{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9212022745735174,
  "eval_steps": 500,
  "global_step": 1134,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.004061738424045491,
      "grad_norm": 14.399637222290039,
      "learning_rate": 2.162496240601504e-06,
      "loss": 0.3612,
      "mean_token_accuracy": 0.89127117395401,
      "num_tokens": 992011.0,
      "step": 5
    },
    {
      "epoch": 0.008123476848090982,
      "grad_norm": 2.374295473098755,
      "learning_rate": 4.865616541353383e-06,
      "loss": 0.1828,
      "mean_token_accuracy": 0.9513086259365082,
      "num_tokens": 1994871.0,
      "step": 10
    },
    {
      "epoch": 0.012185215272136474,
      "grad_norm": 1.1915149688720703,
      "learning_rate": 7.568736842105265e-06,
      "loss": 0.143,
      "mean_token_accuracy": 0.9534980177879333,
      "num_tokens": 3000078.0,
      "step": 15
    },
    {
      "epoch": 0.016246953696181964,
      "grad_norm": 1.179621696472168,
      "learning_rate": 1.0271857142857143e-05,
      "loss": 0.1383,
      "mean_token_accuracy": 0.9597418487071991,
      "num_tokens": 3998846.0,
      "step": 20
    },
    {
      "epoch": 0.020308692120227456,
      "grad_norm": 0.770680844783783,
      "learning_rate": 1.2974977443609024e-05,
      "loss": 0.1168,
      "mean_token_accuracy": 0.9617276728153229,
      "num_tokens": 5004373.0,
      "step": 25
    },
    {
      "epoch": 0.024370430544272948,
      "grad_norm": 0.9632413983345032,
      "learning_rate": 1.5678097744360906e-05,
      "loss": 0.1015,
      "mean_token_accuracy": 0.9644747316837311,
      "num_tokens": 6004570.0,
      "step": 30
    },
    {
      "epoch": 0.02843216896831844,
      "grad_norm": 0.9619526267051697,
      "learning_rate": 1.8381218045112784e-05,
      "loss": 0.1268,
      "mean_token_accuracy": 0.9606809973716736,
      "num_tokens": 6995433.0,
      "step": 35
    },
    {
      "epoch": 0.03249390739236393,
      "grad_norm": 0.7709339261054993,
      "learning_rate": 1.8921449781511438e-05,
      "loss": 0.0921,
      "mean_token_accuracy": 0.970301216840744,
      "num_tokens": 7997481.0,
      "step": 40
    },
    {
      "epoch": 0.036555645816409424,
      "grad_norm": 1.047613263130188,
      "learning_rate": 1.8919856040623315e-05,
      "loss": 0.0981,
      "mean_token_accuracy": 0.9701705276966095,
      "num_tokens": 8999114.0,
      "step": 45
    },
    {
      "epoch": 0.04061738424045491,
      "grad_norm": 0.9218872785568237,
      "learning_rate": 1.8917036637521706e-05,
      "loss": 0.1076,
      "mean_token_accuracy": 0.9680880308151245,
      "num_tokens": 10002492.0,
      "step": 50
    },
    {
      "epoch": 0.04467912266450041,
      "grad_norm": 0.8852908611297607,
      "learning_rate": 1.8912992059347805e-05,
      "loss": 0.0932,
      "mean_token_accuracy": 0.9707131385803223,
      "num_tokens": 10998705.0,
      "step": 55
    },
    {
      "epoch": 0.048740861088545896,
      "grad_norm": 0.8299311995506287,
      "learning_rate": 1.8907723004930628e-05,
      "loss": 0.0907,
      "mean_token_accuracy": 0.9733425199985504,
      "num_tokens": 12005384.0,
      "step": 60
    },
    {
      "epoch": 0.05280259951259139,
      "grad_norm": 0.4758163392543793,
      "learning_rate": 1.8901230384666237e-05,
      "loss": 0.0757,
      "mean_token_accuracy": 0.9753991425037384,
      "num_tokens": 12998571.0,
      "step": 65
    },
    {
      "epoch": 0.05686433793663688,
      "grad_norm": 0.9689606428146362,
      "learning_rate": 1.8893515320360467e-05,
      "loss": 0.108,
      "mean_token_accuracy": 0.9683278620243072,
      "num_tokens": 13992926.0,
      "step": 70
    },
    {
      "epoch": 0.060926076360682375,
      "grad_norm": 0.5853621363639832,
      "learning_rate": 1.8884579145035085e-05,
      "loss": 0.0947,
      "mean_token_accuracy": 0.9709537088871002,
      "num_tokens": 14981397.0,
      "step": 75
    },
    {
      "epoch": 0.06498781478472786,
      "grad_norm": 0.7402133941650391,
      "learning_rate": 1.8874423402697467e-05,
      "loss": 0.0857,
      "mean_token_accuracy": 0.9716212034225464,
      "num_tokens": 15983775.0,
      "step": 80
    },
    {
      "epoch": 0.06904955320877336,
      "grad_norm": 0.7638604044914246,
      "learning_rate": 1.886304984807382e-05,
      "loss": 0.0944,
      "mean_token_accuracy": 0.9713895976543426,
      "num_tokens": 16995530.0,
      "step": 85
    },
    {
      "epoch": 0.07311129163281885,
      "grad_norm": 0.6803886890411377,
      "learning_rate": 1.885046044630602e-05,
      "loss": 0.0838,
      "mean_token_accuracy": 0.9755734443664551,
      "num_tokens": 18015783.0,
      "step": 90
    },
    {
      "epoch": 0.07717303005686434,
      "grad_norm": 0.5754309296607971,
      "learning_rate": 1.883665737261204e-05,
      "loss": 0.0952,
      "mean_token_accuracy": 0.9705892086029053,
      "num_tokens": 19022877.0,
      "step": 95
    },
    {
      "epoch": 0.08123476848090982,
      "grad_norm": 1.1319469213485718,
      "learning_rate": 1.8821643011910127e-05,
      "loss": 0.0884,
      "mean_token_accuracy": 0.9719919919967651,
      "num_tokens": 20023648.0,
      "step": 100
    },
    {
      "epoch": 0.08529650690495533,
      "grad_norm": 0.6185805201530457,
      "learning_rate": 1.8805419958406738e-05,
      "loss": 0.0992,
      "mean_token_accuracy": 0.970376706123352,
      "num_tokens": 21012205.0,
      "step": 105
    },
    {
      "epoch": 0.08935824532900082,
      "grad_norm": 0.6169642210006714,
      "learning_rate": 1.8787991015148292e-05,
      "loss": 0.0788,
      "mean_token_accuracy": 0.9719614624977112,
      "num_tokens": 22030951.0,
      "step": 110
    },
    {
      "epoch": 0.0934199837530463,
      "grad_norm": 0.9917574524879456,
      "learning_rate": 1.8769359193536876e-05,
      "loss": 0.0934,
      "mean_token_accuracy": 0.9725947380065918,
      "num_tokens": 23020658.0,
      "step": 115
    },
    {
      "epoch": 0.09748172217709179,
      "grad_norm": 0.6746852397918701,
      "learning_rate": 1.8749527712809912e-05,
      "loss": 0.0944,
      "mean_token_accuracy": 0.971796190738678,
      "num_tokens": 24025201.0,
      "step": 120
    },
    {
      "epoch": 0.10154346060113728,
      "grad_norm": 0.5181170701980591,
      "learning_rate": 1.8728499999483935e-05,
      "loss": 0.0771,
      "mean_token_accuracy": 0.9764718532562255,
      "num_tokens": 25031985.0,
      "step": 125
    },
    {
      "epoch": 0.10560519902518278,
      "grad_norm": 0.7674288153648376,
      "learning_rate": 1.8706279686762564e-05,
      "loss": 0.0716,
      "mean_token_accuracy": 0.9769453167915344,
      "num_tokens": 26038417.0,
      "step": 130
    },
    {
      "epoch": 0.10966693744922827,
      "grad_norm": 0.6015259027481079,
      "learning_rate": 1.8682870613908743e-05,
      "loss": 0.0818,
      "mean_token_accuracy": 0.9737950205802918,
      "num_tokens": 27029023.0,
      "step": 135
    },
    {
      "epoch": 0.11372867587327376,
      "grad_norm": 0.6201958656311035,
      "learning_rate": 1.8658276825581385e-05,
      "loss": 0.0799,
      "mean_token_accuracy": 0.9736847519874573,
      "num_tokens": 28027458.0,
