{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.5944244604316546,
  "eval_steps": 500,
  "global_step": 661,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0044964028776978415,
      "grad_norm": 34.8267936706543,
      "learning_rate": 1.6634586466165415e-06,
      "loss": 0.3603,
      "mean_token_accuracy": 0.8805555582046509,
      "num_tokens": 466767.0,
      "step": 5
    },
    {
      "epoch": 0.008992805755395683,
      "grad_norm": 1.9446816444396973,
      "learning_rate": 3.7427819548872183e-06,
      "loss": 0.1555,
      "mean_token_accuracy": 0.9567708373069763,
      "num_tokens": 937804.0,
      "step": 10
    },
    {
      "epoch": 0.013489208633093525,
      "grad_norm": 0.79410719871521,
      "learning_rate": 5.822105263157896e-06,
      "loss": 0.1184,
      "mean_token_accuracy": 0.9576388835906983,
      "num_tokens": 1408105.0,
      "step": 15
    },
    {
      "epoch": 0.017985611510791366,
      "grad_norm": 0.5585373640060425,
      "learning_rate": 7.901428571428572e-06,
      "loss": 0.1104,
      "mean_token_accuracy": 0.9598958492279053,
      "num_tokens": 1875896.0,
      "step": 20
    },
    {
      "epoch": 0.022482014388489208,
      "grad_norm": 0.6362137198448181,
      "learning_rate": 9.98075187969925e-06,
      "loss": 0.1046,
      "mean_token_accuracy": 0.9623263955116272,
      "num_tokens": 2342344.0,
      "step": 25
    },
    {
      "epoch": 0.02697841726618705,
      "grad_norm": 0.817181408405304,
      "learning_rate": 1.2060075187969928e-05,
      "loss": 0.1047,
      "mean_token_accuracy": 0.9623264074325562,
      "num_tokens": 2810563.0,
      "step": 30
    },
    {
      "epoch": 0.03147482014388489,
      "grad_norm": 0.7457462549209595,
      "learning_rate": 1.4139398496240603e-05,
      "loss": 0.1101,
      "mean_token_accuracy": 0.9598958373069764,
      "num_tokens": 3281330.0,
      "step": 35
    },
    {
      "epoch": 0.03597122302158273,
      "grad_norm": 0.5849624872207642,
      "learning_rate": 1.4555173218834344e-05,
      "loss": 0.102,
      "mean_token_accuracy": 0.9642361164093017,
      "num_tokens": 3752355.0,
      "step": 40
    },
    {
      "epoch": 0.040467625899280574,
      "grad_norm": 0.5536819100379944,
      "learning_rate": 1.4554807846481353e-05,
      "loss": 0.1003,
      "mean_token_accuracy": 0.9621527791023254,
      "num_tokens": 4218619.0,
      "step": 45
    },
    {
      "epoch": 0.044964028776978415,
      "grad_norm": 0.5768638253211975,
      "learning_rate": 1.4554161438444059e-05,
      "loss": 0.1054,
      "mean_token_accuracy": 0.9625000119209289,
      "num_tokens": 4687504.0,
      "step": 50
    },
    {
      "epoch": 0.04946043165467626,
      "grad_norm": 0.5772081017494202,
      "learning_rate": 1.455323402800781e-05,
      "loss": 0.0986,
      "mean_token_accuracy": 0.9633680701255798,
      "num_tokens": 5156586.0,
      "step": 55
    },
    {
      "epoch": 0.0539568345323741,
      "grad_norm": 0.6602324843406677,
      "learning_rate": 1.455202566292755e-05,
      "loss": 0.1082,
      "mean_token_accuracy": 0.9595486283302307,
      "num_tokens": 5629757.0,
      "step": 60
    },
    {
      "epoch": 0.05845323741007194,
      "grad_norm": 0.5798962116241455,
      "learning_rate": 1.4550536405425362e-05,
      "loss": 0.0927,
      "mean_token_accuracy": 0.9675347328186035,
      "num_tokens": 6095752.0,
      "step": 65
    },
    {
      "epoch": 0.06294964028776978,
      "grad_norm": 0.6183492541313171,
      "learning_rate": 1.4548766332187261e-05,
      "loss": 0.0997,
      "mean_token_accuracy": 0.9645833373069763,
      "num_tokens": 6568280.0,
      "step": 70
    },
    {
      "epoch": 0.06744604316546762,
      "grad_norm": 0.5590345859527588,
      "learning_rate": 1.4546715534359241e-05,
      "loss": 0.0973,
      "mean_token_accuracy": 0.964062511920929,
      "num_tokens": 7039631.0,
      "step": 75
    },
    {
      "epoch": 0.07194244604316546,
      "grad_norm": 0.5744739770889282,
      "learning_rate": 1.4544384117542591e-05,
      "loss": 0.1004,
      "mean_token_accuracy": 0.965625011920929,
      "num_tokens": 7507809.0,
      "step": 80
    },
    {
      "epoch": 0.0764388489208633,
