{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 49.6,
  "eval_steps": 500,
  "global_step": 3100,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.32,
      "grad_norm": 2.214611053466797,
      "learning_rate": 3.064516129032258e-06,
      "loss": 2.8022,
      "step": 20
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.9293187260627747,
      "learning_rate": 6.290322580645161e-06,
      "loss": 2.6463,
      "step": 40
    },
    {
      "epoch": 0.96,
      "grad_norm": 2.630194902420044,
      "learning_rate": 9.516129032258064e-06,
      "loss": 2.4026,
      "step": 60
    },
    {
      "epoch": 1.28,
      "grad_norm": 1.7517138719558716,
      "learning_rate": 1.2741935483870968e-05,
      "loss": 1.9639,
      "step": 80
    },
    {
      "epoch": 1.6,
      "grad_norm": 1.6100302934646606,
      "learning_rate": 1.596774193548387e-05,
      "loss": 1.7115,
      "step": 100
    },
    {
      "epoch": 1.92,
      "grad_norm": 3.468630790710449,
      "learning_rate": 1.9193548387096774e-05,
      "loss": 1.7534,
      "step": 120
    },
    {
      "epoch": 2.24,
      "grad_norm": 2.6692144870758057,
      "learning_rate": 2.2419354838709678e-05,
      "loss": 1.3422,
      "step": 140
    },
    {
      "epoch": 2.56,
      "grad_norm": 1.5186293125152588,
      "learning_rate": 2.5645161290322582e-05,
      "loss": 1.1582,
      "step": 160
    },
    {
      "epoch": 2.88,
      "grad_norm": 4.727887153625488,
      "learning_rate": 2.8870967741935483e-05,
      "loss": 1.1084,
      "step": 180
    },
    {
      "epoch": 3.2,
      "grad_norm": 2.800290107727051,
      "learning_rate": 3.2096774193548393e-05,
      "loss": 0.844,
      "step": 200
    },
    {
      "epoch": 3.52,
      "grad_norm": 1.1538563966751099,
      "learning_rate": 3.532258064516129e-05,
      "loss": 0.6051,
      "step": 220
    },
    {
      "epoch": 3.84,
      "grad_norm": 3.282315492630005,
      "learning_rate": 3.8548387096774195e-05,
      "loss": 0.7587,
      "step": 240
    },
    {
      "epoch": 4.16,
      "grad_norm": 0.9393389821052551,
      "learning_rate": 4.17741935483871e-05,
      "loss": 0.537,
      "step": 260
    },
    {
      "epoch": 4.48,
      "grad_norm": 1.5951660871505737,
      "learning_rate": 4.5e-05,
      "loss": 0.4063,
      "step": 280
    },
    {
      "epoch": 4.8,
      "grad_norm": 2.4143407344818115,
      "learning_rate": 4.822580645161291e-05,
      "loss": 0.4817,
      "step": 300
    },
    {
      "epoch": 5.12,
      "grad_norm": 1.1503572463989258,
      "learning_rate": 4.9998716243505096e-05,
      "loss": 0.3556,
      "step": 320
    },
    {
      "epoch": 5.44,
      "grad_norm": 1.5514352321624756,
      "learning_rate": 4.9986672191133314e-05,
      "loss": 0.2663,
      "step": 340
    },
    {
      "epoch": 5.76,
      "grad_norm": 2.0045723915100098,
      "learning_rate": 4.9961956248762694e-05,
      "loss": 0.3273,
      "step": 360
    },
    {
      "epoch": 6.08,
      "grad_norm": 0.716200053691864,
      "learning_rate": 4.992458095098368e-05,
      "loss": 0.2174,
      "step": 380
    },
    {
      "epoch": 6.4,
      "grad_norm": 3.2870705127716064,
      "learning_rate": 4.9874565252527765e-05,
      "loss": 0.1886,
      "step": 400
    },
    {
      "epoch": 6.72,
      "grad_norm": 2.3181071281433105,
      "learning_rate": 4.981193451865465e-05,
      "loss": 0.2278,
      "step": 420
    },
    {
      "epoch": 7.04,
      "grad_norm": 1.6296441555023193,
      "learning_rate": 4.9736720512288334e-05,
      "loss": 0.168,
      "step": 440
    },
    {
      "epoch": 7.36,
      "grad_norm": 1.757379174232483,
