{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 208,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 1.195672869682312,
      "epoch": 0.009685230024213076,
      "grad_norm": 1.0686899423599243,
      "learning_rate": 0.0,
      "loss": 2.0517563819885254,
      "mean_token_accuracy": 0.5433094203472137,
      "num_tokens": 97740.0,
      "step": 1
    },
    {
      "entropy": 1.216128945350647,
      "epoch": 0.01937046004842615,
      "grad_norm": 1.0773417949676514,
      "learning_rate": 3.125e-06,
      "loss": 2.066392660140991,
      "mean_token_accuracy": 0.5405685901641846,
      "num_tokens": 194347.0,
      "step": 2
    },
    {
      "entropy": 1.2114249616861343,
      "epoch": 0.029055690072639227,
      "grad_norm": 1.0359230041503906,
      "learning_rate": 6.25e-06,
      "loss": 2.060716152191162,
      "mean_token_accuracy": 0.5376822166144848,
      "num_tokens": 292306.0,
      "step": 3
    },
    {
      "entropy": 1.2836366146802902,
      "epoch": 0.0387409200968523,
      "grad_norm": 0.9555837512016296,
      "learning_rate": 9.375000000000001e-06,
      "loss": 2.048638343811035,
      "mean_token_accuracy": 0.5342461578547955,
      "num_tokens": 389675.0,
      "step": 4
    },
    {
      "entropy": 1.3411734849214554,
      "epoch": 0.048426150121065374,
      "grad_norm": 0.7932096719741821,
      "learning_rate": 1.25e-05,
      "loss": 2.0051188468933105,
      "mean_token_accuracy": 0.5393080413341522,
      "num_tokens": 482548.0,
      "step": 5
    },
    {
      "entropy": 1.3891097456216812,
      "epoch": 0.05811138014527845,
      "grad_norm": 0.6072916388511658,
      "learning_rate": 1.5625e-05,
      "loss": 1.8992807865142822,
      "mean_token_accuracy": 0.548196591436863,
      "num_tokens": 579772.0,
      "step": 6
    },
    {
      "entropy": 1.4907268434762955,
      "epoch": 0.06779661016949153,
      "grad_norm": 0.4963963031768799,
      "learning_rate": 1.8750000000000002e-05,
      "loss": 1.898533582687378,
      "mean_token_accuracy": 0.5459267124533653,
      "num_tokens": 677650.0,
      "step": 7
    },
    {
      "entropy": 1.5116273909807205,
      "epoch": 0.0774818401937046,
      "grad_norm": 0.40073174238204956,
      "learning_rate": 2.1875e-05,
      "loss": 1.8216513395309448,
      "mean_token_accuracy": 0.5549031123518944,
      "num_tokens": 774231.0,
      "step": 8
    },
    {
      "entropy": 1.615136444568634,
      "epoch": 0.08716707021791767,
      "grad_norm": 0.35884714126586914,
      "learning_rate": 2.5e-05,
      "loss": 1.8513133525848389,
      "mean_token_accuracy": 0.5457839816808701,
      "num_tokens": 868913.0,
      "step": 9
    },
    {
      "entropy": 1.6603400856256485,
      "epoch": 0.09685230024213075,
      "grad_norm": 0.27204808592796326,
      "learning_rate": 2.8125000000000003e-05,
      "loss": 1.7579457759857178,
      "mean_token_accuracy": 0.5631033554673195,
      "num_tokens": 966483.0,
      "step": 10
    },
    {
      "entropy": 1.6746521443128586,
      "epoch": 0.10653753026634383,
      "grad_norm": 0.24814128875732422,
      "learning_rate": 3.125e-05,
      "loss": 1.6909434795379639,
      "mean_token_accuracy": 0.5695518478751183,
      "num_tokens": 1062677.0,
      "step": 11
    },
    {
      "entropy": 1.7637551128864288,
      "epoch": 0.1162227602905569,
      "grad_norm": 0.25874465703964233,
      "learning_rate": 3.4375e-05,
      "loss": 1.7197679281234741,
      "mean_token_accuracy": 0.5657209157943726,
      "num_tokens": 1160086.0,
      "step": 12
    },
    {
      "entropy": 1.782708317041397,
      "epoch": 0.12590799031476999,
      "grad_norm": 0.2599520683288574,
      "learning_rate": 3.7500000000000003e-05,
      "loss": 1.6749770641326904,
      "mean_token_accuracy": 0.5753393173217773,
      "num_tokens": 1257647.0,
      "step": 13
    },
    {
      "entropy": 1.8490994721651077,
      "epoch": 0.13559322033898305,
      "grad_norm": 0.2954244911670685,
      "learning_rate": 4.0625000000000005e-05,
      "loss": 1.6933197975158691,
      "mean_token_accuracy": 0.568934366106987,
      "num_tokens": 1354505.0,
      "step": 14
    },
    {
      "entropy": 1.7893278002738953,
      "epoch": 0.14527845036319612,
      "grad_norm": 0.30461740493774414,
      "learning_rate": 4.375e-05,
      "loss": 1.6047017574310303,
      "mean_token_accuracy": 0.5875631719827652,
      "num_tokens": 1450622.0,
      "step": 15
    },
    {
      "entropy": 1.8156154304742813,
      "epoch": 0.1549636803874092,
      "grad_norm": 0.2959771454334259,
      "learning_rate": 4.6875e-05,
      "loss": 1.6153976917266846,
      "mean_token_accuracy": 0.5827990546822548,
      "num_tokens": 1547908.0,
      "step": 16
    },
    {
      "entropy": 1.7984451651573181,
      "epoch": 0.16464891041162227,
      "grad_norm": 0.3082118630409241,
      "learning_rate": 5e-05,
      "loss": 1.5840697288513184,
      "mean_token_accuracy": 0.588292121887207,
      "num_tokens": 1645426.0,
      "step": 17
    },
    {
      "entropy": 1.7235808968544006,
      "epoch": 0.17433414043583534,
      "grad_norm": 0.27267831563949585,
      "learning_rate": 4.999859193643945e-05,
      "loss": 1.5370752811431885,
      "mean_token_accuracy": 0.5960508286952972,
      "num_tokens": 1742103.0,
      "step": 18
    },
    {
      "entropy": 1.6560853272676468,
      "epoch": 0.18401937046004843,
      "grad_norm": 0.24492555856704712,
      "learning_rate": 4.999436790436924e-05,
      "loss": 1.5075464248657227,
      "mean_token_accuracy": 0.5996944680809975,
      "num_tokens": 1839477.0,
      "step": 19
    },
    {
      "entropy": 1.5435267090797424,
      "epoch": 0.1937046004842615,
      "grad_norm": 0.1962456852197647,
      "learning_rate": 4.9987328379605816e-05,
      "loss": 1.4266257286071777,
      "mean_token_accuracy": 0.6173319146037102,
      "num_tokens": 1936962.0,
      "step": 20
    },
    {
      "entropy": 1.5529854595661163,
      "epoch": 0.2033898305084746,
      "grad_norm": 0.17611059546470642,
      "learning_rate": 4.9977474155117045e-05,
      "loss": 1.4814770221710205,
      "mean_token_accuracy": 0.6056615859270096,
      "num_tokens": 2034780.0,
      "step": 21
    },
    {
      "entropy": 1.4628148823976517,
      "epoch": 0.21307506053268765,
      "grad_norm": 0.16322024166584015,
      "learning_rate": 4.9964806340932865e-05,
      "loss": 1.4442647695541382,
      "mean_token_accuracy": 0.6119256466627121,
      "num_tokens": 2132583.0,
      "step": 22
    },
    {
      "entropy": 1.4318041056394577,
      "epoch": 0.22276029055690072,
      "grad_norm": 0.17170685529708862,
      "learning_rate": 4.9949326364020314e-05,
      "loss": 1.443179965019226,
      "mean_token_accuracy": 0.6123129650950432,
      "num_tokens": 2226009.0,
      "step": 23
    },
    {
      "entropy": 1.3572796285152435,
      "epoch": 0.2324455205811138,
      "grad_norm": 0.21290098130702972,
      "learning_rate": 4.993103596812268e-05,
      "loss": 1.4151618480682373,
      "mean_token_accuracy": 0.6203208416700363,
      "num_tokens": 2323884.0,
      "step": 24
    },
    {
      "entropy": 1.3599127680063248,
      "epoch": 0.24213075060532688,
      "grad_norm": 0.21652130782604218,
      "learning_rate": 4.9909937213563165e-05,
      "loss": 1.4241068363189697,
      "mean_token_accuracy": 0.6187105998396873,
      "num_tokens": 2420260.0,
      "step": 25
    },
    {
      "entropy": 1.3542447090148926,
      "epoch": 0.25181598062953997,
      "grad_norm": 0.20405800640583038,
      "learning_rate": 4.988603247701277e-05,
      "loss": 1.408291220664978,
      "mean_token_accuracy": 0.6216152384877205,
      "num_tokens": 2517078.0,
      "step": 26
    },
    {
      "entropy": 1.3572800755500793,
      "epoch": 0.26150121065375304,
