1{"step": 0, "loss": 0.4129122197628021, "lr": 1e-05, "grad_norm": 0.19972798228263855, "tokens": 157, "tok_per_s": 56.89517054492226, "t": 3.81}2{"step": 50, "loss": 0.38818779587745667, "lr": 0.00051, "grad_norm": 0.12346061319112778, "tokens": 8616, "tok_per_s": 166.60260199939592, "t": 52.77}3{"step": 100, "loss": 0.4031943380832672, "lr": 0.001, "grad_norm": 0.1051541119813919, "tokens": 16635, "tok_per_s": 174.07926037284497, "t": 96.61}4{"step": 150, "loss": 0.5341644287109375, "lr": 0.0009993400362330058, "grad_norm": 0.04611212760210037, "tokens": 25042, "tok_per_s": 179.646800914527, "t": 140.44}5{"step": 200, "loss": 0.4557754695415497, "lr": 0.0009973620807194618, "grad_norm": 0.0525846928358078, "tokens": 33661, "tok_per_s": 184.1627185460032, "t": 183.82}6{"step": 250, "loss": 0.4396377503871918, "lr": 0.000994071935143687, "grad_norm": 0.09422308951616287, "tokens": 42410, "tok_per_s": 187.07889539128917, "t": 227.74}7{"step": 300, "loss": 0.39374226331710815, "lr": 0.000989479250069539, "grad_norm": 0.07766297459602356, "tokens": 50771, "tok_per_s": 188.26679696976342, "t": 270.72}8{"step": 350, "loss": 0.43009835481643677, "lr": 0.0009835974966336503, "grad_norm": 0.3339686989784241, "tokens": 59133, "tok_per_s": 188.6922982284993, "t": 314.43}9{"step": 400, "loss": 0.5913891196250916, "lr": 0.0009764439270322612, "grad_norm": 0.07514749467372894, "tokens": 67299, "tok_per_s": 188.9744470056249, "t": 357.18}10{"step": 450, "loss": 0.4637806713581085, "lr": 0.0009680395239175562, "grad_norm": 0.015639174729585648, "tokens": 75575, "tok_per_s": 189.2339435576049, "t": 400.42}11{"step": 499, "val_loss_b0": 0.4621244837580907, "val_loss_b1": 0.4601967916399343, "val_loss_b2": 0.45925315793010624, "val_loss_b4": 0.459159500437925, "val_loss_b8": 0.4605305776506765, "t": 455.46}12{"step": 500, "loss": 0.4084116518497467, "lr": 0.0009584089388519307, "grad_norm": 0.0653168335556984, "tokens": 84036, "tok_per_s": 184.57829788793546, "t": 456.33}13{"step": 550, "loss": 0.2858820855617523, "lr": 0.0009475804200007103, "grad_norm": 0.10518217086791992, "tokens": 92359, "tok_per_s": 186.1065918587153, "t": 497.32}14{"step": 600, "loss": 0.3125360906124115, "lr": 0.0009355857292754152, "grad_norm": 0.039819370955228806, "tokens": 100627, "tok_per_s": 188.20453202367807, "t": 535.71}15{"step": 650, "loss": 0.5160963535308838, "lr": 0.0009224600491706008, "grad_norm": 0.06123338267207146, "tokens": 108981, "tok_per_s": 190.4373195733616, "t": 573.31}16{"step": 700, "loss": 0.39216944575309753, "lr": 0.0009082418795675396, "grad_norm": 0.06287015974521637, "tokens": 117352, "tok_per_s": 192.2833420474039, "t": 611.35}17{"step": 750, "loss": 0.5033310651779175, "lr": 0.0008929729248074364, "grad_norm": 0.06266418844461441, "tokens": 125499, "tok_per_s": 193.58539670213707, "t": 649.33}18{"step": 800, "loss": 0.38110843300819397, "lr": 0.0008766979713654089, "grad_norm": 