{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 34.89855072463768,
  "eval_steps": 500,
  "global_step": 1505,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.11594202898550725,
      "grad_norm": 7.532149791717529,
      "learning_rate": 4.999863832700438e-05,
      "loss": 3.8116,
      "num_input_tokens_seen": 106929,
      "step": 5
    },
    {
      "epoch": 0.2318840579710145,
      "grad_norm": 4.141716480255127,
      "learning_rate": 4.999455345634978e-05,
      "loss": 3.6928,
      "num_input_tokens_seen": 225964,
      "step": 10
    },
    {
      "epoch": 0.34782608695652173,
      "grad_norm": 3.470097780227661,
      "learning_rate": 4.9987745833016855e-05,
      "loss": 3.6227,
      "num_input_tokens_seen": 362264,
      "step": 15
    },
    {
      "epoch": 0.463768115942029,
      "grad_norm": 3.4544646739959717,
      "learning_rate": 4.9978216198586135e-05,
      "loss": 3.601,
      "num_input_tokens_seen": 477807,
      "step": 20
    },
    {
      "epoch": 0.5797101449275363,
      "grad_norm": 3.249224901199341,
      "learning_rate": 4.996596559115731e-05,
      "loss": 3.539,
      "num_input_tokens_seen": 588900,
      "step": 25
    },
    {
      "epoch": 0.6956521739130435,
      "grad_norm": 3.395056962966919,
      "learning_rate": 4.995099534523607e-05,
      "loss": 3.4956,
      "num_input_tokens_seen": 706077,
      "step": 30
    },
    {
      "epoch": 0.8115942028985508,
      "grad_norm": 3.997875213623047,
      "learning_rate": 4.9933307091588796e-05,
      "loss": 3.5044,
      "num_input_tokens_seen": 853504,
      "step": 35
    },
    {
      "epoch": 0.927536231884058,
      "grad_norm": 3.5168681144714355,
      "learning_rate": 4.991290275706486e-05,
      "loss": 3.4324,
      "num_input_tokens_seen": 990472,
      "step": 40
    },
    {
      "epoch": 1.0434782608695652,
      "grad_norm": 7.144646167755127,
      "learning_rate": 4.988978456438678e-05,
      "loss": 3.2542,
      "num_input_tokens_seen": 1125870,
      "step": 45
    },
    {
      "epoch": 1.1594202898550725,
      "grad_norm": 3.257103681564331,
      "learning_rate": 4.986395503190805e-05,
      "loss": 2.9024,
      "num_input_tokens_seen": 1249877,
      "step": 50
    },
    {
      "epoch": 1.2753623188405796,
      "grad_norm": 3.3208603858947754,
      "learning_rate": 4.983541697333881e-05,
      "loss": 2.8069,
      "num_input_tokens_seen": 1375193,
      "step": 55
    },
    {
      "epoch": 1.391304347826087,
      "grad_norm": 4.378167629241943,
      "learning_rate": 4.980417349743936e-05,
      "loss": 2.75,
      "num_input_tokens_seen": 1489716,
      "step": 60
    },
    {
      "epoch": 1.5072463768115942,
      "grad_norm": 4.321849822998047,
      "learning_rate": 4.9770228007681494e-05,
      "loss": 2.7329,
      "num_input_tokens_seen": 1600483,
      "step": 65
    },
    {
      "epoch": 1.6231884057971016,
      "grad_norm": 3.6365067958831787,
      "learning_rate": 4.973358420187776e-05,
      "loss": 2.8212,
      "num_input_tokens_seen": 1731315,
      "step": 70
    },
    {
      "epoch": 1.7391304347826086,
      "grad_norm": 3.74035906791687,
      "learning_rate": 4.9694246071778604e-05,
      "loss": 2.7935,
      "num_input_tokens_seen": 1858269,
      "step": 75
    },
    {
      "epoch": 1.855072463768116,
      "grad_norm": 23.26426124572754,
      "learning_rate": 4.9652217902637596e-05,
      "loss": 2.7305,
      "num_input_tokens_seen": 1984587,
      "step": 80
    },
    {
      "epoch": 1.971014492753623,
      "grad_norm": 4.870578289031982,
      "learning_rate": 4.9607504272744575e-05,
      "loss": 2.6482,
      "num_input_tokens_seen": 2109391,
      "step": 85
    },
    {
      "epoch": 2.0869565217391304,
      "grad_norm": 4.419096946716309,
      "learning_rate": 4.956011005292692e-05,
      "loss": 2.4292,
      "num_input_tokens_seen": 2246413,
      "step": 90
    },
    {
      "epoch": 2.2028985507246377,
      "grad_norm": 18.216915130615234,
      "learning_rate": 4.951004040601898e-05,
      "loss": 2.1416,
      "num_input_tokens_seen": 2386890,
      "step": 95
    },
    {
      "epoch": 2.318840579710145,
      "grad_norm": 4.4814581871032715,
      "learning_rate": 4.945730078629964e-05,
      "loss": 2.2847,
      "num_input_tokens_seen": 2522302,
      "step": 100
    },
    {
      "epoch": 2.4347826086956523,
      "grad_norm": 406.9701232910156,
      "learning_rate": 4.9401896938898185e-05,
      "loss": 2.0944,
      "num_input_tokens_seen": 2642208,
      "step": 105
    },
    {
      "epoch": 2.550724637681159,
      "grad_norm": 3.5890581607818604,
      "learning_rate": 4.934383489916843e-05,
      "loss": 2.2862,
      "num_input_tokens_seen": 2780587,
      "step": 110
    },
    {
      "epoch": 2.6666666666666665,
      "grad_norm": 5.334541320800781,
      "learning_rate": 4.928312099203131e-05,
      "loss": 2.105,
      "num_input_tokens_seen": 2885320,
      "step": 115
    },
    {
      "epoch": 2.782608695652174,
      "grad_norm": 5.464664936065674,
      "learning_rate": 4.921976183128585e-05,
      "loss": 2.0287,
      "num_input_tokens_seen": 2996923,
      "step": 120
    },
    {
      "epoch": 2.898550724637681,
      "grad_norm": 4.113780975341797,
      "learning_rate": 4.9153764318888706e-05,
      "loss": 2.0162,
      "num_input_tokens_seen": 3102391,
      "step": 125
    },
    {
      "epoch": 3.0144927536231885,
      "grad_norm": 6.009971618652344,
      "learning_rate": 4.908513564420231e-05,
      "loss": 2.2464,
      "num_input_tokens_seen": 3233443,
      "step": 130
    },
    {
      "epoch": 3.130434782608696,
      "grad_norm": 10.397327423095703,
      "learning_rate": 4.90138832832117e-05,
      "loss": 1.6561,
      "num_input_tokens_seen": 3358966,
      "step": 135
    },
    {
      "epoch": 3.246376811594203,
      "grad_norm": 4.9139556884765625,
      "learning_rate": 4.894001499771015e-05,
      "loss": 1.6113,
      "num_input_tokens_seen": 3490069,
      "step": 140
    },
    {
      "epoch": 3.36231884057971,
      "grad_norm": 4.146034240722656,
      "learning_rate": 4.886353883445363e-05,
      "loss": 1.6235,
      "num_input_tokens_seen": 3609842,
      "step": 145
    },
    {
      "epoch": 3.4782608695652173,
      "grad_norm": 4.301880359649658,
      "learning_rate": 4.878446312428424e-05,
      "loss": 1.7873,
      "num_input_tokens_seen": 3751570,
      "step": 150
    },
    {
      "epoch": 3.5942028985507246,
      "grad_norm": 3.9485158920288086,
      "learning_rate": 4.8702796481222714e-05,
      "loss": 1.3723,
      "num_input_tokens_seen": 3865303,
      "step": 155
    },
    {
      "epoch": 3.710144927536232,
      "grad_norm": 4.183668613433838,
      "learning_rate": 4.861854780153004e-05,
      "loss": 1.6512,
      "num_input_tokens_seen": 3991347,
      "step": 160
    },
    {
      "epoch": 3.8260869565217392,
      "grad_norm": 5.1000471115112305,
      "learning_rate": 4.853172626273841e-05,
      "loss": 1.5524,
      "num_input_tokens_seen": 4113654,
      "step": 165
    },
    {
      "epoch": 3.942028985507246,
      "grad_norm": 4.142239570617676,
      "learning_rate": 4.8442341322651385e-05,
      "loss": 1.5954,
      "num_input_tokens_seen": 4236348,
      "step": 170
    },
    {
      "epoch": 4.057971014492754,
      "grad_norm": 3.8976669311523438,
      "learning_rate": 4.83504027183137e-05,
      "loss": 1.1652,
      "num_input_tokens_seen": 4340378,
      "step": 175
    },
    {
      "epoch": 4.173913043478261,
      "grad_norm": 5.923389911651611,
      "learning_rate": 4.825592046495054e-05,
      "loss": 1.1995,
      "num_input_tokens_seen": 4473601,
      "step": 180
    },
    {
      "epoch": 4.2898550724637685,
      "grad_norm": 4.220530033111572,
      "learning_rate": 4.8158904854876555e-05,
      "loss": 0.9431,
      "num_input_tokens_seen": 4586911,
      "step": 185
    },
    {
      "epoch": 4.405797101449275,
      "grad_norm": 5.896139144897461,
      "learning_rate": 4.805936645637463e-05,
      "loss": 1.1136,
