{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.4995091477019188,
  "eval_steps": 500,
  "global_step": 5597,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.000892458723784025,
      "grad_norm": 3.045154571533203,
      "learning_rate": 1.1061797752808989e-06,
      "loss": 1.3662,
      "step": 10
    },
    {
      "epoch": 0.00178491744756805,
      "grad_norm": 2.2175650596618652,
      "learning_rate": 2.2241573033707866e-06,
      "loss": 1.4403,
      "step": 20
    },
    {
      "epoch": 0.002677376171352075,
      "grad_norm": 1.860337734222412,
      "learning_rate": 3.342134831460675e-06,
      "loss": 1.1652,
      "step": 30
    },
    {
      "epoch": 0.0035698348951361,
      "grad_norm": 0.9244102239608765,
      "learning_rate": 4.4601123595505624e-06,
      "loss": 0.9807,
      "step": 40
    },
    {
      "epoch": 0.004462293618920125,
      "grad_norm": 0.6792669296264648,
      "learning_rate": 5.5780898876404495e-06,
      "loss": 0.8353,
      "step": 50
    },
    {
      "epoch": 0.00535475234270415,
      "grad_norm": 0.47331929206848145,
      "learning_rate": 6.6960674157303374e-06,
      "loss": 0.8033,
      "step": 60
    },
    {
      "epoch": 0.006247211066488175,
      "grad_norm": 0.44425472617149353,
      "learning_rate": 7.814044943820226e-06,
      "loss": 0.7722,
      "step": 70
    },
    {
      "epoch": 0.0071396697902722,
      "grad_norm": 0.48765331506729126,
      "learning_rate": 8.932022471910112e-06,
      "loss": 0.7988,
      "step": 80
    },
    {
      "epoch": 0.008032128514056224,
      "grad_norm": 0.5558447241783142,
      "learning_rate": 1.005e-05,
      "loss": 0.7813,
      "step": 90
    },
    {
      "epoch": 0.00892458723784025,
      "grad_norm": 0.4752425253391266,
      "learning_rate": 1.1167977528089888e-05,
      "loss": 0.7892,
      "step": 100
    },
    {
      "epoch": 0.009817045961624275,
      "grad_norm": 0.4644387364387512,
      "learning_rate": 1.2285955056179776e-05,
      "loss": 0.7592,
      "step": 110
    },
    {
      "epoch": 0.0107095046854083,
      "grad_norm": 0.5231781601905823,
      "learning_rate": 1.3403932584269664e-05,
      "loss": 0.7798,
      "step": 120
    },
    {
      "epoch": 0.011601963409192326,
      "grad_norm": 0.5505983233451843,
      "learning_rate": 1.4521910112359552e-05,
      "loss": 0.778,
      "step": 130
    },
    {
      "epoch": 0.01249442213297635,
      "grad_norm": 0.805553138256073,
      "learning_rate": 1.563988764044944e-05,
      "loss": 0.7573,
      "step": 140
    },
    {
      "epoch": 0.013386880856760375,
      "grad_norm": 0.4206564724445343,
      "learning_rate": 1.6757865168539328e-05,
      "loss": 0.7692,
      "step": 150
    },
    {
      "epoch": 0.0142793395805444,
      "grad_norm": 0.49315696954727173,
      "learning_rate": 1.7875842696629214e-05,
      "loss": 0.7773,
      "step": 160
    },
    {
      "epoch": 0.015171798304328426,
      "grad_norm": 0.5224491953849792,
      "learning_rate": 1.89938202247191e-05,
      "loss": 0.7728,
      "step": 170
    },
    {
      "epoch": 0.01606425702811245,
      "grad_norm": 0.43197882175445557,
      "learning_rate": 2e-05,
      "loss": 0.7919,
      "step": 180
    },
    {
      "epoch": 0.016956715751896476,
      "grad_norm": 0.49200326204299927,
      "learning_rate": 2e-05,
      "loss": 0.7639,
      "step": 190
    },
    {
      "epoch": 0.0178491744756805,
      "grad_norm": 0.44408541917800903,
      "learning_rate": 2e-05,
      "loss": 0.7361,
      "step": 200
    },
    {
      "epoch": 0.018741633199464525,
      "grad_norm": 0.5795125961303711,
      "learning_rate": 2e-05,
      "loss": 0.8062,
      "step": 210
    },
    {
      "epoch": 0.01963409192324855,
      "grad_norm": 0.4349335730075836,
      "learning_rate": 2e-05,
      "loss": 0.7885,
      "step": 220
    },
    {
      "epoch": 0.020526550647032574,
      "grad_norm": 0.4409690201282501,
      "learning_rate": 2e-05,
      "loss": 0.7942,
      "step": 230
    },
    {
      "epoch": 0.0214190093708166,
      "grad_norm": 0.43803665041923523,
      "learning_rate": 2e-05,
      "loss": 0.7597,
      "step": 240
    },
    {
      "epoch": 0.022311468094600623,
      "grad_norm": 0.43824803829193115,
      "learning_rate": 2e-05,
      "loss": 0.7669,
      "step": 250
    },
    {
      "epoch": 0.02320392681838465,
      "grad_norm": 0.43911799788475037,
      "learning_rate": 2e-05,
      "loss": 0.7453,
      "step": 260
    },
    {
      "epoch": 0.024096385542168676,
      "grad_norm": 0.39943230152130127,
      "learning_rate": 2e-05,
      "loss": 0.7293,
      "step": 270
    },
    {
      "epoch": 0.0249888442659527,
      "grad_norm": 0.44129323959350586,
      "learning_rate": 2e-05,
      "loss": 0.8173,
      "step": 280
    },
    {
      "epoch": 0.025881302989736725,
      "grad_norm": 0.40437787771224976,
      "learning_rate": 2e-05,
      "loss": 0.7626,
      "step": 290
    },
    {
      "epoch": 0.02677376171352075,
      "grad_norm": 0.46044057607650757,
      "learning_rate": 2e-05,
      "loss": 0.7705,
      "step": 300
    },
    {
      "epoch": 0.027666220437304774,
      "grad_norm": 0.5197446346282959,
      "learning_rate": 2e-05,
      "loss": 0.7439,
      "step": 310
    },
    {
      "epoch": 0.0285586791610888,
      "grad_norm": 0.4308314323425293,
      "learning_rate": 2e-05,
      "loss": 0.8261,
      "step": 320
    },
    {
      "epoch": 0.029451137884872823,
      "grad_norm": 0.43585020303726196,
      "learning_rate": 2e-05,
      "loss": 0.7583,
      "step": 330
    },
    {
      "epoch": 0.03034359660865685,
      "grad_norm": 0.43804678320884705,
      "learning_rate": 2e-05,
      "loss": 0.7712,
      "step": 340
    },
    {
      "epoch": 0.031236055332440876,
      "grad_norm": 0.4321230351924896,
      "learning_rate": 2e-05,
      "loss": 0.7558,
      "step": 350
    },
    {
      "epoch": 0.0321285140562249,
      "grad_norm": 0.426249623298645,
      "learning_rate": 2e-05,
      "loss": 0.7652,
      "step": 360
    },
    {
      "epoch": 0.03302097278000892,
      "grad_norm": 0.4081008732318878,
      "learning_rate": 2e-05,
      "loss": 0.7536,
      "step": 370
    },
    {
      "epoch": 0.03391343150379295,
      "grad_norm": 0.6208111047744751,
      "learning_rate": 2e-05,
      "loss": 0.7511,
      "step": 380
