{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 25.0,
  "eval_steps": 500,
  "global_step": 1250,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.2,
      "grad_norm": 1.459547519683838,
      "learning_rate": 1.6000000000000001e-06,
      "loss": 1.2185,
      "step": 10
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.5558891296386719,
      "learning_rate": 3.2000000000000003e-06,
      "loss": 0.892,
      "step": 20
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.5772704482078552,
      "learning_rate": 4.800000000000001e-06,
      "loss": 0.7887,
      "step": 30
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.40292349457740784,
      "learning_rate": 6.4000000000000006e-06,
      "loss": 0.7111,
      "step": 40
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.35770532488822937,
      "learning_rate": 8.000000000000001e-06,
      "loss": 0.6489,
      "step": 50
    },
    {
      "epoch": 1.2,
      "grad_norm": 0.43718403577804565,
      "learning_rate": 9.600000000000001e-06,
      "loss": 0.5936,
      "step": 60
    },
    {
      "epoch": 1.4,
      "grad_norm": 0.4591154158115387,
      "learning_rate": 1.1200000000000001e-05,
      "loss": 0.6121,
      "step": 70
    },
    {
      "epoch": 1.6,
      "grad_norm": 0.6591519713401794,
      "learning_rate": 1.2800000000000001e-05,
      "loss": 0.6059,
      "step": 80
    },
    {
      "epoch": 1.8,
      "grad_norm": 0.6931334733963013,
      "learning_rate": 1.4400000000000001e-05,
      "loss": 0.6062,
      "step": 90
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.6464695334434509,
      "learning_rate": 1.6000000000000003e-05,
      "loss": 0.5929,
      "step": 100
    },
    {
      "epoch": 2.2,
      "grad_norm": 0.7549647688865662,
      "learning_rate": 1.76e-05,
      "loss": 0.5138,
      "step": 110
    },
    {
      "epoch": 2.4,
      "grad_norm": 0.6728184819221497,
      "learning_rate": 1.9200000000000003e-05,
      "loss": 0.4737,
      "step": 120
    },
    {
      "epoch": 2.6,
      "grad_norm": 0.7162397503852844,
      "learning_rate": 1.9999025240093045e-05,
      "loss": 0.4891,
      "step": 130
    },
    {
      "epoch": 2.8,
      "grad_norm": 0.9795845150947571,
      "learning_rate": 1.9991228300988586e-05,
      "loss": 0.4663,
      "step": 140
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.6311202049255371,
      "learning_rate": 1.9975640502598243e-05,
      "loss": 0.5164,
      "step": 150
    },
    {
      "epoch": 3.2,
      "grad_norm": 0.9567041993141174,
      "learning_rate": 1.9952273999818312e-05,
      "loss": 0.3981,
      "step": 160
    },
    {
      "epoch": 3.4,
      "grad_norm": 0.8767746090888977,
      "learning_rate": 1.9921147013144782e-05,
      "loss": 0.3761,
      "step": 170
    },
    {
      "epoch": 3.6,
      "grad_norm": 1.8661500215530396,
      "learning_rate": 1.988228381446553e-05,
      "loss": 0.374,
      "step": 180
    },
    {
      "epoch": 3.8,
      "grad_norm": 0.7793363332748413,
      "learning_rate": 1.983571470813386e-05,
      "loss": 0.3908,
      "step": 190
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.8413230180740356,
      "learning_rate": 1.9781476007338058e-05,
      "loss": 0.3527,
      "step": 200
    },
    {
      "epoch": 4.2,
      "grad_norm": 1.5790411233901978,
      "learning_rate": 1.9719610005785466e-05,
      "loss": 0.2756,
      "step": 210