      "grad_norm": 0.19708576798439026,
      "learning_rate": 4.985932445122257e-05,
      "loss": 1.4086432456970215,
      "mean_token_accuracy": 0.618386335670948,
      "num_tokens": 2614502.0,
      "step": 27
    },
    {
      "entropy": 1.297940343618393,
      "epoch": 0.2711864406779661,
      "grad_norm": 0.19344690442085266,
      "learning_rate": 4.982981614472039e-05,
      "loss": 1.3473405838012695,
      "mean_token_accuracy": 0.6318695172667503,
      "num_tokens": 2710926.0,
      "step": 28
    },
    {
      "entropy": 1.3255801796913147,
      "epoch": 0.28087167070217917,
      "grad_norm": 0.15949787199497223,
      "learning_rate": 4.979751088147192e-05,
      "loss": 1.352310061454773,
      "mean_token_accuracy": 0.6299134269356728,
      "num_tokens": 2807646.0,
      "step": 29
    },
    {
      "entropy": 1.337433010339737,
      "epoch": 0.29055690072639223,
      "grad_norm": 0.14451353251934052,
      "learning_rate": 4.97624123005063e-05,
      "loss": 1.3375511169433594,
      "mean_token_accuracy": 0.6326661631464958,
      "num_tokens": 2905309.0,
      "step": 30
    },
    {
      "entropy": 1.3829680979251862,
      "epoch": 0.30024213075060535,
      "grad_norm": 0.14129243791103363,
      "learning_rate": 4.972452435550614e-05,
      "loss": 1.3608760833740234,
      "mean_token_accuracy": 0.6260306239128113,
      "num_tokens": 3002360.0,
      "step": 31
    },
    {
      "entropy": 1.37909235060215,
      "epoch": 0.3099273607748184,
      "grad_norm": 0.13715004920959473,
      "learning_rate": 4.968385131436222e-05,
      "loss": 1.3511570692062378,
      "mean_token_accuracy": 0.6329229399561882,
      "num_tokens": 3098670.0,
      "step": 32
    },
    {
      "entropy": 1.3565758764743805,
      "epoch": 0.3196125907990315,
      "grad_norm": 0.13790789246559143,
      "learning_rate": 4.9640397758692715e-05,
      "loss": 1.3203370571136475,
      "mean_token_accuracy": 0.6328470334410667,
      "num_tokens": 3196614.0,
      "step": 33
    },
    {
      "entropy": 1.3381365239620209,
      "epoch": 0.32929782082324455,
      "grad_norm": 0.1352091133594513,
      "learning_rate": 4.9594168583327094e-05,
      "loss": 1.2988011837005615,
      "mean_token_accuracy": 0.6376497745513916,
      "num_tokens": 3294374.0,
      "step": 34
    },
    {
      "entropy": 1.3610333651304245,
      "epoch": 0.3389830508474576,
      "grad_norm": 0.13593968749046326,
      "learning_rate": 4.954516899575473e-05,
      "loss": 1.33039128780365,
      "mean_token_accuracy": 0.6318908259272575,
      "num_tokens": 3392016.0,
      "step": 35
    },
    {
      "entropy": 1.3339500427246094,
      "epoch": 0.3486682808716707,
      "grad_norm": 0.13187174499034882,
      "learning_rate": 4.949340451553833e-05,
      "loss": 1.3074061870574951,
      "mean_token_accuracy": 0.6397572010755539,
      "num_tokens": 3489070.0,
      "step": 36
    },
    {
      "entropy": 1.30422942340374,
      "epoch": 0.3583535108958838,
      "grad_norm": 0.13068145513534546,
      "learning_rate": 4.943888097369216e-05,
      "loss": 1.2889633178710938,
      "mean_token_accuracy": 0.6402651146054268,
      "num_tokens": 3586431.0,
      "step": 37
    },
    {
      "entropy": 1.3068273216485977,
      "epoch": 0.36803874092009686,
      "grad_norm": 0.13225507736206055,
      "learning_rate": 4.938160451202521e-05,
      "loss": 1.293223261833191,
      "mean_token_accuracy": 0.6415960788726807,
      "num_tokens": 3684062.0,
      "step": 38
    },
    {
      "entropy": 1.2943807691335678,
      "epoch": 0.37772397094430993,
      "grad_norm": 0.12926499545574188,
      "learning_rate": 4.9321581582449365e-05,
      "loss": 1.277047872543335,
      "mean_token_accuracy": 0.6415077224373817,
      "num_tokens": 3780819.0,
      "step": 39
    },
    {
      "entropy": 1.2993432581424713,
      "epoch": 0.387409200968523,
      "grad_norm": 0.12387435138225555,
      "learning_rate": 4.9258818946252624e-05,
      "loss": 1.2970147132873535,
      "mean_token_accuracy": 0.6379837617278099,
      "num_tokens": 3878336.0,
      "step": 40
    },
    {
      "entropy": 1.2892408818006516,
      "epoch": 0.39709443099273606,
      "grad_norm": 0.12141715735197067,
      "learning_rate": 4.9193323673337476e-05,
      "loss": 1.2838318347930908,
      "mean_token_accuracy": 0.6403800770640373,
      "num_tokens": 3975553.0,
      "step": 41
    },
    {
      "entropy": 1.2638924419879913,
      "epoch": 0.4067796610169492,
      "grad_norm": 0.1264447420835495,
      "learning_rate": 4.912510314142448e-05,
      "loss": 1.2587647438049316,
      "mean_token_accuracy": 0.645272970199585,
      "num_tokens": 4072191.0,
      "step": 42
    },
    {
      "entropy": 1.2728796899318695,
      "epoch": 0.41646489104116224,
      "grad_norm": 0.12187587469816208,
      "learning_rate": 4.9054165035221236e-05,
      "loss": 1.2625339031219482,
      "mean_token_accuracy": 0.6477586179971695,
      "num_tokens": 4169972.0,
      "step": 43
    },
    {
      "entropy": 1.2749067693948746,
      "epoch": 0.4261501210653753,
      "grad_norm": 0.11995347589254379,
      "learning_rate": 4.898051734555676e-05,
      "loss": 1.267740249633789,
      "mean_token_accuracy": 0.6439903303980827,
      "num_tokens": 4266802.0,
      "step": 44
    },
    {
      "entropy": 1.260569915175438,
      "epoch": 0.4358353510895884,
      "grad_norm": 0.12481680512428284,
      "learning_rate": 4.890416836848127e-05,
      "loss": 1.2588658332824707,
      "mean_token_accuracy": 0.6484315246343613,
      "num_tokens": 4364720.0,
      "step": 45
    },
    {
      "entropy": 1.2508793324232101,
      "epoch": 0.44552058111380144,
      "grad_norm": 0.11975731700658798,
      "learning_rate": 4.88251267043318e-05,
      "loss": 1.2475931644439697,
      "mean_token_accuracy": 0.6490512862801552,
      "num_tokens": 4460849.0,
      "step": 46
    },
    {
      "entropy": 1.2509469985961914,
      "epoch": 0.4552058111380145,
      "grad_norm": 0.11921220272779465,
      "learning_rate": 4.87434012567633e-05,
      "loss": 1.254028081893921,
      "mean_token_accuracy": 0.6466348469257355,
      "num_tokens": 4558078.0,
      "step": 47
    },
    {
      "entropy": 1.2154233157634735,
      "epoch": 0.4648910411622276,
      "grad_norm": 0.11846459656953812,
      "learning_rate": 4.865900123174574e-05,
      "loss": 1.2160266637802124,
      "mean_token_accuracy": 0.6553884744644165,
      "num_tokens": 4656144.0,
      "step": 48
    },
    {
      "entropy": 1.2299266457557678,
      "epoch": 0.4745762711864407,
      "grad_norm": 0.11931712925434113,
      "learning_rate": 4.857193613652711e-05,
      "loss": 1.235276222229004,
      "mean_token_accuracy": 0.6523211896419525,
      "num_tokens": 4752916.0,
      "step": 49
    },
    {
      "entropy": 1.195318415760994,
      "epoch": 0.48426150121065376,
      "grad_norm": 0.1206982359290123,
      "learning_rate": 4.848221577856243e-05,
      "loss": 1.2011010646820068,
      "mean_token_accuracy": 0.6564221084117889,
      "num_tokens": 4850493.0,
      "step": 50
    },
    {
      "entropy": 1.20865198969841,
      "epoch": 0.4939467312348668,
      "grad_norm": 0.11764053255319595,
      "learning_rate": 4.838985026440905e-05,
      "loss": 1.2125191688537598,
      "mean_token_accuracy": 0.659000813961029,
      "num_tokens": 4948257.0,
      "step": 51
    },
    {
      "entropy": 1.2480878382921219,
      "epoch": 0.5036319612590799,
      "grad_norm": 0.12446491420269012,
      "learning_rate": 4.829484999858815e-05,
      "loss": 1.2488330602645874,
      "mean_token_accuracy": 0.6441424265503883,
      "num_tokens": 5045927.0,
      "step": 52
    },
    {
      "entropy": 1.264208972454071,
      "epoch": 0.513317191283293,
      "grad_norm": 0.12414462119340897,
