{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9992337164750957,
  "eval_steps": 500,
  "global_step": 652,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.007662835249042145,
      "grad_norm": 11.775691032409668,
      "learning_rate": 2.162496240601504e-06,
      "loss": 0.318,
      "mean_token_accuracy": 0.9071180582046509,
      "num_tokens": 1077246.0,
      "step": 5
    },
    {
      "epoch": 0.01532567049808429,
      "grad_norm": 2.0914440155029297,
      "learning_rate": 4.865616541353383e-06,
      "loss": 0.1405,
      "mean_token_accuracy": 0.9591145873069763,
      "num_tokens": 2172444.0,
      "step": 10
    },
    {
      "epoch": 0.022988505747126436,
      "grad_norm": 0.6351454257965088,
      "learning_rate": 7.568736842105265e-06,
      "loss": 0.1095,
      "mean_token_accuracy": 0.9618923664093018,
      "num_tokens": 3254491.0,
      "step": 15
    },
    {
      "epoch": 0.03065134099616858,
      "grad_norm": 0.5055913329124451,
      "learning_rate": 1.0271857142857143e-05,
      "loss": 0.0994,
      "mean_token_accuracy": 0.963628476858139,
      "num_tokens": 4344711.0,
      "step": 20
    },
    {
      "epoch": 0.038314176245210725,
      "grad_norm": 0.5738006830215454,
      "learning_rate": 1.2974977443609024e-05,
      "loss": 0.0936,
      "mean_token_accuracy": 0.9665798544883728,
      "num_tokens": 5430861.0,
      "step": 25
    },
    {
      "epoch": 0.04597701149425287,
      "grad_norm": 0.5167019367218018,
      "learning_rate": 1.5678097744360906e-05,
      "loss": 0.0917,
      "mean_token_accuracy": 0.9660590350627899,
      "num_tokens": 6522178.0,
      "step": 30
    },
    {
      "epoch": 0.05363984674329502,
      "grad_norm": 0.7361806035041809,
      "learning_rate": 1.8381218045112784e-05,
      "loss": 0.0884,
      "mean_token_accuracy": 0.9677951395511627,
      "num_tokens": 7603367.0,
      "step": 35
    },
    {
      "epoch": 0.06130268199233716,
      "grad_norm": 0.8272627592086792,
      "learning_rate": 1.8920370473768527e-05,
      "loss": 0.0851,
      "mean_token_accuracy": 0.9677951395511627,
      "num_tokens": 8699280.0,
      "step": 40
    },
    {
      "epoch": 0.06896551724137931,
      "grad_norm": 0.5785582661628723,
      "learning_rate": 1.8914393016996354e-05,
      "loss": 0.0789,
      "mean_token_accuracy": 0.97109375,
      "num_tokens": 9799050.0,
      "step": 45
    },
    {
      "epoch": 0.07662835249042145,
      "grad_norm": 0.8126276135444641,
      "learning_rate": 1.8903821629022412e-05,
      "loss": 0.0767,
      "mean_token_accuracy": 0.9718750178813934,
      "num_tokens": 10888773.0,
      "step": 50
    },
    {
      "epoch": 0.0842911877394636,
      "grad_norm": 0.6835556626319885,
      "learning_rate": 1.8888663161225525e-05,
      "loss": 0.0794,
      "mean_token_accuracy": 0.9729166865348816,
      "num_tokens": 11967768.0,
      "step": 55
    },
    {
      "epoch": 0.09195402298850575,
      "grad_norm": 0.5534384846687317,
      "learning_rate": 1.886892743789796e-05,
      "loss": 0.0771,
      "mean_token_accuracy": 0.9727430760860443,
      "num_tokens": 13065561.0,
      "step": 60
    },
    {
      "epoch": 0.09961685823754789,
      "grad_norm": 0.7145775556564331,
      "learning_rate": 1.8844627249878266e-05,
      "loss": 0.0743,
      "mean_token_accuracy": 0.9715277850627899,
      "num_tokens": 14152437.0,
      "step": 65
    },
    {
      "epoch": 0.10727969348659004,
      "grad_norm": 0.44082367420196533,
      "learning_rate": 1.881577834626145e-05,
      "loss": 0.08,
      "mean_token_accuracy": 0.9718750059604645,
      "num_tokens": 15253862.0,
      "step": 70
    },
    {
      "epoch": 0.11494252873563218,
      "grad_norm": 0.37960800528526306,
      "learning_rate": 1.8782399424191886e-05,
      "loss": 0.0699,
      "mean_token_accuracy": 0.9753472447395325,
      "num_tokens": 16351852.0,
      "step": 75
    },
    {
      "epoch": 0.12260536398467432,
      "grad_norm": 0.41791388392448425,
      "learning_rate": 1.8744512116745595e-05,
      "loss": 0.0731,
      "mean_token_accuracy": 0.9743923783302307,
      "num_tokens": 17442499.0,
