{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 2113,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 1.9664676249027253,
      "epoch": 0.004733167672464797,
      "grad_norm": 2.9172720909118652,
      "learning_rate": 8.490566037735849e-07,
      "loss": 2.4508,
      "mean_token_accuracy": 0.5263931974768639,
      "num_tokens": 3879417.0,
      "step": 10
    },
    {
      "entropy": 2.0464825958013533,
      "epoch": 0.009466335344929595,
      "grad_norm": 0.7783076763153076,
      "learning_rate": 1.7924528301886793e-06,
      "loss": 2.2632,
      "mean_token_accuracy": 0.5380982637405396,
      "num_tokens": 7799425.0,
      "step": 20
    },
    {
      "entropy": 2.0156721025705338,
      "epoch": 0.014199503017394392,
      "grad_norm": 0.38891270756721497,
      "learning_rate": 2.7358490566037738e-06,
      "loss": 2.0189,
      "mean_token_accuracy": 0.568930983543396,
      "num_tokens": 11646570.0,
      "step": 30
    },
    {
      "entropy": 1.8222261607646941,
      "epoch": 0.01893267068985919,
      "grad_norm": 0.2623351812362671,
      "learning_rate": 3.679245283018868e-06,
      "loss": 1.8679,
      "mean_token_accuracy": 0.5922655925154686,
      "num_tokens": 15549095.0,
      "step": 40
    },
    {
      "entropy": 1.7781452298164369,
      "epoch": 0.023665838362323985,
      "grad_norm": 0.19355912506580353,
      "learning_rate": 4.622641509433963e-06,
      "loss": 1.7967,
      "mean_token_accuracy": 0.6030044049024582,
      "num_tokens": 19395583.0,
      "step": 50
    },
    {
      "entropy": 1.756221604347229,
      "epoch": 0.028399006034788784,
      "grad_norm": 0.17072290182113647,
      "learning_rate": 5.566037735849057e-06,
      "loss": 1.7705,
      "mean_token_accuracy": 0.6082265704870224,
      "num_tokens": 23287255.0,
      "step": 60
    },
    {
      "entropy": 1.721426972746849,
      "epoch": 0.033132173707253576,
      "grad_norm": 0.18896205723285675,
      "learning_rate": 6.5094339622641515e-06,
      "loss": 1.7402,
      "mean_token_accuracy": 0.6131706237792969,
      "num_tokens": 27155383.0,
      "step": 70
    },
    {
      "entropy": 1.720760390162468,
      "epoch": 0.03786534137971838,
      "grad_norm": 0.16523988544940948,
      "learning_rate": 7.452830188679246e-06,
      "loss": 1.7276,
      "mean_token_accuracy": 0.615737484395504,
      "num_tokens": 31071816.0,
      "step": 80
    },
    {
      "entropy": 1.696010684967041,
      "epoch": 0.042598509052183174,
      "grad_norm": 0.17002570629119873,
      "learning_rate": 8.39622641509434e-06,
      "loss": 1.7075,
      "mean_token_accuracy": 0.6187006324529648,
      "num_tokens": 34979450.0,
      "step": 90
    },
    {
      "entropy": 1.6817427426576614,
      "epoch": 0.04733167672464797,
      "grad_norm": 0.1671409159898758,
      "learning_rate": 9.339622641509435e-06,
      "loss": 1.689,
      "mean_token_accuracy": 0.6216033041477204,
      "num_tokens": 38828299.0,
      "step": 100
    },
    {
      "entropy": 1.6723841279745102,
      "epoch": 0.052064844397112765,
      "grad_norm": 0.1683770716190338,
      "learning_rate": 9.999944870161476e-06,
      "loss": 1.683,
      "mean_token_accuracy": 0.6222904548048973,
      "num_tokens": 42710858.0,
      "step": 110
    },
    {
      "entropy": 1.6607272058725357,
      "epoch": 0.05679801206957757,
      "grad_norm": 0.16289345920085907,
      "learning_rate": 9.998964817962902e-06,
      "loss": 1.6697,
      "mean_token_accuracy": 0.6255089312791824,
      "num_tokens": 46563041.0,
      "step": 120
    },
    {
      "entropy": 1.6530560374259948,
      "epoch": 0.06153117974204236,
      "grad_norm": 0.1647324562072754,
      "learning_rate": 9.996759934642046e-06,
      "loss": 1.66,
      "mean_token_accuracy": 0.6271017372608185,
      "num_tokens": 50473774.0,
      "step": 130
    },
    {
      "entropy": 1.646892574429512,
      "epoch": 0.06626434741450715,
      "grad_norm": 0.1683024764060974,
      "learning_rate": 9.993330760432702e-06,
      "loss": 1.6506,
      "mean_token_accuracy": 0.6288532495498658,
      "num_tokens": 54345703.0,
      "step": 140
    },
    {
      "entropy": 1.6421253114938736,
      "epoch": 0.07099751508697195,
      "grad_norm": 0.16962556540966034,
      "learning_rate": 9.98867813554068e-06,
      "loss": 1.649,
      "mean_token_accuracy": 0.6289903283119201,
      "num_tokens": 58224397.0,
      "step": 150
    },
    {
      "entropy": 1.6290255516767502,
      "epoch": 0.07573068275943676,
      "grad_norm": 0.16403813660144806,
      "learning_rate": 9.98280319993795e-06,
      "loss": 1.6341,
      "mean_token_accuracy": 0.6317014634609223,
      "num_tokens": 62114712.0,
      "step": 160
    },
    {
      "entropy": 1.6350149989128113,
      "epoch": 0.08046385043190155,
      "grad_norm": 0.16714808344841003,
      "learning_rate": 9.975707393083328e-06,
      "loss": 1.6379,
      "mean_token_accuracy": 0.6297940164804459,
      "num_tokens": 65991061.0,
      "step": 170
    },
    {
      "entropy": 1.6163844734430313,
      "epoch": 0.08519701810436635,
      "grad_norm": 0.17177364230155945,
      "learning_rate": 9.967392453569775e-06,
      "loss": 1.6222,
      "mean_token_accuracy": 0.6339128240942955,
      "num_tokens": 69856575.0,
      "step": 180
    },
    {
      "entropy": 1.6292197763919831,
      "epoch": 0.08993018577683115,
      "grad_norm": 0.16445177793502808,
      "learning_rate": 9.95786041869843e-06,
      "loss": 1.6316,
      "mean_token_accuracy": 0.6325667470693588,
      "num_tokens": 73718253.0,
      "step": 190
    },
    {
      "entropy": 1.6217930316925049,
      "epoch": 0.09466335344929594,
      "grad_norm": 0.15951167047023773,
      "learning_rate": 9.947113623979423e-06,
      "loss": 1.6253,
      "mean_token_accuracy": 0.6325555801391601,
      "num_tokens": 77599124.0,
      "step": 200
    },
    {
      "entropy": 1.6170487314462663,
      "epoch": 0.09939652112176074,
      "grad_norm": 0.15958699584007263,
      "learning_rate": 9.935154702559637e-06,
      "loss": 1.6246,
      "mean_token_accuracy": 0.6333600997924804,
      "num_tokens": 81539199.0,
      "step": 210
    },
    {
      "entropy": 1.6078982055187225,
      "epoch": 0.10412968879422553,
      "grad_norm": 0.16408687829971313,
      "learning_rate": 9.921986584577545e-06,
      "loss": 1.6135,
      "mean_token_accuracy": 0.6353577077388763,
      "num_tokens": 85404474.0,
      "step": 220
    },
    {
      "entropy": 1.6178746074438095,
      "epoch": 0.10886285646669033,
      "grad_norm": 0.16022811830043793,
      "learning_rate": 9.90761249644528e-06,
      "loss": 1.6173,
      "mean_token_accuracy": 0.6340486526489257,
      "num_tokens": 89310569.0,
      "step": 230
    },
    {
      "entropy": 1.6118096917867661,
      "epoch": 0.11359602413915514,
      "grad_norm": 0.16375407576560974,
      "learning_rate": 9.8920359600581e-06,
      "loss": 1.6195,
      "mean_token_accuracy": 0.6345633432269097,
      "num_tokens": 93225240.0,
      "step": 240
    },
    {
      "entropy": 1.6142515003681184,
      "epoch": 0.11832919181161992,
      "grad_norm": 0.16436107456684113,
      "learning_rate": 9.875260791931474e-06,
      "loss": 1.6154,
      "mean_token_accuracy": 0.6353823691606522,
      "num_tokens": 97106446.0,
      "step": 250
    },
    {
      "entropy": 1.6073731541633607,
      "epoch": 0.12306235948408473,
      "grad_norm": 0.16237056255340576,
      "learning_rate": 9.85729110226596e-06,
      "loss": 1.6087,
      "mean_token_accuracy": 0.6353139713406563,
      "num_tokens": 100977975.0,
      "step": 260
    },
    {
      "entropy": 1.6020812511444091,
      "epoch": 0.12779552715654952,