      "num_input_tokens_seen": 4702445,
      "step": 190
    },
    {
      "epoch": 4.521739130434782,
      "grad_norm": 4.467094421386719,
      "learning_rate": 4.795731611254473e-05,
      "loss": 1.1509,
      "num_input_tokens_seen": 4831301,
      "step": 195
    },
    {
      "epoch": 4.63768115942029,
      "grad_norm": 4.232386112213135,
      "learning_rate": 4.785276494012263e-05,
      "loss": 0.9962,
      "num_input_tokens_seen": 4941656,
      "step": 200
    },
    {
      "epoch": 4.753623188405797,
      "grad_norm": 4.829892635345459,
      "learning_rate": 4.7745724328269e-05,
      "loss": 1.2377,
      "num_input_tokens_seen": 5088437,
      "step": 205
    },
    {
      "epoch": 4.869565217391305,
      "grad_norm": 4.1343913078308105,
      "learning_rate": 4.763620593732867e-05,
      "loss": 1.234,
      "num_input_tokens_seen": 5219806,
      "step": 210
    },
    {
      "epoch": 4.9855072463768115,
      "grad_norm": 4.9217729568481445,
      "learning_rate": 4.752422169756048e-05,
      "loss": 1.1453,
      "num_input_tokens_seen": 5340222,
      "step": 215
    },
    {
      "epoch": 5.101449275362318,
      "grad_norm": 4.4605865478515625,
      "learning_rate": 4.740978380783765e-05,
      "loss": 0.9056,
      "num_input_tokens_seen": 5476315,
      "step": 220
    },
    {
      "epoch": 5.217391304347826,
      "grad_norm": 4.396484375,
      "learning_rate": 4.7292904734318924e-05,
      "loss": 0.7349,
      "num_input_tokens_seen": 5589951,
      "step": 225
    },
    {
      "epoch": 5.333333333333333,
      "grad_norm": 4.053436279296875,
      "learning_rate": 4.7173597209090534e-05,
      "loss": 0.6968,
      "num_input_tokens_seen": 5711449,
      "step": 230
    },
    {
      "epoch": 5.449275362318841,
      "grad_norm": 5.303736209869385,
      "learning_rate": 4.70518742287793e-05,
      "loss": 0.851,
      "num_input_tokens_seen": 5852650,
      "step": 235
    },
    {
      "epoch": 5.565217391304348,
      "grad_norm": 3.70810866355896,
      "learning_rate": 4.6927749053136866e-05,
      "loss": 0.716,
      "num_input_tokens_seen": 5972289,
      "step": 240
    },
    {
      "epoch": 5.681159420289855,
      "grad_norm": 3.9204599857330322,
      "learning_rate": 4.6801235203595195e-05,
      "loss": 0.6384,
      "num_input_tokens_seen": 6088707,
      "step": 245
    },
    {
      "epoch": 5.797101449275362,
      "grad_norm": 4.06931209564209,
      "learning_rate": 4.667234646179368e-05,
      "loss": 0.7799,
      "num_input_tokens_seen": 6215471,
      "step": 250
    },
    {
      "epoch": 5.913043478260869,
      "grad_norm": 4.283618450164795,
      "learning_rate": 4.654109686807787e-05,
      "loss": 0.7923,
      "num_input_tokens_seen": 6335935,
      "step": 255
    },
    {
      "epoch": 6.028985507246377,
      "grad_norm": 4.719886302947998,
      "learning_rate": 4.640750071996995e-05,
      "loss": 0.7452,
      "num_input_tokens_seen": 6463689,
      "step": 260
    },
    {
      "epoch": 6.144927536231884,
      "grad_norm": 3.8415334224700928,
      "learning_rate": 4.6271572570611296e-05,
      "loss": 0.4085,
      "num_input_tokens_seen": 6576954,
      "step": 265
    },
    {
      "epoch": 6.260869565217392,
      "grad_norm": 4.19309663772583,
      "learning_rate": 4.613332722717714e-05,
      "loss": 0.5777,
      "num_input_tokens_seen": 6714404,
      "step": 270
    },
    {
      "epoch": 6.3768115942028984,
      "grad_norm": 5.686235427856445,
      "learning_rate": 4.5992779749263546e-05,
      "loss": 0.4718,
      "num_input_tokens_seen": 6840385,
      "step": 275
    },
    {
      "epoch": 6.492753623188406,
      "grad_norm": 3.2365808486938477,
      "learning_rate": 4.584994544724695e-05,
      "loss": 0.3723,
      "num_input_tokens_seen": 6954269,
      "step": 280
    },
    {
      "epoch": 6.608695652173913,
      "grad_norm": 3.530801296234131,
      "learning_rate": 4.5704839880616296e-05,
      "loss": 0.4453,
      "num_input_tokens_seen": 7076143,
      "step": 285
    },
    {
      "epoch": 6.72463768115942,
      "grad_norm": 3.2134931087493896,
      "learning_rate": 4.5557478856278114e-05,
      "loss": 0.5742,
      "num_input_tokens_seen": 7201833,
      "step": 290
    },
    {
      "epoch": 6.840579710144928,
      "grad_norm": 6.281985282897949,
      "learning_rate": 4.5407878426834596e-05,
      "loss": 0.5291,
      "num_input_tokens_seen": 7330479,
      "step": 295
    },
    {
      "epoch": 6.956521739130435,
      "grad_norm": 17.072542190551758,
      "learning_rate": 4.5256054888834934e-05,
      "loss": 0.4968,
      "num_input_tokens_seen": 7449244,
      "step": 300
    },
    {
      "epoch": 7.072463768115942,
      "grad_norm": 3.737456798553467,
      "learning_rate": 4.5102024781000077e-05,
      "loss": 0.421,
      "num_input_tokens_seen": 7578947,
      "step": 305
    },
    {
      "epoch": 7.188405797101449,
      "grad_norm": 2.6586523056030273,
      "learning_rate": 4.4945804882421086e-05,
      "loss": 0.2767,
      "num_input_tokens_seen": 7691948,
      "step": 310
    },
    {
      "epoch": 7.304347826086957,
      "grad_norm": 2.027702808380127,
      "learning_rate": 4.478741221073136e-05,
      "loss": 0.2922,
      "num_input_tokens_seen": 7815786,
      "step": 315
    },
    {
      "epoch": 7.420289855072464,
      "grad_norm": 3.4651787281036377,
      "learning_rate": 4.4626864020252774e-05,
      "loss": 0.2768,
      "num_input_tokens_seen": 7925106,
      "step": 320
    },
    {
      "epoch": 7.536231884057971,
      "grad_norm": 4.559577941894531,
      "learning_rate": 4.446417780011618e-05,
      "loss": 0.3281,
      "num_input_tokens_seen": 8057202,
      "step": 325
    },
    {
      "epoch": 7.6521739130434785,
      "grad_norm": 2.5885751247406006,
      "learning_rate": 4.42993712723562e-05,
      "loss": 0.3374,
      "num_input_tokens_seen": 8187865,
      "step": 330
    },
    {
      "epoch": 7.768115942028985,
      "grad_norm": 2.792222023010254,
      "learning_rate": 4.413246238998069e-05,
      "loss": 0.2491,
      "num_input_tokens_seen": 8304605,
      "step": 335
    },
    {
      "epoch": 7.884057971014493,
      "grad_norm": 3.610206127166748,
      "learning_rate": 4.3963469335015085e-05,
      "loss": 0.3893,
      "num_input_tokens_seen": 8437319,
      "step": 340
    },
    {
      "epoch": 8.0,
      "grad_norm": 4.31643533706665,
      "learning_rate": 4.379241051652174e-05,
      "loss": 0.3761,
      "num_input_tokens_seen": 8573080,
      "step": 345
    },
    {
      "epoch": 8.115942028985508,
      "grad_norm": 2.2834160327911377,
      "learning_rate": 4.361930456859456e-05,
      "loss": 0.236,
      "num_input_tokens_seen": 8707741,
      "step": 350
    },
    {
      "epoch": 8.231884057971014,
      "grad_norm": 2.6929121017456055,
      "learning_rate": 4.34441703483291e-05,
      "loss": 0.1584,
      "num_input_tokens_seen": 8825774,
      "step": 355
    },
    {
      "epoch": 8.347826086956522,
      "grad_norm": 3.8095011711120605,
      "learning_rate": 4.326702693376844e-05,
      "loss": 0.1481,
      "num_input_tokens_seen": 8932249,
      "step": 360
    },
    {
      "epoch": 8.46376811594203,
      "grad_norm": 2.6493489742279053,
      "learning_rate": 4.308789362182492e-05,
      "loss": 0.1743,
      "num_input_tokens_seen": 9051548,
      "step": 365
    },
    {
      "epoch": 8.579710144927537,
      "grad_norm": 30.796459197998047,
      "learning_rate": 4.290678992617798e-05,
      "loss": 0.3162,
      "num_input_tokens_seen": 9197232,
      "step": 370
    },
    {
      "epoch": 8.695652173913043,
      "grad_norm": 3.3164985179901123,
      "learning_rate": 4.272373557514858e-05,
      "loss": 0.2235,
      "num_input_tokens_seen": 9317650,
      "step": 375
    },
    {
      "epoch": 8.81159420289855,
      "grad_norm": 3.1515417098999023,
      "learning_rate": 4.2538750509550054e-05,
      "loss": 0.2504,
      "num_input_tokens_seen": 9450765,
      "step": 380
    },
    {
      "epoch": 8.927536231884059,