      "step": 140
    },
    {
      "epoch": 0.11779041429731925,
      "grad_norm": 0.6044195890426636,
      "learning_rate": 1.8632502571136532e-05,
      "loss": 0.0748,
      "mean_token_accuracy": 0.9762773215770721,
      "num_tokens": 29044517.0,
      "step": 145
    },
    {
      "epoch": 0.12185215272136475,
      "grad_norm": 0.483508825302124,
      "learning_rate": 1.8605552303893147e-05,
      "loss": 0.0687,
      "mean_token_accuracy": 0.977509468793869,
      "num_tokens": 30042395.0,
      "step": 150
    },
    {
      "epoch": 0.12591389114541024,
      "grad_norm": 0.7777503728866577,
      "learning_rate": 1.8577430680363655e-05,
      "loss": 0.0714,
      "mean_token_accuracy": 0.9771440207958222,
      "num_tokens": 31043159.0,
      "step": 155
    },
    {
      "epoch": 0.12997562956945571,
      "grad_norm": 0.45895782113075256,
      "learning_rate": 1.854814255944939e-05,
      "loss": 0.0816,
      "mean_token_accuracy": 0.9754331827163696,
      "num_tokens": 32037901.0,
      "step": 160
    },
    {
      "epoch": 0.13403736799350122,
      "grad_norm": 0.6511008143424988,
      "learning_rate": 1.851769300160106e-05,
      "loss": 0.0961,
      "mean_token_accuracy": 0.9698115229606629,
      "num_tokens": 33042211.0,
      "step": 165
    },
    {
      "epoch": 0.13809910641754672,
      "grad_norm": 0.5093383193016052,
      "learning_rate": 1.8486087267944396e-05,
      "loss": 0.0732,
      "mean_token_accuracy": 0.9775528907775879,
      "num_tokens": 34038769.0,
      "step": 170
    },
    {
      "epoch": 0.1421608448415922,
      "grad_norm": 0.40691277384757996,
      "learning_rate": 1.845333081937114e-05,
      "loss": 0.0768,
      "mean_token_accuracy": 0.9761992156505584,
      "num_tokens": 35041299.0,
      "step": 175
    },
    {
      "epoch": 0.1462225832656377,
      "grad_norm": 0.48236286640167236,
      "learning_rate": 1.8419429315595465e-05,
      "loss": 0.082,
      "mean_token_accuracy": 0.9754073619842529,
      "num_tokens": 36036562.0,
      "step": 180
    },
    {
      "epoch": 0.1502843216896832,
      "grad_norm": 0.38860923051834106,
      "learning_rate": 1.8384388614176123e-05,
      "loss": 0.0709,
      "mean_token_accuracy": 0.9771956145763397,
      "num_tokens": 37037342.0,
      "step": 185
    },
    {
      "epoch": 0.15434606011372867,
      "grad_norm": 0.5228715538978577,
      "learning_rate": 1.8348214769504347e-05,
      "loss": 0.0687,
      "mean_token_accuracy": 0.979205197095871,
      "num_tokens": 38023814.0,
      "step": 190
    },
    {
      "epoch": 0.15840779853777417,
      "grad_norm": 0.8611324429512024,
      "learning_rate": 1.8310914031757748e-05,
      "loss": 0.0629,
      "mean_token_accuracy": 0.9794583082199096,
      "num_tokens": 39042028.0,
      "step": 195
    },
    {
      "epoch": 0.16246953696181965,
      "grad_norm": 0.7325474619865417,
      "learning_rate": 1.8272492845820436e-05,
      "loss": 0.0676,
      "mean_token_accuracy": 0.9778383314609528,
      "num_tokens": 40040090.0,
      "step": 200
    },
    {
      "epoch": 0.16653127538586515,
      "grad_norm": 0.5119726657867432,
      "learning_rate": 1.823295785016943e-05,
      "loss": 0.0593,
      "mean_token_accuracy": 0.979541826248169,
      "num_tokens": 41024687.0,
      "step": 205
    },
    {
      "epoch": 0.17059301380991065,
      "grad_norm": 0.5183334946632385,
      "learning_rate": 1.8192315875727662e-05,
      "loss": 0.0641,
      "mean_token_accuracy": 0.9781121492385865,
      "num_tokens": 42024132.0,
      "step": 210
    },
    {
      "epoch": 0.17465475223395613,
      "grad_norm": 0.4862414002418518,
      "learning_rate": 1.8150573944683715e-05,
      "loss": 0.0653,
      "mean_token_accuracy": 0.9790656626224518,
      "num_tokens": 43021404.0,
      "step": 215
    },
    {
      "epoch": 0.17871649065800163,
      "grad_norm": 0.9441388249397278,
      "learning_rate": 1.8107739269278526e-05,
      "loss": 0.0635,
      "mean_token_accuracy": 0.9787464737892151,
      "num_tokens": 44031923.0,
      "step": 220
    },
    {
      "epoch": 0.1827782290820471,
      "grad_norm": 0.4346712827682495,
      "learning_rate": 1.8063819250559234e-05,
      "loss": 0.0808,
      "mean_token_accuracy": 0.9749934852123261,
      "num_tokens": 45044154.0,
      "step": 225
    },
    {
      "epoch": 0.1868399675060926,
      "grad_norm": 0.5793266296386719,
      "learning_rate": 1.8018821477100423e-05,
      "loss": 0.0912,
      "mean_token_accuracy": 0.9750064134597778,
      "num_tokens": 46043648.0,
      "step": 230
    },
    {
      "epoch": 0.1909017059301381,
      "grad_norm": 0.408683717250824,
      "learning_rate": 1.7972753723692942e-05,
      "loss": 0.0713,
      "mean_token_accuracy": 0.9780198276042938,
      "num_tokens": 47040734.0,
      "step": 235
    },
    {
      "epoch": 0.19496344435418358,
      "grad_norm": 0.41989192366600037,
      "learning_rate": 1.792562395000059e-05,
      "loss": 0.08,
      "mean_token_accuracy": 0.9751070022583008,
      "num_tokens": 48030967.0,
      "step": 240
    },
    {
      "epoch": 0.1990251827782291,
      "grad_norm": 0.3775230944156647,
      "learning_rate": 1.787744029918479e-05,
      "loss": 0.0709,
      "mean_token_accuracy": 0.9799970209598541,
      "num_tokens": 49028792.0,
      "step": 245
    },
    {
      "epoch": 0.20308692120227456,
      "grad_norm": 0.74113929271698,
      "learning_rate": 1.782821109649764e-05,
      "loss": 0.0792,
      "mean_token_accuracy": 0.9767383992671966,
      "num_tokens": 50035856.0,
      "step": 250
    },
    {
      "epoch": 0.20714865962632006,
      "grad_norm": 0.5142791867256165,
      "learning_rate": 1.777794484784344e-05,
      "loss": 0.0704,
      "mean_token_accuracy": 0.9773755609989166,
      "num_tokens": 51037012.0,
      "step": 255
    },
    {
      "epoch": 0.21121039805036557,
      "grad_norm": 0.536046028137207,
      "learning_rate": 1.7726650238309038e-05,
      "loss": 0.0718,
      "mean_token_accuracy": 0.9771702706813812,
      "num_tokens": 52037118.0,
      "step": 260
    },
    {
      "epoch": 0.21527213647441104,
      "grad_norm": 0.7233165502548218,
      "learning_rate": 1.7674336130663197e-05,
      "loss": 0.0813,
      "mean_token_accuracy": 0.9742062032222748,
      "num_tokens": 53033486.0,
      "step": 265
    },
    {
      "epoch": 0.21933387489845654,
      "grad_norm": 0.44622108340263367,
      "learning_rate": 1.7621011563825275e-05,
      "loss": 0.0733,
      "mean_token_accuracy": 0.9779915809631348,
      "num_tokens": 54035992.0,
      "step": 270
    },
    {
      "epoch": 0.22339561332250202,
      "grad_norm": 0.5961236357688904,
      "learning_rate": 1.7566685751303474e-05,
      "loss": 0.0666,
      "mean_token_accuracy": 0.9787745952606202,
      "num_tokens": 55043180.0,
      "step": 275
    },
    {
      "epoch": 0.22745735174654752,
      "grad_norm": 0.47312936186790466,
      "learning_rate": 1.7511368079602895e-05,
      "loss": 0.0675,
      "mean_token_accuracy": 0.978559947013855,
      "num_tokens": 56047482.0,
      "step": 280
    },
    {
      "epoch": 0.23151909017059302,
      "grad_norm": 0.5586518049240112,
