694356918335,
      "num_tokens": 40840174.0,
      "step": 424
    },
    {
      "epoch": 1.9568717653824037,
      "grad_norm": 0.014287742786109447,
      "learning_rate": 1.6213459328950355e-08,
      "loss": 1.1394,
      "mean_token_accuracy": 0.6858852058649063,
      "num_tokens": 40933861.0,
      "step": 425
    },
    {
      "epoch": 1.9614721104082806,
      "grad_norm": 0.015595638193190098,
      "learning_rate": 1.3134251542544774e-08,
      "loss": 1.1298,
      "mean_token_accuracy": 0.6829616725444794,
      "num_tokens": 41032425.0,
      "step": 426
    },
    {
      "epoch": 1.9660724554341575,
      "grad_norm": 0.014190633781254292,
      "learning_rate": 1.0378634328099268e-08,
      "loss": 1.1347,
      "mean_token_accuracy": 0.6856722384691238,
      "num_tokens": 41127692.0,
      "step": 427
    },
    {
      "epoch": 1.9706728004600345,
      "grad_norm": 0.016822172328829765,
      "learning_rate": 7.946786493666647e-09,
      "loss": 1.123,
      "mean_token_accuracy": 0.6841925904154778,
      "num_tokens": 41220510.0,
      "step": 428
    },
    {
      "epoch": 1.9752731454859114,
      "grad_norm": 0.01764860562980175,
      "learning_rate": 5.838865838366792e-09,
      "loss": 1.1343,
      "mean_token_accuracy": 0.6837752312421799,
      "num_tokens": 41312473.0,
      "step": 429
    },
    {
      "epoch": 1.9798734905117885,
      "grad_norm": 0.01497281901538372,
      "learning_rate": 4.055009142152066e-09,
      "loss": 1.1392,
      "mean_token_accuracy": 0.6823102980852127,
      "num_tokens": 41405730.0,
      "step": 430
    },
    {
      "epoch": 1.9844738355376652,
      "grad_norm": 0.015005367808043957,
      "learning_rate": 2.595332156925534e-09,
      "loss": 1.1139,
      "mean_token_accuracy": 0.6865201741456985,
      "num_tokens": 41503861.0,
      "step": 431
    },
    {
      "epoch": 1.9890741805635423,
      "grad_norm": 0.015820927917957306,
      "learning_rate": 1.4599295990352924e-09,
      "loss": 1.1206,
      "mean_token_accuracy": 0.688090018928051,
      "num_tokens": 41599729.0,
      "step": 432
    },
    {
      "epoch": 1.993674525589419,
      "grad_norm": 0.014639221131801605,
      "learning_rate": 6.488751431266149e-10,
      "loss": 1.1003,
      "mean_token_accuracy": 0.6921690627932549,
      "num_tokens": 41699246.0,
      "step": 433
    },
    {
      "epoch": 1.9982748706152962,
      "grad_norm": 0.014356461353600025,
      "learning_rate": 1.622214173602199e-10,
      "loss": 1.1363,
      "mean_token_accuracy": 0.6814094483852386,
      "num_tokens": 41800821.0,
      "step": 434
    }
  ],
  "logging_steps": 1,
  "max_steps": 434,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 3.765793490454184e+18,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}