0.06998791545629501, "tokens": 133880, "tok_per_s": 195.15907395430273, "t": 687.05}19{"step": 850, "loss": 0.47054141759872437, "lr": 0.0008594647564840406, "grad_norm": 0.07112304866313934, "tokens": 142063, "tok_per_s": 196.54094875220125, "t": 723.86}20{"step": 900, "loss": 0.3469170928001404, "lr": 0.0008413238281518224, "grad_norm": 0.04494454339146614, "tokens": 150436, "tok_per_s": 197.6396715955457, "t": 762.21}21{"step": 950, "loss": 0.4069647789001465, "lr": 0.0008223283968371944, "grad_norm": 0.056489791721105576, "tokens": 158527, "tok_per_s": 198.31702733856687, "t": 800.41}22{"step": 999, "val_loss_b0": 0.4621244837580907, "val_loss_b1": 0.45899866742415635, "val_loss_b2": 0.45708128100489676, "val_loss_b4": 0.45597139022924915, "val_loss_b8": 0.45995436095439374, "t": 848.89}23{"step": 1000, "loss": 0.3964732587337494, "lr": 0.000802534179413072, "grad_norm": 0.07305459678173065, "tokens": 166917, "tok_per_s": 196.67266273469204, "t": 849.75}24{"step": 1050, "loss": 0.44622287154197693, "lr": 0.0007819992357296598, "grad_norm": 0.10645992308855057, "tokens": 175091, "tok_per_s": 197.44634244493733, "t": 887.82}25{"step": 1100, "loss": 0.35587042570114136, "lr": 0.0007607837983149056, "grad_norm": 0.07973819971084595, "tokens": 183159, "tok_per_s": 198.8694870671697, "t": 922.05}26{"step": 1150, "loss": 0.29083487391471863, "lr": 0.0007389500957021191, "grad_norm": 0.060245320200920105, "tokens": 191581, "tok_per_s": 203.29678513698946, "t": 943.42}27{"step": 1200, "loss": 0.5135460495948792, "lr": 0.0007165621699029615, "grad_norm": 0.10896871984004974, "tokens": 200185, "tok_per_s": 207.72981439946798, "t": 964.72}28{"step": 1250, "loss": 0.5982564091682434, "lr": 0.000693685688561191, "grad_norm": 0.14849816262722015, "tokens": 208401, "tok_per_s": 211.5977455469407, "t": 985.94}29{"step": 1300, "loss": 0.430452436208725, "lr": 0.0006703877523381495, "grad_norm": 0.08989538997411728, "tokens": 216624, "tok_per_s": 215.31336238084515, "t": 1007.13}30{"step": 1350, "loss": 0.5400709509849548, "lr": 0.0006467366980949609, "grad_norm": 0.1832449585199356, "tokens": 225093, "tok_per_s": 219.16635173410725, "t": 1028.09}31{"step": 1400, "loss": 0.40109074115753174, "lr": 0.0006228018984487442, "grad_norm": 0.13374966382980347, "tokens": 233304, "tok_per_s": 222.59127258671847, "t": 1049.17}32{"step": 1450, "loss": 0.4143458604812622, "lr": 0.0005986535582907738, "grad_norm": 0.25667375326156616, "tokens": 241497, "tok_per_s": 225.87281227067447, "t": 1070.22}33{"step": 1499, "val_loss_b0": 0.4621244837580907, "val_loss_b1": 0.45823711171533277, "val_loss_b2": 0.4556602564511363, "val_loss_b4": 0.4528994430985693, "val_loss_b8": 0.4528715811438488, "t": 1095.32}34{"step": 1500, "loss": 0.44393235445022583, "lr": 0.000574362508863438, "grad_norm": 0.09799527376890182, "tokens": 249582, "tok_per_s": 228.0161183874455, "t": 1095.62}35{"step": 1550, "loss": 