      "learning_rate": 4.964896137790873e-05,
      "loss": 0.1227,
      "step": 460
    },
    {
      "epoch": 7.68,
      "grad_norm": 1.2622524499893188,
      "learning_rate": 4.954870162220679e-05,
      "loss": 0.1261,
      "step": 480
    },
    {
      "epoch": 8.0,
      "grad_norm": 1.1865277290344238,
      "learning_rate": 4.943599209151314e-05,
      "loss": 0.1167,
      "step": 500
    },
    {
      "epoch": 8.32,
      "grad_norm": 1.244333028793335,
      "learning_rate": 4.931088994601157e-05,
      "loss": 0.1049,
      "step": 520
    },
    {
      "epoch": 8.64,
      "grad_norm": 2.6851558685302734,
      "learning_rate": 4.917345863075048e-05,
      "loss": 0.1016,
      "step": 540
    },
    {
      "epoch": 8.96,
      "grad_norm": 1.0457267761230469,
      "learning_rate": 4.902376784346697e-05,
      "loss": 0.1157,
      "step": 560
    },
    {
      "epoch": 9.28,
      "grad_norm": 0.2963043749332428,
      "learning_rate": 4.886189349923992e-05,
      "loss": 0.073,
      "step": 580
    },
    {
      "epoch": 9.6,
      "grad_norm": 0.45336633920669556,
      "learning_rate": 4.868791769198995e-05,
      "loss": 0.0908,
      "step": 600
    },
    {
      "epoch": 9.92,
      "grad_norm": 1.0918829441070557,
      "learning_rate": 4.8501928652845854e-05,
      "loss": 0.0557,
      "step": 620
    },
    {
      "epoch": 10.24,
      "grad_norm": 2.7920358180999756,
      "learning_rate": 4.83040207053985e-05,
      "loss": 0.0779,
      "step": 640
    },
    {
      "epoch": 10.56,
      "grad_norm": 0.8492644429206848,
      "learning_rate": 4.809429421786502e-05,
      "loss": 0.048,
      "step": 660
    },
    {
      "epoch": 10.88,
      "grad_norm": 0.6128495335578918,
      "learning_rate": 4.787285555218748e-05,
      "loss": 0.0747,
      "step": 680
    },
    {
      "epoch": 11.2,
      "grad_norm": 0.5186921954154968,
      "learning_rate": 4.763981701009184e-05,
      "loss": 0.0629,
      "step": 700
    },
    {
      "epoch": 11.52,
      "grad_norm": 1.6753857135772705,
      "learning_rate": 4.739529677613456e-05,
      "loss": 0.051,
      "step": 720
    },
    {
      "epoch": 11.84,
      "grad_norm": 0.18876530230045319,
      "learning_rate": 4.713941885776586e-05,
      "loss": 0.0699,
      "step": 740
    },
    {
      "epoch": 12.16,
      "grad_norm": 0.4428744912147522,
      "learning_rate": 4.687231302243975e-05,
      "loss": 0.0526,
      "step": 760
    },
    {
      "epoch": 12.48,
      "grad_norm": 0.4765178859233856,
      "learning_rate": 4.659411473180304e-05,
      "loss": 0.0412,
      "step": 780
    },
    {
      "epoch": 12.8,
      "grad_norm": 1.4559166431427002,
      "learning_rate": 4.6304965072996495e-05,
      "loss": 0.0495,
      "step": 800
    },
    {
      "epoch": 13.12,
      "grad_norm": 1.1842377185821533,
      "learning_rate": 4.6005010687103076e-05,
      "loss": 0.063,
      "step": 820
    },
    {
      "epoch": 13.44,
      "grad_norm": 0.5502442717552185,
      "learning_rate": 4.569440369477951e-05,
      "loss": 0.0425,
      "step": 840
    },
    {
      "epoch": 13.76,
      "grad_norm": 1.819698452949524,
      "learning_rate": 4.5373301619108854e-05,
      "loss": 0.0451,
      "step": 860
    },
    {
      "epoch": 14.08,
      "grad_norm": 0.045023053884506226,
      "learning_rate": 4.5041867305713384e-05,
      "loss": 0.0445,
      "step": 880
    },
    {
      "epoch": 14.4,
      "grad_norm": 0.10354474931955338,
      "learning_rate": 4.4700268840168045e-05,
      "loss": 0.0214,
      "step": 900
    },
    {
      "epoch": 14.72,
