{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9988023952095808,
  "eval_steps": 500,
  "global_step": 834,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.005988023952095809,
      "grad_norm": 20.0459041595459,
      "learning_rate": 7.904761904761904e-06,
      "loss": 2.0221,
      "mean_token_accuracy": 0.5558732986450196,
      "num_tokens": 366628.0,
      "step": 5
    },
    {
      "epoch": 0.011976047904191617,
      "grad_norm": 2.7792041301727295,
      "learning_rate": 1.7785714285714282e-05,
      "loss": 0.4994,
      "mean_token_accuracy": 0.8186076283454895,
      "num_tokens": 759222.0,
      "step": 10
    },
    {
      "epoch": 0.017964071856287425,
      "grad_norm": 0.8423223495483398,
      "learning_rate": 2.7666666666666667e-05,
      "loss": 0.1783,
      "mean_token_accuracy": 0.9357934117317199,
      "num_tokens": 1136489.0,
      "step": 15
    },
    {
      "epoch": 0.023952095808383235,
      "grad_norm": 0.6088367700576782,
      "learning_rate": 3.7547619047619045e-05,
      "loss": 0.1546,
      "mean_token_accuracy": 0.9425264596939087,
      "num_tokens": 1520542.0,
      "step": 20
    },
    {
      "epoch": 0.029940119760479042,
      "grad_norm": 0.5601025223731995,
      "learning_rate": 4.7428571428571427e-05,
      "loss": 0.1376,
      "mean_token_accuracy": 0.9515837907791138,
      "num_tokens": 1902896.0,
      "step": 25
    },
    {
      "epoch": 0.03592814371257485,
      "grad_norm": 0.6080322265625,
      "learning_rate": 5.730952380952381e-05,
      "loss": 0.1272,
      "mean_token_accuracy": 0.9530677318572998,
      "num_tokens": 2291074.0,
      "step": 30
    },
    {
      "epoch": 0.041916167664670656,
      "grad_norm": 0.7478774785995483,
      "learning_rate": 6.719047619047618e-05,
      "loss": 0.1116,
      "mean_token_accuracy": 0.9604823112487793,
      "num_tokens": 2698749.0,
      "step": 35
    },
    {
      "epoch": 0.04790419161676647,
      "grad_norm": 0.439449667930603,
      "learning_rate": 6.916346682165974e-05,
      "loss": 0.1059,
      "mean_token_accuracy": 0.9638442039489746,
      "num_tokens": 3090835.0,
      "step": 40
    },
    {
      "epoch": 0.05389221556886228,
      "grad_norm": 0.42887377738952637,
      "learning_rate": 6.915046880442918e-05,
      "loss": 0.0961,
      "mean_token_accuracy": 0.9669005274772644,
      "num_tokens": 3479010.0,
      "step": 45
    },
    {
      "epoch": 0.059880239520958084,
      "grad_norm": 0.7219209671020508,
      "learning_rate": 6.912747763179e-05,
      "loss": 0.0788,
      "mean_token_accuracy": 0.9721761226654053,
      "num_tokens": 3867230.0,
      "step": 50
    },
    {
      "epoch": 0.0658682634730539,
      "grad_norm": 0.32698893547058105,
      "learning_rate": 6.909450216727685e-05,
      "loss": 0.0837,
      "mean_token_accuracy": 0.9730605244636535,
      "num_tokens": 4260989.0,
      "step": 55
    },
    {
      "epoch": 0.0718562874251497,
      "grad_norm": 0.6246018409729004,
      "learning_rate": 6.905155512355905e-05,
      "loss": 0.0725,
      "mean_token_accuracy": 0.9747663378715515,
      "num_tokens": 4642635.0,
      "step": 60
    },
    {
      "epoch": 0.07784431137724551,
      "grad_norm": 0.7247958183288574,
      "learning_rate": 6.899865305753948e-05,
      "loss": 0.0608,
      "mean_token_accuracy": 0.9827869415283204,
      "num_tokens": 5033119.0,
      "step": 65
    },
    {
      "epoch": 0.08383233532934131,
      "grad_norm": 0.40797632932662964,
      "learning_rate": 6.893581636397164e-05,
      "loss": 0.0686,
      "mean_token_accuracy": 0.9763871550559997,
      "num_tokens": 5426014.0,
      "step": 70
    },
    {
      "epoch": 0.08982035928143713,
      "grad_norm": 0.7394274473190308,
      "learning_rate": 6.886306926759702e-05,
      "loss": 0.0562,
      "mean_token_accuracy": 0.981475830078125,
      "num_tokens": 5813839.0,
      "step": 75
    },
    {
      "epoch": 0.09580838323353294,
      "grad_norm": 0.3124346435070038,
      "learning_rate": 6.87804398138061e-05,
      "loss": 0.0529,
      "mean_token_accuracy": 0.983836007118225,
      "num_tokens": 6200417.0,
      "step": 80
    },
    {
      "epoch": 0.10179640718562874,
      "grad_norm": 0.33730122447013855,
      "learning_rate": 6.868795985782625e-05,
      "loss": 0.064,
      "mean_token_accuracy": 0.9803187370300293,
      "num_tokens": 6576824.0,
      "step": 85
    },
    {
      "epoch": 0.10778443113772455,
      "grad_norm": 0.30790194869041443,
      "learning_rate": 6.858566505244093e-05,
      "loss": 0.0643,
      "mean_token_accuracy": 0.9791940450668335,
      "num_tokens": 6969822.0,
      "step": 90
    },
    {
      "epoch": 0.11377245508982035,
      "grad_norm": 0.25939929485321045,
      "learning_rate": 6.847359483424486e-05,
      "loss": 0.054,
      "mean_token_accuracy": 0.981829845905304,
      "num_tokens": 7328084.0,
      "step": 95
    },
    {
      "epoch": 0.11976047904191617,
      "grad_norm": 0.34844934940338135,
      "learning_rate": 6.835179240844054e-05,
      "loss": 0.0563,
      "mean_token_accuracy": 0.981998085975647,
      "num_tokens": 7691316.0,
      "step": 100
    },
    {
      "epoch": 0.12574850299401197,
      "grad_norm": 0.35250937938690186,
      "learning_rate": 6.82203047321817e-05,