      "grad_norm": 0.4603578448295593,
      "learning_rate": 1.4541772201788454e-05,
      "loss": 0.0963,
      "mean_token_accuracy": 0.9656250238418579,
      "num_tokens": 7975490.0,
      "step": 85
    },
    {
      "epoch": 0.08093525179856115,
      "grad_norm": 0.47225049138069153,
      "learning_rate": 1.453887992159164e-05,
      "loss": 0.0948,
      "mean_token_accuracy": 0.9659722328186036,
      "num_tokens": 8445461.0,
      "step": 90
    },
    {
      "epoch": 0.08543165467625899,
      "grad_norm": 0.47965773940086365,
      "learning_rate": 1.453570742588371e-05,
      "loss": 0.0972,
      "mean_token_accuracy": 0.9663194417953491,
      "num_tokens": 8912561.0,
      "step": 95
    },
    {
      "epoch": 0.08992805755395683,
      "grad_norm": 0.5365564823150635,
      "learning_rate": 1.4532254878025292e-05,
      "loss": 0.0998,
      "mean_token_accuracy": 0.9631944537162781,
      "num_tokens": 9382565.0,
      "step": 100
    },
    {
      "epoch": 0.09442446043165467,
      "grad_norm": 0.522896409034729,
      "learning_rate": 1.4528522455797692e-05,
      "loss": 0.0962,
      "mean_token_accuracy": 0.964062511920929,
      "num_tokens": 9851148.0,
      "step": 105
    },
    {
      "epoch": 0.09892086330935251,
      "grad_norm": 0.48392146825790405,
      "learning_rate": 1.4524510351393716e-05,
      "loss": 0.0915,
      "mean_token_accuracy": 0.9649305582046509,
      "num_tokens": 10318813.0,
      "step": 110
    },
    {
      "epoch": 0.10341726618705036,
      "grad_norm": 0.42742612957954407,
      "learning_rate": 1.4520218771407784e-05,
      "loss": 0.0968,
      "mean_token_accuracy": 0.9649305582046509,
      "num_tokens": 10785036.0,
      "step": 115
    },
    {
      "epoch": 0.1079136690647482,
      "grad_norm": 0.3914225995540619,
      "learning_rate": 1.4515647936825296e-05,
      "loss": 0.0967,
      "mean_token_accuracy": 0.964062511920929,
      "num_tokens": 11253858.0,
      "step": 120
    },
    {
      "epoch": 0.11241007194244604,
      "grad_norm": 0.4432506263256073,
      "learning_rate": 1.451079808301125e-05,
      "loss": 0.0985,
      "mean_token_accuracy": 0.9640625238418579,
      "num_tokens": 11722148.0,
      "step": 125
    },
    {
      "epoch": 0.11690647482014388,
      "grad_norm": 0.5651817917823792,
      "learning_rate": 1.4505669459698112e-05,
      "loss": 0.0937,
      "mean_token_accuracy": 0.9680555820465088,
      "num_tokens": 12192766.0,
      "step": 130
    },
    {
      "epoch": 0.12140287769784172,
      "grad_norm": 0.46366673707962036,
      "learning_rate": 1.4500262330972978e-05,
      "loss": 0.0912,
      "mean_token_accuracy": 0.9659722208976745,
      "num_tokens": 12660483.0,
      "step": 135
    },
    {
      "epoch": 0.12589928057553956,
      "grad_norm": 0.42186692357063293,
      "learning_rate": 1.4494576975263952e-05,
      "loss": 0.0962,
      "mean_token_accuracy": 0.9659722208976745,
      "num_tokens": 13127525.0,
      "step": 140
    },
    {
      "epoch": 0.1303956834532374,
      "grad_norm": 0.3628046214580536,
      "learning_rate": 1.448861368532582e-05,
      "loss": 0.0979,
      "mean_token_accuracy": 0.9630208373069763,
      "num_tokens": 13596548.0,
      "step": 145
    },
    {
      "epoch": 0.13489208633093525,
      "grad_norm": 0.39447155594825745,
      "learning_rate": 1.4482372768224984e-05,
      "loss": 0.0953,
      "mean_token_accuracy": 0.9638889074325562,
      "num_tokens": 14065589.0,
      "step": 150
    },
    {
      "epoch": 0.1393884892086331,
      "grad_norm": 0.41298848390579224,
      "learning_rate": 1.4475854545323637e-05,
      "loss": 0.0978,
      "mean_token_accuracy": 0.9649305582046509,
      "num_tokens": 14534593.0,
      "step": 155
    },
    {
      "epoch": 0.14388489208633093,
      "grad_norm": 0.5090773701667786,
      "learning_rate": 1.4469059352263213e-05,
      "loss": 0.092,
      "mean_token_accuracy": 0.9671875238418579,
      "num_tokens": 15004794.0,
      "step": 160
    },
    {
      "epoch": 0.14838129496402877,
      "grad_norm": 0.5376242995262146,
      "learning_rate": 1.4461987538947123e-05,
      "loss": 0.0965,
      "mean_token_accuracy": 0.9642361283302308,
      "num_tokens": 15472592.0,