    },
    {
      "epoch": 0.03480589022757698,
      "grad_norm": 0.45165306329727173,
      "learning_rate": 2e-05,
      "loss": 0.7612,
      "step": 390
    },
    {
      "epoch": 0.035698348951361,
      "grad_norm": 0.8442327380180359,
      "learning_rate": 2e-05,
      "loss": 0.7757,
      "step": 400
    },
    {
      "epoch": 0.036590807675145026,
      "grad_norm": 0.43777230381965637,
      "learning_rate": 2e-05,
      "loss": 0.7554,
      "step": 410
    },
    {
      "epoch": 0.03748326639892905,
      "grad_norm": 0.4242396950721741,
      "learning_rate": 2e-05,
      "loss": 0.7437,
      "step": 420
    },
    {
      "epoch": 0.038375725122713075,
      "grad_norm": 0.44561508297920227,
      "learning_rate": 2e-05,
      "loss": 0.7579,
      "step": 430
    },
    {
      "epoch": 0.0392681838464971,
      "grad_norm": 0.4057537615299225,
      "learning_rate": 2e-05,
      "loss": 0.7672,
      "step": 440
    },
    {
      "epoch": 0.040160642570281124,
      "grad_norm": 0.4248247444629669,
      "learning_rate": 2e-05,
      "loss": 0.7517,
      "step": 450
    },
    {
      "epoch": 0.04105310129406515,
      "grad_norm": 0.38674187660217285,
      "learning_rate": 2e-05,
      "loss": 0.7649,
      "step": 460
    },
    {
      "epoch": 0.04194556001784917,
      "grad_norm": 0.4590655267238617,
      "learning_rate": 2e-05,
      "loss": 0.7138,
      "step": 470
    },
    {
      "epoch": 0.0428380187416332,
      "grad_norm": 0.38914060592651367,
      "learning_rate": 2e-05,
      "loss": 0.7533,
      "step": 480
    },
    {
      "epoch": 0.04373047746541722,
      "grad_norm": 0.4345904588699341,
      "learning_rate": 2e-05,
      "loss": 0.7323,
      "step": 490
    },
    {
      "epoch": 0.04462293618920125,
      "grad_norm": 0.372398316860199,
      "learning_rate": 2e-05,
      "loss": 0.722,
      "step": 500
    },
    {
      "epoch": 0.04551539491298527,
      "grad_norm": 0.39427366852760315,
      "learning_rate": 2e-05,
      "loss": 0.7649,
      "step": 510
    },
    {
      "epoch": 0.0464078536367693,
      "grad_norm": 0.45776817202568054,
      "learning_rate": 2e-05,
      "loss": 0.761,
      "step": 520
    },
    {
      "epoch": 0.04730031236055333,
      "grad_norm": 0.4311637282371521,
      "learning_rate": 2e-05,
      "loss": 0.7761,
      "step": 530
    },
    {
      "epoch": 0.04819277108433735,
      "grad_norm": 0.3782254457473755,
      "learning_rate": 2e-05,
      "loss": 0.734,
      "step": 540
    },
    {
      "epoch": 0.049085229808121376,
      "grad_norm": 0.4318030774593353,
      "learning_rate": 2e-05,
      "loss": 0.7658,
      "step": 550
    },
    {
      "epoch": 0.0499776885319054,
      "grad_norm": 0.4350939691066742,
      "learning_rate": 2e-05,
      "loss": 0.7765,
      "step": 560
    },
    {
      "epoch": 0.050870147255689425,
      "grad_norm": 0.4095909595489502,
      "learning_rate": 2e-05,
      "loss": 0.76,
      "step": 570
    },
    {
      "epoch": 0.05176260597947345,
      "grad_norm": 0.3846561908721924,
      "learning_rate": 2e-05,
      "loss": 0.7516,
      "step": 580
    },
    {
      "epoch": 0.052655064703257475,
      "grad_norm": 0.4147850573062897,
      "learning_rate": 2e-05,
      "loss": 0.7674,
      "step": 590
    },
    {
      "epoch": 0.0535475234270415,
      "grad_norm": 0.41225138306617737,
      "learning_rate": 2e-05,
      "loss": 0.7489,
      "step": 600
    },
    {
      "epoch": 0.054439982150825524,
      "grad_norm": 0.4051811099052429,
      "learning_rate": 2e-05,
      "loss": 0.7396,
      "step": 610
    },
    {
      "epoch": 0.05533244087460955,
      "grad_norm": 0.3863374590873718,
      "learning_rate": 2e-05,
      "loss": 0.7477,
      "step": 620
    },
    {
      "epoch": 0.05622489959839357,
      "grad_norm": 0.3682566285133362,
      "learning_rate": 2e-05,
      "loss": 0.7675,
      "step": 630
    },
    {
      "epoch": 0.0571173583221776,
      "grad_norm": 0.38680627942085266,
      "learning_rate": 2e-05,
      "loss": 0.7149,
      "step": 640
    },
    {
      "epoch": 0.05800981704596162,
      "grad_norm": 0.4512251317501068,
      "learning_rate": 2e-05,
      "loss": 0.7598,
      "step": 650
    },
    {
      "epoch": 0.058902275769745646,
      "grad_norm": 0.45277082920074463,
      "learning_rate": 2e-05,
      "loss": 0.7227,
      "step": 660
    },
    {
      "epoch": 0.05979473449352968,
      "grad_norm": 0.4341653287410736,
      "learning_rate": 2e-05,
      "loss": 0.7798,
      "step": 670
    },
    {
      "epoch": 0.0606871932173137,
      "grad_norm": 0.4053642749786377,
      "learning_rate": 2e-05,
      "loss": 0.7448,
      "step": 680
    },
    {
      "epoch": 0.06157965194109773,
      "grad_norm": 0.4231747090816498,
      "learning_rate": 2e-05,
      "loss": 0.7582,
      "step": 690
    },
    {
      "epoch": 0.06247211066488175,
      "grad_norm": 0.3956149220466614,
      "learning_rate": 2e-05,
      "loss": 0.7584,
      "step": 700
    },
    {
      "epoch": 0.06336456938866578,
      "grad_norm": 0.4424019157886505,
      "learning_rate": 2e-05,
      "loss": 0.731,
      "step": 710
    },
    {
      "epoch": 0.0642570281124498,
      "grad_norm": 0.44977128505706787,
      "learning_rate": 2e-05,
      "loss": 0.7346,
      "step": 720
    },
    {
      "epoch": 0.06514948683623382,
      "grad_norm": 0.4080374240875244,
      "learning_rate": 2e-05,
      "loss": 0.7478,
      "step": 730
    },
    {
      "epoch": 0.06604194556001784,
      "grad_norm": 0.40351366996765137,
      "learning_rate": 2e-05,
      "loss": 0.7308,
      "step": 740
    },
    {
      "epoch": 0.06693440428380187,
      "grad_norm": 0.3880046606063843,
      "learning_rate": 2e-05,
      "loss": 0.7466,
      "step": 750
    },
    {
      "epoch": 0.0678268630075859,
      "grad_norm": 0.3946651518344879,
      "learning_rate": 2e-05,
      "loss": 0.7579,
      "step": 760
    },
    {
      "epoch": 0.06871932173136992,
      "grad_norm": 0.3880173861980438,
      "learning_rate": 2e-05,
      "loss": 0.7346,
      "step": 770
    },
    {
      "epoch": 0.06961178045515395,
      "grad_norm": 0.38916459679603577,
      "learning_rate": 2e-05,
      "loss": 0.7219,
      "step": 780
    },
    {
      "epoch": 0.07050423917893797,
      "grad_norm": 0.3937828838825226,