      "loss": 0.0549,
      "mean_token_accuracy": 0.9820721864700317,
      "num_tokens": 8071906.0,
      "step": 105
    },
    {
      "epoch": 0.1317365269461078,
      "grad_norm": 0.2580828368663788,
      "learning_rate": 6.807918249647057e-05,
      "loss": 0.0524,
      "mean_token_accuracy": 0.9835985541343689,
      "num_tokens": 8467039.0,
      "step": 110
    },
    {
      "epoch": 0.1377245508982036,
      "grad_norm": 0.3509603440761566,
      "learning_rate": 6.792848010661536e-05,
      "loss": 0.0437,
      "mean_token_accuracy": 0.9859248995780945,
      "num_tokens": 8857358.0,
      "step": 115
    },
    {
      "epoch": 0.1437125748502994,
      "grad_norm": 0.20500952005386353,
      "learning_rate": 6.776825566125611e-05,
      "loss": 0.0472,
      "mean_token_accuracy": 0.9837791562080384,
      "num_tokens": 9234411.0,
      "step": 120
    },
    {
      "epoch": 0.1497005988023952,
      "grad_norm": 0.33302396535873413,
      "learning_rate": 6.759857092996655e-05,
      "loss": 0.0501,
      "mean_token_accuracy": 0.9837584733963013,
      "num_tokens": 9635726.0,
      "step": 125
    },
    {
      "epoch": 0.15568862275449102,
      "grad_norm": 0.36120888590812683,
      "learning_rate": 6.741949132944066e-05,
      "loss": 0.0477,
      "mean_token_accuracy": 0.9840852737426757,
      "num_tokens": 10007520.0,
      "step": 130
    },
    {
      "epoch": 0.16167664670658682,
      "grad_norm": 0.2940848469734192,
      "learning_rate": 6.72310858982734e-05,
      "loss": 0.0453,
      "mean_token_accuracy": 0.985532033443451,
      "num_tokens": 10382146.0,
      "step": 135
    },
    {
      "epoch": 0.16766467065868262,
      "grad_norm": 0.37795165181159973,
      "learning_rate": 6.703342727034491e-05,
      "loss": 0.0498,
      "mean_token_accuracy": 0.9846987366676331,
      "num_tokens": 10761504.0,
      "step": 140
    },
    {
      "epoch": 0.17365269461077845,
      "grad_norm": 0.2522616684436798,
      "learning_rate": 6.682659164681891e-05,
      "loss": 0.039,
      "mean_token_accuracy": 0.9853670597076416,
      "num_tokens": 11135398.0,
      "step": 145
    },
    {
      "epoch": 0.17964071856287425,
      "grad_norm": 0.3395073413848877,
      "learning_rate": 6.661065876676553e-05,
      "loss": 0.0428,
      "mean_token_accuracy": 0.9845033049583435,
      "num_tokens": 11511360.0,
      "step": 150
    },
    {
      "epoch": 0.18562874251497005,
      "grad_norm": 0.2507222294807434,
      "learning_rate": 6.638571187642052e-05,
      "loss": 0.0382,
      "mean_token_accuracy": 0.9886868715286254,
      "num_tokens": 11890265.0,
      "step": 155
    },
    {
      "epoch": 0.19161676646706588,
      "grad_norm": 0.3042142689228058,
      "learning_rate": 6.615183769709215e-05,
      "loss": 0.0349,
      "mean_token_accuracy": 0.9871665239334106,
      "num_tokens": 12254628.0,
      "step": 160
    },
    {
      "epoch": 0.19760479041916168,
      "grad_norm": 0.3100643754005432,
      "learning_rate": 6.590912639172857e-05,
      "loss": 0.037,
      "mean_token_accuracy": 0.9894456148147583,
      "num_tokens": 12644098.0,
      "step": 165
    },
    {
      "epoch": 0.20359281437125748,
      "grad_norm": 0.322879433631897,
      "learning_rate": 6.565767153015818e-05,
      "loss": 0.036,
      "mean_token_accuracy": 0.9883933186531066,
      "num_tokens": 13037276.0,
      "step": 170
    },
    {
      "epoch": 0.20958083832335328,
      "grad_norm": 0.21397611498832703,
      "learning_rate": 6.539757005301683e-05,
      "loss": 0.0397,
      "mean_token_accuracy": 0.9874719500541687,
      "num_tokens": 13432477.0,
      "step": 175
    },
    {
      "epoch": 0.2155688622754491,
      "grad_norm": 0.3262614607810974,
      "learning_rate": 6.512892223437517e-05,
      "loss": 0.038,
      "mean_token_accuracy": 0.9862341403961181,
      "num_tokens": 13828230.0,
      "step": 180
    },
    {
      "epoch": 0.2215568862275449,
      "grad_norm": 0.3778105080127716,
      "learning_rate": 6.485183164308128e-05,
      "loss": 0.0422,
      "mean_token_accuracy": 0.9874172449111939,
      "num_tokens": 14212298.0,
      "step": 185
    },
    {
      "epoch": 0.2275449101796407,
      "grad_norm": 0.35106056928634644,
      "learning_rate": 6.456640510283271e-05,
      "loss": 0.0373,
      "mean_token_accuracy": 0.9877774953842163,
      "num_tokens": 14593164.0,
      "step": 190
    },
    {
      "epoch": 0.23353293413173654,
      "grad_norm": 0.3846699297428131,
      "learning_rate": 6.427275265099411e-05,
      "loss": 0.0381,
      "mean_token_accuracy": 0.9873203754425048,
      "num_tokens": 14966764.0,
      "step": 195
    },
    {
      "epoch": 0.23952095808383234,
      "grad_norm": 0.3400399386882782,
      "learning_rate": 6.397098749617558e-05,
      "loss": 0.037,
      "mean_token_accuracy": 0.9876260995864868,
      "num_tokens": 15337481.0,
      "step": 200
    },
    {
      "epoch": 0.24550898203592814,
      "grad_norm": 0.19366240501403809,
      "learning_rate": 6.366122597458865e-05,
      "loss": 0.0431,
      "mean_token_accuracy": 0.9846744656562805,
      "num_tokens": 15729039.0,
      "step": 205
    },
    {
      "epoch": 0.25149700598802394,
      "grad_norm": 0.12042815238237381,
      "learning_rate": 6.334358750519649e-05,