      "grad_norm": 0.15944795310497284,
      "learning_rate": 9.838131293940148e-06,
      "loss": 1.6071,
      "mean_token_accuracy": 0.6363679498434067,
      "num_tokens": 104876425.0,
      "step": 270
    },
    {
      "entropy": 1.6033865660429,
      "epoch": 0.1325286948290143,
      "grad_norm": 0.17549684643745422,
      "learning_rate": 9.817786061431874e-06,
      "loss": 1.6091,
      "mean_token_accuracy": 0.6354461714625359,
      "num_tokens": 108709842.0,
      "step": 280
    },
    {
      "entropy": 1.5905151784420013,
      "epoch": 0.13726186250147912,
      "grad_norm": 0.1742265522480011,
      "learning_rate": 9.79626038966799e-06,
      "loss": 1.596,
      "mean_token_accuracy": 0.6382940918207168,
      "num_tokens": 112609412.0,
      "step": 290
    },
    {
      "entropy": 1.5923225700855255,
      "epoch": 0.1419950301739439,
      "grad_norm": 0.16349110007286072,
      "learning_rate": 9.773559552802982e-06,
      "loss": 1.5926,
      "mean_token_accuracy": 0.6394994035363197,
      "num_tokens": 116498424.0,
      "step": 300
    },
    {
      "entropy": 1.5896567821502685,
      "epoch": 0.1467281978464087,
      "grad_norm": 0.1612998992204666,
      "learning_rate": 9.749689112926701e-06,
      "loss": 1.5973,
      "mean_token_accuracy": 0.6379290148615837,
      "num_tokens": 120371695.0,
      "step": 310
    },
    {
      "entropy": 1.596482390165329,
      "epoch": 0.15146136551887351,
      "grad_norm": 0.16588379442691803,
      "learning_rate": 9.724654918701568e-06,
      "loss": 1.6006,
      "mean_token_accuracy": 0.6373828530311585,
      "num_tokens": 124255939.0,
      "step": 320
    },
    {
      "entropy": 1.584341713786125,
      "epoch": 0.1561945331913383,
      "grad_norm": 0.16518791019916534,
      "learning_rate": 9.698463103929542e-06,
      "loss": 1.5887,
      "mean_token_accuracy": 0.6403212890028953,
      "num_tokens": 128177070.0,
      "step": 330
    },
    {
      "entropy": 1.5725542455911636,
      "epoch": 0.1609277008638031,
      "grad_norm": 0.16715767979621887,
      "learning_rate": 9.671120086049246e-06,
      "loss": 1.5799,
      "mean_token_accuracy": 0.6416747942566872,
      "num_tokens": 132121810.0,
      "step": 340
    },
    {
      "entropy": 1.5927457183599472,
      "epoch": 0.1656608685362679,
      "grad_norm": 0.17323796451091766,
      "learning_rate": 9.642632564563576e-06,
      "loss": 1.5939,
      "mean_token_accuracy": 0.6393081307411194,
      "num_tokens": 136012804.0,
      "step": 350
    },
    {
      "entropy": 1.5822071313858033,
      "epoch": 0.1703940362087327,
      "grad_norm": 0.16068372130393982,
      "learning_rate": 9.61300751939821e-06,
      "loss": 1.583,
      "mean_token_accuracy": 0.6409055680036545,
      "num_tokens": 139940284.0,
      "step": 360
    },
    {
      "entropy": 1.568736445903778,
      "epoch": 0.17512720388119749,
      "grad_norm": 0.16828852891921997,
      "learning_rate": 9.582252209191417e-06,
      "loss": 1.5788,
      "mean_token_accuracy": 0.6417408421635628,
      "num_tokens": 143849853.0,
      "step": 370
    },
    {
      "entropy": 1.5762336939573287,
      "epoch": 0.1798603715536623,
      "grad_norm": 0.1591760367155075,
      "learning_rate": 9.550374169515557e-06,
      "loss": 1.5863,
      "mean_token_accuracy": 0.6399773553013801,
      "num_tokens": 147708328.0,
      "step": 380
    },
    {
      "entropy": 1.5851975083351135,
      "epoch": 0.1845935392261271,
      "grad_norm": 0.16111813485622406,
      "learning_rate": 9.517381211030745e-06,
      "loss": 1.5859,
      "mean_token_accuracy": 0.6406419515609741,
      "num_tokens": 151541012.0,
      "step": 390
    },
    {
      "entropy": 1.5762753903865814,
      "epoch": 0.18932670689859188,
      "grad_norm": 0.1650865077972412,
      "learning_rate": 9.4832814175711e-06,
      "loss": 1.583,
      "mean_token_accuracy": 0.6416757836937904,
      "num_tokens": 155434817.0,
      "step": 400
    },
    {
      "entropy": 1.5776998460292817,
      "epoch": 0.19405987457105667,
      "grad_norm": 0.1527196615934372,
      "learning_rate": 9.448083144164078e-06,
      "loss": 1.5798,
      "mean_token_accuracy": 0.6412320747971535,
      "num_tokens": 159311737.0,
      "step": 410
    },
    {
      "entropy": 1.5809161812067032,
      "epoch": 0.19879304224352148,
      "grad_norm": 0.16029658913612366,
      "learning_rate": 9.411795014983338e-06,
      "loss": 1.5847,
      "mean_token_accuracy": 0.64069664478302,
      "num_tokens": 163185815.0,
      "step": 420
    },
    {
      "entropy": 1.582759228348732,
      "epoch": 0.20352620991598627,
      "grad_norm": 0.15568074584007263,
      "learning_rate": 9.374425921235684e-06,
      "loss": 1.5865,
      "mean_token_accuracy": 0.6403465285897255,
      "num_tokens": 167027904.0,
      "step": 430
    },
    {
      "entropy": 1.5735624372959136,
      "epoch": 0.20825937758845106,
      "grad_norm": 0.15200895071029663,
      "learning_rate": 9.33598501898256e-06,
      "loss": 1.5796,
      "mean_token_accuracy": 0.6411942645907402,
      "num_tokens": 170922541.0,
      "step": 440
    },
    {
      "entropy": 1.573996353149414,
      "epoch": 0.21299254526091588,
      "grad_norm": 0.15826307237148285,
      "learning_rate": 9.296481726896663e-06,
      "loss": 1.5784,
      "mean_token_accuracy": 0.6419912785291672,
      "num_tokens": 174791554.0,
      "step": 450
    },
    {
      "entropy": 1.573242038488388,
      "epoch": 0.21772571293338067,
      "grad_norm": 0.16425147652626038,
      "learning_rate": 9.25592572395421e-06,
      "loss": 1.5741,
      "mean_token_accuracy": 0.6426722049713135,
      "num_tokens": 178691330.0,
      "step": 460
    },
    {
      "entropy": 1.5621694713830947,
      "epoch": 0.22245888060584545,
      "grad_norm": 0.15976452827453613,
      "learning_rate": 9.214326947063424e-06,
      "loss": 1.569,
      "mean_token_accuracy": 0.6429914653301239,
      "num_tokens": 182540865.0,
      "step": 470
    },
    {
      "entropy": 1.5637736171483994,
      "epoch": 0.22719204827831027,
      "grad_norm": 0.15400023758411407,
      "learning_rate": 9.171695588629818e-06,
      "loss": 1.5669,
      "mean_token_accuracy": 0.6447574943304062,
      "num_tokens": 186445682.0,
      "step": 480
    },
    {
      "entropy": 1.5614915072917939,
      "epoch": 0.23192521595077506,
      "grad_norm": 0.1612849235534668,
      "learning_rate": 9.128042094058892e-06,
      "loss": 1.5661,
      "mean_token_accuracy": 0.6434317573904991,
      "num_tokens": 190298343.0,
      "step": 490
    },
    {
      "entropy": 1.57402603328228,
      "epoch": 0.23665838362323985,
      "grad_norm": 0.1595270037651062,
      "learning_rate": 9.083377159196824e-06,
      "loss": 1.5783,
      "mean_token_accuracy": 0.6416443169116974,
      "num_tokens": 194193930.0,
      "step": 500
    },
    {
      "entropy": 1.5727292507886887,
      "epoch": 0.24139155129570464,
      "grad_norm": 0.1542680561542511,
      "learning_rate": 9.037711727709812e-06,
      "loss": 1.5708,
      "mean_token_accuracy": 0.6426661461591721,
      "num_tokens": 198099362.0,
      "step": 510
    },
    {
      "entropy": 1.5556272625923158,
      "epoch": 0.24612471896816945,
      "grad_norm": 0.16021884977817535,
      "learning_rate": 8.991056988402682e-06,
      "loss": 1.5594,
      "mean_token_accuracy": 0.6437688574194909,
      "num_tokens": 201971348.0,
      "step": 520
    },
    {
      "entropy": 1.561934667825699,
      "epoch": 0.25085788664063424,
      "grad_norm": 0.160409614443779,
      "learning_rate": 8.943424372477455e-06,
      "loss": 1.5663,
      "mean_token_accuracy": 0.6441657856106758,
