{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 7.9963586709148835,
  "eval_steps": 500,
  "global_step": 4392,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.018206645425580335,
      "grad_norm": 15.35500484249196,
      "learning_rate": 5e-06,
      "loss": 1.0839,
      "step": 10
    },
    {
      "epoch": 0.03641329085116067,
      "grad_norm": 2.618926418245599,
      "learning_rate": 5e-06,
      "loss": 0.9627,
      "step": 20
    },
    {
      "epoch": 0.05461993627674101,
      "grad_norm": 2.080636783209041,
      "learning_rate": 5e-06,
      "loss": 0.9308,
      "step": 30
    },
    {
      "epoch": 0.07282658170232134,
      "grad_norm": 1.436134442652753,
      "learning_rate": 5e-06,
      "loss": 0.897,
      "step": 40
    },
    {
      "epoch": 0.09103322712790168,
      "grad_norm": 1.581598449271548,
      "learning_rate": 5e-06,
      "loss": 0.8783,
      "step": 50
    },
    {
      "epoch": 0.10923987255348203,
      "grad_norm": 1.3764881165967209,
      "learning_rate": 5e-06,
      "loss": 0.8594,
      "step": 60
    },
    {
      "epoch": 0.12744651797906237,
      "grad_norm": 0.8508264971874785,
      "learning_rate": 5e-06,
      "loss": 0.8476,
      "step": 70
    },
    {
      "epoch": 0.14565316340464268,
      "grad_norm": 0.9826078435096594,
      "learning_rate": 5e-06,
      "loss": 0.8351,
      "step": 80
    },
    {
      "epoch": 0.16385980883022302,
      "grad_norm": 0.7948332220104163,
      "learning_rate": 5e-06,
      "loss": 0.8261,
      "step": 90
    },
    {
      "epoch": 0.18206645425580337,
      "grad_norm": 0.7335767518254585,
      "learning_rate": 5e-06,
      "loss": 0.8185,
      "step": 100
    },
    {
      "epoch": 0.2002730996813837,
      "grad_norm": 1.0534204936405038,
      "learning_rate": 5e-06,
      "loss": 0.8204,
      "step": 110
    },
    {
      "epoch": 0.21847974510696405,
      "grad_norm": 0.847321277840478,
      "learning_rate": 5e-06,
      "loss": 0.8108,
      "step": 120
    },
    {
      "epoch": 0.23668639053254437,
      "grad_norm": 0.6210895897293146,
      "learning_rate": 5e-06,
      "loss": 0.8084,
      "step": 130
    },
    {
      "epoch": 0.25489303595812474,
      "grad_norm": 0.5781940572538745,
      "learning_rate": 5e-06,
      "loss": 0.8103,
      "step": 140
    },
    {
      "epoch": 0.27309968138370505,
      "grad_norm": 0.9063984151725509,
      "learning_rate": 5e-06,
      "loss": 0.8077,
      "step": 150
    },
    {
      "epoch": 0.29130632680928537,
      "grad_norm": 0.6571941221537146,
      "learning_rate": 5e-06,
      "loss": 0.799,
      "step": 160
    },
    {
      "epoch": 0.30951297223486574,
      "grad_norm": 0.5952826448438556,
      "learning_rate": 5e-06,
      "loss": 0.7999,
      "step": 170
    },
    {
      "epoch": 0.32771961766044605,
      "grad_norm": 1.4601461457218337,
      "learning_rate": 5e-06,
      "loss": 0.8004,
      "step": 180
    },
    {
      "epoch": 0.3459262630860264,
      "grad_norm": 1.3162465370008012,
      "learning_rate": 5e-06,
      "loss": 0.7973,
      "step": 190
    },
    {
      "epoch": 0.36413290851160673,
      "grad_norm": 1.2972378846813408,
      "learning_rate": 5e-06,
      "loss": 0.7959,
      "step": 200
    },
    {
      "epoch": 0.38233955393718705,
      "grad_norm": 0.9152791184028909,
      "learning_rate": 5e-06,
      "loss": 0.7892,
      "step": 210
    },
    {
      "epoch": 0.4005461993627674,
      "grad_norm": 0.823555803312372,
      "learning_rate": 5e-06,
      "loss": 0.7932,
      "step": 220
    },
    {
      "epoch": 0.41875284478834773,
      "grad_norm": 0.7145315436546384,
      "learning_rate": 5e-06,
      "loss": 0.7848,
      "step": 230
    },
    {
      "epoch": 0.4369594902139281,
      "grad_norm": 0.5776875953876701,
      "learning_rate": 5e-06,
      "loss": 0.791,
      "step": 240
    },
    {
      "epoch": 0.4551661356395084,
      "grad_norm": 0.6846063811362885,
      "learning_rate": 5e-06,
      "loss": 0.7875,
      "step": 250
    },
    {
      "epoch": 0.47337278106508873,
      "grad_norm": 0.7232578149621245,
      "learning_rate": 5e-06,
      "loss": 0.7843,
      "step": 260
    },
    {
      "epoch": 0.4915794264906691,
      "grad_norm": 0.6962364276167124,
      "learning_rate": 5e-06,
      "loss": 0.7839,
      "step": 270
    },
    {
      "epoch": 0.5097860719162495,
      "grad_norm": 0.8881236192032435,
      "learning_rate": 5e-06,
      "loss": 0.779,
      "step": 280
    },
    {
      "epoch": 0.5279927173418297,
      "grad_norm": 0.6192460064199938,
      "learning_rate": 5e-06,
      "loss": 0.7821,
      "step": 290
    },
    {
      "epoch": 0.5461993627674101,
      "grad_norm": 0.7429540921229376,
      "learning_rate": 5e-06,
      "loss": 0.7771,
      "step": 300
    },
    {
      "epoch": 0.5644060081929905,
      "grad_norm": 0.5678182946639223,
      "learning_rate": 5e-06,
      "loss": 0.7837,
      "step": 310
    },
    {
      "epoch": 0.5826126536185707,
      "grad_norm": 0.7927883549153515,
      "learning_rate": 5e-06,
      "loss": 0.7795,
      "step": 320
    },
    {
      "epoch": 0.6008192990441511,
      "grad_norm": 0.6630873675584549,
      "learning_rate": 5e-06,
      "loss": 0.7788,
      "step": 330
    },
    {
      "epoch": 0.6190259444697315,
      "grad_norm": 0.8821281227865143,
      "learning_rate": 5e-06,
      "loss": 0.7817,
      "step": 340
    },
    {
      "epoch": 0.6372325898953118,
      "grad_norm": 0.5862391322505137,
      "learning_rate": 5e-06,
      "loss": 0.7818,
      "step": 350
    },
    {
      "epoch": 0.6554392353208921,
      "grad_norm": 0.7933124674797363,
      "learning_rate": 5e-06,
      "loss": 0.7714,
      "step": 360
    },
    {
      "epoch": 0.6736458807464725,
      "grad_norm": 0.5912540694609474,
      "learning_rate": 5e-06,
      "loss": 0.7732,
      "step": 370
    },
    {
      "epoch": 0.6918525261720528,
      "grad_norm": 0.5222298817170612,
      "learning_rate": 5e-06,
      "loss": 0.7697,
      "step": 380
    },
    {
      "epoch": 0.7100591715976331,
      "grad_norm": 0.6154935564611252,
      "learning_rate": 5e-06,
      "loss": 0.7716,
      "step": 390
    },
    {
      "epoch": 0.7282658170232135,
      "grad_norm": 0.7305900161607543,
      "learning_rate": 5e-06,
      "loss": 0.7708,
      "step": 400
    },
    {
      "epoch": 0.7464724624487938,
      "grad_norm": 0.7798387910594561,
      "learning_rate": 5e-06,
      "loss": 0.7724,
      "step": 410
    },
    {
      "epoch": 0.7646791078743741,
      "grad_norm": 1.100346410427369,
      "learning_rate": 5e-06,
      "loss": 0.7716,
      "step": 420
    },
    {
      "epoch": 0.7828857532999545,
      "grad_norm": 0.657718484418182,
      "learning_rate": 5e-06,
      "loss": 0.7666,
      "step": 430
    },
    {
      "epoch": 0.8010923987255348,
      "grad_norm": 0.6103104885939142,
      "learning_rate": 5e-06,
      "loss": 0.765,
      "step": 440
    },
    {
      "epoch": 0.8192990441511152,
      "grad_norm": 0.6884438533501283,
      "learning_rate": 5e-06,
      "loss": 0.7699,
      "step": 450
    },
    {
      "epoch": 0.8375056895766955,
      "grad_norm": 0.6270481654771743,
      "learning_rate": 5e-06,
      "loss": 0.767,
      "step": 460
    },
    {
      "epoch": 0.8557123350022758,
      "grad_norm": 0.5861612234613632,
      "learning_rate": 5e-06,
      "loss": 0.7645,
      "step": 470
    },
    {
      "epoch": 0.8739189804278562,
      "grad_norm": 0.586876291200088,
      "learning_rate": 5e-06,
      "loss": 0.7686,
      "step": 480
    },
    {
      "epoch": 0.8921256258534365,
      "grad_norm": 0.7466830111479166,
      "learning_rate": 5e-06,
      "loss": 0.7688,
      "step": 490
    },
    {
      "epoch": 0.9103322712790168,
      "grad_norm": 0.6792188281050604,
      "learning_rate": 5e-06,
      "loss": 0.7685,
      "step": 500
    },
    {
      "epoch": 0.9285389167045972,
      "grad_norm": 0.6846050616906328,
      "learning_rate": 5e-06,
      "loss": 0.7616,
      "step": 510
    },
    {
      "epoch": 0.9467455621301775,
      "grad_norm": 0.7324962552104101,
      "learning_rate": 5e-06,
      "loss": 0.762,
      "step": 520
    },
    {
      "epoch": 0.9649522075557578,
      "grad_norm": 0.6893904439545421,
      "learning_rate": 5e-06,
      "loss": 0.7664,
      "step": 530
    },
    {
      "epoch": 0.9831588529813382,
      "grad_norm": 0.6529796330088168,
      "learning_rate": 5e-06,
      "loss": 0.7631,
      "step": 540
    },
    {
      "epoch": 0.9995448338643604,
      "eval_loss": 0.7638739347457886,
      "eval_runtime": 379.9866,
      "eval_samples_per_second": 38.951,
      "eval_steps_per_second": 0.611,
      "step": 549
    },
    {
      "epoch": 1.0013654984069185,
      "grad_norm": 1.2061555179450747,