      "grad_norm": 3.3926901817321777,
      "learning_rate": 4.235185488051585e-05,
      "loss": 0.2136,
      "num_input_tokens_seen": 9582961,
      "step": 385
    },
    {
      "epoch": 9.043478260869565,
      "grad_norm": 4.670753002166748,
      "learning_rate": 4.216306904730447e-05,
      "loss": 0.1047,
      "num_input_tokens_seen": 9678616,
      "step": 390
    },
    {
      "epoch": 9.159420289855072,
      "grad_norm": 2.166652202606201,
      "learning_rate": 4.1972413575081595e-05,
      "loss": 0.1015,
      "num_input_tokens_seen": 9788512,
      "step": 395
    },
    {
      "epoch": 9.27536231884058,
      "grad_norm": 2.1161272525787354,
      "learning_rate": 4.177990923267986e-05,
      "loss": 0.1505,
      "num_input_tokens_seen": 9916229,
      "step": 400
    },
    {
      "epoch": 9.391304347826088,
      "grad_norm": 2.378105401992798,
      "learning_rate": 4.158557699033644e-05,
      "loss": 0.1135,
      "num_input_tokens_seen": 10042697,
      "step": 405
    },
    {
      "epoch": 9.507246376811594,
      "grad_norm": 2.5567331314086914,
      "learning_rate": 4.138943801740865e-05,
      "loss": 0.1832,
      "num_input_tokens_seen": 10171849,
      "step": 410
    },
    {
      "epoch": 9.623188405797102,
      "grad_norm": 2.022610902786255,
      "learning_rate": 4.119151368006793e-05,
      "loss": 0.1178,
      "num_input_tokens_seen": 10281924,
      "step": 415
    },
    {
      "epoch": 9.73913043478261,
      "grad_norm": 2.5578079223632812,
      "learning_rate": 4.099182553897229e-05,
      "loss": 0.1426,
      "num_input_tokens_seen": 10418758,
      "step": 420
    },
    {
      "epoch": 9.855072463768115,
      "grad_norm": 2.7287228107452393,
      "learning_rate": 4.079039534691767e-05,
      "loss": 0.1603,
      "num_input_tokens_seen": 10558322,
      "step": 425
    },
    {
      "epoch": 9.971014492753623,
      "grad_norm": 2.361532688140869,
      "learning_rate": 4.058724504646834e-05,
      "loss": 0.1548,
      "num_input_tokens_seen": 10679536,
      "step": 430
    },
    {
      "epoch": 10.08695652173913,
      "grad_norm": 1.8757002353668213,
      "learning_rate": 4.0382396767566536e-05,
      "loss": 0.1407,
      "num_input_tokens_seen": 10821076,
      "step": 435
    },
    {
      "epoch": 10.202898550724637,
      "grad_norm": 2.352725028991699,
      "learning_rate": 4.017587282512181e-05,
      "loss": 0.0791,
      "num_input_tokens_seen": 10949771,
      "step": 440
    },
    {
      "epoch": 10.318840579710145,
      "grad_norm": 1.7948939800262451,
      "learning_rate": 3.9967695716580224e-05,
      "loss": 0.0722,
      "num_input_tokens_seen": 11072044,
      "step": 445
    },
    {
      "epoch": 10.434782608695652,
      "grad_norm": 1.954727292060852,
      "learning_rate": 3.975788811947351e-05,
      "loss": 0.0655,
      "num_input_tokens_seen": 11182627,
      "step": 450
    },
    {
      "epoch": 10.55072463768116,
      "grad_norm": 2.143941640853882,
      "learning_rate": 3.954647288894883e-05,
      "loss": 0.0723,
      "num_input_tokens_seen": 11303028,
      "step": 455
    },
    {
      "epoch": 10.666666666666666,
      "grad_norm": 2.0527164936065674,
      "learning_rate": 3.933347305527898e-05,
      "loss": 0.0655,
      "num_input_tokens_seen": 11415868,
      "step": 460
    },
    {
      "epoch": 10.782608695652174,
      "grad_norm": 1.6390535831451416,
      "learning_rate": 3.911891182135371e-05,
      "loss": 0.1534,
      "num_input_tokens_seen": 11555653,
      "step": 465
    },
    {
      "epoch": 10.898550724637682,
      "grad_norm": 2.3848719596862793,
      "learning_rate": 3.8902812560152066e-05,
      "loss": 0.0947,
      "num_input_tokens_seen": 11681065,
      "step": 470
    },
    {
      "epoch": 11.014492753623188,
      "grad_norm": 2.2094757556915283,
      "learning_rate": 3.868519881219631e-05,
      "loss": 0.0868,
      "num_input_tokens_seen": 11809957,
      "step": 475
    },
    {
      "epoch": 11.130434782608695,
      "grad_norm": 4.137216567993164,
      "learning_rate": 3.846609428298757e-05,
      "loss": 0.0467,
      "num_input_tokens_seen": 11937881,
      "step": 480
    },
    {
      "epoch": 11.246376811594203,
      "grad_norm": 1.6658189296722412,
      "learning_rate": 3.824552284042351e-05,
      "loss": 0.0521,
      "num_input_tokens_seen": 12048905,
      "step": 485
    },
    {
      "epoch": 11.36231884057971,
      "grad_norm": 1.5732171535491943,
      "learning_rate": 3.8023508512198256e-05,
      "loss": 0.051,
      "num_input_tokens_seen": 12185453,
      "step": 490
    },
    {
      "epoch": 11.478260869565217,
      "grad_norm": 1.8459701538085938,
      "learning_rate": 3.780007548318507e-05,
      "loss": 0.0753,
      "num_input_tokens_seen": 12310911,
      "step": 495
    },
    {
      "epoch": 11.594202898550725,
      "grad_norm": 1.4724109172821045,
      "learning_rate": 3.7575248092801686e-05,
      "loss": 0.0601,
      "num_input_tokens_seen": 12439708,
      "step": 500
    },
    {
      "epoch": 11.710144927536232,
      "grad_norm": 2.4690322875976562,
      "learning_rate": 3.734905083235901e-05,
      "loss": 0.0533,
      "num_input_tokens_seen": 12554467,
      "step": 505
    },
    {
      "epoch": 11.826086956521738,
      "grad_norm": 2.369218111038208,
      "learning_rate": 3.712150834239313e-05,
      "loss": 0.064,
      "num_input_tokens_seen": 12682329,
      "step": 510
    },
    {
      "epoch": 11.942028985507246,
      "grad_norm": 1.6901100873947144,
      "learning_rate": 3.689264540998116e-05,
      "loss": 0.0755,
      "num_input_tokens_seen": 12800852,
      "step": 515
    },
    {
      "epoch": 12.057971014492754,
      "grad_norm": 1.303114414215088,
      "learning_rate": 3.66624869660411e-05,
      "loss": 0.0553,
      "num_input_tokens_seen": 12917527,
      "step": 520
    },
    {
      "epoch": 12.173913043478262,
      "grad_norm": 1.1986353397369385,
      "learning_rate": 3.6431058082615964e-05,
      "loss": 0.0355,
      "num_input_tokens_seen": 13044774,
      "step": 525
    },
    {
      "epoch": 12.289855072463768,
      "grad_norm": 1.5653026103973389,
      "learning_rate": 3.619838397014263e-05,
      "loss": 0.0413,
      "num_input_tokens_seen": 13175692,
      "step": 530
    },
    {
      "epoch": 12.405797101449275,
      "grad_norm": 1.0767664909362793,
      "learning_rate": 3.5964489974705553e-05,
      "loss": 0.0596,
      "num_input_tokens_seen": 13293164,
      "step": 535
    },
    {
      "epoch": 12.521739130434783,
      "grad_norm": 1.6005312204360962,
      "learning_rate": 3.572940157527572e-05,
      "loss": 0.0479,
      "num_input_tokens_seen": 13417894,
      "step": 540
    },
    {
      "epoch": 12.63768115942029,
      "grad_norm": 1.627121925354004,
      "learning_rate": 3.549314438093515e-05,
      "loss": 0.047,
      "num_input_tokens_seen": 13551913,
      "step": 545
    },
    {
      "epoch": 12.753623188405797,
      "grad_norm": 2.239276647567749,
      "learning_rate": 3.525574412808717e-05,
      "loss": 0.0492,
      "num_input_tokens_seen": 13675309,
      "step": 550
    },
    {
      "epoch": 12.869565217391305,
      "grad_norm": 1.5702998638153076,
      "learning_rate": 3.501722667765286e-05,
      "loss": 0.0471,
      "num_input_tokens_seen": 13797691,
      "step": 555
    },
    {
      "epoch": 12.985507246376812,
      "grad_norm": 1.8216972351074219,
      "learning_rate": 3.47776180122539e-05,
      "loss": 0.1041,
      "num_input_tokens_seen": 13919770,
      "step": 560
    },
    {
      "epoch": 13.101449275362318,
      "grad_norm": 0.9026144742965698,
      "learning_rate": 3.453694423338225e-05,
      "loss": 0.0282,
      "num_input_tokens_seen": 14037673,
      "step": 565
    },
    {
      "epoch": 13.217391304347826,
      "grad_norm": 1.4504765272140503,
      "learning_rate": 3.4295231558556715e-05,
      "loss": 0.0272,
      "num_input_tokens_seen": 14167090,
      "step": 570
    },
    {
      "epoch": 13.333333333333334,
      "grad_norm": 1.4278969764709473,