      "num_tokens": 205887553.0,
      "step": 530
    },
    {
      "entropy": 1.5650932669639588,
      "epoch": 0.25559105431309903,
      "grad_norm": 0.16672302782535553,
      "learning_rate": 8.894825550732491e-06,
      "loss": 1.5651,
      "mean_token_accuracy": 0.6445234566926956,
      "num_tokens": 209773471.0,
      "step": 540
    },
    {
      "entropy": 1.5703383475542068,
      "epoch": 0.2603242219855638,
      "grad_norm": 0.1647508144378662,
      "learning_rate": 8.845272430702962e-06,
      "loss": 1.5719,
      "mean_token_accuracy": 0.6425846204161644,
      "num_tokens": 213667761.0,
      "step": 550
    },
    {
      "entropy": 1.5690355956554414,
      "epoch": 0.2650573896580286,
      "grad_norm": 0.16149023175239563,
      "learning_rate": 8.79477715374329e-06,
      "loss": 1.5733,
      "mean_token_accuracy": 0.6426513850688934,
      "num_tokens": 217473103.0,
      "step": 560
    },
    {
      "entropy": 1.565832221508026,
      "epoch": 0.26979055733049345,
      "grad_norm": 0.1554195135831833,
      "learning_rate": 8.743352092052326e-06,
      "loss": 1.568,
      "mean_token_accuracy": 0.6430184543132782,
      "num_tokens": 221365501.0,
      "step": 570
    },
    {
      "entropy": 1.562074252963066,
      "epoch": 0.27452372500295824,
      "grad_norm": 0.16073721647262573,
      "learning_rate": 8.691009845641931e-06,
      "loss": 1.5701,
      "mean_token_accuracy": 0.6420939102768898,
      "num_tokens": 225235652.0,
      "step": 580
    },
    {
      "entropy": 1.5577911257743835,
      "epoch": 0.27925689267542303,
      "grad_norm": 0.16031447052955627,
      "learning_rate": 8.637763239249778e-06,
      "loss": 1.5627,
      "mean_token_accuracy": 0.6444457039237023,
      "num_tokens": 229134110.0,
      "step": 590
    },
    {
      "entropy": 1.556574183702469,
      "epoch": 0.2839900603478878,
      "grad_norm": 0.1577136367559433,
      "learning_rate": 8.583625319197063e-06,
      "loss": 1.5565,
      "mean_token_accuracy": 0.6455687090754509,
      "num_tokens": 233039906.0,
      "step": 600
    },
    {
      "entropy": 1.5509059250354766,
      "epoch": 0.2887232280203526,
      "grad_norm": 0.15498368442058563,
      "learning_rate": 8.528609350191938e-06,
      "loss": 1.558,
      "mean_token_accuracy": 0.6440837427973747,
      "num_tokens": 236877041.0,
      "step": 610
    },
    {
      "entropy": 1.5675470024347304,
      "epoch": 0.2934563956928174,
      "grad_norm": 0.16251349449157715,
      "learning_rate": 8.472728812079436e-06,
      "loss": 1.5698,
      "mean_token_accuracy": 0.6437425896525383,
      "num_tokens": 240760840.0,
      "step": 620
    },
    {
      "entropy": 1.5483269661664962,
      "epoch": 0.29818956336528224,
      "grad_norm": 0.16416430473327637,
      "learning_rate": 8.415997396538671e-06,
      "loss": 1.5516,
      "mean_token_accuracy": 0.6461272791028023,
      "num_tokens": 244646788.0,
      "step": 630
    },
    {
      "entropy": 1.5591329038143158,
      "epoch": 0.30292273103774703,
      "grad_norm": 0.15344582498073578,
      "learning_rate": 8.358429003728158e-06,
      "loss": 1.5631,
      "mean_token_accuracy": 0.6448241740465164,
      "num_tokens": 248505578.0,
      "step": 640
    },
    {
      "entropy": 1.5571163922548295,
      "epoch": 0.3076558987102118,
      "grad_norm": 0.16142602264881134,
      "learning_rate": 8.30003773888003e-06,
      "loss": 1.561,
      "mean_token_accuracy": 0.6446821197867394,
      "num_tokens": 252442387.0,
      "step": 650
    },
    {
      "entropy": 1.543628853559494,
      "epoch": 0.3123890663826766,
      "grad_norm": 0.16222691535949707,
      "learning_rate": 8.240837908844011e-06,
      "loss": 1.5465,
      "mean_token_accuracy": 0.6476170778274536,
      "num_tokens": 256347145.0,
      "step": 660
    },
    {
      "entropy": 1.5412938565015792,
      "epoch": 0.3171222340551414,
      "grad_norm": 0.1638648957014084,
      "learning_rate": 8.180844018582014e-06,
      "loss": 1.5483,
      "mean_token_accuracy": 0.6471631437540054,
      "num_tokens": 260255832.0,
      "step": 670
    },
    {
      "entropy": 1.5546195566654206,
      "epoch": 0.3218554017276062,
      "grad_norm": 0.1579921394586563,
      "learning_rate": 8.12007076761416e-06,
      "loss": 1.5539,
      "mean_token_accuracy": 0.6463906392455101,
      "num_tokens": 264145360.0,
      "step": 680
    },
    {
      "entropy": 1.538521057367325,
      "epoch": 0.32658856940007097,
      "grad_norm": 0.15421803295612335,
      "learning_rate": 8.05853304641716e-06,
      "loss": 1.5499,
      "mean_token_accuracy": 0.6467000633478165,
      "num_tokens": 268031581.0,
      "step": 690
    },
    {
      "entropy": 1.5507268905639648,
      "epoch": 0.3313217370725358,
      "grad_norm": 0.16604824364185333,
      "learning_rate": 7.996245932775883e-06,
      "loss": 1.5511,
      "mean_token_accuracy": 0.6466175228357315,
      "num_tokens": 271952881.0,
      "step": 700
    },
    {
      "entropy": 1.5473167181015015,
      "epoch": 0.3360549047450006,
      "grad_norm": 0.16262564063072205,
      "learning_rate": 7.933224688089059e-06,
      "loss": 1.5454,
      "mean_token_accuracy": 0.6477824598550797,
      "num_tokens": 275858331.0,
      "step": 710
    },
    {
      "entropy": 1.5481310993433,
      "epoch": 0.3407880724174654,
      "grad_norm": 0.1532575786113739,
      "learning_rate": 7.869484753629963e-06,
      "loss": 1.5541,
      "mean_token_accuracy": 0.6460428267717362,
      "num_tokens": 279739647.0,
      "step": 720
    },
    {
      "entropy": 1.5434869408607483,
      "epoch": 0.3455212400899302,
      "grad_norm": 0.1606515794992447,
      "learning_rate": 7.805041746763052e-06,
      "loss": 1.5495,
      "mean_token_accuracy": 0.6475459963083268,
      "num_tokens": 283656822.0,
      "step": 730
    },
    {
      "entropy": 1.550658145546913,
      "epoch": 0.35025440776239497,
      "grad_norm": 0.16676083207130432,
      "learning_rate": 7.739911457117437e-06,
      "loss": 1.5461,
      "mean_token_accuracy": 0.6471642255783081,
      "num_tokens": 287491081.0,
      "step": 740
    },
    {
      "entropy": 1.5445130974054337,
      "epoch": 0.35498757543485976,
      "grad_norm": 0.15253710746765137,
      "learning_rate": 7.674109842718163e-06,
      "loss": 1.5545,
      "mean_token_accuracy": 0.646132855117321,
      "num_tokens": 291367807.0,
      "step": 750
    },
    {
      "entropy": 1.5350563913583755,
      "epoch": 0.3597207431073246,
      "grad_norm": 0.16385188698768616,
      "learning_rate": 7.607653026076218e-06,
      "loss": 1.5395,
      "mean_token_accuracy": 0.6482170835137367,
      "num_tokens": 295210867.0,
      "step": 760
    },
    {
      "entropy": 1.5343231499195098,
      "epoch": 0.3644539107797894,
      "grad_norm": 0.1605066955089569,
      "learning_rate": 7.540557290238251e-06,
      "loss": 1.5386,
      "mean_token_accuracy": 0.648663455247879,
      "num_tokens": 299093246.0,
      "step": 770
    },
    {
      "entropy": 1.5489379942417145,
      "epoch": 0.3691870784522542,
      "grad_norm": 0.16125887632369995,
      "learning_rate": 7.4728390747969385e-06,
      "loss": 1.554,
      "mean_token_accuracy": 0.646386744081974,
      "num_tokens": 302979174.0,
      "step": 780
    },
    {
      "entropy": 1.5436400204896927,
      "epoch": 0.37392024612471897,
      "grad_norm": 0.16131755709648132,
      "learning_rate": 7.404514971863015e-06,
      "loss": 1.5547,
      "mean_token_accuracy": 0.6458986714482308,
      "num_tokens": 306861334.0,
      "step": 790
    },
    {
      "entropy": 1.5432285010814666,
      "epoch": 0.37865341379718376,
      "grad_norm": 0.15674395859241486,
