{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 116595,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.012865045670912132,
      "grad_norm": 7.8005571365356445,
      "learning_rate": 4.978558257215147e-05,
      "loss": 4.1174,
      "num_input_tokens_seen": 290696,
      "step": 500
    },
    {
      "epoch": 0.025730091341824263,
      "grad_norm": 8.382389068603516,
      "learning_rate": 4.957116514430293e-05,
      "loss": 3.9583,
      "num_input_tokens_seen": 580600,
      "step": 1000
    },
    {
      "epoch": 0.0385951370127364,
      "grad_norm": 7.8291778564453125,
      "learning_rate": 4.935674771645439e-05,
      "loss": 3.859,
      "num_input_tokens_seen": 874800,
      "step": 1500
    },
    {
      "epoch": 0.05146018268364853,
      "grad_norm": 7.659093379974365,
      "learning_rate": 4.914233028860586e-05,
      "loss": 3.8451,
      "num_input_tokens_seen": 1167104,
      "step": 2000
    },
    {
      "epoch": 0.06432522835456066,
      "grad_norm": 9.296012878417969,
      "learning_rate": 4.892791286075732e-05,
      "loss": 3.7631,
      "num_input_tokens_seen": 1460848,
      "step": 2500
    },
    {
      "epoch": 0.0771902740254728,
      "grad_norm": 7.6745710372924805,
      "learning_rate": 4.871349543290879e-05,
      "loss": 3.7213,
      "num_input_tokens_seen": 1756272,
      "step": 3000
    },
    {
      "epoch": 0.09005531969638492,
      "grad_norm": 7.863762378692627,
      "learning_rate": 4.8499078005060254e-05,
      "loss": 3.6711,
      "num_input_tokens_seen": 2051400,
      "step": 3500
    },
    {
      "epoch": 0.10292036536729705,
      "grad_norm": 9.283061027526855,
      "learning_rate": 4.828466057721171e-05,
      "loss": 3.6652,
      "num_input_tokens_seen": 2341024,
      "step": 4000
    },
    {
      "epoch": 0.11578541103820919,
      "grad_norm": 6.726513385772705,
      "learning_rate": 4.807024314936318e-05,
      "loss": 3.6139,
      "num_input_tokens_seen": 2630600,
      "step": 4500
    },
    {
      "epoch": 0.1286504567091213,
      "grad_norm": 8.104506492614746,
      "learning_rate": 4.7855825721514643e-05,
      "loss": 3.6039,
      "num_input_tokens_seen": 2919840,
      "step": 5000
    },
    {
      "epoch": 0.14151550238003344,
      "grad_norm": 7.184753894805908,
      "learning_rate": 4.764140829366611e-05,
      "loss": 3.5984,
      "num_input_tokens_seen": 3211736,
      "step": 5500
    },
    {
      "epoch": 0.1543805480509456,
      "grad_norm": 8.069321632385254,
      "learning_rate": 4.7426990865817574e-05,
      "loss": 3.5789,
      "num_input_tokens_seen": 3504520,
      "step": 6000
    },
    {
      "epoch": 0.1672455937218577,
      "grad_norm": 8.252145767211914,
      "learning_rate": 4.721257343796904e-05,
      "loss": 3.5405,
      "num_input_tokens_seen": 3797624,
      "step": 6500
    },
    {
      "epoch": 0.18011063939276983,
      "grad_norm": 7.835130214691162,
      "learning_rate": 4.6998156010120505e-05,
      "loss": 3.5301,
      "num_input_tokens_seen": 4094448,
      "step": 7000
    },
    {
      "epoch": 0.19297568506368198,
      "grad_norm": 7.696624279022217,
      "learning_rate": 4.678373858227197e-05,
      "loss": 3.5305,
      "num_input_tokens_seen": 4384360,
      "step": 7500
    },
    {
      "epoch": 0.2058407307345941,
      "grad_norm": 6.831740856170654,
      "learning_rate": 4.6569321154423436e-05,
      "loss": 3.5104,
      "num_input_tokens_seen": 4676344,
      "step": 8000
    },
    {
      "epoch": 0.21870577640550623,
      "grad_norm": 8.174054145812988,
      "learning_rate": 4.63549037265749e-05,
      "loss": 3.4725,
      "num_input_tokens_seen": 4964600,
      "step": 8500
    },
    {
      "epoch": 0.23157082207641838,
      "grad_norm": 8.504060745239258,
      "learning_rate": 4.614048629872637e-05,
      "loss": 3.4622,
      "num_input_tokens_seen": 5252472,
      "step": 9000
    },
    {
      "epoch": 0.2444358677473305,
      "grad_norm": 6.625699996948242,
      "learning_rate": 4.5926068870877826e-05,
      "loss": 3.4745,
      "num_input_tokens_seen": 5544856,
      "step": 9500
    },
    {
      "epoch": 0.2573009134182426,
      "grad_norm": 6.5118513107299805,
      "learning_rate": 4.571165144302929e-05,
      "loss": 3.4726,
      "num_input_tokens_seen": 5835512,
      "step": 10000
    },
    {
      "epoch": 0.27016595908915475,
      "grad_norm": 7.566400527954102,
      "learning_rate": 4.549723401518076e-05,
      "loss": 3.4469,
      "num_input_tokens_seen": 6123680,
      "step": 10500
    },
    {
      "epoch": 0.2830310047600669,
      "grad_norm": 7.794484615325928,
      "learning_rate": 4.528281658733222e-05,
      "loss": 3.452,
      "num_input_tokens_seen": 6413064,
      "step": 11000
    },
    {
      "epoch": 0.29589605043097905,
      "grad_norm": 7.974316120147705,
      "learning_rate": 4.506839915948369e-05,
      "loss": 3.4362,
      "num_input_tokens_seen": 6702440,
      "step": 11500
    },
    {
      "epoch": 0.3087610961018912,
      "grad_norm": 7.082245349884033,
      "learning_rate": 4.4853981731635146e-05,
      "loss": 3.4132,
      "num_input_tokens_seen": 6994512,
      "step": 12000
    },
    {
      "epoch": 0.3216261417728033,
      "grad_norm": 7.03355598449707,
      "learning_rate": 4.463956430378661e-05,
      "loss": 3.4169,
      "num_input_tokens_seen": 7288136,
      "step": 12500
    },
    {
      "epoch": 0.3344911874437154,
      "grad_norm": 6.516682147979736,
      "learning_rate": 4.442514687593808e-05,
      "loss": 3.416,
      "num_input_tokens_seen": 7578896,
      "step": 13000
    },
    {
      "epoch": 0.34735623311462754,
      "grad_norm": 6.383710861206055,
      "learning_rate": 4.421072944808954e-05,
      "loss": 3.4127,
      "num_input_tokens_seen": 7868256,
      "step": 13500
    },
    {
      "epoch": 0.36022127878553967,
      "grad_norm": 7.31823205947876,
      "learning_rate": 4.399631202024101e-05,
      "loss": 3.3886,
      "num_input_tokens_seen": 8160560,
      "step": 14000
    },
    {
      "epoch": 0.37308632445645185,
      "grad_norm": 7.401162624359131,
      "learning_rate": 4.378189459239247e-05,
      "loss": 3.4051,
      "num_input_tokens_seen": 8451784,
      "step": 14500
    },
    {
      "epoch": 0.38595137012736397,