      "learning_rate": 1.7455068106603746e-05,
      "loss": 0.0665,
      "mean_token_accuracy": 0.9779742896556854,
      "num_tokens": 57050552.0,
      "step": 285
    },
    {
      "epoch": 0.2355808285946385,
      "grad_norm": 0.6518518328666687,
      "learning_rate": 1.7397795559909887e-05,
      "loss": 0.0909,
      "mean_token_accuracy": 0.9725136160850525,
      "num_tokens": 58062799.0,
      "step": 290
    },
    {
      "epoch": 0.239642567018684,
      "grad_norm": 0.4761049747467041,
      "learning_rate": 1.7339560335168108e-05,
      "loss": 0.0687,
      "mean_token_accuracy": 0.9774059534072876,
      "num_tokens": 59062786.0,
      "step": 295
    },
    {
      "epoch": 0.2437043054427295,
      "grad_norm": 1.136487603187561,
      "learning_rate": 1.7280372494358326e-05,
      "loss": 0.0855,
      "mean_token_accuracy": 0.9744214117527008,
      "num_tokens": 60051266.0,
      "step": 300
    },
    {
      "epoch": 0.24776604386677498,
      "grad_norm": 0.4296221137046814,
      "learning_rate": 1.722024226405505e-05,
      "loss": 0.0655,
      "mean_token_accuracy": 0.978978979587555,
      "num_tokens": 61041820.0,
      "step": 305
    },
    {
      "epoch": 0.2518277822908205,
      "grad_norm": 0.5203496813774109,
      "learning_rate": 1.715918003366045e-05,
      "loss": 0.08,
      "mean_token_accuracy": 0.973987239599228,
      "num_tokens": 62062340.0,
      "step": 310
    },
    {
      "epoch": 0.255889520714866,
      "grad_norm": 0.6146700382232666,
      "learning_rate": 1.7097196353609213e-05,
      "loss": 0.0715,
      "mean_token_accuracy": 0.9770861506462097,
      "num_tokens": 63050727.0,
      "step": 315
    },
    {
      "epoch": 0.25995125913891143,
      "grad_norm": 0.5587770342826843,
      "learning_rate": 1.7034301933545664e-05,
      "loss": 0.0646,
      "mean_token_accuracy": 0.9802396357059479,
      "num_tokens": 64041728.0,
      "step": 320
    },
    {
      "epoch": 0.26401299756295693,
      "grad_norm": 0.3269279897212982,
      "learning_rate": 1.69705076404733e-05,
      "loss": 0.0717,
      "mean_token_accuracy": 0.9762361407279968,
      "num_tokens": 65041809.0,
      "step": 325
    },
    {
      "epoch": 0.26807473598700243,
      "grad_norm": 0.8979080319404602,
      "learning_rate": 1.6905824496877203e-05,
      "loss": 0.0831,
      "mean_token_accuracy": 0.9753478348255158,
      "num_tokens": 66070204.0,
      "step": 330
    },
    {
      "epoch": 0.27213647441104794,
      "grad_norm": 0.5646147727966309,
      "learning_rate": 1.684026367881953e-05,
      "loss": 0.0789,
      "mean_token_accuracy": 0.9754846930503845,
      "num_tokens": 67067978.0,
      "step": 335
    },
    {
      "epoch": 0.27619821283509344,
      "grad_norm": 0.5035107731819153,
      "learning_rate": 1.677383651400851e-05,
      "loss": 0.0758,
      "mean_token_accuracy": 0.9764487504959106,
      "num_tokens": 68064770.0,
      "step": 340
    },
    {
      "epoch": 0.2802599512591389,
      "grad_norm": 0.44015178084373474,
      "learning_rate": 1.670655447984122e-05,
      "loss": 0.0682,
      "mean_token_accuracy": 0.9785989046096801,
      "num_tokens": 69067404.0,
      "step": 345
    },
    {
      "epoch": 0.2843216896831844,
      "grad_norm": 0.43992072343826294,
      "learning_rate": 1.6638429201420494e-05,
      "loss": 0.0697,
      "mean_token_accuracy": 0.9792998969554901,
      "num_tokens": 70051522.0,
      "step": 350
    },
    {
      "epoch": 0.2883834281072299,
      "grad_norm": 0.37727996706962585,
      "learning_rate": 1.656947244954633e-05,
      "loss": 0.0802,
      "mean_token_accuracy": 0.9761293828487396,
      "num_tokens": 71058953.0,
      "step": 355
    },
    {
      "epoch": 0.2924451665312754,
      "grad_norm": 0.4864629805088043,
      "learning_rate": 1.6499696138682102e-05,
      "loss": 0.0626,
      "mean_token_accuracy": 0.9788231253623962,
      "num_tokens": 72064905.0,
      "step": 360
    },
    {
      "epoch": 0.2965069049553209,
      "grad_norm": 0.6466361880302429,
      "learning_rate": 1.6429112324895965e-05,
      "loss": 0.0773,
      "mean_token_accuracy": 0.9750386059284211,
      "num_tokens": 73055913.0,
      "step": 365
    },
    {
      "epoch": 0.3005686433793664,
      "grad_norm": 0.429469496011734,
      "learning_rate": 1.6357733203777778e-05,
      "loss": 0.062,
      "mean_token_accuracy": 0.9791361629962921,
      "num_tokens": 74062715.0,
      "step": 370
    },
    {
      "epoch": 0.30463038180341184,
      "grad_norm": 0.496416836977005,
      "learning_rate": 1.628557110833195e-05,
      "loss": 0.0734,
      "mean_token_accuracy": 0.9773313343524933,
      "num_tokens": 75061770.0,
      "step": 375
    },
    {
      "epoch": 0.30869212022745735,
      "grad_norm": 0.4150291383266449,
      "learning_rate": 1.6212638506846493e-05,
      "loss": 0.0662,
      "mean_token_accuracy": 0.9787569582462311,
      "num_tokens": 76084099.0,
      "step": 380
    },
    {
      "epoch": 0.31275385865150285,
      "grad_norm": 0.7713292241096497,
      "learning_rate": 1.613894800073876e-05,
      "loss": 0.0745,
      "mean_token_accuracy": 0.9783197581768036,
      "num_tokens": 77079407.0,
      "step": 385
    },
    {
      "epoch": 0.31681559707554835,
      "grad_norm": 0.4755105674266815,
      "learning_rate": 1.6064512322378122e-05,
      "loss": 0.0647,
      "mean_token_accuracy": 0.9794311225414276,
      "num_tokens": 78072762.0,
      "step": 390
    },
    {
      "epoch": 0.32087733549959385,
      "grad_norm": 0.5751511454582214,
      "learning_rate": 1.598934433288608e-05,
      "loss": 0.0654,
      "mean_token_accuracy": 0.9776327073574066,
      "num_tokens": 79082918.0,
      "step": 395
    },
    {
      "epoch": 0.3249390739236393,
      "grad_norm": 0.5209650993347168,
      "learning_rate": 1.591345701991407e-05,
      "loss": 0.0717,
      "mean_token_accuracy": 0.9785758316516876,
      "num_tokens": 80087826.0,
      "step": 400
    },
    {
      "epoch": 0.3290008123476848,
      "grad_norm": 0.39954304695129395,
      "learning_rate": 1.5836863495399453e-05,
      "loss": 0.0743,
      "mean_token_accuracy": 0.9768778443336487,
      "num_tokens": 81089906.0,
      "step": 405
    },
    {
      "epoch": 0.3330625507717303,
      "grad_norm": 0.5202496647834778,
      "learning_rate": 1.5759576993300008e-05,
      "loss": 0.0867,
      "mean_token_accuracy": 0.9760882616043091,
      "num_tokens": 82078098.0,
      "step": 410
    },
    {
      "epoch": 0.3371242891957758,
      "grad_norm": 0.3619210720062256,
      "learning_rate": 1.568161086730733e-05,
      "loss": 0.078,
      "mean_token_accuracy": 0.9770406484603882,
      "num_tokens": 83067880.0,
      "step": 415
    },
    {
      "epoch": 0.3411860276198213,
      "grad_norm": 0.5246071219444275,
      "learning_rate": 1.560297858853959e-05,
      "loss": 0.0603,
      "mean_token_accuracy": 0.9798705577850342,
      "num_tokens": 84068722.0,
      "step": 420
    },
    {
      "epoch": 0.34524776604386676,
      "grad_norm": 0.5286214351654053,
      "learning_rate": 1.552369374321394e-05,
      "loss": 0.0686,
      "mean_token_accuracy": 0.978751677274704,