      "grad_norm": 0.23762211203575134,
      "learning_rate": 4.4348679462756556e-05,
      "loss": 0.0552,
      "step": 920
    },
    {
      "epoch": 15.04,
      "grad_norm": 0.845052182674408,
      "learning_rate": 4.398727748061324e-05,
      "loss": 0.0524,
      "step": 940
    },
    {
      "epoch": 15.36,
      "grad_norm": 0.29455024003982544,
      "learning_rate": 4.361624617729536e-05,
      "loss": 0.0318,
      "step": 960
    },
    {
      "epoch": 15.68,
      "grad_norm": 0.05131356045603752,
      "learning_rate": 4.323577371983155e-05,
      "loss": 0.0347,
      "step": 980
    },
    {
      "epoch": 16.0,
      "grad_norm": 0.2307986468076706,
      "learning_rate": 4.28460530632937e-05,
      "loss": 0.0541,
      "step": 1000
    },
    {
      "epoch": 16.32,
      "grad_norm": 0.9982470870018005,
      "learning_rate": 4.2447281852940525e-05,
      "loss": 0.0327,
      "step": 1020
    },
    {
      "epoch": 16.64,
      "grad_norm": 2.9247119426727295,
      "learning_rate": 4.203966232398261e-05,
      "loss": 0.0251,
      "step": 1040
    },
    {
      "epoch": 16.96,
      "grad_norm": 0.5199835896492004,
      "learning_rate": 4.162340119901961e-05,
      "loss": 0.0451,
      "step": 1060
    },
    {
      "epoch": 17.28,
      "grad_norm": 0.6802399754524231,
      "learning_rate": 4.1198709583201754e-05,
      "loss": 0.0272,
      "step": 1080
    },
    {
      "epoch": 17.6,
      "grad_norm": 0.9806874394416809,
      "learning_rate": 4.0765802857168687e-05,
      "loss": 0.0517,
      "step": 1100
    },
    {
      "epoch": 17.92,
      "grad_norm": 0.12717130780220032,
      "learning_rate": 4.0324900567820046e-05,
      "loss": 0.0286,
      "step": 1120
    },
    {
      "epoch": 18.24,
      "grad_norm": 0.4048568308353424,
      "learning_rate": 3.987622631697316e-05,
      "loss": 0.0258,
      "step": 1140
    },
    {
      "epoch": 18.56,
      "grad_norm": 1.2760275602340698,
      "learning_rate": 3.942000764796427e-05,
      "loss": 0.0289,
      "step": 1160
    },
    {
      "epoch": 18.88,
      "grad_norm": 0.30493712425231934,
      "learning_rate": 3.895647593025088e-05,
      "loss": 0.0457,
      "step": 1180
    },
    {
      "epoch": 19.2,
      "grad_norm": 0.6177706122398376,
      "learning_rate": 3.8485866242073584e-05,
      "loss": 0.0316,
      "step": 1200
    },
    {
      "epoch": 19.52,
      "grad_norm": 0.10721703618764877,
      "learning_rate": 3.80084172512372e-05,
      "loss": 0.0326,
      "step": 1220
    },
    {
      "epoch": 19.84,
      "grad_norm": 0.06823062896728516,
      "learning_rate": 3.7524371094071266e-05,
      "loss": 0.0238,
      "step": 1240
    },
    {
      "epoch": 20.16,
      "grad_norm": 0.9792996644973755,
      "learning_rate": 3.703397325263162e-05,
      "loss": 0.0286,
      "step": 1260
    },
    {
      "epoch": 20.48,
      "grad_norm": 0.20131415128707886,
      "learning_rate": 3.653747243020515e-05,
      "loss": 0.0294,
      "step": 1280
    },
    {
      "epoch": 20.8,
      "grad_norm": 0.530823826789856,
      "learning_rate": 3.603512042518093e-05,
      "loss": 0.0364,
      "step": 1300
    },
    {
      "epoch": 21.12,
      "grad_norm": 0.006112121045589447,
      "learning_rate": 3.552717200335171e-05,
      "loss": 0.0265,
      "step": 1320
    },
    {
      "epoch": 21.44,
      "grad_norm": 0.005943021737039089,
      "learning_rate": 3.501388476871039e-05,
      "loss": 0.0319,
      "step": 1340
    },
    {
      "epoch": 21.76,
      "grad_norm": 0.07666248083114624,
      "learning_rate": 3.449551903280729e-05,
      "loss": 0.0137,
      "step": 1360