      "loss": 0.0386,
      "mean_token_accuracy": 0.98653324842453,
      "num_tokens": 16125147.0,
      "step": 210
    },
    {
      "epoch": 0.25748502994011974,
      "grad_norm": 0.1692809909582138,
      "learning_rate": 6.301819454367557e-05,
      "loss": 0.026,
      "mean_token_accuracy": 0.9903317451477051,
      "num_tokens": 16509303.0,
      "step": 215
    },
    {
      "epoch": 0.2634730538922156,
      "grad_norm": 0.17025096714496613,
      "learning_rate": 6.26851725352067e-05,
      "loss": 0.0268,
      "mean_token_accuracy": 0.9908241868019104,
      "num_tokens": 16888245.0,
      "step": 220
    },
    {
      "epoch": 0.2694610778443114,
      "grad_norm": 0.17811234295368195,
      "learning_rate": 6.234464986611352e-05,
      "loss": 0.028,
      "mean_token_accuracy": 0.9904983758926391,
      "num_tokens": 17271305.0,
      "step": 225
    },
    {
      "epoch": 0.2754491017964072,
      "grad_norm": 0.2255362719297409,
      "learning_rate": 6.199675781436707e-05,
      "loss": 0.0228,
      "mean_token_accuracy": 0.9931029319763184,
      "num_tokens": 17667173.0,
      "step": 230
    },
    {
      "epoch": 0.281437125748503,
      "grad_norm": 0.365841805934906,
      "learning_rate": 6.164163049897576e-05,
      "loss": 0.0354,
      "mean_token_accuracy": 0.9877443432807922,
      "num_tokens": 18045084.0,
      "step": 235
    },
    {
      "epoch": 0.2874251497005988,
      "grad_norm": 0.2448989450931549,
      "learning_rate": 6.127940482827982e-05,
      "loss": 0.0365,
      "mean_token_accuracy": 0.9870758295059204,
      "num_tokens": 18446090.0,
      "step": 240
    },
    {
      "epoch": 0.2934131736526946,
      "grad_norm": 0.3074445426464081,
      "learning_rate": 6.091022044717059e-05,
      "loss": 0.0332,
      "mean_token_accuracy": 0.9892607688903808,
      "num_tokens": 18824322.0,
      "step": 245
    },
    {
      "epoch": 0.2994011976047904,
      "grad_norm": 0.2015693336725235,
      "learning_rate": 6.053421968325472e-05,
      "loss": 0.0366,
      "mean_token_accuracy": 0.9873138904571533,
      "num_tokens": 19206557.0,
      "step": 250
    },
    {
      "epoch": 0.30538922155688625,
      "grad_norm": 0.16197246313095093,
      "learning_rate": 6.015154749198421e-05,
      "loss": 0.0332,
      "mean_token_accuracy": 0.9898733973503113,
      "num_tokens": 19585158.0,
      "step": 255
    },
    {
      "epoch": 0.31137724550898205,
      "grad_norm": 0.23897434771060944,
      "learning_rate": 5.97623514007733e-05,
      "loss": 0.0217,
      "mean_token_accuracy": 0.9924596428871155,
      "num_tokens": 19963252.0,
      "step": 260
    },
    {
      "epoch": 0.31736526946107785,
      "grad_norm": 0.29277706146240234,
      "learning_rate": 5.9366781452123836e-05,
      "loss": 0.028,
      "mean_token_accuracy": 0.9900582790374756,
      "num_tokens": 20349149.0,
      "step": 265
    },
    {
      "epoch": 0.32335329341317365,
      "grad_norm": 0.28648537397384644,
      "learning_rate": 5.8964990145781026e-05,
      "loss": 0.0274,
      "mean_token_accuracy": 0.9911342620849609,
      "num_tokens": 20722385.0,
      "step": 270
    },
    {
      "epoch": 0.32934131736526945,
      "grad_norm": 0.19693732261657715,
      "learning_rate": 5.855713237994193e-05,
      "loss": 0.0326,
      "mean_token_accuracy": 0.9893397927284241,
      "num_tokens": 21095997.0,
      "step": 275
    },
    {
      "epoch": 0.33532934131736525,
      "grad_norm": 0.17976205050945282,
      "learning_rate": 5.8143365391539205e-05,
      "loss": 0.0308,
      "mean_token_accuracy": 0.9905115008354187,
      "num_tokens": 21472617.0,
      "step": 280
    },
    {
      "epoch": 0.3413173652694611,
      "grad_norm": 0.18820658326148987,
      "learning_rate": 5.7723848695623276e-05,
      "loss": 0.0243,
      "mean_token_accuracy": 0.9919281005859375,
      "num_tokens": 21845638.0,
      "step": 285
    },
    {
      "epoch": 0.3473053892215569,
      "grad_norm": 0.21537180244922638,
      "learning_rate": 5.7298744023866306e-05,
      "loss": 0.0215,
      "mean_token_accuracy": 0.994717252254486,
      "num_tokens": 22212673.0,
      "step": 290
    },
    {
      "epoch": 0.3532934131736527,
      "grad_norm": 0.17392006516456604,
      "learning_rate": 5.686821526221148e-05,
      "loss": 0.0291,
      "mean_token_accuracy": 0.9916199445724487,
      "num_tokens": 22607908.0,
      "step": 295
    },
    {
      "epoch": 0.3592814371257485,
      "grad_norm": 0.20204167068004608,
      "learning_rate": 5.643242838769192e-05,
      "loss": 0.0285,
      "mean_token_accuracy": 0.9906000375747681,
      "num_tokens": 22982387.0,
      "step": 300
    },
    {
      "epoch": 0.3652694610778443,
      "grad_norm": 0.1749100536108017,
      "learning_rate": 5.5991551404443325e-05,
      "loss": 0.0226,
      "mean_token_accuracy": 0.9934131622314453,
      "num_tokens": 23372699.0,
      "step": 305
    },
    {
      "epoch": 0.3712574850299401,
      "grad_norm": 0.23637597262859344,
      "learning_rate": 5.554575427893521e-05,
      "loss": 0.0282,
      "mean_token_accuracy": 0.9920474648475647,
      "num_tokens": 23773625.0,
      "step": 310
    },
    {
      "epoch": 0.3772455089820359,
      "grad_norm": 0.22615177929401398,
