{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.47512135922330095,
  "eval_steps": 500,
  "global_step": 783,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.003033980582524272,
      "grad_norm": 12.881993293762207,
      "learning_rate": 1.7209638554216869e-06,
      "loss": 1.8952,
      "mean_token_accuracy": 0.6737864851951599,
      "num_tokens": 407386.0,
      "step": 5
    },
    {
      "epoch": 0.006067961165048544,
      "grad_norm": 5.927900791168213,
      "learning_rate": 3.872168674698796e-06,
      "loss": 1.7501,
      "mean_token_accuracy": 0.6814195275306701,
      "num_tokens": 814608.0,
      "step": 10
    },
    {
      "epoch": 0.009101941747572815,
      "grad_norm": 3.0456957817077637,
      "learning_rate": 6.023373493975903e-06,
      "loss": 1.4337,
      "mean_token_accuracy": 0.704640793800354,
      "num_tokens": 1221103.0,
      "step": 15
    },
    {
      "epoch": 0.012135922330097087,
      "grad_norm": 2.4085395336151123,
      "learning_rate": 8.174578313253012e-06,
      "loss": 1.0971,
      "mean_token_accuracy": 0.7540878534317017,
      "num_tokens": 1628633.0,
      "step": 20
    },
    {
      "epoch": 0.01516990291262136,
      "grad_norm": 3.0483357906341553,
      "learning_rate": 1.0325783132530121e-05,
      "loss": 0.7505,
      "mean_token_accuracy": 0.8266023993492126,
      "num_tokens": 2035802.0,
      "step": 25
    },
    {
      "epoch": 0.01820388349514563,
      "grad_norm": 2.5269112586975098,
      "learning_rate": 1.247698795180723e-05,
      "loss": 0.413,
      "mean_token_accuracy": 0.8993165135383606,
      "num_tokens": 2442910.0,
      "step": 30
    },
    {
      "epoch": 0.021237864077669904,
      "grad_norm": 0.5028864741325378,
      "learning_rate": 1.4628192771084337e-05,
      "loss": 0.1641,
      "mean_token_accuracy": 0.9638293981552124,
      "num_tokens": 2849002.0,
      "step": 35
    },
    {
      "epoch": 0.024271844660194174,
      "grad_norm": 0.32503458857536316,
      "learning_rate": 1.6779397590361448e-05,
      "loss": 0.0904,
      "mean_token_accuracy": 0.9782682418823242,
      "num_tokens": 3256206.0,
      "step": 40
    },
    {
      "epoch": 0.027305825242718445,
      "grad_norm": 0.3155650794506073,
      "learning_rate": 1.8930602409638557e-05,
      "loss": 0.0543,
      "mean_token_accuracy": 0.986135733127594,
      "num_tokens": 3663246.0,
      "step": 45
    },
    {
      "epoch": 0.03033980582524272,
      "grad_norm": 0.18562550842761993,
      "learning_rate": 2.1081807228915663e-05,
      "loss": 0.0373,
      "mean_token_accuracy": 0.9897839546203613,
      "num_tokens": 4069706.0,
      "step": 50
    },
    {
      "epoch": 0.03337378640776699,
      "grad_norm": 0.120665542781353,
      "learning_rate": 2.3233012048192772e-05,
      "loss": 0.028,
      "mean_token_accuracy": 0.9917330026626587,
      "num_tokens": 4476831.0,
      "step": 55
    },
    {
      "epoch": 0.03640776699029126,
      "grad_norm": 0.08647274225950241,
      "learning_rate": 2.538421686746988e-05,
      "loss": 0.0246,
      "mean_token_accuracy": 0.9926599621772766,
      "num_tokens": 4883779.0,
      "step": 60
    },
    {
      "epoch": 0.03944174757281554,
      "grad_norm": 0.07028564810752869,
      "learning_rate": 2.753542168674699e-05,
      "loss": 0.0242,
      "mean_token_accuracy": 0.9927438616752624,
      "num_tokens": 5290679.0,
      "step": 65
    },
    {
      "epoch": 0.04247572815533981,
      "grad_norm": 0.10383659601211548,
      "learning_rate": 2.96866265060241e-05,
      "loss": 0.0237,
      "mean_token_accuracy": 0.992747163772583,
      "num_tokens": 5697620.0,
      "step": 70
    },
    {
      "epoch": 0.04550970873786408,
      "grad_norm": 0.09976121038198471,
      "learning_rate": 3.183783132530121e-05,
      "loss": 0.0278,
      "mean_token_accuracy": 0.991703736782074,
      "num_tokens": 6104413.0,
      "step": 75
    },
    {
      "epoch": 0.04854368932038835,
      "grad_norm": 0.06832871586084366,
      "learning_rate": 3.3989036144578313e-05,
      "loss": 0.0243,
      "mean_token_accuracy": 0.992477285861969,
      "num_tokens": 6510558.0,
      "step": 80
    },
    {
      "epoch": 0.05157766990291262,
      "grad_norm": 0.06535604596138,
      "learning_rate": 3.570997301875174e-05,
      "loss": 0.0248,
      "mean_token_accuracy": 0.9922604560852051,
      "num_tokens": 6917082.0,
      "step": 85
    },
    {
      "epoch": 0.05461165048543689,
      "grad_norm": 0.10271787643432617,
      "learning_rate": 3.570902868647871e-05,
      "loss": 0.0224,
      "mean_token_accuracy": 0.9928990721702575,
      "num_tokens": 7324254.0,
      "step": 90
    },
    {
      "epoch": 0.05764563106796117,
      "grad_norm": 0.10438171029090881,
      "learning_rate": 3.5706735400516436e-05,
      "loss": 0.0243,
      "mean_token_accuracy": 0.9923665881156921,
      "num_tokens": 7731206.0,
      "step": 95
    },
    {
      "epoch": 0.06067961165048544,
      "grad_norm": 0.04581941291689873,