    },
    {
      "epoch": 22.08,
      "grad_norm": 0.6763716340065002,
      "learning_rate": 3.397233768273415e-05,
      "loss": 0.0416,
      "step": 1380
    },
    {
      "epoch": 22.4,
      "grad_norm": 0.07458912581205368,
      "learning_rate": 3.344460604780202e-05,
      "loss": 0.0179,
      "step": 1400
    },
    {
      "epoch": 22.72,
      "grad_norm": 0.25514400005340576,
      "learning_rate": 3.291259176498052e-05,
      "loss": 0.0276,
      "step": 1420
    },
    {
      "epoch": 23.04,
      "grad_norm": 0.006252670660614967,
      "learning_rate": 3.237656464316693e-05,
      "loss": 0.0352,
      "step": 1440
    },
    {
      "epoch": 23.36,
      "grad_norm": 0.36337536573410034,
      "learning_rate": 3.183679652635357e-05,
      "loss": 0.0212,
      "step": 1460
    },
    {
      "epoch": 23.68,
      "grad_norm": 0.28176209330558777,
      "learning_rate": 3.129356115576332e-05,
      "loss": 0.0338,
      "step": 1480
    },
    {
      "epoch": 24.0,
      "grad_norm": 0.012418941594660282,
      "learning_rate": 3.074713403102284e-05,
      "loss": 0.0295,
      "step": 1500
    },
    {
      "epoch": 24.32,
      "grad_norm": 0.16376622021198273,
      "learning_rate": 3.0197792270443982e-05,
      "loss": 0.0185,
      "step": 1520
    },
    {
      "epoch": 24.64,
      "grad_norm": 0.45608577132225037,
      "learning_rate": 2.9645814470484452e-05,
      "loss": 0.0328,
      "step": 1540
    },
    {
      "epoch": 24.96,
      "grad_norm": 0.1006656065583229,
      "learning_rate": 2.9091480564458666e-05,
      "loss": 0.025,
      "step": 1560
    },
    {
      "epoch": 25.28,
      "grad_norm": 0.3901682496070862,
      "learning_rate": 2.8535071680570734e-05,
      "loss": 0.0294,
      "step": 1580
    },
    {
      "epoch": 25.6,
      "grad_norm": 0.12243347615003586,
      "learning_rate": 2.7976869999341426e-05,
      "loss": 0.0282,
      "step": 1600
    },
    {
      "epoch": 25.92,
      "grad_norm": 0.0033943182788789272,
      "learning_rate": 2.741715861050143e-05,
      "loss": 0.0294,
      "step": 1620
    },
    {
      "epoch": 26.24,
      "grad_norm": 0.0014275741996243596,
      "learning_rate": 2.685622136942359e-05,
      "loss": 0.0354,
      "step": 1640
    },
    {
      "epoch": 26.56,
      "grad_norm": 0.018641650676727295,
      "learning_rate": 2.629434275316673e-05,
      "loss": 0.0162,
      "step": 1660
    },
    {
      "epoch": 26.88,
      "grad_norm": 0.13316482305526733,
      "learning_rate": 2.573180771620432e-05,
      "loss": 0.0205,
      "step": 1680
    },
    {
      "epoch": 27.2,
      "grad_norm": 0.003175324061885476,
      "learning_rate": 2.516890154591095e-05,
      "loss": 0.0129,
      "step": 1700
    },
    {
      "epoch": 27.52,
      "grad_norm": 0.28820428252220154,
      "learning_rate": 2.4605909717879964e-05,
      "loss": 0.0333,
      "step": 1720
    },
    {
      "epoch": 27.84,
      "grad_norm": 0.08302447199821472,
      "learning_rate": 2.4043117751145694e-05,
      "loss": 0.0261,
      "step": 1740
    },
    {
      "epoch": 28.16,
      "grad_norm": 0.37718892097473145,
      "learning_rate": 2.34808110633836e-05,
      "loss": 0.0418,
      "step": 1760
    },
    {
      "epoch": 28.48,
      "grad_norm": 0.16185913980007172,
      "learning_rate": 2.291927482616191e-05,
      "loss": 0.0111,
      "step": 1780
    },
    {
      "epoch": 28.8,
      "grad_norm": 0.33071696758270264,
      "learning_rate": 2.235879382031794e-05,
      "loss": 0.0263,
      "step": 1800
    },
    {
      "epoch": 29.12,
      "grad_norm": 0.002949915360659361,