      "grad_norm": 6.930325031280518,
      "learning_rate": 4.356747716454393e-05,
      "loss": 3.3821,
      "num_input_tokens_seen": 8743248,
      "step": 15000
    },
    {
      "epoch": 0.3988164157982761,
      "grad_norm": 6.667317867279053,
      "learning_rate": 4.33530597366954e-05,
      "loss": 3.3649,
      "num_input_tokens_seen": 9033824,
      "step": 15500
    },
    {
      "epoch": 0.4116814614691882,
      "grad_norm": 8.706613540649414,
      "learning_rate": 4.313864230884686e-05,
      "loss": 3.3882,
      "num_input_tokens_seen": 9326000,
      "step": 16000
    },
    {
      "epoch": 0.42454650714010034,
      "grad_norm": 6.553988933563232,
      "learning_rate": 4.292422488099833e-05,
      "loss": 3.3477,
      "num_input_tokens_seen": 9615136,
      "step": 16500
    },
    {
      "epoch": 0.43741155281101246,
      "grad_norm": 8.549520492553711,
      "learning_rate": 4.2709807453149794e-05,
      "loss": 3.3446,
      "num_input_tokens_seen": 9907896,
      "step": 17000
    },
    {
      "epoch": 0.4502765984819246,
      "grad_norm": 6.12117862701416,
      "learning_rate": 4.249539002530126e-05,
      "loss": 3.3399,
      "num_input_tokens_seen": 10199424,
      "step": 17500
    },
    {
      "epoch": 0.46314164415283676,
      "grad_norm": 6.233087062835693,
      "learning_rate": 4.2280972597452725e-05,
      "loss": 3.3456,
      "num_input_tokens_seen": 10489184,
      "step": 18000
    },
    {
      "epoch": 0.4760066898237489,
      "grad_norm": 6.620530605316162,
      "learning_rate": 4.206655516960419e-05,
      "loss": 3.3362,
      "num_input_tokens_seen": 10784400,
      "step": 18500
    },
    {
      "epoch": 0.488871735494661,
      "grad_norm": 6.852022647857666,
      "learning_rate": 4.1852137741755656e-05,
      "loss": 3.3319,
      "num_input_tokens_seen": 11076656,
      "step": 19000
    },
    {
      "epoch": 0.5017367811655732,
      "grad_norm": 7.368551254272461,
      "learning_rate": 4.1637720313907114e-05,
      "loss": 3.3271,
      "num_input_tokens_seen": 11371104,
      "step": 19500
    },
    {
      "epoch": 0.5146018268364853,
      "grad_norm": 7.825785160064697,
      "learning_rate": 4.142330288605858e-05,
      "loss": 3.2941,
      "num_input_tokens_seen": 11664608,
      "step": 20000
    },
    {
      "epoch": 0.5274668725073974,
      "grad_norm": 7.652072429656982,
      "learning_rate": 4.1208885458210045e-05,
      "loss": 3.3174,
      "num_input_tokens_seen": 11954656,
      "step": 20500
    },
    {
      "epoch": 0.5403319181783095,
      "grad_norm": 6.378400802612305,
      "learning_rate": 4.099446803036151e-05,
      "loss": 3.3438,
      "num_input_tokens_seen": 12245192,
      "step": 21000
    },
    {
      "epoch": 0.5531969638492217,
      "grad_norm": 6.9614434242248535,
      "learning_rate": 4.0780050602512976e-05,
      "loss": 3.2937,
      "num_input_tokens_seen": 12534328,
      "step": 21500
    },
    {
      "epoch": 0.5660620095201337,
      "grad_norm": 7.3693318367004395,
      "learning_rate": 4.056563317466444e-05,
      "loss": 3.2893,
      "num_input_tokens_seen": 12825376,
      "step": 22000
    },
    {
      "epoch": 0.5789270551910459,
      "grad_norm": 7.541472911834717,
      "learning_rate": 4.03512157468159e-05,
      "loss": 3.2868,
      "num_input_tokens_seen": 13114616,
      "step": 22500
    },
    {
      "epoch": 0.5917921008619581,
      "grad_norm": 7.473111629486084,
      "learning_rate": 4.0136798318967366e-05,
      "loss": 3.2725,
      "num_input_tokens_seen": 13405272,
      "step": 23000
    },
    {
      "epoch": 0.6046571465328702,
      "grad_norm": 6.550891399383545,
      "learning_rate": 3.992238089111883e-05,
      "loss": 3.3067,
      "num_input_tokens_seen": 13699360,
      "step": 23500
    },
    {
      "epoch": 0.6175221922037823,
      "grad_norm": 6.285316467285156,
      "learning_rate": 3.97079634632703e-05,
      "loss": 3.2861,
      "num_input_tokens_seen": 13990416,
      "step": 24000
    },
    {
      "epoch": 0.6303872378746944,
      "grad_norm": 7.406656742095947,
      "learning_rate": 3.949354603542176e-05,
      "loss": 3.2772,
      "num_input_tokens_seen": 14282464,
      "step": 24500
    },
    {
      "epoch": 0.6432522835456066,
      "grad_norm": 7.1049065589904785,
      "learning_rate": 3.927912860757322e-05,
      "loss": 3.257,
      "num_input_tokens_seen": 14575304,
      "step": 25000
    },
    {
      "epoch": 0.6561173292165187,
      "grad_norm": 5.734112739562988,
      "learning_rate": 3.9064711179724686e-05,
      "loss": 3.2361,
      "num_input_tokens_seen": 14863952,
      "step": 25500
    },
    {
      "epoch": 0.6689823748874308,
      "grad_norm": 6.281294822692871,
      "learning_rate": 3.885029375187615e-05,
      "loss": 3.2489,
      "num_input_tokens_seen": 15152296,
      "step": 26000
    },
    {
      "epoch": 0.681847420558343,
      "grad_norm": 8.989299774169922,
      "learning_rate": 3.863587632402762e-05,
      "loss": 3.2436,
      "num_input_tokens_seen": 15442488,
      "step": 26500
    },
    {
      "epoch": 0.6947124662292551,
      "grad_norm": 7.305727481842041,
      "learning_rate": 3.842145889617908e-05,
      "loss": 3.2413,
      "num_input_tokens_seen": 15733104,
      "step": 27000
    },
    {
      "epoch": 0.7075775119001673,
      "grad_norm": 5.735226631164551,
      "learning_rate": 3.820704146833055e-05,
      "loss": 3.2391,
      "num_input_tokens_seen": 16025536,
      "step": 27500
    },
    {
      "epoch": 0.7204425575710793,
      "grad_norm": 6.767416000366211,
      "learning_rate": 3.7992624040482014e-05,
      "loss": 3.2148,
      "num_input_tokens_seen": 16313080,
      "step": 28000
    },
    {
      "epoch": 0.7333076032419915,
      "grad_norm": 6.932997703552246,
      "learning_rate": 3.777820661263348e-05,
      "loss": 3.2245,
      "num_input_tokens_seen": 16600696,
      "step": 28500
    },
    {
      "epoch": 0.7461726489129037,
      "grad_norm": 7.29443359375,
      "learning_rate": 3.7563789184784945e-05,
      "loss": 3.2226,
      "num_input_tokens_seen": 16889416,
      "step": 29000
    },
    {
      "epoch": 0.7590376945838158,
      "grad_norm": 6.85066556930542,
      "learning_rate": 3.734937175693641e-05,
      "loss": 3.2464,
      "num_input_tokens_seen": 17181624,