      "learning_rate": 4.819722568241274e-05,
      "loss": 1.26497483253479,
      "mean_token_accuracy": 0.6444287970662117,
      "num_tokens": 5142165.0,
      "step": 53
    },
    {
      "entropy": 1.2086184322834015,
      "epoch": 0.5230024213075061,
      "grad_norm": 0.11623445153236389,
      "learning_rate": 4.8096988312782174e-05,
      "loss": 1.200068712234497,
      "mean_token_accuracy": 0.6566915512084961,
      "num_tokens": 5239937.0,
      "step": 54
    },
    {
      "entropy": 1.229715496301651,
      "epoch": 0.5326876513317191,
      "grad_norm": 0.1199541836977005,
      "learning_rate": 4.799414918094347e-05,
      "loss": 1.2203726768493652,
      "mean_token_accuracy": 0.6514962241053581,
      "num_tokens": 5336273.0,
      "step": 55
    },
    {
      "entropy": 1.1793622523546219,
      "epoch": 0.5423728813559322,
      "grad_norm": 0.1172776147723198,
      "learning_rate": 4.788871987121937e-05,
      "loss": 1.1629124879837036,
      "mean_token_accuracy": 0.666003905236721,
      "num_tokens": 5434119.0,
      "step": 56
    },
    {
      "entropy": 1.2295389622449875,
      "epoch": 0.5520581113801453,
      "grad_norm": 0.11812405288219452,
      "learning_rate": 4.77807122597034e-05,
      "loss": 1.2235515117645264,
      "mean_token_accuracy": 0.6521660089492798,
      "num_tokens": 5530374.0,
      "step": 57
    },
    {
      "entropy": 1.2132055461406708,
      "epoch": 0.5617433414043583,
      "grad_norm": 0.12084934115409851,
      "learning_rate": 4.767013851292212e-05,
      "loss": 1.2091310024261475,
      "mean_token_accuracy": 0.6549870446324348,
      "num_tokens": 5627067.0,
      "step": 58
    },
    {
      "entropy": 1.2059176713228226,
      "epoch": 0.5714285714285714,
      "grad_norm": 0.11870504170656204,
      "learning_rate": 4.755701108646463e-05,
      "loss": 1.1985912322998047,
      "mean_token_accuracy": 0.6569678112864494,
      "num_tokens": 5721452.0,
      "step": 59
    },
    {
      "entropy": 1.236282542347908,
      "epoch": 0.5811138014527845,
      "grad_norm": 0.11872227489948273,
      "learning_rate": 4.744134272357948e-05,
      "loss": 1.2417306900024414,
      "mean_token_accuracy": 0.6455686613917351,
      "num_tokens": 5818528.0,
      "step": 60
    },
    {
      "entropy": 1.2076577246189117,
      "epoch": 0.5907990314769975,
      "grad_norm": 0.11734182387590408,
      "learning_rate": 4.732314645373921e-05,
      "loss": 1.1962571144104004,
      "mean_token_accuracy": 0.6565083861351013,
      "num_tokens": 5915548.0,
      "step": 61
    },
    {
      "entropy": 1.1657419800758362,
      "epoch": 0.6004842615012107,
      "grad_norm": 0.12059835344552994,
      "learning_rate": 4.7202435591172676e-05,
      "loss": 1.16697096824646,
      "mean_token_accuracy": 0.6659450680017471,
      "num_tokens": 6011742.0,
      "step": 62
    },
    {
      "entropy": 1.1779616475105286,
      "epoch": 0.6101694915254238,
      "grad_norm": 0.11815991997718811,
      "learning_rate": 4.7079223733365237e-05,
      "loss": 1.182944655418396,
      "mean_token_accuracy": 0.661351315677166,
      "num_tokens": 6108610.0,
      "step": 63
    },
    {
      "entropy": 1.2032527029514313,
      "epoch": 0.6198547215496368,
      "grad_norm": 0.11827582865953445,
      "learning_rate": 4.6953524759527054e-05,
      "loss": 1.2088656425476074,
      "mean_token_accuracy": 0.6562395468354225,
      "num_tokens": 6205888.0,
      "step": 64
    },
    {
      "entropy": 1.1818794161081314,
      "epoch": 0.6295399515738499,
      "grad_norm": 0.11876095831394196,
      "learning_rate": 4.6825352829029705e-05,
      "loss": 1.1859397888183594,
      "mean_token_accuracy": 0.6605978757143021,
      "num_tokens": 6303176.0,
      "step": 65
    },
    {
      "entropy": 1.1840116828680038,
      "epoch": 0.639225181598063,
      "grad_norm": 0.11764654517173767,
      "learning_rate": 4.6694722379811185e-05,
      "loss": 1.188291072845459,
      "mean_token_accuracy": 0.6614853143692017,
      "num_tokens": 6400425.0,
      "step": 66
    },
    {
      "entropy": 1.1935285329818726,
      "epoch": 0.648910411622276,
      "grad_norm": 0.11767949163913727,
      "learning_rate": 4.656164812674951e-05,
      "loss": 1.196911096572876,
      "mean_token_accuracy": 0.6572826132178307,
      "num_tokens": 6497165.0,
      "step": 67
    },
    {
      "entropy": 1.2137638479471207,
      "epoch": 0.6585956416464891,
      "grad_norm": 0.11640281975269318,
      "learning_rate": 4.642614506000523e-05,
      "loss": 1.1995229721069336,
      "mean_token_accuracy": 0.6570323258638382,
      "num_tokens": 6594629.0,
      "step": 68
    },
    {
      "entropy": 1.1680169254541397,
      "epoch": 0.6682808716707022,
      "grad_norm": 0.12093165516853333,
      "learning_rate": 4.628822844333278e-05,
      "loss": 1.1655635833740234,
      "mean_token_accuracy": 0.6667345613241196,
      "num_tokens": 6691606.0,
      "step": 69
    },
    {
      "entropy": 1.1821613758802414,
      "epoch": 0.6779661016949152,
      "grad_norm": 0.1191101223230362,
      "learning_rate": 4.614791381236115e-05,
      "loss": 1.1815344095230103,
      "mean_token_accuracy": 0.6611361801624298,
      "num_tokens": 6787144.0,
      "step": 70
    },
    {
      "entropy": 1.180012434720993,
      "epoch": 0.6876513317191283,
      "grad_norm": 0.11644171178340912,
      "learning_rate": 4.6005216972843864e-05,
      "loss": 1.187152624130249,
      "mean_token_accuracy": 0.6607257425785065,
      "num_tokens": 6884460.0,
      "step": 71
    },
    {
      "entropy": 1.1491027027368546,
      "epoch": 0.6973365617433414,
      "grad_norm": 0.11654338240623474,
      "learning_rate": 4.5860153998878494e-05,
      "loss": 1.1466959714889526,
      "mean_token_accuracy": 0.6696180328726768,
      "num_tokens": 6982209.0,
      "step": 72
    },
    {
      "entropy": 1.1703396141529083,
      "epoch": 0.7070217917675545,
      "grad_norm": 0.11882360279560089,
      "learning_rate": 4.571274123109606e-05,
      "loss": 1.168134331703186,
      "mean_token_accuracy": 0.6635337024927139,
      "num_tokens": 7078357.0,
      "step": 73
    },
    {
      "entropy": 1.1620274484157562,
      "epoch": 0.7167070217917676,
      "grad_norm": 0.11732776463031769,
      "learning_rate": 4.5562995274820284e-05,
      "loss": 1.1588343381881714,
      "mean_token_accuracy": 0.6665216088294983,
      "num_tokens": 7174932.0,
      "step": 74
    },
    {
      "entropy": 1.1402729004621506,
      "epoch": 0.7263922518159807,
      "grad_norm": 0.11696197837591171,
      "learning_rate": 4.541093299819714e-05,
      "loss": 1.1440062522888184,
      "mean_token_accuracy": 0.673184260725975,
      "num_tokens": 7271358.0,
      "step": 75
    },
    {
      "entropy": 1.1796619296073914,
      "epoch": 0.7360774818401937,
      "grad_norm": 0.11728471517562866,
      "learning_rate": 4.5256571530294664e-05,
      "loss": 1.1810756921768188,
      "mean_token_accuracy": 0.6604448780417442,
      "num_tokens": 7368532.0,
      "step": 76
    },
    {
      "entropy": 1.1303958296775818,
      "epoch": 0.7457627118644068,
      "grad_norm": 0.11636237055063248,
      "learning_rate": 4.5099928259173516e-05,
      "loss": 1.126218318939209,
      "mean_token_accuracy": 0.6715309843420982,
      "num_tokens": 7465042.0,
      "step": 77
    },
    {
      "entropy": 1.1526209563016891,
      "epoch": 0.7554479418886199,
      "grad_norm": 0.11668036133050919,
      "learning_rate": 4.4941020829928245e-05,
      "loss": 1.1493991613388062,
      "mean_token_accuracy": 0.6683483645319939,
      "num_tokens": 7561958.0,
      "step": 78
    },
    {
      "entropy": 1.1575733572244644,
      "epoch": 0.7651331719128329,
      "grad_norm": 0.12059628218412399,
      "learning_rate": 4.477986714269972e-05,