      "num_tokens": 85057303.0,
      "step": 425
    },
    {
      "epoch": 0.34930950446791226,
      "grad_norm": 0.5156117677688599,
      "learning_rate": 1.5443770030299094e-05,
      "loss": 0.0659,
      "mean_token_accuracy": 0.9783392786979676,
      "num_tokens": 86051489.0,
      "step": 430
    },
    {
      "epoch": 0.35337124289195776,
      "grad_norm": 0.44470447301864624,
      "learning_rate": 1.5363221259148358e-05,
      "loss": 0.0574,
      "mean_token_accuracy": 0.9804697036743164,
      "num_tokens": 87063958.0,
      "step": 435
    },
    {
      "epoch": 0.35743298131600326,
      "grad_norm": 0.5599440932273865,
      "learning_rate": 1.528206134711367e-05,
      "loss": 0.0725,
      "mean_token_accuracy": 0.9779959142208099,
      "num_tokens": 88049361.0,
      "step": 440
    },
    {
      "epoch": 0.36149471974004876,
      "grad_norm": 0.5209068655967712,
      "learning_rate": 1.5200304317140906e-05,
      "loss": 0.0776,
      "mean_token_accuracy": 0.9772964715957642,
      "num_tokens": 89061185.0,
      "step": 445
    },
    {
      "epoch": 0.3655564581640942,
      "grad_norm": 0.27904748916625977,
      "learning_rate": 1.5117964295346982e-05,
      "loss": 0.0569,
      "mean_token_accuracy": 0.9810852766036987,
      "num_tokens": 90053534.0,
      "step": 450
    },
    {
      "epoch": 0.3696181965881397,
      "grad_norm": 0.41418689489364624,
      "learning_rate": 1.5035055508579122e-05,
      "loss": 0.067,
      "mean_token_accuracy": 0.9778187096118927,
      "num_tokens": 91055021.0,
      "step": 455
    },
    {
      "epoch": 0.3736799350121852,
      "grad_norm": 0.3942714333534241,
      "learning_rate": 1.4951592281956722e-05,
      "loss": 0.0722,
      "mean_token_accuracy": 0.9767170548439026,
      "num_tokens": 92070739.0,
      "step": 460
    },
    {
      "epoch": 0.3777416734362307,
      "grad_norm": 0.3835037350654602,
      "learning_rate": 1.4867589036396216e-05,
      "loss": 0.0724,
      "mean_token_accuracy": 0.9770483434200287,
      "num_tokens": 93045231.0,
      "step": 465
    },
    {
      "epoch": 0.3818034118602762,
      "grad_norm": 0.4278547763824463,
      "learning_rate": 1.4783060286119425e-05,
      "loss": 0.0638,
      "mean_token_accuracy": 0.9796122193336487,
      "num_tokens": 94043610.0,
      "step": 470
    },
    {
      "epoch": 0.38586515028432167,
      "grad_norm": 0.3801787793636322,
      "learning_rate": 1.4698020636145757e-05,
      "loss": 0.0779,
      "mean_token_accuracy": 0.9765353560447693,
      "num_tokens": 95045489.0,
      "step": 475
    },
    {
      "epoch": 0.38992688870836717,
      "grad_norm": 0.4804375469684601,
      "learning_rate": 1.4612484779768722e-05,
      "loss": 0.082,
      "mean_token_accuracy": 0.9749348998069763,
      "num_tokens": 96048701.0,
      "step": 480
    },
    {
      "epoch": 0.39398862713241267,
      "grad_norm": 0.569338858127594,
      "learning_rate": 1.4526467496017212e-05,
      "loss": 0.0701,
      "mean_token_accuracy": 0.9771721363067627,
      "num_tokens": 97041359.0,
      "step": 485
    },
    {
      "epoch": 0.3980503655564582,
      "grad_norm": 0.3253653645515442,
      "learning_rate": 1.4439983647101948e-05,
      "loss": 0.068,
      "mean_token_accuracy": 0.9796066105365753,
      "num_tokens": 98051228.0,
      "step": 490
    },
    {
      "epoch": 0.4021121039805037,
      "grad_norm": 0.47800612449645996,
      "learning_rate": 1.4353048175847556e-05,
      "loss": 0.0607,
      "mean_token_accuracy": 0.979815149307251,
      "num_tokens": 99057017.0,
      "step": 495
    },
    {
      "epoch": 0.4061738424045491,
      "grad_norm": 0.527355968952179,
      "learning_rate": 1.426567610311074e-05,
      "loss": 0.0611,
      "mean_token_accuracy": 0.9794078767299652,
      "num_tokens": 100067101.0,
      "step": 500
    },
    {
      "epoch": 0.4102355808285946,
      "grad_norm": 0.43324795365333557,
      "learning_rate": 1.417788252518493e-05,
      "loss": 0.0768,
      "mean_token_accuracy": 0.9763797581195831,
      "num_tokens": 101065634.0,
      "step": 505
    },
    {
      "epoch": 0.41429731925264013,
      "grad_norm": 0.372696191072464,
      "learning_rate": 1.408968261119194e-05,
      "loss": 0.0582,
      "mean_token_accuracy": 0.9820960402488709,
      "num_tokens": 102077893.0,
      "step": 510
    },
    {
      "epoch": 0.41835905767668563,
      "grad_norm": 0.5359270572662354,
      "learning_rate": 1.4001091600461005e-05,
      "loss": 0.0651,
      "mean_token_accuracy": 0.9794639050960541,
      "num_tokens": 103086975.0,
      "step": 515
    },
    {
      "epoch": 0.42242079610073113,
      "grad_norm": 0.43643906712532043,
      "learning_rate": 1.3912124799895712e-05,
      "loss": 0.0558,
      "mean_token_accuracy": 0.9812205255031585,
      "num_tokens": 104092373.0,
      "step": 520
    },
    {
      "epoch": 0.4264825345247766,
      "grad_norm": 0.3926670253276825,
      "learning_rate": 1.3822797581329248e-05,
      "loss": 0.0701,
      "mean_token_accuracy": 0.9781829357147217,
      "num_tokens": 105090458.0,
      "step": 525
    },
    {
      "epoch": 0.4305442729488221,
      "grad_norm": 0.5042207837104797,
      "learning_rate": 1.3733125378868431e-05,
      "loss": 0.07,
      "mean_token_accuracy": 0.9787480890750885,
      "num_tokens": 106092066.0,
      "step": 530
    },
    {
      "epoch": 0.4346060113728676,
      "grad_norm": 0.38992759585380554,
      "learning_rate": 1.3643123686226967e-05,
      "loss": 0.0662,
      "mean_token_accuracy": 0.9801897227764129,
      "num_tokens": 107100108.0,
      "step": 535
    },
    {
      "epoch": 0.4386677497969131,
      "grad_norm": 0.30508944392204285,
      "learning_rate": 1.3552808054048443e-05,
      "loss": 0.0554,
      "mean_token_accuracy": 0.9811424314975739,
      "num_tokens": 108112187.0,
      "step": 540
    },
    {
      "epoch": 0.4427294882209586,
      "grad_norm": 0.47000792622566223,
      "learning_rate": 1.3462194087219425e-05,
      "loss": 0.0708,
      "mean_token_accuracy": 0.9791782617568969,
      "num_tokens": 109114693.0,
      "step": 545
    },
    {
      "epoch": 0.44679122664500404,
      "grad_norm": 0.47066664695739746,
      "learning_rate": 1.337129744217326e-05,
      "loss": 0.0668,
      "mean_token_accuracy": 0.9796491503715515,
      "num_tokens": 110115381.0,
      "step": 550
    },
    {
      "epoch": 0.45085296506904954,
      "grad_norm": 0.3269657492637634,
      "learning_rate": 1.3280133824184903e-05,
      "loss": 0.0455,
      "mean_token_accuracy": 0.983445793390274,
      "num_tokens": 111123116.0,
      "step": 555
    },
    {
      "epoch": 0.45491470349309504,
      "grad_norm": 0.4069489538669586,
      "learning_rate": 1.3188718984657341e-05,
      "loss": 0.053,
      "mean_token_accuracy": 0.982389360666275,
      "num_tokens": 112131658.0,
      "step": 560
    },
    {
      "epoch": 0.45897644191714054,
      "grad_norm": 0.47114020586013794,
      "learning_rate": 1.3097068718400052e-05,
      "loss": 0.0682,
      "mean_token_accuracy": 0.979246187210083,
      "num_tokens": 113117295.0,
      "step": 565
    },
    {
      "epoch": 0.46303818034118605,