0.32693710923194885, "lr": 0.00055, "grad_norm": 0.08381565660238266, "tokens": 257750, "tok_per_s": 232.9872087440326, "t": 1107.33}36{"step": 1600, "loss": 0.4304475784301758, "lr": 0.0005256374911365621, "grad_norm": 0.1895965188741684, "tokens": 266068, "tok_per_s": 238.0215705981498, "t": 1118.88}37{"step": 1650, "loss": 0.4676089584827423, "lr": 0.0005013464417092262, "grad_norm": 0.19817127287387848, "tokens": 274463, "tok_per_s": 242.95748568095814, "t": 1130.72}38{"step": 1700, "loss": 0.309781014919281, "lr": 0.00047719810155125587, "grad_norm": 0.09844126552343369, "tokens": 282527, "tok_per_s": 247.5615834480285, "t": 1142.28}39{"step": 1750, "loss": 0.45513054728507996, "lr": 0.0004532633019050392, "grad_norm": 0.13619910180568695, "tokens": 290793, "tok_per_s": 252.20084187250063, "t": 1154.07}40{"step": 1800, "loss": 0.41488316655158997, "lr": 0.0004296122476618507, "grad_norm": 0.10337869822978973, "tokens": 299230, "tok_per_s": 256.9561458602445, "t": 1165.56}41{"step": 1850, "loss": 0.38761013746261597, "lr": 0.00040631431143880916, "grad_norm": 0.14403648674488068, "tokens": 307566, "tok_per_s": 261.5142966292266, "t": 1177.14}42{"step": 1900, "loss": 0.4000546932220459, "lr": 0.00038343783009703845, "grad_norm": 0.13354885578155518, "tokens": 315780, "tok_per_s": 265.85122036099784, "t": 1188.85}43{"step": 1950, "loss": 0.4173225164413452, "lr": 0.000361049904297881, "grad_norm": 0.1400756537914276, "tokens": 324229, "tok_per_s": 270.29490318392743, "t": 1200.58}44{"step": 1999, "val_loss_b0": 0.4621244837580907, "val_loss_b1": 0.45572436273009864, "val_loss_b2": 0.45319900023872595, "val_loss_b4": 0.45178519710053827, "val_loss_b8": 0.4583407176898112, "t": 1214.86}45{"step": 2000, "loss": 0.46050792932510376, "lr": 0.0003392162016850945, "grad_norm": 0.12699079513549805, "tokens": 332550, "tok_per_s": 273.9170675324794, "t": 1215.1}46{"step": 2050, "loss": 0.5125675201416016, "lr": 0.0003180007642703402, "grad_norm": 0.17297698557376862, "tokens": 340880, "tok_per_s": 278.10336623606287, "t": 1226.78}47{"step": 2100, "loss": 0.3059358298778534, "lr": 0.000297465820586928, "grad_norm": 0.1602209508419037, "tokens": 349421, "tok_per_s": 282.3732950832643, "t": 1238.49}48{"step": 2150, "loss": 0.40952491760253906, "lr": 0.0002776716031628058, "grad_norm": 0.1729360967874527, "tokens": 357735, "tok_per_s": 286.3925053712437, "t": 1250.15}49{"step": 2200, "loss": 0.45865118503570557, "lr": 0.00025867617184817756, "grad_norm": 0.157239630818367, "tokens": 366168, "tok_per_s": 290.387960596787, "t": 1262.01}50{"step": 2250, "loss": 0.38147762417793274, "lr": 0.00024053524351595949, "grad_norm": 0.20153163373470306, "tokens": 374489, "tok_per_s": 294.27452244356635, "t": 1273.63}51{"step": 2300, "loss": 0.31697332859039307, "lr": 0.0002233020286345913, "grad_norm": 0.10010411590337753, "tokens": 382832, "tok_per_s": 298.1013449240844, "t": 1285.28}52{"step": 