      "learning_rate": 3.570309339189348e-05,
      "loss": 0.0224,
      "mean_token_accuracy": 0.9928493976593018,
      "num_tokens": 8137840.0,
      "step": 100
    },
    {
      "epoch": 0.06371359223300971,
      "grad_norm": 0.05334591865539551,
      "learning_rate": 3.569810302751041e-05,
      "loss": 0.0248,
      "mean_token_accuracy": 0.9923072218894958,
      "num_tokens": 8544534.0,
      "step": 105
    },
    {
      "epoch": 0.06674757281553398,
      "grad_norm": 0.052358437329530716,
      "learning_rate": 3.569176481010288e-05,
      "loss": 0.0239,
      "mean_token_accuracy": 0.9925079822540284,
      "num_tokens": 8951215.0,
      "step": 110
    },
    {
      "epoch": 0.06978155339805825,
      "grad_norm": 0.08001242578029633,
      "learning_rate": 3.568407937819095e-05,
      "loss": 0.0264,
      "mean_token_accuracy": 0.9917665481567383,
      "num_tokens": 9357839.0,
      "step": 115
    },
    {
      "epoch": 0.07281553398058252,
      "grad_norm": 0.05479741096496582,
      "learning_rate": 3.5675047506014815e-05,
      "loss": 0.0242,
      "mean_token_accuracy": 0.9923253774642944,
      "num_tokens": 9764643.0,
      "step": 120
    },
    {
      "epoch": 0.07584951456310679,
      "grad_norm": 0.05963903293013573,
      "learning_rate": 3.566467010345674e-05,
      "loss": 0.0231,
      "mean_token_accuracy": 0.9926764607429505,
      "num_tokens": 10170675.0,
      "step": 125
    },
    {
      "epoch": 0.07888349514563107,
      "grad_norm": 0.057525910437107086,
      "learning_rate": 3.5652948215949454e-05,
      "loss": 0.0223,
      "mean_token_accuracy": 0.9926943898200988,
      "num_tokens": 10577296.0,
      "step": 130
    },
    {
      "epoch": 0.08191747572815535,
      "grad_norm": 0.059444062411785126,
      "learning_rate": 3.56398830243708e-05,
      "loss": 0.0228,
      "mean_token_accuracy": 0.9926655173301697,
      "num_tokens": 10983842.0,
      "step": 135
    },
    {
      "epoch": 0.08495145631067962,
      "grad_norm": 0.10497656464576721,
      "learning_rate": 3.562547584492477e-05,
      "loss": 0.021,
      "mean_token_accuracy": 0.9932034730911254,
      "num_tokens": 11391523.0,
      "step": 140
    },
    {
      "epoch": 0.08798543689320389,
      "grad_norm": 0.04242556169629097,
      "learning_rate": 3.5609728129008934e-05,
      "loss": 0.0212,
      "mean_token_accuracy": 0.9929630875587463,
      "num_tokens": 11798319.0,
      "step": 145
    },
    {
      "epoch": 0.09101941747572816,
      "grad_norm": 0.03587333485484123,
      "learning_rate": 3.5592641463068204e-05,
      "loss": 0.0214,
      "mean_token_accuracy": 0.9929973602294921,
      "num_tokens": 12204367.0,
      "step": 150
    },
    {
      "epoch": 0.09405339805825243,
      "grad_norm": 0.03362417593598366,
      "learning_rate": 3.557421756843501e-05,
      "loss": 0.022,
      "mean_token_accuracy": 0.9928773045539856,
      "num_tokens": 12611691.0,
      "step": 155
    },
    {
      "epoch": 0.0970873786407767,
      "grad_norm": 0.05092537775635719,
      "learning_rate": 3.555445830115592e-05,
      "loss": 0.019,
      "mean_token_accuracy": 0.9936353206634522,
      "num_tokens": 13019316.0,
      "step": 160
    },
    {
      "epoch": 0.10012135922330097,
      "grad_norm": 0.051133450120687485,
      "learning_rate": 3.5533365651804625e-05,
      "loss": 0.0231,
      "mean_token_accuracy": 0.9924604773521424,
      "num_tokens": 13426273.0,
      "step": 165
    },
    {
      "epoch": 0.10315533980582524,
      "grad_norm": 0.0654042586684227,
      "learning_rate": 3.5510941745281405e-05,
      "loss": 0.0234,
      "mean_token_accuracy": 0.9927022099494934,
      "num_tokens": 13832090.0,
      "step": 170
    },
    {
      "epoch": 0.10618932038834951,
      "grad_norm": 0.05316481739282608,
      "learning_rate": 3.548718884059912e-05,
      "loss": 0.0173,
      "mean_token_accuracy": 0.9940166115760803,
      "num_tokens": 14239146.0,
      "step": 175
    },
    {
      "epoch": 0.10922330097087378,
      "grad_norm": 0.0432591438293457,
      "learning_rate": 3.5462109330655545e-05,
      "loss": 0.0226,
      "mean_token_accuracy": 0.9927533149719239,
      "num_tokens": 14645842.0,
      "step": 180
    },
    {
      "epoch": 0.11225728155339806,
      "grad_norm": 0.040612801909446716,
      "learning_rate": 3.5435705741992395e-05,
      "loss": 0.0228,
      "mean_token_accuracy": 0.9925273418426513,
      "num_tokens": 15052978.0,
      "step": 185
    },
    {
      "epoch": 0.11529126213592233,
      "grad_norm": 0.03967801108956337,
      "learning_rate": 3.540798073454075e-05,
      "loss": 0.0215,
      "mean_token_accuracy": 0.9930168747901916,
      "num_tokens": 15459755.0,
      "step": 190
    },
    {
      "epoch": 0.1183252427184466,
      "grad_norm": 0.06023363396525383,
      "learning_rate": 3.5378937101353065e-05,
      "loss": 0.0227,
      "mean_token_accuracy": 0.9926756858825684,
      "num_tokens": 15866485.0,