      "learning_rate": 2e-05,
      "loss": 0.76,
      "step": 790
    },
    {
      "epoch": 0.071396697902722,
      "grad_norm": 0.43031296133995056,
      "learning_rate": 2e-05,
      "loss": 0.758,
      "step": 800
    },
    {
      "epoch": 0.07228915662650602,
      "grad_norm": 0.4201333224773407,
      "learning_rate": 2e-05,
      "loss": 0.7449,
      "step": 810
    },
    {
      "epoch": 0.07318161535029005,
      "grad_norm": 0.3999972343444824,
      "learning_rate": 2e-05,
      "loss": 0.7487,
      "step": 820
    },
    {
      "epoch": 0.07407407407407407,
      "grad_norm": 0.39585524797439575,
      "learning_rate": 2e-05,
      "loss": 0.7076,
      "step": 830
    },
    {
      "epoch": 0.0749665327978581,
      "grad_norm": 0.3961842656135559,
      "learning_rate": 2e-05,
      "loss": 0.7031,
      "step": 840
    },
    {
      "epoch": 0.07585899152164212,
      "grad_norm": 0.3905622363090515,
      "learning_rate": 2e-05,
      "loss": 0.7484,
      "step": 850
    },
    {
      "epoch": 0.07675145024542615,
      "grad_norm": 0.40534672141075134,
      "learning_rate": 2e-05,
      "loss": 0.739,
      "step": 860
    },
    {
      "epoch": 0.07764390896921017,
      "grad_norm": 0.38796329498291016,
      "learning_rate": 2e-05,
      "loss": 0.7236,
      "step": 870
    },
    {
      "epoch": 0.0785363676929942,
      "grad_norm": 0.5421860218048096,
      "learning_rate": 2e-05,
      "loss": 0.7263,
      "step": 880
    },
    {
      "epoch": 0.07942882641677822,
      "grad_norm": 0.4691896140575409,
      "learning_rate": 2e-05,
      "loss": 0.7289,
      "step": 890
    },
    {
      "epoch": 0.08032128514056225,
      "grad_norm": 0.4255770742893219,
      "learning_rate": 2e-05,
      "loss": 0.7399,
      "step": 900
    },
    {
      "epoch": 0.08121374386434628,
      "grad_norm": 0.3914809823036194,
      "learning_rate": 2e-05,
      "loss": 0.7908,
      "step": 910
    },
    {
      "epoch": 0.0821062025881303,
      "grad_norm": 0.3992801010608673,
      "learning_rate": 2e-05,
      "loss": 0.7144,
      "step": 920
    },
    {
      "epoch": 0.08299866131191433,
      "grad_norm": 0.3806823194026947,
      "learning_rate": 2e-05,
      "loss": 0.7471,
      "step": 930
    },
    {
      "epoch": 0.08389112003569835,
      "grad_norm": 0.546196460723877,
      "learning_rate": 2e-05,
      "loss": 0.6958,
      "step": 940
    },
    {
      "epoch": 0.08478357875948238,
      "grad_norm": 0.5522550344467163,
      "learning_rate": 2e-05,
      "loss": 0.7481,
      "step": 950
    },
    {
      "epoch": 0.0856760374832664,
      "grad_norm": 0.421906977891922,
      "learning_rate": 2e-05,
      "loss": 0.7771,
      "step": 960
    },
    {
      "epoch": 0.08656849620705043,
      "grad_norm": 0.381605327129364,
      "learning_rate": 2e-05,
      "loss": 0.7512,
      "step": 970
    },
    {
      "epoch": 0.08746095493083444,
      "grad_norm": 0.5547103881835938,
      "learning_rate": 2e-05,
      "loss": 0.7529,
      "step": 980
    },
    {
      "epoch": 0.08835341365461848,
      "grad_norm": 0.41171810030937195,
      "learning_rate": 2e-05,
      "loss": 0.7224,
      "step": 990
    },
    {
      "epoch": 0.0892458723784025,
      "grad_norm": 0.3596498370170593,
      "learning_rate": 2e-05,
      "loss": 0.7414,
      "step": 1000
    },
    {
      "epoch": 0.09013833110218653,
      "grad_norm": 0.46202516555786133,
      "learning_rate": 2e-05,
      "loss": 0.7322,
      "step": 1010
    },
    {
      "epoch": 0.09103078982597054,
      "grad_norm": 0.39855870604515076,
      "learning_rate": 2e-05,
      "loss": 0.7243,
      "step": 1020
    },
    {
      "epoch": 0.09192324854975457,
      "grad_norm": 0.38900578022003174,
      "learning_rate": 2e-05,
      "loss": 0.7178,
      "step": 1030
    },
    {
      "epoch": 0.0928157072735386,
      "grad_norm": 0.4451678395271301,
      "learning_rate": 2e-05,
      "loss": 0.7473,
      "step": 1040
    },
    {
      "epoch": 0.09370816599732262,
      "grad_norm": 0.3663797676563263,
      "learning_rate": 2e-05,
      "loss": 0.7148,
      "step": 1050
    },
    {
      "epoch": 0.09460062472110665,
      "grad_norm": 0.3737003803253174,
      "learning_rate": 2e-05,
      "loss": 0.7308,
      "step": 1060
    },
    {
      "epoch": 0.09549308344489067,
      "grad_norm": 0.5132238864898682,
      "learning_rate": 2e-05,
      "loss": 0.7483,
      "step": 1070
    },
    {
      "epoch": 0.0963855421686747,
      "grad_norm": 0.39757436513900757,
      "learning_rate": 2e-05,
      "loss": 0.7055,
      "step": 1080
    },
    {
      "epoch": 0.09727800089245872,
      "grad_norm": 0.3889594078063965,
      "learning_rate": 2e-05,
      "loss": 0.7647,
      "step": 1090
    },
    {
      "epoch": 0.09817045961624275,
      "grad_norm": 0.40366658568382263,
      "learning_rate": 2e-05,
      "loss": 0.7453,
      "step": 1100
    },
    {
      "epoch": 0.09906291834002677,
      "grad_norm": 0.37995877861976624,
      "learning_rate": 2e-05,
      "loss": 0.7417,
      "step": 1110
    },
    {
      "epoch": 0.0999553770638108,
      "grad_norm": 0.3837803304195404,
      "learning_rate": 2e-05,
      "loss": 0.728,
      "step": 1120
    },
    {
      "epoch": 0.10084783578759482,
      "grad_norm": 0.37044045329093933,
      "learning_rate": 2e-05,
      "loss": 0.7216,
      "step": 1130
    },
    {
      "epoch": 0.10174029451137885,
      "grad_norm": 0.3822389245033264,
      "learning_rate": 2e-05,
      "loss": 0.7075,
      "step": 1140
    },
    {
      "epoch": 0.10263275323516287,
      "grad_norm": 0.3520481288433075,
      "learning_rate": 2e-05,
      "loss": 0.7131,
      "step": 1150
    },
    {
      "epoch": 0.1035252119589469,
      "grad_norm": 0.38390985131263733,
      "learning_rate": 2e-05,
      "loss": 0.7559,
      "step": 1160
    },
    {
      "epoch": 0.10441767068273092,
      "grad_norm": 0.3780132830142975,
      "learning_rate": 2e-05,
      "loss": 0.7224,
      "step": 1170
    },
    {
      "epoch": 0.10531012940651495,
      "grad_norm": 0.4197860062122345,
      "learning_rate": 2e-05,
      "loss": 0.7275,
      "step": 1180
    },
    {
      "epoch": 0.10620258813029898,
      "grad_norm": 0.3551185131072998,
      "learning_rate": 2e-05,
      "loss": 0.7118,
      "step": 1190
    },
    {