      "learning_rate": 5e-06,
      "loss": 0.812,
      "step": 550
    },
    {
      "epoch": 1.019572143832499,
      "grad_norm": 0.8129440249983992,
      "learning_rate": 5e-06,
      "loss": 0.7222,
      "step": 560
    },
    {
      "epoch": 1.0377787892580792,
      "grad_norm": 0.6593013104808815,
      "learning_rate": 5e-06,
      "loss": 0.7205,
      "step": 570
    },
    {
      "epoch": 1.0559854346836595,
      "grad_norm": 0.6107941470930411,
      "learning_rate": 5e-06,
      "loss": 0.7205,
      "step": 580
    },
    {
      "epoch": 1.07419208010924,
      "grad_norm": 0.8276670919993867,
      "learning_rate": 5e-06,
      "loss": 0.7144,
      "step": 590
    },
    {
      "epoch": 1.0923987255348202,
      "grad_norm": 0.6253185542660747,
      "learning_rate": 5e-06,
      "loss": 0.7136,
      "step": 600
    },
    {
      "epoch": 1.1106053709604005,
      "grad_norm": 0.6108900775704884,
      "learning_rate": 5e-06,
      "loss": 0.7189,
      "step": 610
    },
    {
      "epoch": 1.128812016385981,
      "grad_norm": 0.8988785705551109,
      "learning_rate": 5e-06,
      "loss": 0.7122,
      "step": 620
    },
    {
      "epoch": 1.1470186618115612,
      "grad_norm": 0.616453203309773,
      "learning_rate": 5e-06,
      "loss": 0.7138,
      "step": 630
    },
    {
      "epoch": 1.1652253072371415,
      "grad_norm": 0.8816478964333492,
      "learning_rate": 5e-06,
      "loss": 0.7174,
      "step": 640
    },
    {
      "epoch": 1.183431952662722,
      "grad_norm": 0.6581015672880766,
      "learning_rate": 5e-06,
      "loss": 0.7183,
      "step": 650
    },
    {
      "epoch": 1.2016385980883022,
      "grad_norm": 0.6253002121195789,
      "learning_rate": 5e-06,
      "loss": 0.7169,
      "step": 660
    },
    {
      "epoch": 1.2198452435138827,
      "grad_norm": 0.7483871477416271,
      "learning_rate": 5e-06,
      "loss": 0.7136,
      "step": 670
    },
    {
      "epoch": 1.238051888939463,
      "grad_norm": 0.6941445361796307,
      "learning_rate": 5e-06,
      "loss": 0.7114,
      "step": 680
    },
    {
      "epoch": 1.2562585343650432,
      "grad_norm": 0.6143667364149711,
      "learning_rate": 5e-06,
      "loss": 0.7185,
      "step": 690
    },
    {
      "epoch": 1.2744651797906235,
      "grad_norm": 0.6592963882014911,
      "learning_rate": 5e-06,
      "loss": 0.7166,
      "step": 700
    },
    {
      "epoch": 1.292671825216204,
      "grad_norm": 0.6189597729803721,
      "learning_rate": 5e-06,
      "loss": 0.7158,
      "step": 710
    },
    {
      "epoch": 1.3108784706417842,
      "grad_norm": 0.6920980650430207,
      "learning_rate": 5e-06,
      "loss": 0.7144,
      "step": 720
    },
    {
      "epoch": 1.3290851160673647,
      "grad_norm": 0.49530512173626284,
      "learning_rate": 5e-06,
      "loss": 0.718,
      "step": 730
    },
    {
      "epoch": 1.347291761492945,
      "grad_norm": 0.8064053486625922,
      "learning_rate": 5e-06,
      "loss": 0.7131,
      "step": 740
    },
    {
      "epoch": 1.3654984069185252,
      "grad_norm": 0.5821998941102782,
      "learning_rate": 5e-06,
      "loss": 0.7136,
      "step": 750
    },
    {
      "epoch": 1.3837050523441057,
      "grad_norm": 0.5735380427187717,
      "learning_rate": 5e-06,
      "loss": 0.7154,
      "step": 760
    },
    {
      "epoch": 1.401911697769686,
      "grad_norm": 0.5961452096313207,
      "learning_rate": 5e-06,
      "loss": 0.7148,
      "step": 770
    },
    {
      "epoch": 1.4201183431952662,
      "grad_norm": 0.6364200555389256,
      "learning_rate": 5e-06,
      "loss": 0.7152,
      "step": 780
    },
    {
      "epoch": 1.4383249886208467,
      "grad_norm": 0.618866609217087,
      "learning_rate": 5e-06,
      "loss": 0.7139,
      "step": 790
    },
    {
      "epoch": 1.456531634046427,
      "grad_norm": 0.5153453443480923,
      "learning_rate": 5e-06,
      "loss": 0.7169,
      "step": 800
    },
    {
      "epoch": 1.4747382794720072,
      "grad_norm": 0.5921562610346923,
      "learning_rate": 5e-06,
      "loss": 0.7157,
      "step": 810
    },
    {
      "epoch": 1.4929449248975877,
      "grad_norm": 0.6864315768684789,
      "learning_rate": 5e-06,
      "loss": 0.7186,
      "step": 820
    },
    {
      "epoch": 1.511151570323168,
      "grad_norm": 0.6718756873067311,
      "learning_rate": 5e-06,
      "loss": 0.713,
      "step": 830
    },
    {
      "epoch": 1.5293582157487484,
      "grad_norm": 0.5292103286821733,
      "learning_rate": 5e-06,
      "loss": 0.7121,
      "step": 840
    },
    {
      "epoch": 1.5475648611743287,
      "grad_norm": 0.5807337157950755,
      "learning_rate": 5e-06,
      "loss": 0.7161,
      "step": 850
    },
    {
      "epoch": 1.565771506599909,
      "grad_norm": 0.5725693174193143,
      "learning_rate": 5e-06,
      "loss": 0.7144,
      "step": 860
    },
    {
      "epoch": 1.5839781520254892,
      "grad_norm": 0.6693290134206946,
      "learning_rate": 5e-06,
      "loss": 0.7137,
      "step": 870
    },
    {
      "epoch": 1.6021847974510697,
      "grad_norm": 0.5960183936149324,
      "learning_rate": 5e-06,
      "loss": 0.718,
      "step": 880
    },
    {
      "epoch": 1.62039144287665,
      "grad_norm": 0.6717348604009659,
      "learning_rate": 5e-06,
      "loss": 0.7107,
      "step": 890
    },
    {
      "epoch": 1.6385980883022304,
      "grad_norm": 0.6147573815012755,
      "learning_rate": 5e-06,
      "loss": 0.7101,
      "step": 900
    },
    {
      "epoch": 1.6568047337278107,
      "grad_norm": 0.5451936467703803,
      "learning_rate": 5e-06,
      "loss": 0.7128,
      "step": 910
    },
    {
      "epoch": 1.675011379153391,
      "grad_norm": 0.7294513986677604,
      "learning_rate": 5e-06,
      "loss": 0.7104,
      "step": 920
    },
    {
      "epoch": 1.6932180245789712,
      "grad_norm": 0.6756274493055726,
      "learning_rate": 5e-06,
      "loss": 0.7115,
      "step": 930
    },
    {
      "epoch": 1.7114246700045517,
      "grad_norm": 0.6503407532587065,
      "learning_rate": 5e-06,
      "loss": 0.7105,
      "step": 940
    },
    {
      "epoch": 1.7296313154301322,
      "grad_norm": 0.5369254794383076,
      "learning_rate": 5e-06,
      "loss": 0.7116,
      "step": 950
    },
    {
      "epoch": 1.7478379608557124,
      "grad_norm": 0.6658976879506388,
      "learning_rate": 5e-06,
      "loss": 0.7108,
      "step": 960
    },
    {
      "epoch": 1.7660446062812927,
      "grad_norm": 0.5443862825931207,
      "learning_rate": 5e-06,
      "loss": 0.7107,
      "step": 970
    },
    {
      "epoch": 1.784251251706873,
      "grad_norm": 0.6571290419201206,
      "learning_rate": 5e-06,
      "loss": 0.7118,
      "step": 980
    },
    {
      "epoch": 1.8024578971324532,
      "grad_norm": 0.543057907363159,
      "learning_rate": 5e-06,
      "loss": 0.7109,
      "step": 990
    },
    {
      "epoch": 1.8206645425580337,
      "grad_norm": 0.5863888559339765,
      "learning_rate": 5e-06,
      "loss": 0.7128,
      "step": 1000
    },
    {
      "epoch": 1.8388711879836142,
      "grad_norm": 0.7512763814183313,
      "learning_rate": 5e-06,
      "loss": 0.7129,
      "step": 1010
    },
    {
      "epoch": 1.8570778334091944,
      "grad_norm": 0.6875839517913319,
      "learning_rate": 5e-06,
      "loss": 0.7163,
      "step": 1020
    },
    {
      "epoch": 1.8752844788347747,
      "grad_norm": 0.5392039305234562,
      "learning_rate": 5e-06,
      "loss": 0.712,
      "step": 1030
    },
    {
      "epoch": 1.893491124260355,
      "grad_norm": 0.5715846033792595,
      "learning_rate": 5e-06,
      "loss": 0.7106,
      "step": 1040
    },
    {
      "epoch": 1.9116977696859354,
      "grad_norm": 0.554967637717086,
      "learning_rate": 5e-06,
      "loss": 0.7135,
      "step": 1050
    },
    {
      "epoch": 1.9299044151115157,
      "grad_norm": 0.6064174196129051,
      "learning_rate": 5e-06,
      "loss": 0.7113,
      "step": 1060
    },
    {
      "epoch": 1.9481110605370962,
      "grad_norm": 0.57876695036443,
      "learning_rate": 5e-06,
      "loss": 0.7133,
      "step": 1070
    },
    {
      "epoch": 1.9663177059626764,
      "grad_norm": 0.6168643227515411,
      "learning_rate": 5e-06,
      "loss": 0.7107,
      "step": 1080
    },
    {
      "epoch": 1.9845243513882567,
      "grad_norm": 0.5918567965220687,
      "learning_rate": 5e-06,
      "loss": 0.7148,
      "step": 1090
    },
    {
      "epoch": 1.9990896677287209,
      "eval_loss": 0.7494105696678162,
      "eval_runtime": 371.303,
      "eval_samples_per_second": 39.862,
      "eval_steps_per_second": 0.625,
      "step": 1098
    },
    {
      "epoch": 2.002730996813837,
      "grad_norm": 0.7112677344136977,
      "learning_rate": 5e-06,
      "loss": 0.7534,
