{
  "best_global_step": 2800,
  "best_metric": 0.94057297706604,
  "best_model_checkpoint": "models/MNLP_M3_rag_model_test/checkpoint-2800",
  "epoch": 2.923237597911227,
  "eval_steps": 200,
  "global_step": 2800,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.010443864229765013,
      "grad_norm": 10.195783615112305,
      "learning_rate": 9.000000000000001e-07,
      "loss": 3.4273,
      "step": 10
    },
    {
      "epoch": 0.020887728459530026,
      "grad_norm": 7.810600280761719,
      "learning_rate": 1.9000000000000002e-06,
      "loss": 3.2543,
      "step": 20
    },
    {
      "epoch": 0.031331592689295036,
      "grad_norm": 5.423489570617676,
      "learning_rate": 2.9e-06,
      "loss": 3.0848,
      "step": 30
    },
    {
      "epoch": 0.04177545691906005,
      "grad_norm": 6.003882884979248,
      "learning_rate": 3.900000000000001e-06,
      "loss": 3.0378,
      "step": 40
    },
    {
      "epoch": 0.05221932114882506,
      "grad_norm": 5.395635604858398,
      "learning_rate": 4.9000000000000005e-06,
      "loss": 2.935,
      "step": 50
    },
    {
      "epoch": 0.06266318537859007,
      "grad_norm": 5.4613823890686035,
      "learning_rate": 5.9e-06,
      "loss": 2.8095,
      "step": 60
    },
    {
      "epoch": 0.0731070496083551,
      "grad_norm": 5.638515472412109,
      "learning_rate": 6.9e-06,
      "loss": 2.8053,
      "step": 70
    },
    {
      "epoch": 0.0835509138381201,
      "grad_norm": 5.723353385925293,
      "learning_rate": 7.9e-06,
      "loss": 2.7206,
      "step": 80
    },
    {
      "epoch": 0.09399477806788512,
      "grad_norm": 6.676548480987549,
      "learning_rate": 8.900000000000001e-06,
      "loss": 2.6614,
      "step": 90
    },
    {
      "epoch": 0.10443864229765012,
      "grad_norm": 6.09738302230835,
      "learning_rate": 9.9e-06,
      "loss": 2.4909,
      "step": 100
    },
    {
      "epoch": 0.11488250652741515,
      "grad_norm": 6.762812614440918,
      "learning_rate": 9.967520750631542e-06,
      "loss": 2.4897,
      "step": 110
    },
    {
      "epoch": 0.12532637075718014,
      "grad_norm": 6.7795796394348145,
      "learning_rate": 9.931432695777698e-06,
      "loss": 2.3994,
      "step": 120
    },
    {
      "epoch": 0.13577023498694518,
      "grad_norm": 6.7266669273376465,
      "learning_rate": 9.895344640923855e-06,
      "loss": 2.395,
      "step": 130
    },
    {
      "epoch": 0.1462140992167102,
      "grad_norm": 6.685121536254883,
      "learning_rate": 9.859256586070011e-06,
      "loss": 2.3435,
      "step": 140
    },
    {
      "epoch": 0.1566579634464752,
      "grad_norm": 7.3826985359191895,
      "learning_rate": 9.823168531216168e-06,
      "loss": 2.1847,
      "step": 150
    },
    {
      "epoch": 0.1671018276762402,
      "grad_norm": 6.167830467224121,
      "learning_rate": 9.787080476362326e-06,
      "loss": 2.2339,
      "step": 160
    },
    {
      "epoch": 0.17754569190600522,
      "grad_norm": 9.586319923400879,
      "learning_rate": 9.750992421508482e-06,
      "loss": 2.1565,
      "step": 170
    },
    {
      "epoch": 0.18798955613577023,
      "grad_norm": 9.096244812011719,
      "learning_rate": 9.714904366654639e-06,
      "loss": 2.2398,
      "step": 180
    },
    {
      "epoch": 0.19843342036553524,
      "grad_norm": 9.149866104125977,
      "learning_rate": 9.68242511728618e-06,
      "loss": 2.0402,
      "step": 190
    },
    {
      "epoch": 0.20887728459530025,
      "grad_norm": 7.662464141845703,
      "learning_rate": 9.646337062432336e-06,
      "loss": 1.853,
      "step": 200
    },
    {
      "epoch": 0.20887728459530025,
      "eval_loss": 1.9208216667175293,
      "eval_runtime": 23.2773,
      "eval_samples_per_second": 36.559,
      "eval_steps_per_second": 4.597,
      "step": 200
    },
    {
      "epoch": 0.2193211488250653,
      "grad_norm": 7.436556339263916,
      "learning_rate": 9.610249007578492e-06,
      "loss": 1.9207,
      "step": 210
    },
    {
      "epoch": 0.2297650130548303,
      "grad_norm": 7.291353225708008,
      "learning_rate": 9.574160952724649e-06,
      "loss": 1.9513,
      "step": 220
    },
    {
      "epoch": 0.2402088772845953,
      "grad_norm": 7.357730865478516,
      "learning_rate": 9.538072897870805e-06,
      "loss": 1.8785,
      "step": 230
    },
    {
      "epoch": 0.2506527415143603,
      "grad_norm": 7.417892932891846,
      "learning_rate": 9.501984843016962e-06,
      "loss": 1.8304,
      "step": 240
    },
    {
      "epoch": 0.26109660574412535,
      "grad_norm": 7.092027187347412,
      "learning_rate": 9.46589678816312e-06,
      "loss": 1.7244,
      "step": 250
    },
    {
      "epoch": 0.27154046997389036,
      "grad_norm": 8.140602111816406,
      "learning_rate": 9.429808733309276e-06,
      "loss": 1.9412,
      "step": 260
    },
    {
      "epoch": 0.2819843342036554,
      "grad_norm": 6.528462886810303,
      "learning_rate": 9.393720678455433e-06,
      "loss": 1.7654,
      "step": 270
    },
    {
      "epoch": 0.2924281984334204,
      "grad_norm": 6.523935794830322,
      "learning_rate": 9.35763262360159e-06,
      "loss": 1.8401,
      "step": 280
    },
    {
      "epoch": 0.3028720626631854,
      "grad_norm": 6.85488224029541,
      "learning_rate": 9.321544568747746e-06,
      "loss": 1.8402,
      "step": 290
    },
    {
      "epoch": 0.3133159268929504,
      "grad_norm": 6.1414875984191895,
      "learning_rate": 9.285456513893902e-06,
      "loss": 1.682,
      "step": 300
    },
    {
      "epoch": 0.3237597911227154,
      "grad_norm": 7.549343109130859,
      "learning_rate": 9.249368459040059e-06,
      "loss": 1.5124,
      "step": 310
    },
    {
      "epoch": 0.3342036553524804,
      "grad_norm": 7.781056880950928,
      "learning_rate": 9.213280404186215e-06,
      "loss": 1.6474,
      "step": 320
    },
    {
      "epoch": 0.34464751958224543,
      "grad_norm": 6.674058437347412,
      "learning_rate": 9.177192349332372e-06,
      "loss": 1.6971,
      "step": 330
    },
    {
      "epoch": 0.35509138381201044,
      "grad_norm": 8.93964958190918,
      "learning_rate": 9.14110429447853e-06,
      "loss": 1.5545,
      "step": 340
    },
    {
      "epoch": 0.36553524804177545,
      "grad_norm": 7.935058116912842,
      "learning_rate": 9.105016239624686e-06,
      "loss": 1.6633,
      "step": 350
    },
    {
      "epoch": 0.37597911227154046,