      "epoch": 0.107095046854083,
      "grad_norm": 0.36302730441093445,
      "learning_rate": 2e-05,
      "loss": 0.7032,
      "step": 1200
    },
    {
      "epoch": 0.10798750557786703,
      "grad_norm": 0.4428386688232422,
      "learning_rate": 2e-05,
      "loss": 0.7687,
      "step": 1210
    },
    {
      "epoch": 0.10887996430165105,
      "grad_norm": 0.3998299539089203,
      "learning_rate": 2e-05,
      "loss": 0.7204,
      "step": 1220
    },
    {
      "epoch": 0.10977242302543508,
      "grad_norm": 0.3677099347114563,
      "learning_rate": 2e-05,
      "loss": 0.7057,
      "step": 1230
    },
    {
      "epoch": 0.1106648817492191,
      "grad_norm": 0.4453865885734558,
      "learning_rate": 2e-05,
      "loss": 0.7309,
      "step": 1240
    },
    {
      "epoch": 0.11155734047300313,
      "grad_norm": 0.35671019554138184,
      "learning_rate": 2e-05,
      "loss": 0.6845,
      "step": 1250
    },
    {
      "epoch": 0.11244979919678715,
      "grad_norm": 0.38202935457229614,
      "learning_rate": 2e-05,
      "loss": 0.72,
      "step": 1260
    },
    {
      "epoch": 0.11334225792057118,
      "grad_norm": 0.38386300206184387,
      "learning_rate": 2e-05,
      "loss": 0.7228,
      "step": 1270
    },
    {
      "epoch": 0.1142347166443552,
      "grad_norm": 0.41439297795295715,
      "learning_rate": 2e-05,
      "loss": 0.7334,
      "step": 1280
    },
    {
      "epoch": 0.11512717536813923,
      "grad_norm": 0.39205095171928406,
      "learning_rate": 2e-05,
      "loss": 0.7325,
      "step": 1290
    },
    {
      "epoch": 0.11601963409192324,
      "grad_norm": 0.4260752201080322,
      "learning_rate": 2e-05,
      "loss": 0.7129,
      "step": 1300
    },
    {
      "epoch": 0.11691209281570727,
      "grad_norm": 0.35168108344078064,
      "learning_rate": 2e-05,
      "loss": 0.7003,
      "step": 1310
    },
    {
      "epoch": 0.11780455153949129,
      "grad_norm": 0.354591965675354,
      "learning_rate": 2e-05,
      "loss": 0.7081,
      "step": 1320
    },
    {
      "epoch": 0.11869701026327532,
      "grad_norm": 0.3994496762752533,
      "learning_rate": 2e-05,
      "loss": 0.7348,
      "step": 1330
    },
    {
      "epoch": 0.11958946898705936,
      "grad_norm": 0.3687463402748108,
      "learning_rate": 2e-05,
      "loss": 0.7138,
      "step": 1340
    },
    {
      "epoch": 0.12048192771084337,
      "grad_norm": 0.3527894914150238,
      "learning_rate": 2e-05,
      "loss": 0.7331,
      "step": 1350
    },
    {
      "epoch": 0.1213743864346274,
      "grad_norm": 0.39547526836395264,
      "learning_rate": 2e-05,
      "loss": 0.6874,
      "step": 1360
    },
    {
      "epoch": 0.12226684515841142,
      "grad_norm": 0.35508838295936584,
      "learning_rate": 2e-05,
      "loss": 0.7588,
      "step": 1370
    },
    {
      "epoch": 0.12315930388219545,
      "grad_norm": 0.3983374238014221,
      "learning_rate": 2e-05,
      "loss": 0.7383,
      "step": 1380
    },
    {
      "epoch": 0.12405176260597947,
      "grad_norm": 0.35641518235206604,
      "learning_rate": 2e-05,
      "loss": 0.7357,
      "step": 1390
    },
    {
      "epoch": 0.1249442213297635,
      "grad_norm": 0.41379478573799133,
      "learning_rate": 2e-05,
      "loss": 0.7257,
      "step": 1400
    },
    {
      "epoch": 0.12583668005354753,
      "grad_norm": 0.39988917112350464,
      "learning_rate": 2e-05,
      "loss": 0.7272,
      "step": 1410
    },
    {
      "epoch": 0.12672913877733155,
      "grad_norm": 0.3463815152645111,
      "learning_rate": 2e-05,
      "loss": 0.7595,
      "step": 1420
    },
    {
      "epoch": 0.12762159750111557,
      "grad_norm": 0.36326804757118225,
      "learning_rate": 2e-05,
      "loss": 0.6967,
      "step": 1430
    },
    {
      "epoch": 0.1285140562248996,
      "grad_norm": 0.3950536251068115,
      "learning_rate": 2e-05,
      "loss": 0.686,
      "step": 1440
    },
    {
      "epoch": 0.12940651494868363,
      "grad_norm": 0.43041136860847473,
      "learning_rate": 2e-05,
      "loss": 0.7122,
      "step": 1450
    },
    {
      "epoch": 0.13029897367246765,
      "grad_norm": 0.40778934955596924,
      "learning_rate": 2e-05,
      "loss": 0.7107,
      "step": 1460
    },
    {
      "epoch": 0.13119143239625167,
      "grad_norm": 0.37620630860328674,
      "learning_rate": 2e-05,
      "loss": 0.7352,
      "step": 1470
    },
    {
      "epoch": 0.13208389112003568,
      "grad_norm": 0.34521961212158203,
      "learning_rate": 2e-05,
      "loss": 0.725,
      "step": 1480
    },
    {
      "epoch": 0.13297634984381973,
      "grad_norm": 0.4491102397441864,
      "learning_rate": 2e-05,
      "loss": 0.7313,
      "step": 1490
    },
    {
      "epoch": 0.13386880856760375,
      "grad_norm": 0.3915207087993622,
      "learning_rate": 2e-05,
      "loss": 0.7185,
      "step": 1500
    },
    {
      "epoch": 0.13476126729138777,
      "grad_norm": 0.40829235315322876,
      "learning_rate": 2e-05,
      "loss": 0.7266,
      "step": 1510
    },
    {
      "epoch": 0.1356537260151718,
      "grad_norm": 0.4085201323032379,
      "learning_rate": 2e-05,
      "loss": 0.7142,
      "step": 1520
    },
    {
      "epoch": 0.13654618473895583,
      "grad_norm": 0.3529062569141388,
      "learning_rate": 2e-05,
      "loss": 0.7092,
      "step": 1530
    },
    {
      "epoch": 0.13743864346273985,
      "grad_norm": 0.360169917345047,
      "learning_rate": 2e-05,
      "loss": 0.7402,
      "step": 1540
    },
    {
      "epoch": 0.13833110218652386,
      "grad_norm": 0.4295094907283783,
      "learning_rate": 2e-05,
      "loss": 0.7528,
      "step": 1550
    },
    {
      "epoch": 0.1392235609103079,
      "grad_norm": 0.38765519857406616,
      "learning_rate": 2e-05,
      "loss": 0.7173,
      "step": 1560
    },
    {
      "epoch": 0.14011601963409193,
      "grad_norm": 0.4022582769393921,
      "learning_rate": 2e-05,
      "loss": 0.6994,
      "step": 1570
    },
    {
      "epoch": 0.14100847835787594,
      "grad_norm": 0.39251431822776794,
      "learning_rate": 2e-05,
      "loss": 0.7394,
      "step": 1580
    },
    {
      "epoch": 0.14190093708165996,
      "grad_norm": 0.4118332862854004,
      "learning_rate": 2e-05,
      "loss": 0.7361,
      "step": 1590
    },
    {
      "epoch": 0.142793395805444,
      "grad_norm": 0.39799195528030396,