      "loss": 1.155458688735962,
      "mean_token_accuracy": 0.6660623773932457,
      "num_tokens": 7659711.0,
      "step": 79
    },
    {
      "entropy": 1.1208164542913437,
      "epoch": 0.774818401937046,
      "grad_norm": 0.11503883451223373,
      "learning_rate": 4.4616485350658685e-05,
      "loss": 1.1200377941131592,
      "mean_token_accuracy": 0.6769244968891144,
      "num_tokens": 7756475.0,
      "step": 80
    },
    {
      "entropy": 1.1618834882974625,
      "epoch": 0.784503631961259,
      "grad_norm": 0.11844263970851898,
      "learning_rate": 4.445089385796099e-05,
      "loss": 1.1612154245376587,
      "mean_token_accuracy": 0.6636371240019798,
      "num_tokens": 7853572.0,
      "step": 81
    },
    {
      "entropy": 1.177445963025093,
      "epoch": 0.7941888619854721,
      "grad_norm": 0.12128449976444244,
      "learning_rate": 4.4283111317674374e-05,
      "loss": 1.1839146614074707,
      "mean_token_accuracy": 0.6612392961978912,
      "num_tokens": 7951478.0,
      "step": 82
    },
    {
      "entropy": 1.1343424022197723,
      "epoch": 0.8038740920096852,
      "grad_norm": 0.11806600540876389,
      "learning_rate": 4.4113156629677316e-05,
      "loss": 1.1392714977264404,
      "mean_token_accuracy": 0.6691810935735703,
      "num_tokens": 8048090.0,
      "step": 83
    },
    {
      "entropy": 1.1562943011522293,
      "epoch": 0.8135593220338984,
      "grad_norm": 0.11784016340970993,
      "learning_rate": 4.394104893853007e-05,
      "loss": 1.1472623348236084,
      "mean_token_accuracy": 0.665581151843071,
      "num_tokens": 8145495.0,
      "step": 84
    },
    {
      "entropy": 1.1324752867221832,
      "epoch": 0.8232445520581114,
      "grad_norm": 0.11986548453569412,
      "learning_rate": 4.3766807631318106e-05,
      "loss": 1.1278144121170044,
      "mean_token_accuracy": 0.6711176708340645,
      "num_tokens": 8243257.0,
      "step": 85
    },
    {
      "entropy": 1.127712443470955,
      "epoch": 0.8329297820823245,
      "grad_norm": 0.11755618453025818,
      "learning_rate": 4.359045233546827e-05,
      "loss": 1.128782033920288,
      "mean_token_accuracy": 0.6730731576681137,
      "num_tokens": 8338257.0,
      "step": 86
    },
    {
      "entropy": 1.1132700890302658,
      "epoch": 0.8426150121065376,
      "grad_norm": 0.11721386760473251,
      "learning_rate": 4.341200291653781e-05,
      "loss": 1.1207164525985718,
      "mean_token_accuracy": 0.6753561720252037,
      "num_tokens": 8436280.0,
      "step": 87
    },
    {
      "entropy": 1.1623711436986923,
      "epoch": 0.8523002421307506,
      "grad_norm": 0.12041810154914856,
      "learning_rate": 4.323147947597666e-05,
      "loss": 1.1799827814102173,
      "mean_token_accuracy": 0.6590451151132584,
      "num_tokens": 8534296.0,
      "step": 88
    },
    {
      "entropy": 1.0908512026071548,
      "epoch": 0.8619854721549637,
      "grad_norm": 0.11574803292751312,
      "learning_rate": 4.3048902348863116e-05,
      "loss": 1.0909357070922852,
      "mean_token_accuracy": 0.6803726628422737,
      "num_tokens": 8631835.0,
      "step": 89
    },
    {
      "entropy": 1.1286626309156418,
      "epoch": 0.8716707021791767,
      "grad_norm": 0.1222979798913002,
      "learning_rate": 4.2864292101613136e-05,
      "loss": 1.1375240087509155,
      "mean_token_accuracy": 0.6725632026791573,
      "num_tokens": 8728666.0,
      "step": 90
    },
    {
      "entropy": 1.0889050513505936,
      "epoch": 0.8813559322033898,
      "grad_norm": 0.116656094789505,
      "learning_rate": 4.267766952966369e-05,
      "loss": 1.0847504138946533,
      "mean_token_accuracy": 0.6843830943107605,
      "num_tokens": 8826157.0,
      "step": 91
    },
    {
      "entropy": 1.1236678957939148,
      "epoch": 0.8910411622276029,
      "grad_norm": 0.12352480739355087,
      "learning_rate": 4.248905565513023e-05,
      "loss": 1.1190218925476074,
      "mean_token_accuracy": 0.6742537319660187,
      "num_tokens": 8919599.0,
      "step": 92
    },
    {
      "entropy": 1.102852076292038,
      "epoch": 0.9007263922518159,
      "grad_norm": 0.12034481763839722,
      "learning_rate": 4.229847172443866e-05,
      "loss": 1.1002410650253296,
      "mean_token_accuracy": 0.6793207675218582,
      "num_tokens": 9017083.0,
      "step": 93
    },
    {
      "entropy": 1.1279104799032211,
      "epoch": 0.910411622276029,
      "grad_norm": 0.11987902224063873,
      "learning_rate": 4.210593920593201e-05,
      "loss": 1.1197580099105835,
      "mean_token_accuracy": 0.6761348396539688,
      "num_tokens": 9114699.0,
      "step": 94
    },
    {
      "entropy": 1.128271833062172,
      "epoch": 0.9200968523002422,
      "grad_norm": 0.12280263006687164,
      "learning_rate": 4.191147978745218e-05,
      "loss": 1.1215918064117432,
      "mean_token_accuracy": 0.6710417121648788,
      "num_tokens": 9212478.0,
      "step": 95
    },
    {
      "entropy": 1.1278261244297028,
      "epoch": 0.9297820823244553,
      "grad_norm": 0.12078865617513657,
      "learning_rate": 4.171511537389684e-05,
      "loss": 1.1220818758010864,
      "mean_token_accuracy": 0.6726439297199249,
      "num_tokens": 9309385.0,
      "step": 96
    },
    {
      "entropy": 1.1121345162391663,
      "epoch": 0.9394673123486683,
      "grad_norm": 0.12248018383979797,
      "learning_rate": 4.151686808475204e-05,
      "loss": 1.112907886505127,
      "mean_token_accuracy": 0.6733067110180855,
      "num_tokens": 9405674.0,
      "step": 97
    },
    {
      "entropy": 1.1223307251930237,
      "epoch": 0.9491525423728814,
      "grad_norm": 0.11941546201705933,
      "learning_rate": 4.131676025160047e-05,
      "loss": 1.1234936714172363,
      "mean_token_accuracy": 0.6732581779360771,
      "num_tokens": 9502863.0,
      "step": 98
    },
    {
      "entropy": 1.0854229852557182,
      "epoch": 0.9588377723970944,
      "grad_norm": 0.12412574887275696,
      "learning_rate": 4.111481441560598e-05,
      "loss": 1.0869731903076172,
      "mean_token_accuracy": 0.682966060936451,
      "num_tokens": 9600669.0,
      "step": 99
    },
    {
      "entropy": 1.1000574603676796,
      "epoch": 0.9685230024213075,
      "grad_norm": 0.12113860994577408,
      "learning_rate": 4.091105332497439e-05,
      "loss": 1.1088688373565674,
      "mean_token_accuracy": 0.6795928180217743,
      "num_tokens": 9697304.0,
      "step": 100
    },
    {
      "entropy": 1.1247247457504272,
      "epoch": 0.9782082324455206,
      "grad_norm": 0.12144359946250916,
      "learning_rate": 4.070549993239106e-05,
      "loss": 1.1281317472457886,
      "mean_token_accuracy": 0.6737656146287918,
      "num_tokens": 9794066.0,
      "step": 101
    },
    {
      "entropy": 1.102698415517807,
      "epoch": 0.9878934624697336,
      "grad_norm": 0.11878343671560287,
      "learning_rate": 4.049817739243532e-05,
      "loss": 1.0992276668548584,
      "mean_token_accuracy": 0.678807906806469,
      "num_tokens": 9889545.0,
      "step": 102
    },
    {
      "entropy": 1.1041645109653473,
      "epoch": 0.9975786924939467,
      "grad_norm": 0.1187015026807785,
      "learning_rate": 4.028910905897229e-05,
      "loss": 1.103971004486084,
      "mean_token_accuracy": 0.6783908009529114,
      "num_tokens": 9984866.0,
      "step": 103
    },
    {
      "entropy": 1.1201090216636658,
      "epoch": 1.0,
      "grad_norm": 0.232050359249115,
      "learning_rate": 4.007831848252211e-05,
      "loss": 1.108691692352295,
      "mean_token_accuracy": 0.6691955626010895,
      "num_tokens": 10009403.0,
      "step": 104
    },
    {
      "entropy": 1.1104417890310287,
      "epoch": 1.0096852300242132,
      "grad_norm": 0.1235378161072731,
      "learning_rate": 3.986582940760717e-05,
      "loss": 1.0965468883514404,