      "step": 165
    },
    {
      "epoch": 0.1528776978417266,
      "grad_norm": 0.5043388605117798,
      "learning_rate": 1.445463946952271e-05,
      "loss": 0.09,
      "mean_token_accuracy": 0.9692708492279053,
      "num_tokens": 15940501.0,
      "step": 170
    },
    {
      "epoch": 0.15737410071942445,
      "grad_norm": 0.49153539538383484,
      "learning_rate": 1.4447015522362527e-05,
      "loss": 0.0953,
      "mean_token_accuracy": 0.9645833611488343,
      "num_tokens": 16409778.0,
      "step": 175
    },
    {
      "epoch": 0.1618705035971223,
      "grad_norm": 0.5271637439727783,
      "learning_rate": 1.4439116090044831e-05,
      "loss": 0.0937,
      "mean_token_accuracy": 0.9659722447395325,
      "num_tokens": 16880259.0,
      "step": 180
    },
    {
      "epoch": 0.16636690647482014,
      "grad_norm": 0.4684729278087616,
      "learning_rate": 1.4430941579333378e-05,
      "loss": 0.0906,
      "mean_token_accuracy": 0.9690972328186035,
      "num_tokens": 17349465.0,
      "step": 185
    },
    {
      "epoch": 0.17086330935251798,
      "grad_norm": 0.4850688874721527,
      "learning_rate": 1.442249241115648e-05,
      "loss": 0.0902,
      "mean_token_accuracy": 0.9680555701255799,
      "num_tokens": 17816904.0,
      "step": 190
    },
    {
      "epoch": 0.17535971223021582,
      "grad_norm": 0.44540664553642273,
      "learning_rate": 1.4413769020585319e-05,
      "loss": 0.0819,
      "mean_token_accuracy": 0.9680555701255799,
      "num_tokens": 18284016.0,
      "step": 195
    },
    {
      "epoch": 0.17985611510791366,
      "grad_norm": 0.5207117199897766,
      "learning_rate": 1.4404771856811558e-05,
      "loss": 0.0879,
      "mean_token_accuracy": 0.9677083373069764,
      "num_tokens": 18751285.0,
      "step": 200
    },
    {
      "epoch": 0.1843525179856115,
      "grad_norm": 0.4022666811943054,
      "learning_rate": 1.4395501383124202e-05,
      "loss": 0.0861,
      "mean_token_accuracy": 0.9689236164093018,
      "num_tokens": 19219382.0,
      "step": 205
    },
    {
      "epoch": 0.18884892086330934,
      "grad_norm": 0.4309217631816864,
      "learning_rate": 1.4385958076885744e-05,
      "loss": 0.0959,
      "mean_token_accuracy": 0.9657986283302307,
      "num_tokens": 19689263.0,
      "step": 210
    },
    {
      "epoch": 0.19334532374100719,
      "grad_norm": 0.47961878776550293,
      "learning_rate": 1.4376142429507583e-05,
      "loss": 0.0935,
      "mean_token_accuracy": 0.9659722208976745,
      "num_tokens": 20159528.0,
      "step": 215
    },
    {
      "epoch": 0.19784172661870503,
      "grad_norm": 0.688991904258728,
      "learning_rate": 1.4366054946424724e-05,
      "loss": 0.0906,
      "mean_token_accuracy": 0.9696180582046509,
      "num_tokens": 20630783.0,
      "step": 220
    },
    {
      "epoch": 0.20233812949640287,
      "grad_norm": 0.4284147620201111,
      "learning_rate": 1.4355696147069747e-05,
      "loss": 0.0879,
      "mean_token_accuracy": 0.9678819656372071,
      "num_tokens": 21097847.0,
      "step": 225
    },
    {
      "epoch": 0.2068345323741007,
      "grad_norm": 0.43436649441719055,
      "learning_rate": 1.4345066564846065e-05,
      "loss": 0.0982,
      "mean_token_accuracy": 0.9640625,
      "num_tokens": 21568414.0,
      "step": 230
    },
    {
      "epoch": 0.21133093525179855,
      "grad_norm": 0.4775388836860657,
      "learning_rate": 1.4334166747100448e-05,
      "loss": 0.0864,
      "mean_token_accuracy": 0.9689236164093018,
      "num_tokens": 22038261.0,
      "step": 235
    },
    {
      "epoch": 0.2158273381294964,
      "grad_norm": 0.4478053152561188,
      "learning_rate": 1.4322997255094853e-05,
      "loss": 0.0915,
      "mean_token_accuracy": 0.9663194417953491,
      "num_tokens": 22507928.0,
      "step": 240
    },
    {
      "epoch": 0.22032374100719423,
      "grad_norm": 0.5618180632591248,
      "learning_rate": 1.4311558663977512e-05,
      "loss": 0.0948,
      "mean_token_accuracy": 0.9670138955116272,
      "num_tokens": 22977356.0,
      "step": 245
    },
    {
      "epoch": 0.22482014388489208,
      "grad_norm": 0.45267704129219055,
      "learning_rate": 1.4299851562753316e-05,