      "step": 80
    },
    {
      "epoch": 0.13026819923371646,
      "grad_norm": 0.44737985730171204,
      "learning_rate": 1.8702140978909714e-05,
      "loss": 0.0699,
      "mean_token_accuracy": 0.9751736223697662,
      "num_tokens": 18525035.0,
      "step": 85
    },
    {
      "epoch": 0.13793103448275862,
      "grad_norm": 0.3966364562511444,
      "learning_rate": 1.8655313471668277e-05,
      "loss": 0.0698,
      "mean_token_accuracy": 0.9746528029441833,
      "num_tokens": 19631265.0,
      "step": 90
    },
    {
      "epoch": 0.14559386973180077,
      "grad_norm": 0.4190419316291809,
      "learning_rate": 1.8604059944204582e-05,
      "loss": 0.0697,
      "mean_token_accuracy": 0.9736979186534882,
      "num_tokens": 20712851.0,
      "step": 95
    },
    {
      "epoch": 0.1532567049808429,
      "grad_norm": 0.4391917288303375,
      "learning_rate": 1.8548413614231694e-05,
      "loss": 0.0689,
      "mean_token_accuracy": 0.9743923723697663,
      "num_tokens": 21807456.0,
      "step": 100
    },
    {
      "epoch": 0.16091954022988506,
      "grad_norm": 0.42839646339416504,
      "learning_rate": 1.848841054646387e-05,
      "loss": 0.0723,
      "mean_token_accuracy": 0.9734375178813934,
      "num_tokens": 22895468.0,
      "step": 105
    },
    {
      "epoch": 0.1685823754789272,
      "grad_norm": 0.4990848898887634,
      "learning_rate": 1.8424089629242794e-05,
      "loss": 0.0672,
      "mean_token_accuracy": 0.9744791805744171,
      "num_tokens": 23989680.0,
      "step": 110
    },
    {
      "epoch": 0.17624521072796934,
      "grad_norm": 0.4388997554779053,
      "learning_rate": 1.8355492549333818e-05,
      "loss": 0.0669,
      "mean_token_accuracy": 0.9763021051883698,
      "num_tokens": 25072335.0,
      "step": 115
    },
    {
      "epoch": 0.1839080459770115,
      "grad_norm": 0.35158392786979675,
      "learning_rate": 1.8282663764908525e-05,
      "loss": 0.0688,
      "mean_token_accuracy": 0.9746527731418609,
      "num_tokens": 26162308.0,
      "step": 120
    },
    {
      "epoch": 0.19157088122605365,
      "grad_norm": 0.3553743362426758,
      "learning_rate": 1.8205650476731113e-05,
      "loss": 0.0635,
      "mean_token_accuracy": 0.9765625119209289,
      "num_tokens": 27254497.0,
      "step": 125
    },
    {
      "epoch": 0.19923371647509577,
      "grad_norm": 0.48281288146972656,
      "learning_rate": 1.8124502597567315e-05,
      "loss": 0.0639,
      "mean_token_accuracy": 0.9776909708976745,
      "num_tokens": 28349197.0,
      "step": 130
    },
    {
      "epoch": 0.20689655172413793,
      "grad_norm": 0.38537609577178955,
      "learning_rate": 1.8039272719835648e-05,
      "loss": 0.0661,
      "mean_token_accuracy": 0.9774305641651153,
      "num_tokens": 29430193.0,
      "step": 135
    },
    {
      "epoch": 0.21455938697318008,
      "grad_norm": 0.4320482313632965,
      "learning_rate": 1.795001608152194e-05,
      "loss": 0.0688,
      "mean_token_accuracy": 0.9763020873069763,
      "num_tokens": 30532672.0,
      "step": 140
    },
    {
      "epoch": 0.2222222222222222,
      "grad_norm": 0.3345838487148285,
      "learning_rate": 1.7856790530379266e-05,
      "loss": 0.0631,
      "mean_token_accuracy": 0.9763020873069763,
      "num_tokens": 31627374.0,
      "step": 145
    },
    {
      "epoch": 0.22988505747126436,
      "grad_norm": 0.46670979261398315,
      "learning_rate": 1.7759656486436495e-05,
      "loss": 0.0632,
      "mean_token_accuracy": 0.9772569477558136,
      "num_tokens": 32709508.0,
      "step": 150
    },
    {
      "epoch": 0.23754789272030652,
      "grad_norm": 0.4126276671886444,
      "learning_rate": 1.76586769028397e-05,
      "loss": 0.0641,
      "mean_token_accuracy": 0.9767361402511596,
      "num_tokens": 33805592.0,
      "step": 155
    },
    {
      "epoch": 0.24521072796934865,
      "grad_norm": 0.37826210260391235,
      "learning_rate": 1.755391722505186e-05,
      "loss": 0.0654,
      "mean_token_accuracy": 0.977343761920929,
      "num_tokens": 34901401.0,
      "step": 160
    },
    {
      "epoch": 0.25287356321839083,
      "grad_norm": 0.33070090413093567,
