{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 1875,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.008,
      "grad_norm": 5.211891174316406,
      "learning_rate": 4.999943853127351e-05,
      "loss": 4.482,
      "num_input_tokens_seen": 6000,
      "step": 5
    },
    {
      "epoch": 0.016,
      "grad_norm": 2.5028183460235596,
      "learning_rate": 4.999715760779541e-05,
      "loss": 3.7966,
      "num_input_tokens_seen": 12912,
      "step": 10
    },
    {
      "epoch": 0.024,
      "grad_norm": 2.7427284717559814,
      "learning_rate": 4.999312229773022e-05,
      "loss": 3.4993,
      "num_input_tokens_seen": 18512,
      "step": 15
    },
    {
      "epoch": 0.032,
      "grad_norm": 2.573559522628784,
      "learning_rate": 4.998733288428987e-05,
      "loss": 3.1988,
      "num_input_tokens_seen": 25584,
      "step": 20
    },
    {
      "epoch": 0.04,
      "grad_norm": 2.6185526847839355,
      "learning_rate": 4.997978977379536e-05,
      "loss": 2.8839,
      "num_input_tokens_seen": 32128,
      "step": 25
    },
    {
      "epoch": 0.048,
      "grad_norm": 2.775859832763672,
      "learning_rate": 4.997049349564814e-05,
      "loss": 2.7057,
      "num_input_tokens_seen": 37984,
      "step": 30
    },
    {
      "epoch": 0.056,
      "grad_norm": 3.018771171569824,
      "learning_rate": 4.995944470229302e-05,
      "loss": 2.6265,
      "num_input_tokens_seen": 44320,
      "step": 35
    },
    {
      "epoch": 0.064,
      "grad_norm": 2.792170763015747,
      "learning_rate": 4.9946644169172355e-05,
      "loss": 2.5242,
      "num_input_tokens_seen": 51136,
      "step": 40
    },
    {
      "epoch": 0.072,
      "grad_norm": 2.7025442123413086,
      "learning_rate": 4.993209279467164e-05,
      "loss": 2.4338,
      "num_input_tokens_seen": 57552,
      "step": 45
    },
    {
      "epoch": 0.08,
      "grad_norm": 2.5979084968566895,
      "learning_rate": 4.991579160005644e-05,
      "loss": 2.2604,
      "num_input_tokens_seen": 64112,
      "step": 50
    },
    {
      "epoch": 0.088,
      "grad_norm": 3.2108237743377686,
      "learning_rate": 4.9897741729400705e-05,
      "loss": 2.2676,
      "num_input_tokens_seen": 70704,
      "step": 55
    },
    {
      "epoch": 0.096,
      "grad_norm": 3.074040651321411,
      "learning_rate": 4.987794444950651e-05,
      "loss": 2.2402,
      "num_input_tokens_seen": 77600,
      "step": 60
    },
    {
      "epoch": 0.104,
      "grad_norm": 2.9350333213806152,
      "learning_rate": 4.9856401149815126e-05,
      "loss": 2.0383,
      "num_input_tokens_seen": 84512,
      "step": 65
    },
    {
      "epoch": 0.112,
      "grad_norm": 3.459714412689209,
      "learning_rate": 4.98331133423095e-05,
      "loss": 1.9534,
      "num_input_tokens_seen": 91328,
      "step": 70
    },
    {
      "epoch": 0.12,
      "grad_norm": 4.01439905166626,
      "learning_rate": 4.980808266140813e-05,
      "loss": 1.9574,
      "num_input_tokens_seen": 97616,
      "step": 75
    },
    {
      "epoch": 0.128,
      "grad_norm": 3.519948720932007,
      "learning_rate": 4.9781310863850405e-05,
      "loss": 2.0109,
      "num_input_tokens_seen": 103696,
      "step": 80
    },
    {
      "epoch": 0.136,
      "grad_norm": 3.308018684387207,
      "learning_rate": 4.975279982857324e-05,
      "loss": 1.8779,
      "num_input_tokens_seen": 110512,
      "step": 85
    },
    {
      "epoch": 0.144,
      "grad_norm": 3.8108999729156494,
      "learning_rate": 4.972255155657925e-05,
      "loss": 1.8983,
      "num_input_tokens_seen": 116464,
      "step": 90
    },
    {
      "epoch": 0.152,
      "grad_norm": 3.3505210876464844,
      "learning_rate": 4.969056817079633e-05,
      "loss": 1.6856,
      "num_input_tokens_seen": 123968,
      "step": 95
    },
    {
      "epoch": 0.16,
      "grad_norm": 3.2357165813446045,
      "learning_rate": 4.965685191592859e-05,
      "loss": 1.7751,
      "num_input_tokens_seen": 130608,
      "step": 100
    },
    {
      "epoch": 0.168,
      "grad_norm": 3.8320517539978027,
      "learning_rate": 4.96214051582989e-05,
      "loss": 1.6848,
      "num_input_tokens_seen": 136512,
      "step": 105
    },
    {
      "epoch": 0.176,
      "grad_norm": 4.119499206542969,
      "learning_rate": 4.958423038568274e-05,
      "loss": 1.7686,
      "num_input_tokens_seen": 142688,
      "step": 110
    },
    {
      "epoch": 0.184,
      "grad_norm": 3.845122814178467,
      "learning_rate": 4.9545330207133664e-05,
      "loss": 1.5997,
      "num_input_tokens_seen": 149456,
      "step": 115
    },
    {
      "epoch": 0.192,
      "grad_norm": 4.3131537437438965,
      "learning_rate": 4.9504707352800125e-05,
      "loss": 1.6921,
      "num_input_tokens_seen": 156560,
      "step": 120
    },
    {
      "epoch": 0.2,
      "grad_norm": 4.0912089347839355,
      "learning_rate": 4.946236467373392e-05,
      "loss": 1.6171,
      "num_input_tokens_seen": 162960,
      "step": 125
    },
    {
      "epoch": 0.208,
      "grad_norm": 3.5116782188415527,
      "learning_rate": 4.941830514169004e-05,
      "loss": 1.6444,
      "num_input_tokens_seen": 169856,
      "step": 130
    },
    {
      "epoch": 0.216,
      "grad_norm": 4.033109664916992,
      "learning_rate": 4.9372531848918145e-05,
      "loss": 1.6453,
      "num_input_tokens_seen": 176816,
      "step": 135
    },
    {
      "epoch": 0.224,
      "grad_norm": 3.906648874282837,
      "learning_rate": 4.9325048007945526e-05,
      "loss": 1.6189,
      "num_input_tokens_seen": 183024,
      "step": 140
    },
    {
      "epoch": 0.232,
      "grad_norm": 4.060617923736572,
      "learning_rate": 4.927585695135162e-05,
      "loss": 1.5828,
      "num_input_tokens_seen": 190176,
      "step": 145
    },
    {
      "epoch": 0.24,
      "grad_norm": 3.4880213737487793,
      "learning_rate": 4.922496213153416e-05,
      "loss": 1.549,
      "num_input_tokens_seen": 197616,
      "step": 150
    },
    {
      "epoch": 0.248,
      "grad_norm": 4.554350852966309,
      "learning_rate": 4.917236712046682e-05,
      "loss": 1.6037,
      "num_input_tokens_seen": 204080,
      "step": 155
    },
    {
      "epoch": 0.256,
      "grad_norm": 3.911822557449341,
      "learning_rate": 4.911807560944858e-05,
      "loss": 1.5218,
      "num_input_tokens_seen": 210656,
      "step": 160
    },
    {
      "epoch": 0.264,
      "grad_norm": 4.279529094696045,
      "learning_rate": 4.906209140884459e-05,
      "loss": 1.455,
      "num_input_tokens_seen": 217376,
      "step": 165
    },
    {
      "epoch": 0.272,
      "grad_norm": 5.295681953430176,
      "learning_rate": 4.9004418447818815e-05,
      "loss": 1.4421,
      "num_input_tokens_seen": 223968,
      "step": 170
    },
    {
      "epoch": 0.28,
      "grad_norm": 5.107441425323486,
      "learning_rate": 4.894506077405824e-05,
      "loss": 1.5054,
      "num_input_tokens_seen": 230240,
      "step": 175
    },
    {
      "epoch": 0.288,
      "grad_norm": 4.014343738555908,
      "learning_rate": 4.888402255348876e-05,
      "loss": 1.4786,
      "num_input_tokens_seen": 236912,
      "step": 180
    },
    {
      "epoch": 0.296,
      "grad_norm": 4.109804153442383,
      "learning_rate": 4.8821308069982867e-05,
      "loss": 1.5626,
      "num_input_tokens_seen": 243040,
      "step": 185
    },
    {
      "epoch": 0.304,
      "grad_norm": 4.771740913391113,
      "learning_rate": 4.8756921725058934e-05,
      "loss": 1.4438,
      "num_input_tokens_seen": 250000,
      "step": 190
    },
    {
      "epoch": 0.312,
      "grad_norm": 5.005781650543213,
      "learning_rate": 4.8690868037572346e-05,
      "loss": 1.3603,
      "num_input_tokens_seen": 256128,
      "step": 195
    },
    {
      "epoch": 0.32,
      "grad_norm": 5.3310136795043945,
      "learning_rate": 4.862315164339829e-05,
      "loss": 1.4616,
      "num_input_tokens_seen": 263088,
      "step": 200
    },
    {
      "epoch": 0.328,
      "grad_norm": 4.358487129211426,
      "learning_rate": 4.855377729510648e-05,
      "loss": 1.3917,
      "num_input_tokens_seen": 270096,
      "step": 205
    },
    {
      "epoch": 0.336,
      "grad_norm": 4.911283016204834,
      "learning_rate": 4.848274986162754e-05,
      "loss": 1.3146,
      "num_input_tokens_seen": 277040,
      "step": 210
    },
    {
      "epoch": 0.344,
      "grad_norm": 4.217891216278076,
      "learning_rate": 4.841007432791129e-05,
      "loss": 1.4117,
      "num_input_tokens_seen": 283776,
      "step": 215
    },
    {
      "epoch": 0.352,
      "grad_norm": 4.528128623962402,
      "learning_rate": 4.833575579457691e-05,
      "loss": 1.4422,
      "num_input_tokens_seen": 290752,
      "step": 220
    },
    {
      "epoch": 0.36,
      "grad_norm": 4.210456848144531,
      "learning_rate": 4.8259799477554965e-05,
      "loss": 1.4179,
      "num_input_tokens_seen": 297792,
      "step": 225
    },
    {
      "epoch": 0.368,
      "grad_norm": 4.201178550720215,
      "learning_rate": 4.8182210707721284e-05,
      "loss": 1.3685,
      "num_input_tokens_seen": 304464,
      "step": 230
    },
    {
      "epoch": 0.376,
      "grad_norm": 4.0117268562316895,
      "learning_rate": 4.810299493052289e-05,
      "loss": 1.3603,
      "num_input_tokens_seen": 310816,
      "step": 235
    },
    {
      "epoch": 0.384,
      "grad_norm": 4.414295196533203,