      "learning_rate": 7.3356017219999236e-06,
      "loss": 1.543,
      "mean_token_accuracy": 0.6479764103889465,
      "num_tokens": 310763957.0,
      "step": 800
    },
    {
      "entropy": 1.5451003342866898,
      "epoch": 0.38338658146964855,
      "grad_norm": 0.1598990261554718,
      "learning_rate": 7.2661162101220895e-06,
      "loss": 1.5494,
      "mean_token_accuracy": 0.6463793829083443,
      "num_tokens": 314645121.0,
      "step": 810
    },
    {
      "entropy": 1.5456315219402312,
      "epoch": 0.38811974914211333,
      "grad_norm": 0.15369856357574463,
      "learning_rate": 7.196075461357831e-06,
      "loss": 1.5488,
      "mean_token_accuracy": 0.6465605661273003,
      "num_tokens": 318500691.0,
      "step": 820
    },
    {
      "entropy": 1.5482069969177246,
      "epoch": 0.3928529168145782,
      "grad_norm": 0.1560712456703186,
      "learning_rate": 7.125496636877922e-06,
      "loss": 1.5473,
      "mean_token_accuracy": 0.6471294403076172,
      "num_tokens": 322430307.0,
      "step": 830
    },
    {
      "entropy": 1.527587327361107,
      "epoch": 0.39758608448704297,
      "grad_norm": 0.154985710978508,
      "learning_rate": 7.054397029690802e-06,
      "loss": 1.5341,
      "mean_token_accuracy": 0.6498543590307235,
      "num_tokens": 326334568.0,
      "step": 840
    },
    {
      "entropy": 1.5514892965555191,
      "epoch": 0.40231925215950776,
      "grad_norm": 0.15923038125038147,
      "learning_rate": 6.982794060405502e-06,
      "loss": 1.5525,
      "mean_token_accuracy": 0.6457523703575134,
      "num_tokens": 330201587.0,
      "step": 850
    },
    {
      "entropy": 1.533634251356125,
      "epoch": 0.40705241983197255,
      "grad_norm": 0.16257378458976746,
      "learning_rate": 6.910705272963307e-06,
      "loss": 1.5412,
      "mean_token_accuracy": 0.6487752512097359,
      "num_tokens": 334119000.0,
      "step": 860
    },
    {
      "entropy": 1.5422323346138,
      "epoch": 0.41178558750443733,
      "grad_norm": 0.15663348138332367,
      "learning_rate": 6.8381483303391795e-06,
      "loss": 1.5437,
      "mean_token_accuracy": 0.6477985471487046,
      "num_tokens": 337973430.0,
      "step": 870
    },
    {
      "entropy": 1.5455356657505035,
      "epoch": 0.4165187551769021,
      "grad_norm": 0.15920013189315796,
      "learning_rate": 6.765141010214038e-06,
      "loss": 1.5461,
      "mean_token_accuracy": 0.6477194577455521,
      "num_tokens": 341848262.0,
      "step": 880
    },
    {
      "entropy": 1.5300615668296813,
      "epoch": 0.42125192284936697,
      "grad_norm": 0.15260560810565948,
      "learning_rate": 6.691701200618925e-06,
      "loss": 1.5355,
      "mean_token_accuracy": 0.6492922276258468,
      "num_tokens": 345741246.0,
      "step": 890
    },
    {
      "entropy": 1.5360501438379288,
      "epoch": 0.42598509052183176,
      "grad_norm": 0.1613568812608719,
      "learning_rate": 6.617846895552137e-06,
      "loss": 1.5409,
      "mean_token_accuracy": 0.6486483410000801,
      "num_tokens": 349581218.0,
      "step": 900
    },
    {
      "entropy": 1.5319293648004533,
      "epoch": 0.43071825819429654,
      "grad_norm": 0.16074907779693604,
      "learning_rate": 6.543596190570381e-06,
      "loss": 1.5406,
      "mean_token_accuracy": 0.6477162018418312,
      "num_tokens": 353439070.0,
      "step": 910
    },
    {
      "entropy": 1.5505404442548751,
      "epoch": 0.43545142586676133,
      "grad_norm": 0.1633734107017517,
      "learning_rate": 6.4689672783550715e-06,
      "loss": 1.551,
      "mean_token_accuracy": 0.6463420912623405,
      "num_tokens": 357338828.0,
      "step": 920
    },
    {
      "entropy": 1.5329742044210435,
      "epoch": 0.4401845935392261,
      "grad_norm": 0.1549346148967743,
      "learning_rate": 6.393978444254798e-06,
      "loss": 1.5368,
      "mean_token_accuracy": 0.6490220680832863,
      "num_tokens": 361226544.0,
      "step": 930
    },
    {
      "entropy": 1.5339823961257935,
      "epoch": 0.4449177612116909,
      "grad_norm": 0.15460601449012756,
      "learning_rate": 6.3186480618051215e-06,
      "loss": 1.5398,
      "mean_token_accuracy": 0.6480798035860061,
      "num_tokens": 365067789.0,
      "step": 940
    },
    {
      "entropy": 1.5356625199317933,
      "epoch": 0.4496509288841557,
      "grad_norm": 0.1592043936252594,
      "learning_rate": 6.242994588226731e-06,
      "loss": 1.5378,
      "mean_token_accuracy": 0.6489008024334908,
      "num_tokens": 368967569.0,
      "step": 950
    },
    {
      "entropy": 1.532603034377098,
      "epoch": 0.45438409655662054,
      "grad_norm": 0.15242810547351837,
      "learning_rate": 6.1670365599031215e-06,
      "loss": 1.5352,
      "mean_token_accuracy": 0.649759191274643,
      "num_tokens": 372857431.0,
      "step": 960
    },
    {
      "entropy": 1.5232465624809266,
      "epoch": 0.45911726422908533,
      "grad_norm": 0.1570185422897339,
      "learning_rate": 6.090792587838867e-06,
      "loss": 1.5277,
      "mean_token_accuracy": 0.6510468333959579,
      "num_tokens": 376736423.0,
      "step": 970
    },
    {
      "entropy": 1.5315758377313613,
      "epoch": 0.4638504319015501,
      "grad_norm": 0.15817788243293762,
      "learning_rate": 6.014281353099601e-06,
      "loss": 1.5306,
      "mean_token_accuracy": 0.6500695019960403,
      "num_tokens": 380611456.0,
      "step": 980
    },
    {
      "entropy": 1.5315613538026809,
      "epoch": 0.4685835995740149,
      "grad_norm": 0.16947001218795776,
      "learning_rate": 5.937521602234842e-06,
      "loss": 1.5368,
      "mean_token_accuracy": 0.6481669098138809,
      "num_tokens": 384494701.0,
      "step": 990
    },
    {
      "entropy": 1.541380015015602,
      "epoch": 0.4733167672464797,
      "grad_norm": 0.1462167650461197,
      "learning_rate": 5.8605321426847795e-06,
      "loss": 1.5436,
      "mean_token_accuracy": 0.6473192036151886,
      "num_tokens": 388320020.0,
      "step": 1000
    },
    {
      "entropy": 1.5256280928850174,
      "epoch": 0.4780499349189445,
      "grad_norm": 0.16574805974960327,
      "learning_rate": 5.783331838172116e-06,
      "loss": 1.5349,
      "mean_token_accuracy": 0.6488758757710457,
      "num_tokens": 392189228.0,
      "step": 1010
    },
    {
      "entropy": 1.527175721526146,
      "epoch": 0.4827831025914093,
      "grad_norm": 0.15633495151996613,
      "learning_rate": 5.705939604080147e-06,
      "loss": 1.5316,
      "mean_token_accuracy": 0.6492022335529327,
      "num_tokens": 396051728.0,
      "step": 1020
    },
    {
      "entropy": 1.5351572215557099,
      "epoch": 0.4875162702638741,
      "grad_norm": 0.17186181247234344,
      "learning_rate": 5.628374402818173e-06,
      "loss": 1.54,
      "mean_token_accuracy": 0.6493882149457931,
      "num_tokens": 399969073.0,
      "step": 1030
    },
    {
      "entropy": 1.5383628755807877,
      "epoch": 0.4922494379363389,
      "grad_norm": 0.1601661890745163,
      "learning_rate": 5.550655239175377e-06,
      "loss": 1.5395,
      "mean_token_accuracy": 0.6477264970541,
      "num_tokens": 403812928.0,
      "step": 1040
    },
    {
      "entropy": 1.5366960883140564,
      "epoch": 0.4969826056088037,
      "grad_norm": 0.16334351897239685,
      "learning_rate": 5.472801155664339e-06,
      "loss": 1.539,
      "mean_token_accuracy": 0.6490108296275139,
      "num_tokens": 407675957.0,
      "step": 1050
    },
    {
      "entropy": 1.5218679517507554,
      "epoch": 0.5017157732812685,
      "grad_norm": 0.15565453469753265,
      "learning_rate": 5.394831227855291e-06,
      "loss": 1.5295,
      "mean_token_accuracy": 0.64990915954113,
      "num_tokens": 411549219.0,