      "step": 1100
    },
    {
      "epoch": 2.020937642239417,
      "grad_norm": 0.7657125714853397,
      "learning_rate": 5e-06,
      "loss": 0.6594,
      "step": 1110
    },
    {
      "epoch": 2.039144287664998,
      "grad_norm": 0.7269679004826854,
      "learning_rate": 5e-06,
      "loss": 0.6665,
      "step": 1120
    },
    {
      "epoch": 2.057350933090578,
      "grad_norm": 0.5908930298865153,
      "learning_rate": 5e-06,
      "loss": 0.6641,
      "step": 1130
    },
    {
      "epoch": 2.0755575785161584,
      "grad_norm": 0.6685407405530368,
      "learning_rate": 5e-06,
      "loss": 0.6611,
      "step": 1140
    },
    {
      "epoch": 2.0937642239417387,
      "grad_norm": 0.6835975141817704,
      "learning_rate": 5e-06,
      "loss": 0.6631,
      "step": 1150
    },
    {
      "epoch": 2.111970869367319,
      "grad_norm": 0.6561709037357517,
      "learning_rate": 5e-06,
      "loss": 0.6631,
      "step": 1160
    },
    {
      "epoch": 2.1301775147928996,
      "grad_norm": 0.9447854836936921,
      "learning_rate": 5e-06,
      "loss": 0.6641,
      "step": 1170
    },
    {
      "epoch": 2.14838416021848,
      "grad_norm": 0.6554564933004847,
      "learning_rate": 5e-06,
      "loss": 0.6644,
      "step": 1180
    },
    {
      "epoch": 2.16659080564406,
      "grad_norm": 1.0127997383891072,
      "learning_rate": 5e-06,
      "loss": 0.6711,
      "step": 1190
    },
    {
      "epoch": 2.1847974510696404,
      "grad_norm": 0.7206567370516284,
      "learning_rate": 5e-06,
      "loss": 0.6668,
      "step": 1200
    },
    {
      "epoch": 2.2030040964952207,
      "grad_norm": 0.655937928590562,
      "learning_rate": 5e-06,
      "loss": 0.6717,
      "step": 1210
    },
    {
      "epoch": 2.221210741920801,
      "grad_norm": 0.8360483477674778,
      "learning_rate": 5e-06,
      "loss": 0.6689,
      "step": 1220
    },
    {
      "epoch": 2.2394173873463816,
      "grad_norm": 0.6092424114015346,
      "learning_rate": 5e-06,
      "loss": 0.6628,
      "step": 1230
    },
    {
      "epoch": 2.257624032771962,
      "grad_norm": 0.6406882124202334,
      "learning_rate": 5e-06,
      "loss": 0.6651,
      "step": 1240
    },
    {
      "epoch": 2.275830678197542,
      "grad_norm": 0.7614678401383996,
      "learning_rate": 5e-06,
      "loss": 0.6677,
      "step": 1250
    },
    {
      "epoch": 2.2940373236231224,
      "grad_norm": 0.7284934278440872,
      "learning_rate": 5e-06,
      "loss": 0.6649,
      "step": 1260
    },
    {
      "epoch": 2.3122439690487027,
      "grad_norm": 0.6736018159497253,
      "learning_rate": 5e-06,
      "loss": 0.6661,
      "step": 1270
    },
    {
      "epoch": 2.330450614474283,
      "grad_norm": 0.682991318579073,
      "learning_rate": 5e-06,
      "loss": 0.6706,
      "step": 1280
    },
    {
      "epoch": 2.3486572598998636,
      "grad_norm": 0.6067823760231827,
      "learning_rate": 5e-06,
      "loss": 0.6642,
      "step": 1290
    },
    {
      "epoch": 2.366863905325444,
      "grad_norm": 0.6016632433423775,
      "learning_rate": 5e-06,
      "loss": 0.6675,
      "step": 1300
    },
    {
      "epoch": 2.385070550751024,
      "grad_norm": 0.6784572170158043,
      "learning_rate": 5e-06,
      "loss": 0.6658,
      "step": 1310
    },
    {
      "epoch": 2.4032771961766044,
      "grad_norm": 0.6804813510854688,
      "learning_rate": 5e-06,
      "loss": 0.6693,
      "step": 1320
    },
    {
      "epoch": 2.4214838416021847,
      "grad_norm": 0.6667510974726526,
      "learning_rate": 5e-06,
      "loss": 0.6667,
      "step": 1330
    },
    {
      "epoch": 2.4396904870277654,
      "grad_norm": 0.5712701093393895,
      "learning_rate": 5e-06,
      "loss": 0.6693,
      "step": 1340
    },
    {
      "epoch": 2.4578971324533456,
      "grad_norm": 0.5495878729178928,
      "learning_rate": 5e-06,
      "loss": 0.6691,
      "step": 1350
    },
    {
      "epoch": 2.476103777878926,
      "grad_norm": 0.6863248077276337,
      "learning_rate": 5e-06,
      "loss": 0.6697,
      "step": 1360
    },
    {
      "epoch": 2.494310423304506,
      "grad_norm": 0.6281725692693549,
      "learning_rate": 5e-06,
      "loss": 0.6659,
      "step": 1370
    },
    {
      "epoch": 2.5125170687300864,
      "grad_norm": 0.584114706811986,
      "learning_rate": 5e-06,
      "loss": 0.6689,
      "step": 1380
    },
    {
      "epoch": 2.5307237141556667,
      "grad_norm": 0.8069687501187531,
      "learning_rate": 5e-06,
      "loss": 0.6713,
      "step": 1390
    },
    {
      "epoch": 2.548930359581247,
      "grad_norm": 0.5837408246005996,
      "learning_rate": 5e-06,
      "loss": 0.6702,
      "step": 1400
    },
    {
      "epoch": 2.5671370050068276,
      "grad_norm": 0.6602631718487662,
      "learning_rate": 5e-06,
      "loss": 0.6729,
      "step": 1410
    },
    {
      "epoch": 2.585343650432408,
      "grad_norm": 0.5751520967539265,
      "learning_rate": 5e-06,
      "loss": 0.6727,
      "step": 1420
    },
    {
      "epoch": 2.603550295857988,
      "grad_norm": 0.5753704561331634,
      "learning_rate": 5e-06,
      "loss": 0.6638,
      "step": 1430
    },
    {
      "epoch": 2.6217569412835684,
      "grad_norm": 0.5947582650600889,
      "learning_rate": 5e-06,
      "loss": 0.6687,
      "step": 1440
    },
    {
      "epoch": 2.6399635867091487,
      "grad_norm": 0.6421455226025955,
      "learning_rate": 5e-06,
      "loss": 0.6673,
      "step": 1450
    },
    {
      "epoch": 2.6581702321347294,
      "grad_norm": 0.625971391993565,
      "learning_rate": 5e-06,
      "loss": 0.6723,
      "step": 1460
    },
    {
      "epoch": 2.6763768775603096,
      "grad_norm": 0.6230286833589651,
      "learning_rate": 5e-06,
      "loss": 0.6697,
      "step": 1470
    },
    {
      "epoch": 2.69458352298589,
      "grad_norm": 0.6704340829719386,
      "learning_rate": 5e-06,
      "loss": 0.6726,
      "step": 1480
    },
    {
      "epoch": 2.71279016841147,
      "grad_norm": 0.7115047592117129,
      "learning_rate": 5e-06,
      "loss": 0.6697,
      "step": 1490
    },
    {
      "epoch": 2.7309968138370504,
      "grad_norm": 0.5348550201712026,
      "learning_rate": 5e-06,
      "loss": 0.6713,
      "step": 1500
    },
    {
      "epoch": 2.749203459262631,
      "grad_norm": 0.6018887550964896,
      "learning_rate": 5e-06,
      "loss": 0.6713,
      "step": 1510
    },
    {
      "epoch": 2.7674101046882114,
      "grad_norm": 0.5829651029728877,
      "learning_rate": 5e-06,
      "loss": 0.6711,
      "step": 1520
    },
    {
      "epoch": 2.7856167501137916,
      "grad_norm": 0.7257452058424095,
      "learning_rate": 5e-06,
      "loss": 0.6679,
      "step": 1530
    },
    {
      "epoch": 2.803823395539372,
      "grad_norm": 0.7324462033376344,
      "learning_rate": 5e-06,
      "loss": 0.6705,
      "step": 1540
    },
    {
      "epoch": 2.822030040964952,
      "grad_norm": 0.6518477868742856,
      "learning_rate": 5e-06,
      "loss": 0.6714,
      "step": 1550
    },
    {
      "epoch": 2.8402366863905324,
      "grad_norm": 0.5915504419734868,
      "learning_rate": 5e-06,
      "loss": 0.6678,
      "step": 1560
    },
    {
      "epoch": 2.8584433318161127,
      "grad_norm": 0.5596845389686367,
      "learning_rate": 5e-06,
      "loss": 0.6738,
      "step": 1570
    },
    {
      "epoch": 2.8766499772416934,
      "grad_norm": 0.6211271364874157,
      "learning_rate": 5e-06,
      "loss": 0.6682,
      "step": 1580
    },
    {
      "epoch": 2.8948566226672736,
      "grad_norm": 0.6029083750419233,
      "learning_rate": 5e-06,
      "loss": 0.6689,
      "step": 1590
    },
    {
      "epoch": 2.913063268092854,
      "grad_norm": 0.6997154043188799,
      "learning_rate": 5e-06,
      "loss": 0.6654,
      "step": 1600
    },
    {
      "epoch": 2.931269913518434,
      "grad_norm": 0.6561810265940334,
      "learning_rate": 5e-06,
      "loss": 0.6648,
      "step": 1610
    },
    {
      "epoch": 2.9494765589440144,
      "grad_norm": 0.5949195869956836,
      "learning_rate": 5e-06,
      "loss": 0.6674,
      "step": 1620
    },
    {
      "epoch": 2.967683204369595,
      "grad_norm": 0.6572460988263176,
      "learning_rate": 5e-06,
      "loss": 0.6748,
      "step": 1630
    },
    {
      "epoch": 2.9858898497951754,
      "grad_norm": 0.6466132865038215,
      "learning_rate": 5e-06,
      "loss": 0.6707,
      "step": 1640
    },
    {
      "epoch": 2.9986345015930813,
      "eval_loss": 0.7486923933029175,
      "eval_runtime": 371.6394,
      "eval_samples_per_second": 39.826,
      "eval_steps_per_second": 0.624,
      "step": 1647
    },
    {
      "epoch": 3.0040964952207556,
      "grad_norm": 0.9066558799698319,
      "learning_rate": 5e-06,
      "loss": 0.7123,
      "step": 1650
    },
    {