      "step": 29500
    },
    {
      "epoch": 0.7719027402547279,
      "grad_norm": 7.167715072631836,
      "learning_rate": 3.713495432908787e-05,
      "loss": 3.2091,
      "num_input_tokens_seen": 17470040,
      "step": 30000
    },
    {
      "epoch": 0.78476778592564,
      "grad_norm": 6.19227933883667,
      "learning_rate": 3.6920536901239334e-05,
      "loss": 3.2147,
      "num_input_tokens_seen": 17762608,
      "step": 30500
    },
    {
      "epoch": 0.7976328315965522,
      "grad_norm": 6.353000164031982,
      "learning_rate": 3.67061194733908e-05,
      "loss": 3.2076,
      "num_input_tokens_seen": 18054152,
      "step": 31000
    },
    {
      "epoch": 0.8104978772674643,
      "grad_norm": 6.809099197387695,
      "learning_rate": 3.6491702045542265e-05,
      "loss": 3.2129,
      "num_input_tokens_seen": 18348664,
      "step": 31500
    },
    {
      "epoch": 0.8233629229383764,
      "grad_norm": 6.5169677734375,
      "learning_rate": 3.627728461769373e-05,
      "loss": 3.218,
      "num_input_tokens_seen": 18639328,
      "step": 32000
    },
    {
      "epoch": 0.8362279686092886,
      "grad_norm": 7.095029830932617,
      "learning_rate": 3.606286718984519e-05,
      "loss": 3.222,
      "num_input_tokens_seen": 18930776,
      "step": 32500
    },
    {
      "epoch": 0.8490930142802007,
      "grad_norm": 7.541679382324219,
      "learning_rate": 3.5848449761996655e-05,
      "loss": 3.2153,
      "num_input_tokens_seen": 19223592,
      "step": 33000
    },
    {
      "epoch": 0.8619580599511129,
      "grad_norm": 7.510098457336426,
      "learning_rate": 3.563403233414812e-05,
      "loss": 3.1922,
      "num_input_tokens_seen": 19517672,
      "step": 33500
    },
    {
      "epoch": 0.8748231056220249,
      "grad_norm": 6.9184956550598145,
      "learning_rate": 3.5419614906299586e-05,
      "loss": 3.1663,
      "num_input_tokens_seen": 19806520,
      "step": 34000
    },
    {
      "epoch": 0.8876881512929371,
      "grad_norm": 6.282911777496338,
      "learning_rate": 3.520519747845105e-05,
      "loss": 3.1922,
      "num_input_tokens_seen": 20097768,
      "step": 34500
    },
    {
      "epoch": 0.9005531969638492,
      "grad_norm": 7.266313552856445,
      "learning_rate": 3.4990780050602516e-05,
      "loss": 3.1741,
      "num_input_tokens_seen": 20384448,
      "step": 35000
    },
    {
      "epoch": 0.9134182426347613,
      "grad_norm": 5.5496110916137695,
      "learning_rate": 3.4776362622753975e-05,
      "loss": 3.193,
      "num_input_tokens_seen": 20672584,
      "step": 35500
    },
    {
      "epoch": 0.9262832883056735,
      "grad_norm": 7.990514278411865,
      "learning_rate": 3.456194519490544e-05,
      "loss": 3.1539,
      "num_input_tokens_seen": 20965184,
      "step": 36000
    },
    {
      "epoch": 0.9391483339765856,
      "grad_norm": 6.803572177886963,
      "learning_rate": 3.4347527767056906e-05,
      "loss": 3.16,
      "num_input_tokens_seen": 21256728,
      "step": 36500
    },
    {
      "epoch": 0.9520133796474978,
      "grad_norm": 7.202699661254883,
      "learning_rate": 3.413311033920837e-05,
      "loss": 3.1587,
      "num_input_tokens_seen": 21553464,
      "step": 37000
    },
    {
      "epoch": 0.9648784253184098,
      "grad_norm": 7.018800735473633,
      "learning_rate": 3.391869291135984e-05,
      "loss": 3.1395,
      "num_input_tokens_seen": 21847744,
      "step": 37500
    },
    {
      "epoch": 0.977743470989322,
      "grad_norm": 9.715666770935059,
      "learning_rate": 3.37042754835113e-05,
      "loss": 3.1493,
      "num_input_tokens_seen": 22142256,
      "step": 38000
    },
    {
      "epoch": 0.9906085166602342,
      "grad_norm": 6.174098491668701,
      "learning_rate": 3.348985805566277e-05,
      "loss": 3.1744,
      "num_input_tokens_seen": 22433344,
      "step": 38500
    },
    {
      "epoch": 1.0034735623311464,
      "grad_norm": 6.0936503410339355,
      "learning_rate": 3.327544062781423e-05,
      "loss": 3.1358,
      "num_input_tokens_seen": 22726152,
      "step": 39000
    },
    {
      "epoch": 1.0163386080020584,
      "grad_norm": 6.897722244262695,
      "learning_rate": 3.30610231999657e-05,
      "loss": 3.0054,
      "num_input_tokens_seen": 23018744,
      "step": 39500
    },
    {
      "epoch": 1.0292036536729705,
      "grad_norm": 6.8439412117004395,
      "learning_rate": 3.2846605772117164e-05,
      "loss": 2.9973,
      "num_input_tokens_seen": 23307936,
      "step": 40000
    },
    {
      "epoch": 1.0420686993438826,
      "grad_norm": 6.564641952514648,
      "learning_rate": 3.263218834426862e-05,
      "loss": 2.9964,
      "num_input_tokens_seen": 23604824,
      "step": 40500
    },
    {
      "epoch": 1.0549337450147949,
      "grad_norm": 6.155550479888916,
      "learning_rate": 3.241777091642009e-05,
      "loss": 3.0128,
      "num_input_tokens_seen": 23895032,
      "step": 41000
    },
    {
      "epoch": 1.067798790685707,
      "grad_norm": 6.811328887939453,
      "learning_rate": 3.2203353488571554e-05,
      "loss": 3.0221,
      "num_input_tokens_seen": 24185928,
      "step": 41500
    },
    {
      "epoch": 1.080663836356619,
      "grad_norm": 6.012444019317627,
      "learning_rate": 3.198893606072302e-05,
      "loss": 3.0283,
      "num_input_tokens_seen": 24474600,
      "step": 42000
    },
    {
      "epoch": 1.0935288820275313,
      "grad_norm": 6.834691524505615,
      "learning_rate": 3.1774518632874485e-05,
      "loss": 3.001,
      "num_input_tokens_seen": 24765648,
      "step": 42500
    },
    {
      "epoch": 1.1063939276984434,
      "grad_norm": 7.075684070587158,
      "learning_rate": 3.1560101205025943e-05,
      "loss": 2.9998,
      "num_input_tokens_seen": 25057256,
      "step": 43000
    },
    {
      "epoch": 1.1192589733693554,
      "grad_norm": 6.3663716316223145,
      "learning_rate": 3.134568377717741e-05,
      "loss": 3.0107,
      "num_input_tokens_seen": 25348904,
      "step": 43500
    },
    {
      "epoch": 1.1321240190402677,
      "grad_norm": 7.406547546386719,
      "learning_rate": 3.1131266349328874e-05,
      "loss": 3.0029,
      "num_input_tokens_seen": 25640496,
      "step": 44000
    },
    {
      "epoch": 1.1449890647111798,
      "grad_norm": 7.625437259674072,