      "learning_rate": 1.7445445348437266e-05,
      "loss": 0.0604,
      "mean_token_accuracy": 0.9770833611488342,
      "num_tokens": 35990805.0,
      "step": 165
    },
    {
      "epoch": 0.26053639846743293,
      "grad_norm": 0.3684611916542053,
      "learning_rate": 1.7333331574258133e-05,
      "loss": 0.0633,
      "mean_token_accuracy": 0.9767361223697663,
      "num_tokens": 37095539.0,
      "step": 170
    },
    {
      "epoch": 0.2681992337164751,
      "grad_norm": 0.4362129867076874,
      "learning_rate": 1.721764856411194e-05,
      "loss": 0.0635,
      "mean_token_accuracy": 0.9766493141651154,
      "num_tokens": 38185400.0,
      "step": 175
    },
    {
      "epoch": 0.27586206896551724,
      "grad_norm": 0.4917359948158264,
      "learning_rate": 1.7098471292839025e-05,
      "loss": 0.0624,
      "mean_token_accuracy": 0.9771701574325562,
      "num_tokens": 39270809.0,
      "step": 180
    },
    {
      "epoch": 0.2835249042145594,
      "grad_norm": 0.3966314494609833,
      "learning_rate": 1.6975876999930955e-05,
      "loss": 0.0596,
      "mean_token_accuracy": 0.9801215350627899,
      "num_tokens": 40368119.0,
      "step": 185
    },
    {
      "epoch": 0.29118773946360155,
      "grad_norm": 0.3741203248500824,
      "learning_rate": 1.684994513947113e-05,
      "loss": 0.0623,
      "mean_token_accuracy": 0.9770833373069763,
      "num_tokens": 41463186.0,
      "step": 190
    },
    {
      "epoch": 0.2988505747126437,
      "grad_norm": 0.34731847047805786,
      "learning_rate": 1.6720757328640154e-05,
      "loss": 0.0608,
      "mean_token_accuracy": 0.9786458492279053,
      "num_tokens": 42548414.0,
      "step": 195
    },
    {
      "epoch": 0.3065134099616858,
      "grad_norm": 0.3592664301395416,
      "learning_rate": 1.658839729481924e-05,
      "loss": 0.0612,
      "mean_token_accuracy": 0.9782117962837219,
      "num_tokens": 43637070.0,
      "step": 200
    },
    {
      "epoch": 0.31417624521072796,
      "grad_norm": 0.36953556537628174,
      "learning_rate": 1.645295082132601e-05,
      "loss": 0.0574,
      "mean_token_accuracy": 0.979253488779068,
      "num_tokens": 44719233.0,
      "step": 205
    },
    {
      "epoch": 0.3218390804597701,
      "grad_norm": 0.3487086296081543,
      "learning_rate": 1.631450569181782e-05,
      "loss": 0.0608,
      "mean_token_accuracy": 0.9775173604488373,
      "num_tokens": 45805568.0,
      "step": 210
    },
    {
      "epoch": 0.32950191570881227,
      "grad_norm": 0.30979305505752563,
      "learning_rate": 1.6173151633398657e-05,
      "loss": 0.0626,
      "mean_token_accuracy": 0.9780382215976715,
      "num_tokens": 46907262.0,
      "step": 215
    },
    {
      "epoch": 0.3371647509578544,
      "grad_norm": 0.3208833932876587,
      "learning_rate": 1.602898025846646e-05,
      "loss": 0.0625,
      "mean_token_accuracy": 0.9773437678813934,
      "num_tokens": 47996982.0,
      "step": 220
    },
    {
      "epoch": 0.3448275862068966,
      "grad_norm": 0.39099711179733276,
      "learning_rate": 1.588208500533861e-05,
      "loss": 0.0617,
      "mean_token_accuracy": 0.9773437559604645,
      "num_tokens": 49101799.0,
      "step": 225
    },
    {
      "epoch": 0.3524904214559387,
      "grad_norm": 0.34100982546806335,
      "learning_rate": 1.573256107769398e-05,
      "loss": 0.0635,
      "mean_token_accuracy": 0.9769097328186035,
      "num_tokens": 50196721.0,
      "step": 230
    },
    {
      "epoch": 0.36015325670498083,
      "grad_norm": 0.3107825517654419,
      "learning_rate": 1.5580505382870888e-05,
      "loss": 0.0633,
      "mean_token_accuracy": 0.9769965529441833,
      "num_tokens": 51290453.0,
      "step": 235
    },
    {
      "epoch": 0.367816091954023,
      "grad_norm": 0.31218069791793823,
      "learning_rate": 1.5426016469060878e-05,
      "loss": 0.0653,
      "mean_token_accuracy": 0.9751736164093018,
      "num_tokens": 52373930.0,
      "step": 240
    },
    {
      "epoch": 0.37547892720306514,
      "grad_norm": 0.3370981216430664,
      "learning_rate": 1.526919446143905e-05,
      "loss": 0.0599,
      "mean_token_accuracy": 0.9768229424953461,