      "grad_norm": 6.446653842926025,
      "learning_rate": 9.068928184770843e-06,
      "loss": 1.6209,
      "step": 360
    },
    {
      "epoch": 0.38642297650130547,
      "grad_norm": 9.611429214477539,
      "learning_rate": 9.036448935402382e-06,
      "loss": 1.4615,
      "step": 370
    },
    {
      "epoch": 0.3968668407310705,
      "grad_norm": 8.988059043884277,
      "learning_rate": 9.000360880548538e-06,
      "loss": 1.5815,
      "step": 380
    },
    {
      "epoch": 0.4073107049608355,
      "grad_norm": 7.07716178894043,
      "learning_rate": 8.964272825694695e-06,
      "loss": 1.6719,
      "step": 390
    },
    {
      "epoch": 0.4177545691906005,
      "grad_norm": 6.789717674255371,
      "learning_rate": 8.928184770840851e-06,
      "loss": 1.4354,
      "step": 400
    },
    {
      "epoch": 0.4177545691906005,
      "eval_loss": 1.4963077306747437,
      "eval_runtime": 23.1593,
      "eval_samples_per_second": 36.746,
      "eval_steps_per_second": 4.62,
      "step": 400
    },
    {
      "epoch": 0.4281984334203655,
      "grad_norm": 6.676051139831543,
      "learning_rate": 8.89209671598701e-06,
      "loss": 1.695,
      "step": 410
    },
    {
      "epoch": 0.4386422976501306,
      "grad_norm": 6.711126327514648,
      "learning_rate": 8.856008661133166e-06,
      "loss": 1.6562,
      "step": 420
    },
    {
      "epoch": 0.4490861618798956,
      "grad_norm": 7.0567731857299805,
      "learning_rate": 8.819920606279322e-06,
      "loss": 1.6646,
      "step": 430
    },
    {
      "epoch": 0.4595300261096606,
      "grad_norm": 6.944768905639648,
      "learning_rate": 8.783832551425479e-06,
      "loss": 1.4864,
      "step": 440
    },
    {
      "epoch": 0.4699738903394256,
      "grad_norm": 6.025828838348389,
      "learning_rate": 8.747744496571635e-06,
      "loss": 1.4261,
      "step": 450
    },
    {
      "epoch": 0.4804177545691906,
      "grad_norm": 7.318626880645752,
      "learning_rate": 8.711656441717792e-06,
      "loss": 1.4183,
      "step": 460
    },
    {
      "epoch": 0.4908616187989556,
      "grad_norm": 9.207475662231445,
      "learning_rate": 8.675568386863948e-06,
      "loss": 1.6308,
      "step": 470
    },
    {
      "epoch": 0.5013054830287206,
      "grad_norm": 7.152507781982422,
      "learning_rate": 8.639480332010105e-06,
      "loss": 1.5118,
      "step": 480
    },
    {
      "epoch": 0.5117493472584856,
      "grad_norm": 6.89570426940918,
      "learning_rate": 8.603392277156261e-06,
      "loss": 1.6088,
      "step": 490
    },
    {
      "epoch": 0.5221932114882507,
      "grad_norm": 8.016292572021484,
      "learning_rate": 8.567304222302419e-06,
      "loss": 1.6461,
      "step": 500
    },
    {
      "epoch": 0.5326370757180157,
      "grad_norm": 5.700518608093262,
      "learning_rate": 8.531216167448576e-06,
      "loss": 1.5497,
      "step": 510
    },
    {
      "epoch": 0.5430809399477807,
      "grad_norm": 7.127388000488281,
      "learning_rate": 8.495128112594732e-06,
      "loss": 1.4443,
      "step": 520
    },
    {
      "epoch": 0.5535248041775457,
      "grad_norm": 9.201033592224121,
      "learning_rate": 8.459040057740888e-06,
      "loss": 1.3384,
      "step": 530
    },
    {
      "epoch": 0.5639686684073107,
      "grad_norm": 9.565522193908691,
      "learning_rate": 8.422952002887045e-06,
      "loss": 1.3732,
      "step": 540
    },
    {
      "epoch": 0.5744125326370757,
      "grad_norm": 6.29095458984375,
      "learning_rate": 8.386863948033201e-06,
      "loss": 1.5768,
      "step": 550
    },
    {
      "epoch": 0.5848563968668408,
      "grad_norm": 7.279628753662109,
      "learning_rate": 8.350775893179358e-06,
      "loss": 1.5237,
      "step": 560
    },
    {
      "epoch": 0.5953002610966057,
      "grad_norm": 5.866252899169922,
      "learning_rate": 8.314687838325514e-06,
      "loss": 1.491,
      "step": 570
    },
    {
      "epoch": 0.6057441253263708,
      "grad_norm": 6.8628621101379395,
      "learning_rate": 8.27859978347167e-06,
      "loss": 1.3991,
      "step": 580
    },
    {
      "epoch": 0.6161879895561357,
      "grad_norm": 5.9102654457092285,
      "learning_rate": 8.242511728617829e-06,
      "loss": 1.3842,
      "step": 590
    },
    {
      "epoch": 0.6266318537859008,
      "grad_norm": 6.6509199142456055,
      "learning_rate": 8.206423673763985e-06,
      "loss": 1.4015,
      "step": 600
    },
    {
      "epoch": 0.6266318537859008,
      "eval_loss": 1.3116086721420288,
      "eval_runtime": 23.131,
      "eval_samples_per_second": 36.79,
      "eval_steps_per_second": 4.626,
      "step": 600
    },
    {
      "epoch": 0.6370757180156658,
      "grad_norm": 10.004154205322266,
      "learning_rate": 8.170335618910142e-06,
      "loss": 1.4046,
      "step": 610
    },
    {
      "epoch": 0.6475195822454308,
      "grad_norm": 6.851074695587158,
      "learning_rate": 8.134247564056298e-06,
      "loss": 1.3358,
      "step": 620
    },
    {
      "epoch": 0.6579634464751958,
      "grad_norm": 6.676290512084961,
      "learning_rate": 8.098159509202455e-06,
      "loss": 1.2649,
      "step": 630
    },
    {
      "epoch": 0.6684073107049608,
      "grad_norm": 6.854121685028076,
      "learning_rate": 8.062071454348611e-06,
      "loss": 1.4563,
      "step": 640
    },
    {
      "epoch": 0.6788511749347258,
      "grad_norm": 7.104318141937256,
      "learning_rate": 8.025983399494768e-06,
      "loss": 1.2667,
      "step": 650
    },
    {
      "epoch": 0.6892950391644909,
      "grad_norm": 6.3163957595825195,
      "learning_rate": 7.989895344640924e-06,
      "loss": 1.4532,
      "step": 660
    },
    {
      "epoch": 0.6997389033942559,
      "grad_norm": 7.5210652351379395,
      "learning_rate": 7.95380728978708e-06,
      "loss": 1.179,
      "step": 670
    },
    {
      "epoch": 0.7101827676240209,
      "grad_norm": 7.099525451660156,
      "learning_rate": 7.917719234933237e-06,
      "loss": 1.2406,
      "step": 680
    },
    {
      "epoch": 0.720626631853786,
      "grad_norm": 6.814334392547607,
      "learning_rate": 7.881631180079395e-06,
      "loss": 1.1178,
      "step": 690
    },
    {
      "epoch": 0.7310704960835509,
      "grad_norm": 6.895037651062012,
      "learning_rate": 7.845543125225551e-06,
      "loss": 1.3291,
      "step": 700
    },
    {
      "epoch": 0.741514360313316,