      "epoch": 3.022303140646336,
      "grad_norm": 0.6985831692717679,
      "learning_rate": 5e-06,
      "loss": 0.6192,
      "step": 1660
    },
    {
      "epoch": 3.040509786071916,
      "grad_norm": 0.7297757031418619,
      "learning_rate": 5e-06,
      "loss": 0.6192,
      "step": 1670
    },
    {
      "epoch": 3.0587164314974964,
      "grad_norm": 0.7384505423128972,
      "learning_rate": 5e-06,
      "loss": 0.6218,
      "step": 1680
    },
    {
      "epoch": 3.076923076923077,
      "grad_norm": 0.7273583338941372,
      "learning_rate": 5e-06,
      "loss": 0.6211,
      "step": 1690
    },
    {
      "epoch": 3.0951297223486574,
      "grad_norm": 0.7471474278740236,
      "learning_rate": 5e-06,
      "loss": 0.621,
      "step": 1700
    },
    {
      "epoch": 3.1133363677742376,
      "grad_norm": 0.6193936409655777,
      "learning_rate": 5e-06,
      "loss": 0.6201,
      "step": 1710
    },
    {
      "epoch": 3.131543013199818,
      "grad_norm": 0.6710258448973723,
      "learning_rate": 5e-06,
      "loss": 0.6224,
      "step": 1720
    },
    {
      "epoch": 3.149749658625398,
      "grad_norm": 0.7409460472770332,
      "learning_rate": 5e-06,
      "loss": 0.625,
      "step": 1730
    },
    {
      "epoch": 3.1679563040509784,
      "grad_norm": 0.709501536492518,
      "learning_rate": 5e-06,
      "loss": 0.6204,
      "step": 1740
    },
    {
      "epoch": 3.186162949476559,
      "grad_norm": 0.7649819943694469,
      "learning_rate": 5e-06,
      "loss": 0.6163,
      "step": 1750
    },
    {
      "epoch": 3.2043695949021394,
      "grad_norm": 0.7173580779194709,
      "learning_rate": 5e-06,
      "loss": 0.6234,
      "step": 1760
    },
    {
      "epoch": 3.2225762403277196,
      "grad_norm": 0.6457493603144049,
      "learning_rate": 5e-06,
      "loss": 0.6213,
      "step": 1770
    },
    {
      "epoch": 3.2407828857533,
      "grad_norm": 0.6074440281419403,
      "learning_rate": 5e-06,
      "loss": 0.6229,
      "step": 1780
    },
    {
      "epoch": 3.25898953117888,
      "grad_norm": 0.7025871444184026,
      "learning_rate": 5e-06,
      "loss": 0.621,
      "step": 1790
    },
    {
      "epoch": 3.277196176604461,
      "grad_norm": 0.7659855454008233,
      "learning_rate": 5e-06,
      "loss": 0.6244,
      "step": 1800
    },
    {
      "epoch": 3.295402822030041,
      "grad_norm": 0.6532860177479579,
      "learning_rate": 5e-06,
      "loss": 0.6229,
      "step": 1810
    },
    {
      "epoch": 3.3136094674556213,
      "grad_norm": 0.5432923070386573,
      "learning_rate": 5e-06,
      "loss": 0.6226,
      "step": 1820
    },
    {
      "epoch": 3.3318161128812016,
      "grad_norm": 0.6927506071213173,
      "learning_rate": 5e-06,
      "loss": 0.6241,
      "step": 1830
    },
    {
      "epoch": 3.350022758306782,
      "grad_norm": 0.6319912049544428,
      "learning_rate": 5e-06,
      "loss": 0.6234,
      "step": 1840
    },
    {
      "epoch": 3.368229403732362,
      "grad_norm": 0.6730493262615627,
      "learning_rate": 5e-06,
      "loss": 0.6239,
      "step": 1850
    },
    {
      "epoch": 3.386436049157943,
      "grad_norm": 0.624938299600225,
      "learning_rate": 5e-06,
      "loss": 0.6225,
      "step": 1860
    },
    {
      "epoch": 3.404642694583523,
      "grad_norm": 0.623354310700811,
      "learning_rate": 5e-06,
      "loss": 0.6255,
      "step": 1870
    },
    {
      "epoch": 3.4228493400091033,
      "grad_norm": 0.6329249624990303,
      "learning_rate": 5e-06,
      "loss": 0.6253,
      "step": 1880
    },
    {
      "epoch": 3.4410559854346836,
      "grad_norm": 0.691830816858052,
      "learning_rate": 5e-06,
      "loss": 0.6313,
      "step": 1890
    },
    {
      "epoch": 3.459262630860264,
      "grad_norm": 0.6032142068639882,
      "learning_rate": 5e-06,
      "loss": 0.6279,
      "step": 1900
    },
    {
      "epoch": 3.477469276285844,
      "grad_norm": 0.6620277244742816,
      "learning_rate": 5e-06,
      "loss": 0.6279,
      "step": 1910
    },
    {
      "epoch": 3.495675921711425,
      "grad_norm": 0.6334218656408422,
      "learning_rate": 5e-06,
      "loss": 0.6299,
      "step": 1920
    },
    {
      "epoch": 3.513882567137005,
      "grad_norm": 0.7375169169002789,
      "learning_rate": 5e-06,
      "loss": 0.6281,
      "step": 1930
    },
    {
      "epoch": 3.5320892125625853,
      "grad_norm": 0.6448512236031347,
      "learning_rate": 5e-06,
      "loss": 0.6275,
      "step": 1940
    },
    {
      "epoch": 3.5502958579881656,
      "grad_norm": 0.5665149421787481,
      "learning_rate": 5e-06,
      "loss": 0.6299,
      "step": 1950
    },
    {
      "epoch": 3.568502503413746,
      "grad_norm": 0.5999159680891181,
      "learning_rate": 5e-06,
      "loss": 0.6286,
      "step": 1960
    },
    {
      "epoch": 3.5867091488393266,
      "grad_norm": 0.6281107227517486,
      "learning_rate": 5e-06,
      "loss": 0.6248,
      "step": 1970
    },
    {
      "epoch": 3.604915794264907,
      "grad_norm": 0.6353135942879086,
      "learning_rate": 5e-06,
      "loss": 0.6281,
      "step": 1980
    },
    {
      "epoch": 3.623122439690487,
      "grad_norm": 0.6193631021536377,
      "learning_rate": 5e-06,
      "loss": 0.6278,
      "step": 1990
    },
    {
      "epoch": 3.6413290851160673,
      "grad_norm": 0.5918570154763652,
      "learning_rate": 5e-06,
      "loss": 0.6296,
      "step": 2000
    },
    {
      "epoch": 3.6595357305416476,
      "grad_norm": 0.6010286514232586,
      "learning_rate": 5e-06,
      "loss": 0.6312,
      "step": 2010
    },
    {
      "epoch": 3.6777423759672283,
      "grad_norm": 0.556597304136066,
      "learning_rate": 5e-06,
      "loss": 0.6282,
      "step": 2020
    },
    {
      "epoch": 3.695949021392808,
      "grad_norm": 0.7271994139802581,
      "learning_rate": 5e-06,
      "loss": 0.6282,
      "step": 2030
    },
    {
      "epoch": 3.714155666818389,
      "grad_norm": 0.6169419004473696,
      "learning_rate": 5e-06,
      "loss": 0.633,
      "step": 2040
    },
    {
      "epoch": 3.732362312243969,
      "grad_norm": 0.6025566845682745,
      "learning_rate": 5e-06,
      "loss": 0.6299,
      "step": 2050
    },
    {
      "epoch": 3.7505689576695493,
      "grad_norm": 0.5736503199358731,
      "learning_rate": 5e-06,
      "loss": 0.6307,
      "step": 2060
    },
    {
      "epoch": 3.7687756030951296,
      "grad_norm": 0.6112656378611477,
      "learning_rate": 5e-06,
      "loss": 0.6314,
      "step": 2070
    },
    {
      "epoch": 3.78698224852071,
      "grad_norm": 0.683407900900997,
      "learning_rate": 5e-06,
      "loss": 0.6292,
      "step": 2080
    },
    {
      "epoch": 3.8051888939462906,
      "grad_norm": 0.5823312600994301,
      "learning_rate": 5e-06,
      "loss": 0.6302,
      "step": 2090
    },
    {
      "epoch": 3.823395539371871,
      "grad_norm": 0.7722774266919119,
      "learning_rate": 5e-06,
      "loss": 0.6329,
      "step": 2100
    },
    {
      "epoch": 3.841602184797451,
      "grad_norm": 0.5962312514475476,
      "learning_rate": 5e-06,
      "loss": 0.6336,
      "step": 2110
    },
    {
      "epoch": 3.8598088302230313,
      "grad_norm": 0.5704378657470786,
      "learning_rate": 5e-06,
      "loss": 0.6343,
      "step": 2120
    },
    {
      "epoch": 3.8780154756486116,
      "grad_norm": 0.6221717567681257,
      "learning_rate": 5e-06,
      "loss": 0.6323,
      "step": 2130
    },
    {
      "epoch": 3.8962221210741923,
      "grad_norm": 0.8715901410739767,
      "learning_rate": 5e-06,
      "loss": 0.6289,
      "step": 2140
    },
    {
      "epoch": 3.9144287664997726,
      "grad_norm": 0.7264743862880063,
      "learning_rate": 5e-06,
      "loss": 0.6319,
      "step": 2150
    },
    {
      "epoch": 3.932635411925353,
      "grad_norm": 0.5646894741460851,
      "learning_rate": 5e-06,
      "loss": 0.6326,
      "step": 2160
    },
    {
      "epoch": 3.950842057350933,
      "grad_norm": 0.6303177348076553,
      "learning_rate": 5e-06,
      "loss": 0.6347,
      "step": 2170
    },
    {
      "epoch": 3.9690487027765133,
      "grad_norm": 0.5882842156900018,
      "learning_rate": 5e-06,
      "loss": 0.6255,
      "step": 2180
    },
    {
      "epoch": 3.987255348202094,
      "grad_norm": 0.7637770706292029,
      "learning_rate": 5e-06,
      "loss": 0.6324,
      "step": 2190
    },
    {
      "epoch": 4.0,
      "eval_loss": 0.7583181262016296,
      "eval_runtime": 379.7995,
      "eval_samples_per_second": 38.971,
      "eval_steps_per_second": 0.611,
      "step": 2197
    },
    {
      "epoch": 4.005461993627674,
      "grad_norm": 1.3302589404721668,
      "learning_rate": 5e-06,
      "loss": 0.6642,
      "step": 2200
    },
    {
      "epoch": 4.023668639053255,
      "grad_norm": 1.0479601737495694,