      "step": 195
    },
    {
      "epoch": 0.12135922330097088,
      "grad_norm": 0.028881069272756577,
      "learning_rate": 3.534857776832187e-05,
      "loss": 0.0194,
      "mean_token_accuracy": 0.9934703350067139,
      "num_tokens": 16274029.0,
      "step": 200
    },
    {
      "epoch": 0.12439320388349515,
      "grad_norm": 0.04172505810856819,
      "learning_rate": 3.5316905793884944e-05,
      "loss": 0.0217,
      "mean_token_accuracy": 0.9927970051765442,
      "num_tokens": 16681849.0,
      "step": 205
    },
    {
      "epoch": 0.12742718446601942,
      "grad_norm": 0.047556426376104355,
      "learning_rate": 3.528392436871724e-05,
      "loss": 0.0213,
      "mean_token_accuracy": 0.992992353439331,
      "num_tokens": 17089757.0,
      "step": 210
    },
    {
      "epoch": 0.1304611650485437,
      "grad_norm": 0.053197335451841354,
      "learning_rate": 3.5249636815409434e-05,
      "loss": 0.0233,
      "mean_token_accuracy": 0.9926238298416138,
      "num_tokens": 17496241.0,
      "step": 215
    },
    {
      "epoch": 0.13349514563106796,
      "grad_norm": 0.040444646030664444,
      "learning_rate": 3.5214046588133205e-05,
      "loss": 0.0242,
      "mean_token_accuracy": 0.992365038394928,
      "num_tokens": 17902466.0,
      "step": 220
    },
    {
      "epoch": 0.13652912621359223,
      "grad_norm": 0.07247660309076309,
      "learning_rate": 3.5177157272293284e-05,
      "loss": 0.0217,
      "mean_token_accuracy": 0.9929222345352173,
      "num_tokens": 18308520.0,
      "step": 225
    },
    {
      "epoch": 0.1395631067961165,
      "grad_norm": 0.3286735713481903,
      "learning_rate": 3.513897258416623e-05,
      "loss": 0.0211,
      "mean_token_accuracy": 0.9928990960121155,
      "num_tokens": 18715808.0,
      "step": 230
    },
    {
      "epoch": 0.14259708737864077,
      "grad_norm": 1.6682692766189575,
      "learning_rate": 3.509949637052602e-05,
      "loss": 0.0204,
      "mean_token_accuracy": 0.9931188821792603,
      "num_tokens": 19122927.0,
      "step": 235
    },
    {
      "epoch": 0.14563106796116504,
      "grad_norm": 0.3066796362400055,
      "learning_rate": 3.505873260825659e-05,
      "loss": 0.0223,
      "mean_token_accuracy": 0.9929746747016907,
      "num_tokens": 19530080.0,
      "step": 240
    },
    {
      "epoch": 0.1486650485436893,
      "grad_norm": 0.09548434615135193,
      "learning_rate": 3.5016685403951156e-05,
      "loss": 0.0215,
      "mean_token_accuracy": 0.9930257797241211,
      "num_tokens": 19936885.0,
      "step": 245
    },
    {
      "epoch": 0.15169902912621358,
      "grad_norm": 0.04770074412226677,
      "learning_rate": 3.4973358993498494e-05,
      "loss": 0.0206,
      "mean_token_accuracy": 0.993126392364502,
      "num_tokens": 20342832.0,
      "step": 250
    },
    {
      "epoch": 0.15473300970873785,
      "grad_norm": 0.07485935091972351,
      "learning_rate": 3.492875774165625e-05,
      "loss": 0.0232,
      "mean_token_accuracy": 0.9926700353622436,
      "num_tokens": 20748938.0,
      "step": 255
    },
    {
      "epoch": 0.15776699029126215,
      "grad_norm": 0.04704100638628006,
      "learning_rate": 3.48828861416112e-05,
      "loss": 0.0203,
      "mean_token_accuracy": 0.9931455612182617,
      "num_tokens": 21156249.0,
      "step": 260
    },
    {
      "epoch": 0.16080097087378642,
      "grad_norm": 0.04121704027056694,
      "learning_rate": 3.4835748814526635e-05,
      "loss": 0.0231,
      "mean_token_accuracy": 0.9925450682640076,
      "num_tokens": 21562911.0,
      "step": 265
    },
    {
      "epoch": 0.1638349514563107,
      "grad_norm": 0.28638556599617004,
      "learning_rate": 3.478735050907677e-05,
      "loss": 0.0212,
      "mean_token_accuracy": 0.9931713461875915,
      "num_tokens": 21968979.0,
      "step": 270
    },
    {
      "epoch": 0.16686893203883496,
      "grad_norm": 0.12565217912197113,
      "learning_rate": 3.473769610096839e-05,
      "loss": 0.0254,
      "mean_token_accuracy": 0.9920137882232666,
      "num_tokens": 22375386.0,
      "step": 275
    },
    {
      "epoch": 0.16990291262135923,
      "grad_norm": 0.22467869520187378,
      "learning_rate": 3.468679059244967e-05,
      "loss": 0.022,
      "mean_token_accuracy": 0.9929057478904724,
      "num_tokens": 22782306.0,
      "step": 280
    },
    {
      "epoch": 0.1729368932038835,
      "grad_norm": 0.0678594633936882,
      "learning_rate": 3.463463911180622e-05,
      "loss": 0.023,
      "mean_token_accuracy": 0.9927614450454711,
      "num_tokens": 23188938.0,
      "step": 285
    },
    {
      "epoch": 0.17597087378640777,
      "grad_norm": 0.05392945557832718,
      "learning_rate": 3.458124691284445e-05,
      "loss": 0.0205,
      "mean_token_accuracy": 0.9933057188987732,
      "num_tokens": 23595074.0,
      "step": 290
    },
    {
      "epoch": 0.17900485436893204,
      "grad_norm": 0.058865759521722794,
      "learning_rate": 3.452661937436235e-05,