      "grad_norm": 6.383569717407227,
      "learning_rate": 7.809455070371708e-06,
      "loss": 1.2887,
      "step": 710
    },
    {
      "epoch": 0.7519582245430809,
      "grad_norm": 8.13758373260498,
      "learning_rate": 7.773367015517864e-06,
      "loss": 1.2576,
      "step": 720
    },
    {
      "epoch": 0.762402088772846,
      "grad_norm": 5.2739362716674805,
      "learning_rate": 7.73727896066402e-06,
      "loss": 1.2835,
      "step": 730
    },
    {
      "epoch": 0.7728459530026109,
      "grad_norm": 7.7005934715271,
      "learning_rate": 7.701190905810177e-06,
      "loss": 1.2633,
      "step": 740
    },
    {
      "epoch": 0.783289817232376,
      "grad_norm": 6.738245964050293,
      "learning_rate": 7.665102850956334e-06,
      "loss": 1.2129,
      "step": 750
    },
    {
      "epoch": 0.793733681462141,
      "grad_norm": 5.725705146789551,
      "learning_rate": 7.629014796102491e-06,
      "loss": 1.2501,
      "step": 760
    },
    {
      "epoch": 0.804177545691906,
      "grad_norm": 6.256880760192871,
      "learning_rate": 7.5929267412486475e-06,
      "loss": 1.2567,
      "step": 770
    },
    {
      "epoch": 0.814621409921671,
      "grad_norm": 6.835810661315918,
      "learning_rate": 7.556838686394804e-06,
      "loss": 1.2408,
      "step": 780
    },
    {
      "epoch": 0.825065274151436,
      "grad_norm": 5.811525821685791,
      "learning_rate": 7.52075063154096e-06,
      "loss": 1.2409,
      "step": 790
    },
    {
      "epoch": 0.835509138381201,
      "grad_norm": 8.554996490478516,
      "learning_rate": 7.484662576687118e-06,
      "loss": 1.2575,
      "step": 800
    },
    {
      "epoch": 0.835509138381201,
      "eval_loss": 1.2006852626800537,
      "eval_runtime": 23.1272,
      "eval_samples_per_second": 36.796,
      "eval_steps_per_second": 4.627,
      "step": 800
    },
    {
      "epoch": 0.8459530026109661,
      "grad_norm": 7.091418266296387,
      "learning_rate": 7.448574521833274e-06,
      "loss": 1.0256,
      "step": 810
    },
    {
      "epoch": 0.856396866840731,
      "grad_norm": 7.5542683601379395,
      "learning_rate": 7.4124864669794306e-06,
      "loss": 1.0605,
      "step": 820
    },
    {
      "epoch": 0.8668407310704961,
      "grad_norm": 7.177179336547852,
      "learning_rate": 7.376398412125587e-06,
      "loss": 1.1747,
      "step": 830
    },
    {
      "epoch": 0.8772845953002611,
      "grad_norm": 6.226783752441406,
      "learning_rate": 7.3403103572717434e-06,
      "loss": 1.1518,
      "step": 840
    },
    {
      "epoch": 0.8877284595300261,
      "grad_norm": 7.062104225158691,
      "learning_rate": 7.304222302417901e-06,
      "loss": 1.2322,
      "step": 850
    },
    {
      "epoch": 0.8981723237597912,
      "grad_norm": 8.756392478942871,
      "learning_rate": 7.268134247564057e-06,
      "loss": 1.271,
      "step": 860
    },
    {
      "epoch": 0.9086161879895561,
      "grad_norm": 6.364953994750977,
      "learning_rate": 7.232046192710214e-06,
      "loss": 1.1853,
      "step": 870
    },
    {
      "epoch": 0.9190600522193212,
      "grad_norm": 6.629589080810547,
      "learning_rate": 7.19595813785637e-06,
      "loss": 1.2279,
      "step": 880
    },
    {
      "epoch": 0.9295039164490861,
      "grad_norm": 5.4719977378845215,
      "learning_rate": 7.1598700830025265e-06,
      "loss": 1.0789,
      "step": 890
    },
    {
      "epoch": 0.9399477806788512,
      "grad_norm": 5.824141025543213,
      "learning_rate": 7.123782028148684e-06,
      "loss": 1.0435,
      "step": 900
    },
    {
      "epoch": 0.9503916449086162,
      "grad_norm": 9.214339256286621,
      "learning_rate": 7.08769397329484e-06,
      "loss": 1.0518,
      "step": 910
    },
    {
      "epoch": 0.9608355091383812,
      "grad_norm": 6.3007941246032715,
      "learning_rate": 7.051605918440997e-06,
      "loss": 1.0297,
      "step": 920
    },
    {
      "epoch": 0.9712793733681462,
      "grad_norm": 5.431386947631836,
      "learning_rate": 7.015517863587153e-06,
      "loss": 1.1292,
      "step": 930
    },
    {
      "epoch": 0.9817232375979112,
      "grad_norm": 8.941716194152832,
      "learning_rate": 6.9794298087333105e-06,
      "loss": 1.1617,
      "step": 940
    },
    {
      "epoch": 0.9921671018276762,
      "grad_norm": 8.926597595214844,
      "learning_rate": 6.943341753879467e-06,
      "loss": 1.3497,
      "step": 950
    },
    {
      "epoch": 1.002088772845953,
      "grad_norm": 6.556091785430908,
      "learning_rate": 6.907253699025623e-06,
      "loss": 1.176,
      "step": 960
    },
    {
      "epoch": 1.012532637075718,
      "grad_norm": 12.022808074951172,
      "learning_rate": 6.87116564417178e-06,
      "loss": 0.9251,
      "step": 970
    },
    {
      "epoch": 1.022976501305483,
      "grad_norm": 7.0122389793396,
      "learning_rate": 6.835077589317936e-06,
      "loss": 0.9231,
      "step": 980
    },
    {
      "epoch": 1.033420365535248,
      "grad_norm": 6.355090141296387,
      "learning_rate": 6.7989895344640936e-06,
      "loss": 1.0371,
      "step": 990
    },
    {
      "epoch": 1.0438642297650131,
      "grad_norm": 7.36827278137207,
      "learning_rate": 6.76290147961025e-06,
      "loss": 1.0666,
      "step": 1000
    },
    {
      "epoch": 1.0438642297650131,
      "eval_loss": 1.1358416080474854,
      "eval_runtime": 23.1354,
      "eval_samples_per_second": 36.783,
      "eval_steps_per_second": 4.625,
      "step": 1000
    },
    {
      "epoch": 1.054308093994778,
      "grad_norm": 5.1998772621154785,
      "learning_rate": 6.7268134247564065e-06,
      "loss": 0.847,
      "step": 1010
    },
    {
      "epoch": 1.064751958224543,
      "grad_norm": 6.828693866729736,
      "learning_rate": 6.690725369902563e-06,
      "loss": 1.0755,
      "step": 1020
    },
    {
      "epoch": 1.0751958224543081,
      "grad_norm": 6.408750534057617,
      "learning_rate": 6.65463731504872e-06,
      "loss": 0.9039,
      "step": 1030
    },
    {
      "epoch": 1.0856396866840732,
      "grad_norm": 7.197463035583496,
      "learning_rate": 6.618549260194877e-06,
      "loss": 0.959,
      "step": 1040
    },
    {
      "epoch": 1.096083550913838,
      "grad_norm": 5.504906177520752,
      "learning_rate": 6.582461205341033e-06,
      "loss": 0.9467,
      "step": 1050
    },
    {
      "epoch": 1.106527415143603,
      "grad_norm": 7.88711404800415,