      "learning_rate": 3.4052506318467084e-05,
      "loss": 0.0342,
      "num_input_tokens_seen": 14311710,
      "step": 575
    },
    {
      "epoch": 13.44927536231884,
      "grad_norm": 1.1284997463226318,
      "learning_rate": 3.3808794954105716e-05,
      "loss": 0.0855,
      "num_input_tokens_seen": 14404322,
      "step": 580
    },
    {
      "epoch": 13.565217391304348,
      "grad_norm": 1.4915614128112793,
      "learning_rate": 3.356412401388732e-05,
      "loss": 0.0378,
      "num_input_tokens_seen": 14530794,
      "step": 585
    },
    {
      "epoch": 13.681159420289855,
      "grad_norm": 1.372157096862793,
      "learning_rate": 3.3318520150756846e-05,
      "loss": 0.0457,
      "num_input_tokens_seen": 14637342,
      "step": 590
    },
    {
      "epoch": 13.797101449275363,
      "grad_norm": 1.6492116451263428,
      "learning_rate": 3.307201011928616e-05,
      "loss": 0.0453,
      "num_input_tokens_seen": 14787534,
      "step": 595
    },
    {
      "epoch": 13.91304347826087,
      "grad_norm": 1.3583859205245972,
      "learning_rate": 3.282462077275947e-05,
      "loss": 0.0378,
      "num_input_tokens_seen": 14909175,
      "step": 600
    },
    {
      "epoch": 14.028985507246377,
      "grad_norm": 1.0751795768737793,
      "learning_rate": 3.257637906024822e-05,
      "loss": 0.0296,
      "num_input_tokens_seen": 15030530,
      "step": 605
    },
    {
      "epoch": 14.144927536231885,
      "grad_norm": 1.474602222442627,
      "learning_rate": 3.2327312023675287e-05,
      "loss": 0.0216,
      "num_input_tokens_seen": 15148359,
      "step": 610
    },
    {
      "epoch": 14.26086956521739,
      "grad_norm": 1.0749961137771606,
      "learning_rate": 3.2077446794869295e-05,
      "loss": 0.0299,
      "num_input_tokens_seen": 15280749,
      "step": 615
    },
    {
      "epoch": 14.376811594202898,
      "grad_norm": 1.4042794704437256,
      "learning_rate": 3.1826810592609036e-05,
      "loss": 0.0247,
      "num_input_tokens_seen": 15397167,
      "step": 620
    },
    {
      "epoch": 14.492753623188406,
      "grad_norm": 1.2280118465423584,
      "learning_rate": 3.157543071965835e-05,
      "loss": 0.0455,
      "num_input_tokens_seen": 15522794,
      "step": 625
    },
    {
      "epoch": 14.608695652173914,
      "grad_norm": 1.2819784879684448,
      "learning_rate": 3.132333455979202e-05,
      "loss": 0.0262,
      "num_input_tokens_seen": 15637987,
      "step": 630
    },
    {
      "epoch": 14.72463768115942,
      "grad_norm": 1.2691748142242432,
      "learning_rate": 3.107054957481271e-05,
      "loss": 0.0281,
      "num_input_tokens_seen": 15773163,
      "step": 635
    },
    {
      "epoch": 14.840579710144928,
      "grad_norm": 1.2752504348754883,
      "learning_rate": 3.081710330155942e-05,
      "loss": 0.0294,
      "num_input_tokens_seen": 15892659,
      "step": 640
    },
    {
      "epoch": 14.956521739130435,
      "grad_norm": 1.3479197025299072,
      "learning_rate": 3.056302334890786e-05,
      "loss": 0.0291,
      "num_input_tokens_seen": 16024576,
      "step": 645
    },
    {
      "epoch": 15.072463768115941,
      "grad_norm": 1.3151382207870483,
      "learning_rate": 3.030833739476285e-05,
      "loss": 0.0216,
      "num_input_tokens_seen": 16151987,
      "step": 650
    },
    {
      "epoch": 15.18840579710145,
      "grad_norm": 2.3882877826690674,
      "learning_rate": 3.0053073183043256e-05,
      "loss": 0.0218,
      "num_input_tokens_seen": 16278639,
      "step": 655
    },
    {
      "epoch": 15.304347826086957,
      "grad_norm": 0.9794278144836426,
      "learning_rate": 2.979725852065981e-05,
      "loss": 0.0283,
      "num_input_tokens_seen": 16414743,
      "step": 660
    },
    {
      "epoch": 15.420289855072463,
      "grad_norm": 0.8964869976043701,
      "learning_rate": 2.954092127448591e-05,
      "loss": 0.0259,
      "num_input_tokens_seen": 16529298,
      "step": 665
    },
    {
      "epoch": 15.53623188405797,
      "grad_norm": 1.1441810131072998,
      "learning_rate": 2.9284089368322045e-05,
      "loss": 0.0716,
      "num_input_tokens_seen": 16655909,
      "step": 670
    },
    {
      "epoch": 15.652173913043478,
      "grad_norm": 1.0959213972091675,
      "learning_rate": 2.9026790779853874e-05,
      "loss": 0.025,
      "num_input_tokens_seen": 16798263,
      "step": 675
    },
    {
      "epoch": 15.768115942028986,
      "grad_norm": 1.0119343996047974,
      "learning_rate": 2.876905353760459e-05,
      "loss": 0.0218,
      "num_input_tokens_seen": 16916827,
      "step": 680
    },
    {
      "epoch": 15.884057971014492,
      "grad_norm": 1.1373978853225708,
      "learning_rate": 2.8510905717881614e-05,
      "loss": 0.0231,
      "num_input_tokens_seen": 17040247,
      "step": 685
    },
    {
      "epoch": 16.0,
      "grad_norm": 1.2512497901916504,
      "learning_rate": 2.8252375441718137e-05,
      "loss": 0.0228,
      "num_input_tokens_seen": 17146160,
      "step": 690
    },
    {
      "epoch": 16.115942028985508,
      "grad_norm": 0.7410117387771606,
      "learning_rate": 2.7993490871809808e-05,
      "loss": 0.029,
      "num_input_tokens_seen": 17284643,
      "step": 695
    },
    {
      "epoch": 16.231884057971016,
      "grad_norm": 1.0934263467788696,
      "learning_rate": 2.7734280209446865e-05,
      "loss": 0.0199,
      "num_input_tokens_seen": 17426644,
      "step": 700
    },
    {
      "epoch": 16.347826086956523,
      "grad_norm": 1.0034395456314087,
      "learning_rate": 2.7474771691442018e-05,
      "loss": 0.0259,
      "num_input_tokens_seen": 17541812,
      "step": 705
    },
    {
      "epoch": 16.463768115942027,
      "grad_norm": 1.4287781715393066,
      "learning_rate": 2.721499358705458e-05,
      "loss": 0.021,
      "num_input_tokens_seen": 17667755,
      "step": 710
    },
    {
      "epoch": 16.579710144927535,
      "grad_norm": 1.0989606380462646,
      "learning_rate": 2.6954974194910888e-05,
      "loss": 0.0199,
      "num_input_tokens_seen": 17788162,
      "step": 715
    },
    {
      "epoch": 16.695652173913043,
      "grad_norm": 0.9687130451202393,
      "learning_rate": 2.6694741839921732e-05,
      "loss": 0.0189,
      "num_input_tokens_seen": 17911718,
      "step": 720
    },
    {
      "epoch": 16.81159420289855,
      "grad_norm": 1.143617033958435,
      "learning_rate": 2.6434324870196748e-05,
      "loss": 0.0169,
      "num_input_tokens_seen": 18018729,
      "step": 725
    },
    {
      "epoch": 16.92753623188406,
      "grad_norm": 1.1395140886306763,
      "learning_rate": 2.617375165395634e-05,
      "loss": 0.0209,
      "num_input_tokens_seen": 18139681,
      "step": 730
    },
    {
      "epoch": 17.043478260869566,
      "grad_norm": 0.881986677646637,
      "learning_rate": 2.5913050576441477e-05,
      "loss": 0.0201,
      "num_input_tokens_seen": 18278544,
      "step": 735
    },
    {
      "epoch": 17.159420289855074,
      "grad_norm": 0.8654409050941467,
      "learning_rate": 2.5652250036821523e-05,
      "loss": 0.017,
      "num_input_tokens_seen": 18396700,
      "step": 740
    },
    {
      "epoch": 17.27536231884058,
      "grad_norm": 0.9699842929840088,
      "learning_rate": 2.5391378445100644e-05,
      "loss": 0.0187,
      "num_input_tokens_seen": 18506229,
      "step": 745
    },
    {
      "epoch": 17.391304347826086,
      "grad_norm": 0.8799194693565369,
      "learning_rate": 2.5130464219022992e-05,
      "loss": 0.0242,
      "num_input_tokens_seen": 18621580,
      "step": 750
    },
    {
      "epoch": 17.507246376811594,
      "grad_norm": 0.9715821146965027,
      "learning_rate": 2.486953578097702e-05,
      "loss": 0.0153,
      "num_input_tokens_seen": 18748382,
      "step": 755
    },
    {
      "epoch": 17.6231884057971,
      "grad_norm": 0.8819458484649658,
      "learning_rate": 2.4608621554899362e-05,
      "loss": 0.0182,
      "num_input_tokens_seen": 18884730,
      "step": 760
    },
    {
      "epoch": 17.73913043478261,
      "grad_norm": 0.8835431933403015,