      "learning_rate": 3.091684892148034e-05,
      "loss": 3.0045,
      "num_input_tokens_seen": 25934664,
      "step": 44500
    },
    {
      "epoch": 1.1578541103820919,
      "grad_norm": 6.509032726287842,
      "learning_rate": 3.0702431493631805e-05,
      "loss": 3.0089,
      "num_input_tokens_seen": 26225392,
      "step": 45000
    },
    {
      "epoch": 1.170719156053004,
      "grad_norm": 5.867618560791016,
      "learning_rate": 3.0488014065783267e-05,
      "loss": 2.9789,
      "num_input_tokens_seen": 26516360,
      "step": 45500
    },
    {
      "epoch": 1.1835842017239162,
      "grad_norm": 6.742827892303467,
      "learning_rate": 3.0273596637934733e-05,
      "loss": 2.9921,
      "num_input_tokens_seen": 26811464,
      "step": 46000
    },
    {
      "epoch": 1.1964492473948283,
      "grad_norm": 6.05054235458374,
      "learning_rate": 3.0059179210086198e-05,
      "loss": 3.0073,
      "num_input_tokens_seen": 27102928,
      "step": 46500
    },
    {
      "epoch": 1.2093142930657403,
      "grad_norm": 6.148196220397949,
      "learning_rate": 2.9844761782237664e-05,
      "loss": 3.0191,
      "num_input_tokens_seen": 27389024,
      "step": 47000
    },
    {
      "epoch": 1.2221793387366526,
      "grad_norm": 6.119533538818359,
      "learning_rate": 2.963034435438913e-05,
      "loss": 2.9884,
      "num_input_tokens_seen": 27680744,
      "step": 47500
    },
    {
      "epoch": 1.2350443844075647,
      "grad_norm": 5.91504430770874,
      "learning_rate": 2.9415926926540588e-05,
      "loss": 2.97,
      "num_input_tokens_seen": 27972728,
      "step": 48000
    },
    {
      "epoch": 1.2479094300784768,
      "grad_norm": 7.024054050445557,
      "learning_rate": 2.9201509498692053e-05,
      "loss": 3.0079,
      "num_input_tokens_seen": 28267000,
      "step": 48500
    },
    {
      "epoch": 1.2607744757493888,
      "grad_norm": 6.829626560211182,
      "learning_rate": 2.898709207084352e-05,
      "loss": 2.9945,
      "num_input_tokens_seen": 28557872,
      "step": 49000
    },
    {
      "epoch": 1.2736395214203011,
      "grad_norm": 6.172780990600586,
      "learning_rate": 2.8772674642994984e-05,
      "loss": 3.0175,
      "num_input_tokens_seen": 28850024,
      "step": 49500
    },
    {
      "epoch": 1.2865045670912132,
      "grad_norm": 6.888333320617676,
      "learning_rate": 2.855825721514645e-05,
      "loss": 2.9914,
      "num_input_tokens_seen": 29142640,
      "step": 50000
    },
    {
      "epoch": 1.2993696127621253,
      "grad_norm": 7.5561394691467285,
      "learning_rate": 2.8343839787297915e-05,
      "loss": 3.0081,
      "num_input_tokens_seen": 29436480,
      "step": 50500
    },
    {
      "epoch": 1.3122346584330375,
      "grad_norm": 6.160106182098389,
      "learning_rate": 2.8129422359449377e-05,
      "loss": 2.9796,
      "num_input_tokens_seen": 29728032,
      "step": 51000
    },
    {
      "epoch": 1.3250997041039496,
      "grad_norm": 6.359186172485352,
      "learning_rate": 2.7915004931600843e-05,
      "loss": 2.9663,
      "num_input_tokens_seen": 30015064,
      "step": 51500
    },
    {
      "epoch": 1.3379647497748617,
      "grad_norm": 5.363637447357178,
      "learning_rate": 2.7700587503752308e-05,
      "loss": 3.0251,
      "num_input_tokens_seen": 30307168,
      "step": 52000
    },
    {
      "epoch": 1.3508297954457738,
      "grad_norm": 6.702362537384033,
      "learning_rate": 2.7486170075903773e-05,
      "loss": 3.0027,
      "num_input_tokens_seen": 30597808,
      "step": 52500
    },
    {
      "epoch": 1.363694841116686,
      "grad_norm": 6.282807350158691,
      "learning_rate": 2.727175264805524e-05,
      "loss": 2.9954,
      "num_input_tokens_seen": 30888840,
      "step": 53000
    },
    {
      "epoch": 1.376559886787598,
      "grad_norm": 6.255498886108398,
      "learning_rate": 2.7057335220206698e-05,
      "loss": 3.0016,
      "num_input_tokens_seen": 31178408,
      "step": 53500
    },
    {
      "epoch": 1.3894249324585102,
      "grad_norm": 5.822259426116943,
      "learning_rate": 2.6842917792358163e-05,
      "loss": 2.9864,
      "num_input_tokens_seen": 31466920,
      "step": 54000
    },
    {
      "epoch": 1.4022899781294225,
      "grad_norm": 6.1176652908325195,
      "learning_rate": 2.662850036450963e-05,
      "loss": 2.9769,
      "num_input_tokens_seen": 31757920,
      "step": 54500
    },
    {
      "epoch": 1.4151550238003345,
      "grad_norm": 7.492593765258789,
      "learning_rate": 2.6414082936661094e-05,
      "loss": 2.9921,
      "num_input_tokens_seen": 32049864,
      "step": 55000
    },
    {
      "epoch": 1.4280200694712466,
      "grad_norm": 6.679434299468994,
      "learning_rate": 2.619966550881256e-05,
      "loss": 2.9667,
      "num_input_tokens_seen": 32342072,
      "step": 55500
    },
    {
      "epoch": 1.4408851151421587,
      "grad_norm": 6.818129539489746,
      "learning_rate": 2.5985248080964018e-05,
      "loss": 2.9678,
      "num_input_tokens_seen": 32632672,
      "step": 56000
    },
    {
      "epoch": 1.453750160813071,
      "grad_norm": 5.836079120635986,
      "learning_rate": 2.5770830653115487e-05,
      "loss": 2.9515,
      "num_input_tokens_seen": 32927752,
      "step": 56500
    },
    {
      "epoch": 1.466615206483983,
      "grad_norm": 7.332676410675049,
      "learning_rate": 2.5556413225266952e-05,
      "loss": 2.9725,
      "num_input_tokens_seen": 33220536,
      "step": 57000
    },
    {
      "epoch": 1.479480252154895,
      "grad_norm": 6.780636787414551,
      "learning_rate": 2.5341995797418418e-05,
      "loss": 2.9507,
      "num_input_tokens_seen": 33511536,
      "step": 57500
    },
    {
      "epoch": 1.4923452978258074,
      "grad_norm": 7.856017589569092,
      "learning_rate": 2.5127578369569883e-05,
      "loss": 2.9431,
      "num_input_tokens_seen": 33801088,
      "step": 58000
    },
    {
      "epoch": 1.5052103434967194,
      "grad_norm": 6.466399192810059,
      "learning_rate": 2.4913160941721345e-05,
      "loss": 2.9826,
      "num_input_tokens_seen": 34091576,
      "step": 58500
    },
    {
      "epoch": 1.5180753891676315,
      "grad_norm": 7.123006343841553,
      "learning_rate": 2.4698743513872807e-05,
      "loss": 2.9551,
      "num_input_tokens_seen": 34385640,