      "grad_norm": 0.500640869140625,
      "learning_rate": 1.3005198860899923e-05,
      "loss": 0.0667,
      "mean_token_accuracy": 0.9778048694133759,
      "num_tokens": 114109691.0,
      "step": 570
    },
    {
      "epoch": 0.4670999187652315,
      "grad_norm": 0.5766412019729614,
      "learning_rate": 1.2913125285585197e-05,
      "loss": 0.0689,
      "mean_token_accuracy": 0.9785227596759796,
      "num_tokens": 115115022.0,
      "step": 575
    },
    {
      "epoch": 0.471161657189277,
      "grad_norm": 0.4441542625427246,
      "learning_rate": 1.2820863901082816e-05,
      "loss": 0.0714,
      "mean_token_accuracy": 0.9776635289192199,
      "num_tokens": 116122504.0,
      "step": 580
    },
    {
      "epoch": 0.4752233956133225,
      "grad_norm": 0.27584049105644226,
      "learning_rate": 1.2728430648469712e-05,
      "loss": 0.0564,
      "mean_token_accuracy": 0.9820808231830597,
      "num_tokens": 117127778.0,
      "step": 585
    },
    {
      "epoch": 0.479285134037368,
      "grad_norm": 0.4740694463253021,
      "learning_rate": 1.263584149851848e-05,
      "loss": 0.0583,
      "mean_token_accuracy": 0.9806334555149079,
      "num_tokens": 118133172.0,
      "step": 590
    },
    {
      "epoch": 0.4833468724614135,
      "grad_norm": 0.3616255223751068,
      "learning_rate": 1.2543112448937912e-05,
      "loss": 0.069,
      "mean_token_accuracy": 0.9772207081317902,
      "num_tokens": 119127106.0,
      "step": 595
    },
    {
      "epoch": 0.487408610885459,
      "grad_norm": 0.48199892044067383,
      "learning_rate": 1.2450259521608902e-05,
      "loss": 0.0561,
      "mean_token_accuracy": 0.9815092027187348,
      "num_tokens": 120127921.0,
      "step": 600
    },
    {
      "epoch": 0.49147034930950445,
      "grad_norm": 0.44914308190345764,
      "learning_rate": 1.2357298759816144e-05,
      "loss": 0.0696,
      "mean_token_accuracy": 0.9784136474132538,
      "num_tokens": 121113617.0,
      "step": 605
    },
    {
      "epoch": 0.49553208773354995,
      "grad_norm": 0.6228646039962769,
      "learning_rate": 1.2264246225476145e-05,
      "loss": 0.0718,
      "mean_token_accuracy": 0.97708158493042,
      "num_tokens": 122104372.0,
      "step": 610
    },
    {
      "epoch": 0.49959382615759546,
      "grad_norm": 0.6243708729743958,
      "learning_rate": 1.2171117996362045e-05,
      "loss": 0.0827,
      "mean_token_accuracy": 0.9765641748905182,
      "num_tokens": 123108223.0,
      "step": 615
    },
    {
      "epoch": 0.503655564581641,
      "grad_norm": 0.4215443730354309,
      "learning_rate": 1.2077930163325635e-05,
      "loss": 0.067,
      "mean_token_accuracy": 0.9782631993293762,
      "num_tokens": 124118187.0,
      "step": 620
    },
    {
      "epoch": 0.5077173030056864,
      "grad_norm": 0.4475902318954468,
      "learning_rate": 1.1984698827517183e-05,
      "loss": 0.0617,
      "mean_token_accuracy": 0.9806814730167389,
      "num_tokens": 125114685.0,
      "step": 625
    },
    {
      "epoch": 0.511779041429732,
      "grad_norm": 0.32930293679237366,
      "learning_rate": 1.1891440097603434e-05,
      "loss": 0.0623,
      "mean_token_accuracy": 0.979959774017334,
      "num_tokens": 126115241.0,
      "step": 630
    },
    {
      "epoch": 0.5158407798537774,
      "grad_norm": 0.4772458076477051,
      "learning_rate": 1.1798170086984339e-05,
      "loss": 0.0637,
      "mean_token_accuracy": 0.9797438502311706,
      "num_tokens": 127111697.0,
      "step": 635
    },
    {
      "epoch": 0.5199025182778229,
      "grad_norm": 0.4190440773963928,
      "learning_rate": 1.1704904911008935e-05,
      "loss": 0.0586,
      "mean_token_accuracy": 0.9803499579429626,
      "num_tokens": 128116084.0,
      "step": 640
    },
    {
      "epoch": 0.5239642567018684,
      "grad_norm": 0.47672557830810547,
      "learning_rate": 1.161166068419094e-05,
      "loss": 0.0717,
      "mean_token_accuracy": 0.97803236246109,
      "num_tokens": 129124304.0,
      "step": 645
    },
    {
      "epoch": 0.5280259951259139,
      "grad_norm": 0.45728600025177,
      "learning_rate": 1.1518453517424426e-05,
      "loss": 0.0649,
      "mean_token_accuracy": 0.9799485743045807,
      "num_tokens": 130130941.0,
      "step": 650
    },
    {
      "epoch": 0.5320877335499594,
      "grad_norm": 0.5543997883796692,
      "learning_rate": 1.1425299515200173e-05,
      "loss": 0.1053,
      "mean_token_accuracy": 0.9713131427764893,
      "num_tokens": 131138247.0,
      "step": 655
    },
    {
      "epoch": 0.5361494719740049,
      "grad_norm": 0.4355736970901489,
      "learning_rate": 1.1332214772823103e-05,
      "loss": 0.0561,
      "mean_token_accuracy": 0.9821233510971069,
      "num_tokens": 132126842.0,
      "step": 660
    },
    {
      "epoch": 0.5402112103980503,
      "grad_norm": 0.3682590425014496,
      "learning_rate": 1.1239215373631297e-05,
      "loss": 0.0626,
      "mean_token_accuracy": 0.9794946372509002,
      "num_tokens": 133123218.0,
      "step": 665
    },
    {
      "epoch": 0.5442729488220959,
      "grad_norm": 0.26944273710250854,
      "learning_rate": 1.1146317386217115e-05,
      "loss": 0.0616,
      "mean_token_accuracy": 0.9809187352657318,
      "num_tokens": 134129997.0,
      "step": 670
    },
    {
      "epoch": 0.5483346872461413,
      "grad_norm": 0.7072899341583252,
      "learning_rate": 1.1053536861650798e-05,
      "loss": 0.0628,
      "mean_token_accuracy": 0.9790135264396668,
      "num_tokens": 135124661.0,
      "step": 675
    },
    {
      "epoch": 0.5523964256701869,
      "grad_norm": 0.46994104981422424,
      "learning_rate": 1.0960889830707175e-05,
      "loss": 0.0673,
      "mean_token_accuracy": 0.9778666079044342,
      "num_tokens": 136135364.0,
      "step": 680
    },
    {
      "epoch": 0.5564581640942323,
      "grad_norm": 0.4814133048057556,
      "learning_rate": 1.086839230109582e-05,
      "loss": 0.0675,
      "mean_token_accuracy": 0.9789254486560821,
      "num_tokens": 137142278.0,
      "step": 685
    },
    {
      "epoch": 0.5605199025182778,
      "grad_norm": 0.44070592522621155,
      "learning_rate": 1.0776060254695205e-05,
      "loss": 0.0658,
      "mean_token_accuracy": 0.9797971844673157,
      "num_tokens": 138149535.0,
      "step": 690
    },
    {
      "epoch": 0.5645816409423233,
      "grad_norm": 0.3803769648075104,
      "learning_rate": 1.0683909644791352e-05,
      "loss": 0.062,
      "mean_token_accuracy": 0.9803054690361023,
      "num_tokens": 139140136.0,
      "step": 695
    },
    {
      "epoch": 0.5686433793663688,
      "grad_norm": 0.3313233256340027,
      "learning_rate": 1.0591956393321364e-05,
      "loss": 0.0598,
      "mean_token_accuracy": 0.9789064943790435,
      "num_tokens": 140151980.0,
      "step": 700
    },
    {
      "epoch": 0.5727051177904143,
      "grad_norm": 0.33801141381263733,
      "learning_rate": 1.0500216388122435e-05,
      "loss": 0.0516,
      "mean_token_accuracy": 0.9833789706230164,
      "num_tokens": 141160974.0,
      "step": 705
    },
    {
      "epoch": 0.5767668562144598,
      "grad_norm": 0.3508951961994171,
      "learning_rate": 1.0408705480186703e-05,