      "step": 1060
    },
    {
      "entropy": 1.5312676399946212,
      "epoch": 0.5064489409537333,
      "grad_norm": 0.16980209946632385,
      "learning_rate": 5.3167645597022855e-06,
      "loss": 1.5335,
      "mean_token_accuracy": 0.6499560788273812,
      "num_tokens": 415463466.0,
      "step": 1070
    },
    {
      "entropy": 1.5292291462421417,
      "epoch": 0.5111821086261981,
      "grad_norm": 0.1765119880437851,
      "learning_rate": 5.238620278862397e-06,
      "loss": 1.5346,
      "mean_token_accuracy": 0.649413350224495,
      "num_tokens": 419305289.0,
      "step": 1080
    },
    {
      "entropy": 1.5442948073148728,
      "epoch": 0.5159152762986629,
      "grad_norm": 0.1539527326822281,
      "learning_rate": 5.160417532009122e-06,
      "loss": 1.5398,
      "mean_token_accuracy": 0.6489475637674331,
      "num_tokens": 423200780.0,
      "step": 1090
    },
    {
      "entropy": 1.5311467558145524,
      "epoch": 0.5206484439711276,
      "grad_norm": 0.15670068562030792,
      "learning_rate": 5.082175480141126e-06,
      "loss": 1.5399,
      "mean_token_accuracy": 0.6483099237084389,
      "num_tokens": 427105709.0,
      "step": 1100
    },
    {
      "entropy": 1.5239627480506897,
      "epoch": 0.5253816116435924,
      "grad_norm": 0.15966466069221497,
      "learning_rate": 5.003913293887467e-06,
      "loss": 1.5261,
      "mean_token_accuracy": 0.6510903730988502,
      "num_tokens": 430984481.0,
      "step": 1110
    },
    {
      "entropy": 1.521809309720993,
      "epoch": 0.5301147793160572,
      "grad_norm": 0.15411856770515442,
      "learning_rate": 4.925650148810465e-06,
      "loss": 1.5239,
      "mean_token_accuracy": 0.6515027031302452,
      "num_tokens": 434868955.0,
      "step": 1120
    },
    {
      "entropy": 1.5308598577976227,
      "epoch": 0.5348479469885221,
      "grad_norm": 0.16343414783477783,
      "learning_rate": 4.847405220707378e-06,
      "loss": 1.537,
      "mean_token_accuracy": 0.6491450101137162,
      "num_tokens": 438789538.0,
      "step": 1130
    },
    {
      "entropy": 1.522998222708702,
      "epoch": 0.5395811146609869,
      "grad_norm": 0.1585935652256012,
      "learning_rate": 4.7691976809119835e-06,
      "loss": 1.5227,
      "mean_token_accuracy": 0.6509381130337715,
      "num_tokens": 442659205.0,
      "step": 1140
    },
    {
      "entropy": 1.5199230432510376,
      "epoch": 0.5443142823334517,
      "grad_norm": 0.15242497622966766,
      "learning_rate": 4.691046691597303e-06,
      "loss": 1.5277,
      "mean_token_accuracy": 0.6500678122043609,
      "num_tokens": 446546475.0,
      "step": 1150
    },
    {
      "entropy": 1.5328166007995605,
      "epoch": 0.5490474500059165,
      "grad_norm": 0.1544242948293686,
      "learning_rate": 4.612971401080521e-06,
      "loss": 1.5344,
      "mean_token_accuracy": 0.6496497511863708,
      "num_tokens": 450440495.0,
      "step": 1160
    },
    {
      "entropy": 1.5248778969049455,
      "epoch": 0.5537806176783813,
      "grad_norm": 0.15791839361190796,
      "learning_rate": 4.5349909391313384e-06,
      "loss": 1.529,
      "mean_token_accuracy": 0.6500067859888077,
      "num_tokens": 454368931.0,
      "step": 1170
    },
    {
      "entropy": 1.5288578271865845,
      "epoch": 0.5585137853508461,
      "grad_norm": 0.1565057933330536,
      "learning_rate": 4.457124412284849e-06,
      "loss": 1.5328,
      "mean_token_accuracy": 0.6499333620071411,
      "num_tokens": 458289128.0,
      "step": 1180
    },
    {
      "entropy": 1.5237576812505722,
      "epoch": 0.5632469530233108,
      "grad_norm": 0.16522431373596191,
      "learning_rate": 4.379390899160116e-06,
      "loss": 1.5242,
      "mean_token_accuracy": 0.6513824790716172,
      "num_tokens": 462166520.0,
      "step": 1190
    },
    {
      "entropy": 1.5189936101436614,
      "epoch": 0.5679801206957756,
      "grad_norm": 0.1568940430879593,
      "learning_rate": 4.301809445785582e-06,
      "loss": 1.5221,
      "mean_token_accuracy": 0.6510216951370239,
      "num_tokens": 466060024.0,
      "step": 1200
    },
    {
      "entropy": 1.52001011967659,
      "epoch": 0.5727132883682404,
      "grad_norm": 0.16081492602825165,
      "learning_rate": 4.224399060932477e-06,
      "loss": 1.5253,
      "mean_token_accuracy": 0.6509154736995697,
      "num_tokens": 469988704.0,
      "step": 1210
    },
    {
      "entropy": 1.518327683210373,
      "epoch": 0.5774464560407052,
      "grad_norm": 0.1593891978263855,
      "learning_rate": 4.147178711457343e-06,
      "loss": 1.5268,
      "mean_token_accuracy": 0.6499305665493011,
      "num_tokens": 473843546.0,
      "step": 1220
    },
    {
      "entropy": 1.5157315611839295,
      "epoch": 0.58217962371317,
      "grad_norm": 0.1540728062391281,
      "learning_rate": 4.070167317654829e-06,
      "loss": 1.5215,
      "mean_token_accuracy": 0.6516207203269004,
      "num_tokens": 477705517.0,
      "step": 1230
    },
    {
      "entropy": 1.5173570811748505,
      "epoch": 0.5869127913856348,
      "grad_norm": 0.1539110541343689,
      "learning_rate": 3.9933837486219065e-06,
      "loss": 1.5212,
      "mean_token_accuracy": 0.6517001137137413,
      "num_tokens": 481587746.0,
      "step": 1240
    },
    {
      "entropy": 1.519597727060318,
      "epoch": 0.5916459590580996,
      "grad_norm": 0.1510864943265915,
      "learning_rate": 3.916846817634618e-06,
      "loss": 1.5254,
      "mean_token_accuracy": 0.6502514213323594,
      "num_tokens": 485420622.0,
      "step": 1250
    },
    {
      "entropy": 1.5250845789909362,
      "epoch": 0.5963791267305645,
      "grad_norm": 0.15542279183864594,
      "learning_rate": 3.840575277538495e-06,
      "loss": 1.5268,
      "mean_token_accuracy": 0.6506903722882271,
      "num_tokens": 489322234.0,
      "step": 1260
    },
    {
      "entropy": 1.530037170648575,
      "epoch": 0.6011122944030293,
      "grad_norm": 0.16025404632091522,
      "learning_rate": 3.764587816153813e-06,
      "loss": 1.5328,
      "mean_token_accuracy": 0.6497552871704102,
      "num_tokens": 493213867.0,
      "step": 1270
    },
    {
      "entropy": 1.5264522224664687,
      "epoch": 0.6058454620754941,
      "grad_norm": 0.1563606560230255,
      "learning_rate": 3.688903051696737e-06,
      "loss": 1.5271,
      "mean_token_accuracy": 0.6511508211493492,
      "num_tokens": 497079317.0,
      "step": 1280
    },
    {
      "entropy": 1.517073431611061,
      "epoch": 0.6105786297479588,
      "grad_norm": 0.15244126319885254,
      "learning_rate": 3.61353952821756e-06,
      "loss": 1.519,
      "mean_token_accuracy": 0.6525383740663528,
      "num_tokens": 501003584.0,
      "step": 1290
    },
    {
      "entropy": 1.520869755744934,
      "epoch": 0.6153117974204236,
      "grad_norm": 0.15403762459754944,
      "learning_rate": 3.5385157110570917e-06,
      "loss": 1.5296,
      "mean_token_accuracy": 0.6494565770030022,
      "num_tokens": 504882648.0,
      "step": 1300
    },
    {
      "entropy": 1.5278783291578293,
      "epoch": 0.6200449650928884,
      "grad_norm": 0.1558082103729248,
      "learning_rate": 3.463849982322326e-06,
      "loss": 1.532,
      "mean_token_accuracy": 0.6506448686122894,
      "num_tokens": 508796374.0,
      "step": 1310
    },
    {
      "entropy": 1.5135444432497025,
      "epoch": 0.6247781327653532,
      "grad_norm": 0.16050703823566437,
      "learning_rate": 3.389560636382531e-06,
      "loss": 1.5184,
      "mean_token_accuracy": 0.6528948619961739,
      "num_tokens": 512676763.0,
      "step": 1320
    },
    {
      "entropy": 1.5228564977645873,
      "epoch": 0.629511300437818,
      "grad_norm": 0.1465277224779129,