      "learning_rate": 2e-05,
      "loss": 0.6995,
      "step": 1600
    },
    {
      "epoch": 0.14368585452922802,
      "grad_norm": 0.35336437821388245,
      "learning_rate": 2e-05,
      "loss": 0.755,
      "step": 1610
    },
    {
      "epoch": 0.14457831325301204,
      "grad_norm": 0.39616140723228455,
      "learning_rate": 2e-05,
      "loss": 0.7043,
      "step": 1620
    },
    {
      "epoch": 0.14547077197679606,
      "grad_norm": 0.3775366246700287,
      "learning_rate": 2e-05,
      "loss": 0.723,
      "step": 1630
    },
    {
      "epoch": 0.1463632307005801,
      "grad_norm": 0.38130348920822144,
      "learning_rate": 2e-05,
      "loss": 0.7029,
      "step": 1640
    },
    {
      "epoch": 0.14725568942436412,
      "grad_norm": 0.7423177361488342,
      "learning_rate": 2e-05,
      "loss": 0.7276,
      "step": 1650
    },
    {
      "epoch": 0.14814814814814814,
      "grad_norm": 0.3703177273273468,
      "learning_rate": 2e-05,
      "loss": 0.7094,
      "step": 1660
    },
    {
      "epoch": 0.14904060687193219,
      "grad_norm": 0.36464789509773254,
      "learning_rate": 2e-05,
      "loss": 0.6968,
      "step": 1670
    },
    {
      "epoch": 0.1499330655957162,
      "grad_norm": 0.3765050172805786,
      "learning_rate": 2e-05,
      "loss": 0.73,
      "step": 1680
    },
    {
      "epoch": 0.15082552431950022,
      "grad_norm": 0.36146077513694763,
      "learning_rate": 2e-05,
      "loss": 0.6911,
      "step": 1690
    },
    {
      "epoch": 0.15171798304328424,
      "grad_norm": 0.3474405109882355,
      "learning_rate": 2e-05,
      "loss": 0.7045,
      "step": 1700
    },
    {
      "epoch": 0.15261044176706828,
      "grad_norm": 0.396933913230896,
      "learning_rate": 2e-05,
      "loss": 0.7363,
      "step": 1710
    },
    {
      "epoch": 0.1535029004908523,
      "grad_norm": 0.37580183148384094,
      "learning_rate": 2e-05,
      "loss": 0.7219,
      "step": 1720
    },
    {
      "epoch": 0.15439535921463632,
      "grad_norm": 0.3710784614086151,
      "learning_rate": 2e-05,
      "loss": 0.7089,
      "step": 1730
    },
    {
      "epoch": 0.15528781793842034,
      "grad_norm": 0.36403852701187134,
      "learning_rate": 2e-05,
      "loss": 0.7111,
      "step": 1740
    },
    {
      "epoch": 0.15618027666220438,
      "grad_norm": 0.5121102333068848,
      "learning_rate": 2e-05,
      "loss": 0.7,
      "step": 1750
    },
    {
      "epoch": 0.1570727353859884,
      "grad_norm": 0.38486865162849426,
      "learning_rate": 2e-05,
      "loss": 0.7475,
      "step": 1760
    },
    {
      "epoch": 0.15796519410977242,
      "grad_norm": 0.35487449169158936,
      "learning_rate": 2e-05,
      "loss": 0.7121,
      "step": 1770
    },
    {
      "epoch": 0.15885765283355643,
      "grad_norm": 0.36509183049201965,
      "learning_rate": 2e-05,
      "loss": 0.7385,
      "step": 1780
    },
    {
      "epoch": 0.15975011155734048,
      "grad_norm": 0.43019792437553406,
      "learning_rate": 2e-05,
      "loss": 0.7413,
      "step": 1790
    },
    {
      "epoch": 0.1606425702811245,
      "grad_norm": 0.3725093901157379,
      "learning_rate": 2e-05,
      "loss": 0.7075,
      "step": 1800
    },
    {
      "epoch": 0.16153502900490851,
      "grad_norm": 0.4121997356414795,
      "learning_rate": 2e-05,
      "loss": 0.7195,
      "step": 1810
    },
    {
      "epoch": 0.16242748772869256,
      "grad_norm": 0.36837446689605713,
      "learning_rate": 2e-05,
      "loss": 0.7398,
      "step": 1820
    },
    {
      "epoch": 0.16331994645247658,
      "grad_norm": 0.3591003715991974,
      "learning_rate": 2e-05,
      "loss": 0.7011,
      "step": 1830
    },
    {
      "epoch": 0.1642124051762606,
      "grad_norm": 0.38240692019462585,
      "learning_rate": 2e-05,
      "loss": 0.7141,
      "step": 1840
    },
    {
      "epoch": 0.1651048639000446,
      "grad_norm": 0.3967023491859436,
      "learning_rate": 2e-05,
      "loss": 0.7348,
      "step": 1850
    },
    {
      "epoch": 0.16599732262382866,
      "grad_norm": 0.5831686854362488,
      "learning_rate": 2e-05,
      "loss": 0.7106,
      "step": 1860
    },
    {
      "epoch": 0.16688978134761268,
      "grad_norm": 0.3728710412979126,
      "learning_rate": 2e-05,
      "loss": 0.6938,
      "step": 1870
    },
    {
      "epoch": 0.1677822400713967,
      "grad_norm": 0.4233781695365906,
      "learning_rate": 2e-05,
      "loss": 0.7282,
      "step": 1880
    },
    {
      "epoch": 0.1686746987951807,
      "grad_norm": 0.38467922806739807,
      "learning_rate": 2e-05,
      "loss": 0.7001,
      "step": 1890
    },
    {
      "epoch": 0.16956715751896476,
      "grad_norm": 0.35904461145401,
      "learning_rate": 2e-05,
      "loss": 0.713,
      "step": 1900
    },
    {
      "epoch": 0.17045961624274877,
      "grad_norm": 0.3828829526901245,
      "learning_rate": 2e-05,
      "loss": 0.6828,
      "step": 1910
    },
    {
      "epoch": 0.1713520749665328,
      "grad_norm": 0.35955435037612915,
      "learning_rate": 2e-05,
      "loss": 0.7209,
      "step": 1920
    },
    {
      "epoch": 0.17224453369031684,
      "grad_norm": 0.3535810708999634,
      "learning_rate": 2e-05,
      "loss": 0.7072,
      "step": 1930
    },
    {
      "epoch": 0.17313699241410085,
      "grad_norm": 0.3620419502258301,
      "learning_rate": 2e-05,
      "loss": 0.7045,
      "step": 1940
    },
    {
      "epoch": 0.17402945113788487,
      "grad_norm": 0.37288907170295715,
      "learning_rate": 2e-05,
      "loss": 0.7701,
      "step": 1950
    },
    {
      "epoch": 0.1749219098616689,
      "grad_norm": 0.36177289485931396,
      "learning_rate": 2e-05,
      "loss": 0.695,
      "step": 1960
    },
    {
      "epoch": 0.17581436858545293,
      "grad_norm": 0.46705108880996704,
      "learning_rate": 2e-05,
      "loss": 0.7036,
      "step": 1970
    },
    {
      "epoch": 0.17670682730923695,
      "grad_norm": 0.38864728808403015,
      "learning_rate": 2e-05,
      "loss": 0.695,
      "step": 1980
    },
    {
      "epoch": 0.17759928603302097,
      "grad_norm": 0.3607434332370758,
      "learning_rate": 2e-05,
      "loss": 0.7111,
      "step": 1990
    },
    {
      "epoch": 0.178491744756805,
      "grad_norm": 0.36717599630355835,
      "learning_rate": 2e-05,
      "loss": 0.691,