      "mean_token_accuracy": 0.6793974786996841,
      "num_tokens": 10106335.0,
      "step": 105
    },
    {
      "entropy": 1.0799510329961777,
      "epoch": 1.0193704600484261,
      "grad_norm": 0.12456927448511124,
      "learning_rate": 3.965166577007733e-05,
      "loss": 1.0694092512130737,
      "mean_token_accuracy": 0.6838481202721596,
      "num_tokens": 10202191.0,
      "step": 106
    },
    {
      "entropy": 1.081312820315361,
      "epoch": 1.0290556900726393,
      "grad_norm": 0.12274689227342606,
      "learning_rate": 3.94358516944137e-05,
      "loss": 1.0728635787963867,
      "mean_token_accuracy": 0.6833763420581818,
      "num_tokens": 10299771.0,
      "step": 107
    },
    {
      "entropy": 1.0861759409308434,
      "epoch": 1.0387409200968523,
      "grad_norm": 0.12080616503953934,
      "learning_rate": 3.9218411491011176e-05,
      "loss": 1.0843992233276367,
      "mean_token_accuracy": 0.6849978193640709,
      "num_tokens": 10397225.0,
      "step": 108
    },
    {
      "entropy": 1.0708721578121185,
      "epoch": 1.0484261501210654,
      "grad_norm": 0.12346586585044861,
      "learning_rate": 3.899936965343989e-05,
      "loss": 1.0690574645996094,
      "mean_token_accuracy": 0.6835706457495689,
      "num_tokens": 10495072.0,
      "step": 109
    },
    {
      "entropy": 1.0603735968470573,
      "epoch": 1.0581113801452784,
      "grad_norm": 0.12792955338954926,
      "learning_rate": 3.877875085568623e-05,
      "loss": 1.0686826705932617,
      "mean_token_accuracy": 0.6861361637711525,
      "num_tokens": 10593014.0,
      "step": 110
    },
    {
      "entropy": 1.0885941982269287,
      "epoch": 1.0677966101694916,
      "grad_norm": 0.1284114122390747,
      "learning_rate": 3.8556579949373384e-05,
      "loss": 1.0875612497329712,
      "mean_token_accuracy": 0.6810608729720116,
      "num_tokens": 10690703.0,
      "step": 111
    },
    {
      "entropy": 1.0722324550151825,
      "epoch": 1.0774818401937045,
      "grad_norm": 0.12488774955272675,
      "learning_rate": 3.833288196096194e-05,
      "loss": 1.0810635089874268,
      "mean_token_accuracy": 0.6832909435033798,
      "num_tokens": 10788092.0,
      "step": 112
    },
    {
      "entropy": 1.0997016876935959,
      "epoch": 1.0871670702179177,
      "grad_norm": 0.127477765083313,
      "learning_rate": 3.8107682088930794e-05,
      "loss": 1.0950262546539307,
      "mean_token_accuracy": 0.6799236834049225,
      "num_tokens": 10884533.0,
      "step": 113
    },
    {
      "entropy": 1.0732601955533028,
      "epoch": 1.0968523002421307,
      "grad_norm": 0.12743300199508667,
      "learning_rate": 3.7881005700938635e-05,
      "loss": 1.0789378881454468,
      "mean_token_accuracy": 0.6835982874035835,
      "num_tokens": 10980390.0,
      "step": 114
    },
    {
      "entropy": 1.1172342747449875,
      "epoch": 1.1065375302663438,
      "grad_norm": 0.13004903495311737,
      "learning_rate": 3.7652878330966415e-05,
      "loss": 1.111201286315918,
      "mean_token_accuracy": 0.6758027002215385,
      "num_tokens": 11076573.0,
      "step": 115
    },
    {
      "entropy": 1.077336147427559,
      "epoch": 1.116222760290557,
      "grad_norm": 0.12165925651788712,
      "learning_rate": 3.742332567644107e-05,
      "loss": 1.0708179473876953,
      "mean_token_accuracy": 0.6829231157898903,
      "num_tokens": 11173211.0,
      "step": 116
    },
    {
      "entropy": 1.0420297384262085,
      "epoch": 1.12590799031477,
      "grad_norm": 0.12317481637001038,
      "learning_rate": 3.719237359534087e-05,
      "loss": 1.0381829738616943,
      "mean_token_accuracy": 0.6936260089278221,
      "num_tokens": 11269149.0,
      "step": 117
    },
    {
      "entropy": 1.0763309448957443,
      "epoch": 1.1355932203389831,
      "grad_norm": 0.12462957948446274,
      "learning_rate": 3.6960048103282566e-05,
      "loss": 1.0847418308258057,
      "mean_token_accuracy": 0.6818151772022247,
      "num_tokens": 11366538.0,
      "step": 118
    },
    {
      "entropy": 1.0835078358650208,
      "epoch": 1.145278450363196,
      "grad_norm": 0.1280977874994278,
      "learning_rate": 3.672637537059093e-05,
      "loss": 1.0795782804489136,
      "mean_token_accuracy": 0.6846858412027359,
      "num_tokens": 11464379.0,
      "step": 119
    },
    {
      "entropy": 1.099198043346405,
      "epoch": 1.1549636803874093,
      "grad_norm": 0.12681786715984344,
      "learning_rate": 3.6491381719350756e-05,
      "loss": 1.1035975217819214,
      "mean_token_accuracy": 0.6763817369937897,
      "num_tokens": 11561148.0,
      "step": 120
    },
    {
      "entropy": 1.0874132961034775,
      "epoch": 1.1646489104116222,
      "grad_norm": 0.12863373756408691,
      "learning_rate": 3.6255093620441834e-05,
      "loss": 1.0856728553771973,
      "mean_token_accuracy": 0.6814989522099495,
      "num_tokens": 11658242.0,
      "step": 121
    },
    {
      "entropy": 1.0825031399726868,
      "epoch": 1.1743341404358354,
      "grad_norm": 0.1271398812532425,
      "learning_rate": 3.6017537690557115e-05,
      "loss": 1.0711510181427002,
      "mean_token_accuracy": 0.684900738298893,
      "num_tokens": 11754691.0,
      "step": 122
    },
    {
      "entropy": 1.0565595179796219,
      "epoch": 1.1840193704600483,
      "grad_norm": 0.1259683221578598,
      "learning_rate": 3.577874068920446e-05,
      "loss": 1.042423129081726,
      "mean_token_accuracy": 0.6939693465828896,
      "num_tokens": 11851957.0,
      "step": 123
    },
    {
      "entropy": 1.085630625486374,
      "epoch": 1.1937046004842615,
      "grad_norm": 0.1252395212650299,
      "learning_rate": 3.553872951569236e-05,
      "loss": 1.0735639333724976,
      "mean_token_accuracy": 0.6810360923409462,
      "num_tokens": 11949158.0,
      "step": 124
    },
    {
      "entropy": 1.0829370319843292,
      "epoch": 1.2033898305084745,
      "grad_norm": 0.1221250519156456,
      "learning_rate": 3.529753120609982e-05,
      "loss": 1.0764325857162476,
      "mean_token_accuracy": 0.6828718408942223,
      "num_tokens": 12047065.0,
      "step": 125
    },
    {
      "entropy": 1.08208729326725,
      "epoch": 1.2130750605326877,
      "grad_norm": 0.1301564872264862,
      "learning_rate": 3.505517293023088e-05,
      "loss": 1.0983633995056152,
      "mean_token_accuracy": 0.6789150461554527,
      "num_tokens": 12140843.0,
      "step": 126
    },
    {
      "entropy": 1.0690992325544357,
      "epoch": 1.2227602905569008,
      "grad_norm": 0.1276232749223709,
      "learning_rate": 3.481168198855409e-05,
      "loss": 1.0886598825454712,
      "mean_token_accuracy": 0.6814791783690453,
      "num_tokens": 12238169.0,
      "step": 127
    },
    {
      "entropy": 1.0706257745623589,
      "epoch": 1.2324455205811138,
      "grad_norm": 0.1339784413576126,
      "learning_rate": 3.456708580912725e-05,
      "loss": 1.0642316341400146,
      "mean_token_accuracy": 0.6837696209549904,
      "num_tokens": 12336055.0,
      "step": 128
    },
    {
      "entropy": 1.0820157527923584,
      "epoch": 1.242130750605327,
      "grad_norm": 0.12442120909690857,
      "learning_rate": 3.432141194450772e-05,
      "loss": 1.089120626449585,
      "mean_token_accuracy": 0.6795459613204002,
      "num_tokens": 12433292.0,
      "step": 129
    },
    {
      "entropy": 1.061967819929123,
      "epoch": 1.25181598062954,
      "grad_norm": 0.1252843737602234,
      "learning_rate": 3.407468806864883e-05,
      "loss": 1.061234951019287,
      "mean_token_accuracy": 0.6862892434000969,
      "num_tokens": 12529632.0,
      "step": 130
    },
    {
      "entropy": 1.0756078958511353,
      "epoch": 1.261501210653753,
      "grad_norm": 0.12569892406463623,
      "learning_rate": 3.382694197378252e-05,
      "loss": 1.0803606510162354,