    },
    {
      "epoch": 4.4,
      "grad_norm": 1.1819688081741333,
      "learning_rate": 1.9650164944723116e-05,
      "loss": 0.2622,
      "step": 220
    },
    {
      "epoch": 4.6,
      "grad_norm": 1.2115598917007446,
      "learning_rate": 1.9573194975320672e-05,
      "loss": 0.2475,
      "step": 230
    },
    {
      "epoch": 4.8,
      "grad_norm": 1.0625479221343994,
      "learning_rate": 1.9488760116444966e-05,
      "loss": 0.2682,
      "step": 240
    },
    {
      "epoch": 5.0,
      "grad_norm": 1.2376039028167725,
      "learning_rate": 1.9396926207859085e-05,
      "loss": 0.2514,
      "step": 250
    },
    {
      "epoch": 5.2,
      "grad_norm": 0.6343020796775818,
      "learning_rate": 1.9297764858882516e-05,
      "loss": 0.1448,
      "step": 260
    },
    {
      "epoch": 5.4,
      "grad_norm": 1.1657440662384033,
      "learning_rate": 1.9191353392552346e-05,
      "loss": 0.1725,
      "step": 270
    },
    {
      "epoch": 5.6,
      "grad_norm": 1.4713112115859985,
      "learning_rate": 1.907777478532909e-05,
      "loss": 0.1753,
      "step": 280
    },
    {
      "epoch": 5.8,
      "grad_norm": 1.9010937213897705,
      "learning_rate": 1.895711760239413e-05,
      "loss": 0.1697,
      "step": 290
    },
    {
      "epoch": 6.0,
      "grad_norm": 2.2462244033813477,
      "learning_rate": 1.8829475928589272e-05,
      "loss": 0.157,
      "step": 300
    },
    {
      "epoch": 6.2,
      "grad_norm": 1.5734026432037354,
      "learning_rate": 1.869494929505219e-05,
      "loss": 0.0987,
      "step": 310
    },
    {
      "epoch": 6.4,
      "grad_norm": 1.6871922016143799,
      "learning_rate": 1.855364260160507e-05,
      "loss": 0.1122,
      "step": 320
    },
    {
      "epoch": 6.6,
      "grad_norm": 0.8505676984786987,
      "learning_rate": 1.8405666034956842e-05,
      "loss": 0.0999,
      "step": 330
    },
    {
      "epoch": 6.8,
      "grad_norm": 1.9714100360870361,
      "learning_rate": 1.8251134982782952e-05,
      "loss": 0.0925,
      "step": 340
    },
    {
      "epoch": 7.0,
      "grad_norm": 1.3289463520050049,
      "learning_rate": 1.8090169943749477e-05,
      "loss": 0.1034,
      "step": 350
    },
    {
      "epoch": 7.2,
      "grad_norm": 1.7757859230041504,
      "learning_rate": 1.792289643355191e-05,
      "loss": 0.0538,
      "step": 360
    },
    {
      "epoch": 7.4,
      "grad_norm": 1.5128904581069946,
      "learning_rate": 1.7749444887041797e-05,
      "loss": 0.0562,
      "step": 370
    },
    {
      "epoch": 7.6,
      "grad_norm": 1.7095279693603516,
      "learning_rate": 1.7569950556517566e-05,
      "loss": 0.0656,
      "step": 380
    },
    {
      "epoch": 7.8,
      "grad_norm": 1.7054250240325928,
      "learning_rate": 1.7384553406258842e-05,
      "loss": 0.0783,
      "step": 390
    },
    {
      "epoch": 8.0,
      "grad_norm": 1.2087674140930176,
      "learning_rate": 1.7193398003386514e-05,
      "loss": 0.0832,
      "step": 400
    },
    {
      "epoch": 8.2,
      "grad_norm": 1.2171767950057983,
      "learning_rate": 1.6996633405133656e-05,
      "loss": 0.0389,
      "step": 410
    },
    {
      "epoch": 8.4,
      "grad_norm": 1.2036916017532349,
      "learning_rate": 1.6794413042615168e-05,
      "loss": 0.0456,
      "step": 420
    },
    {
      "epoch": 8.6,
      "grad_norm": 1.6378772258758545,
      "learning_rate": 1.6586894601186804e-05,