      "learning_rate": 6.5463731504871896e-06,
      "loss": 1.0635,
      "step": 1060
    },
    {
      "epoch": 1.1169712793733682,
      "grad_norm": 5.1561055183410645,
      "learning_rate": 6.510285095633346e-06,
      "loss": 0.9635,
      "step": 1070
    },
    {
      "epoch": 1.1274151436031332,
      "grad_norm": 5.088303565979004,
      "learning_rate": 6.474197040779503e-06,
      "loss": 0.9605,
      "step": 1080
    },
    {
      "epoch": 1.137859007832898,
      "grad_norm": 4.595157146453857,
      "learning_rate": 6.43810898592566e-06,
      "loss": 1.0494,
      "step": 1090
    },
    {
      "epoch": 1.1483028720626631,
      "grad_norm": 6.682077884674072,
      "learning_rate": 6.402020931071816e-06,
      "loss": 0.888,
      "step": 1100
    },
    {
      "epoch": 1.1587467362924282,
      "grad_norm": 9.08073902130127,
      "learning_rate": 6.365932876217973e-06,
      "loss": 1.1239,
      "step": 1110
    },
    {
      "epoch": 1.1691906005221933,
      "grad_norm": 7.480021953582764,
      "learning_rate": 6.329844821364129e-06,
      "loss": 1.0337,
      "step": 1120
    },
    {
      "epoch": 1.1796344647519583,
      "grad_norm": 5.750502109527588,
      "learning_rate": 6.293756766510286e-06,
      "loss": 1.132,
      "step": 1130
    },
    {
      "epoch": 1.1900783289817232,
      "grad_norm": 6.151831150054932,
      "learning_rate": 6.257668711656443e-06,
      "loss": 0.9985,
      "step": 1140
    },
    {
      "epoch": 1.2005221932114882,
      "grad_norm": 6.567698001861572,
      "learning_rate": 6.221580656802599e-06,
      "loss": 1.1177,
      "step": 1150
    },
    {
      "epoch": 1.2109660574412533,
      "grad_norm": 6.1963887214660645,
      "learning_rate": 6.185492601948756e-06,
      "loss": 0.9328,
      "step": 1160
    },
    {
      "epoch": 1.2214099216710184,
      "grad_norm": 5.044003009796143,
      "learning_rate": 6.149404547094913e-06,
      "loss": 0.8986,
      "step": 1170
    },
    {
      "epoch": 1.2318537859007832,
      "grad_norm": 6.648619651794434,
      "learning_rate": 6.113316492241068e-06,
      "loss": 0.9905,
      "step": 1180
    },
    {
      "epoch": 1.2422976501305483,
      "grad_norm": 9.110396385192871,
      "learning_rate": 6.077228437387225e-06,
      "loss": 1.0126,
      "step": 1190
    },
    {
      "epoch": 1.2527415143603133,
      "grad_norm": 8.707374572753906,
      "learning_rate": 6.0411403825333815e-06,
      "loss": 0.8611,
      "step": 1200
    },
    {
      "epoch": 1.2527415143603133,
      "eval_loss": 1.0788438320159912,
      "eval_runtime": 23.1258,
      "eval_samples_per_second": 36.799,
      "eval_steps_per_second": 4.627,
      "step": 1200
    },
    {
      "epoch": 1.2631853785900784,
      "grad_norm": 5.457468509674072,
      "learning_rate": 6.005052327679538e-06,
      "loss": 0.8327,
      "step": 1210
    },
    {
      "epoch": 1.2736292428198435,
      "grad_norm": 6.050765037536621,
      "learning_rate": 5.9689642728256944e-06,
      "loss": 1.0782,
      "step": 1220
    },
    {
      "epoch": 1.2840731070496083,
      "grad_norm": 6.254447937011719,
      "learning_rate": 5.932876217971852e-06,
      "loss": 0.9388,
      "step": 1230
    },
    {
      "epoch": 1.2945169712793734,
      "grad_norm": 5.181304931640625,
      "learning_rate": 5.896788163118008e-06,
      "loss": 0.9711,
      "step": 1240
    },
    {
      "epoch": 1.3049608355091384,
      "grad_norm": 6.832638263702393,
      "learning_rate": 5.860700108264165e-06,
      "loss": 1.0333,
      "step": 1250
    },
    {
      "epoch": 1.3154046997389033,
      "grad_norm": 8.406023025512695,
      "learning_rate": 5.824612053410321e-06,
      "loss": 1.0902,
      "step": 1260
    },
    {
      "epoch": 1.3258485639686683,
      "grad_norm": 6.346268653869629,
      "learning_rate": 5.7885239985564775e-06,
      "loss": 0.8651,
      "step": 1270
    },
    {
      "epoch": 1.3362924281984334,
      "grad_norm": 8.447615623474121,
      "learning_rate": 5.752435943702635e-06,
      "loss": 1.046,
      "step": 1280
    },
    {
      "epoch": 1.3467362924281985,
      "grad_norm": 8.351264953613281,
      "learning_rate": 5.716347888848791e-06,
      "loss": 1.029,
      "step": 1290
    },
    {
      "epoch": 1.3571801566579635,
      "grad_norm": 6.036417007446289,
      "learning_rate": 5.680259833994948e-06,
      "loss": 1.0089,
      "step": 1300
    },
    {
      "epoch": 1.3676240208877284,
      "grad_norm": 5.646811485290527,
      "learning_rate": 5.644171779141104e-06,
      "loss": 0.9346,
      "step": 1310
    },
    {
      "epoch": 1.3780678851174935,
      "grad_norm": 6.4004950523376465,
      "learning_rate": 5.608083724287261e-06,
      "loss": 0.9489,
      "step": 1320
    },
    {
      "epoch": 1.3885117493472585,
      "grad_norm": 5.746732234954834,
      "learning_rate": 5.571995669433418e-06,
      "loss": 0.9356,
      "step": 1330
    },
    {
      "epoch": 1.3989556135770234,
      "grad_norm": 5.3405632972717285,
      "learning_rate": 5.535907614579574e-06,
      "loss": 0.7099,
      "step": 1340
    },
    {
      "epoch": 1.4093994778067884,
      "grad_norm": 6.424510955810547,
      "learning_rate": 5.499819559725731e-06,
      "loss": 0.6955,
      "step": 1350
    },
    {
      "epoch": 1.4198433420365535,
      "grad_norm": 5.751818656921387,
      "learning_rate": 5.463731504871887e-06,
      "loss": 0.9094,
      "step": 1360
    },
    {
      "epoch": 1.4302872062663186,
      "grad_norm": 7.787084579467773,
      "learning_rate": 5.4276434500180445e-06,
      "loss": 0.9618,
      "step": 1370
    },
    {
      "epoch": 1.4407310704960836,
      "grad_norm": 6.467785835266113,
      "learning_rate": 5.391555395164201e-06,
      "loss": 0.8549,
      "step": 1380
    },
    {
      "epoch": 1.4511749347258487,
      "grad_norm": 6.860886573791504,
      "learning_rate": 5.3554673403103574e-06,
      "loss": 0.8342,
      "step": 1390
    },
    {
      "epoch": 1.4616187989556135,
      "grad_norm": 5.627669811248779,
      "learning_rate": 5.319379285456514e-06,
      "loss": 0.9012,
      "step": 1400
    },
    {
      "epoch": 1.4616187989556135,
      "eval_loss": 1.0375572443008423,
      "eval_runtime": 23.1318,
      "eval_samples_per_second": 36.789,
      "eval_steps_per_second": 4.626,
      "step": 1400
    },
    {
      "epoch": 1.4720626631853786,
      "grad_norm": 7.144196033477783,