      "step": 59000
    },
    {
      "epoch": 1.5309404348385436,
      "grad_norm": 6.76849889755249,
      "learning_rate": 2.4484326086024273e-05,
      "loss": 2.9577,
      "num_input_tokens_seen": 34677472,
      "step": 59500
    },
    {
      "epoch": 1.5438054805094557,
      "grad_norm": 6.654425144195557,
      "learning_rate": 2.426990865817574e-05,
      "loss": 2.9669,
      "num_input_tokens_seen": 34967440,
      "step": 60000
    },
    {
      "epoch": 1.556670526180368,
      "grad_norm": 5.956307411193848,
      "learning_rate": 2.40554912303272e-05,
      "loss": 2.9519,
      "num_input_tokens_seen": 35256688,
      "step": 60500
    },
    {
      "epoch": 1.5695355718512802,
      "grad_norm": 6.195157527923584,
      "learning_rate": 2.3841073802478666e-05,
      "loss": 2.9379,
      "num_input_tokens_seen": 35545424,
      "step": 61000
    },
    {
      "epoch": 1.5824006175221923,
      "grad_norm": 6.981558322906494,
      "learning_rate": 2.362665637463013e-05,
      "loss": 2.9557,
      "num_input_tokens_seen": 35834248,
      "step": 61500
    },
    {
      "epoch": 1.5952656631931044,
      "grad_norm": 6.045523166656494,
      "learning_rate": 2.3412238946781593e-05,
      "loss": 2.9469,
      "num_input_tokens_seen": 36122472,
      "step": 62000
    },
    {
      "epoch": 1.6081307088640164,
      "grad_norm": 7.8723297119140625,
      "learning_rate": 2.319782151893306e-05,
      "loss": 2.9537,
      "num_input_tokens_seen": 36413912,
      "step": 62500
    },
    {
      "epoch": 1.6209957545349285,
      "grad_norm": 7.2311296463012695,
      "learning_rate": 2.2983404091084524e-05,
      "loss": 2.9507,
      "num_input_tokens_seen": 36706248,
      "step": 63000
    },
    {
      "epoch": 1.6338608002058406,
      "grad_norm": 6.698265552520752,
      "learning_rate": 2.276898666323599e-05,
      "loss": 2.9525,
      "num_input_tokens_seen": 36995928,
      "step": 63500
    },
    {
      "epoch": 1.6467258458767529,
      "grad_norm": 5.823113918304443,
      "learning_rate": 2.2554569235387455e-05,
      "loss": 2.9689,
      "num_input_tokens_seen": 37285032,
      "step": 64000
    },
    {
      "epoch": 1.6595908915476651,
      "grad_norm": 6.0338134765625,
      "learning_rate": 2.2340151807538917e-05,
      "loss": 2.9312,
      "num_input_tokens_seen": 37575280,
      "step": 64500
    },
    {
      "epoch": 1.6724559372185772,
      "grad_norm": 7.287151336669922,
      "learning_rate": 2.2125734379690383e-05,
      "loss": 2.9272,
      "num_input_tokens_seen": 37866864,
      "step": 65000
    },
    {
      "epoch": 1.6853209828894893,
      "grad_norm": 6.978508949279785,
      "learning_rate": 2.1911316951841845e-05,
      "loss": 2.9399,
      "num_input_tokens_seen": 38160464,
      "step": 65500
    },
    {
      "epoch": 1.6981860285604014,
      "grad_norm": 6.498319625854492,
      "learning_rate": 2.169689952399331e-05,
      "loss": 2.9323,
      "num_input_tokens_seen": 38450864,
      "step": 66000
    },
    {
      "epoch": 1.7110510742313134,
      "grad_norm": 6.914408206939697,
      "learning_rate": 2.1482482096144776e-05,
      "loss": 2.9378,
      "num_input_tokens_seen": 38741808,
      "step": 66500
    },
    {
      "epoch": 1.7239161199022257,
      "grad_norm": 7.851312160491943,
      "learning_rate": 2.1268064668296238e-05,
      "loss": 2.9346,
      "num_input_tokens_seen": 39034008,
      "step": 67000
    },
    {
      "epoch": 1.7367811655731378,
      "grad_norm": 7.065863132476807,
      "learning_rate": 2.1053647240447703e-05,
      "loss": 2.9458,
      "num_input_tokens_seen": 39326632,
      "step": 67500
    },
    {
      "epoch": 1.74964621124405,
      "grad_norm": 7.006825923919678,
      "learning_rate": 2.083922981259917e-05,
      "loss": 2.9241,
      "num_input_tokens_seen": 39616152,
      "step": 68000
    },
    {
      "epoch": 1.7625112569149621,
      "grad_norm": 6.096765041351318,
      "learning_rate": 2.0624812384750634e-05,
      "loss": 2.9341,
      "num_input_tokens_seen": 39909336,
      "step": 68500
    },
    {
      "epoch": 1.7753763025858742,
      "grad_norm": 6.418511390686035,
      "learning_rate": 2.04103949569021e-05,
      "loss": 2.9602,
      "num_input_tokens_seen": 40201392,
      "step": 69000
    },
    {
      "epoch": 1.7882413482567863,
      "grad_norm": 8.561145782470703,
      "learning_rate": 2.019597752905356e-05,
      "loss": 2.8944,
      "num_input_tokens_seen": 40494488,
      "step": 69500
    },
    {
      "epoch": 1.8011063939276983,
      "grad_norm": 6.326451778411865,
      "learning_rate": 1.9981560101205027e-05,
      "loss": 2.9622,
      "num_input_tokens_seen": 40783600,
      "step": 70000
    },
    {
      "epoch": 1.8139714395986106,
      "grad_norm": 7.819330215454102,
      "learning_rate": 1.9767142673356493e-05,
      "loss": 2.9218,
      "num_input_tokens_seen": 41076096,
      "step": 70500
    },
    {
      "epoch": 1.8268364852695227,
      "grad_norm": 7.052221298217773,
      "learning_rate": 1.9552725245507955e-05,
      "loss": 2.8852,
      "num_input_tokens_seen": 41371488,
      "step": 71000
    },
    {
      "epoch": 1.839701530940435,
      "grad_norm": 6.763613700866699,
      "learning_rate": 1.933830781765942e-05,
      "loss": 2.9176,
      "num_input_tokens_seen": 41660952,
      "step": 71500
    },
    {
      "epoch": 1.852566576611347,
      "grad_norm": 6.716059684753418,
      "learning_rate": 1.9123890389810882e-05,
      "loss": 2.9128,
      "num_input_tokens_seen": 41952344,
      "step": 72000
    },
    {
      "epoch": 1.8654316222822591,
      "grad_norm": 5.6197381019592285,
      "learning_rate": 1.8909472961962348e-05,
      "loss": 2.9254,
      "num_input_tokens_seen": 42240960,
      "step": 72500
    },
    {
      "epoch": 1.8782966679531712,
      "grad_norm": 7.378303050994873,
      "learning_rate": 1.8695055534113813e-05,
      "loss": 2.9128,
      "num_input_tokens_seen": 42531912,
      "step": 73000
    },
    {
      "epoch": 1.8911617136240833,
      "grad_norm": 6.518546104431152,
      "learning_rate": 1.848063810626528e-05,
      "loss": 2.9244,
      "num_input_tokens_seen": 42824144,
      "step": 73500
    },
    {
      "epoch": 1.9040267592949955,
      "grad_norm": 6.920441627502441,