      "loss": 0.0539,
      "step": 430
    },
    {
      "epoch": 8.8,
      "grad_norm": 1.062146544456482,
      "learning_rate": 1.63742398974869e-05,
      "loss": 0.0351,
      "step": 440
    },
    {
      "epoch": 9.0,
      "grad_norm": 0.7032294273376465,
      "learning_rate": 1.6156614753256583e-05,
      "loss": 0.0331,
      "step": 450
    },
    {
      "epoch": 9.2,
      "grad_norm": 1.2708150148391724,
      "learning_rate": 1.5934188866037017e-05,
      "loss": 0.0298,
      "step": 460
    },
    {
      "epoch": 9.4,
      "grad_norm": 0.5238479375839233,
      "learning_rate": 1.570713567684432e-05,
      "loss": 0.0271,
      "step": 470
    },
    {
      "epoch": 9.6,
      "grad_norm": 0.6896675229072571,
      "learning_rate": 1.5475632234925505e-05,
      "loss": 0.0148,
      "step": 480
    },
    {
      "epoch": 9.8,
      "grad_norm": 0.8893882632255554,
      "learning_rate": 1.5239859059700794e-05,
      "loss": 0.0384,
      "step": 490
    },
    {
      "epoch": 10.0,
      "grad_norm": 1.3057194948196411,
      "learning_rate": 1.5000000000000002e-05,
      "loss": 0.0216,
      "step": 500
    },
    {
      "epoch": 10.2,
      "grad_norm": 0.4277481436729431,
      "learning_rate": 1.4756242090702756e-05,
      "loss": 0.0186,
      "step": 510
    },
    {
      "epoch": 10.4,
      "grad_norm": 2.500697374343872,
      "learning_rate": 1.4508775406894308e-05,
      "loss": 0.0289,
      "step": 520
    },
    {
      "epoch": 10.6,
      "grad_norm": 0.4189341366291046,
      "learning_rate": 1.4257792915650728e-05,
      "loss": 0.0161,
      "step": 530
    },
    {
      "epoch": 10.8,
      "grad_norm": 0.6834658980369568,
      "learning_rate": 1.4003490325568953e-05,
      "loss": 0.016,
      "step": 540
    },
    {
      "epoch": 11.0,
      "grad_norm": 0.32209300994873047,
      "learning_rate": 1.3746065934159123e-05,
      "loss": 0.0214,
      "step": 550
    },
    {
      "epoch": 11.2,
      "grad_norm": 0.6142628788948059,
      "learning_rate": 1.3485720473218153e-05,
      "loss": 0.0169,
      "step": 560
    },
    {
      "epoch": 11.4,
      "grad_norm": 0.020493434742093086,
      "learning_rate": 1.3222656952305113e-05,
      "loss": 0.0072,
      "step": 570
    },
    {
      "epoch": 11.6,
      "grad_norm": 0.6618867516517639,
      "learning_rate": 1.2957080500440469e-05,
      "loss": 0.0134,
      "step": 580
    },
    {
      "epoch": 11.8,
      "grad_norm": 2.014190196990967,
      "learning_rate": 1.2689198206152657e-05,
      "loss": 0.0141,
      "step": 590
    },
    {
      "epoch": 12.0,
      "grad_norm": 0.3526052236557007,
      "learning_rate": 1.2419218955996677e-05,
      "loss": 0.0152,
      "step": 600
    },
    {
      "epoch": 12.2,
      "grad_norm": 0.39659520983695984,
      "learning_rate": 1.2147353271670634e-05,
      "loss": 0.0082,
      "step": 610
    },
    {
      "epoch": 12.4,
      "grad_norm": 1.6676621437072754,
      "learning_rate": 1.187381314585725e-05,
      "loss": 0.0086,
      "step": 620
    },
    {
      "epoch": 12.6,
      "grad_norm": 0.48944175243377686,
      "learning_rate": 1.159881187691835e-05,
      "loss": 0.0073,
      "step": 630
    },
    {
      "epoch": 12.8,
      "grad_norm": 0.11040916293859482,
      "learning_rate": 1.1322563902571227e-05,
      "loss": 0.0111,
      "step": 640
    },
    {
      "epoch": 13.0,