      "learning_rate": 5e-06,
      "loss": 0.5747,
      "step": 2210
    },
    {
      "epoch": 4.041875284478834,
      "grad_norm": 0.8236224363789262,
      "learning_rate": 5e-06,
      "loss": 0.5739,
      "step": 2220
    },
    {
      "epoch": 4.060081929904415,
      "grad_norm": 0.6514312376533014,
      "learning_rate": 5e-06,
      "loss": 0.5717,
      "step": 2230
    },
    {
      "epoch": 4.078288575329996,
      "grad_norm": 0.6470033460056318,
      "learning_rate": 5e-06,
      "loss": 0.577,
      "step": 2240
    },
    {
      "epoch": 4.096495220755576,
      "grad_norm": 0.6842183198102436,
      "learning_rate": 5e-06,
      "loss": 0.5772,
      "step": 2250
    },
    {
      "epoch": 4.114701866181156,
      "grad_norm": 0.7773595484089195,
      "learning_rate": 5e-06,
      "loss": 0.5787,
      "step": 2260
    },
    {
      "epoch": 4.132908511606736,
      "grad_norm": 0.8512328139098053,
      "learning_rate": 5e-06,
      "loss": 0.5782,
      "step": 2270
    },
    {
      "epoch": 4.151115157032317,
      "grad_norm": 0.7556248928216761,
      "learning_rate": 5e-06,
      "loss": 0.5794,
      "step": 2280
    },
    {
      "epoch": 4.1693218024578975,
      "grad_norm": 0.649345136898707,
      "learning_rate": 5e-06,
      "loss": 0.5806,
      "step": 2290
    },
    {
      "epoch": 4.187528447883477,
      "grad_norm": 0.7218760078584716,
      "learning_rate": 5e-06,
      "loss": 0.5797,
      "step": 2300
    },
    {
      "epoch": 4.205735093309058,
      "grad_norm": 0.6931763868077586,
      "learning_rate": 5e-06,
      "loss": 0.5767,
      "step": 2310
    },
    {
      "epoch": 4.223941738734638,
      "grad_norm": 0.764188364334872,
      "learning_rate": 5e-06,
      "loss": 0.5778,
      "step": 2320
    },
    {
      "epoch": 4.2421483841602186,
      "grad_norm": 0.7472509039059302,
      "learning_rate": 5e-06,
      "loss": 0.5786,
      "step": 2330
    },
    {
      "epoch": 4.260355029585799,
      "grad_norm": 0.7681897599903893,
      "learning_rate": 5e-06,
      "loss": 0.5824,
      "step": 2340
    },
    {
      "epoch": 4.278561675011379,
      "grad_norm": 0.6853235525321896,
      "learning_rate": 5e-06,
      "loss": 0.578,
      "step": 2350
    },
    {
      "epoch": 4.29676832043696,
      "grad_norm": 0.6371627957254833,
      "learning_rate": 5e-06,
      "loss": 0.5767,
      "step": 2360
    },
    {
      "epoch": 4.31497496586254,
      "grad_norm": 0.62518793171934,
      "learning_rate": 5e-06,
      "loss": 0.5828,
      "step": 2370
    },
    {
      "epoch": 4.33318161128812,
      "grad_norm": 0.6835225300328861,
      "learning_rate": 5e-06,
      "loss": 0.5845,
      "step": 2380
    },
    {
      "epoch": 4.3513882567137,
      "grad_norm": 0.6635739669645049,
      "learning_rate": 5e-06,
      "loss": 0.5831,
      "step": 2390
    },
    {
      "epoch": 4.369594902139281,
      "grad_norm": 0.6905352305966937,
      "learning_rate": 5e-06,
      "loss": 0.5838,
      "step": 2400
    },
    {
      "epoch": 4.3878015475648615,
      "grad_norm": 0.675578669245968,
      "learning_rate": 5e-06,
      "loss": 0.5838,
      "step": 2410
    },
    {
      "epoch": 4.406008192990441,
      "grad_norm": 0.6472341173499976,
      "learning_rate": 5e-06,
      "loss": 0.5873,
      "step": 2420
    },
    {
      "epoch": 4.424214838416022,
      "grad_norm": 0.6312793186052579,
      "learning_rate": 5e-06,
      "loss": 0.5862,
      "step": 2430
    },
    {
      "epoch": 4.442421483841602,
      "grad_norm": 0.6451118193568525,
      "learning_rate": 5e-06,
      "loss": 0.5825,
      "step": 2440
    },
    {
      "epoch": 4.4606281292671826,
      "grad_norm": 0.6570443368465694,
      "learning_rate": 5e-06,
      "loss": 0.5854,
      "step": 2450
    },
    {
      "epoch": 4.478834774692763,
      "grad_norm": 0.658937290588797,
      "learning_rate": 5e-06,
      "loss": 0.5866,
      "step": 2460
    },
    {
      "epoch": 4.497041420118343,
      "grad_norm": 0.7077581907222236,
      "learning_rate": 5e-06,
      "loss": 0.5869,
      "step": 2470
    },
    {
      "epoch": 4.515248065543924,
      "grad_norm": 0.6792893518112693,
      "learning_rate": 5e-06,
      "loss": 0.5873,
      "step": 2480
    },
    {
      "epoch": 4.533454710969504,
      "grad_norm": 0.7399002369000202,
      "learning_rate": 5e-06,
      "loss": 0.5832,
      "step": 2490
    },
    {
      "epoch": 4.551661356395084,
      "grad_norm": 0.7316128957109573,
      "learning_rate": 5e-06,
      "loss": 0.5878,
      "step": 2500
    },
    {
      "epoch": 4.569868001820664,
      "grad_norm": 0.709723130607844,
      "learning_rate": 5e-06,
      "loss": 0.5844,
      "step": 2510
    },
    {
      "epoch": 4.588074647246245,
      "grad_norm": 0.6462822194584583,
      "learning_rate": 5e-06,
      "loss": 0.5883,
      "step": 2520
    },
    {
      "epoch": 4.6062812926718255,
      "grad_norm": 0.7572060951703973,
      "learning_rate": 5e-06,
      "loss": 0.5914,
      "step": 2530
    },
    {
      "epoch": 4.624487938097405,
      "grad_norm": 0.7698033683902142,
      "learning_rate": 5e-06,
      "loss": 0.5882,
      "step": 2540
    },
    {
      "epoch": 4.642694583522986,
      "grad_norm": 0.7913214081907007,
      "learning_rate": 5e-06,
      "loss": 0.5865,
      "step": 2550
    },
    {
      "epoch": 4.660901228948566,
      "grad_norm": 0.652908829617188,
      "learning_rate": 5e-06,
      "loss": 0.5885,
      "step": 2560
    },
    {
      "epoch": 4.6791078743741465,
      "grad_norm": 0.7196290964249576,
      "learning_rate": 5e-06,
      "loss": 0.585,
      "step": 2570
    },
    {
      "epoch": 4.697314519799727,
      "grad_norm": 0.6963685189277059,
      "learning_rate": 5e-06,
      "loss": 0.5874,
      "step": 2580
    },
    {
      "epoch": 4.715521165225307,
      "grad_norm": 0.6213064117674314,
      "learning_rate": 5e-06,
      "loss": 0.5886,
      "step": 2590
    },
    {
      "epoch": 4.733727810650888,
      "grad_norm": 0.6931094000594287,
      "learning_rate": 5e-06,
      "loss": 0.5931,
      "step": 2600
    },
    {
      "epoch": 4.751934456076468,
      "grad_norm": 0.6421262946164188,
      "learning_rate": 5e-06,
      "loss": 0.5877,
      "step": 2610
    },
    {
      "epoch": 4.770141101502048,
      "grad_norm": 0.7592504356723478,
      "learning_rate": 5e-06,
      "loss": 0.5901,
      "step": 2620
    },
    {
      "epoch": 4.788347746927629,
      "grad_norm": 0.5755276719477025,
      "learning_rate": 5e-06,
      "loss": 0.5877,
      "step": 2630
    },
    {
      "epoch": 4.806554392353209,
      "grad_norm": 0.7579206553363019,
      "learning_rate": 5e-06,
      "loss": 0.593,
      "step": 2640
    },
    {
      "epoch": 4.8247610377787895,
      "grad_norm": 0.7227783750503406,
      "learning_rate": 5e-06,
      "loss": 0.592,
      "step": 2650
    },
    {
      "epoch": 4.842967683204369,
      "grad_norm": 0.6143720314418375,
      "learning_rate": 5e-06,
      "loss": 0.5923,
      "step": 2660
    },
    {
      "epoch": 4.86117432862995,
      "grad_norm": 0.684711113595212,
      "learning_rate": 5e-06,
      "loss": 0.5906,
      "step": 2670
    },
    {
      "epoch": 4.879380974055531,
      "grad_norm": 0.6041365723044803,
      "learning_rate": 5e-06,
      "loss": 0.5943,
      "step": 2680
    },
    {
      "epoch": 4.8975876194811105,
      "grad_norm": 0.6690754978083606,
      "learning_rate": 5e-06,
      "loss": 0.5909,
      "step": 2690
    },
    {
      "epoch": 4.915794264906691,
      "grad_norm": 0.6408949623705275,
      "learning_rate": 5e-06,
      "loss": 0.5938,
      "step": 2700
    },
    {
      "epoch": 4.934000910332271,
      "grad_norm": 0.6310380336524642,
      "learning_rate": 5e-06,
      "loss": 0.5896,
      "step": 2710
    },
    {
      "epoch": 4.952207555757852,
      "grad_norm": 0.7658719466797836,
      "learning_rate": 5e-06,
      "loss": 0.5925,
      "step": 2720
    },
    {
      "epoch": 4.970414201183432,
      "grad_norm": 0.6764346949045434,
      "learning_rate": 5e-06,
      "loss": 0.587,
      "step": 2730
    },
    {
      "epoch": 4.988620846609012,
      "grad_norm": 0.6578929361436391,
      "learning_rate": 5e-06,
      "loss": 0.5947,
      "step": 2740
    },
    {
      "epoch": 4.999544833864361,
      "eval_loss": 0.7756755352020264,
      "eval_runtime": 379.7908,
      "eval_samples_per_second": 38.971,
      "eval_steps_per_second": 0.611,
      "step": 2746
    },
    {
      "epoch": 5.006827492034593,
      "grad_norm": 1.6971405469705954,
      "learning_rate": 5e-06,
      "loss": 0.6175,
      "step": 2750
    },
    {
      "epoch": 5.025034137460173,
      "grad_norm": 1.1249548156787617,
      "learning_rate": 5e-06,
      "loss": 0.5284,