      "loss": 0.0216,
      "mean_token_accuracy": 0.9929778814315796,
      "num_tokens": 24001568.0,
      "step": 295
    },
    {
      "epoch": 0.1820388349514563,
      "grad_norm": 0.05121828243136406,
      "learning_rate": 3.447076199960755e-05,
      "loss": 0.0223,
      "mean_token_accuracy": 0.9928174257278443,
      "num_tokens": 24407977.0,
      "step": 300
    },
    {
      "epoch": 0.18507281553398058,
      "grad_norm": 0.046074215322732925,
      "learning_rate": 3.4413680415722955e-05,
      "loss": 0.0211,
      "mean_token_accuracy": 0.9932069063186646,
      "num_tokens": 24813624.0,
      "step": 305
    },
    {
      "epoch": 0.18810679611650485,
      "grad_norm": 0.03264041990041733,
      "learning_rate": 3.435538037317986e-05,
      "loss": 0.0204,
      "mean_token_accuracy": 0.9931853413581848,
      "num_tokens": 25219688.0,
      "step": 310
    },
    {
      "epoch": 0.19114077669902912,
      "grad_norm": 0.03584812954068184,
      "learning_rate": 3.429586774519863e-05,
      "loss": 0.02,
      "mean_token_accuracy": 0.9934399485588074,
      "num_tokens": 25626392.0,
      "step": 315
    },
    {
      "epoch": 0.1941747572815534,
      "grad_norm": 0.03796568140387535,
      "learning_rate": 3.423514852715701e-05,
      "loss": 0.0198,
      "mean_token_accuracy": 0.9932878255844116,
      "num_tokens": 26033554.0,
      "step": 320
    },
    {
      "epoch": 0.19720873786407767,
      "grad_norm": 0.030003419145941734,
      "learning_rate": 3.417322883598621e-05,
      "loss": 0.0183,
      "mean_token_accuracy": 0.9937293291091919,
      "num_tokens": 26439949.0,
      "step": 325
    },
    {
      "epoch": 0.20024271844660194,
      "grad_norm": 0.03972536325454712,
      "learning_rate": 3.411011490955457e-05,
      "loss": 0.0215,
      "mean_token_accuracy": 0.9929001450538635,
      "num_tokens": 26847039.0,
      "step": 330
    },
    {
      "epoch": 0.2032766990291262,
      "grad_norm": 0.04523684084415436,
      "learning_rate": 3.4045813106039256e-05,
      "loss": 0.0211,
      "mean_token_accuracy": 0.9929434895515442,
      "num_tokens": 27253524.0,
      "step": 335
    },
    {
      "epoch": 0.20631067961165048,
      "grad_norm": 0.05554588511586189,
      "learning_rate": 3.398032990328565e-05,
      "loss": 0.0198,
      "mean_token_accuracy": 0.9933468461036682,
      "num_tokens": 27660579.0,
      "step": 340
    },
    {
      "epoch": 0.20934466019417475,
      "grad_norm": 0.0418703593313694,
      "learning_rate": 3.3913671898154795e-05,
      "loss": 0.0254,
      "mean_token_accuracy": 0.9919080615043641,
      "num_tokens": 28066764.0,
      "step": 345
    },
    {
      "epoch": 0.21237864077669902,
      "grad_norm": 0.03904140740633011,
      "learning_rate": 3.384584580585883e-05,
      "loss": 0.0222,
      "mean_token_accuracy": 0.9925149798393249,
      "num_tokens": 28474314.0,
      "step": 350
    },
    {
      "epoch": 0.2154126213592233,
      "grad_norm": 0.036088310182094574,
      "learning_rate": 3.377685845928448e-05,
      "loss": 0.0208,
      "mean_token_accuracy": 0.9931179046630859,
      "num_tokens": 28881394.0,
      "step": 355
    },
    {
      "epoch": 0.21844660194174756,
      "grad_norm": 0.04056026414036751,
      "learning_rate": 3.370671680830471e-05,
      "loss": 0.0218,
      "mean_token_accuracy": 0.992766273021698,
      "num_tokens": 29287460.0,
      "step": 360
    },
    {
      "epoch": 0.22148058252427186,
      "grad_norm": 0.0464637354016304,
      "learning_rate": 3.363542791907854e-05,
      "loss": 0.0197,
      "mean_token_accuracy": 0.993265700340271,
      "num_tokens": 29694534.0,
      "step": 365
    },
    {
      "epoch": 0.22451456310679613,
      "grad_norm": 0.036997150629758835,
      "learning_rate": 3.356299897333927e-05,
      "loss": 0.0219,
      "mean_token_accuracy": 0.9928918242454529,
      "num_tokens": 30101092.0,
      "step": 370
    },
    {
      "epoch": 0.2275485436893204,
      "grad_norm": 0.04870880767703056,
      "learning_rate": 3.348943726767093e-05,
      "loss": 0.0204,
      "mean_token_accuracy": 0.9932775974273682,
      "num_tokens": 30507917.0,
      "step": 375
    },
    {
      "epoch": 0.23058252427184467,
      "grad_norm": 0.07872658967971802,
      "learning_rate": 3.3414750212773206e-05,
      "loss": 0.02,
      "mean_token_accuracy": 0.9932668089866639,
      "num_tokens": 30914865.0,
      "step": 380
    },
    {
      "epoch": 0.23361650485436894,
      "grad_norm": 0.03615431487560272,
      "learning_rate": 3.333894533271491e-05,
      "loss": 0.0193,
      "mean_token_accuracy": 0.9935308218002319,
      "num_tokens": 31321893.0,
      "step": 385
    },
    {
      "epoch": 0.2366504854368932,
      "grad_norm": 0.033446334302425385,
      "learning_rate": 3.326203026417599e-05,
      "loss": 0.0199,
      "mean_token_accuracy": 0.9932949662208557,
      "num_tokens": 31729278.0,
      "step": 390
    },
    {
      "epoch": 0.23968446601941748,