      "learning_rate": 2.179965229153265e-05,
      "loss": 0.0299,
      "step": 1820
    },
    {
      "epoch": 29.44,
      "grad_norm": 0.0035819699987769127,
      "learning_rate": 2.1242133806176667e-05,
      "loss": 0.0267,
      "step": 1840
    },
    {
      "epoch": 29.76,
      "grad_norm": 0.0039703804068267345,
      "learning_rate": 2.0686521107500638e-05,
      "loss": 0.0204,
      "step": 1860
    },
    {
      "epoch": 30.08,
      "grad_norm": 0.004021900240331888,
      "learning_rate": 2.0133095972243233e-05,
      "loss": 0.0308,
      "step": 1880
    },
    {
      "epoch": 30.4,
      "grad_norm": 0.2821226418018341,
      "learning_rate": 1.9582139067729117e-05,
      "loss": 0.024,
      "step": 1900
    },
    {
      "epoch": 30.72,
      "grad_norm": 0.5529562830924988,
      "learning_rate": 1.90339298095297e-05,
      "loss": 0.0113,
      "step": 1920
    },
    {
      "epoch": 31.04,
      "grad_norm": 0.1492016613483429,
      "learning_rate": 1.8488746219758674e-05,
      "loss": 0.0301,
      "step": 1940
    },
    {
      "epoch": 31.36,
      "grad_norm": 0.19194553792476654,
      "learning_rate": 1.7946864786074165e-05,
      "loss": 0.0293,
      "step": 1960
    },
    {
      "epoch": 31.68,
      "grad_norm": 0.1448647677898407,
      "learning_rate": 1.740856032145917e-05,
      "loss": 0.0242,
      "step": 1980
    },
    {
      "epoch": 32.0,
      "grad_norm": 0.06869282573461533,
      "learning_rate": 1.6874105824851267e-05,
      "loss": 0.022,
      "step": 2000
    },
    {
      "epoch": 32.32,
      "grad_norm": 0.002635813085362315,
      "learning_rate": 1.634377234269226e-05,
      "loss": 0.0264,
      "step": 2020
    },
    {
      "epoch": 32.64,
      "grad_norm": 0.0027471587527543306,
      "learning_rate": 1.5817828831468144e-05,
      "loss": 0.0155,
      "step": 2040
    },
    {
      "epoch": 32.96,
      "grad_norm": 0.28300318121910095,
      "learning_rate": 1.5296542021308825e-05,
      "loss": 0.0208,
      "step": 2060
    },
    {
      "epoch": 33.28,
      "grad_norm": 0.0016769981011748314,
      "learning_rate": 1.478017628071706e-05,
      "loss": 0.0264,
      "step": 2080
    },
    {
      "epoch": 33.6,
      "grad_norm": 0.0014146752655506134,
      "learning_rate": 1.4268993482495055e-05,
      "loss": 0.0174,
      "step": 2100
    },
    {
      "epoch": 33.92,
      "grad_norm": 0.05964767187833786,
      "learning_rate": 1.3763252870936649e-05,
      "loss": 0.0214,
      "step": 2120
    },
    {
      "epoch": 34.24,
      "grad_norm": 0.0016291196225211024,
      "learning_rate": 1.3263210930352737e-05,
      "loss": 0.0306,
      "step": 2140
    },
    {
      "epoch": 34.56,
      "grad_norm": 0.034806057810783386,
      "learning_rate": 1.2769121254996159e-05,
      "loss": 0.0146,
      "step": 2160
    },
    {
      "epoch": 34.88,
      "grad_norm": 0.315729558467865,
      "learning_rate": 1.228123442045249e-05,
      "loss": 0.0254,
      "step": 2180
    },
    {
      "epoch": 35.2,
      "grad_norm": 0.5809018015861511,
      "learning_rate": 1.1799797856561606e-05,
      "loss": 0.0176,
      "step": 2200
    },
    {
      "epoch": 35.52,
      "grad_norm": 0.18777510523796082,
      "learning_rate": 1.1325055721934637e-05,
      "loss": 0.0205,
      "step": 2220
    },
    {
      "epoch": 35.84,
      "grad_norm": 0.11027589440345764,
      "learning_rate": 1.0857248780129928e-05,
      "loss": 0.0153,
      "step": 2240
    },
    {
      "epoch": 36.16,
      "grad_norm": 0.18985402584075928,
      "learning_rate": 1.0396614277550752e-05,
      "loss": 0.0251,
      "step": 2260
    },