      "loss": 0.0939,
      "mean_token_accuracy": 0.9675347447395325,
      "num_tokens": 23445012.0,
      "step": 250
    },
    {
      "epoch": 0.22931654676258992,
      "grad_norm": 0.4265112578868866,
      "learning_rate": 1.4287876554253494e-05,
      "loss": 0.0923,
      "mean_token_accuracy": 0.9666666746139526,
      "num_tokens": 23918138.0,
      "step": 255
    },
    {
      "epoch": 0.23381294964028776,
      "grad_norm": 0.46962136030197144,
      "learning_rate": 1.4275634255104564e-05,
      "loss": 0.0903,
      "mean_token_accuracy": 0.9654513955116272,
      "num_tokens": 24389671.0,
      "step": 260
    },
    {
      "epoch": 0.2383093525179856,
      "grad_norm": 0.3906201720237732,
      "learning_rate": 1.4263125295696585e-05,
      "loss": 0.0854,
      "mean_token_accuracy": 0.9684027791023254,
      "num_tokens": 24862914.0,
      "step": 265
    },
    {
      "epoch": 0.24280575539568344,
      "grad_norm": 0.4527030289173126,
      "learning_rate": 1.4250350320150692e-05,
      "loss": 0.089,
      "mean_token_accuracy": 0.9682291626930237,
      "num_tokens": 25331420.0,
      "step": 270
    },
    {
      "epoch": 0.24730215827338128,
      "grad_norm": 0.40908169746398926,
      "learning_rate": 1.4237309986285931e-05,
      "loss": 0.0876,
      "mean_token_accuracy": 0.9682291626930237,
      "num_tokens": 25799898.0,
      "step": 275
    },
    {
      "epoch": 0.2517985611510791,
      "grad_norm": 0.4546169638633728,
      "learning_rate": 1.422400496558539e-05,
      "loss": 0.0875,
      "mean_token_accuracy": 0.9692708253860474,
      "num_tokens": 26272817.0,
      "step": 280
    },
    {
      "epoch": 0.256294964028777,
      "grad_norm": 0.4130357503890991,
      "learning_rate": 1.4210435943161616e-05,
      "loss": 0.0903,
      "mean_token_accuracy": 0.9682291746139526,
      "num_tokens": 26740515.0,
      "step": 285
    },
    {
      "epoch": 0.2607913669064748,
      "grad_norm": 0.4100114405155182,
      "learning_rate": 1.4196603617721347e-05,
      "loss": 0.0901,
      "mean_token_accuracy": 0.9684027910232544,
      "num_tokens": 27210026.0,
      "step": 290
    },
    {
      "epoch": 0.2652877697841727,
      "grad_norm": 0.46927815675735474,
      "learning_rate": 1.4182508701529513e-05,
      "loss": 0.0839,
      "mean_token_accuracy": 0.9690972447395325,
      "num_tokens": 27678965.0,
      "step": 295
    },
    {
      "epoch": 0.2697841726618705,
      "grad_norm": 0.4076975882053375,
      "learning_rate": 1.4168151920372579e-05,
      "loss": 0.0821,
      "mean_token_accuracy": 0.969444465637207,
      "num_tokens": 28149371.0,
      "step": 300
    },
    {
      "epoch": 0.27428057553956836,
      "grad_norm": 0.36228322982788086,
      "learning_rate": 1.415353401352117e-05,
      "loss": 0.088,
      "mean_token_accuracy": 0.9684027791023254,
      "num_tokens": 28614761.0,
      "step": 305
    },
    {
      "epoch": 0.2787769784172662,
      "grad_norm": 0.4045368731021881,
      "learning_rate": 1.4138655733691993e-05,
      "loss": 0.0876,
      "mean_token_accuracy": 0.9696180582046509,
      "num_tokens": 29084694.0,
      "step": 310
    },
    {
      "epoch": 0.28327338129496404,
      "grad_norm": 0.41623398661613464,
      "learning_rate": 1.4123517847009096e-05,
      "loss": 0.0812,
      "mean_token_accuracy": 0.970312488079071,
      "num_tokens": 29554086.0,
      "step": 315
    },
    {
      "epoch": 0.28776978417266186,
      "grad_norm": 0.4583008289337158,
      "learning_rate": 1.4108121132964388e-05,
      "loss": 0.0883,
      "mean_token_accuracy": 0.9680555582046508,
      "num_tokens": 30024350.0,
      "step": 320
    },
    {
      "epoch": 0.2922661870503597,
      "grad_norm": 0.4604341685771942,
      "learning_rate": 1.4092466384377532e-05,
      "loss": 0.0791,
      "mean_token_accuracy": 0.971006965637207,
      "num_tokens": 30491759.0,
      "step": 325
    },
    {
      "epoch": 0.29676258992805754,
      "grad_norm": 0.4965105950832367,
      "learning_rate": 1.4076554407355108e-05,
      "loss": 0.0854,
      "mean_token_accuracy": 0.9703125,
      "num_tokens": 30959969.0,
      "step": 330
    },
    {
      "epoch": 0.3012589928057554,