      "mean_token_accuracy": 0.6807011142373085,
      "num_tokens": 12627363.0,
      "step": 131
    },
    {
      "entropy": 1.080217882990837,
      "epoch": 1.271186440677966,
      "grad_norm": 0.13052122294902802,
      "learning_rate": 3.357820156728867e-05,
      "loss": 1.0745978355407715,
      "mean_token_accuracy": 0.6857112869620323,
      "num_tokens": 12722873.0,
      "step": 132
    },
    {
      "entropy": 1.0762659460306168,
      "epoch": 1.2808716707021792,
      "grad_norm": 0.12694182991981506,
      "learning_rate": 3.332849486855144e-05,
      "loss": 1.0733556747436523,
      "mean_token_accuracy": 0.6827760264277458,
      "num_tokens": 12818842.0,
      "step": 133
    },
    {
      "entropy": 1.061683513224125,
      "epoch": 1.2905569007263922,
      "grad_norm": 0.12459179759025574,
      "learning_rate": 3.307785000580313e-05,
      "loss": 1.0536550283432007,
      "mean_token_accuracy": 0.68765439838171,
      "num_tokens": 12916424.0,
      "step": 134
    },
    {
      "entropy": 1.0600966960191727,
      "epoch": 1.3002421307506054,
      "grad_norm": 0.12580938637256622,
      "learning_rate": 3.282629521295556e-05,
      "loss": 1.0634956359863281,
      "mean_token_accuracy": 0.6860811784863472,
      "num_tokens": 13012418.0,
      "step": 135
    },
    {
      "entropy": 1.042241021990776,
      "epoch": 1.3099273607748185,
      "grad_norm": 0.12430591136217117,
      "learning_rate": 3.257385882641971e-05,
      "loss": 1.0436371564865112,
      "mean_token_accuracy": 0.6909345462918282,
      "num_tokens": 13109173.0,
      "step": 136
    },
    {
      "entropy": 1.0537392497062683,
      "epoch": 1.3196125907990315,
      "grad_norm": 0.1255572885274887,
      "learning_rate": 3.232056928191376e-05,
      "loss": 1.0527831315994263,
      "mean_token_accuracy": 0.6868399828672409,
      "num_tokens": 13207226.0,
      "step": 137
    },
    {
      "entropy": 1.0547452345490456,
      "epoch": 1.3292978208232444,
      "grad_norm": 0.12568867206573486,
      "learning_rate": 3.2066455111259945e-05,
      "loss": 1.061872959136963,
      "mean_token_accuracy": 0.6875149309635162,
      "num_tokens": 13304690.0,
      "step": 138
    },
    {
      "entropy": 1.054403394460678,
      "epoch": 1.3389830508474576,
      "grad_norm": 0.12822820246219635,
      "learning_rate": 3.1811544939170575e-05,
      "loss": 1.0644862651824951,
      "mean_token_accuracy": 0.6821185201406479,
      "num_tokens": 13402318.0,
      "step": 139
    },
    {
      "entropy": 1.0602877885103226,
      "epoch": 1.3486682808716708,
      "grad_norm": 0.12605498731136322,
      "learning_rate": 3.155586748002362e-05,
      "loss": 1.0644670724868774,
      "mean_token_accuracy": 0.6855698898434639,
      "num_tokens": 13498911.0,
      "step": 140
    },
    {
      "entropy": 1.0703921020030975,
      "epoch": 1.3583535108958837,
      "grad_norm": 0.12941165268421173,
      "learning_rate": 3.1299451534628135e-05,
      "loss": 1.068335771560669,
      "mean_token_accuracy": 0.6835189908742905,
      "num_tokens": 13596489.0,
      "step": 141
    },
    {
      "entropy": 1.0610421299934387,
      "epoch": 1.368038740920097,
      "grad_norm": 0.12477688491344452,
      "learning_rate": 3.1042325986980064e-05,
      "loss": 1.0563592910766602,
      "mean_token_accuracy": 0.6854637488722801,
      "num_tokens": 13693181.0,
      "step": 142
    },
    {
      "entropy": 1.075249344110489,
      "epoch": 1.3777239709443099,
      "grad_norm": 0.128186896443367,
      "learning_rate": 3.0784519801008546e-05,
      "loss": 1.068171739578247,
      "mean_token_accuracy": 0.6825502961874008,
      "num_tokens": 13790207.0,
      "step": 143
    },
    {
      "entropy": 1.0562544986605644,
      "epoch": 1.387409200968523,
      "grad_norm": 0.12701760232448578,
      "learning_rate": 3.0526062017313254e-05,
      "loss": 1.0526593923568726,
      "mean_token_accuracy": 0.6873726323246956,
      "num_tokens": 13886824.0,
      "step": 144
    },
    {
      "entropy": 1.0743589103221893,
      "epoch": 1.397094430992736,
      "grad_norm": 0.12756222486495972,
      "learning_rate": 3.0266981749893157e-05,
      "loss": 1.0770368576049805,
      "mean_token_accuracy": 0.6815621331334114,
      "num_tokens": 13984711.0,
      "step": 145
    },
    {
      "entropy": 1.0754915922880173,
      "epoch": 1.4067796610169492,
      "grad_norm": 0.12397201359272003,
      "learning_rate": 3.0007308182866985e-05,
      "loss": 1.070479393005371,
      "mean_token_accuracy": 0.6830718889832497,
      "num_tokens": 14082491.0,
      "step": 146
    },
    {
      "entropy": 1.0787562653422356,
      "epoch": 1.4164648910411621,
      "grad_norm": 0.12594996392726898,
      "learning_rate": 2.974707056718571e-05,
      "loss": 1.0885682106018066,
      "mean_token_accuracy": 0.6782692447304726,
      "num_tokens": 14178367.0,
      "step": 147
    },
    {
      "entropy": 1.0706654638051987,
      "epoch": 1.4261501210653753,
      "grad_norm": 0.12751780450344086,
      "learning_rate": 2.9486298217337637e-05,
      "loss": 1.084141731262207,
      "mean_token_accuracy": 0.6798113733530045,
      "num_tokens": 14275669.0,
      "step": 148
    },
    {
      "entropy": 1.0344142615795135,
      "epoch": 1.4358353510895885,
      "grad_norm": 0.12485277652740479,
      "learning_rate": 2.9225020508046232e-05,
      "loss": 1.0383057594299316,
      "mean_token_accuracy": 0.691432312130928,
      "num_tokens": 14372400.0,
      "step": 149
    },
    {
      "entropy": 1.0746425688266754,
      "epoch": 1.4455205811138014,
      "grad_norm": 0.13012577593326569,
      "learning_rate": 2.8963266870961226e-05,
      "loss": 1.0656825304031372,
      "mean_token_accuracy": 0.6830078810453415,
      "num_tokens": 14468895.0,
      "step": 150
    },
    {
      "entropy": 1.061945840716362,
      "epoch": 1.4552058111380144,
      "grad_norm": 0.12334053218364716,
      "learning_rate": 2.8701066791343288e-05,
      "loss": 1.0595557689666748,
      "mean_token_accuracy": 0.6865139305591583,
      "num_tokens": 14566496.0,
      "step": 151
    },
    {
      "entropy": 1.0778489708900452,
      "epoch": 1.4648910411622276,
      "grad_norm": 0.125824972987175,
      "learning_rate": 2.8438449804742628e-05,
      "loss": 1.0726466178894043,
      "mean_token_accuracy": 0.680848240852356,
      "num_tokens": 14664097.0,
      "step": 152
    },
    {
      "entropy": 1.0415344908833504,
      "epoch": 1.4745762711864407,
      "grad_norm": 0.12552373111248016,
      "learning_rate": 2.8175445493671972e-05,
      "loss": 1.0375463962554932,
      "mean_token_accuracy": 0.6916451752185822,
      "num_tokens": 14761778.0,
      "step": 153
    },
    {
      "entropy": 1.072740688920021,
      "epoch": 1.4842615012106537,
      "grad_norm": 0.12327924370765686,
      "learning_rate": 2.7912083484274266e-05,
      "loss": 1.0719635486602783,
      "mean_token_accuracy": 0.6858290657401085,
      "num_tokens": 14858797.0,
      "step": 154
    },
    {
      "entropy": 1.082845151424408,
      "epoch": 1.4939467312348669,
      "grad_norm": 0.12639667093753815,
      "learning_rate": 2.7648393442985403e-05,
      "loss": 1.067570686340332,
      "mean_token_accuracy": 0.6845085397362709,
      "num_tokens": 14955777.0,
      "step": 155
    },
    {
      "entropy": 1.0947035253047943,
      "epoch": 1.5036319612590798,
      "grad_norm": 0.12689661979675293,
      "learning_rate": 2.7384405073192454e-05,
      "loss": 1.0844032764434814,
      "mean_token_accuracy": 0.679810144007206,
      "num_tokens": 15052841.0,
      "step": 156
    },
    {
      "entropy": 1.0554356947541237,
      "epoch": 1.513317191283293,
      "grad_norm": 0.12555629014968872,
      "learning_rate": 2.7120148111887732e-05,