      "learning_rate": 4.802215770559577e-05,
      "loss": 1.3605,
      "num_input_tokens_seen": 317632,
      "step": 240
    },
    {
      "epoch": 0.392,
      "grad_norm": 5.311755657196045,
      "learning_rate": 4.793970470637469e-05,
      "loss": 1.3752,
      "num_input_tokens_seen": 324400,
      "step": 245
    },
    {
      "epoch": 0.4,
      "grad_norm": 5.168431758880615,
      "learning_rate": 4.7855641719695023e-05,
      "loss": 1.1998,
      "num_input_tokens_seen": 331232,
      "step": 250
    },
    {
      "epoch": 0.408,
      "grad_norm": 2.856295585632324,
      "learning_rate": 4.776997464538662e-05,
      "loss": 1.2035,
      "num_input_tokens_seen": 338624,
      "step": 255
    },
    {
      "epoch": 0.416,
      "grad_norm": 5.690861225128174,
      "learning_rate": 4.768270949585968e-05,
      "loss": 1.3473,
      "num_input_tokens_seen": 344752,
      "step": 260
    },
    {
      "epoch": 0.424,
      "grad_norm": 4.342827320098877,
      "learning_rate": 4.759385239568289e-05,
      "loss": 1.4272,
      "num_input_tokens_seen": 350608,
      "step": 265
    },
    {
      "epoch": 0.432,
      "grad_norm": 4.6672468185424805,
      "learning_rate": 4.750340958115346e-05,
      "loss": 1.3359,
      "num_input_tokens_seen": 356848,
      "step": 270
    },
    {
      "epoch": 0.44,
      "grad_norm": 5.1840081214904785,
      "learning_rate": 4.741138739985951e-05,
      "loss": 1.2609,
      "num_input_tokens_seen": 363312,
      "step": 275
    },
    {
      "epoch": 0.448,
      "grad_norm": 4.333302974700928,
      "learning_rate": 4.731779231023456e-05,
      "loss": 1.2936,
      "num_input_tokens_seen": 370512,
      "step": 280
    },
    {
      "epoch": 0.456,
      "grad_norm": 4.599977493286133,
      "learning_rate": 4.722263088110426e-05,
      "loss": 1.171,
      "num_input_tokens_seen": 377456,
      "step": 285
    },
    {
      "epoch": 0.464,
      "grad_norm": 5.032026767730713,
      "learning_rate": 4.712590979122534e-05,
      "loss": 1.3206,
      "num_input_tokens_seen": 383584,
      "step": 290
    },
    {
      "epoch": 0.472,
      "grad_norm": 4.346674919128418,
      "learning_rate": 4.702763582881692e-05,
      "loss": 1.2582,
      "num_input_tokens_seen": 391120,
      "step": 295
    },
    {
      "epoch": 0.48,
      "grad_norm": 4.144872665405273,
      "learning_rate": 4.692781589108402e-05,
      "loss": 1.2324,
      "num_input_tokens_seen": 397648,
      "step": 300
    },
    {
      "epoch": 0.488,
      "grad_norm": 5.145121097564697,
      "learning_rate": 4.682645698373357e-05,
      "loss": 1.2978,
      "num_input_tokens_seen": 403872,
      "step": 305
    },
    {
      "epoch": 0.496,
      "grad_norm": 5.409262180328369,
      "learning_rate": 4.6723566220482664e-05,
      "loss": 1.2644,
      "num_input_tokens_seen": 410192,
      "step": 310
    },
    {
      "epoch": 0.504,
      "grad_norm": 4.336021900177002,
      "learning_rate": 4.661915082255932e-05,
      "loss": 1.1224,
      "num_input_tokens_seen": 417472,
      "step": 315
    },
    {
      "epoch": 0.512,
      "grad_norm": 4.864579677581787,
      "learning_rate": 4.651321811819568e-05,
      "loss": 1.1621,
      "num_input_tokens_seen": 424880,
      "step": 320
    },
    {
      "epoch": 0.52,
      "grad_norm": 4.952954292297363,
      "learning_rate": 4.640577554211366e-05,
      "loss": 1.2365,
      "num_input_tokens_seen": 431488,
      "step": 325
    },
    {
      "epoch": 0.528,
      "grad_norm": 5.332362174987793,
      "learning_rate": 4.629683063500319e-05,
      "loss": 1.2764,
      "num_input_tokens_seen": 438064,
      "step": 330
    },
    {
      "epoch": 0.536,
      "grad_norm": 5.671170234680176,
      "learning_rate": 4.618639104299294e-05,
      "loss": 1.2609,
      "num_input_tokens_seen": 444096,
      "step": 335
    },
    {
      "epoch": 0.544,
      "grad_norm": 5.552740097045898,
      "learning_rate": 4.607446451711372e-05,
      "loss": 1.1504,
      "num_input_tokens_seen": 450368,
      "step": 340
    },
    {
      "epoch": 0.552,
      "grad_norm": 5.54111385345459,
      "learning_rate": 4.596105891275449e-05,
      "loss": 1.1347,
      "num_input_tokens_seen": 456496,
      "step": 345
    },
    {
      "epoch": 0.56,
      "grad_norm": 4.788923263549805,
      "learning_rate": 4.5846182189111035e-05,
      "loss": 1.183,
      "num_input_tokens_seen": 463408,
      "step": 350
    },
    {
      "epoch": 0.568,
      "grad_norm": 4.818716049194336,
      "learning_rate": 4.5729842408627334e-05,
      "loss": 1.1623,
      "num_input_tokens_seen": 470096,
      "step": 355
    },
    {
      "epoch": 0.576,
      "grad_norm": 4.24908971786499,
      "learning_rate": 4.561204773642974e-05,
      "loss": 1.1498,
      "num_input_tokens_seen": 476096,
      "step": 360
    },
    {
      "epoch": 0.584,
      "grad_norm": 5.277526378631592,
      "learning_rate": 4.5492806439753935e-05,
      "loss": 1.1961,
      "num_input_tokens_seen": 482832,
      "step": 365
    },
    {
      "epoch": 0.592,
      "grad_norm": 5.597021579742432,
      "learning_rate": 4.5372126887364655e-05,
      "loss": 1.1888,
      "num_input_tokens_seen": 489504,
      "step": 370
    },
    {
      "epoch": 0.6,
      "grad_norm": 4.754395008087158,
      "learning_rate": 4.5250017548968404e-05,
      "loss": 1.2106,
      "num_input_tokens_seen": 495456,
      "step": 375
    },
    {
      "epoch": 0.608,
      "grad_norm": 5.067101001739502,
      "learning_rate": 4.512648699461897e-05,
      "loss": 1.1692,
      "num_input_tokens_seen": 501472,
      "step": 380
    },
    {
      "epoch": 0.616,
      "grad_norm": 4.505124568939209,
      "learning_rate": 4.5001543894115975e-05,
      "loss": 1.1807,
      "num_input_tokens_seen": 507920,
      "step": 385
    },
    {
      "epoch": 0.624,
      "grad_norm": 4.827609062194824,
      "learning_rate": 4.487519701639641e-05,
      "loss": 1.2016,
      "num_input_tokens_seen": 514016,
      "step": 390
    },
    {
      "epoch": 0.632,
      "grad_norm": 5.530371189117432,
      "learning_rate": 4.4747455228919146e-05,
      "loss": 1.1268,
      "num_input_tokens_seen": 520672,
      "step": 395
    },
    {
      "epoch": 0.64,
      "grad_norm": 5.6696038246154785,
      "learning_rate": 4.461832749704268e-05,
      "loss": 1.1203,
      "num_input_tokens_seen": 527248,
      "step": 400
    },
    {
      "epoch": 0.648,
      "grad_norm": 4.725157737731934,
      "learning_rate": 4.4487822883395805e-05,
      "loss": 1.1945,
      "num_input_tokens_seen": 533696,
      "step": 405
    },
    {
      "epoch": 0.656,
      "grad_norm": 6.170118808746338,
      "learning_rate": 4.4355950547241645e-05,
      "loss": 1.1527,
      "num_input_tokens_seen": 539968,
      "step": 410
    },
    {
      "epoch": 0.664,
      "grad_norm": 5.390254020690918,
      "learning_rate": 4.422271974383479e-05,
      "loss": 1.0895,
      "num_input_tokens_seen": 546112,
      "step": 415
    },
    {
      "epoch": 0.672,
      "grad_norm": 5.188884735107422,
      "learning_rate": 4.4088139823771744e-05,
      "loss": 1.202,
      "num_input_tokens_seen": 552112,
      "step": 420
    },
    {
      "epoch": 0.68,
      "grad_norm": 4.36733341217041,
      "learning_rate": 4.395222023233466e-05,
      "loss": 1.104,
      "num_input_tokens_seen": 559008,
      "step": 425
    },
    {
      "epoch": 0.688,
      "grad_norm": 5.411820888519287,
      "learning_rate": 4.381497050882845e-05,
      "loss": 1.0959,
      "num_input_tokens_seen": 564784,
      "step": 430
    },
    {
      "epoch": 0.696,
      "grad_norm": 5.965424537658691,
      "learning_rate": 4.3676400285911256e-05,
      "loss": 1.158,
      "num_input_tokens_seen": 570736,
      "step": 435
    },
    {
      "epoch": 0.704,
      "grad_norm": 5.538514614105225,
      "learning_rate": 4.353651928891842e-05,
      "loss": 1.1841,
      "num_input_tokens_seen": 576992,
      "step": 440
    },
    {
      "epoch": 0.712,
      "grad_norm": 3.329176664352417,
      "learning_rate": 4.3395337335179945e-05,
      "loss": 1.0173,
      "num_input_tokens_seen": 583888,
      "step": 445
    },
    {
      "epoch": 0.72,
      "grad_norm": 5.082017421722412,
      "learning_rate": 4.325286433333142e-05,
      "loss": 1.1729,
      "num_input_tokens_seen": 589856,
      "step": 450
    },
    {
      "epoch": 0.728,
      "grad_norm": 4.892393112182617,
      "learning_rate": 4.310911028261867e-05,
      "loss": 1.105,
      "num_input_tokens_seen": 596624,
      "step": 455
    },
    {
      "epoch": 0.736,
      "grad_norm": 4.692873001098633,
      "learning_rate": 4.296408527219592e-05,
      "loss": 1.1137,
      "num_input_tokens_seen": 603184,
      "step": 460
    },
    {
      "epoch": 0.744,
      "grad_norm": 4.720251560211182,
      "learning_rate": 4.281779948041772e-05,
      "loss": 0.9949,
      "num_input_tokens_seen": 609280,
      "step": 465
    },
    {
      "epoch": 0.752,
      "grad_norm": 5.242442607879639,
      "learning_rate": 4.267026317412461e-05,
      "loss": 1.0397,
      "num_input_tokens_seen": 616160,
      "step": 470
    },
    {
      "epoch": 0.76,
      "grad_norm": 5.894191265106201,
      "learning_rate": 4.252148670792254e-05,
      "loss": 1.1702,
      "num_input_tokens_seen": 622464,