      "step": 2000
    },
    {
      "epoch": 0.17938420348058903,
      "grad_norm": 0.36441174149513245,
      "learning_rate": 2e-05,
      "loss": 0.6795,
      "step": 2010
    },
    {
      "epoch": 0.18027666220437305,
      "grad_norm": 0.42372533679008484,
      "learning_rate": 2e-05,
      "loss": 0.7215,
      "step": 2020
    },
    {
      "epoch": 0.18116912092815707,
      "grad_norm": 0.41000911593437195,
      "learning_rate": 2e-05,
      "loss": 0.7257,
      "step": 2030
    },
    {
      "epoch": 0.18206157965194109,
      "grad_norm": 0.40014421939849854,
      "learning_rate": 2e-05,
      "loss": 0.702,
      "step": 2040
    },
    {
      "epoch": 0.18295403837572513,
      "grad_norm": 0.3575685918331146,
      "learning_rate": 2e-05,
      "loss": 0.7076,
      "step": 2050
    },
    {
      "epoch": 0.18384649709950915,
      "grad_norm": 0.3880053460597992,
      "learning_rate": 2e-05,
      "loss": 0.6628,
      "step": 2060
    },
    {
      "epoch": 0.18473895582329317,
      "grad_norm": 0.3879825472831726,
      "learning_rate": 2e-05,
      "loss": 0.711,
      "step": 2070
    },
    {
      "epoch": 0.1856314145470772,
      "grad_norm": 0.3911679983139038,
      "learning_rate": 2e-05,
      "loss": 0.7285,
      "step": 2080
    },
    {
      "epoch": 0.18652387327086123,
      "grad_norm": 0.35667601227760315,
      "learning_rate": 2e-05,
      "loss": 0.7385,
      "step": 2090
    },
    {
      "epoch": 0.18741633199464525,
      "grad_norm": 0.4055507779121399,
      "learning_rate": 2e-05,
      "loss": 0.7087,
      "step": 2100
    },
    {
      "epoch": 0.18830879071842926,
      "grad_norm": 0.34139570593833923,
      "learning_rate": 2e-05,
      "loss": 0.7122,
      "step": 2110
    },
    {
      "epoch": 0.1892012494422133,
      "grad_norm": 0.40417465567588806,
      "learning_rate": 2e-05,
      "loss": 0.7251,
      "step": 2120
    },
    {
      "epoch": 0.19009370816599733,
      "grad_norm": 0.41374412178993225,
      "learning_rate": 2e-05,
      "loss": 0.6949,
      "step": 2130
    },
    {
      "epoch": 0.19098616688978134,
      "grad_norm": 0.3738286793231964,
      "learning_rate": 2e-05,
      "loss": 0.6991,
      "step": 2140
    },
    {
      "epoch": 0.19187862561356536,
      "grad_norm": 0.36283162236213684,
      "learning_rate": 2e-05,
      "loss": 0.7033,
      "step": 2150
    },
    {
      "epoch": 0.1927710843373494,
      "grad_norm": 0.40305814146995544,
      "learning_rate": 2e-05,
      "loss": 0.7036,
      "step": 2160
    },
    {
      "epoch": 0.19366354306113343,
      "grad_norm": 0.4250462055206299,
      "learning_rate": 2e-05,
      "loss": 0.7248,
      "step": 2170
    },
    {
      "epoch": 0.19455600178491744,
      "grad_norm": 0.3769831359386444,
      "learning_rate": 2e-05,
      "loss": 0.7055,
      "step": 2180
    },
    {
      "epoch": 0.19544846050870146,
      "grad_norm": 0.35918813943862915,
      "learning_rate": 2e-05,
      "loss": 0.676,
      "step": 2190
    },
    {
      "epoch": 0.1963409192324855,
      "grad_norm": 0.3661561906337738,
      "learning_rate": 2e-05,
      "loss": 0.7058,
      "step": 2200
    },
    {
      "epoch": 0.19723337795626952,
      "grad_norm": 0.32479768991470337,
      "learning_rate": 2e-05,
      "loss": 0.6975,
      "step": 2210
    },
    {
      "epoch": 0.19812583668005354,
      "grad_norm": 0.36116334795951843,
      "learning_rate": 2e-05,
      "loss": 0.7006,
      "step": 2220
    },
    {
      "epoch": 0.1990182954038376,
      "grad_norm": 0.3757176995277405,
      "learning_rate": 2e-05,
      "loss": 0.6867,
      "step": 2230
    },
    {
      "epoch": 0.1999107541276216,
      "grad_norm": 0.3349403440952301,
      "learning_rate": 2e-05,
      "loss": 0.7119,
      "step": 2240
    },
    {
      "epoch": 0.20080321285140562,
      "grad_norm": 0.41477563977241516,
      "learning_rate": 2e-05,
      "loss": 0.6737,
      "step": 2250
    },
    {
      "epoch": 0.20169567157518964,
      "grad_norm": 0.7723550796508789,
      "learning_rate": 2e-05,
      "loss": 0.7063,
      "step": 2260
    },
    {
      "epoch": 0.20258813029897368,
      "grad_norm": 0.3693602979183197,
      "learning_rate": 2e-05,
      "loss": 0.7431,
      "step": 2270
    },
    {
      "epoch": 0.2034805890227577,
      "grad_norm": 0.4255744218826294,
      "learning_rate": 2e-05,
      "loss": 0.7051,
      "step": 2280
    },
    {
      "epoch": 0.20437304774654172,
      "grad_norm": 0.3678276836872101,
      "learning_rate": 2e-05,
      "loss": 0.6955,
      "step": 2290
    },
    {
      "epoch": 0.20526550647032574,
      "grad_norm": 0.39403510093688965,
      "learning_rate": 2e-05,
      "loss": 0.6889,
      "step": 2300
    },
    {
      "epoch": 0.20615796519410978,
      "grad_norm": 0.3570137023925781,
      "learning_rate": 2e-05,
      "loss": 0.7089,
      "step": 2310
    },
    {
      "epoch": 0.2070504239178938,
      "grad_norm": 0.3645356297492981,
      "learning_rate": 2e-05,
      "loss": 0.7132,
      "step": 2320
    },
    {
      "epoch": 0.20794288264167782,
      "grad_norm": 0.36887142062187195,
      "learning_rate": 2e-05,
      "loss": 0.7095,
      "step": 2330
    },
    {
      "epoch": 0.20883534136546184,
      "grad_norm": 0.3418295085430145,
      "learning_rate": 2e-05,
      "loss": 0.7006,
      "step": 2340
    },
    {
      "epoch": 0.20972780008924588,
      "grad_norm": 0.3889702260494232,
      "learning_rate": 2e-05,
      "loss": 0.7043,
      "step": 2350
    },
    {
      "epoch": 0.2106202588130299,
      "grad_norm": 0.3959186375141144,
      "learning_rate": 2e-05,
      "loss": 0.7222,
      "step": 2360
    },
    {
      "epoch": 0.21151271753681392,
      "grad_norm": 0.34344956278800964,
      "learning_rate": 2e-05,
      "loss": 0.708,
      "step": 2370
    },
    {
      "epoch": 0.21240517626059796,
      "grad_norm": 0.3525504171848297,
      "learning_rate": 2e-05,
      "loss": 0.7142,
      "step": 2380
    },
    {
      "epoch": 0.21329763498438198,
      "grad_norm": 0.36897340416908264,
      "learning_rate": 2e-05,
      "loss": 0.7233,
      "step": 2390
    },
    {
      "epoch": 0.214190093708166,
      "grad_norm": 0.3377510607242584,
      "learning_rate": 2e-05,
      "loss": 0.72,
      "step": 2400
    },
    {
      "epoch": 0.21508255243195,