      "learning_rate": 2.4347749963178486e-05,
      "loss": 0.0143,
      "num_input_tokens_seen": 19003589,
      "step": 765
    },
    {
      "epoch": 17.855072463768117,
      "grad_norm": 0.780754566192627,
      "learning_rate": 2.4086949423558526e-05,
      "loss": 0.0164,
      "num_input_tokens_seen": 19136411,
      "step": 770
    },
    {
      "epoch": 17.971014492753625,
      "grad_norm": 0.7591371536254883,
      "learning_rate": 2.3826248346043663e-05,
      "loss": 0.0157,
      "num_input_tokens_seen": 19260436,
      "step": 775
    },
    {
      "epoch": 18.08695652173913,
      "grad_norm": 0.673797070980072,
      "learning_rate": 2.356567512980326e-05,
      "loss": 0.0304,
      "num_input_tokens_seen": 19388733,
      "step": 780
    },
    {
      "epoch": 18.202898550724637,
      "grad_norm": 0.4008718729019165,
      "learning_rate": 2.3305258160078274e-05,
      "loss": 0.009,
      "num_input_tokens_seen": 19531204,
      "step": 785
    },
    {
      "epoch": 18.318840579710145,
      "grad_norm": 0.6676005125045776,
      "learning_rate": 2.3045025805089118e-05,
      "loss": 0.0105,
      "num_input_tokens_seen": 19624608,
      "step": 790
    },
    {
      "epoch": 18.434782608695652,
      "grad_norm": 0.6956990957260132,
      "learning_rate": 2.278500641294543e-05,
      "loss": 0.0104,
      "num_input_tokens_seen": 19751062,
      "step": 795
    },
    {
      "epoch": 18.55072463768116,
      "grad_norm": 0.80479896068573,
      "learning_rate": 2.252522830855798e-05,
      "loss": 0.0103,
      "num_input_tokens_seen": 19879837,
      "step": 800
    },
    {
      "epoch": 18.666666666666668,
      "grad_norm": 0.7206840515136719,
      "learning_rate": 2.2265719790553147e-05,
      "loss": 0.0107,
      "num_input_tokens_seen": 20019385,
      "step": 805
    },
    {
      "epoch": 18.782608695652176,
      "grad_norm": 0.6994977593421936,
      "learning_rate": 2.2006509128190195e-05,
      "loss": 0.0269,
      "num_input_tokens_seen": 20138003,
      "step": 810
    },
    {
      "epoch": 18.89855072463768,
      "grad_norm": 0.5642988681793213,
      "learning_rate": 2.174762455828187e-05,
      "loss": 0.0086,
      "num_input_tokens_seen": 20260523,
      "step": 815
    },
    {
      "epoch": 19.014492753623188,
      "grad_norm": 0.5547834038734436,
      "learning_rate": 2.1489094282118395e-05,
      "loss": 0.0133,
      "num_input_tokens_seen": 20375322,
      "step": 820
    },
    {
      "epoch": 19.130434782608695,
      "grad_norm": 0.48678871989250183,
      "learning_rate": 2.123094646239541e-05,
      "loss": 0.0114,
      "num_input_tokens_seen": 20477407,
      "step": 825
    },
    {
      "epoch": 19.246376811594203,
      "grad_norm": 0.4791460633277893,
      "learning_rate": 2.0973209220146135e-05,
      "loss": 0.007,
      "num_input_tokens_seen": 20605728,
      "step": 830
    },
    {
      "epoch": 19.36231884057971,
      "grad_norm": 1.1198338270187378,
      "learning_rate": 2.0715910631677968e-05,
      "loss": 0.0088,
      "num_input_tokens_seen": 20725799,
      "step": 835
    },
    {
      "epoch": 19.47826086956522,
      "grad_norm": 0.6645247936248779,
      "learning_rate": 2.0459078725514092e-05,
      "loss": 0.007,
      "num_input_tokens_seen": 20865534,
      "step": 840
    },
    {
      "epoch": 19.594202898550726,
      "grad_norm": 0.5324479341506958,
      "learning_rate": 2.020274147934019e-05,
      "loss": 0.0059,
      "num_input_tokens_seen": 20977913,
      "step": 845
    },
    {
      "epoch": 19.71014492753623,
      "grad_norm": 0.6183504462242126,
      "learning_rate": 1.9946926816956743e-05,
      "loss": 0.0069,
      "num_input_tokens_seen": 21102848,
      "step": 850
    },
    {
      "epoch": 19.82608695652174,
      "grad_norm": 0.6665703058242798,
      "learning_rate": 1.9691662605237166e-05,
      "loss": 0.008,
      "num_input_tokens_seen": 21243679,
      "step": 855
    },
    {
      "epoch": 19.942028985507246,
      "grad_norm": 0.3298584222793579,
      "learning_rate": 1.9436976651092144e-05,
      "loss": 0.0127,
      "num_input_tokens_seen": 21364202,
      "step": 860
    },
    {
      "epoch": 20.057971014492754,
      "grad_norm": 0.2818591296672821,
      "learning_rate": 1.9182896698440584e-05,
      "loss": 0.0059,
      "num_input_tokens_seen": 21496089,
      "step": 865
    },
    {
      "epoch": 20.17391304347826,
      "grad_norm": 0.6906440258026123,
      "learning_rate": 1.89294504251873e-05,
      "loss": 0.0046,
      "num_input_tokens_seen": 21603193,
      "step": 870
    },
    {
      "epoch": 20.28985507246377,
      "grad_norm": 0.33482542634010315,
      "learning_rate": 1.867666544020798e-05,
      "loss": 0.0058,
      "num_input_tokens_seen": 21742062,
      "step": 875
    },
    {
      "epoch": 20.405797101449274,
      "grad_norm": 2.443847417831421,
      "learning_rate": 1.8424569280341653e-05,
      "loss": 0.0082,
      "num_input_tokens_seen": 21869307,
      "step": 880
    },
    {
      "epoch": 20.52173913043478,
      "grad_norm": 0.43886587023735046,
      "learning_rate": 1.817318940739098e-05,
      "loss": 0.0148,
      "num_input_tokens_seen": 21992573,
      "step": 885
    },
    {
      "epoch": 20.63768115942029,
      "grad_norm": 0.93570876121521,
      "learning_rate": 1.7922553205130707e-05,
      "loss": 0.0064,
      "num_input_tokens_seen": 22101845,
      "step": 890
    },
    {
      "epoch": 20.753623188405797,
      "grad_norm": 1176.9595947265625,
      "learning_rate": 1.767268797632472e-05,
      "loss": 0.008,
      "num_input_tokens_seen": 22230253,
      "step": 895
    },
    {
      "epoch": 20.869565217391305,
      "grad_norm": 0.35642215609550476,
      "learning_rate": 1.7423620939751788e-05,
      "loss": 0.0053,
      "num_input_tokens_seen": 22373454,
      "step": 900
    },
    {
      "epoch": 20.985507246376812,
      "grad_norm": 0.39736178517341614,
      "learning_rate": 1.7175379227240523e-05,
      "loss": 0.0054,
      "num_input_tokens_seen": 22493123,
      "step": 905
    },
    {
      "epoch": 21.10144927536232,
      "grad_norm": 0.5092463493347168,
      "learning_rate": 1.692798988071385e-05,
      "loss": 0.0044,
      "num_input_tokens_seen": 22629005,
      "step": 910
    },
    {
      "epoch": 21.217391304347824,
      "grad_norm": 0.26361697912216187,
      "learning_rate": 1.6681479849243153e-05,
      "loss": 0.0043,
      "num_input_tokens_seen": 22752358,
      "step": 915
    },
    {
      "epoch": 21.333333333333332,
      "grad_norm": 0.19933666288852692,
      "learning_rate": 1.6435875986112685e-05,
      "loss": 0.0035,
      "num_input_tokens_seen": 22880349,
      "step": 920
    },
    {
      "epoch": 21.44927536231884,
      "grad_norm": 0.22622954845428467,
      "learning_rate": 1.6191205045894283e-05,
      "loss": 0.0044,
      "num_input_tokens_seen": 22987343,
      "step": 925
    },
    {
      "epoch": 21.565217391304348,
      "grad_norm": 0.30199098587036133,
      "learning_rate": 1.594749368153292e-05,
      "loss": 0.0178,
      "num_input_tokens_seen": 23113462,
      "step": 930
    },
    {
      "epoch": 21.681159420289855,
      "grad_norm": 0.9627483487129211,
      "learning_rate": 1.570476844144329e-05,
      "loss": 0.0089,
      "num_input_tokens_seen": 23221714,
      "step": 935
    },
    {
      "epoch": 21.797101449275363,
      "grad_norm": 0.27791452407836914,
      "learning_rate": 1.546305576661776e-05,
      "loss": 0.004,
      "num_input_tokens_seen": 23368857,
      "step": 940
    },
    {
      "epoch": 21.91304347826087,
      "grad_norm": 0.3269965648651123,
      "learning_rate": 1.5222381987746104e-05,
      "loss": 0.004,
      "num_input_tokens_seen": 23494483,
      "step": 945
    },
    {
      "epoch": 22.028985507246375,
      "grad_norm": 0.15966826677322388,
      "learning_rate": 1.4982773322347144e-05,
      "loss": 0.0034,
      "num_input_tokens_seen": 23605463,
      "step": 950
    },
    {
      "epoch": 22.144927536231883,