      "num_tokens": 53467690.0,
      "step": 245
    },
    {
      "epoch": 0.3831417624521073,
      "grad_norm": 0.3400161862373352,
      "learning_rate": 1.5110140997272344e-05,
      "loss": 0.0561,
      "mean_token_accuracy": 0.9788194596767426,
      "num_tokens": 54572016.0,
      "step": 250
    },
    {
      "epoch": 0.39080459770114945,
      "grad_norm": 0.3365417718887329,
      "learning_rate": 1.494895916004783e-05,
      "loss": 0.0595,
      "mean_token_accuracy": 0.9775173664093018,
      "num_tokens": 55667785.0,
      "step": 255
    },
    {
      "epoch": 0.39846743295019155,
      "grad_norm": 0.27660876512527466,
      "learning_rate": 1.4785753412663708e-05,
      "loss": 0.0601,
      "mean_token_accuracy": 0.9776041746139527,
      "num_tokens": 56751073.0,
      "step": 260
    },
    {
      "epoch": 0.4061302681992337,
      "grad_norm": 0.30515599250793457,
      "learning_rate": 1.4620629529726256e-05,
      "loss": 0.0562,
      "mean_token_accuracy": 0.9784722268581391,
      "num_tokens": 57843072.0,
      "step": 265
    },
    {
      "epoch": 0.41379310344827586,
      "grad_norm": 0.4340786039829254,
      "learning_rate": 1.4453694528996721e-05,
      "loss": 0.061,
      "mean_token_accuracy": 0.9785590589046478,
      "num_tokens": 58943997.0,
      "step": 270
    },
    {
      "epoch": 0.421455938697318,
      "grad_norm": 0.38081300258636475,
      "learning_rate": 1.4285056602032446e-05,
      "loss": 0.0565,
      "mean_token_accuracy": 0.9790798723697662,
      "num_tokens": 60045338.0,
      "step": 275
    },
    {
      "epoch": 0.42911877394636017,
      "grad_norm": 0.28883832693099976,
      "learning_rate": 1.4114825044067308e-05,
      "loss": 0.0621,
      "mean_token_accuracy": 0.9771701335906983,
      "num_tokens": 61133724.0,
      "step": 280
    },
    {
      "epoch": 0.4367816091954023,
      "grad_norm": 0.28922542929649353,
      "learning_rate": 1.3943110183176817e-05,
      "loss": 0.0593,
      "mean_token_accuracy": 0.9784722328186035,
      "num_tokens": 62214177.0,
      "step": 285
    },
    {
      "epoch": 0.4444444444444444,
      "grad_norm": 0.34836313128471375,
      "learning_rate": 1.3770023308773852e-05,
      "loss": 0.0581,
      "mean_token_accuracy": 0.97734375,
      "num_tokens": 63298786.0,
      "step": 290
    },
    {
      "epoch": 0.4521072796934866,
      "grad_norm": 0.31483110785484314,
      "learning_rate": 1.3595676599481324e-05,
      "loss": 0.058,
      "mean_token_accuracy": 0.9768229246139526,
      "num_tokens": 64384875.0,
      "step": 295
    },
    {
      "epoch": 0.45977011494252873,
      "grad_norm": 0.4621759355068207,
      "learning_rate": 1.3420183050428566e-05,
      "loss": 0.0577,
      "mean_token_accuracy": 0.97890625,
      "num_tokens": 65477649.0,
      "step": 300
    },
    {
      "epoch": 0.4674329501915709,
      "grad_norm": 0.38731318712234497,
      "learning_rate": 1.324365640001851e-05,
      "loss": 0.0545,
      "mean_token_accuracy": 0.980121523141861,
      "num_tokens": 66557809.0,
      "step": 305
    },
    {
      "epoch": 0.47509578544061304,
      "grad_norm": 0.3266485035419464,
      "learning_rate": 1.306621105621317e-05,
      "loss": 0.0555,
      "mean_token_accuracy": 0.9790798723697662,
      "num_tokens": 67651417.0,
      "step": 310
    },
    {
      "epoch": 0.4827586206896552,
      "grad_norm": 0.3696978688240051,
      "learning_rate": 1.2887962022385146e-05,
      "loss": 0.0575,
      "mean_token_accuracy": 0.9781250059604645,
      "num_tokens": 68748506.0,
      "step": 315
    },
    {
      "epoch": 0.4904214559386973,
      "grad_norm": 0.3244847059249878,
      "learning_rate": 1.2709024822783297e-05,
      "loss": 0.0597,
      "mean_token_accuracy": 0.9786458492279053,
      "num_tokens": 69831103.0,
      "step": 320
    },
    {
      "epoch": 0.49808429118773945,
      "grad_norm": 0.2741115093231201,
      "learning_rate": 1.2529515427660755e-05,
      "loss": 0.054,
      "mean_token_accuracy": 0.9805555701255798,
      "num_tokens": 70922262.0,
      "step": 325
    },
    {
      "epoch": 0.5057471264367817,