      "step": 475
    },
    {
      "epoch": 0.768,
      "grad_norm": 4.998919486999512,
      "learning_rate": 4.237148052345616e-05,
      "loss": 1.0943,
      "num_input_tokens_seen": 628832,
      "step": 480
    },
    {
      "epoch": 0.776,
      "grad_norm": 5.455171585083008,
      "learning_rate": 4.2220255148675956e-05,
      "loss": 1.0487,
      "num_input_tokens_seen": 634464,
      "step": 485
    },
    {
      "epoch": 0.784,
      "grad_norm": 5.683940410614014,
      "learning_rate": 4.206782119709942e-05,
      "loss": 1.1097,
      "num_input_tokens_seen": 641504,
      "step": 490
    },
    {
      "epoch": 0.792,
      "grad_norm": 5.903459548950195,
      "learning_rate": 4.1914189367066094e-05,
      "loss": 1.047,
      "num_input_tokens_seen": 647760,
      "step": 495
    },
    {
      "epoch": 0.8,
      "grad_norm": 5.085323333740234,
      "learning_rate": 4.1759370440986775e-05,
      "loss": 1.1237,
      "num_input_tokens_seen": 654592,
      "step": 500
    },
    {
      "epoch": 0.808,
      "grad_norm": 5.392106533050537,
      "learning_rate": 4.160337528458676e-05,
      "loss": 1.1501,
      "num_input_tokens_seen": 661360,
      "step": 505
    },
    {
      "epoch": 0.816,
      "grad_norm": 5.11650276184082,
      "learning_rate": 4.144621484614319e-05,
      "loss": 1.1284,
      "num_input_tokens_seen": 668080,
      "step": 510
    },
    {
      "epoch": 0.824,
      "grad_norm": 5.24700927734375,
      "learning_rate": 4.1287900155716784e-05,
      "loss": 1.0633,
      "num_input_tokens_seen": 673472,
      "step": 515
    },
    {
      "epoch": 0.832,
      "grad_norm": 6.16030740737915,
      "learning_rate": 4.112844232437757e-05,
      "loss": 1.0798,
      "num_input_tokens_seen": 680128,
      "step": 520
    },
    {
      "epoch": 0.84,
      "grad_norm": 4.552190780639648,
      "learning_rate": 4.0967852543425175e-05,
      "loss": 0.7796,
      "num_input_tokens_seen": 688768,
      "step": 525
    },
    {
      "epoch": 0.848,
      "grad_norm": 3.760404109954834,
      "learning_rate": 4.080614208360336e-05,
      "loss": 1.051,
      "num_input_tokens_seen": 696416,
      "step": 530
    },
    {
      "epoch": 0.856,
      "grad_norm": 6.2114386558532715,
      "learning_rate": 4.064332229430895e-05,
      "loss": 1.0647,
      "num_input_tokens_seen": 702592,
      "step": 535
    },
    {
      "epoch": 0.864,
      "grad_norm": 5.2313032150268555,
      "learning_rate": 4.047940460279537e-05,
      "loss": 1.0644,
      "num_input_tokens_seen": 709136,
      "step": 540
    },
    {
      "epoch": 0.872,
      "grad_norm": 5.142989158630371,
      "learning_rate": 4.031440051337056e-05,
      "loss": 1.0907,
      "num_input_tokens_seen": 715216,
      "step": 545
    },
    {
      "epoch": 0.88,
      "grad_norm": 4.840134620666504,
      "learning_rate": 4.0148321606589656e-05,
      "loss": 1.009,
      "num_input_tokens_seen": 721712,
      "step": 550
    },
    {
      "epoch": 0.888,
      "grad_norm": 5.025780200958252,
      "learning_rate": 3.9981179538442146e-05,
      "loss": 0.9835,
      "num_input_tokens_seen": 728112,
      "step": 555
    },
    {
      "epoch": 0.896,
      "grad_norm": 7.77101993560791,
      "learning_rate": 3.981298603953385e-05,
      "loss": 0.9688,
      "num_input_tokens_seen": 734608,
      "step": 560
    },
    {
      "epoch": 0.904,
      "grad_norm": 5.2934770584106445,
      "learning_rate": 3.964375291426361e-05,
      "loss": 0.9882,
      "num_input_tokens_seen": 740864,
      "step": 565
    },
    {
      "epoch": 0.912,
      "grad_norm": 5.84243106842041,
      "learning_rate": 3.947349203999484e-05,
      "loss": 0.9927,
      "num_input_tokens_seen": 746512,
      "step": 570
    },
    {
      "epoch": 0.92,
      "grad_norm": 4.940598964691162,
      "learning_rate": 3.930221536622191e-05,
      "loss": 1.0481,
      "num_input_tokens_seen": 753280,
      "step": 575
    },
    {
      "epoch": 0.928,
      "grad_norm": 4.853553295135498,
      "learning_rate": 3.91299349137315e-05,
      "loss": 0.9097,
      "num_input_tokens_seen": 759872,
      "step": 580
    },
    {
      "epoch": 0.936,
      "grad_norm": 4.950205326080322,
      "learning_rate": 3.895666277375892e-05,
      "loss": 0.9216,
      "num_input_tokens_seen": 766144,
      "step": 585
    },
    {
      "epoch": 0.944,
      "grad_norm": 5.953805446624756,
      "learning_rate": 3.8782411107139564e-05,
      "loss": 1.0334,
      "num_input_tokens_seen": 772448,
      "step": 590
    },
    {
      "epoch": 0.952,
      "grad_norm": 5.650152206420898,
      "learning_rate": 3.8607192143455326e-05,
      "loss": 0.9875,
      "num_input_tokens_seen": 779136,
      "step": 595
    },
    {
      "epoch": 0.96,
      "grad_norm": 4.282099723815918,
      "learning_rate": 3.843101818017637e-05,
      "loss": 1.0499,
      "num_input_tokens_seen": 786320,
      "step": 600
    },
    {
      "epoch": 0.968,
      "grad_norm": 5.714658260345459,
      "learning_rate": 3.8253901581798016e-05,
      "loss": 1.0884,
      "num_input_tokens_seen": 793136,
      "step": 605
    },
    {
      "epoch": 0.976,
      "grad_norm": 4.641557693481445,
      "learning_rate": 3.8075854778972955e-05,
      "loss": 1.0082,
      "num_input_tokens_seen": 799792,
      "step": 610
    },
    {
      "epoch": 0.984,
      "grad_norm": 5.510296821594238,
      "learning_rate": 3.789689026763883e-05,
      "loss": 0.9715,
      "num_input_tokens_seen": 806208,
      "step": 615
    },
    {
      "epoch": 0.992,
      "grad_norm": 5.0314106941223145,
      "learning_rate": 3.771702060814123e-05,
      "loss": 0.9564,
      "num_input_tokens_seen": 812480,
      "step": 620
    },
    {
      "epoch": 1.0,
      "grad_norm": 5.7377028465271,
      "learning_rate": 3.753625842435216e-05,
      "loss": 1.0693,
      "num_input_tokens_seen": 818608,
      "step": 625
    },
    {
      "epoch": 1.008,
      "grad_norm": 4.714173316955566,
      "learning_rate": 3.7354616402784035e-05,
      "loss": 0.9512,
      "num_input_tokens_seen": 825824,
      "step": 630
    },
    {
      "epoch": 1.016,
      "grad_norm": 5.2931036949157715,
      "learning_rate": 3.717210729169935e-05,
      "loss": 0.8803,
      "num_input_tokens_seen": 832832,
      "step": 635
    },
    {
      "epoch": 1.024,
      "grad_norm": 5.152520179748535,
      "learning_rate": 3.6988743900215894e-05,
      "loss": 0.9242,
      "num_input_tokens_seen": 838464,
      "step": 640
    },
    {
      "epoch": 1.032,
      "grad_norm": 4.291209697723389,
      "learning_rate": 3.680453909740782e-05,
      "loss": 0.9662,
      "num_input_tokens_seen": 845296,
      "step": 645
    },
    {
      "epoch": 1.04,
      "grad_norm": 4.950625896453857,
      "learning_rate": 3.661950581140239e-05,
      "loss": 0.8432,
      "num_input_tokens_seen": 851904,
      "step": 650
    },
    {
      "epoch": 1.048,
      "grad_norm": 5.483580589294434,
      "learning_rate": 3.643365702847272e-05,
      "loss": 0.8771,
      "num_input_tokens_seen": 859632,
      "step": 655
    },
    {
      "epoch": 1.056,
      "grad_norm": 4.987343788146973,
      "learning_rate": 3.624700579212626e-05,
      "loss": 0.8642,
      "num_input_tokens_seen": 865776,
      "step": 660
    },
    {
      "epoch": 1.064,
      "grad_norm": 5.25522518157959,
      "learning_rate": 3.6059565202189435e-05,
      "loss": 0.8519,
      "num_input_tokens_seen": 872336,
      "step": 665
    },
    {
      "epoch": 1.072,
      "grad_norm": 5.969356536865234,
      "learning_rate": 3.5871348413888204e-05,
      "loss": 1.0019,
      "num_input_tokens_seen": 878272,
      "step": 670
    },
    {
      "epoch": 1.08,
      "grad_norm": 4.862826824188232,
      "learning_rate": 3.568236863692482e-05,
      "loss": 0.9431,
      "num_input_tokens_seen": 885280,
      "step": 675
    },
    {
      "epoch": 1.088,
      "grad_norm": 6.21727991104126,
      "learning_rate": 3.5492639134550695e-05,
      "loss": 0.8563,
      "num_input_tokens_seen": 891392,
      "step": 680
    },
    {
      "epoch": 1.096,
      "grad_norm": 4.942775726318359,
      "learning_rate": 3.5302173222635524e-05,
      "loss": 0.908,
      "num_input_tokens_seen": 897456,
      "step": 685
    },
    {
      "epoch": 1.104,
      "grad_norm": 5.569796562194824,
      "learning_rate": 3.511098426873283e-05,
      "loss": 0.905,
      "num_input_tokens_seen": 904016,
      "step": 690
    },
    {
      "epoch": 1.112,
      "grad_norm": 5.717967987060547,
      "learning_rate": 3.491908569114164e-05,
      "loss": 0.8745,
      "num_input_tokens_seen": 910496,
      "step": 695
    },
    {
      "epoch": 1.12,
      "grad_norm": 5.90877628326416,
      "learning_rate": 3.4726490957964834e-05,
      "loss": 0.8864,
      "num_input_tokens_seen": 916640,
      "step": 700
    },
    {
      "epoch": 1.1280000000000001,
      "grad_norm": 5.11007833480835,
      "learning_rate": 3.453321358616393e-05,
      "loss": 0.8327,
      "num_input_tokens_seen": 922800,
      "step": 705
    },
    {
      "epoch": 1.1360000000000001,
      "grad_norm": 5.624155044555664,
      "learning_rate": 3.433926714061032e-05,
      "loss": 0.9515,
      "num_input_tokens_seen": 928896,
      "step": 710
    },
    {
      "epoch": 1.144,
      "grad_norm": 5.189602851867676,