      "learning_rate": 5.28329123060267e-06,
      "loss": 0.9949,
      "step": 1410
    },
    {
      "epoch": 1.4825065274151437,
      "grad_norm": 6.208961486816406,
      "learning_rate": 5.247203175748828e-06,
      "loss": 1.0719,
      "step": 1420
    },
    {
      "epoch": 1.4929503916449085,
      "grad_norm": 7.110988140106201,
      "learning_rate": 5.211115120894984e-06,
      "loss": 0.9162,
      "step": 1430
    },
    {
      "epoch": 1.5033942558746736,
      "grad_norm": 6.903599739074707,
      "learning_rate": 5.1750270660411405e-06,
      "loss": 0.9974,
      "step": 1440
    },
    {
      "epoch": 1.5138381201044386,
      "grad_norm": 5.059232234954834,
      "learning_rate": 5.138939011187297e-06,
      "loss": 0.9602,
      "step": 1450
    },
    {
      "epoch": 1.5242819843342037,
      "grad_norm": 3.6045143604278564,
      "learning_rate": 5.1028509563334534e-06,
      "loss": 0.8448,
      "step": 1460
    },
    {
      "epoch": 1.5347258485639688,
      "grad_norm": 8.628995895385742,
      "learning_rate": 5.066762901479611e-06,
      "loss": 0.914,
      "step": 1470
    },
    {
      "epoch": 1.5451697127937338,
      "grad_norm": 7.227733612060547,
      "learning_rate": 5.030674846625767e-06,
      "loss": 0.9206,
      "step": 1480
    },
    {
      "epoch": 1.5556135770234987,
      "grad_norm": 7.930648326873779,
      "learning_rate": 4.994586791771924e-06,
      "loss": 1.1221,
      "step": 1490
    },
    {
      "epoch": 1.5660574412532637,
      "grad_norm": 6.340338706970215,
      "learning_rate": 4.95849873691808e-06,
      "loss": 0.8298,
      "step": 1500
    },
    {
      "epoch": 1.5765013054830286,
      "grad_norm": 5.558096408843994,
      "learning_rate": 4.922410682064237e-06,
      "loss": 0.9457,
      "step": 1510
    },
    {
      "epoch": 1.5869451697127936,
      "grad_norm": 7.608903408050537,
      "learning_rate": 4.886322627210394e-06,
      "loss": 0.9164,
      "step": 1520
    },
    {
      "epoch": 1.5973890339425587,
      "grad_norm": 9.46885871887207,
      "learning_rate": 4.85023457235655e-06,
      "loss": 0.9498,
      "step": 1530
    },
    {
      "epoch": 1.6078328981723238,
      "grad_norm": 7.6691107749938965,
      "learning_rate": 4.814146517502707e-06,
      "loss": 0.9764,
      "step": 1540
    },
    {
      "epoch": 1.6182767624020888,
      "grad_norm": 8.231538772583008,
      "learning_rate": 4.778058462648863e-06,
      "loss": 0.8314,
      "step": 1550
    },
    {
      "epoch": 1.628720626631854,
      "grad_norm": 5.868668556213379,
      "learning_rate": 4.7419704077950205e-06,
      "loss": 0.9127,
      "step": 1560
    },
    {
      "epoch": 1.6391644908616188,
      "grad_norm": 7.236291885375977,
      "learning_rate": 4.705882352941177e-06,
      "loss": 0.9485,
      "step": 1570
    },
    {
      "epoch": 1.6496083550913838,
      "grad_norm": 6.871162414550781,
      "learning_rate": 4.669794298087333e-06,
      "loss": 1.0603,
      "step": 1580
    },
    {
      "epoch": 1.6600522193211487,
      "grad_norm": 6.303982734680176,
      "learning_rate": 4.63370624323349e-06,
      "loss": 0.945,
      "step": 1590
    },
    {
      "epoch": 1.6704960835509137,
      "grad_norm": 7.167915344238281,
      "learning_rate": 4.597618188379647e-06,
      "loss": 0.7351,
      "step": 1600
    },
    {
      "epoch": 1.6704960835509137,
      "eval_loss": 1.0095878839492798,
      "eval_runtime": 23.1917,
      "eval_samples_per_second": 36.694,
      "eval_steps_per_second": 4.614,
      "step": 1600
    },
    {
      "epoch": 1.6809399477806788,
      "grad_norm": 6.626043796539307,
      "learning_rate": 4.5615301335258035e-06,
      "loss": 0.9369,
      "step": 1610
    },
    {
      "epoch": 1.6913838120104439,
      "grad_norm": 6.645303726196289,
      "learning_rate": 4.52544207867196e-06,
      "loss": 0.8509,
      "step": 1620
    },
    {
      "epoch": 1.701827676240209,
      "grad_norm": 7.132906913757324,
      "learning_rate": 4.4893540238181164e-06,
      "loss": 0.9522,
      "step": 1630
    },
    {
      "epoch": 1.712271540469974,
      "grad_norm": 6.155941009521484,
      "learning_rate": 4.453265968964273e-06,
      "loss": 0.9662,
      "step": 1640
    },
    {
      "epoch": 1.722715404699739,
      "grad_norm": 5.0147705078125,
      "learning_rate": 4.41717791411043e-06,
      "loss": 0.8755,
      "step": 1650
    },
    {
      "epoch": 1.733159268929504,
      "grad_norm": 7.039682388305664,
      "learning_rate": 4.381089859256587e-06,
      "loss": 0.8761,
      "step": 1660
    },
    {
      "epoch": 1.743603133159269,
      "grad_norm": 8.536678314208984,
      "learning_rate": 4.345001804402743e-06,
      "loss": 0.8054,
      "step": 1670
    },
    {
      "epoch": 1.7540469973890338,
      "grad_norm": 6.860482215881348,
      "learning_rate": 4.3089137495488995e-06,
      "loss": 0.9095,
      "step": 1680
    },
    {
      "epoch": 1.7644908616187989,
      "grad_norm": 6.3796563148498535,
      "learning_rate": 4.272825694695056e-06,
      "loss": 1.08,
      "step": 1690
    },
    {
      "epoch": 1.774934725848564,
      "grad_norm": 11.013219833374023,
      "learning_rate": 4.236737639841213e-06,
      "loss": 0.834,
      "step": 1700
    },
    {
      "epoch": 1.785378590078329,
      "grad_norm": 6.231220722198486,
      "learning_rate": 4.20064958498737e-06,
      "loss": 0.8488,
      "step": 1710
    },
    {
      "epoch": 1.795822454308094,
      "grad_norm": 7.019144535064697,
      "learning_rate": 4.164561530133526e-06,
      "loss": 0.7922,
      "step": 1720
    },
    {
      "epoch": 1.8062663185378591,
      "grad_norm": 6.586852550506592,
      "learning_rate": 4.128473475279683e-06,
      "loss": 0.8326,
      "step": 1730
    },
    {
      "epoch": 1.816710182767624,
      "grad_norm": 6.34022855758667,
      "learning_rate": 4.09238542042584e-06,
      "loss": 0.906,
      "step": 1740
    },
    {
      "epoch": 1.827154046997389,
      "grad_norm": 7.541686058044434,
      "learning_rate": 4.056297365571996e-06,
      "loss": 0.9029,
      "step": 1750
    },
    {
      "epoch": 1.837597911227154,
      "grad_norm": 5.867885589599609,
      "learning_rate": 4.020209310718153e-06,
      "loss": 0.8072,
      "step": 1760
    },
    {
      "epoch": 1.848041775456919,
      "grad_norm": 4.805484771728516,
      "learning_rate": 3.984121255864309e-06,