2350, "loss": 0.5400072336196899, "lr": 0.00020702707519256364, "grad_norm": 0.1387370526790619, "tokens": 391097, "tok_per_s": 301.81386734538813, "t": 1296.87}53{"step": 2400, "loss": 0.28169769048690796, "lr": 0.00019175812043246033, "grad_norm": 0.17468968033790588, "tokens": 399648, "tok_per_s": 305.6659463526755, "t": 1308.51}54{"step": 2450, "loss": 0.36093592643737793, "lr": 0.00017753995082939931, "grad_norm": 0.12739917635917664, "tokens": 407914, "tok_per_s": 309.19108045260367, "t": 1320.34}55{"step": 2499, "val_loss_b0": 0.4621244837580907, "val_loss_b1": 0.45343796592325103, "val_loss_b2": 0.45010219835306753, "val_loss_b4": 0.4473517019799215, "val_loss_b8": 0.45191617056676003, "t": 1334.59}56{"step": 2500, "loss": 0.257260799407959, "lr": 0.00016441427072458494, "grad_norm": 0.1651025414466858, "tokens": 416524, "tok_per_s": 312.2891441534228, "t": 1334.82}57{"step": 2550, "loss": 0.3216046988964081, "lr": 0.0001524195799992897, "grad_norm": 0.09912362694740295, "tokens": 424563, "tok_per_s": 315.5798639893656, "t": 1346.39}58{"step": 2600, "loss": 0.2849016487598419, "lr": 0.0001415910611480694, "grad_norm": 0.16812288761138916, "tokens": 432991, "tok_per_s": 319.1076161491276, "t": 1357.92}59{"step": 2650, "loss": 0.39032068848609924, "lr": 0.00013196047608244387, "grad_norm": 0.16742543876171112, "tokens": 441165, "tok_per_s": 322.3642078125968, "t": 1369.57}60{"step": 2700, "loss": 0.34877675771713257, "lr": 0.00012355607296773896, "grad_norm": 0.07702798396348953, "tokens": 449578, "tok_per_s": 325.7434251377033, "t": 1381.2}61{"step": 2750, "loss": 0.37251514196395874, "lr": 0.00011640250336634969, "grad_norm": 0.16345788538455963, "tokens": 457759, "tok_per_s": 328.8850480125528, "t": 1392.9}62{"step": 2800, "loss": 0.3682067096233368, "lr": 0.00011052074993046101, "grad_norm": 0.12354344129562378, "tokens": 465905, "tok_per_s": 332.00457726291967, "t": 1404.35}63{"step": 2850, "loss": 0.40392011404037476, "lr": 0.00010592806485631325, "grad_norm": 0.2048843801021576, "tokens": 474059, "tok_per_s": 335.0476177780821, "t": 1415.94}64{"step": 2900, "loss": 0.5353833436965942, "lr": 0.00010263791928053818, "grad_norm": 0.10990317910909653, "tokens": 482466, "tok_per_s": 338.2304145965071, "t": 1427.49}65{"step": 2950, "loss": 0.3792978823184967, "lr": 0.0001006599637669943, "grad_norm": 0.20534870028495789, "tokens": 490971, "tok_per_s": 341.41793606416, "t": 1439.08}66{"step": 2999, "loss": 0.359876811504364, "lr": 0.0001000002640500325, "grad_norm": 0.1537003368139267, "tokens": 499017, "tok_per_s": 344.28147910427066, "t": 1450.49}67{"step": 2999, "val_loss_b0": 0.4621244837580907, "val_loss_b1": 0.45207495146516863, "val_loss_b2": 0.44882025420225097, "val_loss_b4": 0.44789463924973866, "val_loss_b8": 0.45654735499382487, "t": 1453.31}68{"step": 2999, "checkpoint_sha256": "a359b2a0ce1fd5fc872d7c87fcf343373e4a9151aa6652a8b4257dc82dc909d5", "t": 1453.36}69