      "learning_rate": 3.414466523313332e-05,
      "loss": 0.9265,
      "num_input_tokens_seen": 936064,
      "step": 715
    },
    {
      "epoch": 1.152,
      "grad_norm": 5.5833587646484375,
      "learning_rate": 3.394942152156482e-05,
      "loss": 0.9191,
      "num_input_tokens_seen": 942720,
      "step": 720
    },
    {
      "epoch": 1.16,
      "grad_norm": 5.759437561035156,
      "learning_rate": 3.375354970878073e-05,
      "loss": 0.9103,
      "num_input_tokens_seen": 949120,
      "step": 725
    },
    {
      "epoch": 1.168,
      "grad_norm": 5.949998378753662,
      "learning_rate": 3.355706354173928e-05,
      "loss": 0.8963,
      "num_input_tokens_seen": 955424,
      "step": 730
    },
    {
      "epoch": 1.176,
      "grad_norm": 5.6423420906066895,
      "learning_rate": 3.3359976810516164e-05,
      "loss": 0.9571,
      "num_input_tokens_seen": 961680,
      "step": 735
    },
    {
      "epoch": 1.184,
      "grad_norm": 4.929549217224121,
      "learning_rate": 3.3162303347336764e-05,
      "loss": 0.8842,
      "num_input_tokens_seen": 968832,
      "step": 740
    },
    {
      "epoch": 1.192,
      "grad_norm": 4.750211238861084,
      "learning_rate": 3.296405702560532e-05,
      "loss": 0.9295,
      "num_input_tokens_seen": 975232,
      "step": 745
    },
    {
      "epoch": 1.2,
      "grad_norm": 5.610837459564209,
      "learning_rate": 3.276525175893126e-05,
      "loss": 0.9305,
      "num_input_tokens_seen": 981552,
      "step": 750
    },
    {
      "epoch": 1.208,
      "grad_norm": 6.68215799331665,
      "learning_rate": 3.25659015001527e-05,
      "loss": 0.8855,
      "num_input_tokens_seen": 987936,
      "step": 755
    },
    {
      "epoch": 1.216,
      "grad_norm": 5.899706840515137,
      "learning_rate": 3.236602024035716e-05,
      "loss": 0.8349,
      "num_input_tokens_seen": 994480,
      "step": 760
    },
    {
      "epoch": 1.224,
      "grad_norm": 5.103348731994629,
      "learning_rate": 3.2165622007899676e-05,
      "loss": 0.888,
      "num_input_tokens_seen": 1001488,
      "step": 765
    },
    {
      "epoch": 1.232,
      "grad_norm": 5.507425308227539,
      "learning_rate": 3.196472086741815e-05,
      "loss": 0.884,
      "num_input_tokens_seen": 1007200,
      "step": 770
    },
    {
      "epoch": 1.24,
      "grad_norm": 5.364018440246582,
      "learning_rate": 3.176333091884635e-05,
      "loss": 0.8897,
      "num_input_tokens_seen": 1013744,
      "step": 775
    },
    {
      "epoch": 1.248,
      "grad_norm": 6.075136661529541,
      "learning_rate": 3.156146629642425e-05,
      "loss": 0.8845,
      "num_input_tokens_seen": 1021072,
      "step": 780
    },
    {
      "epoch": 1.256,
      "grad_norm": 5.49012565612793,
      "learning_rate": 3.135914116770609e-05,
      "loss": 0.9178,
      "num_input_tokens_seen": 1028064,
      "step": 785
    },
    {
      "epoch": 1.264,
      "grad_norm": 4.626290798187256,
      "learning_rate": 3.1156369732566006e-05,
      "loss": 0.9141,
      "num_input_tokens_seen": 1035184,
      "step": 790
    },
    {
      "epoch": 1.272,
      "grad_norm": 6.4841437339782715,
      "learning_rate": 3.0953166222201476e-05,
      "loss": 0.8258,
      "num_input_tokens_seen": 1041280,
      "step": 795
    },
    {
      "epoch": 1.28,
      "grad_norm": 6.494681358337402,
      "learning_rate": 3.074954489813449e-05,
      "loss": 0.8271,
      "num_input_tokens_seen": 1047520,
      "step": 800
    },
    {
      "epoch": 1.288,
      "grad_norm": 5.340859889984131,
      "learning_rate": 3.054552005121064e-05,
      "loss": 0.8407,
      "num_input_tokens_seen": 1053920,
      "step": 805
    },
    {
      "epoch": 1.296,
      "grad_norm": 5.7333879470825195,
      "learning_rate": 3.034110600059616e-05,
      "loss": 0.9351,
      "num_input_tokens_seen": 1060480,
      "step": 810
    },
    {
      "epoch": 1.304,
      "grad_norm": 5.079939365386963,
      "learning_rate": 3.013631709277292e-05,
      "loss": 0.8622,
      "num_input_tokens_seen": 1067312,
      "step": 815
    },
    {
      "epoch": 1.312,
      "grad_norm": 5.47499418258667,
      "learning_rate": 2.9931167700531578e-05,
      "loss": 0.7646,
      "num_input_tokens_seen": 1073856,
      "step": 820
    },
    {
      "epoch": 1.32,
      "grad_norm": 5.8759379386901855,
      "learning_rate": 2.97256722219628e-05,
      "loss": 0.8869,
      "num_input_tokens_seen": 1080144,
      "step": 825
    },
    {
      "epoch": 1.328,
      "grad_norm": 5.712034225463867,
      "learning_rate": 2.9519845079446823e-05,
      "loss": 0.8583,
      "num_input_tokens_seen": 1086800,
      "step": 830
    },
    {
      "epoch": 1.336,
      "grad_norm": 5.631429672241211,
      "learning_rate": 2.9313700718641167e-05,
      "loss": 0.9651,
      "num_input_tokens_seen": 1094192,
      "step": 835
    },
    {
      "epoch": 1.3439999999999999,
      "grad_norm": 5.5782694816589355,
      "learning_rate": 2.9107253607466832e-05,
      "loss": 0.9531,
      "num_input_tokens_seen": 1101520,
      "step": 840
    },
    {
      "epoch": 1.3519999999999999,
      "grad_norm": 5.170969009399414,
      "learning_rate": 2.8900518235092905e-05,
      "loss": 0.8686,
      "num_input_tokens_seen": 1108352,
      "step": 845
    },
    {
      "epoch": 1.3599999999999999,
      "grad_norm": 6.310078144073486,
      "learning_rate": 2.8693509110919598e-05,
      "loss": 0.8434,
      "num_input_tokens_seen": 1114512,
      "step": 850
    },
    {
      "epoch": 1.3679999999999999,
      "grad_norm": 6.386317253112793,
      "learning_rate": 2.8486240763559986e-05,
      "loss": 0.8637,
      "num_input_tokens_seen": 1120912,
      "step": 855
    },
    {
      "epoch": 1.376,
      "grad_norm": 5.666493892669678,
      "learning_rate": 2.8278727739820333e-05,
      "loss": 0.9294,
      "num_input_tokens_seen": 1127360,
      "step": 860
    },
    {
      "epoch": 1.384,
      "grad_norm": 5.921559810638428,
      "learning_rate": 2.8070984603679107e-05,
      "loss": 0.8557,
      "num_input_tokens_seen": 1134112,
      "step": 865
    },
    {
      "epoch": 1.392,
      "grad_norm": 6.2435383796691895,
      "learning_rate": 2.7863025935264875e-05,
      "loss": 0.9056,
      "num_input_tokens_seen": 1140592,
      "step": 870
    },
    {
      "epoch": 1.4,
      "grad_norm": 5.0157246589660645,
      "learning_rate": 2.7654866329833002e-05,
      "loss": 0.8511,
      "num_input_tokens_seen": 1147568,
      "step": 875
    },
    {
      "epoch": 1.408,
      "grad_norm": 5.033071994781494,
      "learning_rate": 2.744652039674129e-05,
      "loss": 0.884,
      "num_input_tokens_seen": 1154256,
      "step": 880
    },
    {
      "epoch": 1.416,
      "grad_norm": 5.921014308929443,
      "learning_rate": 2.723800275842468e-05,
      "loss": 0.9172,
      "num_input_tokens_seen": 1160864,
      "step": 885
    },
    {
      "epoch": 1.424,
      "grad_norm": 5.19156551361084,
      "learning_rate": 2.702932804936894e-05,
      "loss": 0.8077,
      "num_input_tokens_seen": 1168096,
      "step": 890
    },
    {
      "epoch": 1.432,
      "grad_norm": 5.669256210327148,
      "learning_rate": 2.6820510915083648e-05,
      "loss": 0.8873,
      "num_input_tokens_seen": 1174224,
      "step": 895
    },
    {
      "epoch": 1.44,
      "grad_norm": 7.450194358825684,
      "learning_rate": 2.661156601107424e-05,
      "loss": 0.9441,
      "num_input_tokens_seen": 1180240,
      "step": 900
    },
    {
      "epoch": 1.448,
      "grad_norm": 6.64365291595459,
      "learning_rate": 2.6402508001813496e-05,
      "loss": 0.9357,
      "num_input_tokens_seen": 1186800,
      "step": 905
    },
    {
      "epoch": 1.456,
      "grad_norm": 6.2755126953125,
      "learning_rate": 2.6193351559712292e-05,
      "loss": 0.8376,
      "num_input_tokens_seen": 1193184,
      "step": 910
    },
    {
      "epoch": 1.464,
      "grad_norm": 5.1714324951171875,
      "learning_rate": 2.5984111364089876e-05,
      "loss": 0.8699,
      "num_input_tokens_seen": 1199120,
      "step": 915
    },
    {
      "epoch": 1.472,
      "grad_norm": 5.131344318389893,
      "learning_rate": 2.5774802100143592e-05,
      "loss": 0.7691,
      "num_input_tokens_seen": 1205472,
      "step": 920
    },
    {
      "epoch": 1.48,
      "grad_norm": 5.99024772644043,
      "learning_rate": 2.5565438457918244e-05,
      "loss": 0.825,
      "num_input_tokens_seen": 1212144,
      "step": 925
    },
    {
      "epoch": 1.488,
      "grad_norm": 6.630756378173828,
      "learning_rate": 2.5356035131275096e-05,
      "loss": 0.8829,
      "num_input_tokens_seen": 1218272,
      "step": 930
    },
    {
      "epoch": 1.496,
      "grad_norm": 5.101822376251221,
      "learning_rate": 2.5146606816860597e-05,
      "loss": 0.8135,
      "num_input_tokens_seen": 1224880,
      "step": 935
    },
    {
      "epoch": 1.504,
      "grad_norm": 5.636932373046875,
      "learning_rate": 2.4937168213074907e-05,
      "loss": 0.8377,
      "num_input_tokens_seen": 1231760,
      "step": 940
    },
    {
      "epoch": 1.512,
      "grad_norm": 5.158182621002197,
      "learning_rate": 2.472773401904037e-05,
      "loss": 0.7784,
      "num_input_tokens_seen": 1239136,
      "step": 945
    },
    {
      "epoch": 1.52,
      "grad_norm": 5.6569976806640625,
      "learning_rate": 2.4518318933569785e-05,