      "grad_norm": 0.39910000562667847,
      "learning_rate": 2e-05,
      "loss": 0.7211,
      "step": 2410
    },
    {
      "epoch": 0.21597501115573406,
      "grad_norm": 0.3923272490501404,
      "learning_rate": 2e-05,
      "loss": 0.6979,
      "step": 2420
    },
    {
      "epoch": 0.21686746987951808,
      "grad_norm": 0.4154103398323059,
      "learning_rate": 2e-05,
      "loss": 0.7031,
      "step": 2430
    },
    {
      "epoch": 0.2177599286033021,
      "grad_norm": 0.40713241696357727,
      "learning_rate": 2e-05,
      "loss": 0.6988,
      "step": 2440
    },
    {
      "epoch": 0.2186523873270861,
      "grad_norm": 0.3655557930469513,
      "learning_rate": 2e-05,
      "loss": 0.6843,
      "step": 2450
    },
    {
      "epoch": 0.21954484605087016,
      "grad_norm": 0.3539635241031647,
      "learning_rate": 2e-05,
      "loss": 0.7085,
      "step": 2460
    },
    {
      "epoch": 0.22043730477465417,
      "grad_norm": 0.41050246357917786,
      "learning_rate": 2e-05,
      "loss": 0.6647,
      "step": 2470
    },
    {
      "epoch": 0.2213297634984382,
      "grad_norm": 0.36172372102737427,
      "learning_rate": 2e-05,
      "loss": 0.706,
      "step": 2480
    },
    {
      "epoch": 0.2222222222222222,
      "grad_norm": 0.5249997973442078,
      "learning_rate": 2e-05,
      "loss": 0.7144,
      "step": 2490
    },
    {
      "epoch": 0.22311468094600626,
      "grad_norm": 0.3864399194717407,
      "learning_rate": 2e-05,
      "loss": 0.7189,
      "step": 2500
    },
    {
      "epoch": 0.22400713966979027,
      "grad_norm": 0.4319363236427307,
      "learning_rate": 2e-05,
      "loss": 0.7015,
      "step": 2510
    },
    {
      "epoch": 0.2248995983935743,
      "grad_norm": 0.3635087013244629,
      "learning_rate": 2e-05,
      "loss": 0.6695,
      "step": 2520
    },
    {
      "epoch": 0.22579205711735834,
      "grad_norm": 0.35824811458587646,
      "learning_rate": 2e-05,
      "loss": 0.6949,
      "step": 2530
    },
    {
      "epoch": 0.22668451584114235,
      "grad_norm": 0.375846803188324,
      "learning_rate": 2e-05,
      "loss": 0.7237,
      "step": 2540
    },
    {
      "epoch": 0.22757697456492637,
      "grad_norm": 0.35899844765663147,
      "learning_rate": 2e-05,
      "loss": 0.6959,
      "step": 2550
    },
    {
      "epoch": 0.2284694332887104,
      "grad_norm": 0.3737565279006958,
      "learning_rate": 2e-05,
      "loss": 0.7104,
      "step": 2560
    },
    {
      "epoch": 0.22936189201249443,
      "grad_norm": 0.35111722350120544,
      "learning_rate": 2e-05,
      "loss": 0.6941,
      "step": 2570
    },
    {
      "epoch": 0.23025435073627845,
      "grad_norm": 0.3938366770744324,
      "learning_rate": 2e-05,
      "loss": 0.7205,
      "step": 2580
    },
    {
      "epoch": 0.23114680946006247,
      "grad_norm": 0.35663139820098877,
      "learning_rate": 2e-05,
      "loss": 0.7184,
      "step": 2590
    },
    {
      "epoch": 0.2320392681838465,
      "grad_norm": 0.42633557319641113,
      "learning_rate": 2e-05,
      "loss": 0.7065,
      "step": 2600
    },
    {
      "epoch": 0.23293172690763053,
      "grad_norm": 0.4921680986881256,
      "learning_rate": 2e-05,
      "loss": 0.706,
      "step": 2610
    },
    {
      "epoch": 0.23382418563141455,
      "grad_norm": 0.3506428301334381,
      "learning_rate": 2e-05,
      "loss": 0.721,
      "step": 2620
    },
    {
      "epoch": 0.23471664435519857,
      "grad_norm": 0.34051036834716797,
      "learning_rate": 2e-05,
      "loss": 0.6939,
      "step": 2630
    },
    {
      "epoch": 0.23560910307898258,
      "grad_norm": 0.36223891377449036,
      "learning_rate": 2e-05,
      "loss": 0.7122,
      "step": 2640
    },
    {
      "epoch": 0.23650156180276663,
      "grad_norm": 0.35965049266815186,
      "learning_rate": 2e-05,
      "loss": 0.709,
      "step": 2650
    },
    {
      "epoch": 0.23739402052655065,
      "grad_norm": 0.3486102223396301,
      "learning_rate": 2e-05,
      "loss": 0.693,
      "step": 2660
    },
    {
      "epoch": 0.23828647925033467,
      "grad_norm": 0.38577350974082947,
      "learning_rate": 2e-05,
      "loss": 0.7067,
      "step": 2670
    },
    {
      "epoch": 0.2391789379741187,
      "grad_norm": 0.40134578943252563,
      "learning_rate": 2e-05,
      "loss": 0.7132,
      "step": 2680
    },
    {
      "epoch": 0.24007139669790273,
      "grad_norm": 0.41890060901641846,
      "learning_rate": 2e-05,
      "loss": 0.7044,
      "step": 2690
    },
    {
      "epoch": 0.24096385542168675,
      "grad_norm": 0.3403701186180115,
      "learning_rate": 2e-05,
      "loss": 0.7058,
      "step": 2700
    },
    {
      "epoch": 0.24185631414547076,
      "grad_norm": 0.33793163299560547,
      "learning_rate": 2e-05,
      "loss": 0.7181,
      "step": 2710
    },
    {
      "epoch": 0.2427487728692548,
      "grad_norm": 0.5403984785079956,
      "learning_rate": 2e-05,
      "loss": 0.7062,
      "step": 2720
    },
    {
      "epoch": 0.24364123159303883,
      "grad_norm": 0.4084964394569397,
      "learning_rate": 2e-05,
      "loss": 0.6615,
      "step": 2730
    },
    {
      "epoch": 0.24453369031682284,
      "grad_norm": 0.3381352722644806,
      "learning_rate": 2e-05,
      "loss": 0.7106,
      "step": 2740
    },
    {
      "epoch": 0.24542614904060686,
      "grad_norm": 0.38011088967323303,
      "learning_rate": 2e-05,
      "loss": 0.6913,
      "step": 2750
    },
    {
      "epoch": 0.2463186077643909,
      "grad_norm": 0.3438170552253723,
      "learning_rate": 2e-05,
      "loss": 0.6822,
      "step": 2760
    },
    {
      "epoch": 0.24721106648817492,
      "grad_norm": 0.35488688945770264,
      "learning_rate": 2e-05,
      "loss": 0.7069,
      "step": 2770
    },
    {
      "epoch": 0.24810352521195894,
      "grad_norm": 0.3474864661693573,
      "learning_rate": 2e-05,
      "loss": 0.6961,
      "step": 2780
    },
    {
      "epoch": 0.24899598393574296,
      "grad_norm": 0.37238046526908875,
      "learning_rate": 2e-05,
      "loss": 0.6934,
      "step": 2790
    },
    {
      "epoch": 0.249888442659527,
      "grad_norm": 0.361629456281662,
      "learning_rate": 2e-05,
      "loss": 0.7176,
      "step": 2800
    },
    {
      "epoch": 0.250780901383311,
      "grad_norm": 0.36093610525131226,
      "learning_rate": 2e-05,