      "grad_norm": 0.03931819647550583,
      "learning_rate": 3.3184012755678186e-05,
      "loss": 0.0199,
      "mean_token_accuracy": 0.9932982683181762,
      "num_tokens": 32135795.0,
      "step": 395
    },
    {
      "epoch": 0.24271844660194175,
      "grad_norm": 0.036338433623313904,
      "learning_rate": 3.310490066680442e-05,
      "loss": 0.0238,
      "mean_token_accuracy": 0.9924678921699523,
      "num_tokens": 32542773.0,
      "step": 400
    },
    {
      "epoch": 0.24575242718446602,
      "grad_norm": 0.033566638827323914,
      "learning_rate": 3.302470196740707e-05,
      "loss": 0.0192,
      "mean_token_accuracy": 0.9934674024581909,
      "num_tokens": 32949527.0,
      "step": 405
    },
    {
      "epoch": 0.2487864077669903,
      "grad_norm": 0.044052544981241226,
      "learning_rate": 3.2943424736805e-05,
      "loss": 0.0208,
      "mean_token_accuracy": 0.9930399060249329,
      "num_tokens": 33356981.0,
      "step": 410
    },
    {
      "epoch": 0.2518203883495146,
      "grad_norm": 0.0594814233481884,
      "learning_rate": 3.28610771629697e-05,
      "loss": 0.0193,
      "mean_token_accuracy": 0.9934720158576965,
      "num_tokens": 33763975.0,
      "step": 415
    },
    {
      "epoch": 0.25485436893203883,
      "grad_norm": 0.06194014474749565,
      "learning_rate": 3.2777667541700405e-05,
      "loss": 0.0204,
      "mean_token_accuracy": 0.9930955171585083,
      "num_tokens": 34171521.0,
      "step": 420
    },
    {
      "epoch": 0.25788834951456313,
      "grad_norm": 0.03647923097014427,
      "learning_rate": 3.269320427578833e-05,
      "loss": 0.0217,
      "mean_token_accuracy": 0.9928809881210328,
      "num_tokens": 34578265.0,
      "step": 425
    },
    {
      "epoch": 0.2609223300970874,
      "grad_norm": 0.06534451991319656,
      "learning_rate": 3.260769587417021e-05,
      "loss": 0.0203,
      "mean_token_accuracy": 0.9933925867080688,
      "num_tokens": 34984780.0,
      "step": 430
    },
    {
      "epoch": 0.26395631067961167,
      "grad_norm": 0.03720612823963165,
      "learning_rate": 3.252115095107108e-05,
      "loss": 0.0222,
      "mean_token_accuracy": 0.9927129507064819,
      "num_tokens": 35390481.0,
      "step": 435
    },
    {
      "epoch": 0.2669902912621359,
      "grad_norm": 0.052582353353500366,
      "learning_rate": 3.243357822513647e-05,
      "loss": 0.0206,
      "mean_token_accuracy": 0.9931320190429688,
      "num_tokens": 35797895.0,
      "step": 440
    },
    {
      "epoch": 0.2700242718446602,
      "grad_norm": 0.035469647496938705,
      "learning_rate": 3.234498651855404e-05,
      "loss": 0.0227,
      "mean_token_accuracy": 0.9924560546875,
      "num_tokens": 36205302.0,
      "step": 445
    },
    {
      "epoch": 0.27305825242718446,
      "grad_norm": 0.030852986499667168,
      "learning_rate": 3.2255384756164905e-05,
      "loss": 0.02,
      "mean_token_accuracy": 0.9932003140449523,
      "num_tokens": 36611860.0,
      "step": 450
    },
    {
      "epoch": 0.27609223300970875,
      "grad_norm": 0.03644876554608345,
      "learning_rate": 3.216478196456445e-05,
      "loss": 0.019,
      "mean_token_accuracy": 0.9935380101203919,
      "num_tokens": 37018765.0,
      "step": 455
    },
    {
      "epoch": 0.279126213592233,
      "grad_norm": 0.037993211299180984,
      "learning_rate": 3.207318727119302e-05,
      "loss": 0.0219,
      "mean_token_accuracy": 0.992822015285492,
      "num_tokens": 37426212.0,
      "step": 460
    },
    {
      "epoch": 0.2821601941747573,
      "grad_norm": 0.03461167588829994,
      "learning_rate": 3.1980609903416415e-05,
      "loss": 0.0199,
      "mean_token_accuracy": 0.9932843446731567,
      "num_tokens": 37832810.0,
      "step": 465
    },
    {
      "epoch": 0.28519417475728154,
      "grad_norm": 0.03228127583861351,
      "learning_rate": 3.1887059187596306e-05,
      "loss": 0.0223,
      "mean_token_accuracy": 0.992818820476532,
      "num_tokens": 38239386.0,
      "step": 470
    },
    {
      "epoch": 0.28822815533980584,
      "grad_norm": 0.04857522249221802,
      "learning_rate": 3.179254454815066e-05,
      "loss": 0.0211,
      "mean_token_accuracy": 0.9930593371391296,
      "num_tokens": 38646309.0,
      "step": 475
    },
    {
      "epoch": 0.2912621359223301,
      "grad_norm": 0.04640183225274086,
      "learning_rate": 3.1697075506604365e-05,
      "loss": 0.0205,
      "mean_token_accuracy": 0.9930596828460694,
      "num_tokens": 39054017.0,
      "step": 480
    },
    {
      "epoch": 0.2942961165048544,
      "grad_norm": 0.07724732160568237,
      "learning_rate": 3.160066168062996e-05,
      "loss": 0.0184,
      "mean_token_accuracy": 0.9936440348625183,
      "num_tokens": 39461269.0,
      "step": 485
    },
    {
      "epoch": 0.2973300970873786,
      "grad_norm": 0.02800828032195568,
      "learning_rate": 3.150331278307879e-05,
      "loss": 0.0221,
      "mean_token_accuracy": 0.99279283285141,
      "num_tokens": 39868650.0,