      "learning_rate": 3.315665875386807e-06,
      "loss": 1.5259,
      "mean_token_accuracy": 0.6503187119960785,
      "num_tokens": 516588257.0,
      "step": 1330
    },
    {
      "entropy": 1.5219411462545396,
      "epoch": 0.6342444681102828,
      "grad_norm": 0.15017631649971008,
      "learning_rate": 3.2421838048042516e-06,
      "loss": 1.5266,
      "mean_token_accuracy": 0.6507244795560837,
      "num_tokens": 520496855.0,
      "step": 1340
    },
    {
      "entropy": 1.5131709426641464,
      "epoch": 0.6389776357827476,
      "grad_norm": 0.1609673649072647,
      "learning_rate": 3.169132428987819e-06,
      "loss": 1.5171,
      "mean_token_accuracy": 0.652211906015873,
      "num_tokens": 524368593.0,
      "step": 1350
    },
    {
      "entropy": 1.5158830434083939,
      "epoch": 0.6437108034552124,
      "grad_norm": 0.1541507989168167,
      "learning_rate": 3.0965296467629413e-06,
      "loss": 1.5207,
      "mean_token_accuracy": 0.6517833650112153,
      "num_tokens": 528292613.0,
      "step": 1360
    },
    {
      "entropy": 1.5173752516508103,
      "epoch": 0.6484439711276772,
      "grad_norm": 0.14874839782714844,
      "learning_rate": 3.024393247042018e-06,
      "loss": 1.5186,
      "mean_token_accuracy": 0.6526758641004562,
      "num_tokens": 532197090.0,
      "step": 1370
    },
    {
      "entropy": 1.5261268466711044,
      "epoch": 0.6531771388001419,
      "grad_norm": 0.1556578129529953,
      "learning_rate": 2.9527409044658286e-06,
      "loss": 1.529,
      "mean_token_accuracy": 0.6497259110212326,
      "num_tokens": 536037282.0,
      "step": 1380
    },
    {
      "entropy": 1.5166011482477189,
      "epoch": 0.6579103064726068,
      "grad_norm": 0.15171046555042267,
      "learning_rate": 2.881590175072948e-06,
      "loss": 1.5189,
      "mean_token_accuracy": 0.6521017536520958,
      "num_tokens": 539892183.0,
      "step": 1390
    },
    {
      "entropy": 1.5153537422418595,
      "epoch": 0.6626434741450716,
      "grad_norm": 0.1611650586128235,
      "learning_rate": 2.8109584919982145e-06,
      "loss": 1.5221,
      "mean_token_accuracy": 0.6510555073618889,
      "num_tokens": 543786829.0,
      "step": 1400
    },
    {
      "entropy": 1.5176646322011949,
      "epoch": 0.6673766418175364,
      "grad_norm": 0.16072982549667358,
      "learning_rate": 2.7408631612013228e-06,
      "loss": 1.5198,
      "mean_token_accuracy": 0.6510667949914932,
      "num_tokens": 547697306.0,
      "step": 1410
    },
    {
      "entropy": 1.516287124156952,
      "epoch": 0.6721098094900012,
      "grad_norm": 0.15958984196186066,
      "learning_rate": 2.671321357226566e-06,
      "loss": 1.5191,
      "mean_token_accuracy": 0.6518496558070183,
      "num_tokens": 551553208.0,
      "step": 1420
    },
    {
      "entropy": 1.5146547496318816,
      "epoch": 0.676842977162466,
      "grad_norm": 0.156447172164917,
      "learning_rate": 2.602350118994782e-06,
      "loss": 1.5178,
      "mean_token_accuracy": 0.6530700176954269,
      "num_tokens": 555450956.0,
      "step": 1430
    },
    {
      "entropy": 1.5136778473854064,
      "epoch": 0.6815761448349308,
      "grad_norm": 0.15450631082057953,
      "learning_rate": 2.5339663456285302e-06,
      "loss": 1.5173,
      "mean_token_accuracy": 0.6531704932451248,
      "num_tokens": 559347478.0,
      "step": 1440
    },
    {
      "entropy": 1.5194635063409805,
      "epoch": 0.6863093125073956,
      "grad_norm": 0.1552649736404419,
      "learning_rate": 2.466186792311522e-06,
      "loss": 1.5282,
      "mean_token_accuracy": 0.6514062762260437,
      "num_tokens": 563180332.0,
      "step": 1450
    },
    {
      "entropy": 1.5103699028491975,
      "epoch": 0.6910424801798604,
      "grad_norm": 0.15950508415699005,
      "learning_rate": 2.399028066183306e-06,
      "loss": 1.515,
      "mean_token_accuracy": 0.6529736325144768,
      "num_tokens": 567111777.0,
      "step": 1460
    },
    {
      "entropy": 1.515543583035469,
      "epoch": 0.6957756478523252,
      "grad_norm": 0.1573014110326767,
      "learning_rate": 2.332506622270256e-06,
      "loss": 1.5143,
      "mean_token_accuracy": 0.6537506148219109,
      "num_tokens": 570991505.0,
      "step": 1470
    },
    {
      "entropy": 1.5154441595077515,
      "epoch": 0.7005088155247899,
      "grad_norm": 0.1603464037179947,
      "learning_rate": 2.2666387594537896e-06,
      "loss": 1.5204,
      "mean_token_accuracy": 0.652016893029213,
      "num_tokens": 574883883.0,
      "step": 1480
    },
    {
      "entropy": 1.5132777452468873,
      "epoch": 0.7052419831972547,
      "grad_norm": 0.15662123262882233,
      "learning_rate": 2.201440616476885e-06,
      "loss": 1.5177,
      "mean_token_accuracy": 0.6527584567666054,
      "num_tokens": 578724891.0,
      "step": 1490
    },
    {
      "entropy": 1.5139577269554139,
      "epoch": 0.7099751508697195,
      "grad_norm": 0.15398600697517395,
      "learning_rate": 2.136928167989803e-06,
      "loss": 1.5188,
      "mean_token_accuracy": 0.6523755803704262,
      "num_tokens": 582630649.0,
      "step": 1500
    },
    {
      "entropy": 1.5097831279039382,
      "epoch": 0.7147083185421843,
      "grad_norm": 0.1604987233877182,
      "learning_rate": 2.073117220636027e-06,
      "loss": 1.517,
      "mean_token_accuracy": 0.653074149787426,
      "num_tokens": 586509280.0,
      "step": 1510
    },
    {
      "entropy": 1.5180349677801133,
      "epoch": 0.7194414862146492,
      "grad_norm": 0.15329141914844513,
      "learning_rate": 2.0100234091793778e-06,
      "loss": 1.5201,
      "mean_token_accuracy": 0.6520388692617416,
      "num_tokens": 590387510.0,
      "step": 1520
    },
    {
      "entropy": 1.5136069297790526,
      "epoch": 0.724174653887114,
      "grad_norm": 0.15793868899345398,
      "learning_rate": 1.9476621926732227e-06,
      "loss": 1.5221,
      "mean_token_accuracy": 0.6520983532071114,
      "num_tokens": 594314494.0,
      "step": 1530
    },
    {
      "entropy": 1.5157446891069413,
      "epoch": 0.7289078215595788,
      "grad_norm": 0.15694737434387207,
      "learning_rate": 1.8860488506727409e-06,
      "loss": 1.5175,
      "mean_token_accuracy": 0.6533299177885056,
      "num_tokens": 598258201.0,
      "step": 1540
    },
    {
      "entropy": 1.522775200009346,
      "epoch": 0.7336409892320436,
      "grad_norm": 0.15210026502609253,
      "learning_rate": 1.8251984794911887e-06,
      "loss": 1.526,
      "mean_token_accuracy": 0.6508068576455116,
      "num_tokens": 602110882.0,
      "step": 1550
    },
    {
      "entropy": 1.5086775600910187,
      "epoch": 0.7383741569045084,
      "grad_norm": 0.15608255565166473,
      "learning_rate": 1.765125988501034e-06,
      "loss": 1.513,
      "mean_token_accuracy": 0.6531857639551163,
      "num_tokens": 606013390.0,
      "step": 1560
    },
    {
      "entropy": 1.5057473361492157,
      "epoch": 0.7431073245769731,
      "grad_norm": 0.1536129117012024,
      "learning_rate": 1.7058460964809086e-06,
      "loss": 1.5105,
      "mean_token_accuracy": 0.6533811703324318,
      "num_tokens": 609866962.0,
      "step": 1570
    },
    {
      "entropy": 1.5125353664159775,
      "epoch": 0.7478404922494379,
      "grad_norm": 0.15434998273849487,
      "learning_rate": 1.6473733280092658e-06,
      "loss": 1.5163,
      "mean_token_accuracy": 0.6529958829283714,
      "num_tokens": 613748468.0,
      "step": 1580
    },
    {
      "entropy": 1.5094835668802262,
      "epoch": 0.7525736599219027,
      "grad_norm": 0.16267992556095123,
      "learning_rate": 1.589722009905606e-06,
      "loss": 1.512,
      "mean_token_accuracy": 0.6539744392037392,
      "num_tokens": 617586252.0,