    {
      "epoch": 36.48,
      "grad_norm": 0.011078303679823875,
      "learning_rate": 9.943385823126775e-06,
      "loss": 0.0224,
      "step": 2280
    },
    {
      "epoch": 36.8,
      "grad_norm": 0.15364831686019897,
      "learning_rate": 9.497793269840211e-06,
      "loss": 0.0219,
      "step": 2300
    },
    {
      "epoch": 37.12,
      "grad_norm": 0.16373781859874725,
      "learning_rate": 9.06006259815683e-06,
      "loss": 0.021,
      "step": 2320
    },
    {
      "epoch": 37.44,
      "grad_norm": 0.41937291622161865,
      "learning_rate": 8.630415801420835e-06,
      "loss": 0.0236,
      "step": 2340
    },
    {
      "epoch": 37.76,
      "grad_norm": 0.4237123727798462,
      "learning_rate": 8.209070773271894e-06,
      "loss": 0.0216,
      "step": 2360
    },
    {
      "epoch": 38.08,
      "grad_norm": 0.08532612025737762,
      "learning_rate": 7.79624119714121e-06,
      "loss": 0.0268,
      "step": 2380
    },
    {
      "epoch": 38.4,
      "grad_norm": 0.2872686982154846,
      "learning_rate": 7.392136437882855e-06,
      "loss": 0.0306,
      "step": 2400
    },
    {
      "epoch": 38.72,
      "grad_norm": 0.2834513485431671,
      "learning_rate": 6.996961435595223e-06,
      "loss": 0.0224,
      "step": 2420
    },
    {
      "epoch": 39.04,
      "grad_norm": 0.0006759735988453031,
      "learning_rate": 6.610916601686481e-06,
      "loss": 0.0138,
      "step": 2440
    },
    {
      "epoch": 39.36,
      "grad_norm": 0.24796371161937714,
      "learning_rate": 6.234197717236742e-06,
      "loss": 0.0234,
      "step": 2460
    },
    {
      "epoch": 39.68,
      "grad_norm": 0.2657662332057953,
      "learning_rate": 5.866995833708464e-06,
      "loss": 0.0164,
      "step": 2480
    },
    {
      "epoch": 40.0,
      "grad_norm": 0.0017559522530063987,
      "learning_rate": 5.509497176055492e-06,
      "loss": 0.0169,
      "step": 2500
    },
    {
      "epoch": 40.32,
      "grad_norm": 0.18896016478538513,
      "learning_rate": 5.161883048279817e-06,
      "loss": 0.0116,
      "step": 2520
    },
    {
      "epoch": 40.64,
      "grad_norm": 0.2336779683828354,
      "learning_rate": 4.824329741483949e-06,
      "loss": 0.0219,
      "step": 2540
    },
    {
      "epoch": 40.96,
      "grad_norm": 0.16521525382995605,
      "learning_rate": 4.497008444465681e-06,
      "loss": 0.0366,
      "step": 2560
    },
    {
      "epoch": 41.28,
      "grad_norm": 0.0013411182444542646,
      "learning_rate": 4.180085156900274e-06,
      "loss": 0.0228,
      "step": 2580
    },
    {
      "epoch": 41.6,
      "grad_norm": 0.0015802403213456273,
      "learning_rate": 3.873720605154468e-06,
      "loss": 0.0135,
      "step": 2600
    },
    {
      "epoch": 41.92,
      "grad_norm": 0.002210975391790271,
      "learning_rate": 3.578070160774724e-06,
      "loss": 0.0267,
      "step": 2620
    },
    {
      "epoch": 42.24,
      "grad_norm": 0.23047080636024475,
      "learning_rate": 3.293283761691182e-06,
      "loss": 0.0218,
      "step": 2640
    },
    {
      "epoch": 42.56,
      "grad_norm": 0.2625073492527008,
      "learning_rate": 3.0195058361772277e-06,
      "loss": 0.0315,
      "step": 2660
    },
    {
      "epoch": 42.88,
      "grad_norm": 0.22485554218292236,
      "learning_rate": 2.756875229603295e-06,
      "loss": 0.022,
      "step": 2680
    },
    {
      "epoch": 43.2,
      "grad_norm": 0.001550053246319294,
      "learning_rate": 2.5055251340219855e-06,
      "loss": 0.016,
      "step": 2700
    },
    {
      "epoch": 43.52,
      "grad_norm": 0.22042883932590485,
      "learning_rate": 2.2655830206202655e-06,