      "learning_rate": 5.50952088744456e-05,
      "loss": 0.0184,
      "mean_token_accuracy": 0.9927815079689026,
      "num_tokens": 24146851.0,
      "step": 315
    },
    {
      "epoch": 0.38323353293413176,
      "grad_norm": 0.19745571911334991,
      "learning_rate": 5.464008888480446e-05,
      "loss": 0.0268,
      "mean_token_accuracy": 0.9912487864494324,
      "num_tokens": 24529743.0,
      "step": 320
    },
    {
      "epoch": 0.38922155688622756,
      "grad_norm": 0.19386577606201172,
      "learning_rate": 5.418056976743139e-05,
      "loss": 0.0244,
      "mean_token_accuracy": 0.9926334738731384,
      "num_tokens": 24934687.0,
      "step": 325
    },
    {
      "epoch": 0.39520958083832336,
      "grad_norm": 0.27069705724716187,
      "learning_rate": 5.3716828675693487e-05,
      "loss": 0.0225,
      "mean_token_accuracy": 0.992877197265625,
      "num_tokens": 25306420.0,
      "step": 330
    },
    {
      "epoch": 0.40119760479041916,
      "grad_norm": 0.21653445065021515,
      "learning_rate": 5.324904439060933e-05,
      "loss": 0.0256,
      "mean_token_accuracy": 0.992726719379425,
      "num_tokens": 25682865.0,
      "step": 335
    },
    {
      "epoch": 0.40718562874251496,
      "grad_norm": 0.19806110858917236,
      "learning_rate": 5.27773972519255e-05,
      "loss": 0.0204,
      "mean_token_accuracy": 0.9935464262962341,
      "num_tokens": 26055096.0,
      "step": 340
    },
    {
      "epoch": 0.41317365269461076,
      "grad_norm": 0.13250744342803955,
      "learning_rate": 5.2302069088592176e-05,
      "loss": 0.0218,
      "mean_token_accuracy": 0.9927313923835754,
      "num_tokens": 26445726.0,
      "step": 345
    },
    {
      "epoch": 0.41916167664670656,
      "grad_norm": 0.26525649428367615,
      "learning_rate": 5.1823243148664595e-05,
      "loss": 0.0197,
      "mean_token_accuracy": 0.9940995812416077,
      "num_tokens": 26823023.0,
      "step": 350
    },
    {
      "epoch": 0.4251497005988024,
      "grad_norm": 0.4039984941482544,
      "learning_rate": 5.1341104028657496e-05,
      "loss": 0.022,
      "mean_token_accuracy": 0.9927559137344361,
      "num_tokens": 27213794.0,
      "step": 355
    },
    {
      "epoch": 0.4311377245508982,
      "grad_norm": 0.2599278688430786,
      "learning_rate": 5.0855837602379545e-05,
      "loss": 0.0224,
      "mean_token_accuracy": 0.992340886592865,
      "num_tokens": 27604116.0,
      "step": 360
    },
    {
      "epoch": 0.437125748502994,
      "grad_norm": 0.17638690769672394,
      "learning_rate": 5.0367630949275574e-05,
      "loss": 0.0272,
      "mean_token_accuracy": 0.991790771484375,
      "num_tokens": 27968846.0,
      "step": 365
    },
    {
      "epoch": 0.4431137724550898,
      "grad_norm": 0.12005963176488876,
      "learning_rate": 4.987667228230376e-05,
      "loss": 0.0273,
      "mean_token_accuracy": 0.9912943959236145,
      "num_tokens": 28339114.0,
      "step": 370
    },
    {
      "epoch": 0.4491017964071856,
      "grad_norm": 0.24598518013954163,
      "learning_rate": 4.938315087537608e-05,
      "loss": 0.0274,
      "mean_token_accuracy": 0.9911744117736816,
      "num_tokens": 28737903.0,
      "step": 375
    },
    {
      "epoch": 0.4550898203592814,
      "grad_norm": 0.11921826750040054,
      "learning_rate": 4.888725699038949e-05,
      "loss": 0.0243,
      "mean_token_accuracy": 0.9933557271957397,
      "num_tokens": 29126414.0,
      "step": 380
    },
    {
      "epoch": 0.46107784431137727,
      "grad_norm": 0.2723204791545868,
      "learning_rate": 4.838918180387643e-05,
      "loss": 0.03,
      "mean_token_accuracy": 0.9915509223937988,
      "num_tokens": 29525652.0,
      "step": 385
    },
    {
      "epoch": 0.46706586826347307,
      "grad_norm": 0.14287085831165314,
      "learning_rate": 4.7889117333302594e-05,
      "loss": 0.0181,
      "mean_token_accuracy": 0.9940540671348572,
      "num_tokens": 29921040.0,
      "step": 390
    },
    {
      "epoch": 0.47305389221556887,
      "grad_norm": 0.2439376562833786,
      "learning_rate": 4.738725636304056e-05,
      "loss": 0.0221,
      "mean_token_accuracy": 0.9937701940536499,
      "num_tokens": 30313012.0,
      "step": 395
    },
    {
      "epoch": 0.47904191616766467,
      "grad_norm": 0.2295493334531784,
      "learning_rate": 4.688379237004771e-05,
      "loss": 0.0247,
      "mean_token_accuracy": 0.9914366722106933,
      "num_tokens": 30686119.0,
      "step": 400
    },
    {
      "epoch": 0.48502994011976047,
      "grad_norm": 0.16021516919136047,
      "learning_rate": 4.637891944927724e-05,
      "loss": 0.0215,
      "mean_token_accuracy": 0.9929421782493592,
      "num_tokens": 31066195.0,
      "step": 405
    },
    {
      "epoch": 0.49101796407185627,
      "grad_norm": 0.20455706119537354,
      "learning_rate": 4.5872832238850775e-05,
      "loss": 0.0192,
      "mean_token_accuracy": 0.9942737460136414,
      "num_tokens": 31438081.0,
      "step": 410
    },
    {
      "epoch": 0.49700598802395207,
      "grad_norm": 0.2634243667125702,
      "learning_rate": 4.536572584502176e-05,
      "loss": 0.0164,
      "mean_token_accuracy": 0.9939553380012512,
      "num_tokens": 31836583.0,
      "step": 415
    },
    {
      "epoch": 0.5029940119760479,