      "learning_rate": 1.8266220678416744e-05,
      "loss": 2.9321,
      "num_input_tokens_seen": 43116312,
      "step": 74000
    },
    {
      "epoch": 1.9168918049659076,
      "grad_norm": 8.135106086730957,
      "learning_rate": 1.805180325056821e-05,
      "loss": 2.9242,
      "num_input_tokens_seen": 43408792,
      "step": 74500
    },
    {
      "epoch": 1.92975685063682,
      "grad_norm": 6.130031585693359,
      "learning_rate": 1.783738582271967e-05,
      "loss": 2.9233,
      "num_input_tokens_seen": 43702768,
      "step": 75000
    },
    {
      "epoch": 1.942621896307732,
      "grad_norm": 6.980623245239258,
      "learning_rate": 1.7622968394871137e-05,
      "loss": 2.9175,
      "num_input_tokens_seen": 43991200,
      "step": 75500
    },
    {
      "epoch": 1.955486941978644,
      "grad_norm": 6.8675456047058105,
      "learning_rate": 1.74085509670226e-05,
      "loss": 2.9169,
      "num_input_tokens_seen": 44281424,
      "step": 76000
    },
    {
      "epoch": 1.968351987649556,
      "grad_norm": 6.631172180175781,
      "learning_rate": 1.7194133539174064e-05,
      "loss": 2.9198,
      "num_input_tokens_seen": 44569856,
      "step": 76500
    },
    {
      "epoch": 1.9812170333204682,
      "grad_norm": 6.995504379272461,
      "learning_rate": 1.697971611132553e-05,
      "loss": 2.9182,
      "num_input_tokens_seen": 44862968,
      "step": 77000
    },
    {
      "epoch": 1.9940820789913805,
      "grad_norm": 7.234148979187012,
      "learning_rate": 1.6765298683476992e-05,
      "loss": 2.9226,
      "num_input_tokens_seen": 45153272,
      "step": 77500
    },
    {
      "epoch": 2.0069471246622927,
      "grad_norm": 6.20206356048584,
      "learning_rate": 1.6550881255628457e-05,
      "loss": 2.8863,
      "num_input_tokens_seen": 45443152,
      "step": 78000
    },
    {
      "epoch": 2.019812170333205,
      "grad_norm": 6.416888236999512,
      "learning_rate": 1.6336463827779923e-05,
      "loss": 2.7699,
      "num_input_tokens_seen": 45732040,
      "step": 78500
    },
    {
      "epoch": 2.032677216004117,
      "grad_norm": 7.381237506866455,
      "learning_rate": 1.612204639993139e-05,
      "loss": 2.7847,
      "num_input_tokens_seen": 46022712,
      "step": 79000
    },
    {
      "epoch": 2.045542261675029,
      "grad_norm": 6.199607849121094,
      "learning_rate": 1.5907628972082854e-05,
      "loss": 2.7676,
      "num_input_tokens_seen": 46315064,
      "step": 79500
    },
    {
      "epoch": 2.058407307345941,
      "grad_norm": 6.7973456382751465,
      "learning_rate": 1.5693211544234316e-05,
      "loss": 2.7909,
      "num_input_tokens_seen": 46608096,
      "step": 80000
    },
    {
      "epoch": 2.071272353016853,
      "grad_norm": 5.843527317047119,
      "learning_rate": 1.547879411638578e-05,
      "loss": 2.7965,
      "num_input_tokens_seen": 46896800,
      "step": 80500
    },
    {
      "epoch": 2.084137398687765,
      "grad_norm": 6.95015811920166,
      "learning_rate": 1.5264376688537247e-05,
      "loss": 2.7969,
      "num_input_tokens_seen": 47190120,
      "step": 81000
    },
    {
      "epoch": 2.0970024443586777,
      "grad_norm": 7.737337589263916,
      "learning_rate": 1.5049959260688709e-05,
      "loss": 2.7876,
      "num_input_tokens_seen": 47483520,
      "step": 81500
    },
    {
      "epoch": 2.1098674900295897,
      "grad_norm": 7.859601974487305,
      "learning_rate": 1.4835541832840174e-05,
      "loss": 2.7888,
      "num_input_tokens_seen": 47769600,
      "step": 82000
    },
    {
      "epoch": 2.122732535700502,
      "grad_norm": 5.804116725921631,
      "learning_rate": 1.4621124404991638e-05,
      "loss": 2.7868,
      "num_input_tokens_seen": 48064320,
      "step": 82500
    },
    {
      "epoch": 2.135597581371414,
      "grad_norm": 6.726932048797607,
      "learning_rate": 1.4406706977143104e-05,
      "loss": 2.8175,
      "num_input_tokens_seen": 48355560,
      "step": 83000
    },
    {
      "epoch": 2.148462627042326,
      "grad_norm": 6.097367763519287,
      "learning_rate": 1.4192289549294569e-05,
      "loss": 2.7876,
      "num_input_tokens_seen": 48644608,
      "step": 83500
    },
    {
      "epoch": 2.161327672713238,
      "grad_norm": 8.123278617858887,
      "learning_rate": 1.3977872121446031e-05,
      "loss": 2.7795,
      "num_input_tokens_seen": 48934720,
      "step": 84000
    },
    {
      "epoch": 2.17419271838415,
      "grad_norm": 6.270246505737305,
      "learning_rate": 1.3763454693597497e-05,
      "loss": 2.7723,
      "num_input_tokens_seen": 49227264,
      "step": 84500
    },
    {
      "epoch": 2.1870577640550626,
      "grad_norm": 7.330353260040283,
      "learning_rate": 1.354903726574896e-05,
      "loss": 2.7704,
      "num_input_tokens_seen": 49518072,
      "step": 85000
    },
    {
      "epoch": 2.1999228097259746,
      "grad_norm": 7.461730003356934,
      "learning_rate": 1.3334619837900426e-05,
      "loss": 2.7744,
      "num_input_tokens_seen": 49809432,
      "step": 85500
    },
    {
      "epoch": 2.2127878553968867,
      "grad_norm": 5.806447505950928,
      "learning_rate": 1.3120202410051891e-05,
      "loss": 2.7776,
      "num_input_tokens_seen": 50093416,
      "step": 86000
    },
    {
      "epoch": 2.225652901067799,
      "grad_norm": 5.978988170623779,
      "learning_rate": 1.2905784982203353e-05,
      "loss": 2.7773,
      "num_input_tokens_seen": 50386496,
      "step": 86500
    },
    {
      "epoch": 2.238517946738711,
      "grad_norm": 6.581029891967773,
      "learning_rate": 1.2691367554354819e-05,
      "loss": 2.8044,
      "num_input_tokens_seen": 50680456,
      "step": 87000
    },
    {
      "epoch": 2.251382992409623,
      "grad_norm": 6.622259616851807,
      "learning_rate": 1.2476950126506282e-05,
      "loss": 2.7959,
      "num_input_tokens_seen": 50973488,
      "step": 87500
    },
    {
      "epoch": 2.2642480380805354,
      "grad_norm": 9.183111190795898,
      "learning_rate": 1.2262532698657748e-05,
      "loss": 2.7865,
      "num_input_tokens_seen": 51266744,
      "step": 88000
    },
    {
      "epoch": 2.2771130837514475,
      "grad_norm": 6.282793045043945,
      "learning_rate": 1.2048115270809212e-05,
      "loss": 2.7768,
      "num_input_tokens_seen": 51556696,