      "grad_norm": 0.2509593963623047,
      "learning_rate": 1.2349550178113955e-05,
      "loss": 0.0584,
      "mean_token_accuracy": 0.978038203716278,
      "num_tokens": 72011671.0,
      "step": 330
    },
    {
      "epoch": 0.5134099616858238,
      "grad_norm": 0.2601114809513092,
      "learning_rate": 1.216924571068126e-05,
      "loss": 0.0557,
      "mean_token_accuracy": 0.9801215469837189,
      "num_tokens": 73115781.0,
      "step": 335
    },
    {
      "epoch": 0.5210727969348659,
      "grad_norm": 0.3080248236656189,
      "learning_rate": 1.1988718881750141e-05,
      "loss": 0.0555,
      "mean_token_accuracy": 0.9796875059604645,
      "num_tokens": 74197772.0,
      "step": 340
    },
    {
      "epoch": 0.5287356321839081,
      "grad_norm": 0.44369372725486755,
      "learning_rate": 1.1808086691821876e-05,
      "loss": 0.0544,
      "mean_token_accuracy": 0.9800347208976745,
      "num_tokens": 75300928.0,
      "step": 345
    },
    {
      "epoch": 0.5363984674329502,
      "grad_norm": 0.31000152230262756,
      "learning_rate": 1.1627466209682822e-05,
      "loss": 0.056,
      "mean_token_accuracy": 0.9795138955116272,
      "num_tokens": 76387213.0,
      "step": 350
    },
    {
      "epoch": 0.5440613026819924,
      "grad_norm": 0.2904181480407715,
      "learning_rate": 1.1446974496531438e-05,
      "loss": 0.0561,
      "mean_token_accuracy": 0.9789930582046509,
      "num_tokens": 77470917.0,
      "step": 355
    },
    {
      "epoch": 0.5517241379310345,
      "grad_norm": 0.2386808842420578,
      "learning_rate": 1.1266728530110266e-05,
      "loss": 0.062,
      "mean_token_accuracy": 0.9775173604488373,
      "num_tokens": 78568625.0,
      "step": 360
    },
    {
      "epoch": 0.5593869731800766,
      "grad_norm": 0.2828770577907562,
      "learning_rate": 1.1086845128891916e-05,
      "loss": 0.0587,
      "mean_token_accuracy": 0.97890625,
      "num_tokens": 79650226.0,
      "step": 365
    },
    {
      "epoch": 0.5670498084291188,
      "grad_norm": 0.2870100736618042,
      "learning_rate": 1.0907440876368388e-05,
      "loss": 0.0583,
      "mean_token_accuracy": 0.9782118260860443,
      "num_tokens": 80742756.0,
      "step": 370
    },
    {
      "epoch": 0.5747126436781609,
      "grad_norm": 0.27211377024650574,
      "learning_rate": 1.0728632045492557e-05,
      "loss": 0.0551,
      "mean_token_accuracy": 0.9795138895511627,
      "num_tokens": 81823129.0,
      "step": 375
    },
    {
      "epoch": 0.5823754789272031,
      "grad_norm": 0.26771602034568787,
      "learning_rate": 1.0550534523321044e-05,
      "loss": 0.0547,
      "mean_token_accuracy": 0.9801215291023254,
      "num_tokens": 82912796.0,
      "step": 380
    },
    {
      "epoch": 0.5900383141762452,
      "grad_norm": 0.3738754689693451,
      "learning_rate": 1.0373263735907098e-05,
      "loss": 0.0532,
      "mean_token_accuracy": 0.9796007096767425,
      "num_tokens": 83997407.0,
      "step": 385
    },
    {
      "epoch": 0.5977011494252874,
      "grad_norm": 0.3444598913192749,
      "learning_rate": 1.0196934573492302e-05,
      "loss": 0.0546,
      "mean_token_accuracy": 0.9808159708976746,
      "num_tokens": 85097209.0,
      "step": 390
    },
    {
      "epoch": 0.6053639846743295,
      "grad_norm": 0.21151889860630035,
      "learning_rate": 1.002166131604553e-05,
      "loss": 0.0538,
      "mean_token_accuracy": 0.9802083313465119,
      "num_tokens": 86191701.0,
      "step": 395
    },
    {
      "epoch": 0.6130268199233716,
      "grad_norm": 0.2923840582370758,
      "learning_rate": 9.84755755919744e-06,
      "loss": 0.0589,
      "mean_token_accuracy": 0.9780381977558136,
      "num_tokens": 87290541.0,
      "step": 400
    },
    {
      "epoch": 0.6206896551724138,
      "grad_norm": 0.23207074403762817,
      "learning_rate": 9.674736140618434e-06,
      "loss": 0.0606,
      "mean_token_accuracy": 0.9776041865348816,
      "num_tokens": 88398792.0,
      "step": 405
    },
    {
      "epoch": 0.6283524904214559,
      "grad_norm": 0.33359429240226746,
      "learning_rate": 9.503309066887955e-06,
      "loss": 0.056,