      "grad_norm": 0.25241556763648987,
      "learning_rate": 4.485779576695823e-05,
      "loss": 0.0267,
      "mean_token_accuracy": 0.9913934826850891,
      "num_tokens": 32218980.0,
      "step": 420
    },
    {
      "epoch": 0.5089820359281437,
      "grad_norm": 0.12369759380817413,
      "learning_rate": 4.4349237821374164e-05,
      "loss": 0.0212,
      "mean_token_accuracy": 0.9931016206741333,
      "num_tokens": 32595667.0,
      "step": 425
    },
    {
      "epoch": 0.5149700598802395,
      "grad_norm": 0.2500801682472229,
      "learning_rate": 4.3840248067038435e-05,
      "loss": 0.0181,
      "mean_token_accuracy": 0.995021378993988,
      "num_tokens": 32999092.0,
      "step": 430
    },
    {
      "epoch": 0.5209580838323353,
      "grad_norm": 0.1701110452413559,
      "learning_rate": 4.3331022729190385e-05,
      "loss": 0.0221,
      "mean_token_accuracy": 0.9927478671073914,
      "num_tokens": 33394580.0,
      "step": 435
    },
    {
      "epoch": 0.5269461077844312,
      "grad_norm": 0.15505260229110718,
      "learning_rate": 4.282175812389132e-05,
      "loss": 0.0203,
      "mean_token_accuracy": 0.9935660481452941,
      "num_tokens": 33783326.0,
      "step": 440
    },
    {
      "epoch": 0.5329341317365269,
      "grad_norm": 0.1670973300933838,
      "learning_rate": 4.2312650582340866e-05,
      "loss": 0.0294,
      "mean_token_accuracy": 0.9918290734291076,
      "num_tokens": 34178239.0,
      "step": 445
    },
    {
      "epoch": 0.5389221556886228,
      "grad_norm": 0.1728363186120987,
      "learning_rate": 4.18038963751876e-05,
      "loss": 0.0216,
      "mean_token_accuracy": 0.9923473358154297,
      "num_tokens": 34564304.0,
      "step": 450
    },
    {
      "epoch": 0.5449101796407185,
      "grad_norm": 0.18342676758766174,
      "learning_rate": 4.1295691636862965e-05,
      "loss": 0.024,
      "mean_token_accuracy": 0.9924877882003784,
      "num_tokens": 34941941.0,
      "step": 455
    },
    {
      "epoch": 0.5508982035928144,
      "grad_norm": 0.23771464824676514,
      "learning_rate": 4.07882322899677e-05,
      "loss": 0.022,
      "mean_token_accuracy": 0.993938660621643,
      "num_tokens": 35333400.0,
      "step": 460
    },
    {
      "epoch": 0.5568862275449101,
      "grad_norm": 0.1375587284564972,
      "learning_rate": 4.0281713969739954e-05,
      "loss": 0.0167,
      "mean_token_accuracy": 0.9948888421058655,
      "num_tokens": 35704725.0,
      "step": 465
    },
    {
      "epoch": 0.562874251497006,
      "grad_norm": 0.20477965474128723,
      "learning_rate": 3.9776331948634216e-05,
      "loss": 0.0207,
      "mean_token_accuracy": 0.992275619506836,
      "num_tokens": 36082900.0,
      "step": 470
    },
    {
      "epoch": 0.5688622754491018,
      "grad_norm": 0.15638668835163116,
      "learning_rate": 3.927228106103999e-05,
      "loss": 0.0226,
      "mean_token_accuracy": 0.9923402309417725,
      "num_tokens": 36458521.0,
      "step": 475
    },
    {
      "epoch": 0.5748502994011976,
      "grad_norm": 0.18565590679645538,
      "learning_rate": 3.876975562816948e-05,
      "loss": 0.0203,
      "mean_token_accuracy": 0.9925201058387756,
      "num_tokens": 36841501.0,
      "step": 480
    },
    {
      "epoch": 0.5808383233532934,
      "grad_norm": 0.2210000902414322,
      "learning_rate": 3.8268949383143085e-05,
      "loss": 0.0263,
      "mean_token_accuracy": 0.9906040310859681,
      "num_tokens": 37232809.0,
      "step": 485
    },
    {
      "epoch": 0.5868263473053892,
      "grad_norm": 0.20876674354076385,
      "learning_rate": 3.777005539630155e-05,
      "loss": 0.0209,
      "mean_token_accuracy": 0.9923003077507019,
      "num_tokens": 37611412.0,
      "step": 490
    },
    {
      "epoch": 0.592814371257485,
      "grad_norm": 0.163838192820549,
      "learning_rate": 3.727326600077367e-05,
      "loss": 0.0194,
      "mean_token_accuracy": 0.9930386304855346,
      "num_tokens": 37986228.0,
      "step": 495
    },
    {
      "epoch": 0.5988023952095808,
      "grad_norm": 0.2008194774389267,
      "learning_rate": 3.677877271832825e-05,
      "loss": 0.0234,
      "mean_token_accuracy": 0.9930065631866455,
      "num_tokens": 38380451.0,
      "step": 500
    },
    {
      "epoch": 0.6047904191616766,
      "grad_norm": 0.14270946383476257,
      "learning_rate": 3.628676618553876e-05,
      "loss": 0.0225,
      "mean_token_accuracy": 0.9921725034713745,
      "num_tokens": 38762838.0,
      "step": 505
    },
    {
      "epoch": 0.6107784431137725,
      "grad_norm": 0.18238236010074615,
      "learning_rate": 3.5797436080289395e-05,
      "loss": 0.0172,
      "mean_token_accuracy": 0.9943962931632996,
      "num_tokens": 39147058.0,
      "step": 510
    },
    {
      "epoch": 0.6167664670658682,
      "grad_norm": 0.39257997274398804,
      "learning_rate": 3.531097104865054e-05,
      "loss": 0.016,
      "mean_token_accuracy": 0.9946008443832397,
      "num_tokens": 39528886.0,
      "step": 515
    },
    {
      "epoch": 0.6227544910179641,
      "grad_norm": 0.21363849937915802,
      "learning_rate": 3.482755863215217e-05,
      "loss": 0.0181,
      "mean_token_accuracy": 0.9943473219871521,
      "num_tokens": 39905988.0,
      "step": 520
    },
    {
      "epoch": 0.6287425149700598,