      "loss": 0.8143,
      "num_input_tokens_seen": 1246096,
      "step": 950
    },
    {
      "epoch": 1.528,
      "grad_norm": 4.830045223236084,
      "learning_rate": 2.4308937654134893e-05,
      "loss": 0.7414,
      "num_input_tokens_seen": 1252608,
      "step": 955
    },
    {
      "epoch": 1.536,
      "grad_norm": 5.795775413513184,
      "learning_rate": 2.4099604875834795e-05,
      "loss": 0.7988,
      "num_input_tokens_seen": 1258688,
      "step": 960
    },
    {
      "epoch": 1.544,
      "grad_norm": 6.66441011428833,
      "learning_rate": 2.3890335290364595e-05,
      "loss": 0.8191,
      "num_input_tokens_seen": 1264736,
      "step": 965
    },
    {
      "epoch": 1.552,
      "grad_norm": 5.739951133728027,
      "learning_rate": 2.368114358498434e-05,
      "loss": 0.8813,
      "num_input_tokens_seen": 1271360,
      "step": 970
    },
    {
      "epoch": 1.56,
      "grad_norm": 5.536952018737793,
      "learning_rate": 2.3472044441488174e-05,
      "loss": 0.8435,
      "num_input_tokens_seen": 1276832,
      "step": 975
    },
    {
      "epoch": 1.568,
      "grad_norm": 6.003225803375244,
      "learning_rate": 2.32630525351739e-05,
      "loss": 0.876,
      "num_input_tokens_seen": 1283136,
      "step": 980
    },
    {
      "epoch": 1.576,
      "grad_norm": 6.880493640899658,
      "learning_rate": 2.3054182533813087e-05,
      "loss": 0.8883,
      "num_input_tokens_seen": 1289552,
      "step": 985
    },
    {
      "epoch": 1.584,
      "grad_norm": 5.1923675537109375,
      "learning_rate": 2.284544909662158e-05,
      "loss": 0.8653,
      "num_input_tokens_seen": 1296512,
      "step": 990
    },
    {
      "epoch": 1.592,
      "grad_norm": 6.762387275695801,
      "learning_rate": 2.2636866873230677e-05,
      "loss": 0.9483,
      "num_input_tokens_seen": 1302512,
      "step": 995
    },
    {
      "epoch": 1.6,
      "grad_norm": 4.309864521026611,
      "learning_rate": 2.2428450502658967e-05,
      "loss": 0.7619,
      "num_input_tokens_seen": 1309984,
      "step": 1000
    },
    {
      "epoch": 1.608,
      "grad_norm": 6.295886039733887,
      "learning_rate": 2.2220214612284924e-05,
      "loss": 0.8426,
      "num_input_tokens_seen": 1315872,
      "step": 1005
    },
    {
      "epoch": 1.616,
      "grad_norm": 4.982956886291504,
      "learning_rate": 2.2012173816820297e-05,
      "loss": 0.8076,
      "num_input_tokens_seen": 1322672,
      "step": 1010
    },
    {
      "epoch": 1.624,
      "grad_norm": 6.953153133392334,
      "learning_rate": 2.1804342717284415e-05,
      "loss": 0.8432,
      "num_input_tokens_seen": 1329408,
      "step": 1015
    },
    {
      "epoch": 1.6320000000000001,
      "grad_norm": 6.00987434387207,
      "learning_rate": 2.1596735899979396e-05,
      "loss": 0.8961,
      "num_input_tokens_seen": 1334880,
      "step": 1020
    },
    {
      "epoch": 1.6400000000000001,
      "grad_norm": 6.879395484924316,
      "learning_rate": 2.138936793546649e-05,
      "loss": 0.9016,
      "num_input_tokens_seen": 1341584,
      "step": 1025
    },
    {
      "epoch": 1.6480000000000001,
      "grad_norm": 5.388754844665527,
      "learning_rate": 2.1182253377543425e-05,
      "loss": 0.8779,
      "num_input_tokens_seen": 1348240,
      "step": 1030
    },
    {
      "epoch": 1.6560000000000001,
      "grad_norm": 5.502389430999756,
      "learning_rate": 2.0975406762222966e-05,
      "loss": 0.809,
      "num_input_tokens_seen": 1354480,
      "step": 1035
    },
    {
      "epoch": 1.6640000000000001,
      "grad_norm": 3.7813355922698975,
      "learning_rate": 2.076884260671276e-05,
      "loss": 0.7541,
      "num_input_tokens_seen": 1361456,
      "step": 1040
    },
    {
      "epoch": 1.6720000000000002,
      "grad_norm": 6.419345855712891,
      "learning_rate": 2.056257540839647e-05,
      "loss": 0.8278,
      "num_input_tokens_seen": 1367984,
      "step": 1045
    },
    {
      "epoch": 1.6800000000000002,
      "grad_norm": 4.702061653137207,
      "learning_rate": 2.0356619643816234e-05,
      "loss": 0.8217,
      "num_input_tokens_seen": 1374480,
      "step": 1050
    },
    {
      "epoch": 1.688,
      "grad_norm": 6.7628984451293945,
      "learning_rate": 2.0150989767656728e-05,
      "loss": 0.7578,
      "num_input_tokens_seen": 1382160,
      "step": 1055
    },
    {
      "epoch": 1.696,
      "grad_norm": 5.897040843963623,
      "learning_rate": 1.994570021173067e-05,
      "loss": 0.8402,
      "num_input_tokens_seen": 1389120,
      "step": 1060
    },
    {
      "epoch": 1.704,
      "grad_norm": 6.665550231933594,
      "learning_rate": 1.9740765383965893e-05,
      "loss": 0.8512,
      "num_input_tokens_seen": 1395712,
      "step": 1065
    },
    {
      "epoch": 1.712,
      "grad_norm": 5.517580509185791,
      "learning_rate": 1.9536199667394215e-05,
      "loss": 0.8197,
      "num_input_tokens_seen": 1402208,
      "step": 1070
    },
    {
      "epoch": 1.72,
      "grad_norm": 5.793928146362305,
      "learning_rate": 1.9332017419141962e-05,
      "loss": 0.8453,
      "num_input_tokens_seen": 1408640,
      "step": 1075
    },
    {
      "epoch": 1.728,
      "grad_norm": 5.839468479156494,
      "learning_rate": 1.9128232969422315e-05,
      "loss": 0.816,
      "num_input_tokens_seen": 1414848,
      "step": 1080
    },
    {
      "epoch": 1.736,
      "grad_norm": 6.3471479415893555,
      "learning_rate": 1.8924860620529594e-05,
      "loss": 0.8817,
      "num_input_tokens_seen": 1421456,
      "step": 1085
    },
    {
      "epoch": 1.744,
      "grad_norm": 6.201384544372559,
      "learning_rate": 1.872191464583547e-05,
      "loss": 0.7613,
      "num_input_tokens_seen": 1427872,
      "step": 1090
    },
    {
      "epoch": 1.752,
      "grad_norm": 6.5023064613342285,
      "learning_rate": 1.851940928878718e-05,
      "loss": 0.8203,
      "num_input_tokens_seen": 1434592,
      "step": 1095
    },
    {
      "epoch": 1.76,
      "grad_norm": 5.858229160308838,
      "learning_rate": 1.8317358761907942e-05,
      "loss": 0.7657,
      "num_input_tokens_seen": 1440800,
      "step": 1100
    },
    {
      "epoch": 1.768,
      "grad_norm": 5.838942527770996,
      "learning_rate": 1.811577724579938e-05,
      "loss": 0.8285,
      "num_input_tokens_seen": 1446976,
      "step": 1105
    },
    {
      "epoch": 1.776,
      "grad_norm": 5.380393981933594,
      "learning_rate": 1.7914678888146347e-05,
      "loss": 0.8087,
      "num_input_tokens_seen": 1453280,
      "step": 1110
    },
    {
      "epoch": 1.784,
      "grad_norm": 4.861492156982422,
      "learning_rate": 1.7714077802723994e-05,
      "loss": 0.8048,
      "num_input_tokens_seen": 1459600,
      "step": 1115
    },
    {
      "epoch": 1.792,
      "grad_norm": 6.196881294250488,
      "learning_rate": 1.7513988068407146e-05,
      "loss": 0.8194,
      "num_input_tokens_seen": 1466544,
      "step": 1120
    },
    {
      "epoch": 1.8,
      "grad_norm": 6.481358051300049,
      "learning_rate": 1.7314423728182283e-05,
      "loss": 0.8346,
      "num_input_tokens_seen": 1473856,
      "step": 1125
    },
    {
      "epoch": 1.808,
      "grad_norm": 5.957236289978027,
      "learning_rate": 1.7115398788161925e-05,
      "loss": 0.8259,
      "num_input_tokens_seen": 1479712,
      "step": 1130
    },
    {
      "epoch": 1.8159999999999998,
      "grad_norm": 4.8695573806762695,
      "learning_rate": 1.6916927216601593e-05,
      "loss": 0.8222,
      "num_input_tokens_seen": 1486000,
      "step": 1135
    },
    {
      "epoch": 1.8239999999999998,
      "grad_norm": 5.079285144805908,
      "learning_rate": 1.6719022942919527e-05,
      "loss": 0.8857,
      "num_input_tokens_seen": 1492720,
      "step": 1140
    },
    {
      "epoch": 1.8319999999999999,
      "grad_norm": 4.920922756195068,
      "learning_rate": 1.6521699856719062e-05,
      "loss": 0.8005,
      "num_input_tokens_seen": 1499312,
      "step": 1145
    },
    {
      "epoch": 1.8399999999999999,
      "grad_norm": 6.130289077758789,
      "learning_rate": 1.6324971806813767e-05,
      "loss": 0.855,
      "num_input_tokens_seen": 1505648,
      "step": 1150
    },
    {
      "epoch": 1.8479999999999999,
      "grad_norm": 7.0988030433654785,
      "learning_rate": 1.612885260025552e-05,
      "loss": 0.8266,
      "num_input_tokens_seen": 1513024,
      "step": 1155
    },
    {
      "epoch": 1.8559999999999999,
      "grad_norm": 6.3503594398498535,
      "learning_rate": 1.59333560013655e-05,
      "loss": 0.8106,
      "num_input_tokens_seen": 1518960,
      "step": 1160
    },
    {
      "epoch": 1.8639999999999999,
      "grad_norm": 7.253221035003662,
      "learning_rate": 1.5738495730768105e-05,
      "loss": 0.8073,
      "num_input_tokens_seen": 1525136,
      "step": 1165
    },
    {
      "epoch": 1.8719999999999999,
      "grad_norm": 6.441041469573975,
      "learning_rate": 1.5544285464428045e-05,
      "loss": 0.7954,
      "num_input_tokens_seen": 1532000,
      "step": 1170
    },
    {
      "epoch": 1.88,
      "grad_norm": 5.565557956695557,
      "learning_rate": 1.535073883269048e-05,
      "loss": 0.8736,
      "num_input_tokens_seen": 1538416,
      "step": 1175
    },
    {
      "epoch": 1.888,
      "grad_norm": 4.04221773147583,
      "learning_rate": 1.515786941932441e-05,
      "loss": 0.7713,