      "grad_norm": 0.3009255826473236,
      "learning_rate": 1.4744255871912823e-05,
      "loss": 0.0066,
      "num_input_tokens_seen": 23715776,
      "step": 955
    },
    {
      "epoch": 22.26086956521739,
      "grad_norm": 0.4215935170650482,
      "learning_rate": 1.4506855619064846e-05,
      "loss": 0.0034,
      "num_input_tokens_seen": 23841669,
      "step": 960
    },
    {
      "epoch": 22.3768115942029,
      "grad_norm": 0.20214155316352844,
      "learning_rate": 1.4270598424724292e-05,
      "loss": 0.0032,
      "num_input_tokens_seen": 23960567,
      "step": 965
    },
    {
      "epoch": 22.492753623188406,
      "grad_norm": 7.0683207511901855,
      "learning_rate": 1.4035510025294462e-05,
      "loss": 0.0124,
      "num_input_tokens_seen": 24074628,
      "step": 970
    },
    {
      "epoch": 22.608695652173914,
      "grad_norm": 0.20178793370723724,
      "learning_rate": 1.3801616029857378e-05,
      "loss": 0.0027,
      "num_input_tokens_seen": 24214324,
      "step": 975
    },
    {
      "epoch": 22.72463768115942,
      "grad_norm": 1.3855236768722534,
      "learning_rate": 1.3568941917384036e-05,
      "loss": 0.0037,
      "num_input_tokens_seen": 24326727,
      "step": 980
    },
    {
      "epoch": 22.840579710144926,
      "grad_norm": 0.18420317769050598,
      "learning_rate": 1.3337513033958904e-05,
      "loss": 0.0029,
      "num_input_tokens_seen": 24456961,
      "step": 985
    },
    {
      "epoch": 22.956521739130434,
      "grad_norm": 0.15907694399356842,
      "learning_rate": 1.310735459001884e-05,
      "loss": 0.0035,
      "num_input_tokens_seen": 24606652,
      "step": 990
    },
    {
      "epoch": 23.07246376811594,
      "grad_norm": 0.2548115849494934,
      "learning_rate": 1.2878491657606872e-05,
      "loss": 0.002,
      "num_input_tokens_seen": 24710410,
      "step": 995
    },
    {
      "epoch": 23.18840579710145,
      "grad_norm": 0.36587971448898315,
      "learning_rate": 1.2650949167640993e-05,
      "loss": 0.0023,
      "num_input_tokens_seen": 24831908,
      "step": 1000
    },
    {
      "epoch": 23.304347826086957,
      "grad_norm": 0.13662408292293549,
      "learning_rate": 1.2424751907198312e-05,
      "loss": 0.0031,
      "num_input_tokens_seen": 24951342,
      "step": 1005
    },
    {
      "epoch": 23.420289855072465,
      "grad_norm": 0.19979843497276306,
      "learning_rate": 1.2199924516814939e-05,
      "loss": 0.0027,
      "num_input_tokens_seen": 25088309,
      "step": 1010
    },
    {
      "epoch": 23.536231884057973,
      "grad_norm": 0.14170995354652405,
      "learning_rate": 1.1976491487801748e-05,
      "loss": 0.0124,
      "num_input_tokens_seen": 25216080,
      "step": 1015
    },
    {
      "epoch": 23.652173913043477,
      "grad_norm": 0.06863216310739517,
      "learning_rate": 1.1754477159576499e-05,
      "loss": 0.0023,
      "num_input_tokens_seen": 25326581,
      "step": 1020
    },
    {
      "epoch": 23.768115942028984,
      "grad_norm": 0.25133436918258667,
      "learning_rate": 1.1533905717012428e-05,
      "loss": 0.0027,
      "num_input_tokens_seen": 25477500,
      "step": 1025
    },
    {
      "epoch": 23.884057971014492,
      "grad_norm": 0.28348398208618164,
      "learning_rate": 1.1314801187803686e-05,
      "loss": 0.0041,
      "num_input_tokens_seen": 25601354,
      "step": 1030
    },
    {
      "epoch": 24.0,
      "grad_norm": 0.5024954676628113,
      "learning_rate": 1.1097187439847939e-05,
      "loss": 0.0021,
      "num_input_tokens_seen": 25719240,
      "step": 1035
    },
    {
      "epoch": 24.115942028985508,
      "grad_norm": 0.1774568408727646,
      "learning_rate": 1.088108817864629e-05,
      "loss": 0.0039,
      "num_input_tokens_seen": 25834910,
      "step": 1040
    },
    {
      "epoch": 24.231884057971016,
      "grad_norm": 0.08105342090129852,
      "learning_rate": 1.0666526944721016e-05,
      "loss": 0.0025,
      "num_input_tokens_seen": 25974530,
      "step": 1045
    },
    {
      "epoch": 24.347826086956523,
      "grad_norm": 0.13048779964447021,
      "learning_rate": 1.0453527111051184e-05,
      "loss": 0.002,
      "num_input_tokens_seen": 26104464,
      "step": 1050
    },
    {
      "epoch": 24.463768115942027,
      "grad_norm": 0.10774020105600357,
      "learning_rate": 1.0242111880526495e-05,
      "loss": 0.0024,
      "num_input_tokens_seen": 26251334,
      "step": 1055
    },
    {
      "epoch": 24.579710144927535,
      "grad_norm": 0.7494776248931885,
      "learning_rate": 1.003230428341979e-05,
      "loss": 0.0031,
      "num_input_tokens_seen": 26366561,
      "step": 1060
    },
    {
      "epoch": 24.695652173913043,
      "grad_norm": 0.3580308258533478,
      "learning_rate": 9.824127174878195e-06,
      "loss": 0.0022,
      "num_input_tokens_seen": 26486437,
      "step": 1065
    },
    {
      "epoch": 24.81159420289855,
      "grad_norm": 0.1473228931427002,
      "learning_rate": 9.617603232433475e-06,
      "loss": 0.0022,
      "num_input_tokens_seen": 26601526,
      "step": 1070
    },
    {
      "epoch": 24.92753623188406,
      "grad_norm": 0.11716706305742264,
      "learning_rate": 9.412754953531663e-06,
      "loss": 0.0109,
      "num_input_tokens_seen": 26727922,
      "step": 1075
    },
    {
      "epoch": 25.043478260869566,
      "grad_norm": 0.12043190747499466,
      "learning_rate": 9.209604653082326e-06,
      "loss": 0.0019,
      "num_input_tokens_seen": 26835621,
      "step": 1080
    },
    {
      "epoch": 25.159420289855074,
      "grad_norm": 0.1277165412902832,
      "learning_rate": 9.008174461027724e-06,
      "loss": 0.0016,
      "num_input_tokens_seen": 26955101,
      "step": 1085
    },
    {
      "epoch": 25.27536231884058,
      "grad_norm": 0.08892516791820526,
      "learning_rate": 8.808486319932083e-06,
      "loss": 0.002,
      "num_input_tokens_seen": 27077833,
      "step": 1090
    },
    {
      "epoch": 25.391304347826086,
      "grad_norm": 0.30754807591438293,
      "learning_rate": 8.610561982591357e-06,
      "loss": 0.0018,
      "num_input_tokens_seen": 27192758,
      "step": 1095
    },
    {
      "epoch": 25.507246376811594,
      "grad_norm": 0.7194050550460815,
      "learning_rate": 8.414423009663563e-06,
      "loss": 0.0028,
      "num_input_tokens_seen": 27324970,
      "step": 1100
    },
    {
      "epoch": 25.6231884057971,
      "grad_norm": 4777.61328125,
      "learning_rate": 8.220090767320137e-06,
      "loss": 0.0021,
      "num_input_tokens_seen": 27477531,
      "step": 1105
    },
    {
      "epoch": 25.73913043478261,
      "grad_norm": 2.280327081680298,
      "learning_rate": 8.027586424918412e-06,
      "loss": 0.0057,
      "num_input_tokens_seen": 27592035,
      "step": 1110
    },
    {
      "epoch": 25.855072463768117,
      "grad_norm": 0.13882993161678314,
      "learning_rate": 7.836930952695533e-06,
      "loss": 0.0067,
      "num_input_tokens_seen": 27712377,
      "step": 1115
    },
    {
      "epoch": 25.971014492753625,
      "grad_norm": 0.20987676084041595,
      "learning_rate": 7.648145119484153e-06,
      "loss": 0.002,
      "num_input_tokens_seen": 27834613,
      "step": 1120
    },
    {
      "epoch": 26.08695652173913,
      "grad_norm": 0.09795770049095154,
      "learning_rate": 7.461249490449954e-06,
      "loss": 0.0021,
      "num_input_tokens_seen": 27966996,
      "step": 1125
    },
    {
      "epoch": 26.202898550724637,
      "grad_norm": 0.14506971836090088,
      "learning_rate": 7.276264424851423e-06,
      "loss": 0.002,
      "num_input_tokens_seen": 28093538,
      "step": 1130
    },
    {
      "epoch": 26.318840579710145,
      "grad_norm": 0.08091314136981964,
      "learning_rate": 7.0932100738220265e-06,
      "loss": 0.0017,
      "num_input_tokens_seen": 28215579,
      "step": 1135
    },
    {
      "epoch": 26.434782608695652,
      "grad_norm": 0.22550061345100403,
      "learning_rate": 6.912106378175098e-06,
      "loss": 0.0014,
      "num_input_tokens_seen": 28344144,
      "step": 1140
    },
    {