      "loss": 0.0267,
      "step": 2720
    },
    {
      "epoch": 43.84,
      "grad_norm": 0.08822102099657059,
      "learning_rate": 2.037170575072944e-06,
      "loss": 0.0167,
      "step": 2740
    },
    {
      "epoch": 44.16,
      "grad_norm": 0.0016505387611687183,
      "learning_rate": 1.8204036358303173e-06,
      "loss": 0.0492,
      "step": 2760
    },
    {
      "epoch": 44.48,
      "grad_norm": 0.3908143639564514,
      "learning_rate": 1.615392135371116e-06,
      "loss": 0.0254,
      "step": 2780
    },
    {
      "epoch": 44.8,
      "grad_norm": 0.3156016767024994,
      "learning_rate": 1.4222400444507318e-06,
      "loss": 0.0136,
      "step": 2800
    },
    {
      "epoch": 45.12,
      "grad_norm": 0.001897096517495811,
      "learning_rate": 1.2410453193728493e-06,
      "loss": 0.0116,
      "step": 2820
    },
    {
      "epoch": 45.44,
      "grad_norm": 0.20051489770412445,
      "learning_rate": 1.0718998523113004e-06,
      "loss": 0.0311,
      "step": 2840
    },
    {
      "epoch": 45.76,
      "grad_norm": 0.0016040581976994872,
      "learning_rate": 9.148894247073298e-07,
      "loss": 0.0283,
      "step": 2860
    },
    {
      "epoch": 46.08,
      "grad_norm": 0.0015667045954614878,
      "learning_rate": 7.700936637658779e-07,
      "loss": 0.0186,
      "step": 2880
    },
    {
      "epoch": 46.4,
      "grad_norm": 0.04447433352470398,
      "learning_rate": 6.375860020729541e-07,
      "loss": 0.0229,
      "step": 2900
    },
    {
      "epoch": 46.72,
      "grad_norm": 0.20941714942455292,
      "learning_rate": 5.174336403546226e-07,
      "loss": 0.0268,
      "step": 2920
    },
    {
      "epoch": 47.04,
      "grad_norm": 0.0015096565475687385,
      "learning_rate": 4.096975133963954e-07,
      "loss": 0.0161,
      "step": 2940
    },
    {
      "epoch": 47.36,
      "grad_norm": 0.19307351112365723,
      "learning_rate": 3.144322591404292e-07,
      "loss": 0.0113,
      "step": 2960
    },
    {
      "epoch": 47.68,
      "grad_norm": 0.002177381655201316,
      "learning_rate": 2.316861909760909e-07,
      "loss": 0.0222,
      "step": 2980
    },
    {
      "epoch": 48.0,
      "grad_norm": 0.09334852546453476,
      "learning_rate": 1.6150127323803222e-07,
      "loss": 0.028,
      "step": 3000
    },
    {
      "epoch": 48.32,
      "grad_norm": 0.09507758915424347,
      "learning_rate": 1.0391309992413833e-07,
      "loss": 0.0217,
      "step": 3020
    },
    {
      "epoch": 48.64,
      "grad_norm": 0.0010601489339023829,
      "learning_rate": 5.895087664417876e-08,
      "loss": 0.0191,
      "step": 3040
    },
    {
      "epoch": 48.96,
      "grad_norm": 0.001158875529654324,
      "learning_rate": 2.6637405808302428e-08,
      "loss": 0.0162,
      "step": 3060
    },
    {
      "epoch": 49.28,
      "grad_norm": 0.0015347091248258948,
      "learning_rate": 6.989075062879824e-09,
      "loss": 0.0146,
      "step": 3080
    },
    {
      "epoch": 49.6,
      "grad_norm": 0.0010455228621140122,
      "learning_rate": 1.584897958428755e-11,
      "loss": 0.0217,
      "step": 3100
    },
    {
      "epoch": 49.6,
      "step": 3100,
      "total_flos": 1.7939255986343117e+17,
      "train_loss": 0.172668604437382,
      "train_runtime": 11274.7291,
      "train_samples_per_second": 2.217,
      "train_steps_per_second": 0.275
    }
  ],
  "logging_steps": 20,
  "max_steps": 3100,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 50,
  "save_steps": 1000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.7939255986343117e+17,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}