      "num_input_tokens_seen": 1544992,
      "step": 1180
    },
    {
      "epoch": 1.896,
      "grad_norm": 5.468989372253418,
      "learning_rate": 1.4965690760569346e-05,
      "loss": 0.7552,
      "num_input_tokens_seen": 1551360,
      "step": 1185
    },
    {
      "epoch": 1.904,
      "grad_norm": 7.29986047744751,
      "learning_rate": 1.4774216344185205e-05,
      "loss": 0.7731,
      "num_input_tokens_seen": 1557616,
      "step": 1190
    },
    {
      "epoch": 1.912,
      "grad_norm": 5.466056823730469,
      "learning_rate": 1.4583459608505801e-05,
      "loss": 0.7138,
      "num_input_tokens_seen": 1565504,
      "step": 1195
    },
    {
      "epoch": 1.92,
      "grad_norm": 6.537974834442139,
      "learning_rate": 1.4393433941495637e-05,
      "loss": 0.8888,
      "num_input_tokens_seen": 1571808,
      "step": 1200
    },
    {
      "epoch": 1.928,
      "grad_norm": 5.771127223968506,
      "learning_rate": 1.4204152679810258e-05,
      "loss": 0.7852,
      "num_input_tokens_seen": 1578320,
      "step": 1205
    },
    {
      "epoch": 1.936,
      "grad_norm": 5.833364963531494,
      "learning_rate": 1.401562910786034e-05,
      "loss": 0.8193,
      "num_input_tokens_seen": 1585888,
      "step": 1210
    },
    {
      "epoch": 1.944,
      "grad_norm": 6.242034435272217,
      "learning_rate": 1.3827876456879246e-05,
      "loss": 0.7761,
      "num_input_tokens_seen": 1591712,
      "step": 1215
    },
    {
      "epoch": 1.952,
      "grad_norm": 6.540128707885742,
      "learning_rate": 1.3640907903994454e-05,
      "loss": 0.8357,
      "num_input_tokens_seen": 1598560,
      "step": 1220
    },
    {
      "epoch": 1.96,
      "grad_norm": 6.001108646392822,
      "learning_rate": 1.3454736571302763e-05,
      "loss": 0.7871,
      "num_input_tokens_seen": 1605568,
      "step": 1225
    },
    {
      "epoch": 1.968,
      "grad_norm": 7.643476963043213,
      "learning_rate": 1.3269375524949284e-05,
      "loss": 0.8225,
      "num_input_tokens_seen": 1611984,
      "step": 1230
    },
    {
      "epoch": 1.976,
      "grad_norm": 5.949811935424805,
      "learning_rate": 1.308483777421046e-05,
      "loss": 0.7597,
      "num_input_tokens_seen": 1618528,
      "step": 1235
    },
    {
      "epoch": 1.984,
      "grad_norm": 6.608103275299072,
      "learning_rate": 1.2901136270580993e-05,
      "loss": 0.7157,
      "num_input_tokens_seen": 1625840,
      "step": 1240
    },
    {
      "epoch": 1.992,
      "grad_norm": 6.3687849044799805,
      "learning_rate": 1.2718283906864939e-05,
      "loss": 0.8312,
      "num_input_tokens_seen": 1632304,
      "step": 1245
    },
    {
      "epoch": 2.0,
      "grad_norm": 5.552835941314697,
      "learning_rate": 1.2536293516270703e-05,
      "loss": 0.8579,
      "num_input_tokens_seen": 1638512,
      "step": 1250
    },
    {
      "epoch": 2.008,
      "grad_norm": 5.662571430206299,
      "learning_rate": 1.2355177871510538e-05,
      "loss": 0.7455,
      "num_input_tokens_seen": 1644576,
      "step": 1255
    },
    {
      "epoch": 2.016,
      "grad_norm": 5.270674705505371,
      "learning_rate": 1.2174949683903941e-05,
      "loss": 0.7716,
      "num_input_tokens_seen": 1651344,
      "step": 1260
    },
    {
      "epoch": 2.024,
      "grad_norm": 5.09323787689209,
      "learning_rate": 1.1995621602485685e-05,
      "loss": 0.739,
      "num_input_tokens_seen": 1657872,
      "step": 1265
    },
    {
      "epoch": 2.032,
      "grad_norm": 4.8315749168396,
      "learning_rate": 1.1817206213117942e-05,
      "loss": 0.7423,
      "num_input_tokens_seen": 1664400,
      "step": 1270
    },
    {
      "epoch": 2.04,
      "grad_norm": 5.925480842590332,
      "learning_rate": 1.1639716037607035e-05,
      "loss": 0.7219,
      "num_input_tokens_seen": 1670576,
      "step": 1275
    },
    {
      "epoch": 2.048,
      "grad_norm": 5.623491287231445,
      "learning_rate": 1.1463163532824572e-05,
      "loss": 0.6975,
      "num_input_tokens_seen": 1676592,
      "step": 1280
    },
    {
      "epoch": 2.056,
      "grad_norm": 7.131720542907715,
      "learning_rate": 1.1287561089833248e-05,
      "loss": 0.7536,
      "num_input_tokens_seen": 1683152,
      "step": 1285
    },
    {
      "epoch": 2.064,
      "grad_norm": 5.964690685272217,
      "learning_rate": 1.1112921033017079e-05,
      "loss": 0.7828,
      "num_input_tokens_seen": 1689808,
      "step": 1290
    },
    {
      "epoch": 2.072,
      "grad_norm": 5.969583511352539,
      "learning_rate": 1.093925561921657e-05,
      "loss": 0.7186,
      "num_input_tokens_seen": 1696352,
      "step": 1295
    },
    {
      "epoch": 2.08,
      "grad_norm": 5.476303577423096,
      "learning_rate": 1.0766577036868395e-05,
      "loss": 0.689,
      "num_input_tokens_seen": 1703056,
      "step": 1300
    },
    {
      "epoch": 2.088,
      "grad_norm": 6.045198440551758,
      "learning_rate": 1.0594897405149995e-05,
      "loss": 0.7234,
      "num_input_tokens_seen": 1709040,
      "step": 1305
    },
    {
      "epoch": 2.096,
      "grad_norm": 5.853973865509033,
      "learning_rate": 1.0424228773129019e-05,
      "loss": 0.7442,
      "num_input_tokens_seen": 1715616,
      "step": 1310
    },
    {
      "epoch": 2.104,
      "grad_norm": 5.016544342041016,
      "learning_rate": 1.0254583118917698e-05,
      "loss": 0.6908,
      "num_input_tokens_seen": 1722192,
      "step": 1315
    },
    {
      "epoch": 2.112,
      "grad_norm": 5.601842880249023,
      "learning_rate": 1.0085972348832137e-05,
      "loss": 0.6929,
      "num_input_tokens_seen": 1729232,
      "step": 1320
    },
    {
      "epoch": 2.12,
      "grad_norm": 7.124587059020996,
      "learning_rate": 9.918408296556706e-06,
      "loss": 0.7966,
      "num_input_tokens_seen": 1735408,
      "step": 1325
    },
    {
      "epoch": 2.128,
      "grad_norm": 6.398017406463623,
      "learning_rate": 9.751902722313527e-06,
      "loss": 0.6879,
      "num_input_tokens_seen": 1742608,
      "step": 1330
    },
    {
      "epoch": 2.136,
      "grad_norm": 4.915652751922607,
      "learning_rate": 9.586467312037075e-06,
      "loss": 0.7194,
      "num_input_tokens_seen": 1748704,
      "step": 1335
    },
    {
      "epoch": 2.144,
      "grad_norm": 6.361541271209717,
      "learning_rate": 9.422113676554073e-06,
      "loss": 0.7192,
      "num_input_tokens_seen": 1755024,
      "step": 1340
    },
    {
      "epoch": 2.152,
      "grad_norm": 6.177706241607666,
      "learning_rate": 9.258853350768499e-06,
      "loss": 0.7604,
      "num_input_tokens_seen": 1761808,
      "step": 1345
    },
    {
      "epoch": 2.16,
      "grad_norm": 6.024658203125,
      "learning_rate": 9.096697792852155e-06,
      "loss": 0.7169,
      "num_input_tokens_seen": 1768688,
      "step": 1350
    },
    {
      "epoch": 2.168,
      "grad_norm": 5.89205265045166,
      "learning_rate": 8.93565838344039e-06,
      "loss": 0.8025,
      "num_input_tokens_seen": 1774928,
      "step": 1355
    },
    {
      "epoch": 2.176,
      "grad_norm": 6.651025772094727,
      "learning_rate": 8.775746424833427e-06,
      "loss": 0.6733,
      "num_input_tokens_seen": 1781408,
      "step": 1360
    },
    {
      "epoch": 2.184,
      "grad_norm": 6.420565605163574,
      "learning_rate": 8.616973140203098e-06,
      "loss": 0.7122,
      "num_input_tokens_seen": 1787872,
      "step": 1365
    },
    {
      "epoch": 2.192,
      "grad_norm": 6.494694709777832,
      "learning_rate": 8.459349672805198e-06,
      "loss": 0.7446,
      "num_input_tokens_seen": 1794176,
      "step": 1370
    },
    {
      "epoch": 2.2,
      "grad_norm": 5.824810028076172,
      "learning_rate": 8.302887085197341e-06,
      "loss": 0.6692,
      "num_input_tokens_seen": 1801088,
      "step": 1375
    },
    {
      "epoch": 2.208,
      "grad_norm": 5.81256103515625,
      "learning_rate": 8.147596358462662e-06,
      "loss": 0.6532,
      "num_input_tokens_seen": 1808864,
      "step": 1380
    },
    {
      "epoch": 2.216,
      "grad_norm": 5.595151424407959,
      "learning_rate": 7.993488391439025e-06,
      "loss": 0.735,
      "num_input_tokens_seen": 1815424,
      "step": 1385
    },
    {
      "epoch": 2.224,
      "grad_norm": 6.604226589202881,
      "learning_rate": 7.840573999954153e-06,
      "loss": 0.7757,
      "num_input_tokens_seen": 1821680,
      "step": 1390
    },
    {
      "epoch": 2.232,
      "grad_norm": 6.494365692138672,
      "learning_rate": 7.688863916066524e-06,
      "loss": 0.7276,
      "num_input_tokens_seen": 1828528,
      "step": 1395
    },
    {
      "epoch": 2.24,
      "grad_norm": 7.171821594238281,
      "learning_rate": 7.538368787312186e-06,
      "loss": 0.7284,
      "num_input_tokens_seen": 1834592,
      "step": 1400
    },
    {
      "epoch": 2.248,
      "grad_norm": 6.457537651062012,
      "learning_rate": 7.389099175957426e-06,
      "loss": 0.7064,
      "num_input_tokens_seen": 1840960,
      "step": 1405
    },
    {
      "epoch": 2.2560000000000002,
      "grad_norm": 5.4786906242370605,
      "learning_rate": 7.241065558257512e-06,
      "loss": 0.8045,
      "num_input_tokens_seen": 1847328,
      "step": 1410
    },
    {
      "epoch": 2.2640000000000002,
      "grad_norm": 6.286975383758545,
      "learning_rate": 7.094278323721417e-06,
      "loss": 0.7438,