      "step": 1590
    },
    {
      "entropy": 1.5102200895547866,
      "epoch": 0.7573068275943675,
      "grad_norm": 0.15085026621818542,
      "learning_rate": 1.5329062677201594e-06,
      "loss": 1.5152,
      "mean_token_accuracy": 0.652720046043396,
      "num_tokens": 621473159.0,
      "step": 1600
    },
    {
      "entropy": 1.5119531810283662,
      "epoch": 0.7620399952668323,
      "grad_norm": 0.15508662164211273,
      "learning_rate": 1.4769400222728974e-06,
      "loss": 1.5176,
      "mean_token_accuracy": 0.6533014222979545,
      "num_tokens": 625366303.0,
      "step": 1610
    },
    {
      "entropy": 1.5167303442955018,
      "epoch": 0.7667731629392971,
      "grad_norm": 0.14820463955402374,
      "learning_rate": 1.4218369862426896e-06,
      "loss": 1.5186,
      "mean_token_accuracy": 0.6529193088412285,
      "num_tokens": 629214314.0,
      "step": 1620
    },
    {
      "entropy": 1.510753783583641,
      "epoch": 0.7715063306117619,
      "grad_norm": 0.1586351990699768,
      "learning_rate": 1.3676106608074602e-06,
      "loss": 1.5144,
      "mean_token_accuracy": 0.653745847940445,
      "num_tokens": 633129219.0,
      "step": 1630
    },
    {
      "entropy": 1.5246556103229523,
      "epoch": 0.7762394982842267,
      "grad_norm": 0.15720601379871368,
      "learning_rate": 1.3142743323361818e-06,
      "loss": 1.5293,
      "mean_token_accuracy": 0.6503741011023522,
      "num_tokens": 636956832.0,
      "step": 1640
    },
    {
      "entropy": 1.518925702571869,
      "epoch": 0.7809726659566916,
      "grad_norm": 0.15746639668941498,
      "learning_rate": 1.2618410691334805e-06,
      "loss": 1.5244,
      "mean_token_accuracy": 0.6520501941442489,
      "num_tokens": 640891051.0,
      "step": 1650
    },
    {
      "entropy": 1.51315179169178,
      "epoch": 0.7857058336291564,
      "grad_norm": 0.15106885135173798,
      "learning_rate": 1.2103237182376848e-06,
      "loss": 1.5184,
      "mean_token_accuracy": 0.6530534133315087,
      "num_tokens": 644765813.0,
      "step": 1660
    },
    {
      "entropy": 1.5101027220487595,
      "epoch": 0.7904390013016211,
      "grad_norm": 0.15605495870113373,
      "learning_rate": 1.159734902273078e-06,
      "loss": 1.5137,
      "mean_token_accuracy": 0.6541712284088135,
      "num_tokens": 648685831.0,
      "step": 1670
    },
    {
      "entropy": 1.5037204265594482,
      "epoch": 0.7951721689740859,
      "grad_norm": 0.15608017146587372,
      "learning_rate": 1.1100870163571486e-06,
      "loss": 1.5085,
      "mean_token_accuracy": 0.6543751254677772,
      "num_tokens": 652549548.0,
      "step": 1680
    },
    {
      "entropy": 1.5166430234909059,
      "epoch": 0.7999053366465507,
      "grad_norm": 0.15508434176445007,
      "learning_rate": 1.061392225063571e-06,
      "loss": 1.524,
      "mean_token_accuracy": 0.6517237603664399,
      "num_tokens": 656455494.0,
      "step": 1690
    },
    {
      "entropy": 1.5094672977924346,
      "epoch": 0.8046385043190155,
      "grad_norm": 0.1599401831626892,
      "learning_rate": 1.0136624594416828e-06,
      "loss": 1.5145,
      "mean_token_accuracy": 0.6521976351737976,
      "num_tokens": 660321875.0,
      "step": 1700
    },
    {
      "entropy": 1.524332845211029,
      "epoch": 0.8093716719914803,
      "grad_norm": 0.15658311545848846,
      "learning_rate": 9.669094140931678e-07,
      "loss": 1.5249,
      "mean_token_accuracy": 0.651316037774086,
      "num_tokens": 664173208.0,
      "step": 1710
    },
    {
      "entropy": 1.5061070144176483,
      "epoch": 0.8141048396639451,
      "grad_norm": 0.15877439081668854,
      "learning_rate": 9.211445443066896e-07,
      "loss": 1.5095,
      "mean_token_accuracy": 0.6547744259238243,
      "num_tokens": 668090711.0,
      "step": 1720
    },
    {
      "entropy": 1.5110326170921327,
      "epoch": 0.8188380073364099,
      "grad_norm": 0.15988118946552277,
      "learning_rate": 8.763790632511482e-07,
      "loss": 1.52,
      "mean_token_accuracy": 0.6521027386188507,
      "num_tokens": 671930932.0,
      "step": 1730
    },
    {
      "entropy": 1.5050281405448913,
      "epoch": 0.8235711750088747,
      "grad_norm": 0.15555670857429504,
      "learning_rate": 8.326239392282586e-07,
      "loss": 1.5105,
      "mean_token_accuracy": 0.6531902849674225,
      "num_tokens": 675792125.0,
      "step": 1740
    },
    {
      "entropy": 1.5129504561424256,
      "epoch": 0.8283043426813395,
      "grad_norm": 0.16341622173786163,
      "learning_rate": 7.898898929851406e-07,
      "loss": 1.5162,
      "mean_token_accuracy": 0.6542815431952477,
      "num_tokens": 679648075.0,
      "step": 1750
    },
    {
      "entropy": 1.5132198542356492,
      "epoch": 0.8330375103538042,
      "grad_norm": 0.1575479656457901,
      "learning_rate": 7.481873950875468e-07,
      "loss": 1.5167,
      "mean_token_accuracy": 0.6529857292771339,
      "num_tokens": 683496935.0,
      "step": 1760
    },
    {
      "entropy": 1.514217236638069,
      "epoch": 0.837770678026269,
      "grad_norm": 0.16133052110671997,
      "learning_rate": 7.075266633543959e-07,
      "loss": 1.5176,
      "mean_token_accuracy": 0.6526405304670334,
      "num_tokens": 687412196.0,
      "step": 1770
    },
    {
      "entropy": 1.5245876908302307,
      "epoch": 0.8425038456987339,
      "grad_norm": 0.1658501774072647,
      "learning_rate": 6.679176603542398e-07,
      "loss": 1.5267,
      "mean_token_accuracy": 0.6507634833455086,
      "num_tokens": 691261759.0,
      "step": 1780
    },
    {
      "entropy": 1.520870679616928,
      "epoch": 0.8472370133711987,
      "grad_norm": 0.15542222559452057,
      "learning_rate": 6.29370090964262e-07,
      "loss": 1.5237,
      "mean_token_accuracy": 0.6512942254543305,
      "num_tokens": 695146169.0,
      "step": 1790
    },
    {
      "entropy": 1.5049445867538451,
      "epoch": 0.8519701810436635,
      "grad_norm": 0.15088842809200287,
      "learning_rate": 5.918933999924143e-07,
      "loss": 1.507,
      "mean_token_accuracy": 0.6546556517481804,
      "num_tokens": 699030254.0,
      "step": 1800
    },
    {
      "entropy": 1.5152593225240707,
      "epoch": 0.8567033487161283,
      "grad_norm": 0.15672406554222107,
      "learning_rate": 5.55496769863288e-07,
      "loss": 1.5197,
      "mean_token_accuracy": 0.652678707242012,
      "num_tokens": 702916640.0,
      "step": 1810
    },
    {
      "entropy": 1.5112317591905593,
      "epoch": 0.8614365163885931,
      "grad_norm": 0.15444321930408478,
      "learning_rate": 5.201891183682545e-07,
      "loss": 1.5167,
      "mean_token_accuracy": 0.6519044652581215,
      "num_tokens": 706729238.0,
      "step": 1820
    },
    {
      "entropy": 1.5071990102529527,
      "epoch": 0.8661696840610579,
      "grad_norm": 0.1569969356060028,
      "learning_rate": 4.859790964804617e-07,
      "loss": 1.5104,
      "mean_token_accuracy": 0.6534604758024216,
      "num_tokens": 710608784.0,
      "step": 1830
    },
    {
      "entropy": 1.5036645025014876,
      "epoch": 0.8709028517335227,
      "grad_norm": 0.16283676028251648,
      "learning_rate": 4.5287508623519773e-07,
      "loss": 1.5068,
      "mean_token_accuracy": 0.6550710022449493,
      "num_tokens": 714514707.0,
      "step": 1840
    },
    {
      "entropy": 1.5189965814352036,
      "epoch": 0.8756360194059875,
      "grad_norm": 0.15639755129814148,
      "learning_rate": 4.2088519867614585e-07,
      "loss": 1.5222,
      "mean_token_accuracy": 0.6517624288797379,
      "num_tokens": 718366975.0,
      "step": 1850
    },
    {
      "entropy": 1.5078352093696594,
      "epoch": 0.8803691870784522,