      "epoch": 26.55072463768116,
      "grad_norm": 0.23987355828285217,
      "learning_rate": 6.732973066231563e-06,
      "loss": 0.0022,
      "num_input_tokens_seen": 28478650,
      "step": 1145
    },
    {
      "epoch": 26.666666666666668,
      "grad_norm": 0.1993756741285324,
      "learning_rate": 6.555829651670911e-06,
      "loss": 0.0023,
      "num_input_tokens_seen": 28593004,
      "step": 1150
    },
    {
      "epoch": 26.782608695652176,
      "grad_norm": 0.7184757590293884,
      "learning_rate": 6.380695431405456e-06,
      "loss": 0.0028,
      "num_input_tokens_seen": 28707392,
      "step": 1155
    },
    {
      "epoch": 26.89855072463768,
      "grad_norm": 0.06247011199593544,
      "learning_rate": 6.207589483478266e-06,
      "loss": 0.006,
      "num_input_tokens_seen": 28834902,
      "step": 1160
    },
    {
      "epoch": 27.014492753623188,
      "grad_norm": 0.11046591401100159,
      "learning_rate": 6.0365306649849214e-06,
      "loss": 0.0045,
      "num_input_tokens_seen": 28948812,
      "step": 1165
    },
    {
      "epoch": 27.130434782608695,
      "grad_norm": 0.12309098988771439,
      "learning_rate": 5.867537610019317e-06,
      "loss": 0.0019,
      "num_input_tokens_seen": 29078309,
      "step": 1170
    },
    {
      "epoch": 27.246376811594203,
      "grad_norm": 0.11428932845592499,
      "learning_rate": 5.700628727643806e-06,
      "loss": 0.002,
      "num_input_tokens_seen": 29211503,
      "step": 1175
    },
    {
      "epoch": 27.36231884057971,
      "grad_norm": 0.1093268170952797,
      "learning_rate": 5.53582219988382e-06,
      "loss": 0.0019,
      "num_input_tokens_seen": 29344489,
      "step": 1180
    },
    {
      "epoch": 27.47826086956522,
      "grad_norm": 0.2166384607553482,
      "learning_rate": 5.373135979747227e-06,
      "loss": 0.006,
      "num_input_tokens_seen": 29464082,
      "step": 1185
    },
    {
      "epoch": 27.594202898550726,
      "grad_norm": 0.15387850999832153,
      "learning_rate": 5.2125877892686496e-06,
      "loss": 0.0043,
      "num_input_tokens_seen": 29581124,
      "step": 1190
    },
    {
      "epoch": 27.71014492753623,
      "grad_norm": 0.11962082982063293,
      "learning_rate": 5.054195117578914e-06,
      "loss": 0.0019,
      "num_input_tokens_seen": 29696346,
      "step": 1195
    },
    {
      "epoch": 27.82608695652174,
      "grad_norm": 0.18724732100963593,
      "learning_rate": 4.897975218999926e-06,
      "loss": 0.002,
      "num_input_tokens_seen": 29815117,
      "step": 1200
    },
    {
      "epoch": 27.942028985507246,
      "grad_norm": 0.09917350113391876,
      "learning_rate": 4.743945111165068e-06,
      "loss": 0.0022,
      "num_input_tokens_seen": 29939175,
      "step": 1205
    },
    {
      "epoch": 28.057971014492754,
      "grad_norm": 0.08235369622707367,
      "learning_rate": 4.592121573165414e-06,
      "loss": 0.0016,
      "num_input_tokens_seen": 30079840,
      "step": 1210
    },
    {
      "epoch": 28.17391304347826,
      "grad_norm": 0.20488996803760529,
      "learning_rate": 4.442521143721892e-06,
      "loss": 0.0033,
      "num_input_tokens_seen": 30192219,
      "step": 1215
    },
    {
      "epoch": 28.28985507246377,
      "grad_norm": 0.05383768677711487,
      "learning_rate": 4.295160119383712e-06,
      "loss": 0.0018,
      "num_input_tokens_seen": 30330969,
      "step": 1220
    },
    {
      "epoch": 28.405797101449274,
      "grad_norm": 0.14237363636493683,
      "learning_rate": 4.150054552753055e-06,
      "loss": 0.0018,
      "num_input_tokens_seen": 30453302,
      "step": 1225
    },
    {
      "epoch": 28.52173913043478,
      "grad_norm": 0.12487669289112091,
      "learning_rate": 4.007220250736454e-06,
      "loss": 0.0078,
      "num_input_tokens_seen": 30568943,
      "step": 1230
    },
    {
      "epoch": 28.63768115942029,
      "grad_norm": 0.1423855572938919,
      "learning_rate": 3.866672772822863e-06,
      "loss": 0.0019,
      "num_input_tokens_seen": 30696057,
      "step": 1235
    },
    {
      "epoch": 28.753623188405797,
      "grad_norm": 0.1543101817369461,
      "learning_rate": 3.7284274293887115e-06,
      "loss": 0.0019,
      "num_input_tokens_seen": 30815506,
      "step": 1240
    },
    {
      "epoch": 28.869565217391305,
      "grad_norm": 0.1402539610862732,
      "learning_rate": 3.592499280030057e-06,
      "loss": 0.0027,
      "num_input_tokens_seen": 30916446,
      "step": 1245
    },
    {
      "epoch": 28.985507246376812,
      "grad_norm": 0.26191645860671997,
      "learning_rate": 3.458903131922134e-06,
      "loss": 0.0023,
      "num_input_tokens_seen": 31054242,
      "step": 1250
    },
    {
      "epoch": 29.10144927536232,
      "grad_norm": 0.09874732792377472,
      "learning_rate": 3.3276535382063213e-06,
      "loss": 0.0029,
      "num_input_tokens_seen": 31189078,
      "step": 1255
    },
    {
      "epoch": 29.217391304347824,
      "grad_norm": 0.11677820980548859,
      "learning_rate": 3.198764796404807e-06,
      "loss": 0.0018,
      "num_input_tokens_seen": 31311374,
      "step": 1260
    },
    {
      "epoch": 29.333333333333332,
      "grad_norm": 0.05459802597761154,
      "learning_rate": 3.0722509468631392e-06,
      "loss": 0.0018,
      "num_input_tokens_seen": 31444681,
      "step": 1265
    },
    {
      "epoch": 29.44927536231884,
      "grad_norm": 0.1113714948296547,
      "learning_rate": 2.948125771220697e-06,
      "loss": 0.0018,
      "num_input_tokens_seen": 31567569,
      "step": 1270
    },
    {
      "epoch": 29.565217391304348,
      "grad_norm": 0.1816156655550003,
      "learning_rate": 2.8264027909094715e-06,
      "loss": 0.0019,
      "num_input_tokens_seen": 31697338,
      "step": 1275
    },
    {
      "epoch": 29.681159420289855,
      "grad_norm": 0.13639949262142181,
      "learning_rate": 2.707095265681081e-06,
      "loss": 0.0018,
      "num_input_tokens_seen": 31826661,
      "step": 1280
    },
    {
      "epoch": 29.797101449275363,
      "grad_norm": 0.05292365327477455,
      "learning_rate": 2.5902161921623454e-06,
      "loss": 0.0023,
      "num_input_tokens_seen": 31944680,
      "step": 1285
    },
    {
      "epoch": 29.91304347826087,
      "grad_norm": 0.16608740389347076,
      "learning_rate": 2.475778302439524e-06,
      "loss": 0.0078,
      "num_input_tokens_seen": 32067106,
      "step": 1290
    },
    {
      "epoch": 30.028985507246375,
      "grad_norm": 0.09277443587779999,
      "learning_rate": 2.3637940626713346e-06,
      "loss": 0.0018,
      "num_input_tokens_seen": 32184526,
      "step": 1295
    },
    {
      "epoch": 30.144927536231883,
      "grad_norm": 0.18832191824913025,
      "learning_rate": 2.254275671731007e-06,
      "loss": 0.0017,
      "num_input_tokens_seen": 32309423,
      "step": 1300
    },
    {
      "epoch": 30.26086956521739,
      "grad_norm": 0.1828456073999405,
      "learning_rate": 2.14723505987737e-06,
      "loss": 0.0071,
      "num_input_tokens_seen": 32429445,
      "step": 1305
    },
    {
      "epoch": 30.3768115942029,
      "grad_norm": 0.07503814995288849,
      "learning_rate": 2.0426838874552714e-06,
      "loss": 0.0016,
      "num_input_tokens_seen": 32540571,
      "step": 1310
    },
    {
      "epoch": 30.492753623188406,
      "grad_norm": 0.19047732651233673,
      "learning_rate": 1.9406335436253724e-06,
      "loss": 0.0018,
      "num_input_tokens_seen": 32665528,
      "step": 1315
    },
    {
      "epoch": 30.608695652173914,
      "grad_norm": 0.17791509628295898,
      "learning_rate": 1.8410951451234533e-06,
      "loss": 0.0017,
      "num_input_tokens_seen": 32800773,
      "step": 1320
    },
    {
      "epoch": 30.72463768115942,
      "grad_norm": 0.10698456317186356,
      "learning_rate": 1.7440795350494588e-06,
      "loss": 0.0017,
      "num_input_tokens_seen": 32928397,
      "step": 1325
    },
    {
      "epoch": 30.840579710144926,
      "grad_norm": 0.0963551327586174,
      "learning_rate": 1.649597281686302e-06,
      "loss": 0.0019,