      "grad_norm": 0.36111050844192505,
      "learning_rate": 1.4060386021249099e-05,
      "loss": 0.0844,
      "mean_token_accuracy": 0.9715278029441834,
      "num_tokens": 31430407.0,
      "step": 335
    },
    {
      "epoch": 0.3057553956834532,
      "grad_norm": 0.4066219925880432,
      "learning_rate": 1.4043962058614708e-05,
      "loss": 0.0788,
      "mean_token_accuracy": 0.9732638835906983,
      "num_tokens": 31900742.0,
      "step": 340
    },
    {
      "epoch": 0.3102517985611511,
      "grad_norm": 0.40719839930534363,
      "learning_rate": 1.4027283365167485e-05,
      "loss": 0.0951,
      "mean_token_accuracy": 0.9659722328186036,
      "num_tokens": 32373021.0,
      "step": 345
    },
    {
      "epoch": 0.3147482014388489,
      "grad_norm": 0.46199360489845276,
      "learning_rate": 1.401035079973978e-05,
      "loss": 0.0883,
      "mean_token_accuracy": 0.9671875,
      "num_tokens": 32840444.0,
      "step": 350
    },
    {
      "epoch": 0.3192446043165468,
      "grad_norm": 0.5061129331588745,
      "learning_rate": 1.399316523423651e-05,
      "loss": 0.0906,
      "mean_token_accuracy": 0.9661458492279053,
      "num_tokens": 33311175.0,
      "step": 355
    },
    {
      "epoch": 0.3237410071942446,
      "grad_norm": 0.39469435811042786,
      "learning_rate": 1.397572755359028e-05,
      "loss": 0.0879,
      "mean_token_accuracy": 0.9696180582046509,
      "num_tokens": 33781258.0,
      "step": 360
    },
    {
      "epoch": 0.32823741007194246,
      "grad_norm": 0.39724910259246826,
      "learning_rate": 1.39580386557158e-05,
      "loss": 0.0872,
      "mean_token_accuracy": 0.9685764074325561,
      "num_tokens": 34250531.0,
      "step": 365
    },
    {
      "epoch": 0.3327338129496403,
      "grad_norm": 0.45710304379463196,
      "learning_rate": 1.3940099451463658e-05,
      "loss": 0.0855,
      "mean_token_accuracy": 0.9699652791023254,
      "num_tokens": 34720407.0,
      "step": 370
    },
    {
      "epoch": 0.33723021582733814,
      "grad_norm": 0.45646730065345764,
      "learning_rate": 1.392191086457341e-05,
      "loss": 0.0785,
      "mean_token_accuracy": 0.9746527910232544,
      "num_tokens": 35186721.0,
      "step": 375
    },
    {
      "epoch": 0.34172661870503596,
      "grad_norm": 0.4192124605178833,
      "learning_rate": 1.3903473831626019e-05,
      "loss": 0.0891,
      "mean_token_accuracy": 0.9710069537162781,
      "num_tokens": 35654890.0,
      "step": 380
    },
    {
      "epoch": 0.3462230215827338,
      "grad_norm": 0.394729346036911,
      "learning_rate": 1.388478930199563e-05,
      "loss": 0.0872,
      "mean_token_accuracy": 0.9710069298744202,
      "num_tokens": 36124738.0,
      "step": 385
    },
    {
      "epoch": 0.35071942446043164,
      "grad_norm": 0.33801376819610596,
      "learning_rate": 1.3865858237800677e-05,
      "loss": 0.085,
      "mean_token_accuracy": 0.9694444537162781,
      "num_tokens": 36592823.0,
      "step": 390
    },
    {
      "epoch": 0.3552158273381295,
      "grad_norm": 0.36717891693115234,
      "learning_rate": 1.3846681613854347e-05,
      "loss": 0.0852,
      "mean_token_accuracy": 0.9701388835906982,
      "num_tokens": 37060806.0,
      "step": 395
    },
    {
      "epoch": 0.3597122302158273,
      "grad_norm": 0.41675055027008057,
      "learning_rate": 1.3827260417614385e-05,
      "loss": 0.0864,
      "mean_token_accuracy": 0.9703125119209289,
      "num_tokens": 37533546.0,
      "step": 400
    },
    {
      "epoch": 0.3642086330935252,
      "grad_norm": 0.5478615760803223,
      "learning_rate": 1.3807595649132233e-05,
      "loss": 0.0856,
      "mean_token_accuracy": 0.9692708373069763,
      "num_tokens": 38003799.0,
      "step": 405
    },
    {
      "epoch": 0.368705035971223,
      "grad_norm": 0.35957762598991394,
      "learning_rate": 1.3787688321001566e-05,
      "loss": 0.0816,
      "mean_token_accuracy": 0.9704861164093017,
      "num_tokens": 38473226.0,
      "step": 410
    },
    {
      "epoch": 0.3732014388489209,
      "grad_norm": 0.4207388758659363,
      "learning_rate": 1.3767539458306107e-05,
      "loss": 0.0872,
      "mean_token_accuracy": 0.9694444417953492,
      "num_tokens": 38942060.0,