      "grad_norm": 0.13765949010849,
      "learning_rate": 3.434738519548303e-05,
      "loss": 0.0181,
      "mean_token_accuracy": 0.9954151630401611,
      "num_tokens": 40302048.0,
      "step": 525
    },
    {
      "epoch": 0.6347305389221557,
      "grad_norm": 0.27964115142822266,
      "learning_rate": 3.387063585464345e-05,
      "loss": 0.0153,
      "mean_token_accuracy": 0.9943761229515076,
      "num_tokens": 40679444.0,
      "step": 530
    },
    {
      "epoch": 0.6407185628742516,
      "grad_norm": 0.12370190769433975,
      "learning_rate": 3.339749440557971e-05,
      "loss": 0.0225,
      "mean_token_accuracy": 0.9921257019042968,
      "num_tokens": 41053135.0,
      "step": 535
    },
    {
      "epoch": 0.6467065868263473,
      "grad_norm": 0.1074637621641159,
      "learning_rate": 3.292814325332705e-05,
      "loss": 0.0144,
      "mean_token_accuracy": 0.9947722315788269,
      "num_tokens": 41444077.0,
      "step": 540
    },
    {
      "epoch": 0.6526946107784432,
      "grad_norm": 0.17561455070972443,
      "learning_rate": 3.246276334168917e-05,
      "loss": 0.0191,
      "mean_token_accuracy": 0.9938403844833374,
      "num_tokens": 41811115.0,
      "step": 545
    },
    {
      "epoch": 0.6586826347305389,
      "grad_norm": 0.21135373413562775,
      "learning_rate": 3.200153408348094e-05,
      "loss": 0.0216,
      "mean_token_accuracy": 0.9925486445426941,
      "num_tokens": 42200440.0,
      "step": 550
    },
    {
      "epoch": 0.6646706586826348,
      "grad_norm": 0.16747206449508667,
      "learning_rate": 3.154463329136126e-05,
      "loss": 0.0239,
      "mean_token_accuracy": 0.9930933713912964,
      "num_tokens": 42597406.0,
      "step": 555
    },
    {
      "epoch": 0.6706586826347305,
      "grad_norm": 0.10462242364883423,
      "learning_rate": 3.109223710928305e-05,
      "loss": 0.0181,
      "mean_token_accuracy": 0.9946050405502319,
      "num_tokens": 42965424.0,
      "step": 560
    },
    {
      "epoch": 0.6766467065868264,
      "grad_norm": 0.1582065373659134,
      "learning_rate": 3.064451994458621e-05,
      "loss": 0.0142,
      "mean_token_accuracy": 0.9950128436088562,
      "num_tokens": 43331718.0,
      "step": 565
    },
    {
      "epoch": 0.6826347305389222,
      "grad_norm": 0.18342357873916626,
      "learning_rate": 3.020165440076044e-05,
      "loss": 0.0147,
      "mean_token_accuracy": 0.994437038898468,
      "num_tokens": 43705052.0,
      "step": 570
    },
    {
      "epoch": 0.688622754491018,
      "grad_norm": 0.20606689155101776,
      "learning_rate": 2.976381121090322e-05,
      "loss": 0.0155,
      "mean_token_accuracy": 0.9951949715614319,
      "num_tokens": 44102415.0,
      "step": 575
    },
    {
      "epoch": 0.6946107784431138,
      "grad_norm": 0.36050134897232056,
      "learning_rate": 2.9331159171898936e-05,
      "loss": 0.014,
      "mean_token_accuracy": 0.995421290397644,
      "num_tokens": 44482224.0,
      "step": 580
    },
    {
      "epoch": 0.7005988023952096,
      "grad_norm": 0.2575097680091858,
      "learning_rate": 2.890386507934457e-05,
      "loss": 0.0207,
      "mean_token_accuracy": 0.9926421880722046,
      "num_tokens": 44884852.0,
      "step": 585
    },
    {
      "epoch": 0.7065868263473054,
      "grad_norm": 0.1434570848941803,
      "learning_rate": 2.8482093663246693e-05,
      "loss": 0.0131,
      "mean_token_accuracy": 0.995238995552063,
      "num_tokens": 45278732.0,
      "step": 590
    },
    {
      "epoch": 0.7125748502994012,
      "grad_norm": 0.12609131634235382,
      "learning_rate": 2.8066007524515013e-05,
      "loss": 0.0155,
      "mean_token_accuracy": 0.9951435565948487,
      "num_tokens": 45666594.0,
      "step": 595
    },
    {
      "epoch": 0.718562874251497,
      "grad_norm": 0.174350768327713,
      "learning_rate": 2.7655767072276552e-05,
      "loss": 0.0142,
      "mean_token_accuracy": 0.9950149536132813,
      "num_tokens": 46059546.0,
      "step": 600
    },
    {
      "epoch": 0.7245508982035929,
      "grad_norm": 0.14378827810287476,
      "learning_rate": 2.725153046203494e-05,
      "loss": 0.019,
      "mean_token_accuracy": 0.9934304833412171,
      "num_tokens": 46466567.0,
      "step": 605
    },
    {
      "epoch": 0.7305389221556886,
      "grad_norm": 0.14289937913417816,
      "learning_rate": 2.685345353469843e-05,
      "loss": 0.0116,
      "mean_token_accuracy": 0.9964966535568237,
      "num_tokens": 46868301.0,
      "step": 610
    },
    {
      "epoch": 0.7365269461077845,
      "grad_norm": 0.2372882068157196,
      "learning_rate": 2.646168975650025e-05,
      "loss": 0.0192,
      "mean_token_accuracy": 0.9930261492729187,
      "num_tokens": 47259594.0,
      "step": 615
    },
    {
      "epoch": 0.7425149700598802,
      "grad_norm": 0.09207688271999359,
      "learning_rate": 2.607639015983445e-05,
      "loss": 0.0125,
      "mean_token_accuracy": 0.9965265989303589,
      "num_tokens": 47621984.0,
      "step": 620
    },
    {
      "epoch": 0.7485029940119761,
      "grad_norm": 0.16538254916667938,
      "learning_rate": 2.569770328503006e-05,
      "loss": 0.0145,
      "mean_token_accuracy": 0.9955265760421753,
      "num_tokens": 48003750.0,
      "step": 625
    },
    {