      "num_input_tokens_seen": 1853504,
      "step": 1415
    },
    {
      "epoch": 2.2720000000000002,
      "grad_norm": 6.678887844085693,
      "learning_rate": 6.94874777438265e-06,
      "loss": 0.6711,
      "num_input_tokens_seen": 1860560,
      "step": 1420
    },
    {
      "epoch": 2.2800000000000002,
      "grad_norm": 6.511438846588135,
      "learning_rate": 6.804484124076249e-06,
      "loss": 0.6982,
      "num_input_tokens_seen": 1867120,
      "step": 1425
    },
    {
      "epoch": 2.288,
      "grad_norm": 5.361278057098389,
      "learning_rate": 6.661497497721872e-06,
      "loss": 0.7343,
      "num_input_tokens_seen": 1873552,
      "step": 1430
    },
    {
      "epoch": 2.296,
      "grad_norm": 5.182500839233398,
      "learning_rate": 6.519797930613289e-06,
      "loss": 0.681,
      "num_input_tokens_seen": 1880048,
      "step": 1435
    },
    {
      "epoch": 2.304,
      "grad_norm": 5.267027378082275,
      "learning_rate": 6.379395367713984e-06,
      "loss": 0.7301,
      "num_input_tokens_seen": 1886320,
      "step": 1440
    },
    {
      "epoch": 2.312,
      "grad_norm": 6.636712074279785,
      "learning_rate": 6.240299662959237e-06,
      "loss": 0.7269,
      "num_input_tokens_seen": 1892512,
      "step": 1445
    },
    {
      "epoch": 2.32,
      "grad_norm": 6.594943046569824,
      "learning_rate": 6.102520578564508e-06,
      "loss": 0.7214,
      "num_input_tokens_seen": 1899088,
      "step": 1450
    },
    {
      "epoch": 2.328,
      "grad_norm": 6.010622978210449,
      "learning_rate": 5.966067784340346e-06,
      "loss": 0.7283,
      "num_input_tokens_seen": 1906000,
      "step": 1455
    },
    {
      "epoch": 2.336,
      "grad_norm": 6.823734283447266,
      "learning_rate": 5.8309508570136284e-06,
      "loss": 0.7583,
      "num_input_tokens_seen": 1912016,
      "step": 1460
    },
    {
      "epoch": 2.344,
      "grad_norm": 5.9860310554504395,
      "learning_rate": 5.6971792795555505e-06,
      "loss": 0.7206,
      "num_input_tokens_seen": 1918688,
      "step": 1465
    },
    {
      "epoch": 2.352,
      "grad_norm": 5.339646339416504,
      "learning_rate": 5.5647624405159945e-06,
      "loss": 0.7089,
      "num_input_tokens_seen": 1925664,
      "step": 1470
    },
    {
      "epoch": 2.36,
      "grad_norm": 5.947216987609863,
      "learning_rate": 5.4337096333646365e-06,
      "loss": 0.7013,
      "num_input_tokens_seen": 1932080,
      "step": 1475
    },
    {
      "epoch": 2.368,
      "grad_norm": 5.9478912353515625,
      "learning_rate": 5.304030055838705e-06,
      "loss": 0.6382,
      "num_input_tokens_seen": 1938992,
      "step": 1480
    },
    {
      "epoch": 2.376,
      "grad_norm": 5.693755626678467,
      "learning_rate": 5.175732809297434e-06,
      "loss": 0.7408,
      "num_input_tokens_seen": 1945280,
      "step": 1485
    },
    {
      "epoch": 2.384,
      "grad_norm": 6.069830417633057,
      "learning_rate": 5.048826898083331e-06,
      "loss": 0.7486,
      "num_input_tokens_seen": 1952112,
      "step": 1490
    },
    {
      "epoch": 2.392,
      "grad_norm": 5.075676441192627,
      "learning_rate": 4.9233212288901845e-06,
      "loss": 0.7354,
      "num_input_tokens_seen": 1958560,
      "step": 1495
    },
    {
      "epoch": 2.4,
      "grad_norm": 6.984851837158203,
      "learning_rate": 4.799224610137976e-06,
      "loss": 0.7147,
      "num_input_tokens_seen": 1964480,
      "step": 1500
    },
    {
      "epoch": 2.408,
      "grad_norm": 6.267239570617676,
      "learning_rate": 4.676545751354674e-06,
      "loss": 0.7311,
      "num_input_tokens_seen": 1971200,
      "step": 1505
    },
    {
      "epoch": 2.416,
      "grad_norm": 6.076528549194336,
      "learning_rate": 4.5552932625649944e-06,
      "loss": 0.6918,
      "num_input_tokens_seen": 1977824,
      "step": 1510
    },
    {
      "epoch": 2.424,
      "grad_norm": 5.985647201538086,
      "learning_rate": 4.43547565368605e-06,
      "loss": 0.6621,
      "num_input_tokens_seen": 1985088,
      "step": 1515
    },
    {
      "epoch": 2.432,
      "grad_norm": 5.833768844604492,
      "learning_rate": 4.317101333930191e-06,
      "loss": 0.7025,
      "num_input_tokens_seen": 1991344,
      "step": 1520
    },
    {
      "epoch": 2.44,
      "grad_norm": 6.63688325881958,
      "learning_rate": 4.200178611214736e-06,
      "loss": 0.6912,
      "num_input_tokens_seen": 1997984,
      "step": 1525
    },
    {
      "epoch": 2.448,
      "grad_norm": 5.703419208526611,
      "learning_rate": 4.084715691578939e-06,
      "loss": 0.7216,
      "num_input_tokens_seen": 2005152,
      "step": 1530
    },
    {
      "epoch": 2.456,
      "grad_norm": 5.383528709411621,
      "learning_rate": 3.970720678608034e-06,
      "loss": 0.6081,
      "num_input_tokens_seen": 2011696,
      "step": 1535
    },
    {
      "epoch": 2.464,
      "grad_norm": 7.719223976135254,
      "learning_rate": 3.858201572864537e-06,
      "loss": 0.663,
      "num_input_tokens_seen": 2017904,
      "step": 1540
    },
    {
      "epoch": 2.472,
      "grad_norm": 6.381964683532715,
      "learning_rate": 3.7471662713266744e-06,
      "loss": 0.6668,
      "num_input_tokens_seen": 2023824,
      "step": 1545
    },
    {
      "epoch": 2.48,
      "grad_norm": 6.049680233001709,
      "learning_rate": 3.6376225668342284e-06,
      "loss": 0.6697,
      "num_input_tokens_seen": 2030096,
      "step": 1550
    },
    {
      "epoch": 2.488,
      "grad_norm": 5.877965927124023,
      "learning_rate": 3.529578147541532e-06,
      "loss": 0.5921,
      "num_input_tokens_seen": 2036672,
      "step": 1555
    },
    {
      "epoch": 2.496,
      "grad_norm": 5.371738433837891,
      "learning_rate": 3.4230405963779355e-06,
      "loss": 0.7071,
      "num_input_tokens_seen": 2044016,
      "step": 1560
    },
    {
      "epoch": 2.504,
      "grad_norm": 6.318173885345459,
      "learning_rate": 3.3180173905155902e-06,
      "loss": 0.6577,
      "num_input_tokens_seen": 2051216,
      "step": 1565
    },
    {
      "epoch": 2.512,
      "grad_norm": 6.8952484130859375,
      "learning_rate": 3.2145159008446807e-06,
      "loss": 0.7241,
      "num_input_tokens_seen": 2057504,
      "step": 1570
    },
    {
      "epoch": 2.52,
      "grad_norm": 5.24618673324585,
      "learning_rate": 3.1125433914561186e-06,
      "loss": 0.6901,
      "num_input_tokens_seen": 2064208,
      "step": 1575
    },
    {
      "epoch": 2.528,
      "grad_norm": 6.14894962310791,
      "learning_rate": 3.0121070191317074e-06,
      "loss": 0.7656,
      "num_input_tokens_seen": 2070528,
      "step": 1580
    },
    {
      "epoch": 2.536,
      "grad_norm": 5.479616165161133,
      "learning_rate": 2.9132138328418573e-06,
      "loss": 0.5938,
      "num_input_tokens_seen": 2078016,
      "step": 1585
    },
    {
      "epoch": 2.544,
      "grad_norm": 5.223906993865967,
      "learning_rate": 2.8158707732508727e-06,
      "loss": 0.7392,
      "num_input_tokens_seen": 2083568,
      "step": 1590
    },
    {
      "epoch": 2.552,
      "grad_norm": 6.1376519203186035,
      "learning_rate": 2.72008467222985e-06,
      "loss": 0.6447,
      "num_input_tokens_seen": 2089856,
      "step": 1595
    },
    {
      "epoch": 2.56,
      "grad_norm": 3.5924127101898193,
      "learning_rate": 2.6258622523771287e-06,
      "loss": 0.6891,
      "num_input_tokens_seen": 2097120,
      "step": 1600
    },
    {
      "epoch": 2.568,
      "grad_norm": 7.396603584289551,
      "learning_rate": 2.5332101265465648e-06,
      "loss": 0.7287,
      "num_input_tokens_seen": 2102960,
      "step": 1605
    },
    {
      "epoch": 2.576,
      "grad_norm": 5.440645217895508,
      "learning_rate": 2.442134797383344e-06,
      "loss": 0.6739,
      "num_input_tokens_seen": 2108912,
      "step": 1610
    },
    {
      "epoch": 2.584,
      "grad_norm": 5.821516036987305,
      "learning_rate": 2.3526426568676483e-06,
      "loss": 0.6185,
      "num_input_tokens_seen": 2115824,
      "step": 1615
    },
    {
      "epoch": 2.592,
      "grad_norm": 6.806615829467773,
      "learning_rate": 2.2647399858660156e-06,
      "loss": 0.7561,
      "num_input_tokens_seen": 2122048,
      "step": 1620
    },
    {
      "epoch": 2.6,
      "grad_norm": 5.234946250915527,
      "learning_rate": 2.178432953690565e-06,
      "loss": 0.7218,
      "num_input_tokens_seen": 2129312,
      "step": 1625
    },
    {
      "epoch": 2.608,
      "grad_norm": 6.068792819976807,
      "learning_rate": 2.093727617665955e-06,
      "loss": 0.6861,
      "num_input_tokens_seen": 2135488,
      "step": 1630
    },
    {
      "epoch": 2.616,
      "grad_norm": 6.392036437988281,
      "learning_rate": 2.0106299227043297e-06,
      "loss": 0.712,
      "num_input_tokens_seen": 2141072,
      "step": 1635
    },
    {
      "epoch": 2.624,
      "grad_norm": 5.817321300506592,
      "learning_rate": 1.929145700888008e-06,
      "loss": 0.7248,
      "num_input_tokens_seen": 2148144,
      "step": 1640
    },
    {
      "epoch": 2.632,
      "grad_norm": 6.509500026702881,
      "learning_rate": 1.8492806710602496e-06,
      "loss": 0.7942,
      "num_input_tokens_seen": 2155152,
      "step": 1645
    },
    {
      "epoch": 2.64,
      "grad_norm": 7.039202690124512,
      "learning_rate": 1.7710404384238155e-06,
      "loss": 0.7315,