      "step": 415
    },
    {
      "epoch": 0.3776978417266187,
      "grad_norm": 0.34111288189888,
      "learning_rate": 1.3747150098566884e-05,
      "loss": 0.0887,
      "mean_token_accuracy": 0.9689236164093018,
      "num_tokens": 39412384.0,
      "step": 420
    },
    {
      "epoch": 0.38219424460431656,
      "grad_norm": 0.3971599340438843,
      "learning_rate": 1.3726521291688778e-05,
      "loss": 0.0892,
      "mean_token_accuracy": 0.9678819656372071,
      "num_tokens": 39883643.0,
      "step": 425
    },
    {
      "epoch": 0.38669064748201437,
      "grad_norm": 0.3130061626434326,
      "learning_rate": 1.3705654099906471e-05,
      "loss": 0.0809,
      "mean_token_accuracy": 0.9723958373069763,
      "num_tokens": 40353746.0,
      "step": 430
    },
    {
      "epoch": 0.39118705035971224,
      "grad_norm": 0.4201053977012634,
      "learning_rate": 1.368454959772975e-05,
      "loss": 0.084,
      "mean_token_accuracy": 0.9708333492279053,
      "num_tokens": 40825196.0,
      "step": 435
    },
    {
      "epoch": 0.39568345323741005,
      "grad_norm": 0.41213706135749817,
      "learning_rate": 1.366320887188817e-05,
      "loss": 0.0911,
      "mean_token_accuracy": 0.9704861164093017,
      "num_tokens": 41294879.0,
      "step": 440
    },
    {
      "epoch": 0.4001798561151079,
      "grad_norm": 0.38970908522605896,
      "learning_rate": 1.3641633021275108e-05,
      "loss": 0.0852,
      "mean_token_accuracy": 0.9677083492279053,
      "num_tokens": 41762397.0,
      "step": 445
    },
    {
      "epoch": 0.40467625899280574,
      "grad_norm": 0.3463442921638489,
      "learning_rate": 1.361982315689116e-05,
      "loss": 0.0804,
      "mean_token_accuracy": 0.9704861283302307,
      "num_tokens": 42229511.0,
      "step": 450
    },
    {
      "epoch": 0.4091726618705036,
      "grad_norm": 0.32913315296173096,
      "learning_rate": 1.3597780401786952e-05,
      "loss": 0.0872,
      "mean_token_accuracy": 0.9701388835906982,
      "num_tokens": 42697235.0,
      "step": 455
    },
    {
      "epoch": 0.4136690647482014,
      "grad_norm": 0.44308772683143616,
      "learning_rate": 1.3575505891005293e-05,
      "loss": 0.0769,
      "mean_token_accuracy": 0.9710069417953491,
      "num_tokens": 43162193.0,
      "step": 460
    },
    {
      "epoch": 0.4181654676258993,
      "grad_norm": 0.42118602991104126,
      "learning_rate": 1.355300077152274e-05,
      "loss": 0.0874,
      "mean_token_accuracy": 0.9685763716697693,
      "num_tokens": 43632286.0,
      "step": 465
    },
    {
      "epoch": 0.4226618705035971,
      "grad_norm": 0.3417890965938568,
      "learning_rate": 1.3530266202190536e-05,
      "loss": 0.0818,
      "mean_token_accuracy": 0.9715277791023255,
      "num_tokens": 44100455.0,
      "step": 470
    },
    {
      "epoch": 0.427158273381295,
      "grad_norm": 0.3577599823474884,
      "learning_rate": 1.3507303353674935e-05,
      "loss": 0.0855,
      "mean_token_accuracy": 0.9692708492279053,
      "num_tokens": 44569998.0,
      "step": 475
    },
    {
      "epoch": 0.4316546762589928,
      "grad_norm": 0.34517139196395874,
      "learning_rate": 1.348411340839692e-05,
      "loss": 0.0907,
      "mean_token_accuracy": 0.9680555462837219,
      "num_tokens": 45039376.0,
      "step": 480
    },
    {
      "epoch": 0.43615107913669066,
      "grad_norm": 0.46521875262260437,
      "learning_rate": 1.3460697560471315e-05,
      "loss": 0.0833,
      "mean_token_accuracy": 0.9710069417953491,
      "num_tokens": 45507862.0,
      "step": 485
    },
    {
      "epoch": 0.44064748201438847,
      "grad_norm": 0.5018951296806335,
      "learning_rate": 1.3437057015645308e-05,
      "loss": 0.083,
      "mean_token_accuracy": 0.9689236164093018,
      "num_tokens": 45977282.0,
      "step": 490
    },
    {
      "epoch": 0.44514388489208634,
      "grad_norm": 0.37957438826560974,
      "learning_rate": 1.341319299123635e-05,
      "loss": 0.0792,
      "mean_token_accuracy": 0.9710069417953491,
      "num_tokens": 46441970.0,
      "step": 495
    },
    {
      "epoch": 0.44964028776978415,
      "grad_norm": 0.4474797248840332,
      "learning_rate": 1.3389106716069478e-05,