      "mean_token_accuracy": 0.979687511920929,
      "num_tokens": 89486208.0,
      "step": 410
    },
    {
      "epoch": 0.6360153256704981,
      "grad_norm": 0.2546365559101105,
      "learning_rate": 9.333387440902325e-06,
      "loss": 0.0549,
      "mean_token_accuracy": 0.980381965637207,
      "num_tokens": 90584558.0,
      "step": 415
    },
    {
      "epoch": 0.6436781609195402,
      "grad_norm": 0.2766530215740204,
      "learning_rate": 9.165081389868311e-06,
      "loss": 0.0577,
      "mean_token_accuracy": 0.9791666746139527,
      "num_tokens": 91682674.0,
      "step": 420
    },
    {
      "epoch": 0.6513409961685823,
      "grad_norm": 0.3182806074619293,
      "learning_rate": 8.998499993929042e-06,
      "loss": 0.0566,
      "mean_token_accuracy": 0.9783854365348816,
      "num_tokens": 92783006.0,
      "step": 425
    },
    {
      "epoch": 0.6590038314176245,
      "grad_norm": 0.23961606621742249,
      "learning_rate": 8.833751215468525e-06,
      "loss": 0.0554,
      "mean_token_accuracy": 0.9798611342906952,
      "num_tokens": 93875140.0,
      "step": 430
    },
    {
      "epoch": 0.6666666666666666,
      "grad_norm": 0.29564839601516724,
      "learning_rate": 8.67094182914057e-06,
      "loss": 0.0535,
      "mean_token_accuracy": 0.9807291805744172,
      "num_tokens": 94969575.0,
      "step": 435
    },
    {
      "epoch": 0.6743295019157088,
      "grad_norm": 0.3239465355873108,
      "learning_rate": 8.510177352667533e-06,
      "loss": 0.0546,
      "mean_token_accuracy": 0.9793403029441834,
      "num_tokens": 96060549.0,
      "step": 440
    },
    {
      "epoch": 0.6819923371647509,
      "grad_norm": 0.39116764068603516,
      "learning_rate": 8.351561978453641e-06,
      "loss": 0.0567,
      "mean_token_accuracy": 0.978819465637207,
      "num_tokens": 97156334.0,
      "step": 445
    },
    {
      "epoch": 0.6896551724137931,
      "grad_norm": 0.36669617891311646,
      "learning_rate": 8.195198506057288e-06,
      "loss": 0.0558,
      "mean_token_accuracy": 0.9782986104488373,
      "num_tokens": 98246337.0,
      "step": 450
    },
    {
      "epoch": 0.6973180076628352,
      "grad_norm": 0.3902563750743866,
      "learning_rate": 8.041188275566041e-06,
      "loss": 0.0581,
      "mean_token_accuracy": 0.9772569537162781,
      "num_tokens": 99342179.0,
      "step": 455
    },
    {
      "epoch": 0.7049808429118773,
      "grad_norm": 0.2951866686344147,
      "learning_rate": 7.889631101917516e-06,
      "loss": 0.0546,
      "mean_token_accuracy": 0.978472238779068,
      "num_tokens": 100444505.0,
      "step": 460
    },
    {
      "epoch": 0.7126436781609196,
      "grad_norm": 0.3099174499511719,
      "learning_rate": 7.740625210208746e-06,
      "loss": 0.0559,
      "mean_token_accuracy": 0.9793402850627899,
      "num_tokens": 101533570.0,
      "step": 465
    },
    {
      "epoch": 0.7203065134099617,
      "grad_norm": 0.25526103377342224,
      "learning_rate": 7.594267172035906e-06,
      "loss": 0.0511,
      "mean_token_accuracy": 0.9812500178813934,
      "num_tokens": 102633640.0,
      "step": 470
    },
    {
      "epoch": 0.7279693486590039,
      "grad_norm": 0.29758980870246887,
      "learning_rate": 7.450651842905699e-06,
      "loss": 0.053,
      "mean_token_accuracy": 0.9807291805744172,
      "num_tokens": 103723814.0,
      "step": 475
    },
    {
      "epoch": 0.735632183908046,
      "grad_norm": 0.298353374004364,
      "learning_rate": 7.309872300758933e-06,
      "loss": 0.054,
      "mean_token_accuracy": 0.979600703716278,
      "num_tokens": 104807793.0,
      "step": 480
    },
    {
      "epoch": 0.7432950191570882,
      "grad_norm": 0.36888647079467773,
      "learning_rate": 7.172019785646182e-06,
      "loss": 0.0555,
      "mean_token_accuracy": 0.9802951574325561,
      "num_tokens": 105896118.0,
      "step": 485
    },
    {
      "epoch": 0.7509578544061303,
      "grad_norm": 0.3105420172214508,
      "learning_rate": 7.037183640594544e-06,
      "loss": 0.0508,
      "mean_token_accuracy": 0.9802083313465119,
      "num_tokens": 106982558.0,
      "step": 490
    },
    {