      "epoch": 0.7544910179640718,
      "grad_norm": 0.17553557455539703,
      "learning_rate": 2.532577512308595e-05,
      "loss": 0.0155,
      "mean_token_accuracy": 0.9939799427986145,
      "num_tokens": 48394172.0,
      "step": 630
    },
    {
      "epoch": 0.7604790419161677,
      "grad_norm": 0.18478022515773773,
      "learning_rate": 2.4960749059388533e-05,
      "loss": 0.0153,
      "mean_token_accuracy": 0.9951601982116699,
      "num_tokens": 48784517.0,
      "step": 635
    },
    {
      "epoch": 0.7664670658682635,
      "grad_norm": 0.11673842370510101,
      "learning_rate": 2.4602765818433956e-05,
      "loss": 0.0152,
      "mean_token_accuracy": 0.9948717355728149,
      "num_tokens": 49172953.0,
      "step": 640
    },
    {
      "epoch": 0.7724550898203593,
      "grad_norm": 0.26270660758018494,
      "learning_rate": 2.4251963409576117e-05,
      "loss": 0.0155,
      "mean_token_accuracy": 0.9952125310897827,
      "num_tokens": 49556262.0,
      "step": 645
    },
    {
      "epoch": 0.7784431137724551,
      "grad_norm": 0.15634196996688843,
      "learning_rate": 2.3908477073821473e-05,
      "loss": 0.0158,
      "mean_token_accuracy": 0.99416583776474,
      "num_tokens": 49937579.0,
      "step": 650
    },
    {
      "epoch": 0.7844311377245509,
      "grad_norm": 0.13335254788398743,
      "learning_rate": 2.3572439231691e-05,
      "loss": 0.0145,
      "mean_token_accuracy": 0.9946424961090088,
      "num_tokens": 50313593.0,
      "step": 655
    },
    {
      "epoch": 0.7904191616766467,
      "grad_norm": 0.2822599709033966,
      "learning_rate": 2.3243979432169632e-05,
      "loss": 0.0176,
      "mean_token_accuracy": 0.9955491185188293,
      "num_tokens": 50685536.0,
      "step": 660
    },
    {
      "epoch": 0.7964071856287425,
      "grad_norm": 0.07648531347513199,
      "learning_rate": 2.2923224302762596e-05,
      "loss": 0.0133,
      "mean_token_accuracy": 0.9955049633979798,
      "num_tokens": 51090191.0,
      "step": 665
    },
    {
      "epoch": 0.8023952095808383,
      "grad_norm": 0.1840098351240158,
      "learning_rate": 2.2610297500678124e-05,
      "loss": 0.0122,
      "mean_token_accuracy": 0.996946370601654,
      "num_tokens": 51463229.0,
      "step": 670
    },
    {
      "epoch": 0.8083832335329342,
      "grad_norm": 0.1478729248046875,
      "learning_rate": 2.2305319665155268e-05,
      "loss": 0.0125,
      "mean_token_accuracy": 0.9957447528839112,
      "num_tokens": 51826253.0,
      "step": 675
    },
    {
      "epoch": 0.8143712574850299,
      "grad_norm": 0.18469828367233276,
      "learning_rate": 2.2008408370955114e-05,
      "loss": 0.0129,
      "mean_token_accuracy": 0.9962447047233581,
      "num_tokens": 52204145.0,
      "step": 680
    },
    {
      "epoch": 0.8203592814371258,
      "grad_norm": 0.3210107982158661,
      "learning_rate": 2.1719678083033577e-05,
      "loss": 0.0157,
      "mean_token_accuracy": 0.9948264122009277,
      "num_tokens": 52574030.0,
      "step": 685
    },
    {
      "epoch": 0.8263473053892215,
      "grad_norm": 0.1561156064271927,
      "learning_rate": 2.143924011241286e-05,
      "loss": 0.0121,
      "mean_token_accuracy": 0.9962893486022949,
      "num_tokens": 52951648.0,
      "step": 690
    },
    {
      "epoch": 0.8323353293413174,
      "grad_norm": 0.19974002242088318,
      "learning_rate": 2.1167202573269027e-05,
      "loss": 0.0121,
      "mean_token_accuracy": 0.9963161826133728,
      "num_tokens": 53349988.0,
      "step": 695
    },
    {
      "epoch": 0.8383233532934131,
      "grad_norm": 0.21415163576602936,
      "learning_rate": 2.090367034125188e-05,
      "loss": 0.0182,
      "mean_token_accuracy": 0.9934315085411072,
      "num_tokens": 53735179.0,
      "step": 700
    },
    {
      "epoch": 0.844311377245509,
      "grad_norm": 0.16792823374271393,
      "learning_rate": 2.0648745013053432e-05,
      "loss": 0.0176,
      "mean_token_accuracy": 0.9943677663803101,
      "num_tokens": 54128630.0,
      "step": 705
    },
    {
      "epoch": 0.8502994011976048,
      "grad_norm": 0.13437584042549133,
      "learning_rate": 2.040252486724043e-05,
      "loss": 0.0111,
      "mean_token_accuracy": 0.9968718528747559,
      "num_tokens": 54513191.0,
      "step": 710
    },
    {
      "epoch": 0.8562874251497006,
      "grad_norm": 0.16742588579654694,
      "learning_rate": 2.01651048263661e-05,
      "loss": 0.0156,
      "mean_token_accuracy": 0.9953626275062561,
      "num_tokens": 54893314.0,
      "step": 715
    },
    {
      "epoch": 0.8622754491017964,
      "grad_norm": 0.18785876035690308,
      "learning_rate": 1.9936576420375702e-05,
      "loss": 0.0121,
      "mean_token_accuracy": 0.9964333415031433,
      "num_tokens": 55271598.0,
      "step": 720
    },
    {
      "epoch": 0.8682634730538922,
      "grad_norm": 0.11614938825368881,
      "learning_rate": 1.9717027751319967e-05,
      "loss": 0.0112,
      "mean_token_accuracy": 0.9966801166534424,
      "num_tokens": 55667199.0,
      "step": 725
    },
    {
      "epoch": 0.874251497005988,
      "grad_norm": 0.20819735527038574,
      "learning_rate": 1.9506543459390126e-05,
      "loss": 0.0098,
      "mean_token_accuracy": 0.9964311361312866,
      "num_tokens": 56033335.0,