      "grad_norm": 0.1577681601047516,
      "learning_rate": 3.900172718680484e-07,
      "loss": 1.5131,
      "mean_token_accuracy": 0.6530882626771927,
      "num_tokens": 722251285.0,
      "step": 1860
    },
    {
      "entropy": 1.5013979017734527,
      "epoch": 0.885102354750917,
      "grad_norm": 0.16839657723903656,
      "learning_rate": 3.6027886897624376e-07,
      "loss": 1.5079,
      "mean_token_accuracy": 0.6545991912484169,
      "num_tokens": 726162080.0,
      "step": 1870
    },
    {
      "entropy": 1.5258357018232345,
      "epoch": 0.8898355224233818,
      "grad_norm": 0.1610095053911209,
      "learning_rate": 3.3167727641356064e-07,
      "loss": 1.529,
      "mean_token_accuracy": 0.6494020417332649,
      "num_tokens": 730048788.0,
      "step": 1880
    },
    {
      "entropy": 1.5098802983760833,
      "epoch": 0.8945686900958466,
      "grad_norm": 0.15612637996673584,
      "learning_rate": 3.0421950205502813e-07,
      "loss": 1.5136,
      "mean_token_accuracy": 0.6532332092523575,
      "num_tokens": 733943282.0,
      "step": 1890
    },
    {
      "entropy": 1.5100417882204056,
      "epoch": 0.8993018577683114,
      "grad_norm": 0.154027059674263,
      "learning_rate": 2.779122735208267e-07,
      "loss": 1.5139,
      "mean_token_accuracy": 0.6538966566324234,
      "num_tokens": 737803762.0,
      "step": 1900
    },
    {
      "entropy": 1.511214417219162,
      "epoch": 0.9040350254407762,
      "grad_norm": 0.1588786244392395,
      "learning_rate": 2.527620365279021e-07,
      "loss": 1.5141,
      "mean_token_accuracy": 0.6533711299300193,
      "num_tokens": 741703649.0,
      "step": 1910
    },
    {
      "entropy": 1.5116032123565675,
      "epoch": 0.9087681931132411,
      "grad_norm": 0.15595540404319763,
      "learning_rate": 2.2877495331066312e-07,
      "loss": 1.5156,
      "mean_token_accuracy": 0.6534523665904999,
      "num_tokens": 745612891.0,
      "step": 1920
    },
    {
      "entropy": 1.5098071992397308,
      "epoch": 0.9135013607857059,
      "grad_norm": 0.160483717918396,
      "learning_rate": 2.0595690111112398e-07,
      "loss": 1.5142,
      "mean_token_accuracy": 0.6534514829516411,
      "num_tokens": 749494675.0,
      "step": 1930
    },
    {
      "entropy": 1.5071230858564377,
      "epoch": 0.9182345284581707,
      "grad_norm": 0.15629717707633972,
      "learning_rate": 1.8431347073888217e-07,
      "loss": 1.5111,
      "mean_token_accuracy": 0.6530539333820343,
      "num_tokens": 753341723.0,
      "step": 1940
    },
    {
      "entropy": 1.5019009202718734,
      "epoch": 0.9229676961306355,
      "grad_norm": 0.15971773862838745,
      "learning_rate": 1.638499652012754e-07,
      "loss": 1.5059,
      "mean_token_accuracy": 0.6544443607330322,
      "num_tokens": 757233014.0,
      "step": 1950
    },
    {
      "entropy": 1.5118261843919754,
      "epoch": 0.9277008638031002,
      "grad_norm": 0.15779262781143188,
      "learning_rate": 1.44571398404057e-07,
      "loss": 1.515,
      "mean_token_accuracy": 0.6532472312450409,
      "num_tokens": 761115255.0,
      "step": 1960
    },
    {
      "entropy": 1.5036541551351548,
      "epoch": 0.932434031475565,
      "grad_norm": 0.1535780280828476,
      "learning_rate": 1.2648249392289925e-07,
      "loss": 1.5061,
      "mean_token_accuracy": 0.6545781284570694,
      "num_tokens": 765028626.0,
      "step": 1970
    },
    {
      "entropy": 1.50391583442688,
      "epoch": 0.9371671991480298,
      "grad_norm": 0.15700839459896088,
      "learning_rate": 1.095876838460408e-07,
      "loss": 1.506,
      "mean_token_accuracy": 0.6557769432663918,
      "num_tokens": 768939705.0,
      "step": 1980
    },
    {
      "entropy": 1.5056363612413406,
      "epoch": 0.9419003668204946,
      "grad_norm": 0.15855862200260162,
      "learning_rate": 9.38911076883503e-08,
      "loss": 1.5082,
      "mean_token_accuracy": 0.6543473988771439,
      "num_tokens": 772833689.0,
      "step": 1990
    },
    {
      "entropy": 1.525366762280464,
      "epoch": 0.9466335344929594,
      "grad_norm": 0.15759994089603424,
      "learning_rate": 7.939661137707167e-08,
      "loss": 1.5292,
      "mean_token_accuracy": 0.650723172724247,
      "num_tokens": 776737000.0,
      "step": 2000
    },
    {
      "entropy": 1.5107537627220153,
      "epoch": 0.9513667021654242,
      "grad_norm": 0.14716793596744537,
      "learning_rate": 6.610774630951167e-08,
      "loss": 1.514,
      "mean_token_accuracy": 0.6538170203566551,
      "num_tokens": 780623496.0,
      "step": 2010
    },
    {
      "entropy": 1.510766288638115,
      "epoch": 0.956099869837889,
      "grad_norm": 0.15204967558383942,
      "learning_rate": 5.402776848288216e-08,
      "loss": 1.5127,
      "mean_token_accuracy": 0.6531319335103035,
      "num_tokens": 784511197.0,
      "step": 2020
    },
    {
      "entropy": 1.5045825749635697,
      "epoch": 0.9608330375103538,
      "grad_norm": 0.162211611866951,
      "learning_rate": 4.315963769652931e-08,
      "loss": 1.5063,
      "mean_token_accuracy": 0.6537961974740029,
      "num_tokens": 788407251.0,
      "step": 2030
    },
    {
      "entropy": 1.5063252121210098,
      "epoch": 0.9655662051828185,
      "grad_norm": 0.15293368697166443,
      "learning_rate": 3.350601682673094e-08,
      "loss": 1.5086,
      "mean_token_accuracy": 0.654248197376728,
      "num_tokens": 792249083.0,
      "step": 2040
    },
    {
      "entropy": 1.5172866135835648,
      "epoch": 0.9702993728552834,
      "grad_norm": 0.15927255153656006,
      "learning_rate": 2.5069271174247844e-08,
      "loss": 1.5187,
      "mean_token_accuracy": 0.6526034936308861,
      "num_tokens": 796138376.0,
      "step": 2050
    },
    {
      "entropy": 1.515240240097046,
      "epoch": 0.9750325405277482,
      "grad_norm": 0.15734337270259857,
      "learning_rate": 1.785146788478298e-08,
      "loss": 1.5168,
      "mean_token_accuracy": 0.6522197261452675,
      "num_tokens": 800048611.0,
      "step": 2060
    },
    {
      "entropy": 1.505976414680481,
      "epoch": 0.979765708200213,
      "grad_norm": 0.15182293951511383,
      "learning_rate": 1.1854375442498234e-08,
      "loss": 1.5083,
      "mean_token_accuracy": 0.653746984899044,
      "num_tokens": 803953901.0,
      "step": 2070
    },
    {
      "entropy": 1.5067311346530914,
      "epoch": 0.9844988758726778,
      "grad_norm": 0.15344145894050598,
      "learning_rate": 7.0794632367032765e-09,
      "loss": 1.5107,
      "mean_token_accuracy": 0.6533343732357025,
      "num_tokens": 807795934.0,
      "step": 2080
    },
    {
      "entropy": 1.5039544194936751,
      "epoch": 0.9892320435451426,
      "grad_norm": 0.15625813603401184,
      "learning_rate": 3.5279012018324397e-09,
      "loss": 1.5083,
      "mean_token_accuracy": 0.6547796651721001,
      "num_tokens": 811672777.0,
      "step": 2090
    },
    {
      "entropy": 1.5102417439222335,
      "epoch": 0.9939652112176074,
      "grad_norm": 0.14735041558742523,
      "learning_rate": 1.2005595307906792e-09,
      "loss": 1.5121,
      "mean_token_accuracy": 0.6533789873123169,
      "num_tokens": 815540486.0,
      "step": 2100
    },
    {
      "entropy": 1.5077374666929244,
      "epoch": 0.9986983788900722,
      "grad_norm": 0.15531828999519348,
      "learning_rate": 9.800846173968481e-11,
      "loss": 1.5117,
      "mean_token_accuracy": 0.6538635641336441,
      "num_tokens": 819362988.0,
      "step": 2110
    }
  ],
  "logging_steps": 10,
  "max_steps": 2113,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 100,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 3.6414323458230452e+19,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}