      "epoch": 0.7586206896551724,
      "grad_norm": 0.3314758539199829,
      "learning_rate": 6.905451253703918e-06,
      "loss": 0.0552,
      "mean_token_accuracy": 0.9802083373069763,
      "num_tokens": 108076197.0,
      "step": 495
    },
    {
      "epoch": 0.7662835249042146,
      "grad_norm": 0.3398136496543884,
      "learning_rate": 6.776908001510251e-06,
      "loss": 0.0523,
      "mean_token_accuracy": 0.98046875,
      "num_tokens": 109161850.0,
      "step": 500
    },
    {
      "epoch": 0.7739463601532567,
      "grad_norm": 0.2903926968574524,
      "learning_rate": 6.651637193652513e-06,
      "loss": 0.056,
      "mean_token_accuracy": 0.979340273141861,
      "num_tokens": 110253260.0,
      "step": 505
    },
    {
      "epoch": 0.7816091954022989,
      "grad_norm": 0.24769453704357147,
      "learning_rate": 6.52972001887921e-06,
      "loss": 0.0597,
      "mean_token_accuracy": 0.9774305641651153,
      "num_tokens": 111348891.0,
      "step": 510
    },
    {
      "epoch": 0.789272030651341,
      "grad_norm": 0.2901223301887512,
      "learning_rate": 6.411235492429508e-06,
      "loss": 0.0575,
      "mean_token_accuracy": 0.97855903506279,
      "num_tokens": 112437384.0,
      "step": 515
    },
    {
      "epoch": 0.7969348659003831,
      "grad_norm": 0.41612568497657776,
      "learning_rate": 6.296260404822958e-06,
      "loss": 0.055,
      "mean_token_accuracy": 0.9800347447395324,
      "num_tokens": 113519456.0,
      "step": 520
    },
    {
      "epoch": 0.8045977011494253,
      "grad_norm": 0.3250349462032318,
      "learning_rate": 6.184869272091133e-06,
      "loss": 0.0551,
      "mean_token_accuracy": 0.9795139074325562,
      "num_tokens": 114603903.0,
      "step": 525
    },
    {
      "epoch": 0.8122605363984674,
      "grad_norm": 0.29371413588523865,
      "learning_rate": 6.077134287483337e-06,
      "loss": 0.0547,
      "mean_token_accuracy": 0.9800347208976745,
      "num_tokens": 115699227.0,
      "step": 530
    },
    {
      "epoch": 0.8199233716475096,
      "grad_norm": 0.35530513525009155,
      "learning_rate": 5.973125274677734e-06,
      "loss": 0.0551,
      "mean_token_accuracy": 0.979687511920929,
      "num_tokens": 116787851.0,
      "step": 535
    },
    {
      "epoch": 0.8275862068965517,
      "grad_norm": 0.40407872200012207,
      "learning_rate": 5.872909642528218e-06,
      "loss": 0.0573,
      "mean_token_accuracy": 0.979340273141861,
      "num_tokens": 117873854.0,
      "step": 540
    },
    {
      "epoch": 0.8352490421455939,
      "grad_norm": 0.2643112540245056,
      "learning_rate": 5.776552341376333e-06,
      "loss": 0.0517,
      "mean_token_accuracy": 0.9805555701255798,
      "num_tokens": 118967125.0,
      "step": 545
    },
    {
      "epoch": 0.842911877394636,
      "grad_norm": 0.25208526849746704,
      "learning_rate": 5.6841158209565735e-06,
      "loss": 0.0541,
      "mean_token_accuracy": 0.9786458313465118,
      "num_tokens": 120042343.0,
      "step": 550
    },
    {
      "epoch": 0.8505747126436781,
      "grad_norm": 0.324431836605072,
      "learning_rate": 5.595659989922345e-06,
      "loss": 0.0573,
      "mean_token_accuracy": 0.9793402850627899,
      "num_tokens": 121137305.0,
      "step": 555
    },
    {
      "epoch": 0.8582375478927203,
      "grad_norm": 0.28662794828414917,
      "learning_rate": 5.511242177018818e-06,
      "loss": 0.0526,
      "mean_token_accuracy": 0.9802083432674408,
      "num_tokens": 122226343.0,
      "step": 560
    },
    {
      "epoch": 0.8659003831417624,
      "grad_norm": 0.2520166039466858,
      "learning_rate": 5.430917093927816e-06,
      "loss": 0.0572,
      "mean_token_accuracy": 0.9789930582046509,
      "num_tokens": 123320384.0,
      "step": 565
    },
    {
      "epoch": 0.8735632183908046,
      "grad_norm": 0.2630253732204437,
      "learning_rate": 5.3547367998088715e-06,
      "loss": 0.0487,
      "mean_token_accuracy": 0.9818576514720917,
      "num_tokens": 124404287.0,
      "step": 570
    },
    {
      "epoch": 0.8812260536398467,
      "grad_norm": 0.24426770210266113,
      "learning_rate": 5.282750667559361e-06,