      "grad_norm": 0.19332852959632874,
      "learning_rate": 1.1045284632676535e-05,
      "loss": 0.0084,
      "step": 650
    },
    {
      "epoch": 13.2,
      "grad_norm": 0.5781747698783875,
      "learning_rate": 1.0767190281268187e-05,
      "loss": 0.0088,
      "step": 660
    },
    {
      "epoch": 13.4,
      "grad_norm": 0.5372014045715332,
      "learning_rate": 1.0488497697956134e-05,
      "loss": 0.0129,
      "step": 670
    },
    {
      "epoch": 13.6,
      "grad_norm": 0.0674474835395813,
      "learning_rate": 1.0209424198833571e-05,
      "loss": 0.0047,
      "step": 680
    },
    {
      "epoch": 13.8,
      "grad_norm": 0.0646585077047348,
      "learning_rate": 9.930187397020385e-06,
      "loss": 0.0079,
      "step": 690
    },
    {
      "epoch": 14.0,
      "grad_norm": 0.16117724776268005,
      "learning_rate": 9.651005032974994e-06,
      "loss": 0.0137,
      "step": 700
    },
    {
      "epoch": 14.2,
      "grad_norm": 0.031529657542705536,
      "learning_rate": 9.372094804706867e-06,
      "loss": 0.0195,
      "step": 710
    },
    {
      "epoch": 14.4,
      "grad_norm": 0.7396082878112793,
      "learning_rate": 9.093674198022201e-06,
      "loss": 0.0038,
      "step": 720
    },
    {
      "epoch": 14.6,
      "grad_norm": 0.131903737783432,
      "learning_rate": 8.815960316934991e-06,
      "loss": 0.009,
      "step": 730
    },
    {
      "epoch": 14.8,
      "grad_norm": 0.031985681504011154,
      "learning_rate": 8.539169714375885e-06,
      "loss": 0.0013,
      "step": 740
    },
    {
      "epoch": 15.0,
      "grad_norm": 0.12234848737716675,
      "learning_rate": 8.263518223330698e-06,
      "loss": 0.0043,
      "step": 750
    },
    {
      "epoch": 15.2,
      "grad_norm": 0.10590848326683044,
      "learning_rate": 7.989220788540356e-06,
      "loss": 0.0016,
      "step": 760
    },
    {
      "epoch": 15.4,
      "grad_norm": 0.20384103059768677,
      "learning_rate": 7.716491298893443e-06,
      "loss": 0.0011,
      "step": 770
    },
    {
      "epoch": 15.6,
      "grad_norm": 0.16497106850147247,
      "learning_rate": 7.445542420642097e-06,
      "loss": 0.002,
      "step": 780
    },
    {
      "epoch": 15.8,
      "grad_norm": 0.0461721308529377,
      "learning_rate": 7.176585431571235e-06,
      "loss": 0.0017,
      "step": 790
    },
    {
      "epoch": 16.0,
      "grad_norm": 0.023166760802268982,
      "learning_rate": 6.909830056250527e-06,
      "loss": 0.0008,
      "step": 800
    },
    {
      "epoch": 16.2,
      "grad_norm": 0.032082173973321915,
      "learning_rate": 6.645484302497452e-06,
      "loss": 0.0003,
      "step": 810
    },
    {
      "epoch": 16.4,
      "grad_norm": 0.08265218138694763,
      "learning_rate": 6.383754299179079e-06,
      "loss": 0.0007,
      "step": 820
    },
    {
      "epoch": 16.6,
      "grad_norm": 0.024636510759592056,
      "learning_rate": 6.124844135478971e-06,
      "loss": 0.005,
      "step": 830
    },
    {
      "epoch": 16.8,
      "grad_norm": 0.019458631053566933,
      "learning_rate": 5.868955701754584e-06,
      "loss": 0.0004,
      "step": 840
    },
    {
      "epoch": 17.0,
      "grad_norm": 0.01352341752499342,
      "learning_rate": 5.616288532109225e-06,
      "loss": 0.0007,
      "step": 850
    },
    {
      "epoch": 17.2,
      "grad_norm": 0.009102717973291874,
      "learning_rate": 5.367039648801386e-06,