      "step": 2760
    },
    {
      "epoch": 5.0432407828857535,
      "grad_norm": 0.8687787258109687,
      "learning_rate": 5e-06,
      "loss": 0.5308,
      "step": 2770
    },
    {
      "epoch": 5.061447428311333,
      "grad_norm": 0.8717275105088685,
      "learning_rate": 5e-06,
      "loss": 0.5312,
      "step": 2780
    },
    {
      "epoch": 5.079654073736914,
      "grad_norm": 0.8465418731492865,
      "learning_rate": 5e-06,
      "loss": 0.5304,
      "step": 2790
    },
    {
      "epoch": 5.097860719162495,
      "grad_norm": 0.7886283884740087,
      "learning_rate": 5e-06,
      "loss": 0.5279,
      "step": 2800
    },
    {
      "epoch": 5.1160673645880745,
      "grad_norm": 0.8398942783728177,
      "learning_rate": 5e-06,
      "loss": 0.5363,
      "step": 2810
    },
    {
      "epoch": 5.134274010013655,
      "grad_norm": 0.769923514886696,
      "learning_rate": 5e-06,
      "loss": 0.5338,
      "step": 2820
    },
    {
      "epoch": 5.152480655439235,
      "grad_norm": 0.7868390424124061,
      "learning_rate": 5e-06,
      "loss": 0.5321,
      "step": 2830
    },
    {
      "epoch": 5.170687300864816,
      "grad_norm": 0.7950518134725912,
      "learning_rate": 5e-06,
      "loss": 0.534,
      "step": 2840
    },
    {
      "epoch": 5.188893946290396,
      "grad_norm": 0.9279901061572416,
      "learning_rate": 5e-06,
      "loss": 0.5363,
      "step": 2850
    },
    {
      "epoch": 5.207100591715976,
      "grad_norm": 0.7594304483347795,
      "learning_rate": 5e-06,
      "loss": 0.534,
      "step": 2860
    },
    {
      "epoch": 5.225307237141557,
      "grad_norm": 0.8252531851052891,
      "learning_rate": 5e-06,
      "loss": 0.5349,
      "step": 2870
    },
    {
      "epoch": 5.243513882567137,
      "grad_norm": 0.7626524916862092,
      "learning_rate": 5e-06,
      "loss": 0.534,
      "step": 2880
    },
    {
      "epoch": 5.2617205279927175,
      "grad_norm": 0.78634458156834,
      "learning_rate": 5e-06,
      "loss": 0.5368,
      "step": 2890
    },
    {
      "epoch": 5.279927173418297,
      "grad_norm": 0.7800860821607554,
      "learning_rate": 5e-06,
      "loss": 0.5348,
      "step": 2900
    },
    {
      "epoch": 5.298133818843878,
      "grad_norm": 0.7198214581579875,
      "learning_rate": 5e-06,
      "loss": 0.539,
      "step": 2910
    },
    {
      "epoch": 5.316340464269459,
      "grad_norm": 0.8018455337446776,
      "learning_rate": 5e-06,
      "loss": 0.5354,
      "step": 2920
    },
    {
      "epoch": 5.3345471096950385,
      "grad_norm": 0.6375711217756587,
      "learning_rate": 5e-06,
      "loss": 0.538,
      "step": 2930
    },
    {
      "epoch": 5.352753755120619,
      "grad_norm": 0.7615279567325839,
      "learning_rate": 5e-06,
      "loss": 0.5388,
      "step": 2940
    },
    {
      "epoch": 5.370960400546199,
      "grad_norm": 0.8634503306148038,
      "learning_rate": 5e-06,
      "loss": 0.5387,
      "step": 2950
    },
    {
      "epoch": 5.38916704597178,
      "grad_norm": 0.728328161040519,
      "learning_rate": 5e-06,
      "loss": 0.5392,
      "step": 2960
    },
    {
      "epoch": 5.4073736913973605,
      "grad_norm": 0.6714611489114068,
      "learning_rate": 5e-06,
      "loss": 0.5393,
      "step": 2970
    },
    {
      "epoch": 5.42558033682294,
      "grad_norm": 0.864160433406934,
      "learning_rate": 5e-06,
      "loss": 0.5406,
      "step": 2980
    },
    {
      "epoch": 5.443786982248521,
      "grad_norm": 0.7597928888037461,
      "learning_rate": 5e-06,
      "loss": 0.5368,
      "step": 2990
    },
    {
      "epoch": 5.461993627674101,
      "grad_norm": 0.7636646866418205,
      "learning_rate": 5e-06,
      "loss": 0.5403,
      "step": 3000
    },
    {
      "epoch": 5.4802002730996815,
      "grad_norm": 0.7478936763176718,
      "learning_rate": 5e-06,
      "loss": 0.5411,
      "step": 3010
    },
    {
      "epoch": 5.498406918525261,
      "grad_norm": 0.7764354445181351,
      "learning_rate": 5e-06,
      "loss": 0.5411,
      "step": 3020
    },
    {
      "epoch": 5.516613563950842,
      "grad_norm": 0.8232735346509414,
      "learning_rate": 5e-06,
      "loss": 0.5433,
      "step": 3030
    },
    {
      "epoch": 5.534820209376423,
      "grad_norm": 0.7742332026679888,
      "learning_rate": 5e-06,
      "loss": 0.5459,
      "step": 3040
    },
    {
      "epoch": 5.5530268548020025,
      "grad_norm": 0.8806327502433587,
      "learning_rate": 5e-06,
      "loss": 0.5456,
      "step": 3050
    },
    {
      "epoch": 5.571233500227583,
      "grad_norm": 0.8857449291804894,
      "learning_rate": 5e-06,
      "loss": 0.5434,
      "step": 3060
    },
    {
      "epoch": 5.589440145653163,
      "grad_norm": 0.9161045429573774,
      "learning_rate": 5e-06,
      "loss": 0.5381,
      "step": 3070
    },
    {
      "epoch": 5.607646791078744,
      "grad_norm": 0.7948317610189037,
      "learning_rate": 5e-06,
      "loss": 0.5424,
      "step": 3080
    },
    {
      "epoch": 5.6258534365043245,
      "grad_norm": 0.7867485103330927,
      "learning_rate": 5e-06,
      "loss": 0.5434,
      "step": 3090
    },
    {
      "epoch": 5.644060081929904,
      "grad_norm": 0.7569175444017539,
      "learning_rate": 5e-06,
      "loss": 0.5456,
      "step": 3100
    },
    {
      "epoch": 5.662266727355485,
      "grad_norm": 0.7425402095430522,
      "learning_rate": 5e-06,
      "loss": 0.5437,
      "step": 3110
    },
    {
      "epoch": 5.680473372781065,
      "grad_norm": 0.8566468508292944,
      "learning_rate": 5e-06,
      "loss": 0.543,
      "step": 3120
    },
    {
      "epoch": 5.6986800182066455,
      "grad_norm": 0.7291267040452126,
      "learning_rate": 5e-06,
      "loss": 0.5456,
      "step": 3130
    },
    {
      "epoch": 5.716886663632225,
      "grad_norm": 0.9339173337499906,
      "learning_rate": 5e-06,
      "loss": 0.5488,
      "step": 3140
    },
    {
      "epoch": 5.735093309057806,
      "grad_norm": 0.668254133197116,
      "learning_rate": 5e-06,
      "loss": 0.5461,
      "step": 3150
    },
    {
      "epoch": 5.753299954483387,
      "grad_norm": 0.733779005468029,
      "learning_rate": 5e-06,
      "loss": 0.548,
      "step": 3160
    },
    {
      "epoch": 5.7715065999089665,
      "grad_norm": 0.6817898908805843,
      "learning_rate": 5e-06,
      "loss": 0.5471,
      "step": 3170
    },
    {
      "epoch": 5.789713245334547,
      "grad_norm": 0.7061838303703719,
      "learning_rate": 5e-06,
      "loss": 0.5435,
      "step": 3180
    },
    {
      "epoch": 5.807919890760127,
      "grad_norm": 0.6788561279223211,
      "learning_rate": 5e-06,
      "loss": 0.5485,
      "step": 3190
    },
    {
      "epoch": 5.826126536185708,
      "grad_norm": 0.743542468839561,
      "learning_rate": 5e-06,
      "loss": 0.5442,
      "step": 3200
    },
    {
      "epoch": 5.8443331816112885,
      "grad_norm": 0.7126719114866883,
      "learning_rate": 5e-06,
      "loss": 0.545,
      "step": 3210
    },
    {
      "epoch": 5.862539827036868,
      "grad_norm": 0.7700757305885421,
      "learning_rate": 5e-06,
      "loss": 0.5476,
      "step": 3220
    },
    {
      "epoch": 5.880746472462449,
      "grad_norm": 0.8580615971421329,
      "learning_rate": 5e-06,
      "loss": 0.5496,
      "step": 3230
    },
    {
      "epoch": 5.898953117888029,
      "grad_norm": 0.6651944277970823,
      "learning_rate": 5e-06,
      "loss": 0.5467,
      "step": 3240
    },
    {
      "epoch": 5.9171597633136095,
      "grad_norm": 0.7351660383841082,
      "learning_rate": 5e-06,
      "loss": 0.5498,
      "step": 3250
    },
    {
      "epoch": 5.93536640873919,
      "grad_norm": 0.7432057163838226,
      "learning_rate": 5e-06,
      "loss": 0.5457,
      "step": 3260
    },
    {
      "epoch": 5.95357305416477,
      "grad_norm": 0.7880258422896124,
      "learning_rate": 5e-06,
      "loss": 0.5484,
      "step": 3270
    },
    {
      "epoch": 5.971779699590351,
      "grad_norm": 0.7829267364761494,
      "learning_rate": 5e-06,
      "loss": 0.5522,
      "step": 3280
    },
    {
      "epoch": 5.9899863450159305,
      "grad_norm": 0.755975814943294,
      "learning_rate": 5e-06,
      "loss": 0.5504,
      "step": 3290
    },
    {
      "epoch": 5.999089667728721,
      "eval_loss": 0.8121561408042908,
      "eval_runtime": 385.6914,
      "eval_samples_per_second": 38.375,
      "eval_steps_per_second": 0.602,
      "step": 3295
    },
    {
      "epoch": 6.008192990441511,
      "grad_norm": 1.229518253610835,
      "learning_rate": 5e-06,
      "loss": 0.5616,
      "step": 3300
    },
    {
      "epoch": 6.026399635867092,
      "grad_norm": 0.9690039493999913,
      "learning_rate": 5e-06,
      "loss": 0.4806,
      "step": 3310
    },
    {
      "epoch": 6.044606281292672,