      "loss": 0.0527,
      "mean_token_accuracy": 0.9807291686534881,
      "num_tokens": 125499483.0,
      "step": 575
    },
    {
      "epoch": 0.8888888888888888,
      "grad_norm": 0.2494325190782547,
      "learning_rate": 5.215005351815651e-06,
      "loss": 0.0505,
      "mean_token_accuracy": 0.9824652910232544,
      "num_tokens": 126595035.0,
      "step": 580
    },
    {
      "epoch": 0.896551724137931,
      "grad_norm": 0.3279549777507782,
      "learning_rate": 5.151544758715945e-06,
      "loss": 0.0511,
      "mean_token_accuracy": 0.9807291865348816,
      "num_tokens": 127687237.0,
      "step": 585
    },
    {
      "epoch": 0.9042145593869731,
      "grad_norm": 0.2407781481742859,
      "learning_rate": 5.092410017444466e-06,
      "loss": 0.0507,
      "mean_token_accuracy": 0.9813368141651153,
      "num_tokens": 128784622.0,
      "step": 590
    },
    {
      "epoch": 0.9118773946360154,
      "grad_norm": 0.2758333683013916,
      "learning_rate": 5.037639453575379e-06,
      "loss": 0.0514,
      "mean_token_accuracy": 0.981250011920929,
      "num_tokens": 129878622.0,
      "step": 595
    },
    {
      "epoch": 0.9195402298850575,
      "grad_norm": 0.44546186923980713,
      "learning_rate": 4.987268564233768e-06,
      "loss": 0.0501,
      "mean_token_accuracy": 0.9814236283302307,
      "num_tokens": 130970165.0,
      "step": 600
    },
    {
      "epoch": 0.9272030651340997,
      "grad_norm": 0.3608750104904175,
      "learning_rate": 4.941329995089741e-06,
      "loss": 0.0515,
      "mean_token_accuracy": 0.9806423664093018,
      "num_tokens": 132065030.0,
      "step": 605
    },
    {
      "epoch": 0.9348659003831418,
      "grad_norm": 0.3858118951320648,
      "learning_rate": 4.899853519200571e-06,
      "loss": 0.055,
      "mean_token_accuracy": 0.9790798604488373,
      "num_tokens": 133161259.0,
      "step": 610
    },
    {
      "epoch": 0.9425287356321839,
      "grad_norm": 0.23960553109645844,
      "learning_rate": 4.862866017714614e-06,
      "loss": 0.0517,
      "mean_token_accuracy": 0.9822048664093017,
      "num_tokens": 134253608.0,
      "step": 615
    },
    {
      "epoch": 0.9501915708812261,
      "grad_norm": 0.3373887538909912,
      "learning_rate": 4.830391462449465e-06,
      "loss": 0.0516,
      "mean_token_accuracy": 0.9809027791023255,
      "num_tokens": 135348208.0,
      "step": 620
    },
    {
      "epoch": 0.9578544061302682,
      "grad_norm": 0.3087356388568878,
      "learning_rate": 4.8024509003556915e-06,
      "loss": 0.0566,
      "mean_token_accuracy": 0.9789062678813935,
      "num_tokens": 136445414.0,
      "step": 625
    },
    {
      "epoch": 0.9655172413793104,
      "grad_norm": 0.30412212014198303,
      "learning_rate": 4.77906243987618e-06,
      "loss": 0.051,
      "mean_token_accuracy": 0.9806423783302307,
      "num_tokens": 137537027.0,
      "step": 630
    },
    {
      "epoch": 0.9731800766283525,
      "grad_norm": 0.3232341706752777,
      "learning_rate": 4.760241239209941e-06,
      "loss": 0.0581,
      "mean_token_accuracy": 0.9768229246139526,
      "num_tokens": 138621895.0,
      "step": 635
    },
    {
      "epoch": 0.9808429118773946,
      "grad_norm": 0.28698432445526123,
      "learning_rate": 4.745999496488e-06,
      "loss": 0.0486,
      "mean_token_accuracy": 0.9817708313465119,
      "num_tokens": 139717205.0,
      "step": 640
    },
    {
      "epoch": 0.9885057471264368,
      "grad_norm": 0.25341159105300903,
      "learning_rate": 4.736346441867709e-06,
      "loss": 0.0533,
      "mean_token_accuracy": 0.9797743141651154,
      "num_tokens": 140814683.0,
      "step": 645
    },
    {
      "epoch": 0.9961685823754789,
      "grad_norm": 0.315576434135437,
      "learning_rate": 4.731288331550632e-06,
      "loss": 0.0512,
      "mean_token_accuracy": 0.9809895813465118,
      "num_tokens": 141895145.0,
      "step": 650
    }
  ],
  "logging_steps": 5,
  "max_steps": 652,
  "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": true
      },
      "attributes": {}
    }
  },
  "total_flos": 8.554477504520258e+18,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}