      "step": 490
    },
    {
      "epoch": 0.3003640776699029,
      "grad_norm": 0.037310052663087845,
      "learning_rate": 3.140503862100249e-05,
      "loss": 0.0201,
      "mean_token_accuracy": 0.9932762980461121,
      "num_tokens": 40274804.0,
      "step": 495
    },
    {
      "epoch": 0.30339805825242716,
      "grad_norm": 0.06045287847518921,
      "learning_rate": 3.130584909466502e-05,
      "loss": 0.0224,
      "mean_token_accuracy": 0.9927901029586792,
      "num_tokens": 40680743.0,
      "step": 500
    },
    {
      "epoch": 0.30643203883495146,
      "grad_norm": 0.037607740610837936,
      "learning_rate": 3.1205754196545294e-05,
      "loss": 0.0222,
      "mean_token_accuracy": 0.9926772832870483,
      "num_tokens": 41087347.0,
      "step": 505
    },
    {
      "epoch": 0.3094660194174757,
      "grad_norm": 0.036764051765203476,
      "learning_rate": 3.110476401033054e-05,
      "loss": 0.0217,
      "mean_token_accuracy": 0.9927445411682129,
      "num_tokens": 41494644.0,
      "step": 510
    },
    {
      "epoch": 0.3125,
      "grad_norm": 0.03375716134905815,
      "learning_rate": 3.100288870990043e-05,
      "loss": 0.0196,
      "mean_token_accuracy": 0.9934267640113831,
      "num_tokens": 41902084.0,
      "step": 515
    },
    {
      "epoch": 0.3155339805825243,
      "grad_norm": 0.03154360502958298,
      "learning_rate": 3.0900138558302194e-05,
      "loss": 0.0191,
      "mean_token_accuracy": 0.9935309648513794,
      "num_tokens": 42309300.0,
      "step": 520
    },
    {
      "epoch": 0.31856796116504854,
      "grad_norm": 0.025135209783911705,
      "learning_rate": 3.079652390671665e-05,
      "loss": 0.0193,
      "mean_token_accuracy": 0.9934475421905518,
      "num_tokens": 42716542.0,
      "step": 525
    },
    {
      "epoch": 0.32160194174757284,
      "grad_norm": 0.03982224315404892,
      "learning_rate": 3.069205519341546e-05,
      "loss": 0.0193,
      "mean_token_accuracy": 0.9934385299682618,
      "num_tokens": 43123256.0,
      "step": 530
    },
    {
      "epoch": 0.3246359223300971,
      "grad_norm": 0.04023640230298042,
      "learning_rate": 3.058674294270955e-05,
      "loss": 0.0186,
      "mean_token_accuracy": 0.9937353134155273,
      "num_tokens": 43529652.0,
      "step": 535
    },
    {
      "epoch": 0.3276699029126214,
      "grad_norm": 0.028163542971014977,
      "learning_rate": 3.0480597763888863e-05,
      "loss": 0.02,
      "mean_token_accuracy": 0.993327522277832,
      "num_tokens": 43935601.0,
      "step": 540
    },
    {
      "epoch": 0.3307038834951456,
      "grad_norm": 0.03361319378018379,
      "learning_rate": 3.0373630350153597e-05,
      "loss": 0.0175,
      "mean_token_accuracy": 0.993952751159668,
      "num_tokens": 44342453.0,
      "step": 545
    },
    {
      "epoch": 0.3337378640776699,
      "grad_norm": 0.035180699080228806,
      "learning_rate": 3.026585147753692e-05,
      "loss": 0.0226,
      "mean_token_accuracy": 0.9926326274871826,
      "num_tokens": 44748951.0,
      "step": 550
    },
    {
      "epoch": 0.33677184466019416,
      "grad_norm": 0.04069613292813301,
      "learning_rate": 3.0157272003819414e-05,
      "loss": 0.0207,
      "mean_token_accuracy": 0.9931874632835388,
      "num_tokens": 45155188.0,
      "step": 555
    },
    {
      "epoch": 0.33980582524271846,
      "grad_norm": 0.029416020959615707,
      "learning_rate": 3.0047902867435223e-05,
      "loss": 0.0224,
      "mean_token_accuracy": 0.9927819728851318,
      "num_tokens": 45561807.0,
      "step": 560
    },
    {
      "epoch": 0.3428398058252427,
      "grad_norm": 0.035775549709796906,
      "learning_rate": 2.993775508637012e-05,
      "loss": 0.0227,
      "mean_token_accuracy": 0.9924745202064514,
      "num_tokens": 45968255.0,
      "step": 565
    },
    {
      "epoch": 0.345873786407767,
      "grad_norm": 0.039437539875507355,
      "learning_rate": 2.9826839757051526e-05,
      "loss": 0.0197,
      "mean_token_accuracy": 0.9934724569320679,
      "num_tokens": 46375144.0,
      "step": 570
    },
    {
      "epoch": 0.34890776699029125,
      "grad_norm": 0.03149618208408356,
      "learning_rate": 2.971516805323066e-05,
      "loss": 0.0214,
      "mean_token_accuracy": 0.992820394039154,
      "num_tokens": 46782323.0,
      "step": 575
    },
    {
      "epoch": 0.35194174757281554,
      "grad_norm": 0.03710157796740532,
      "learning_rate": 2.9602751224856852e-05,
      "loss": 0.0199,
      "mean_token_accuracy": 0.9933429002761841,
      "num_tokens": 47188896.0,
      "step": 580
    },
    {
      "epoch": 0.3549757281553398,
      "grad_norm": 0.03251684457063675,
      "learning_rate": 2.9489600596944236e-05,
      "loss": 0.0198,
      "mean_token_accuracy": 0.9933558225631713,
      "num_tokens": 47595402.0,
      "step": 585
    },
    {
      "epoch": 0.3580097087378641,
      "grad_norm": 0.024846093729138374,
      "learning_rate": 2.937572756843084e-05,
      "loss": 0.0208,