      "grad_norm": 0.975061524739041,
      "learning_rate": 5e-06,
      "loss": 0.4807,
      "step": 3320
    },
    {
      "epoch": 6.0628129267182524,
      "grad_norm": 0.8693905009394387,
      "learning_rate": 5e-06,
      "loss": 0.48,
      "step": 3330
    },
    {
      "epoch": 6.081019572143832,
      "grad_norm": 0.8533904059239436,
      "learning_rate": 5e-06,
      "loss": 0.4814,
      "step": 3340
    },
    {
      "epoch": 6.099226217569413,
      "grad_norm": 0.9743084553071109,
      "learning_rate": 5e-06,
      "loss": 0.4832,
      "step": 3350
    },
    {
      "epoch": 6.117432862994993,
      "grad_norm": 0.8428587686314428,
      "learning_rate": 5e-06,
      "loss": 0.4836,
      "step": 3360
    },
    {
      "epoch": 6.1356395084205735,
      "grad_norm": 0.8485721574906323,
      "learning_rate": 5e-06,
      "loss": 0.4842,
      "step": 3370
    },
    {
      "epoch": 6.153846153846154,
      "grad_norm": 0.9049634307737825,
      "learning_rate": 5e-06,
      "loss": 0.4871,
      "step": 3380
    },
    {
      "epoch": 6.172052799271734,
      "grad_norm": 0.8485250729351949,
      "learning_rate": 5e-06,
      "loss": 0.4891,
      "step": 3390
    },
    {
      "epoch": 6.190259444697315,
      "grad_norm": 0.8677063446883995,
      "learning_rate": 5e-06,
      "loss": 0.4863,
      "step": 3400
    },
    {
      "epoch": 6.2084660901228945,
      "grad_norm": 0.8067384460711394,
      "learning_rate": 5e-06,
      "loss": 0.4847,
      "step": 3410
    },
    {
      "epoch": 6.226672735548475,
      "grad_norm": 0.8491856381367863,
      "learning_rate": 5e-06,
      "loss": 0.4872,
      "step": 3420
    },
    {
      "epoch": 6.244879380974056,
      "grad_norm": 0.8434795569738153,
      "learning_rate": 5e-06,
      "loss": 0.4893,
      "step": 3430
    },
    {
      "epoch": 6.263086026399636,
      "grad_norm": 0.877180400082389,
      "learning_rate": 5e-06,
      "loss": 0.488,
      "step": 3440
    },
    {
      "epoch": 6.2812926718252164,
      "grad_norm": 0.7544909031927692,
      "learning_rate": 5e-06,
      "loss": 0.4926,
      "step": 3450
    },
    {
      "epoch": 6.299499317250796,
      "grad_norm": 0.7786685673980743,
      "learning_rate": 5e-06,
      "loss": 0.4903,
      "step": 3460
    },
    {
      "epoch": 6.317705962676377,
      "grad_norm": 0.8813259287698524,
      "learning_rate": 5e-06,
      "loss": 0.4928,
      "step": 3470
    },
    {
      "epoch": 6.335912608101957,
      "grad_norm": 0.8143534255862093,
      "learning_rate": 5e-06,
      "loss": 0.489,
      "step": 3480
    },
    {
      "epoch": 6.3541192535275375,
      "grad_norm": 0.813093240140628,
      "learning_rate": 5e-06,
      "loss": 0.4895,
      "step": 3490
    },
    {
      "epoch": 6.372325898953118,
      "grad_norm": 0.8234651442993571,
      "learning_rate": 5e-06,
      "loss": 0.4874,
      "step": 3500
    },
    {
      "epoch": 6.390532544378698,
      "grad_norm": 0.7583974230123129,
      "learning_rate": 5e-06,
      "loss": 0.4894,
      "step": 3510
    },
    {
      "epoch": 6.408739189804279,
      "grad_norm": 0.8045709791368885,
      "learning_rate": 5e-06,
      "loss": 0.4924,
      "step": 3520
    },
    {
      "epoch": 6.4269458352298585,
      "grad_norm": 0.8021037971072071,
      "learning_rate": 5e-06,
      "loss": 0.4958,
      "step": 3530
    },
    {
      "epoch": 6.445152480655439,
      "grad_norm": 0.8860947638978545,
      "learning_rate": 5e-06,
      "loss": 0.4936,
      "step": 3540
    },
    {
      "epoch": 6.46335912608102,
      "grad_norm": 0.778005059457562,
      "learning_rate": 5e-06,
      "loss": 0.4904,
      "step": 3550
    },
    {
      "epoch": 6.4815657715066,
      "grad_norm": 0.8879225211432363,
      "learning_rate": 5e-06,
      "loss": 0.4996,
      "step": 3560
    },
    {
      "epoch": 6.49977241693218,
      "grad_norm": 0.86115853799121,
      "learning_rate": 5e-06,
      "loss": 0.4958,
      "step": 3570
    },
    {
      "epoch": 6.51797906235776,
      "grad_norm": 0.821923783763917,
      "learning_rate": 5e-06,
      "loss": 0.4998,
      "step": 3580
    },
    {
      "epoch": 6.536185707783341,
      "grad_norm": 0.7488573715925901,
      "learning_rate": 5e-06,
      "loss": 0.4942,
      "step": 3590
    },
    {
      "epoch": 6.554392353208922,
      "grad_norm": 0.766478109512254,
      "learning_rate": 5e-06,
      "loss": 0.4927,
      "step": 3600
    },
    {
      "epoch": 6.5725989986345015,
      "grad_norm": 0.9993393195281365,
      "learning_rate": 5e-06,
      "loss": 0.5009,
      "step": 3610
    },
    {
      "epoch": 6.590805644060082,
      "grad_norm": 1.0105814992493984,
      "learning_rate": 5e-06,
      "loss": 0.4902,
      "step": 3620
    },
    {
      "epoch": 6.609012289485662,
      "grad_norm": 0.918270960348232,
      "learning_rate": 5e-06,
      "loss": 0.5004,
      "step": 3630
    },
    {
      "epoch": 6.627218934911243,
      "grad_norm": 0.7543985269097522,
      "learning_rate": 5e-06,
      "loss": 0.4972,
      "step": 3640
    },
    {
      "epoch": 6.645425580336823,
      "grad_norm": 0.9591362111868698,
      "learning_rate": 5e-06,
      "loss": 0.4978,
      "step": 3650
    },
    {
      "epoch": 6.663632225762403,
      "grad_norm": 0.8732535087129343,
      "learning_rate": 5e-06,
      "loss": 0.4962,
      "step": 3660
    },
    {
      "epoch": 6.681838871187984,
      "grad_norm": 0.8907840024813226,
      "learning_rate": 5e-06,
      "loss": 0.4984,
      "step": 3670
    },
    {
      "epoch": 6.700045516613564,
      "grad_norm": 0.8081950411988741,
      "learning_rate": 5e-06,
      "loss": 0.4995,
      "step": 3680
    },
    {
      "epoch": 6.718252162039144,
      "grad_norm": 0.7957053619070132,
      "learning_rate": 5e-06,
      "loss": 0.5007,
      "step": 3690
    },
    {
      "epoch": 6.736458807464724,
      "grad_norm": 1.0396693696127743,
      "learning_rate": 5e-06,
      "loss": 0.5,
      "step": 3700
    },
    {
      "epoch": 6.754665452890305,
      "grad_norm": 0.8276731947592968,
      "learning_rate": 5e-06,
      "loss": 0.5003,
      "step": 3710
    },
    {
      "epoch": 6.772872098315886,
      "grad_norm": 0.740304498205183,
      "learning_rate": 5e-06,
      "loss": 0.498,
      "step": 3720
    },
    {
      "epoch": 6.7910787437414655,
      "grad_norm": 0.7463156387752775,
      "learning_rate": 5e-06,
      "loss": 0.4994,
      "step": 3730
    },
    {
      "epoch": 6.809285389167046,
      "grad_norm": 0.8016121539442179,
      "learning_rate": 5e-06,
      "loss": 0.4997,
      "step": 3740
    },
    {
      "epoch": 6.827492034592626,
      "grad_norm": 0.8418329142580222,
      "learning_rate": 5e-06,
      "loss": 0.4982,
      "step": 3750
    },
    {
      "epoch": 6.845698680018207,
      "grad_norm": 0.8007293534037732,
      "learning_rate": 5e-06,
      "loss": 0.5008,
      "step": 3760
    },
    {
      "epoch": 6.8639053254437865,
      "grad_norm": 0.7962992985732563,
      "learning_rate": 5e-06,
      "loss": 0.4983,
      "step": 3770
    },
    {
      "epoch": 6.882111970869367,
      "grad_norm": 0.8192800444422219,
      "learning_rate": 5e-06,
      "loss": 0.5012,
      "step": 3780
    },
    {
      "epoch": 6.900318616294948,
      "grad_norm": 0.7933819169487161,
      "learning_rate": 5e-06,
      "loss": 0.4993,
      "step": 3790
    },
    {
      "epoch": 6.918525261720528,
      "grad_norm": 0.7719557878224939,
      "learning_rate": 5e-06,
      "loss": 0.4969,
      "step": 3800
    },
    {
      "epoch": 6.936731907146108,
      "grad_norm": 0.7794726223761498,
      "learning_rate": 5e-06,
      "loss": 0.4997,
      "step": 3810
    },
    {
      "epoch": 6.954938552571688,
      "grad_norm": 0.7490795619365134,
      "learning_rate": 5e-06,
      "loss": 0.4995,
      "step": 3820
    },
    {
      "epoch": 6.973145197997269,
      "grad_norm": 0.8534935347128176,
      "learning_rate": 5e-06,
      "loss": 0.5045,
      "step": 3830
    },
    {
      "epoch": 6.99135184342285,
      "grad_norm": 0.8266159513619395,
      "learning_rate": 5e-06,
      "loss": 0.5002,
      "step": 3840
    },
    {
      "epoch": 6.998634501593082,
      "eval_loss": 0.8599892854690552,
      "eval_runtime": 382.211,
      "eval_samples_per_second": 38.725,
      "eval_steps_per_second": 0.607,
      "step": 3844
    },
    {
      "epoch": 7.0095584888484295,
      "grad_norm": 1.4929469131394506,
      "learning_rate": 5e-06,
      "loss": 0.507,
      "step": 3850
    },
    {
      "epoch": 7.02776513427401,
      "grad_norm": 0.9806727302100092,
      "learning_rate": 5e-06,
      "loss": 0.4258,
      "step": 3860
    },
    {
      "epoch": 7.04597177969959,
      "grad_norm": 1.0186437284049983,
      "learning_rate": 5e-06,