      "loss": 0.0521,
      "mean_token_accuracy": 0.9830210030078887,
      "num_tokens": 142162480.0,
      "step": 710
    },
    {
      "epoch": 0.5808285946385053,
      "grad_norm": 0.4663684666156769,
      "learning_rate": 1.0317439480922496e-05,
      "loss": 0.064,
      "mean_token_accuracy": 0.9799285590648651,
      "num_tokens": 143156903.0,
      "step": 715
    },
    {
      "epoch": 0.5848903330625508,
      "grad_norm": 0.446106880903244,
      "learning_rate": 1.0226434159422418e-05,
      "loss": 0.0635,
      "mean_token_accuracy": 0.9789406716823578,
      "num_tokens": 144144740.0,
      "step": 720
    },
    {
      "epoch": 0.5889520714865962,
      "grad_norm": 0.411443829536438,
      "learning_rate": 1.0135705239738724e-05,
      "loss": 0.0536,
      "mean_token_accuracy": 0.983045494556427,
      "num_tokens": 145151972.0,
      "step": 725
    },
    {
      "epoch": 0.5930138099106418,
      "grad_norm": 0.3873978853225708,
      "learning_rate": 1.0045268398166498e-05,
      "loss": 0.0556,
      "mean_token_accuracy": 0.9815871775150299,
      "num_tokens": 146159772.0,
      "step": 730
    },
    {
      "epoch": 0.5970755483346872,
      "grad_norm": 0.39525410532951355,
      "learning_rate": 9.955139260535084e-06,
      "loss": 0.0646,
      "mean_token_accuracy": 0.9803644895553589,
      "num_tokens": 147153505.0,
      "step": 735
    },
    {
      "epoch": 0.6011372867587328,
      "grad_norm": 0.5776126980781555,
      "learning_rate": 9.865333399508215e-06,
      "loss": 0.0844,
      "mean_token_accuracy": 0.9744473040103913,
      "num_tokens": 148152507.0,
      "step": 740
    },
    {
      "epoch": 0.6051990251827782,
      "grad_norm": 0.47703781723976135,
      "learning_rate": 9.775866331893363e-06,
      "loss": 0.0745,
      "mean_token_accuracy": 0.978210324048996,
      "num_tokens": 149143516.0,
      "step": 745
    },
    {
      "epoch": 0.6092607636068237,
      "grad_norm": 0.2773439884185791,
      "learning_rate": 9.686753515960682e-06,
      "loss": 0.0474,
      "mean_token_accuracy": 0.9832984507083893,
      "num_tokens": 150153753.0,
      "step": 750
    },
    {
      "epoch": 0.6133225020308692,
      "grad_norm": 0.2977774739265442,
      "learning_rate": 9.59801034877215e-06,
      "loss": 0.055,
      "mean_token_accuracy": 0.9833644390106201,
      "num_tokens": 151157935.0,
      "step": 755
    },
    {
      "epoch": 0.6173842404549147,
      "grad_norm": 0.5760729908943176,
      "learning_rate": 9.509652163521203e-06,
      "loss": 0.0661,
      "mean_token_accuracy": 0.979590117931366,
      "num_tokens": 152146336.0,
      "step": 760
    },
    {
      "epoch": 0.6214459788789602,
      "grad_norm": 0.3095396161079407,
      "learning_rate": 9.421694226883443e-06,
      "loss": 0.0467,
      "mean_token_accuracy": 0.9841603338718414,
      "num_tokens": 153155482.0,
      "step": 765
    },
    {
      "epoch": 0.6255077173030057,
      "grad_norm": 0.5612005591392517,
      "learning_rate": 9.334151736378866e-06,
      "loss": 0.0686,
      "mean_token_accuracy": 0.9793064653873443,
      "num_tokens": 154147094.0,
      "step": 770
    },
    {
      "epoch": 0.6295694557270511,
      "grad_norm": 0.2929650843143463,
      "learning_rate": 9.247039817745963e-06,
      "loss": 0.0547,
      "mean_token_accuracy": 0.9809925734996796,
      "num_tokens": 155150669.0,
      "step": 775
    },
    {
      "epoch": 0.6336311941510967,
      "grad_norm": 0.4886499047279358,
      "learning_rate": 9.160373522328317e-06,
      "loss": 0.066,
      "mean_token_accuracy": 0.9795759618282318,
      "num_tokens": 156151667.0,
      "step": 780
    },
    {
      "epoch": 0.6376929325751421,
      "grad_norm": 0.3836148977279663,
      "learning_rate": 9.074167824473968e-06,
      "loss": 0.0658,
      "mean_token_accuracy": 0.9785556554794311,
      "num_tokens": 157140574.0,
      "step": 785
    },
    {
      "epoch": 0.6417546709991877,
      "grad_norm": 0.29880833625793457,
      "learning_rate": 8.988437618948137e-06,
      "loss": 0.0665,
      "mean_token_accuracy": 0.9799898028373718,
      "num_tokens": 158150511.0,
      "step": 790
    },
    {
      "epoch": 0.6458164094232332,
      "grad_norm": 0.5122634768486023,
      "learning_rate": 8.903197718359691e-06,
      "loss": 0.0634,
      "mean_token_accuracy": 0.9797806441783905,
      "num_tokens": 159142424.0,
      "step": 795
    },
    {
      "epoch": 0.6498781478472786,
      "grad_norm": 0.42663833498954773,
      "learning_rate": 8.818462850601776e-06,
      "loss": 0.0625,
      "mean_token_accuracy": 0.9785671591758728,
      "num_tokens": 160144444.0,
      "step": 800
    },
    {
      "epoch": 0.6539398862713242,
      "grad_norm": 0.5549625158309937,
      "learning_rate": 8.734247656307128e-06,
      "loss": 0.0692,
      "mean_token_accuracy": 0.9773725211620331,
      "num_tokens": 161135545.0,
      "step": 805
    },
    {
      "epoch": 0.6580016246953696,
      "grad_norm": 0.30039548873901367,
      "learning_rate": 8.650566686318425e-06,
      "loss": 0.0635,
      "mean_token_accuracy": 0.980705839395523,
      "num_tokens": 162142031.0,
      "step": 810
    },
    {
      "epoch": 0.6620633631194152,
      "grad_norm": 0.44756248593330383,
      "learning_rate": 8.567434399174163e-06,
      "loss": 0.0564,
      "mean_token_accuracy": 0.9830853104591369,
      "num_tokens": 163135791.0,
      "step": 815
    },
    {
      "epoch": 0.6661251015434606,
      "grad_norm": 0.36431410908699036,
      "learning_rate": 8.484865158610512e-06,
      "loss": 0.056,
      "mean_token_accuracy": 0.9825919508934021,
      "num_tokens": 164120104.0,
      "step": 820
    },
    {
      "epoch": 0.670186839967506,
      "grad_norm": 0.3484279215335846,
      "learning_rate": 8.402873231079494e-06,
      "loss": 0.0637,
      "mean_token_accuracy": 0.9789206087589264,
      "num_tokens": 165105725.0,
      "step": 825
    },
    {
      "epoch": 0.6742485783915516,
      "grad_norm": 0.35079893469810486,
      "learning_rate": 8.321472783284026e-06,
      "loss": 0.0638,
      "mean_token_accuracy": 0.9801722347736359,
      "num_tokens": 166100814.0,
      "step": 830
    },
    {
      "epoch": 0.6783103168155971,
      "grad_norm": 0.4066307544708252,
      "learning_rate": 8.240677879730155e-06,
      "loss": 0.0572,
      "mean_token_accuracy": 0.9807899057865143,
      "num_tokens": 167088158.0,
      "step": 835
    },
    {
      "epoch": 0.6823720552396426,
      "grad_norm": 0.5358293652534485,
      "learning_rate": 8.16050248029698e-06,
      "loss": 0.0549,
      "mean_token_accuracy": 0.9824902653694153,
      "num_tokens": 168098614.0,
      "step": 840
    },
    {
      "epoch": 0.6864337936636881,
      "grad_norm": 0.2488897442817688,
      "learning_rate": 8.080960437824644e-06,
      "loss": 0.0609,
      "mean_token_accuracy": 0.9807324767112732,
      "num_tokens": 169095522.0,
      "step": 845
    },
    {
      "epoch": 0.6904955320877335,
      "grad_norm": 0.42293569445610046,
      "learning_rate": 8.0020654957208e-06,
      "loss": 0.0505,
      "mean_token_accuracy": 0.9821749269962311,
      "num_tokens": 170088680.0,
      "step": 850