      "num_input_tokens_seen": 33054819,
      "step": 1330
    },
    {
      "epoch": 30.956521739130434,
      "grad_norm": 0.24703514575958252,
      "learning_rate": 1.5576586773486195e-06,
      "loss": 0.0018,
      "num_input_tokens_seen": 33180616,
      "step": 1335
    },
    {
      "epoch": 31.07246376811594,
      "grad_norm": 0.12497910857200623,
      "learning_rate": 1.4682737372615967e-06,
      "loss": 0.0038,
      "num_input_tokens_seen": 33298041,
      "step": 1340
    },
    {
      "epoch": 31.18840579710145,
      "grad_norm": 0.18260960280895233,
      "learning_rate": 1.3814521984699596e-06,
      "loss": 0.0052,
      "num_input_tokens_seen": 33408343,
      "step": 1345
    },
    {
      "epoch": 31.304347826086957,
      "grad_norm": 0.13422255218029022,
      "learning_rate": 1.297203518777293e-06,
      "loss": 0.0018,
      "num_input_tokens_seen": 33545364,
      "step": 1350
    },
    {
      "epoch": 31.420289855072465,
      "grad_norm": 0.1285027116537094,
      "learning_rate": 1.2155368757157643e-06,
      "loss": 0.0019,
      "num_input_tokens_seen": 33652900,
      "step": 1355
    },
    {
      "epoch": 31.536231884057973,
      "grad_norm": 0.12832242250442505,
      "learning_rate": 1.1364611655463736e-06,
      "loss": 0.0019,
      "num_input_tokens_seen": 33768791,
      "step": 1360
    },
    {
      "epoch": 31.652173913043477,
      "grad_norm": 0.12093157321214676,
      "learning_rate": 1.0599850022898539e-06,
      "loss": 0.0017,
      "num_input_tokens_seen": 33892837,
      "step": 1365
    },
    {
      "epoch": 31.768115942028984,
      "grad_norm": 0.7227018475532532,
      "learning_rate": 9.861167167883046e-07,
      "loss": 0.0022,
      "num_input_tokens_seen": 34015288,
      "step": 1370
    },
    {
      "epoch": 31.884057971014492,
      "grad_norm": 2.143653631210327,
      "learning_rate": 9.148643557976955e-07,
      "loss": 0.0037,
      "num_input_tokens_seen": 34154884,
      "step": 1375
    },
    {
      "epoch": 32.0,
      "grad_norm": 0.17518474161624908,
      "learning_rate": 8.462356811112987e-07,
      "loss": 0.0019,
      "num_input_tokens_seen": 34292320,
      "step": 1380
    },
    {
      "epoch": 32.11594202898551,
      "grad_norm": 0.1274159997701645,
      "learning_rate": 7.802381687141535e-07,
      "loss": 0.0017,
      "num_input_tokens_seen": 34413850,
      "step": 1385
    },
    {
      "epoch": 32.231884057971016,
      "grad_norm": 0.11443401873111725,
      "learning_rate": 7.168790079686932e-07,
      "loss": 0.0018,
      "num_input_tokens_seen": 34547127,
      "step": 1390
    },
    {
      "epoch": 32.34782608695652,
      "grad_norm": 0.08239752799272537,
      "learning_rate": 6.561651008315738e-07,
      "loss": 0.0035,
      "num_input_tokens_seen": 34685112,
      "step": 1395
    },
    {
      "epoch": 32.46376811594203,
      "grad_norm": 0.7361220717430115,
      "learning_rate": 5.981030611018234e-07,
      "loss": 0.0063,
      "num_input_tokens_seen": 34810484,
      "step": 1400
    },
    {
      "epoch": 32.57971014492754,
      "grad_norm": 0.20323431491851807,
      "learning_rate": 5.426992137003622e-07,
      "loss": 0.0018,
      "num_input_tokens_seen": 34920531,
      "step": 1405
    },
    {
      "epoch": 32.69565217391305,
      "grad_norm": 0.11165229231119156,
      "learning_rate": 4.899595939810236e-07,
      "loss": 0.002,
      "num_input_tokens_seen": 35035657,
      "step": 1410
    },
    {
      "epoch": 32.81159420289855,
      "grad_norm": 0.15023387968540192,
      "learning_rate": 4.398899470730827e-07,
      "loss": 0.0017,
      "num_input_tokens_seen": 35167466,
      "step": 1415
    },
    {
      "epoch": 32.927536231884055,
      "grad_norm": 0.18479810655117035,
      "learning_rate": 3.9249572725543196e-07,
      "loss": 0.0016,
      "num_input_tokens_seen": 35296818,
      "step": 1420
    },
    {
      "epoch": 33.04347826086956,
      "grad_norm": 0.08527754247188568,
      "learning_rate": 3.477820973624063e-07,
      "loss": 0.0015,
      "num_input_tokens_seen": 35430399,
      "step": 1425
    },
    {
      "epoch": 33.15942028985507,
      "grad_norm": 0.16888481378555298,
      "learning_rate": 3.0575392822139726e-07,
      "loss": 0.0057,
      "num_input_tokens_seen": 35551540,
      "step": 1430
    },
    {
      "epoch": 33.27536231884058,
      "grad_norm": 0.18187086284160614,
      "learning_rate": 2.664157981222437e-07,
      "loss": 0.0016,
      "num_input_tokens_seen": 35676077,
      "step": 1435
    },
    {
      "epoch": 33.391304347826086,
      "grad_norm": 0.15047162771224976,
      "learning_rate": 2.297719923185032e-07,
      "loss": 0.0016,
      "num_input_tokens_seen": 35785127,
      "step": 1440
    },
    {
      "epoch": 33.507246376811594,
      "grad_norm": 0.12288761883974075,
      "learning_rate": 1.9582650256064205e-07,
      "loss": 0.0019,
      "num_input_tokens_seen": 35911682,
      "step": 1445
    },
    {
      "epoch": 33.6231884057971,
      "grad_norm": 0.22509098052978516,
      "learning_rate": 1.645830266611914e-07,
      "loss": 0.0017,
      "num_input_tokens_seen": 36030754,
      "step": 1450
    },
    {
      "epoch": 33.73913043478261,
      "grad_norm": 2.0408618450164795,
      "learning_rate": 1.3604496809195288e-07,
      "loss": 0.0042,
      "num_input_tokens_seen": 36146749,
      "step": 1455
    },
    {
      "epoch": 33.85507246376812,
      "grad_norm": 0.10705255717039108,
      "learning_rate": 1.1021543561322012e-07,
      "loss": 0.0017,
      "num_input_tokens_seen": 36278454,
      "step": 1460
    },
    {
      "epoch": 33.971014492753625,
      "grad_norm": 1876.0384521484375,
      "learning_rate": 8.709724293513854e-08,
      "loss": 0.0017,
      "num_input_tokens_seen": 36408834,
      "step": 1465
    },
    {
      "epoch": 34.08695652173913,
      "grad_norm": 0.1927630454301834,
      "learning_rate": 6.66929084112089e-08,
      "loss": 0.0015,
      "num_input_tokens_seen": 36550538,
      "step": 1470
    },
    {
      "epoch": 34.20289855072464,
      "grad_norm": 0.1668202131986618,
      "learning_rate": 4.900465476393168e-08,
      "loss": 0.0018,
      "num_input_tokens_seen": 36647436,
      "step": 1475
    },
    {
      "epoch": 34.31884057971015,
      "grad_norm": 0.7123565673828125,
      "learning_rate": 3.403440884269526e-08,
      "loss": 0.0024,
      "num_input_tokens_seen": 36785387,
      "step": 1480
    },
    {
      "epoch": 34.43478260869565,
      "grad_norm": 0.16973845660686493,
      "learning_rate": 2.1783801413866046e-08,
      "loss": 0.0021,
      "num_input_tokens_seen": 36915606,
      "step": 1485
    },
    {
      "epoch": 34.55072463768116,
      "grad_norm": 2.034724473953247,
      "learning_rate": 1.2254166983152737e-08,
      "loss": 0.0035,
      "num_input_tokens_seen": 37036117,
      "step": 1490
    },
    {
      "epoch": 34.666666666666664,
      "grad_norm": 0.155415877699852,
      "learning_rate": 5.446543650219904e-09,
      "loss": 0.0016,
      "num_input_tokens_seen": 37165587,
      "step": 1495
    },
    {
      "epoch": 34.78260869565217,
      "grad_norm": 0.10199662297964096,
      "learning_rate": 1.3616729956228425e-09,
      "loss": 0.0015,
      "num_input_tokens_seen": 37290827,
      "step": 1500
    },
    {
      "epoch": 34.89855072463768,
      "grad_norm": 0.14740267395973206,
      "learning_rate": 0.0,
      "loss": 0.0053,
      "num_input_tokens_seen": 37412688,
      "step": 1505
    },
    {
      "epoch": 34.89855072463768,
      "num_input_tokens_seen": 37412688,
      "step": 1505,
      "total_flos": 8.033958240027034e+16,
      "train_loss": 0.3889684765070578,
      "train_runtime": 37510.9602,
      "train_samples_per_second": 0.322,
      "train_steps_per_second": 0.04
    }
  ],
  "logging_steps": 5,
  "max_steps": 1505,
  "num_input_tokens_seen": 37412688,
  "num_train_epochs": 35,
  "save_steps": 100,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 8.033958240027034e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}