      "loss": 0.9158,
      "step": 1770
    },
    {
      "epoch": 1.858485639686684,
      "grad_norm": 5.949447154998779,
      "learning_rate": 3.948033201010466e-06,
      "loss": 0.8116,
      "step": 1780
    },
    {
      "epoch": 1.868929503916449,
      "grad_norm": 7.641289234161377,
      "learning_rate": 3.911945146156623e-06,
      "loss": 0.8824,
      "step": 1790
    },
    {
      "epoch": 1.8793733681462141,
      "grad_norm": 5.530484199523926,
      "learning_rate": 3.8758570913027795e-06,
      "loss": 0.9264,
      "step": 1800
    },
    {
      "epoch": 1.8793733681462141,
      "eval_loss": 0.9786838293075562,
      "eval_runtime": 23.1205,
      "eval_samples_per_second": 36.807,
      "eval_steps_per_second": 4.628,
      "step": 1800
    },
    {
      "epoch": 1.8898172323759792,
      "grad_norm": 5.981470584869385,
      "learning_rate": 3.839769036448936e-06,
      "loss": 0.7929,
      "step": 1810
    },
    {
      "epoch": 1.9002610966057443,
      "grad_norm": 8.019466400146484,
      "learning_rate": 3.8036809815950928e-06,
      "loss": 0.7616,
      "step": 1820
    },
    {
      "epoch": 1.9107049608355091,
      "grad_norm": 7.649405479431152,
      "learning_rate": 3.7675929267412492e-06,
      "loss": 0.8661,
      "step": 1830
    },
    {
      "epoch": 1.9211488250652742,
      "grad_norm": 8.259872436523438,
      "learning_rate": 3.7315048718874057e-06,
      "loss": 1.0303,
      "step": 1840
    },
    {
      "epoch": 1.931592689295039,
      "grad_norm": 5.947850227355957,
      "learning_rate": 3.695416817033562e-06,
      "loss": 0.9326,
      "step": 1850
    },
    {
      "epoch": 1.942036553524804,
      "grad_norm": 5.193607330322266,
      "learning_rate": 3.6593287621797186e-06,
      "loss": 0.8527,
      "step": 1860
    },
    {
      "epoch": 1.9524804177545692,
      "grad_norm": 6.412321090698242,
      "learning_rate": 3.623240707325875e-06,
      "loss": 0.8768,
      "step": 1870
    },
    {
      "epoch": 1.9629242819843342,
      "grad_norm": 6.859325408935547,
      "learning_rate": 3.587152652472032e-06,
      "loss": 0.8131,
      "step": 1880
    },
    {
      "epoch": 1.9733681462140993,
      "grad_norm": 4.910820484161377,
      "learning_rate": 3.5510645976181883e-06,
      "loss": 0.7232,
      "step": 1890
    },
    {
      "epoch": 1.9838120104438643,
      "grad_norm": 6.052480220794678,
      "learning_rate": 3.5149765427643452e-06,
      "loss": 0.7567,
      "step": 1900
    },
    {
      "epoch": 1.9942558746736292,
      "grad_norm": 6.609030246734619,
      "learning_rate": 3.4788884879105017e-06,
      "loss": 0.8219,
      "step": 1910
    },
    {
      "epoch": 2.004177545691906,
      "grad_norm": 4.539740562438965,
      "learning_rate": 3.4428004330566585e-06,
      "loss": 0.8498,
      "step": 1920
    },
    {
      "epoch": 2.014621409921671,
      "grad_norm": 5.981167316436768,
      "learning_rate": 3.406712378202815e-06,
      "loss": 0.7024,
      "step": 1930
    },
    {
      "epoch": 2.025065274151436,
      "grad_norm": 6.600665092468262,
      "learning_rate": 3.3706243233489714e-06,
      "loss": 0.7467,
      "step": 1940
    },
    {
      "epoch": 2.035509138381201,
      "grad_norm": 5.560609340667725,
      "learning_rate": 3.3345362684951283e-06,
      "loss": 0.7434,
      "step": 1950
    },
    {
      "epoch": 2.045953002610966,
      "grad_norm": 4.679533004760742,
      "learning_rate": 3.2984482136412848e-06,
      "loss": 0.7125,
      "step": 1960
    },
    {
      "epoch": 2.056396866840731,
      "grad_norm": 7.177086353302002,
      "learning_rate": 3.2623601587874416e-06,
      "loss": 0.6745,
      "step": 1970
    },
    {
      "epoch": 2.066840731070496,
      "grad_norm": 6.860986709594727,
      "learning_rate": 3.226272103933598e-06,
      "loss": 0.6814,
      "step": 1980
    },
    {
      "epoch": 2.077284595300261,
      "grad_norm": 8.40719223022461,
      "learning_rate": 3.190184049079755e-06,
      "loss": 0.7077,
      "step": 1990
    },
    {
      "epoch": 2.0877284595300263,
      "grad_norm": 5.830367088317871,
      "learning_rate": 3.1540959942259114e-06,
      "loss": 0.8435,
      "step": 2000
    },
    {
      "epoch": 2.0877284595300263,
      "eval_loss": 0.9753687381744385,
      "eval_runtime": 23.1964,
      "eval_samples_per_second": 36.687,
      "eval_steps_per_second": 4.613,
      "step": 2000
    },
    {
      "epoch": 2.0981723237597913,
      "grad_norm": 6.964289665222168,
      "learning_rate": 3.118007939372068e-06,
      "loss": 0.6964,
      "step": 2010
    },
    {
      "epoch": 2.108616187989556,
      "grad_norm": 6.764989852905273,
      "learning_rate": 3.0819198845182247e-06,
      "loss": 0.7107,
      "step": 2020
    },
    {
      "epoch": 2.119060052219321,
      "grad_norm": 7.992194652557373,
      "learning_rate": 3.045831829664381e-06,
      "loss": 0.7146,
      "step": 2030
    },
    {
      "epoch": 2.129503916449086,
      "grad_norm": 7.102138996124268,
      "learning_rate": 3.009743774810538e-06,
      "loss": 0.686,
      "step": 2040
    },
    {
      "epoch": 2.139947780678851,
      "grad_norm": 7.446751117706299,
      "learning_rate": 2.9736557199566945e-06,
      "loss": 0.7681,
      "step": 2050
    },
    {
      "epoch": 2.1503916449086162,
      "grad_norm": 7.091776371002197,
      "learning_rate": 2.9375676651028514e-06,
      "loss": 0.7674,
      "step": 2060
    },
    {
      "epoch": 2.1608355091383813,
      "grad_norm": 7.994192123413086,
      "learning_rate": 2.901479610249008e-06,
      "loss": 0.7187,
      "step": 2070
    },
    {
      "epoch": 2.1712793733681464,
      "grad_norm": 4.8329386711120605,
      "learning_rate": 2.8653915553951643e-06,
      "loss": 0.7501,
      "step": 2080
    },
    {
      "epoch": 2.1817232375979114,
      "grad_norm": 6.802753925323486,
      "learning_rate": 2.829303500541321e-06,
      "loss": 0.5658,
      "step": 2090
    },
    {
      "epoch": 2.192167101827676,
      "grad_norm": 7.07351541519165,
      "learning_rate": 2.7932154456874776e-06,
      "loss": 0.8106,
      "step": 2100
    },
    {
      "epoch": 2.202610966057441,
      "grad_norm": 6.761138916015625,
      "learning_rate": 2.7571273908336344e-06,
      "loss": 0.7434,
      "step": 2110
    },
    {
      "epoch": 2.213054830287206,
      "grad_norm": 4.220724582672119,