    },
    {
      "epoch": 0.6945572705117791,
      "grad_norm": 0.44050925970077515,
      "learning_rate": 7.923831285586022e-06,
      "loss": 0.0681,
      "mean_token_accuracy": 0.9774746835231781,
      "num_tokens": 171089517.0,
      "step": 855
    },
    {
      "epoch": 0.6986190089358245,
      "grad_norm": 0.7163667678833008,
      "learning_rate": 7.846271324858501e-06,
      "loss": 0.0704,
      "mean_token_accuracy": 0.9801386773586274,
      "num_tokens": 172078474.0,
      "step": 860
    },
    {
      "epoch": 0.7026807473598701,
      "grad_norm": 0.45846736431121826,
      "learning_rate": 7.769399014478478e-06,
      "loss": 0.06,
      "mean_token_accuracy": 0.9795966267585754,
      "num_tokens": 173070609.0,
      "step": 865
    },
    {
      "epoch": 0.7067424857839155,
      "grad_norm": 0.2678920328617096,
      "learning_rate": 7.693227636572826e-06,
      "loss": 0.0595,
      "mean_token_accuracy": 0.9809933125972747,
      "num_tokens": 174085368.0,
      "step": 870
    },
    {
      "epoch": 0.710804224207961,
      "grad_norm": 0.5795628428459167,
      "learning_rate": 7.617770352160114e-06,
      "loss": 0.0608,
      "mean_token_accuracy": 0.9820488452911377,
      "num_tokens": 175081043.0,
      "step": 875
    },
    {
      "epoch": 0.7148659626320065,
      "grad_norm": 0.30526280403137207,
      "learning_rate": 7.543040198876647e-06,
      "loss": 0.0495,
      "mean_token_accuracy": 0.9833504259586334,
      "num_tokens": 176077491.0,
      "step": 880
    },
    {
      "epoch": 0.718927701056052,
      "grad_norm": 0.37808308005332947,
      "learning_rate": 7.469050088723789e-06,
      "loss": 0.0654,
      "mean_token_accuracy": 0.9787123501300812,
      "num_tokens": 177065004.0,
      "step": 885
    },
    {
      "epoch": 0.7229894394800975,
      "grad_norm": 0.5893298983573914,
      "learning_rate": 7.3958128058370085e-06,
      "loss": 0.0642,
      "mean_token_accuracy": 0.978962916135788,
      "num_tokens": 178057013.0,
      "step": 890
    },
    {
      "epoch": 0.727051177904143,
      "grad_norm": 0.43597474694252014,
      "learning_rate": 7.323341004277018e-06,
      "loss": 0.0687,
      "mean_token_accuracy": 0.9791975021362305,
      "num_tokens": 179052588.0,
      "step": 895
    },
    {
      "epoch": 0.7311129163281884,
      "grad_norm": 0.37778469920158386,
      "learning_rate": 7.251647205843377e-06,
      "loss": 0.0644,
      "mean_token_accuracy": 0.9794808864593506,
      "num_tokens": 180049819.0,
      "step": 900
    },
    {
      "epoch": 0.735174654752234,
      "grad_norm": 0.3205077350139618,
      "learning_rate": 7.1807437979109645e-06,
      "loss": 0.0503,
      "mean_token_accuracy": 0.9838395118713379,
      "num_tokens": 181048461.0,
      "step": 905
    },
    {
      "epoch": 0.7392363931762794,
      "grad_norm": 0.29493004083633423,
      "learning_rate": 7.110643031289658e-06,
      "loss": 0.0568,
      "mean_token_accuracy": 0.9822260618209839,
      "num_tokens": 182040723.0,
      "step": 910
    },
    {
      "epoch": 0.743298131600325,
      "grad_norm": 0.3917006552219391,
      "learning_rate": 7.041357018107616e-06,
      "loss": 0.057,
      "mean_token_accuracy": 0.9810117006301879,
      "num_tokens": 183048448.0,
      "step": 915
    },
    {
      "epoch": 0.7473598700243704,
      "grad_norm": 0.5472274422645569,
      "learning_rate": 6.972897729718543e-06,
      "loss": 0.0626,
      "mean_token_accuracy": 0.9806600928306579,
      "num_tokens": 184052714.0,
      "step": 920
    },
    {
      "epoch": 0.7514216084484159,
      "grad_norm": 0.3639894127845764,
      "learning_rate": 6.90527699463323e-06,
      "loss": 0.063,
      "mean_token_accuracy": 0.9794709146022796,
      "num_tokens": 185044196.0,
      "step": 925
    },
    {
      "epoch": 0.7554833468724614,
      "grad_norm": 0.6877843737602234,
      "learning_rate": 6.838506496475833e-06,
      "loss": 0.0599,
      "mean_token_accuracy": 0.982112580537796,
      "num_tokens": 186051636.0,
      "step": 930
    },
    {
      "epoch": 0.7595450852965069,
      "grad_norm": 0.580552875995636,
      "learning_rate": 6.772597771965139e-06,
      "loss": 0.0575,
      "mean_token_accuracy": 0.9812103390693665,
      "num_tokens": 187053272.0,
      "step": 935
    },
    {
      "epoch": 0.7636068237205524,
      "grad_norm": 0.4036097824573517,
      "learning_rate": 6.707562208921234e-06,
      "loss": 0.0562,
      "mean_token_accuracy": 0.9822783052921296,
      "num_tokens": 188064207.0,
      "step": 940
    },
    {
      "epoch": 0.7676685621445979,
      "grad_norm": 0.39085477590560913,
      "learning_rate": 6.643411044297905e-06,
      "loss": 0.0664,
      "mean_token_accuracy": 0.9777272641658783,
      "num_tokens": 189067036.0,
      "step": 945
    },
    {
      "epoch": 0.7717303005686433,
      "grad_norm": 0.307657390832901,
      "learning_rate": 6.580155362241091e-06,
      "loss": 0.0578,
      "mean_token_accuracy": 0.9805595517158509,
      "num_tokens": 190074345.0,
      "step": 950
    },
    {
      "epoch": 0.7757920389926889,
      "grad_norm": 0.3961021900177002,
      "learning_rate": 6.517806092173753e-06,
      "loss": 0.0631,
      "mean_token_accuracy": 0.9791252434253692,
      "num_tokens": 191070785.0,
      "step": 955
    },
    {
      "epoch": 0.7798537774167343,
      "grad_norm": 0.35744354128837585,
      "learning_rate": 6.456374006907465e-06,
      "loss": 0.0514,
      "mean_token_accuracy": 0.9818819284439086,
      "num_tokens": 192075996.0,
      "step": 960
    },
    {
      "epoch": 0.7839155158407799,
      "grad_norm": 0.46781015396118164,
      "learning_rate": 6.395869720781067e-06,
      "loss": 0.0574,
      "mean_token_accuracy": 0.9805688560009003,
      "num_tokens": 193087138.0,
      "step": 965
    },
    {
      "epoch": 0.7879772542648253,
      "grad_norm": 0.36879926919937134,
      "learning_rate": 6.3363036878267145e-06,
      "loss": 0.0613,
      "mean_token_accuracy": 0.9804566144943238,
      "num_tokens": 194094465.0,
      "step": 970
    },
    {
      "epoch": 0.7920389926888708,
      "grad_norm": 0.42734435200691223,
      "learning_rate": 6.2776861999635945e-06,
      "loss": 0.0558,
      "mean_token_accuracy": 0.9807806551456452,
      "num_tokens": 195096915.0,
      "step": 975
    },
    {
      "epoch": 0.7961007311129163,
      "grad_norm": 0.5116071701049805,
      "learning_rate": 6.220027385219691e-06,
      "loss": 0.0481,
      "mean_token_accuracy": 0.9837054967880249,
      "num_tokens": 196111974.0,
      "step": 980
    },
    {
      "epoch": 0.8001624695369618,
      "grad_norm": 0.4918130040168762,
      "learning_rate": 6.163337205981827e-06,
      "loss": 0.0697,
      "mean_token_accuracy": 0.9781570851802825,
      "num_tokens": 197106408.0,
      "step": 985
    },
    {
      "epoch": 0.8042242079610074,
      "grad_norm": 0.3182407319545746,
      "learning_rate": 6.1076254572743625e-06,
      "loss": 0.0519,
      "mean_token_accuracy": 0.9842983186244965,
      "num_tokens": 198118832.0,
      "step": 990
    },
    {
      "epoch": 0.8082859463850528,
      "grad_norm": 0.5030460953712463,