      "step": 730
    },
    {
      "epoch": 0.8802395209580839,
      "grad_norm": 0.33518239855766296,
      "learning_rate": 1.9305204690287468e-05,
      "loss": 0.0113,
      "mean_token_accuracy": 0.9957466006278992,
      "num_tokens": 56410804.0,
      "step": 735
    },
    {
      "epoch": 0.8862275449101796,
      "grad_norm": 0.20654872059822083,
      "learning_rate": 1.9113089063940147e-05,
      "loss": 0.0143,
      "mean_token_accuracy": 0.9959073305130005,
      "num_tokens": 56802987.0,
      "step": 740
    },
    {
      "epoch": 0.8922155688622755,
      "grad_norm": 0.35063081979751587,
      "learning_rate": 1.8930270644579197e-05,
      "loss": 0.0118,
      "mean_token_accuracy": 0.9965811729431152,
      "num_tokens": 57188693.0,
      "step": 745
    },
    {
      "epoch": 0.8982035928143712,
      "grad_norm": 0.31076228618621826,
      "learning_rate": 1.8756819912185395e-05,
      "loss": 0.0119,
      "mean_token_accuracy": 0.9958955526351929,
      "num_tokens": 57580049.0,
      "step": 750
    },
    {
      "epoch": 0.9041916167664671,
      "grad_norm": 0.19703641533851624,
      "learning_rate": 1.8592803735317864e-05,
      "loss": 0.0119,
      "mean_token_accuracy": 0.995406424999237,
      "num_tokens": 57944462.0,
      "step": 755
    },
    {
      "epoch": 0.9101796407185628,
      "grad_norm": 0.09438392519950867,
      "learning_rate": 1.843828534533498e-05,
      "loss": 0.0129,
      "mean_token_accuracy": 0.9964313268661499,
      "num_tokens": 58307275.0,
      "step": 760
    },
    {
      "epoch": 0.9161676646706587,
      "grad_norm": 0.1967303454875946,
      "learning_rate": 1.8293324312017467e-05,
      "loss": 0.0143,
      "mean_token_accuracy": 0.995577871799469,
      "num_tokens": 58688837.0,
      "step": 765
    },
    {
      "epoch": 0.9221556886227545,
      "grad_norm": 0.16829943656921387,
      "learning_rate": 1.8157976520603135e-05,
      "loss": 0.016,
      "mean_token_accuracy": 0.9941723346710205,
      "num_tokens": 59061863.0,
      "step": 770
    },
    {
      "epoch": 0.9281437125748503,
      "grad_norm": 0.15253153443336487,
      "learning_rate": 1.803229415024203e-05,
      "loss": 0.0095,
      "mean_token_accuracy": 0.9960687398910523,
      "num_tokens": 59440224.0,
      "step": 775
    },
    {
      "epoch": 0.9341317365269461,
      "grad_norm": 0.4071504771709442,
      "learning_rate": 1.7916325653880404e-05,
      "loss": 0.013,
      "mean_token_accuracy": 0.9964562654495239,
      "num_tokens": 59816422.0,
      "step": 780
    },
    {
      "epoch": 0.9401197604790419,
      "grad_norm": 0.14573310315608978,
      "learning_rate": 1.781011573958117e-05,
      "loss": 0.0114,
      "mean_token_accuracy": 0.9962265610694885,
      "num_tokens": 60197610.0,
      "step": 785
    },
    {
      "epoch": 0.9461077844311377,
      "grad_norm": 0.20949053764343262,
      "learning_rate": 1.771370535328808e-05,
      "loss": 0.0095,
      "mean_token_accuracy": 0.9965647101402283,
      "num_tokens": 60583281.0,
      "step": 790
    },
    {
      "epoch": 0.9520958083832335,
      "grad_norm": 0.187347874045372,
      "learning_rate": 1.762713166304031e-05,
      "loss": 0.0161,
      "mean_token_accuracy": 0.9939463973045349,
      "num_tokens": 60946880.0,
      "step": 795
    },
    {
      "epoch": 0.9580838323353293,
      "grad_norm": 0.10063041001558304,
      "learning_rate": 1.7550428044643455e-05,
      "loss": 0.0133,
      "mean_token_accuracy": 0.9964974045753479,
      "num_tokens": 61338197.0,
      "step": 800
    },
    {
      "epoch": 0.9640718562874252,
      "grad_norm": 0.15240474045276642,
      "learning_rate": 1.7483624068802527e-05,
      "loss": 0.0103,
      "mean_token_accuracy": 0.9961925983428955,
      "num_tokens": 61695359.0,
      "step": 805
    },
    {
      "epoch": 0.9700598802395209,
      "grad_norm": 0.1674066036939621,
      "learning_rate": 1.7426745489721884e-05,
      "loss": 0.0081,
      "mean_token_accuracy": 0.9972934484481811,
      "num_tokens": 62076251.0,
      "step": 810
    },
    {
      "epoch": 0.9760479041916168,
      "grad_norm": 0.2227143794298172,
      "learning_rate": 1.7379814235176503e-05,
      "loss": 0.0105,
      "mean_token_accuracy": 0.9966707110404969,
      "num_tokens": 62452562.0,
      "step": 815
    },
    {
      "epoch": 0.9820359281437125,
      "grad_norm": 0.10142319649457932,
      "learning_rate": 1.7342848398058374e-05,
      "loss": 0.0108,
      "mean_token_accuracy": 0.996941351890564,
      "num_tokens": 62843611.0,
      "step": 820
    },
    {
      "epoch": 0.9880239520958084,
      "grad_norm": 0.1786433905363083,
      "learning_rate": 1.731586222940139e-05,
      "loss": 0.0154,
      "mean_token_accuracy": 0.9946131467819214,
      "num_tokens": 63243009.0,
      "step": 825
    },
    {
      "epoch": 0.9940119760479041,
      "grad_norm": 0.1379786878824234,
      "learning_rate": 1.7298866132887277e-05,
      "loss": 0.0093,
      "mean_token_accuracy": 0.996778404712677,
      "num_tokens": 63630631.0,
      "step": 830
    }
  ],
  "logging_steps": 5,
  "max_steps": 835,
  "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": 3.4130534528408617e+18,
  "train_batch_size": 14,
  "trial_name": null,
  "trial_params": null
}