      "loss": 0.4304,
      "step": 3870
    },
    {
      "epoch": 7.064178425125171,
      "grad_norm": 0.8346902224337589,
      "learning_rate": 5e-06,
      "loss": 0.4315,
      "step": 3880
    },
    {
      "epoch": 7.082385070550751,
      "grad_norm": 0.8895988414802425,
      "learning_rate": 5e-06,
      "loss": 0.4321,
      "step": 3890
    },
    {
      "epoch": 7.100591715976331,
      "grad_norm": 0.8968273584319876,
      "learning_rate": 5e-06,
      "loss": 0.4316,
      "step": 3900
    },
    {
      "epoch": 7.118798361401912,
      "grad_norm": 1.0430199362279293,
      "learning_rate": 5e-06,
      "loss": 0.4352,
      "step": 3910
    },
    {
      "epoch": 7.137005006827492,
      "grad_norm": 0.9925445246056604,
      "learning_rate": 5e-06,
      "loss": 0.4335,
      "step": 3920
    },
    {
      "epoch": 7.155211652253072,
      "grad_norm": 0.9236834017922617,
      "learning_rate": 5e-06,
      "loss": 0.4344,
      "step": 3930
    },
    {
      "epoch": 7.173418297678653,
      "grad_norm": 0.9006887934639467,
      "learning_rate": 5e-06,
      "loss": 0.4354,
      "step": 3940
    },
    {
      "epoch": 7.191624943104233,
      "grad_norm": 1.021746418559959,
      "learning_rate": 5e-06,
      "loss": 0.4337,
      "step": 3950
    },
    {
      "epoch": 7.209831588529814,
      "grad_norm": 0.8820115554192361,
      "learning_rate": 5e-06,
      "loss": 0.4356,
      "step": 3960
    },
    {
      "epoch": 7.2280382339553935,
      "grad_norm": 0.9450373403984087,
      "learning_rate": 5e-06,
      "loss": 0.4346,
      "step": 3970
    },
    {
      "epoch": 7.246244879380974,
      "grad_norm": 0.9599720579833242,
      "learning_rate": 5e-06,
      "loss": 0.437,
      "step": 3980
    },
    {
      "epoch": 7.264451524806554,
      "grad_norm": 0.944022553759221,
      "learning_rate": 5e-06,
      "loss": 0.4373,
      "step": 3990
    },
    {
      "epoch": 7.282658170232135,
      "grad_norm": 1.1470515929421334,
      "learning_rate": 5e-06,
      "loss": 0.437,
      "step": 4000
    },
    {
      "epoch": 7.300864815657715,
      "grad_norm": 0.9229924188228826,
      "learning_rate": 5e-06,
      "loss": 0.4412,
      "step": 4010
    },
    {
      "epoch": 7.319071461083295,
      "grad_norm": 1.1004139942729736,
      "learning_rate": 5e-06,
      "loss": 0.4404,
      "step": 4020
    },
    {
      "epoch": 7.337278106508876,
      "grad_norm": 0.9234895336641952,
      "learning_rate": 5e-06,
      "loss": 0.4413,
      "step": 4030
    },
    {
      "epoch": 7.355484751934456,
      "grad_norm": 0.9191069378971881,
      "learning_rate": 5e-06,
      "loss": 0.4374,
      "step": 4040
    },
    {
      "epoch": 7.373691397360036,
      "grad_norm": 0.9565597630772228,
      "learning_rate": 5e-06,
      "loss": 0.4427,
      "step": 4050
    },
    {
      "epoch": 7.391898042785617,
      "grad_norm": 0.8991823052622302,
      "learning_rate": 5e-06,
      "loss": 0.4408,
      "step": 4060
    },
    {
      "epoch": 7.410104688211197,
      "grad_norm": 0.9463545273318638,
      "learning_rate": 5e-06,
      "loss": 0.4428,
      "step": 4070
    },
    {
      "epoch": 7.428311333636778,
      "grad_norm": 0.8995180480466028,
      "learning_rate": 5e-06,
      "loss": 0.439,
      "step": 4080
    },
    {
      "epoch": 7.4465179790623575,
      "grad_norm": 0.8848211155155944,
      "learning_rate": 5e-06,
      "loss": 0.4408,
      "step": 4090
    },
    {
      "epoch": 7.464724624487938,
      "grad_norm": 0.9344565804246009,
      "learning_rate": 5e-06,
      "loss": 0.4426,
      "step": 4100
    },
    {
      "epoch": 7.482931269913518,
      "grad_norm": 0.9397047554338785,
      "learning_rate": 5e-06,
      "loss": 0.4422,
      "step": 4110
    },
    {
      "epoch": 7.501137915339099,
      "grad_norm": 1.0544561679078621,
      "learning_rate": 5e-06,
      "loss": 0.4433,
      "step": 4120
    },
    {
      "epoch": 7.519344560764679,
      "grad_norm": 1.018669421540764,
      "learning_rate": 5e-06,
      "loss": 0.4425,
      "step": 4130
    },
    {
      "epoch": 7.537551206190259,
      "grad_norm": 0.884634669314439,
      "learning_rate": 5e-06,
      "loss": 0.4447,
      "step": 4140
    },
    {
      "epoch": 7.55575785161584,
      "grad_norm": 0.8973368938963501,
      "learning_rate": 5e-06,
      "loss": 0.4414,
      "step": 4150
    },
    {
      "epoch": 7.57396449704142,
      "grad_norm": 1.0298222416647689,
      "learning_rate": 5e-06,
      "loss": 0.4475,
      "step": 4160
    },
    {
      "epoch": 7.592171142467,
      "grad_norm": 0.9090837637474475,
      "learning_rate": 5e-06,
      "loss": 0.4427,
      "step": 4170
    },
    {
      "epoch": 7.610377787892581,
      "grad_norm": 0.9114871413052325,
      "learning_rate": 5e-06,
      "loss": 0.4464,
      "step": 4180
    },
    {
      "epoch": 7.628584433318161,
      "grad_norm": 0.8765155235059887,
      "learning_rate": 5e-06,
      "loss": 0.4459,
      "step": 4190
    },
    {
      "epoch": 7.646791078743742,
      "grad_norm": 0.8572465195574555,
      "learning_rate": 5e-06,
      "loss": 0.4466,
      "step": 4200
    },
    {
      "epoch": 7.6649977241693215,
      "grad_norm": 0.8619953574827645,
      "learning_rate": 5e-06,
      "loss": 0.4449,
      "step": 4210
    },
    {
      "epoch": 7.683204369594902,
      "grad_norm": 0.9047350844927557,
      "learning_rate": 5e-06,
      "loss": 0.4491,
      "step": 4220
    },
    {
      "epoch": 7.701411015020483,
      "grad_norm": 0.8935472556665444,
      "learning_rate": 5e-06,
      "loss": 0.4483,
      "step": 4230
    },
    {
      "epoch": 7.719617660446063,
      "grad_norm": 0.8653782388568667,
      "learning_rate": 5e-06,
      "loss": 0.4475,
      "step": 4240
    },
    {
      "epoch": 7.737824305871643,
      "grad_norm": 0.881016800265456,
      "learning_rate": 5e-06,
      "loss": 0.451,
      "step": 4250
    },
    {
      "epoch": 7.756030951297223,
      "grad_norm": 0.9020813961697224,
      "learning_rate": 5e-06,
      "loss": 0.448,
      "step": 4260
    },
    {
      "epoch": 7.774237596722804,
      "grad_norm": 0.9716168118712624,
      "learning_rate": 5e-06,
      "loss": 0.4523,
      "step": 4270
    },
    {
      "epoch": 7.792444242148385,
      "grad_norm": 0.9179534064504127,
      "learning_rate": 5e-06,
      "loss": 0.4515,
      "step": 4280
    },
    {
      "epoch": 7.810650887573964,
      "grad_norm": 0.9614798912890371,
      "learning_rate": 5e-06,
      "loss": 0.448,
      "step": 4290
    },
    {
      "epoch": 7.828857532999545,
      "grad_norm": 0.9044673066961919,
      "learning_rate": 5e-06,
      "loss": 0.452,
      "step": 4300
    },
    {
      "epoch": 7.847064178425125,
      "grad_norm": 0.9051177289315938,
      "learning_rate": 5e-06,
      "loss": 0.4502,
      "step": 4310
    },
    {
      "epoch": 7.865270823850706,
      "grad_norm": 0.9751136269675025,
      "learning_rate": 5e-06,
      "loss": 0.4508,
      "step": 4320
    },
    {
      "epoch": 7.883477469276286,
      "grad_norm": 0.9479807771925007,
      "learning_rate": 5e-06,
      "loss": 0.4481,
      "step": 4330
    },
    {
      "epoch": 7.901684114701866,
      "grad_norm": 1.0019388262723266,
      "learning_rate": 5e-06,
      "loss": 0.4508,
      "step": 4340
    },
    {
      "epoch": 7.919890760127447,
      "grad_norm": 0.8485770165242006,
      "learning_rate": 5e-06,
      "loss": 0.4488,
      "step": 4350
    },
    {
      "epoch": 7.938097405553027,
      "grad_norm": 0.9440844580758145,
      "learning_rate": 5e-06,
      "loss": 0.453,
      "step": 4360
    },
    {
      "epoch": 7.956304050978607,
      "grad_norm": 0.9450193155443942,
      "learning_rate": 5e-06,
      "loss": 0.4519,
      "step": 4370
    },
    {
      "epoch": 7.974510696404187,
      "grad_norm": 0.9124151727103043,
      "learning_rate": 5e-06,
      "loss": 0.4542,
      "step": 4380
    },
    {
      "epoch": 7.992717341829768,
      "grad_norm": 0.9474262270163373,
      "learning_rate": 5e-06,
      "loss": 0.4505,
      "step": 4390
    },
    {
      "epoch": 7.9963586709148835,
      "eval_loss": 0.9232881665229797,
      "eval_runtime": 369.711,
      "eval_samples_per_second": 40.034,
      "eval_steps_per_second": 0.628,
      "step": 4392
    },
    {
      "epoch": 7.9963586709148835,
      "step": 4392,
      "total_flos": 7356554202316800.0,
      "train_loss": 0.6099752168050664,
      "train_runtime": 144891.7905,
      "train_samples_per_second": 15.527,
      "train_steps_per_second": 0.03
    }
  ],
  "logging_steps": 10,
  "max_steps": 4392,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 8,
  "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": 7356554202316800.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}