      "loss": 0.0004,
      "step": 860
    },
    {
      "epoch": 17.4,
      "grad_norm": 0.01286560669541359,
      "learning_rate": 5.121403408612672e-06,
      "loss": 0.0002,
      "step": 870
    },
    {
      "epoch": 17.6,
      "grad_norm": 0.36686578392982483,
      "learning_rate": 4.879571351294287e-06,
      "loss": 0.0007,
      "step": 880
    },
    {
      "epoch": 17.8,
      "grad_norm": 0.005845635663717985,
      "learning_rate": 4.641732050210032e-06,
      "loss": 0.0002,
      "step": 890
    },
    {
      "epoch": 18.0,
      "grad_norm": 0.06807386130094528,
      "learning_rate": 4.408070965292534e-06,
      "loss": 0.0002,
      "step": 900
    },
    {
      "epoch": 18.2,
      "grad_norm": 0.02204430289566517,
      "learning_rate": 4.178770298427107e-06,
      "loss": 0.0001,
      "step": 910
    },
    {
      "epoch": 18.4,
      "grad_norm": 0.008754129521548748,
      "learning_rate": 3.954008851376252e-06,
      "loss": 0.0001,
      "step": 920
    },
    {
      "epoch": 18.6,
      "grad_norm": 0.00636293226853013,
      "learning_rate": 3.7339618863553983e-06,
      "loss": 0.0001,
      "step": 930
    },
    {
      "epoch": 18.8,
      "grad_norm": 0.004866347182542086,
      "learning_rate": 3.5188009893686916e-06,
      "loss": 0.0006,
      "step": 940
    },
    {
      "epoch": 19.0,
      "grad_norm": 0.004850468598306179,
      "learning_rate": 3.308693936411421e-06,
      "loss": 0.0001,
      "step": 950
    },
    {
      "epoch": 19.2,
      "grad_norm": 0.0016120151849463582,
      "learning_rate": 3.103804562643302e-06,
      "loss": 0.0001,
      "step": 960
    },
    {
      "epoch": 19.4,
      "grad_norm": 0.08115795254707336,
      "learning_rate": 2.9042926346347932e-06,
      "loss": 0.0002,
      "step": 970
    },
    {
      "epoch": 19.6,
      "grad_norm": 0.0031273579224944115,
      "learning_rate": 2.7103137257858867e-06,
      "loss": 0.0001,
      "step": 980
    },
    {
      "epoch": 19.8,
      "grad_norm": 0.005434623919427395,
      "learning_rate": 2.522019095014683e-06,
      "loss": 0.0001,
      "step": 990
    },
    {
      "epoch": 20.0,
      "grad_norm": 0.003434584243223071,
      "learning_rate": 2.339555568810221e-06,
      "loss": 0.0003,
      "step": 1000
    },
    {
      "epoch": 20.2,
      "grad_norm": 0.0036388696171343327,
      "learning_rate": 2.163065426741603e-06,
      "loss": 0.0003,
      "step": 1010
    },
    {
      "epoch": 20.4,
      "grad_norm": 0.005258283577859402,
      "learning_rate": 1.9926862905126663e-06,
      "loss": 0.0001,
      "step": 1020
    },
    {
      "epoch": 20.6,
      "grad_norm": 0.0052086408250033855,
      "learning_rate": 1.8285510166487154e-06,
      "loss": 0.0001,
      "step": 1030
    },
    {
      "epoch": 20.8,
      "grad_norm": 0.007011784706264734,
      "learning_rate": 1.6707875928990059e-06,
      "loss": 0.0001,
      "step": 1040
    },
    {
      "epoch": 21.0,
      "grad_norm": 0.008353537879884243,
      "learning_rate": 1.5195190384357405e-06,
      "loss": 0.0001,
      "step": 1050
    },
    {
      "epoch": 21.2,
      "grad_norm": 0.005364676937460899,
      "learning_rate": 1.3748633079274254e-06,
      "loss": 0.0001,
      "step": 1060
    },
    {
      "epoch": 21.4,
      "grad_norm": 0.07464922219514847,
      "learning_rate": 1.2369331995613664e-06,
      "loss": 0.0003,
      "step": 1070