      "mean_token_accuracy": 0.9929834604263306,
      "num_tokens": 48002676.0,
      "step": 590
    },
    {
      "epoch": 0.36104368932038833,
      "grad_norm": 0.03847256675362587,
      "learning_rate": 2.9261143611030253e-05,
      "loss": 0.0197,
      "mean_token_accuracy": 0.9932600855827332,
      "num_tokens": 48408909.0,
      "step": 595
    },
    {
      "epoch": 0.3640776699029126,
      "grad_norm": 0.03794528916478157,
      "learning_rate": 2.914586026807594e-05,
      "loss": 0.0201,
      "mean_token_accuracy": 0.9932417273521423,
      "num_tokens": 48816050.0,
      "step": 600
    },
    {
      "epoch": 0.36711165048543687,
      "grad_norm": 0.03913505747914314,
      "learning_rate": 2.902988915335836e-05,
      "loss": 0.0207,
      "mean_token_accuracy": 0.9929919958114624,
      "num_tokens": 49223367.0,
      "step": 605
    },
    {
      "epoch": 0.37014563106796117,
      "grad_norm": 0.03473582863807678,
      "learning_rate": 2.8913241949954957e-05,
      "loss": 0.0221,
      "mean_token_accuracy": 0.9925926566123963,
      "num_tokens": 49630760.0,
      "step": 610
    },
    {
      "epoch": 0.3731796116504854,
      "grad_norm": 0.027801785618066788,
      "learning_rate": 2.8795930409053225e-05,
      "loss": 0.02,
      "mean_token_accuracy": 0.9932288885116577,
      "num_tokens": 50037598.0,
      "step": 615
    },
    {
      "epoch": 0.3762135922330097,
      "grad_norm": 0.0316888689994812,
      "learning_rate": 2.867796634876686e-05,
      "loss": 0.0184,
      "mean_token_accuracy": 0.9937638401985168,
      "num_tokens": 50444771.0,
      "step": 620
    },
    {
      "epoch": 0.379247572815534,
      "grad_norm": 0.07504960894584656,
      "learning_rate": 2.8559361652945176e-05,
      "loss": 0.0199,
      "mean_token_accuracy": 0.9932967185974121,
      "num_tokens": 50851588.0,
      "step": 625
    },
    {
      "epoch": 0.38228155339805825,
      "grad_norm": 0.045697640627622604,
      "learning_rate": 2.8440128269975945e-05,
      "loss": 0.0187,
      "mean_token_accuracy": 0.9934902310371398,
      "num_tokens": 51258940.0,
      "step": 630
    },
    {
      "epoch": 0.38531553398058255,
      "grad_norm": 0.03645659238100052,
      "learning_rate": 2.8320278211581675e-05,
      "loss": 0.0208,
      "mean_token_accuracy": 0.9932755589485168,
      "num_tokens": 51665862.0,
      "step": 635
    },
    {
      "epoch": 0.3883495145631068,
      "grad_norm": 0.03291779011487961,
      "learning_rate": 2.8199823551609523e-05,
      "loss": 0.0242,
      "mean_token_accuracy": 0.9920920610427857,
      "num_tokens": 52072805.0,
      "step": 640
    },
    {
      "epoch": 0.3913834951456311,
      "grad_norm": 0.03070950135588646,
      "learning_rate": 2.8078776424815e-05,
      "loss": 0.0209,
      "mean_token_accuracy": 0.9930874943733216,
      "num_tokens": 52479343.0,
      "step": 645
    },
    {
      "epoch": 0.39441747572815533,
      "grad_norm": 0.040879931300878525,
      "learning_rate": 2.7957149025639455e-05,
      "loss": 0.0214,
      "mean_token_accuracy": 0.9928395271301269,
      "num_tokens": 52886647.0,
      "step": 650
    },
    {
      "epoch": 0.39745145631067963,
      "grad_norm": 0.035599879920482635,
      "learning_rate": 2.7834953606981647e-05,
      "loss": 0.0213,
      "mean_token_accuracy": 0.9928174138069152,
      "num_tokens": 53292846.0,
      "step": 655
    },
    {
      "epoch": 0.40048543689320387,
      "grad_norm": 0.03400155529379845,
      "learning_rate": 2.77122024789633e-05,
      "loss": 0.0199,
      "mean_token_accuracy": 0.9932538151741028,
      "num_tokens": 53700258.0,
      "step": 660
    },
    {
      "epoch": 0.40351941747572817,
      "grad_norm": 0.028687704354524612,
      "learning_rate": 2.7588908007689038e-05,
      "loss": 0.0186,
      "mean_token_accuracy": 0.9934426546096802,
      "num_tokens": 54107958.0,
      "step": 665
    },
    {
      "epoch": 0.4065533980582524,
      "grad_norm": 0.03738875687122345,
      "learning_rate": 2.7465082614000553e-05,
      "loss": 0.0224,
      "mean_token_accuracy": 0.9925989031791687,
      "num_tokens": 54514336.0,
      "step": 670
    },
    {
      "epoch": 0.4095873786407767,
      "grad_norm": 0.03134342283010483,
      "learning_rate": 2.734073877222535e-05,
      "loss": 0.0192,
      "mean_token_accuracy": 0.9933988332748414,
      "num_tokens": 54921552.0,
      "step": 675
    },
    {
      "epoch": 0.41262135922330095,
      "grad_norm": 0.026389850303530693,
      "learning_rate": 2.7215889008920038e-05,
      "loss": 0.0214,
      "mean_token_accuracy": 0.9930248975753784,
      "num_tokens": 55327241.0,
      "step": 680
    },
    {
      "epoch": 0.41565533980582525,
      "grad_norm": 0.022707246243953705,
      "learning_rate": 2.7090545901608423e-05,
      "loss": 0.0197,
      "mean_token_accuracy": 0.9931993603706359,
      "num_tokens": 55734366.0,
      "step": 685
    },
    {
      "epoch": 0.4186893203883495,