      "loss": 1.0534987449645996,
      "mean_token_accuracy": 0.6883429363369942,
      "num_tokens": 15146917.0,
      "step": 157
    },
    {
      "entropy": 1.0337524116039276,
      "epoch": 1.5230024213075062,
      "grad_norm": 0.12047123908996582,
      "learning_rate": 2.6855652326319063e-05,
      "loss": 1.0310931205749512,
      "mean_token_accuracy": 0.6945987418293953,
      "num_tokens": 15244027.0,
      "step": 158
    },
    {
      "entropy": 1.0539740845561028,
      "epoch": 1.5326876513317191,
      "grad_norm": 0.1266130805015564,
      "learning_rate": 2.659094751063666e-05,
      "loss": 1.0601729154586792,
      "mean_token_accuracy": 0.6876813471317291,
      "num_tokens": 15340804.0,
      "step": 159
    },
    {
      "entropy": 1.063611924648285,
      "epoch": 1.542372881355932,
      "grad_norm": 0.1284509301185608,
      "learning_rate": 2.6326063482536934e-05,
      "loss": 1.0666879415512085,
      "mean_token_accuracy": 0.6829489767551422,
      "num_tokens": 15437706.0,
      "step": 160
    },
    {
      "entropy": 1.0739876180887222,
      "epoch": 1.5520581113801453,
      "grad_norm": 0.128132164478302,
      "learning_rate": 2.606103007990371e-05,
      "loss": 1.0747418403625488,
      "mean_token_accuracy": 0.6830653175711632,
      "num_tokens": 15531879.0,
      "step": 161
    },
    {
      "entropy": 1.0638732314109802,
      "epoch": 1.5617433414043584,
      "grad_norm": 0.12959082424640656,
      "learning_rate": 2.579587715744712e-05,
      "loss": 1.0661025047302246,
      "mean_token_accuracy": 0.6848382279276848,
      "num_tokens": 15629721.0,
      "step": 162
    },
    {
      "entropy": 1.103687047958374,
      "epoch": 1.5714285714285714,
      "grad_norm": 0.1291181743144989,
      "learning_rate": 2.5530634583340592e-05,
      "loss": 1.109142541885376,
      "mean_token_accuracy": 0.6757278963923454,
      "num_tokens": 15727030.0,
      "step": 163
    },
    {
      "entropy": 1.0502248406410217,
      "epoch": 1.5811138014527844,
      "grad_norm": 0.12710939347743988,
      "learning_rate": 2.526533223585641e-05,
      "loss": 1.0568492412567139,
      "mean_token_accuracy": 0.6880576387047768,
      "num_tokens": 15823642.0,
      "step": 164
    },
    {
      "entropy": 1.0711619928479195,
      "epoch": 1.5907990314769975,
      "grad_norm": 0.12909084558486938,
      "learning_rate": 2.5e-05,
      "loss": 1.0641920566558838,
      "mean_token_accuracy": 0.6830467730760574,
      "num_tokens": 15920894.0,
      "step": 165
    },
    {
      "entropy": 1.0699793994426727,
      "epoch": 1.6004842615012107,
      "grad_norm": 0.1268342286348343,
      "learning_rate": 2.47346677641436e-05,
      "loss": 1.0631070137023926,
      "mean_token_accuracy": 0.6858616694808006,
      "num_tokens": 16018678.0,
      "step": 166
    },
    {
      "entropy": 1.064658597111702,
      "epoch": 1.6101694915254239,
      "grad_norm": 0.12978197634220123,
      "learning_rate": 2.446936541665941e-05,
      "loss": 1.0541661977767944,
      "mean_token_accuracy": 0.6868418604135513,
      "num_tokens": 16115100.0,
      "step": 167
    },
    {
      "entropy": 1.0778633058071136,
      "epoch": 1.6198547215496368,
      "grad_norm": 0.12875650823116302,
      "learning_rate": 2.4204122842552896e-05,
      "loss": 1.074495792388916,
      "mean_token_accuracy": 0.6796489804983139,
      "num_tokens": 16212185.0,
      "step": 168
    },
    {
      "entropy": 1.0641062408685684,
      "epoch": 1.6295399515738498,
      "grad_norm": 0.12701085209846497,
      "learning_rate": 2.39389699200963e-05,
      "loss": 1.0605846643447876,
      "mean_token_accuracy": 0.6864253729581833,
      "num_tokens": 16309607.0,
      "step": 169
    },
    {
      "entropy": 1.044494405388832,
      "epoch": 1.639225181598063,
      "grad_norm": 0.12824107706546783,
      "learning_rate": 2.3673936517463075e-05,
      "loss": 1.0462117195129395,
      "mean_token_accuracy": 0.6917339637875557,
      "num_tokens": 16403478.0,
      "step": 170
    },
    {
      "entropy": 1.0691992864012718,
      "epoch": 1.6489104116222761,
      "grad_norm": 0.1296340823173523,
      "learning_rate": 2.3409052489363342e-05,
      "loss": 1.0736641883850098,
      "mean_token_accuracy": 0.684188686311245,
      "num_tokens": 16499376.0,
      "step": 171
    },
    {
      "entropy": 1.0633979439735413,
      "epoch": 1.658595641646489,
      "grad_norm": 0.1293969452381134,
      "learning_rate": 2.3144347673680936e-05,
      "loss": 1.0691418647766113,
      "mean_token_accuracy": 0.6826125979423523,
      "num_tokens": 16597152.0,
      "step": 172
    },
    {
      "entropy": 1.0623700246214867,
      "epoch": 1.668280871670702,
      "grad_norm": 0.1276922971010208,
      "learning_rate": 2.287985188811228e-05,
      "loss": 1.0653769969940186,
      "mean_token_accuracy": 0.6850112453103065,
      "num_tokens": 16693593.0,
      "step": 173
    },
    {
      "entropy": 1.0490839704871178,
      "epoch": 1.6779661016949152,
      "grad_norm": 0.13035964965820312,
      "learning_rate": 2.261559492680755e-05,
      "loss": 1.052870750427246,
      "mean_token_accuracy": 0.6873865127563477,
      "num_tokens": 16791648.0,
      "step": 174
    },
    {
      "entropy": 1.0602750182151794,
      "epoch": 1.6876513317191284,
      "grad_norm": 0.13328254222869873,
      "learning_rate": 2.23516065570146e-05,
      "loss": 1.0508127212524414,
      "mean_token_accuracy": 0.6879312992095947,
      "num_tokens": 16887565.0,
      "step": 175
    },
    {
      "entropy": 1.0494377687573433,
      "epoch": 1.6973365617433414,
      "grad_norm": 0.13337542116641998,
      "learning_rate": 2.2087916515725737e-05,
      "loss": 1.0443730354309082,
      "mean_token_accuracy": 0.6917697265744209,
      "num_tokens": 16983866.0,
      "step": 176
    },
    {
      "entropy": 1.0237534046173096,
      "epoch": 1.7070217917675545,
      "grad_norm": 0.1268254965543747,
      "learning_rate": 2.182455450632803e-05,
      "loss": 1.020442008972168,
      "mean_token_accuracy": 0.6936786100268364,
      "num_tokens": 17080511.0,
      "step": 177
    },
    {
      "entropy": 1.062799796462059,
      "epoch": 1.7167070217917675,
      "grad_norm": 0.13072562217712402,
      "learning_rate": 2.1561550195257378e-05,
      "loss": 1.0620009899139404,
      "mean_token_accuracy": 0.6849319636821747,
      "num_tokens": 17177327.0,
      "step": 178
    },
    {
      "entropy": 1.023113876581192,
      "epoch": 1.7263922518159807,
      "grad_norm": 0.12836919724941254,
      "learning_rate": 2.1298933208656718e-05,
      "loss": 1.0207545757293701,
      "mean_token_accuracy": 0.6944043785333633,
      "num_tokens": 17272025.0,
      "step": 179
    },
    {
      "entropy": 1.0572519600391388,
      "epoch": 1.7360774818401938,
      "grad_norm": 0.13082647323608398,
      "learning_rate": 2.103673312903878e-05,
      "loss": 1.0634784698486328,
      "mean_token_accuracy": 0.6851047798991203,
      "num_tokens": 17369047.0,
      "step": 180
    },
    {
      "entropy": 1.0616972222924232,
      "epoch": 1.7457627118644068,
      "grad_norm": 0.13112697005271912,
      "learning_rate": 2.0774979491953777e-05,
      "loss": 1.0689513683319092,
      "mean_token_accuracy": 0.6840093061327934,
      "num_tokens": 17466402.0,
      "step": 181
    },
    {
      "entropy": 1.0349392369389534,
      "epoch": 1.7554479418886197,
      "grad_norm": 0.12918621301651,
      "learning_rate": 2.0513701782662366e-05,
      "loss": 1.0368030071258545,
      "mean_token_accuracy": 0.6913178861141205,
      "num_tokens": 17563946.0,
      "step": 182
    },
    {
      "entropy": 1.043944090604782,
      "epoch": 1.765133171912833,
      "grad_norm": 0.12798823416233063,
      "learning_rate": 2.025292943281429e-05,