      "loss": 0.0798,
      "mean_token_accuracy": 0.971875,
      "num_tokens": 46912092.0,
      "step": 500
    },
    {
      "epoch": 0.454136690647482,
      "grad_norm": 0.423255980014801,
      "learning_rate": 1.3364799430414044e-05,
      "loss": 0.0834,
      "mean_token_accuracy": 0.9720486164093017,
      "num_tokens": 47381568.0,
      "step": 505
    },
    {
      "epoch": 0.45863309352517984,
      "grad_norm": 0.33100345730781555,
      "learning_rate": 1.3340272385919837e-05,
      "loss": 0.0803,
      "mean_token_accuracy": 0.9711805582046509,
      "num_tokens": 47848779.0,
      "step": 510
    },
    {
      "epoch": 0.4631294964028777,
      "grad_norm": 0.4213719367980957,
      "learning_rate": 1.3315526845552651e-05,
      "loss": 0.0832,
      "mean_token_accuracy": 0.9704861164093017,
      "num_tokens": 48318817.0,
      "step": 515
    },
    {
      "epoch": 0.4676258992805755,
      "grad_norm": 0.35615554451942444,
      "learning_rate": 1.3290564083529227e-05,
      "loss": 0.0847,
      "mean_token_accuracy": 0.9685763955116272,
      "num_tokens": 48786694.0,
      "step": 520
    },
    {
      "epoch": 0.4721223021582734,
      "grad_norm": 0.35342615842819214,
      "learning_rate": 1.326538538525167e-05,
      "loss": 0.0802,
      "mean_token_accuracy": 0.9671875,
      "num_tokens": 49254590.0,
      "step": 525
    },
    {
      "epoch": 0.4766187050359712,
      "grad_norm": 0.3581298291683197,
      "learning_rate": 1.3239992047241228e-05,
      "loss": 0.0821,
      "mean_token_accuracy": 0.9699652791023254,
      "num_tokens": 49726301.0,
      "step": 530
    },
    {
      "epoch": 0.48111510791366907,
      "grad_norm": 0.3920028507709503,
      "learning_rate": 1.3214385377071554e-05,
      "loss": 0.0776,
      "mean_token_accuracy": 0.9711805701255798,
      "num_tokens": 50195878.0,
      "step": 535
    },
    {
      "epoch": 0.4856115107913669,
      "grad_norm": 0.3756016790866852,
      "learning_rate": 1.3188566693301371e-05,
      "loss": 0.0874,
      "mean_token_accuracy": 0.9678819417953491,
      "num_tokens": 50668008.0,
      "step": 540
    },
    {
      "epoch": 0.49010791366906475,
      "grad_norm": 0.3724729120731354,
      "learning_rate": 1.3162537325406566e-05,
      "loss": 0.0861,
      "mean_token_accuracy": 0.9678819417953491,
      "num_tokens": 51134853.0,
      "step": 545
    },
    {
      "epoch": 0.49460431654676257,
      "grad_norm": 0.39227744936943054,
      "learning_rate": 1.313629861371174e-05,
      "loss": 0.0871,
      "mean_token_accuracy": 0.9685764074325561,
      "num_tokens": 51605543.0,
      "step": 550
    },
    {
      "epoch": 0.49910071942446044,
      "grad_norm": 0.3155370056629181,
      "learning_rate": 1.310985190932119e-05,
      "loss": 0.0758,
      "mean_token_accuracy": 0.9730902671813965,
      "num_tokens": 52071731.0,
      "step": 555
    },
    {
      "epoch": 0.5035971223021583,
      "grad_norm": 0.4646400511264801,
      "learning_rate": 1.3083198574049334e-05,
      "loss": 0.0802,
      "mean_token_accuracy": 0.9692708373069763,
      "num_tokens": 52540868.0,
      "step": 560
    },
    {
      "epoch": 0.5080935251798561,
      "grad_norm": 0.5570467114448547,
      "learning_rate": 1.305633998035059e-05,
      "loss": 0.0799,
      "mean_token_accuracy": 0.9722222208976745,
      "num_tokens": 53006734.0,
      "step": 565
    },
    {
      "epoch": 0.512589928057554,
      "grad_norm": 0.4304635524749756,
      "learning_rate": 1.3029277511248705e-05,
      "loss": 0.0797,
      "mean_token_accuracy": 0.9722222447395324,
      "num_tokens": 53477238.0,
      "step": 570
    },
    {
      "epoch": 0.5170863309352518,
      "grad_norm": 0.39134058356285095,
      "learning_rate": 1.3002012560265538e-05,
      "loss": 0.0776,
      "mean_token_accuracy": 0.9729166746139526,
      "num_tokens": 53944632.0,
      "step": 575
    },
    {
      "epoch": 0.5215827338129496,
      "grad_norm": 0.37532541155815125,
      "learning_rate": 1.2974546531349302e-05,
      "loss": 0.0812,
      "mean_token_accuracy": 0.9713541746139527,
      "num_tokens": 54413657.0,
      "step": 580
    },
    {
      "epoch": 0.5260791366906474,
      "grad_norm": 0.43509700894355774,