      "learning_rate": 2.721039335979791e-06,
      "loss": 0.7526,
      "step": 2120
    },
    {
      "epoch": 2.2234986945169712,
      "grad_norm": 6.14243745803833,
      "learning_rate": 2.6849512811259478e-06,
      "loss": 0.6486,
      "step": 2130
    },
    {
      "epoch": 2.2339425587467363,
      "grad_norm": 8.640827178955078,
      "learning_rate": 2.6488632262721042e-06,
      "loss": 0.5949,
      "step": 2140
    },
    {
      "epoch": 2.2443864229765014,
      "grad_norm": 6.576625823974609,
      "learning_rate": 2.6127751714182607e-06,
      "loss": 0.6923,
      "step": 2150
    },
    {
      "epoch": 2.2548302872062664,
      "grad_norm": 6.136504173278809,
      "learning_rate": 2.5766871165644175e-06,
      "loss": 0.6255,
      "step": 2160
    },
    {
      "epoch": 2.2652741514360315,
      "grad_norm": 7.2910847663879395,
      "learning_rate": 2.540599061710574e-06,
      "loss": 0.6843,
      "step": 2170
    },
    {
      "epoch": 2.275718015665796,
      "grad_norm": 6.936916351318359,
      "learning_rate": 2.504511006856731e-06,
      "loss": 0.6751,
      "step": 2180
    },
    {
      "epoch": 2.286161879895561,
      "grad_norm": 5.899853229522705,
      "learning_rate": 2.4684229520028873e-06,
      "loss": 0.6584,
      "step": 2190
    },
    {
      "epoch": 2.2966057441253263,
      "grad_norm": 4.891731262207031,
      "learning_rate": 2.4323348971490438e-06,
      "loss": 0.6373,
      "step": 2200
    },
    {
      "epoch": 2.2966057441253263,
      "eval_loss": 0.958043098449707,
      "eval_runtime": 23.1881,
      "eval_samples_per_second": 36.7,
      "eval_steps_per_second": 4.614,
      "step": 2200
    },
    {
      "epoch": 2.3070496083550913,
      "grad_norm": 6.206886291503906,
      "learning_rate": 2.3962468422952e-06,
      "loss": 0.6725,
      "step": 2210
    },
    {
      "epoch": 2.3174934725848564,
      "grad_norm": 4.663551330566406,
      "learning_rate": 2.360158787441357e-06,
      "loss": 0.5578,
      "step": 2220
    },
    {
      "epoch": 2.3279373368146214,
      "grad_norm": 6.175649166107178,
      "learning_rate": 2.3240707325875135e-06,
      "loss": 0.6835,
      "step": 2230
    },
    {
      "epoch": 2.3383812010443865,
      "grad_norm": 6.676774501800537,
      "learning_rate": 2.2879826777336704e-06,
      "loss": 0.7966,
      "step": 2240
    },
    {
      "epoch": 2.3488250652741516,
      "grad_norm": 8.847614288330078,
      "learning_rate": 2.251894622879827e-06,
      "loss": 0.7679,
      "step": 2250
    },
    {
      "epoch": 2.3592689295039166,
      "grad_norm": 6.491757869720459,
      "learning_rate": 2.2158065680259837e-06,
      "loss": 0.6274,
      "step": 2260
    },
    {
      "epoch": 2.3697127937336813,
      "grad_norm": 6.540876388549805,
      "learning_rate": 2.17971851317214e-06,
      "loss": 0.674,
      "step": 2270
    },
    {
      "epoch": 2.3801566579634463,
      "grad_norm": 7.067712306976318,
      "learning_rate": 2.1436304583182966e-06,
      "loss": 0.6716,
      "step": 2280
    },
    {
      "epoch": 2.3906005221932114,
      "grad_norm": 4.959332466125488,
      "learning_rate": 2.1075424034644535e-06,
      "loss": 0.6729,
      "step": 2290
    },
    {
      "epoch": 2.4010443864229765,
      "grad_norm": 4.016025066375732,
      "learning_rate": 2.07145434861061e-06,
      "loss": 0.6358,
      "step": 2300
    },
    {
      "epoch": 2.4114882506527415,
      "grad_norm": 4.044537544250488,
      "learning_rate": 2.035366293756767e-06,
      "loss": 0.6867,
      "step": 2310
    },
    {
      "epoch": 2.4219321148825066,
      "grad_norm": 4.88841438293457,
      "learning_rate": 1.9992782389029233e-06,
      "loss": 0.7561,
      "step": 2320
    },
    {
      "epoch": 2.4323759791122717,
      "grad_norm": 7.33749532699585,
      "learning_rate": 1.96319018404908e-06,
      "loss": 0.6579,
      "step": 2330
    },
    {
      "epoch": 2.4428198433420367,
      "grad_norm": 6.818521976470947,
      "learning_rate": 1.9271021291952366e-06,
      "loss": 0.7322,
      "step": 2340
    },
    {
      "epoch": 2.453263707571802,
      "grad_norm": 5.549405097961426,
      "learning_rate": 1.8910140743413932e-06,
      "loss": 0.634,
      "step": 2350
    },
    {
      "epoch": 2.4637075718015664,
      "grad_norm": 6.154874801635742,
      "learning_rate": 1.85492601948755e-06,
      "loss": 0.6044,
      "step": 2360
    },
    {
      "epoch": 2.4741514360313315,
      "grad_norm": 5.5303521156311035,
      "learning_rate": 1.8188379646337066e-06,
      "loss": 0.6833,
      "step": 2370
    },
    {
      "epoch": 2.4845953002610965,
      "grad_norm": 6.135169982910156,
      "learning_rate": 1.7827499097798632e-06,
      "loss": 0.7765,
      "step": 2380
    },
    {
      "epoch": 2.4950391644908616,
      "grad_norm": 7.397289752960205,
      "learning_rate": 1.7466618549260197e-06,
      "loss": 0.6748,
      "step": 2390
    },
    {
      "epoch": 2.5054830287206267,
      "grad_norm": 5.909689426422119,
      "learning_rate": 1.7105738000721763e-06,
      "loss": 0.6791,
      "step": 2400
    },
    {
      "epoch": 2.5054830287206267,
      "eval_loss": 0.955007791519165,
      "eval_runtime": 23.2476,
      "eval_samples_per_second": 36.606,
      "eval_steps_per_second": 4.603,
      "step": 2400
    },
    {
      "epoch": 2.5159268929503917,
      "grad_norm": 6.320558547973633,
      "learning_rate": 1.6744857452183328e-06,
      "loss": 0.6219,
      "step": 2410
    },
    {
      "epoch": 2.526370757180157,
      "grad_norm": 7.978168487548828,
      "learning_rate": 1.6383976903644894e-06,
      "loss": 0.8254,
      "step": 2420
    },
    {
      "epoch": 2.5368146214099214,
      "grad_norm": 6.5808210372924805,
      "learning_rate": 1.602309635510646e-06,
      "loss": 0.7243,
      "step": 2430
    },
    {
      "epoch": 2.547258485639687,
      "grad_norm": 4.769480228424072,
      "learning_rate": 1.5662215806568025e-06,
      "loss": 0.7497,
      "step": 2440
    },
    {
      "epoch": 2.5577023498694516,
      "grad_norm": 5.738780975341797,
      "learning_rate": 1.5301335258029592e-06,
      "loss": 0.7048,
      "step": 2450
    },
    {
      "epoch": 2.5681462140992166,
      "grad_norm": 5.658013343811035,
      "learning_rate": 1.4940454709491159e-06,
      "loss": 0.739,
      "step": 2460
    },
    {
      "epoch": 2.5785900783289817,