      "num_input_tokens_seen": 2162048,
      "step": 1650
    },
    {
      "epoch": 2.648,
      "grad_norm": 6.398265361785889,
      "learning_rate": 1.6944304941476225e-06,
      "loss": 0.7056,
      "num_input_tokens_seen": 2168304,
      "step": 1655
    },
    {
      "epoch": 2.656,
      "grad_norm": 4.7756524085998535,
      "learning_rate": 1.6194562149813242e-06,
      "loss": 0.7023,
      "num_input_tokens_seen": 2175504,
      "step": 1660
    },
    {
      "epoch": 2.664,
      "grad_norm": 7.072215557098389,
      "learning_rate": 1.54612286287798e-06,
      "loss": 0.7725,
      "num_input_tokens_seen": 2182048,
      "step": 1665
    },
    {
      "epoch": 2.672,
      "grad_norm": 5.781796455383301,
      "learning_rate": 1.4744355846247255e-06,
      "loss": 0.6212,
      "num_input_tokens_seen": 2188704,
      "step": 1670
    },
    {
      "epoch": 2.68,
      "grad_norm": 6.597018718719482,
      "learning_rate": 1.4043994114815661e-06,
      "loss": 0.7874,
      "num_input_tokens_seen": 2195072,
      "step": 1675
    },
    {
      "epoch": 2.6879999999999997,
      "grad_norm": 5.874861717224121,
      "learning_rate": 1.3360192588282832e-06,
      "loss": 0.7656,
      "num_input_tokens_seen": 2202352,
      "step": 1680
    },
    {
      "epoch": 2.6959999999999997,
      "grad_norm": 4.774026393890381,
      "learning_rate": 1.2692999258194088e-06,
      "loss": 0.7421,
      "num_input_tokens_seen": 2209120,
      "step": 1685
    },
    {
      "epoch": 2.7039999999999997,
      "grad_norm": 6.998686790466309,
      "learning_rate": 1.2042460950474648e-06,
      "loss": 0.6949,
      "num_input_tokens_seen": 2215648,
      "step": 1690
    },
    {
      "epoch": 2.7119999999999997,
      "grad_norm": 5.336740493774414,
      "learning_rate": 1.1408623322142736e-06,
      "loss": 0.7663,
      "num_input_tokens_seen": 2222496,
      "step": 1695
    },
    {
      "epoch": 2.7199999999999998,
      "grad_norm": 5.413997650146484,
      "learning_rate": 1.0791530858105387e-06,
      "loss": 0.6896,
      "num_input_tokens_seen": 2229312,
      "step": 1700
    },
    {
      "epoch": 2.7279999999999998,
      "grad_norm": 5.846436977386475,
      "learning_rate": 1.0191226868036418e-06,
      "loss": 0.6872,
      "num_input_tokens_seen": 2235344,
      "step": 1705
    },
    {
      "epoch": 2.7359999999999998,
      "grad_norm": 6.408481597900391,
      "learning_rate": 9.607753483336811e-07,
      "loss": 0.721,
      "num_input_tokens_seen": 2242000,
      "step": 1710
    },
    {
      "epoch": 2.7439999999999998,
      "grad_norm": 7.655921459197998,
      "learning_rate": 9.041151654177487e-07,
      "loss": 0.7264,
      "num_input_tokens_seen": 2249040,
      "step": 1715
    },
    {
      "epoch": 2.752,
      "grad_norm": 5.736323833465576,
      "learning_rate": 8.491461146625773e-07,
      "loss": 0.7402,
      "num_input_tokens_seen": 2256000,
      "step": 1720
    },
    {
      "epoch": 2.76,
      "grad_norm": 6.008438587188721,
      "learning_rate": 7.95872053985397e-07,
      "loss": 0.7244,
      "num_input_tokens_seen": 2262512,
      "step": 1725
    },
    {
      "epoch": 2.768,
      "grad_norm": 5.405884265899658,
      "learning_rate": 7.442967223432212e-07,
      "loss": 0.7318,
      "num_input_tokens_seen": 2269744,
      "step": 1730
    },
    {
      "epoch": 2.776,
      "grad_norm": 6.758678436279297,
      "learning_rate": 6.944237394703984e-07,
      "loss": 0.7264,
      "num_input_tokens_seen": 2276112,
      "step": 1735
    },
    {
      "epoch": 2.784,
      "grad_norm": 5.777829170227051,
      "learning_rate": 6.462566056245761e-07,
      "loss": 0.7297,
      "num_input_tokens_seen": 2282912,
      "step": 1740
    },
    {
      "epoch": 2.792,
      "grad_norm": 5.518516540527344,
      "learning_rate": 5.997987013410533e-07,
      "loss": 0.648,
      "num_input_tokens_seen": 2289504,
      "step": 1745
    },
    {
      "epoch": 2.8,
      "grad_norm": 6.5723557472229,
      "learning_rate": 5.550532871955061e-07,
      "loss": 0.7339,
      "num_input_tokens_seen": 2296048,
      "step": 1750
    },
    {
      "epoch": 2.808,
      "grad_norm": 5.465515613555908,
      "learning_rate": 5.120235035751653e-07,
      "loss": 0.7185,
      "num_input_tokens_seen": 2301968,
      "step": 1755
    },
    {
      "epoch": 2.816,
      "grad_norm": 5.395256996154785,
      "learning_rate": 4.7071237045839266e-07,
      "loss": 0.7423,
      "num_input_tokens_seen": 2308400,
      "step": 1760
    },
    {
      "epoch": 2.824,
      "grad_norm": 6.215610504150391,
      "learning_rate": 4.311227872027479e-07,
      "loss": 0.6996,
      "num_input_tokens_seen": 2314784,
      "step": 1765
    },
    {
      "epoch": 2.832,
      "grad_norm": 6.787163734436035,
      "learning_rate": 3.932575323414928e-07,
      "loss": 0.6314,
      "num_input_tokens_seen": 2321376,
      "step": 1770
    },
    {
      "epoch": 2.84,
      "grad_norm": 3.62884259223938,
      "learning_rate": 3.571192633885834e-07,
      "loss": 0.6817,
      "num_input_tokens_seen": 2328576,
      "step": 1775
    },
    {
      "epoch": 2.848,
      "grad_norm": 5.678812503814697,
      "learning_rate": 3.227105166521638e-07,
      "loss": 0.671,
      "num_input_tokens_seen": 2334800,
      "step": 1780
    },
    {
      "epoch": 2.856,
      "grad_norm": 6.032224178314209,
      "learning_rate": 2.900337070565473e-07,
      "loss": 0.753,
      "num_input_tokens_seen": 2340608,
      "step": 1785
    },
    {
      "epoch": 2.864,
      "grad_norm": 6.061573505401611,
      "learning_rate": 2.5909112797274093e-07,
      "loss": 0.7097,
      "num_input_tokens_seen": 2347536,
      "step": 1790
    },
    {
      "epoch": 2.872,
      "grad_norm": 5.2953009605407715,
      "learning_rate": 2.298849510574824e-07,
      "loss": 0.7588,
      "num_input_tokens_seen": 2353888,
      "step": 1795
    },
    {
      "epoch": 2.88,
      "grad_norm": 6.579641819000244,
      "learning_rate": 2.024172261008178e-07,
      "loss": 0.68,
      "num_input_tokens_seen": 2360128,
      "step": 1800
    },
    {
      "epoch": 2.888,
      "grad_norm": 8.196795463562012,
      "learning_rate": 1.7668988088226924e-07,
      "loss": 0.7627,
      "num_input_tokens_seen": 2366624,
      "step": 1805
    },
    {
      "epoch": 2.896,
      "grad_norm": 5.645856857299805,
      "learning_rate": 1.5270472103549315e-07,
      "loss": 0.704,
      "num_input_tokens_seen": 2373424,
      "step": 1810
    },
    {
      "epoch": 2.904,
      "grad_norm": 5.634349822998047,
      "learning_rate": 1.3046342992159566e-07,
      "loss": 0.6698,
      "num_input_tokens_seen": 2379872,
      "step": 1815
    },
    {
      "epoch": 2.912,
      "grad_norm": 3.3688719272613525,
      "learning_rate": 1.0996756851096579e-07,
      "loss": 0.6955,
      "num_input_tokens_seen": 2386592,
      "step": 1820
    },
    {
      "epoch": 2.92,
      "grad_norm": 5.799430847167969,
      "learning_rate": 9.121857527372158e-08,
      "loss": 0.7248,
      "num_input_tokens_seen": 2392912,
      "step": 1825
    },
    {
      "epoch": 2.928,
      "grad_norm": 6.198770523071289,
      "learning_rate": 7.421776607876019e-08,
      "loss": 0.6922,
      "num_input_tokens_seen": 2399360,
      "step": 1830
    },
    {
      "epoch": 2.936,
      "grad_norm": 5.737548828125,
      "learning_rate": 5.896633410141239e-08,
      "loss": 0.6992,
      "num_input_tokens_seen": 2405808,
      "step": 1835
    },
    {
      "epoch": 2.944,
      "grad_norm": 6.3686652183532715,
      "learning_rate": 4.546534973968175e-08,
      "loss": 0.6978,
      "num_input_tokens_seen": 2412656,
      "step": 1840
    },
    {
      "epoch": 2.952,
      "grad_norm": 5.940762042999268,
      "learning_rate": 3.37157605391325e-08,
      "loss": 0.7273,
      "num_input_tokens_seen": 2418448,
      "step": 1845
    },
    {
      "epoch": 2.96,
      "grad_norm": 6.598605632781982,
      "learning_rate": 2.3718391126392738e-08,
      "loss": 0.676,
      "num_input_tokens_seen": 2426192,
      "step": 1850
    },
    {
      "epoch": 2.968,
      "grad_norm": 4.683955192565918,
      "learning_rate": 1.5473943151270153e-08,
      "loss": 0.6869,
      "num_input_tokens_seen": 2432480,
      "step": 1855
    },
    {
      "epoch": 2.976,
      "grad_norm": 5.59522008895874,
      "learning_rate": 8.982995237505342e-09,
      "loss": 0.6852,
      "num_input_tokens_seen": 2438576,
      "step": 1860
    },
    {
      "epoch": 2.984,
      "grad_norm": 6.2029218673706055,
      "learning_rate": 4.246002942173699e-09,
      "loss": 0.7215,
      "num_input_tokens_seen": 2445056,
      "step": 1865
    },
    {
      "epoch": 2.992,
      "grad_norm": 5.723230361938477,
      "learning_rate": 1.2632987237054528e-09,
      "loss": 0.7314,
      "num_input_tokens_seen": 2451360,
      "step": 1870
    },
    {
      "epoch": 3.0,
      "grad_norm": 5.415881156921387,
      "learning_rate": 3.509191854877969e-11,
      "loss": 0.6885,
      "num_input_tokens_seen": 2458144,
      "step": 1875
    }
  ],
  "logging_steps": 5,
  "max_steps": 1875,
  "num_input_tokens_seen": 2458144,
  "num_train_epochs": 3,
  "save_steps": 100,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.4435925671411712e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}