      "learning_rate": 1.294688083880227e-05,
      "loss": 0.088,
      "mean_token_accuracy": 0.9684027910232544,
      "num_tokens": 54882059.0,
      "step": 585
    },
    {
      "epoch": 0.5305755395683454,
      "grad_norm": 0.3687283396720886,
      "learning_rate": 1.2919016907207958e-05,
      "loss": 0.0782,
      "mean_token_accuracy": 0.9723958492279052,
      "num_tokens": 55352248.0,
      "step": 590
    },
    {
      "epoch": 0.5350719424460432,
      "grad_norm": 0.30725207924842834,
      "learning_rate": 1.2890956171357757e-05,
      "loss": 0.0832,
      "mean_token_accuracy": 0.9715277910232544,
      "num_tokens": 55821249.0,
      "step": 595
    },
    {
      "epoch": 0.539568345323741,
      "grad_norm": 0.33535632491111755,
      "learning_rate": 1.2862700076177056e-05,
      "loss": 0.086,
      "mean_token_accuracy": 0.9697916746139527,
      "num_tokens": 56292110.0,
      "step": 600
    },
    {
      "epoch": 0.5440647482014388,
      "grad_norm": 0.320467084646225,
      "learning_rate": 1.2834250076650847e-05,
      "loss": 0.0835,
      "mean_token_accuracy": 0.9694444537162781,
      "num_tokens": 56758533.0,
      "step": 605
    },
    {
      "epoch": 0.5485611510791367,
      "grad_norm": 0.31769710779190063,
      "learning_rate": 1.2805607637748786e-05,
      "loss": 0.0873,
      "mean_token_accuracy": 0.9689236164093018,
      "num_tokens": 57229675.0,
      "step": 610
    },
    {
      "epoch": 0.5530575539568345,
      "grad_norm": 0.468817800283432,
      "learning_rate": 1.2776774234349778e-05,
      "loss": 0.0813,
      "mean_token_accuracy": 0.9723958373069763,
      "num_tokens": 57697738.0,
      "step": 615
    },
    {
      "epoch": 0.5575539568345323,
      "grad_norm": 0.36219796538352966,
      "learning_rate": 1.2747751351166015e-05,
      "loss": 0.0885,
      "mean_token_accuracy": 0.9684027910232544,
      "num_tokens": 58164798.0,
      "step": 620
    },
    {
      "epoch": 0.5620503597122302,
      "grad_norm": 0.34549543261528015,
      "learning_rate": 1.271854048266653e-05,
      "loss": 0.0866,
      "mean_token_accuracy": 0.9701388835906982,
      "num_tokens": 58636586.0,
      "step": 625
    },
    {
      "epoch": 0.5665467625899281,
      "grad_norm": 0.33623769879341125,
      "learning_rate": 1.2689143133000252e-05,
      "loss": 0.0881,
      "mean_token_accuracy": 0.9687500119209289,
      "num_tokens": 59106342.0,
      "step": 630
    },
    {
      "epoch": 0.5710431654676259,
      "grad_norm": 0.3239581286907196,
      "learning_rate": 1.2659560815918532e-05,
      "loss": 0.0809,
      "mean_token_accuracy": 0.9708333373069763,
      "num_tokens": 59573612.0,
      "step": 635
    },
    {
      "epoch": 0.5755395683453237,
      "grad_norm": 0.37120115756988525,
      "learning_rate": 1.2629795054697216e-05,
      "loss": 0.0849,
      "mean_token_accuracy": 0.9715277552604675,
      "num_tokens": 60043487.0,
      "step": 640
    },
    {
      "epoch": 0.5800359712230215,
      "grad_norm": 0.40190795063972473,
      "learning_rate": 1.2599847382058198e-05,
      "loss": 0.0833,
      "mean_token_accuracy": 0.969444465637207,
      "num_tokens": 60511760.0,
      "step": 645
    },
    {
      "epoch": 0.5845323741007195,
      "grad_norm": 0.3242078423500061,
      "learning_rate": 1.2569719340090496e-05,
      "loss": 0.0839,
      "mean_token_accuracy": 0.9699652791023254,
      "num_tokens": 60979675.0,
      "step": 650
    },
    {
      "epoch": 0.5890287769784173,
      "grad_norm": 0.32616937160491943,
      "learning_rate": 1.2539412480170847e-05,
      "loss": 0.0839,
      "mean_token_accuracy": 0.9699652910232544,
      "num_tokens": 61444272.0,
      "step": 655
    },
    {
      "epoch": 0.5935251798561151,
      "grad_norm": 0.3729708194732666,
      "learning_rate": 1.2508928362883822e-05,
      "loss": 0.0806,
      "mean_token_accuracy": 0.9697916746139527,
      "num_tokens": 61916908.0,
      "step": 660
    }
  ],
  "logging_steps": 5,
  "max_steps": 2224,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "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": 3.2552287361354957e+18,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}