      "grad_norm": 0.042725060135126114,
      "learning_rate": 2.6964722077514394e-05,
      "loss": 0.024,
      "mean_token_accuracy": 0.9923825621604919,
      "num_tokens": 56140445.0,
      "step": 690
    },
    {
      "epoch": 0.4217233009708738,
      "grad_norm": 0.055999163538217545,
      "learning_rate": 2.6838430212289874e-05,
      "loss": 0.0236,
      "mean_token_accuracy": 0.9923421740531921,
      "num_tokens": 56546727.0,
      "step": 695
    },
    {
      "epoch": 0.42475728155339804,
      "grad_norm": 0.09104075282812119,
      "learning_rate": 2.6711683028737836e-05,
      "loss": 0.0211,
      "mean_token_accuracy": 0.9931105375289917,
      "num_tokens": 56953430.0,
      "step": 700
    },
    {
      "epoch": 0.42779126213592233,
      "grad_norm": 0.03057127818465233,
      "learning_rate": 2.6584493295530605e-05,
      "loss": 0.0195,
      "mean_token_accuracy": 0.9935212850570678,
      "num_tokens": 57359890.0,
      "step": 705
    },
    {
      "epoch": 0.4308252427184466,
      "grad_norm": 0.039992984384298325,
      "learning_rate": 2.6456873825923518e-05,
      "loss": 0.0203,
      "mean_token_accuracy": 0.9931409597396851,
      "num_tokens": 57766561.0,
      "step": 710
    },
    {
      "epoch": 0.4338592233009709,
      "grad_norm": 0.03372696414589882,
      "learning_rate": 2.63288374764641e-05,
      "loss": 0.0205,
      "mean_token_accuracy": 0.9930363297462463,
      "num_tokens": 58173113.0,
      "step": 715
    },
    {
      "epoch": 0.4368932038834951,
      "grad_norm": 0.020909098908305168,
      "learning_rate": 2.620039714569688e-05,
      "loss": 0.0194,
      "mean_token_accuracy": 0.9933443427085876,
      "num_tokens": 58580501.0,
      "step": 720
    },
    {
      "epoch": 0.4399271844660194,
      "grad_norm": 0.02915063500404358,
      "learning_rate": 2.6071565772863995e-05,
      "loss": 0.0192,
      "mean_token_accuracy": 0.9934536218643188,
      "num_tokens": 58986906.0,
      "step": 725
    },
    {
      "epoch": 0.4429611650485437,
      "grad_norm": 0.03697626665234566,
      "learning_rate": 2.594235633660163e-05,
      "loss": 0.0231,
      "mean_token_accuracy": 0.9925421714782715,
      "num_tokens": 59393310.0,
      "step": 730
    },
    {
      "epoch": 0.44599514563106796,
      "grad_norm": 0.028796004131436348,
      "learning_rate": 2.5812781853632587e-05,
      "loss": 0.0219,
      "mean_token_accuracy": 0.9928050398826599,
      "num_tokens": 59799208.0,
      "step": 735
    },
    {
      "epoch": 0.44902912621359226,
      "grad_norm": 0.03447737544775009,
      "learning_rate": 2.5682855377454918e-05,
      "loss": 0.0184,
      "mean_token_accuracy": 0.9936539769172669,
      "num_tokens": 60206416.0,
      "step": 740
    },
    {
      "epoch": 0.4520631067961165,
      "grad_norm": 0.09332480281591415,
      "learning_rate": 2.555258999702692e-05,
      "loss": 0.0231,
      "mean_token_accuracy": 0.992510449886322,
      "num_tokens": 60612501.0,
      "step": 745
    },
    {
      "epoch": 0.4550970873786408,
      "grad_norm": 0.13943925499916077,
      "learning_rate": 2.5421998835448545e-05,
      "loss": 0.0214,
      "mean_token_accuracy": 0.9927754759788513,
      "num_tokens": 61019113.0,
      "step": 750
    },
    {
      "epoch": 0.45813106796116504,
      "grad_norm": 0.029267804697155952,
      "learning_rate": 2.5291095048639348e-05,
      "loss": 0.019,
      "mean_token_accuracy": 0.9934199333190918,
      "num_tokens": 61425726.0,
      "step": 755
    },
    {
      "epoch": 0.46116504854368934,
      "grad_norm": 0.03230291232466698,
      "learning_rate": 2.5159891824013134e-05,
      "loss": 0.0218,
      "mean_token_accuracy": 0.9925872683525085,
      "num_tokens": 61831973.0,
      "step": 760
    },
    {
      "epoch": 0.4641990291262136,
      "grad_norm": 0.024981655180454254,
      "learning_rate": 2.502840237914948e-05,
      "loss": 0.0189,
      "mean_token_accuracy": 0.9935161352157593,
      "num_tokens": 62238418.0,
      "step": 765
    },
    {
      "epoch": 0.4672330097087379,
      "grad_norm": 0.022092267870903015,
      "learning_rate": 2.4896639960462126e-05,
      "loss": 0.0185,
      "mean_token_accuracy": 0.9936355710029602,
      "num_tokens": 62645878.0,
      "step": 770
    },
    {
      "epoch": 0.4702669902912621,
      "grad_norm": 0.03437012434005737,
      "learning_rate": 2.476461784186456e-05,
      "loss": 0.0202,
      "mean_token_accuracy": 0.9932804703712463,
      "num_tokens": 63053187.0,
      "step": 775
    },
    {
      "epoch": 0.4733009708737864,
      "grad_norm": 0.039246056228876114,
      "learning_rate": 2.463234932343276e-05,
      "loss": 0.022,
      "mean_token_accuracy": 0.9929169893264771,
      "num_tokens": 63459015.0,
      "step": 780
    }
  ],
  "logging_steps": 5,
  "max_steps": 1648,
  "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": 1.4897968655735194e+18,
  "train_batch_size": 20,
  "trial_name": null,
  "trial_params": null
}