      "loss": 1.0485461950302124,
      "mean_token_accuracy": 0.6885135471820831,
      "num_tokens": 17661994.0,
      "step": 183
    },
    {
      "entropy": 1.030234456062317,
      "epoch": 1.774818401937046,
      "grad_norm": 0.1295764446258545,
      "learning_rate": 1.9992691817133024e-05,
      "loss": 1.0321520566940308,
      "mean_token_accuracy": 0.6933487504720688,
      "num_tokens": 17759062.0,
      "step": 184
    },
    {
      "entropy": 1.0458166152238846,
      "epoch": 1.784503631961259,
      "grad_norm": 0.12951599061489105,
      "learning_rate": 1.973301825010685e-05,
      "loss": 1.0535242557525635,
      "mean_token_accuracy": 0.6879439502954483,
      "num_tokens": 17857030.0,
      "step": 185
    },
    {
      "entropy": 1.048134982585907,
      "epoch": 1.794188861985472,
      "grad_norm": 0.1298118531703949,
      "learning_rate": 1.947393798268676e-05,
      "loss": 1.0419211387634277,
      "mean_token_accuracy": 0.6897439807653427,
      "num_tokens": 17954718.0,
      "step": 186
    },
    {
      "entropy": 1.0557890236377716,
      "epoch": 1.8038740920096852,
      "grad_norm": 0.12609189748764038,
      "learning_rate": 1.9215480198991466e-05,
      "loss": 1.0479167699813843,
      "mean_token_accuracy": 0.687422975897789,
      "num_tokens": 18052307.0,
      "step": 187
    },
    {
      "entropy": 1.0465160012245178,
      "epoch": 1.8135593220338984,
      "grad_norm": 0.13359543681144714,
      "learning_rate": 1.895767401301994e-05,
      "loss": 1.0366110801696777,
      "mean_token_accuracy": 0.6899124383926392,
      "num_tokens": 18149449.0,
      "step": 188
    },
    {
      "entropy": 1.0489887967705727,
      "epoch": 1.8232445520581115,
      "grad_norm": 0.12966713309288025,
      "learning_rate": 1.8700548465371874e-05,
      "loss": 1.0362615585327148,
      "mean_token_accuracy": 0.6921318620443344,
      "num_tokens": 18247120.0,
      "step": 189
    },
    {
      "entropy": 1.0712166279554367,
      "epoch": 1.8329297820823245,
      "grad_norm": 0.12812435626983643,
      "learning_rate": 1.84441325199764e-05,
      "loss": 1.0638632774353027,
      "mean_token_accuracy": 0.6849293485283852,
      "num_tokens": 18344157.0,
      "step": 190
    },
    {
      "entropy": 1.0090139284729958,
      "epoch": 1.8426150121065374,
      "grad_norm": 0.12794391810894012,
      "learning_rate": 1.818845506082943e-05,
      "loss": 1.0082288980484009,
      "mean_token_accuracy": 0.6983787342905998,
      "num_tokens": 18441268.0,
      "step": 191
    },
    {
      "entropy": 1.0693828985095024,
      "epoch": 1.8523002421307506,
      "grad_norm": 0.13046976923942566,
      "learning_rate": 1.793354488874006e-05,
      "loss": 1.0742506980895996,
      "mean_token_accuracy": 0.682611919939518,
      "num_tokens": 18537788.0,
      "step": 192
    },
    {
      "entropy": 1.0317746996879578,
      "epoch": 1.8619854721549638,
      "grad_norm": 0.12896262109279633,
      "learning_rate": 1.7679430718086243e-05,
      "loss": 1.0313422679901123,
      "mean_token_accuracy": 0.6922419369220734,
      "num_tokens": 18634210.0,
      "step": 193
    },
    {
      "entropy": 1.0496741384267807,
      "epoch": 1.8716707021791767,
      "grad_norm": 0.1299387514591217,
      "learning_rate": 1.742614117358029e-05,
      "loss": 1.0520509481430054,
      "mean_token_accuracy": 0.687789648771286,
      "num_tokens": 18730773.0,
      "step": 194
    },
    {
      "entropy": 1.054203875362873,
      "epoch": 1.8813559322033897,
      "grad_norm": 0.1265571415424347,
      "learning_rate": 1.7173704787044446e-05,
      "loss": 1.060340404510498,
      "mean_token_accuracy": 0.6852798983454704,
      "num_tokens": 18828751.0,
      "step": 195
    },
    {
      "entropy": 1.0286369621753693,
      "epoch": 1.8910411622276029,
      "grad_norm": 0.12889890372753143,
      "learning_rate": 1.6922149994196878e-05,
      "loss": 1.0369319915771484,
      "mean_token_accuracy": 0.6926497370004654,
      "num_tokens": 18924854.0,
      "step": 196
    },
    {
      "entropy": 1.0430027171969414,
      "epoch": 1.900726392251816,
      "grad_norm": 0.1313585788011551,
      "learning_rate": 1.667150513144856e-05,
      "loss": 1.0536088943481445,
      "mean_token_accuracy": 0.6877090930938721,
      "num_tokens": 19022636.0,
      "step": 197
    },
    {
      "entropy": 1.0726237446069717,
      "epoch": 1.910411622276029,
      "grad_norm": 0.12799973785877228,
      "learning_rate": 1.6421798432711345e-05,
      "loss": 1.0721254348754883,
      "mean_token_accuracy": 0.6813898906111717,
      "num_tokens": 19119897.0,
      "step": 198
    },
    {
      "entropy": 1.0780348926782608,
      "epoch": 1.9200968523002422,
      "grad_norm": 0.1289709508419037,
      "learning_rate": 1.617305802621748e-05,
      "loss": 1.07821786403656,
      "mean_token_accuracy": 0.6835106313228607,
      "num_tokens": 19216975.0,
      "step": 199
    },
    {
      "entropy": 1.0513859540224075,
      "epoch": 1.9297820823244551,
      "grad_norm": 0.1273392140865326,
      "learning_rate": 1.5925311931351172e-05,
      "loss": 1.047814130783081,
      "mean_token_accuracy": 0.689472034573555,
      "num_tokens": 19314144.0,
      "step": 200
    },
    {
      "entropy": 1.0340349897742271,
      "epoch": 1.9394673123486683,
      "grad_norm": 0.12711293995380402,
      "learning_rate": 1.567858805549229e-05,
      "loss": 1.028080940246582,
      "mean_token_accuracy": 0.6955339759588242,
      "num_tokens": 19411168.0,
      "step": 201
    },
    {
      "entropy": 1.016765370965004,
      "epoch": 1.9491525423728815,
      "grad_norm": 0.12909752130508423,
      "learning_rate": 1.5432914190872757e-05,
      "loss": 1.0147216320037842,
      "mean_token_accuracy": 0.6978951469063759,
      "num_tokens": 19508050.0,
      "step": 202
    },
    {
      "entropy": 1.0415372103452682,
      "epoch": 1.9588377723970944,
      "grad_norm": 0.12862662971019745,
      "learning_rate": 1.5188318011445906e-05,
      "loss": 1.0378230810165405,
      "mean_token_accuracy": 0.691890224814415,
      "num_tokens": 19604931.0,
      "step": 203
    },
    {
      "entropy": 1.0212347209453583,
      "epoch": 1.9685230024213074,
      "grad_norm": 0.12980778515338898,
      "learning_rate": 1.4944827069769123e-05,
      "loss": 1.0138835906982422,
      "mean_token_accuracy": 0.6966540440917015,
      "num_tokens": 19700943.0,
      "step": 204
    },
    {
      "entropy": 1.018036775290966,
      "epoch": 1.9782082324455206,
      "grad_norm": 0.12775421142578125,
      "learning_rate": 1.4702468793900188e-05,
      "loss": 1.0206420421600342,
      "mean_token_accuracy": 0.6944147497415543,
      "num_tokens": 19798974.0,
      "step": 205
    },
    {
      "entropy": 1.0431870892643929,
      "epoch": 1.9878934624697338,
      "grad_norm": 0.12499149888753891,
      "learning_rate": 1.4461270484307643e-05,
      "loss": 1.0473066568374634,
      "mean_token_accuracy": 0.6883841976523399,
      "num_tokens": 19896619.0,
      "step": 206
    },
    {
      "entropy": 1.0695895552635193,
      "epoch": 1.9975786924939467,
      "grad_norm": 0.12734684348106384,
      "learning_rate": 1.4221259310795543e-05,
      "loss": 1.065197229385376,
      "mean_token_accuracy": 0.6831630617380142,
      "num_tokens": 19994525.0,
      "step": 207
    },
    {
      "entropy": 1.0691354870796204,
      "epoch": 2.0,
      "grad_norm": 0.260182648897171,
      "learning_rate": 1.3982462309442889e-05,
      "loss": 1.0698096752166748,
      "mean_token_accuracy": 0.6822833716869354,
      "num_tokens": 20018806.0,
      "step": 208
    }
  ],
  "logging_steps": 1,
  "max_steps": 312,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.3021108207332884e+18,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}