    },
    {
      "epoch": 21.6,
      "grad_norm": 0.005178565625101328,
      "learning_rate": 1.1058362670870248e-06,
      "loss": 0.0001,
      "step": 1080
    },
    {
      "epoch": 21.8,
      "grad_norm": 0.003679296700283885,
      "learning_rate": 9.816747359488632e-07,
      "loss": 0.0001,
      "step": 1090
    },
    {
      "epoch": 22.0,
      "grad_norm": 0.008506166748702526,
      "learning_rate": 8.645454235739903e-07,
      "loss": 0.0001,
      "step": 1100
    },
    {
      "epoch": 22.2,
      "grad_norm": 0.0045352838933467865,
      "learning_rate": 7.545396638768698e-07,
      "loss": 0.0001,
      "step": 1110
    },
    {
      "epoch": 22.4,
      "grad_norm": 0.0041460818611085415,
      "learning_rate": 6.517432360398556e-07,
      "loss": 0.0002,
      "step": 1120
    },
    {
      "epoch": 22.6,
      "grad_norm": 0.00789156835526228,
      "learning_rate": 5.562362976251901e-07,
      "loss": 0.0001,
      "step": 1130
    },
    {
      "epoch": 22.8,
      "grad_norm": 0.004850171506404877,
      "learning_rate": 4.6809332207053083e-07,
      "loss": 0.0002,
      "step": 1140
    },
    {
      "epoch": 23.0,
      "grad_norm": 0.0024761913809925318,
      "learning_rate": 3.8738304061681107e-07,
      "loss": 0.0001,
      "step": 1150
    },
    {
      "epoch": 23.2,
      "grad_norm": 0.004848626907914877,
      "learning_rate": 3.1416838871368925e-07,
      "loss": 0.0002,
      "step": 1160
    },
    {
      "epoch": 23.4,
      "grad_norm": 0.008169041946530342,
      "learning_rate": 2.4850645694436736e-07,
      "loss": 0.0001,
      "step": 1170
    },
    {
      "epoch": 23.6,
      "grad_norm": 0.008230429142713547,
      "learning_rate": 1.9044844650808468e-07,
      "loss": 0.0001,
      "step": 1180
    },
    {
      "epoch": 23.8,
      "grad_norm": 0.008880352601408958,
      "learning_rate": 1.400396292949513e-07,
      "loss": 0.0001,
      "step": 1190
    },
    {
      "epoch": 24.0,
      "grad_norm": 0.003499081125482917,
      "learning_rate": 9.731931258429638e-08,
      "loss": 0.0002,
      "step": 1200
    },
    {
      "epoch": 24.2,
      "grad_norm": 0.002381326397880912,
      "learning_rate": 6.232080839403631e-08,
      "loss": 0.0001,
      "step": 1210
    },
    {
      "epoch": 24.4,
      "grad_norm": 0.003195591736584902,
      "learning_rate": 3.50714075049563e-08,
      "loss": 0.0001,
      "step": 1220
    },
    {
      "epoch": 24.6,
      "grad_norm": 0.003270337823778391,
      "learning_rate": 1.5592358180189782e-08,
      "loss": 0.0001,
      "step": 1230
    },
    {
      "epoch": 24.8,
      "grad_norm": 0.00868391152471304,
      "learning_rate": 3.898849596456477e-09,
      "loss": 0.0001,
      "step": 1240
    },
    {
      "epoch": 25.0,
      "grad_norm": 0.0035359435714781284,
      "learning_rate": 0.0,
      "loss": 0.0002,
      "step": 1250
    },
    {
      "epoch": 25.0,
      "step": 1250,
      "total_flos": 2.502555032784732e+17,
      "train_loss": 0.12197050291108899,
      "train_runtime": 13777.8614,
      "train_samples_per_second": 0.907,
      "train_steps_per_second": 0.091
    }
  ],
  "logging_steps": 10,
  "max_steps": 1250,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 25,
  "save_steps": 1000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 2.502555032784732e+17,
  "train_batch_size": 5,
  "trial_name": null,
  "trial_params": null
}