      "grad_norm": 6.587325096130371,
      "learning_rate": 1.4579574160952725e-06,
      "loss": 0.7076,
      "step": 2470
    },
    {
      "epoch": 2.5890339425587467,
      "grad_norm": 5.956645965576172,
      "learning_rate": 1.4218693612414292e-06,
      "loss": 0.6372,
      "step": 2480
    },
    {
      "epoch": 2.599477806788512,
      "grad_norm": 5.966655731201172,
      "learning_rate": 1.3857813063875859e-06,
      "loss": 0.6934,
      "step": 2490
    },
    {
      "epoch": 2.609921671018277,
      "grad_norm": 5.313653945922852,
      "learning_rate": 1.3496932515337425e-06,
      "loss": 0.7163,
      "step": 2500
    },
    {
      "epoch": 2.620365535248042,
      "grad_norm": 6.935596466064453,
      "learning_rate": 1.313605196679899e-06,
      "loss": 0.7104,
      "step": 2510
    },
    {
      "epoch": 2.6308093994778066,
      "grad_norm": 4.822442054748535,
      "learning_rate": 1.2775171418260556e-06,
      "loss": 0.6378,
      "step": 2520
    },
    {
      "epoch": 2.641253263707572,
      "grad_norm": 5.288422107696533,
      "learning_rate": 1.2414290869722123e-06,
      "loss": 0.6463,
      "step": 2530
    },
    {
      "epoch": 2.6516971279373367,
      "grad_norm": 6.668851852416992,
      "learning_rate": 1.205341032118369e-06,
      "loss": 0.7505,
      "step": 2540
    },
    {
      "epoch": 2.6621409921671018,
      "grad_norm": 5.71054220199585,
      "learning_rate": 1.1692529772645256e-06,
      "loss": 0.5856,
      "step": 2550
    },
    {
      "epoch": 2.672584856396867,
      "grad_norm": 6.284550666809082,
      "learning_rate": 1.1331649224106823e-06,
      "loss": 0.8122,
      "step": 2560
    },
    {
      "epoch": 2.683028720626632,
      "grad_norm": 8.781463623046875,
      "learning_rate": 1.097076867556839e-06,
      "loss": 0.7063,
      "step": 2570
    },
    {
      "epoch": 2.693472584856397,
      "grad_norm": 7.29454231262207,
      "learning_rate": 1.0609888127029954e-06,
      "loss": 0.7429,
      "step": 2580
    },
    {
      "epoch": 2.703916449086162,
      "grad_norm": 5.689371109008789,
      "learning_rate": 1.024900757849152e-06,
      "loss": 0.6658,
      "step": 2590
    },
    {
      "epoch": 2.714360313315927,
      "grad_norm": 7.286506175994873,
      "learning_rate": 9.888127029953087e-07,
      "loss": 0.748,
      "step": 2600
    },
    {
      "epoch": 2.714360313315927,
      "eval_loss": 0.9431054592132568,
      "eval_runtime": 23.1747,
      "eval_samples_per_second": 36.721,
      "eval_steps_per_second": 4.617,
      "step": 2600
    },
    {
      "epoch": 2.7248041775456917,
      "grad_norm": 5.635782241821289,
      "learning_rate": 9.527246481414652e-07,
      "loss": 0.673,
      "step": 2610
    },
    {
      "epoch": 2.7352480417754568,
      "grad_norm": 5.282413959503174,
      "learning_rate": 9.166365932876219e-07,
      "loss": 0.9037,
      "step": 2620
    },
    {
      "epoch": 2.745691906005222,
      "grad_norm": 7.922749042510986,
      "learning_rate": 8.805485384337785e-07,
      "loss": 0.6862,
      "step": 2630
    },
    {
      "epoch": 2.756135770234987,
      "grad_norm": 5.463962078094482,
      "learning_rate": 8.444604835799351e-07,
      "loss": 0.6,
      "step": 2640
    },
    {
      "epoch": 2.766579634464752,
      "grad_norm": 8.0007963180542,
      "learning_rate": 8.083724287260918e-07,
      "loss": 0.6688,
      "step": 2650
    },
    {
      "epoch": 2.777023498694517,
      "grad_norm": 7.617900371551514,
      "learning_rate": 7.722843738722483e-07,
      "loss": 0.7277,
      "step": 2660
    },
    {
      "epoch": 2.787467362924282,
      "grad_norm": 5.969784259796143,
      "learning_rate": 7.36196319018405e-07,
      "loss": 0.7085,
      "step": 2670
    },
    {
      "epoch": 2.7979112271540467,
      "grad_norm": 5.169407367706299,
      "learning_rate": 7.001082641645617e-07,
      "loss": 0.7344,
      "step": 2680
    },
    {
      "epoch": 2.8083550913838122,
      "grad_norm": 8.009687423706055,
      "learning_rate": 6.640202093107181e-07,
      "loss": 0.6457,
      "step": 2690
    },
    {
      "epoch": 2.818798955613577,
      "grad_norm": 7.4137187004089355,
      "learning_rate": 6.279321544568748e-07,
      "loss": 0.6416,
      "step": 2700
    },
    {
      "epoch": 2.829242819843342,
      "grad_norm": 5.35453462600708,
      "learning_rate": 5.918440996030314e-07,
      "loss": 0.7867,
      "step": 2710
    },
    {
      "epoch": 2.839686684073107,
      "grad_norm": 7.469908237457275,
      "learning_rate": 5.557560447491881e-07,
      "loss": 0.7955,
      "step": 2720
    },
    {
      "epoch": 2.850130548302872,
      "grad_norm": 6.326605319976807,
      "learning_rate": 5.196679898953446e-07,
      "loss": 0.6995,
      "step": 2730
    },
    {
      "epoch": 2.860574412532637,
      "grad_norm": 7.096553802490234,
      "learning_rate": 4.835799350415013e-07,
      "loss": 0.6625,
      "step": 2740
    },
    {
      "epoch": 2.871018276762402,
      "grad_norm": 5.128674507141113,
      "learning_rate": 4.474918801876579e-07,
      "loss": 0.6791,
      "step": 2750
    },
    {
      "epoch": 2.8814621409921672,
      "grad_norm": 6.457350254058838,
      "learning_rate": 4.1140382533381457e-07,
      "loss": 0.7586,
      "step": 2760
    },
    {
      "epoch": 2.891906005221932,
      "grad_norm": 6.231655597686768,
      "learning_rate": 3.753157704799711e-07,
      "loss": 0.677,
      "step": 2770
    },
    {
      "epoch": 2.9023498694516974,
      "grad_norm": 6.412544250488281,
      "learning_rate": 3.392277156261278e-07,
      "loss": 0.7449,
      "step": 2780
    },
    {
      "epoch": 2.912793733681462,
      "grad_norm": 3.725374698638916,
      "learning_rate": 3.031396607722844e-07,
      "loss": 0.5804,
      "step": 2790
    },
    {
      "epoch": 2.923237597911227,
      "grad_norm": 5.284286975860596,
      "learning_rate": 2.67051605918441e-07,
      "loss": 0.571,
      "step": 2800
    },
    {
      "epoch": 2.923237597911227,
      "eval_loss": 0.94057297706604,
      "eval_runtime": 23.1841,
      "eval_samples_per_second": 36.706,
      "eval_steps_per_second": 4.615,
      "step": 2800
    }
  ],
  "logging_steps": 10,
  "max_steps": 2871,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 200,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 3.029626779598848e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}