      "step": 88500
    },
    {
      "epoch": 2.2899781294223596,
      "grad_norm": 6.839522838592529,
      "learning_rate": 1.1833697842960677e-05,
      "loss": 2.7936,
      "num_input_tokens_seen": 51848936,
      "step": 89000
    },
    {
      "epoch": 2.3028431750932716,
      "grad_norm": 5.936497211456299,
      "learning_rate": 1.1619280415112141e-05,
      "loss": 2.807,
      "num_input_tokens_seen": 52140680,
      "step": 89500
    },
    {
      "epoch": 2.3157082207641837,
      "grad_norm": 6.789340972900391,
      "learning_rate": 1.1404862987263605e-05,
      "loss": 2.7843,
      "num_input_tokens_seen": 52429888,
      "step": 90000
    },
    {
      "epoch": 2.3285732664350958,
      "grad_norm": 6.576247692108154,
      "learning_rate": 1.1190445559415068e-05,
      "loss": 2.7724,
      "num_input_tokens_seen": 52721144,
      "step": 90500
    },
    {
      "epoch": 2.341438312106008,
      "grad_norm": 6.2103166580200195,
      "learning_rate": 1.0976028131566536e-05,
      "loss": 2.7929,
      "num_input_tokens_seen": 53013648,
      "step": 91000
    },
    {
      "epoch": 2.35430335777692,
      "grad_norm": 7.041256904602051,
      "learning_rate": 1.0761610703718e-05,
      "loss": 2.7657,
      "num_input_tokens_seen": 53302912,
      "step": 91500
    },
    {
      "epoch": 2.3671684034478324,
      "grad_norm": 6.339066028594971,
      "learning_rate": 1.0547193275869463e-05,
      "loss": 2.7841,
      "num_input_tokens_seen": 53596464,
      "step": 92000
    },
    {
      "epoch": 2.3800334491187445,
      "grad_norm": 7.0336833000183105,
      "learning_rate": 1.0332775848020927e-05,
      "loss": 2.7859,
      "num_input_tokens_seen": 53886496,
      "step": 92500
    },
    {
      "epoch": 2.3928984947896565,
      "grad_norm": 5.826415538787842,
      "learning_rate": 1.0118358420172392e-05,
      "loss": 2.8007,
      "num_input_tokens_seen": 54176264,
      "step": 93000
    },
    {
      "epoch": 2.4057635404605686,
      "grad_norm": 6.1079487800598145,
      "learning_rate": 9.903940992323856e-06,
      "loss": 2.7775,
      "num_input_tokens_seen": 54466856,
      "step": 93500
    },
    {
      "epoch": 2.4186285861314807,
      "grad_norm": 5.203562259674072,
      "learning_rate": 9.689523564475322e-06,
      "loss": 2.7554,
      "num_input_tokens_seen": 54759176,
      "step": 94000
    },
    {
      "epoch": 2.4314936318023928,
      "grad_norm": 6.122252941131592,
      "learning_rate": 9.475106136626785e-06,
      "loss": 2.7709,
      "num_input_tokens_seen": 55051272,
      "step": 94500
    },
    {
      "epoch": 2.4443586774733053,
      "grad_norm": 6.208186149597168,
      "learning_rate": 9.260688708778249e-06,
      "loss": 2.7717,
      "num_input_tokens_seen": 55344160,
      "step": 95000
    },
    {
      "epoch": 2.4572237231442173,
      "grad_norm": 6.000925064086914,
      "learning_rate": 9.046271280929714e-06,
      "loss": 2.7627,
      "num_input_tokens_seen": 55636256,
      "step": 95500
    },
    {
      "epoch": 2.4700887688151294,
      "grad_norm": 7.118528842926025,
      "learning_rate": 8.831853853081178e-06,
      "loss": 2.7891,
      "num_input_tokens_seen": 55926232,
      "step": 96000
    },
    {
      "epoch": 2.4829538144860415,
      "grad_norm": 6.965567111968994,
      "learning_rate": 8.617436425232644e-06,
      "loss": 2.7854,
      "num_input_tokens_seen": 56217216,
      "step": 96500
    },
    {
      "epoch": 2.4958188601569535,
      "grad_norm": 6.334046840667725,
      "learning_rate": 8.403018997384107e-06,
      "loss": 2.7741,
      "num_input_tokens_seen": 56502960,
      "step": 97000
    },
    {
      "epoch": 2.5086839058278656,
      "grad_norm": 6.356966495513916,
      "learning_rate": 8.188601569535573e-06,
      "loss": 2.7821,
      "num_input_tokens_seen": 56796832,
      "step": 97500
    },
    {
      "epoch": 2.5215489514987777,
      "grad_norm": 6.108860015869141,
      "learning_rate": 7.974184141687037e-06,
      "loss": 2.7624,
      "num_input_tokens_seen": 57090056,
      "step": 98000
    },
    {
      "epoch": 2.5344139971696897,
      "grad_norm": 6.559696197509766,
      "learning_rate": 7.7597667138385e-06,
      "loss": 2.7728,
      "num_input_tokens_seen": 57378920,
      "step": 98500
    },
    {
      "epoch": 2.5472790428406022,
      "grad_norm": 7.9952473640441895,
      "learning_rate": 7.545349285989965e-06,
      "loss": 2.7741,
      "num_input_tokens_seen": 57672696,
      "step": 99000
    },
    {
      "epoch": 2.5601440885115143,
      "grad_norm": 5.8723320960998535,
      "learning_rate": 7.3309318581414305e-06,
      "loss": 2.7708,
      "num_input_tokens_seen": 57962544,
      "step": 99500
    },
    {
      "epoch": 2.5730091341824264,
      "grad_norm": 5.872036933898926,
      "learning_rate": 7.116514430292895e-06,
      "loss": 2.754,
      "num_input_tokens_seen": 58254712,
      "step": 100000
    },
    {
      "epoch": 2.5858741798533385,
      "grad_norm": 6.854528427124023,
      "learning_rate": 6.902097002444359e-06,
      "loss": 2.7553,
      "num_input_tokens_seen": 58547432,
      "step": 100500
    },
    {
      "epoch": 2.5987392255242505,
      "grad_norm": 6.242632865905762,
      "learning_rate": 6.6876795745958235e-06,
      "loss": 2.7555,
      "num_input_tokens_seen": 58839640,
      "step": 101000
    },
    {
      "epoch": 2.6116042711951626,
      "grad_norm": 5.893955230712891,
      "learning_rate": 6.473262146747287e-06,
      "loss": 2.743,
      "num_input_tokens_seen": 59126568,
      "step": 101500
    },
    {
      "epoch": 2.624469316866075,
      "grad_norm": 7.543076038360596,
      "learning_rate": 6.258844718898753e-06,
      "loss": 2.7901,
      "num_input_tokens_seen": 59416160,
      "step": 102000
    },
    {
      "epoch": 2.637334362536987,
      "grad_norm": 6.379047870635986,
      "learning_rate": 6.044427291050217e-06,
      "loss": 2.7763,
      "num_input_tokens_seen": 59705440,
      "step": 102500
    },
    {
      "epoch": 2.6501994082078992,
      "grad_norm": 7.814338684082031,
      "learning_rate": 5.830009863201681e-06,
      "loss": 2.7587,
      "num_input_tokens_seen": 59997352,
      "step": 103000
    },
    {
      "epoch": 2.6630644538788113,
      "grad_norm": 6.58440637588501,