      "learning_rate": 6.052901765066786e-06,
      "loss": 0.0704,
      "mean_token_accuracy": 0.9800339102745056,
      "num_tokens": 199115719.0,
      "step": 995
    },
    {
      "epoch": 0.8123476848090982,
      "grad_norm": 0.37998536229133606,
      "learning_rate": 5.999175584610519e-06,
      "loss": 0.0569,
      "mean_token_accuracy": 0.9805939435958863,
      "num_tokens": 200127338.0,
      "step": 1000
    },
    {
      "epoch": 0.8164094232331438,
      "grad_norm": 0.3664904236793518,
      "learning_rate": 5.946456198805239e-06,
      "loss": 0.0513,
      "mean_token_accuracy": 0.9824835896492005,
      "num_tokens": 201138604.0,
      "step": 1005
    },
    {
      "epoch": 0.8204711616571893,
      "grad_norm": 0.33573564887046814,
      "learning_rate": 5.894752716594947e-06,
      "loss": 0.0669,
      "mean_token_accuracy": 0.979474687576294,
      "num_tokens": 202129177.0,
      "step": 1010
    },
    {
      "epoch": 0.8245329000812348,
      "grad_norm": 0.4686684012413025,
      "learning_rate": 5.844074071394111e-06,
      "loss": 0.0664,
      "mean_token_accuracy": 0.9795205295085907,
      "num_tokens": 203146381.0,
      "step": 1015
    },
    {
      "epoch": 0.8285946385052803,
      "grad_norm": 0.3460758924484253,
      "learning_rate": 5.794429019544149e-06,
      "loss": 0.0663,
      "mean_token_accuracy": 0.9791534006595611,
      "num_tokens": 204152596.0,
      "step": 1020
    },
    {
      "epoch": 0.8326563769293257,
      "grad_norm": 0.3609422743320465,
      "learning_rate": 5.745826138800474e-06,
      "loss": 0.0604,
      "mean_token_accuracy": 0.980061823129654,
      "num_tokens": 205152169.0,
      "step": 1025
    },
    {
      "epoch": 0.8367181153533713,
      "grad_norm": 0.3568183481693268,
      "learning_rate": 5.698273826850436e-06,
      "loss": 0.0622,
      "mean_token_accuracy": 0.9805078387260437,
      "num_tokens": 206147857.0,
      "step": 1030
    },
    {
      "epoch": 0.8407798537774167,
      "grad_norm": 0.40073588490486145,
      "learning_rate": 5.651780299862338e-06,
      "loss": 0.0665,
      "mean_token_accuracy": 0.9787575542926789,
      "num_tokens": 207139657.0,
      "step": 1035
    },
    {
      "epoch": 0.8448415922014623,
      "grad_norm": 0.40601271390914917,
      "learning_rate": 5.6063535910658375e-06,
      "loss": 0.0566,
      "mean_token_accuracy": 0.9825579047203064,
      "num_tokens": 208132208.0,
      "step": 1040
    },
    {
      "epoch": 0.8489033306255077,
      "grad_norm": 0.38374555110931396,
      "learning_rate": 5.562001549363961e-06,
      "loss": 0.0657,
      "mean_token_accuracy": 0.9789287865161895,
      "num_tokens": 209146041.0,
      "step": 1045
    },
    {
      "epoch": 0.8529650690495532,
      "grad_norm": 0.5932730436325073,
      "learning_rate": 5.518731837976942e-06,
      "loss": 0.0705,
      "mean_token_accuracy": 0.9791133642196655,
      "num_tokens": 210150616.0,
      "step": 1050
    },
    {
      "epoch": 0.8570268074735987,
      "grad_norm": 0.3028200566768646,
      "learning_rate": 5.476551933118173e-06,
      "loss": 0.0474,
      "mean_token_accuracy": 0.9840903222560883,
      "num_tokens": 211156350.0,
      "step": 1055
    },
    {
      "epoch": 0.8610885458976442,
      "grad_norm": 0.3962152898311615,
      "learning_rate": 5.435469122702438e-06,
      "loss": 0.0609,
      "mean_token_accuracy": 0.9799316108226777,
      "num_tokens": 212157383.0,
      "step": 1060
    },
    {
      "epoch": 0.8651502843216897,
      "grad_norm": 0.5666550397872925,
      "learning_rate": 5.395490505086702e-06,
      "loss": 0.0676,
      "mean_token_accuracy": 0.9790563762187958,
      "num_tokens": 213137853.0,
      "step": 1065
    },
    {
      "epoch": 0.8692120227457352,
      "grad_norm": 0.6989259719848633,
      "learning_rate": 5.356622987843649e-06,
      "loss": 0.0703,
      "mean_token_accuracy": 0.9787014901638031,
      "num_tokens": 214139586.0,
      "step": 1070
    },
    {
      "epoch": 0.8732737611697806,
      "grad_norm": 0.43771055340766907,
      "learning_rate": 5.318873286568167e-06,
      "loss": 0.0497,
      "mean_token_accuracy": 0.9833686113357544,
      "num_tokens": 215134019.0,
      "step": 1075
    },
    {
      "epoch": 0.8773354995938262,
      "grad_norm": 0.5760379433631897,
      "learning_rate": 5.282247923717032e-06,
      "loss": 0.0528,
      "mean_token_accuracy": 0.9836243689060211,
      "num_tokens": 216120175.0,
      "step": 1080
    },
    {
      "epoch": 0.8813972380178716,
      "grad_norm": 0.3596481382846832,
      "learning_rate": 5.246753227481934e-06,
      "loss": 0.0619,
      "mean_token_accuracy": 0.9801043391227722,
      "num_tokens": 217113238.0,
      "step": 1085
    },
    {
      "epoch": 0.8854589764419172,
      "grad_norm": 0.38112854957580566,
      "learning_rate": 5.212395330696087e-06,
      "loss": 0.0656,
      "mean_token_accuracy": 0.9779551386833191,
      "num_tokens": 218119999.0,
      "step": 1090
    },
    {
      "epoch": 0.8895207148659626,
      "grad_norm": 0.4955122768878937,
      "learning_rate": 5.17918016977459e-06,
      "loss": 0.0658,
      "mean_token_accuracy": 0.9795345664024353,
      "num_tokens": 219106406.0,
      "step": 1095
    },
    {
      "epoch": 0.8935824532900081,
      "grad_norm": 0.29065176844596863,
      "learning_rate": 5.147113483688719e-06,
      "loss": 0.0537,
      "mean_token_accuracy": 0.981332677602768,
      "num_tokens": 220100206.0,
      "step": 1100
    },
    {
      "epoch": 0.8976441917140536,
      "grad_norm": 0.5007052421569824,
      "learning_rate": 5.11620081297434e-06,
      "loss": 0.0553,
      "mean_token_accuracy": 0.9821198642253876,
      "num_tokens": 221105374.0,
      "step": 1105
    },
    {
      "epoch": 0.9017059301380991,
      "grad_norm": 0.3969530463218689,
      "learning_rate": 5.086447498774603e-06,
      "loss": 0.0608,
      "mean_token_accuracy": 0.9799468398094178,
      "num_tokens": 222094453.0,
      "step": 1110
    },
    {
      "epoch": 0.9057676685621446,
      "grad_norm": 0.9423429369926453,
      "learning_rate": 5.057858681917095e-06,
      "loss": 0.0653,
      "mean_token_accuracy": 0.9788918614387512,
      "num_tokens": 223094079.0,
      "step": 1115
    },
    {
      "epoch": 0.9098294069861901,
      "grad_norm": 0.4476824402809143,
      "learning_rate": 5.030439302025599e-06,
      "loss": 0.0729,
      "mean_token_accuracy": 0.9781366765499115,
      "num_tokens": 224081076.0,
      "step": 1120
    },
    {
      "epoch": 0.9138911454102355,
      "grad_norm": 0.2742629945278168,
      "learning_rate": 5.004194096666609e-06,
      "loss": 0.055,
      "mean_token_accuracy": 0.9808351457118988,
      "num_tokens": 225079413.0,
      "step": 1125
    },
    {
      "epoch": 0.9179528838342811,
      "grad_norm": 0.31730109453201294,
      "learning_rate": 4.979127600530779e-06,
      "loss": 0.0585,
      "mean_token_accuracy": 0.9803479611873627,
      "num_tokens": 226078527.0,
      "step": 1130
    }
  ],
  "logging_steps": 5,
  "max_steps": 1230,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.4474025309185245e+19,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}