      "learning_rate": 5.615592435353146e-06,
      "loss": 2.7562,
      "num_input_tokens_seen": 60289912,
      "step": 103500
    },
    {
      "epoch": 2.6759294995497234,
      "grad_norm": 5.680325984954834,
      "learning_rate": 5.40117500750461e-06,
      "loss": 2.7502,
      "num_input_tokens_seen": 60580920,
      "step": 104000
    },
    {
      "epoch": 2.6887945452206354,
      "grad_norm": 6.846250534057617,
      "learning_rate": 5.186757579656075e-06,
      "loss": 2.7138,
      "num_input_tokens_seen": 60874824,
      "step": 104500
    },
    {
      "epoch": 2.7016595908915475,
      "grad_norm": 5.211071014404297,
      "learning_rate": 4.9723401518075395e-06,
      "loss": 2.777,
      "num_input_tokens_seen": 61169240,
      "step": 105000
    },
    {
      "epoch": 2.7145246365624596,
      "grad_norm": 6.492147445678711,
      "learning_rate": 4.757922723959004e-06,
      "loss": 2.7709,
      "num_input_tokens_seen": 61457952,
      "step": 105500
    },
    {
      "epoch": 2.727389682233372,
      "grad_norm": 6.307532787322998,
      "learning_rate": 4.543505296110468e-06,
      "loss": 2.7632,
      "num_input_tokens_seen": 61747000,
      "step": 106000
    },
    {
      "epoch": 2.740254727904284,
      "grad_norm": 6.944429874420166,
      "learning_rate": 4.3290878682619325e-06,
      "loss": 2.7721,
      "num_input_tokens_seen": 62037496,
      "step": 106500
    },
    {
      "epoch": 2.753119773575196,
      "grad_norm": 7.263415813446045,
      "learning_rate": 4.114670440413397e-06,
      "loss": 2.7227,
      "num_input_tokens_seen": 62328752,
      "step": 107000
    },
    {
      "epoch": 2.7659848192461083,
      "grad_norm": 6.571951389312744,
      "learning_rate": 3.900253012564861e-06,
      "loss": 2.7442,
      "num_input_tokens_seen": 62623568,
      "step": 107500
    },
    {
      "epoch": 2.7788498649170204,
      "grad_norm": 5.274496078491211,
      "learning_rate": 3.685835584716326e-06,
      "loss": 2.7609,
      "num_input_tokens_seen": 62911280,
      "step": 108000
    },
    {
      "epoch": 2.791714910587933,
      "grad_norm": 5.882663249969482,
      "learning_rate": 3.47141815686779e-06,
      "loss": 2.7322,
      "num_input_tokens_seen": 63204016,
      "step": 108500
    },
    {
      "epoch": 2.804579956258845,
      "grad_norm": 6.151833534240723,
      "learning_rate": 3.257000729019255e-06,
      "loss": 2.7582,
      "num_input_tokens_seen": 63493976,
      "step": 109000
    },
    {
      "epoch": 2.817445001929757,
      "grad_norm": 6.4147233963012695,
      "learning_rate": 3.0425833011707193e-06,
      "loss": 2.7465,
      "num_input_tokens_seen": 63788848,
      "step": 109500
    },
    {
      "epoch": 2.830310047600669,
      "grad_norm": 7.0969414710998535,
      "learning_rate": 2.8281658733221834e-06,
      "loss": 2.7422,
      "num_input_tokens_seen": 64079248,
      "step": 110000
    },
    {
      "epoch": 2.843175093271581,
      "grad_norm": 6.523220539093018,
      "learning_rate": 2.613748445473648e-06,
      "loss": 2.7598,
      "num_input_tokens_seen": 64370480,
      "step": 110500
    },
    {
      "epoch": 2.856040138942493,
      "grad_norm": 6.571617126464844,
      "learning_rate": 2.3993310176251127e-06,
      "loss": 2.7381,
      "num_input_tokens_seen": 64660624,
      "step": 111000
    },
    {
      "epoch": 2.8689051846134053,
      "grad_norm": 6.383349895477295,
      "learning_rate": 2.1849135897765773e-06,
      "loss": 2.738,
      "num_input_tokens_seen": 64949432,
      "step": 111500
    },
    {
      "epoch": 2.8817702302843173,
      "grad_norm": 6.797569751739502,
      "learning_rate": 1.9704961619280415e-06,
      "loss": 2.7539,
      "num_input_tokens_seen": 65243120,
      "step": 112000
    },
    {
      "epoch": 2.8946352759552294,
      "grad_norm": 6.340763568878174,
      "learning_rate": 1.756078734079506e-06,
      "loss": 2.7239,
      "num_input_tokens_seen": 65535664,
      "step": 112500
    },
    {
      "epoch": 2.907500321626142,
      "grad_norm": 5.992878437042236,
      "learning_rate": 1.5416613062309707e-06,
      "loss": 2.7459,
      "num_input_tokens_seen": 65826080,
      "step": 113000
    },
    {
      "epoch": 2.920365367297054,
      "grad_norm": 6.328054904937744,
      "learning_rate": 1.3272438783824349e-06,
      "loss": 2.7524,
      "num_input_tokens_seen": 66121728,
      "step": 113500
    },
    {
      "epoch": 2.933230412967966,
      "grad_norm": 5.5994486808776855,
      "learning_rate": 1.1128264505338995e-06,
      "loss": 2.7688,
      "num_input_tokens_seen": 66410432,
      "step": 114000
    },
    {
      "epoch": 2.946095458638878,
      "grad_norm": 6.498249053955078,
      "learning_rate": 8.984090226853639e-07,
      "loss": 2.762,
      "num_input_tokens_seen": 66703552,
      "step": 114500
    },
    {
      "epoch": 2.95896050430979,
      "grad_norm": 5.900263786315918,
      "learning_rate": 6.839915948368284e-07,
      "loss": 2.7559,
      "num_input_tokens_seen": 66994216,
      "step": 115000
    },
    {
      "epoch": 2.9718255499807027,
      "grad_norm": 6.3716864585876465,
      "learning_rate": 4.6957416698829286e-07,
      "loss": 2.7289,
      "num_input_tokens_seen": 67284840,
      "step": 115500
    },
    {
      "epoch": 2.9846905956516148,
      "grad_norm": 7.285637378692627,
      "learning_rate": 2.5515673913975725e-07,
      "loss": 2.7593,
      "num_input_tokens_seen": 67578272,
      "step": 116000
    },
    {
      "epoch": 2.997555641322527,
      "grad_norm": 5.382935523986816,
      "learning_rate": 4.073931129122175e-08,
      "loss": 2.7668,
      "num_input_tokens_seen": 67872888,
      "step": 116500
    },
    {
      "epoch": 3.0,
      "num_input_tokens_seen": 67927868,
      "step": 116595,
      "total_flos": 1.7989403580039168e+16,
      "train_loss": 1.1922986107342055,
      "train_runtime": 2155.3414,
      "train_samples_per_second": 432.761,
      "train_steps_per_second": 54.096,
      "train_tokens_per_second": 31520.605
    }
  ],
  "logging_steps": 500,
  "max_steps": 116595,
  "num_input_tokens_seen": 67927868,
  "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": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.7989403580039168e+16,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}