      "loss": 0.6631,
      "step": 2810
    },
    {
      "epoch": 0.25167336010709507,
      "grad_norm": 0.354215145111084,
      "learning_rate": 2e-05,
      "loss": 0.685,
      "step": 2820
    },
    {
      "epoch": 0.2525658188308791,
      "grad_norm": 0.3703426122665405,
      "learning_rate": 2e-05,
      "loss": 0.6811,
      "step": 2830
    },
    {
      "epoch": 0.2534582775546631,
      "grad_norm": 0.3834165036678314,
      "learning_rate": 2e-05,
      "loss": 0.7128,
      "step": 2840
    },
    {
      "epoch": 0.2543507362784471,
      "grad_norm": 0.35115286707878113,
      "learning_rate": 2e-05,
      "loss": 0.6888,
      "step": 2850
    },
    {
      "epoch": 0.25524319500223114,
      "grad_norm": 0.3497035503387451,
      "learning_rate": 2e-05,
      "loss": 0.6511,
      "step": 2860
    },
    {
      "epoch": 0.25613565372601516,
      "grad_norm": 0.3738860487937927,
      "learning_rate": 2e-05,
      "loss": 0.7181,
      "step": 2870
    },
    {
      "epoch": 0.2570281124497992,
      "grad_norm": 0.3604569733142853,
      "learning_rate": 2e-05,
      "loss": 0.6982,
      "step": 2880
    },
    {
      "epoch": 0.25792057117358325,
      "grad_norm": 0.3718276917934418,
      "learning_rate": 2e-05,
      "loss": 0.7142,
      "step": 2890
    },
    {
      "epoch": 0.25881302989736726,
      "grad_norm": 0.3611880838871002,
      "learning_rate": 2e-05,
      "loss": 0.702,
      "step": 2900
    },
    {
      "epoch": 0.2597054886211513,
      "grad_norm": 0.33287131786346436,
      "learning_rate": 2e-05,
      "loss": 0.684,
      "step": 2910
    },
    {
      "epoch": 0.2605979473449353,
      "grad_norm": 0.40636178851127625,
      "learning_rate": 2e-05,
      "loss": 0.6977,
      "step": 2920
    },
    {
      "epoch": 0.2614904060687193,
      "grad_norm": 0.3703216016292572,
      "learning_rate": 2e-05,
      "loss": 0.6841,
      "step": 2930
    },
    {
      "epoch": 0.26238286479250333,
      "grad_norm": 0.3907012343406677,
      "learning_rate": 2e-05,
      "loss": 0.6648,
      "step": 2940
    },
    {
      "epoch": 0.26327532351628735,
      "grad_norm": 0.357471764087677,
      "learning_rate": 2e-05,
      "loss": 0.6518,
      "step": 2950
    },
    {
      "epoch": 0.26416778224007137,
      "grad_norm": 0.38111019134521484,
      "learning_rate": 2e-05,
      "loss": 0.7083,
      "step": 2960
    },
    {
      "epoch": 0.26506024096385544,
      "grad_norm": 0.3443157970905304,
      "learning_rate": 2e-05,
      "loss": 0.68,
      "step": 2970
    },
    {
      "epoch": 0.26595269968763946,
      "grad_norm": 0.3853066563606262,
      "learning_rate": 2e-05,
      "loss": 0.7192,
      "step": 2980
    },
    {
      "epoch": 0.2668451584114235,
      "grad_norm": 0.43045032024383545,
      "learning_rate": 2e-05,
      "loss": 0.7119,
      "step": 2990
    },
    {
      "epoch": 0.2677376171352075,
      "grad_norm": 0.46091946959495544,
      "learning_rate": 2e-05,
      "loss": 0.6947,
      "step": 3000
    },
    {
      "epoch": 0.2686300758589915,
      "grad_norm": 0.3609314262866974,
      "learning_rate": 2e-05,
      "loss": 0.6636,
      "step": 3010
    },
    {
      "epoch": 0.26952253458277553,
      "grad_norm": 0.3863031268119812,
      "learning_rate": 2e-05,
      "loss": 0.6821,
      "step": 3020
    },
    {
      "epoch": 0.27041499330655955,
      "grad_norm": 0.4071292579174042,
      "learning_rate": 2e-05,
      "loss": 0.7093,
      "step": 3030
    },
    {
      "epoch": 0.2713074520303436,
      "grad_norm": 0.34357038140296936,
      "learning_rate": 2e-05,
      "loss": 0.7139,
      "step": 3040
    },
    {
      "epoch": 0.27219991075412764,
      "grad_norm": 0.4047830402851105,
      "learning_rate": 2e-05,
      "loss": 0.6868,
      "step": 3050
    },
    {
      "epoch": 0.27309236947791166,
      "grad_norm": 0.37571510672569275,
      "learning_rate": 2e-05,
      "loss": 0.6568,
      "step": 3060
    },
    {
      "epoch": 0.2739848282016957,
      "grad_norm": 0.339928537607193,
      "learning_rate": 2e-05,
      "loss": 0.6987,
      "step": 3070
    },
    {
      "epoch": 0.2748772869254797,
      "grad_norm": 0.36671870946884155,
      "learning_rate": 2e-05,
      "loss": 0.706,
      "step": 3080
    },
    {
      "epoch": 0.2757697456492637,
      "grad_norm": 0.3619630038738251,
      "learning_rate": 2e-05,
      "loss": 0.6936,
      "step": 3090
    },
    {
      "epoch": 0.2766622043730477,
      "grad_norm": 0.3525165915489197,
      "learning_rate": 2e-05,
      "loss": 0.7032,
      "step": 3100
    },
    {
      "epoch": 0.27755466309683174,
      "grad_norm": 0.3450978100299835,
      "learning_rate": 2e-05,
      "loss": 0.7087,
      "step": 3110
    },
    {
      "epoch": 0.2784471218206158,
      "grad_norm": 0.3459194600582123,
      "learning_rate": 2e-05,
      "loss": 0.6823,
      "step": 3120
    },
    {
      "epoch": 0.27933958054439983,
      "grad_norm": 0.38839390873908997,
      "learning_rate": 2e-05,
      "loss": 0.6859,
      "step": 3130
    },
    {
      "epoch": 0.28023203926818385,
      "grad_norm": 0.3918834328651428,
      "learning_rate": 2e-05,
      "loss": 0.7154,
      "step": 3140
    },
    {
      "epoch": 0.28112449799196787,
      "grad_norm": 0.36024022102355957,
      "learning_rate": 2e-05,
      "loss": 0.6795,
      "step": 3150
    },
    {
      "epoch": 0.2820169567157519,
      "grad_norm": 0.36553332209587097,
      "learning_rate": 2e-05,
      "loss": 0.651,
      "step": 3160
    },
    {
      "epoch": 0.2829094154395359,
      "grad_norm": 0.3750966787338257,
      "learning_rate": 2e-05,
      "loss": 0.6676,
      "step": 3170
    },
    {
      "epoch": 0.2838018741633199,
      "grad_norm": 0.36494848132133484,
      "learning_rate": 2e-05,
      "loss": 0.6919,
      "step": 3180
    },
    {
      "epoch": 0.284694332887104,
      "grad_norm": 0.42444247007369995,
      "learning_rate": 2e-05,
      "loss": 0.697,
      "step": 3190
    },
    {
      "epoch": 0.285586791610888,
      "grad_norm": 0.4227437973022461,
      "learning_rate": 2e-05,
      "loss": 0.6557,
      "step": 3200
    },
    {
      "epoch": 0.28647925033467203,
      "grad_norm": 0.3764297068119049,
      "learning_rate": 2e-05,
      "loss": 0.6852,
      "step": 3210
    },
    {
      "epoch": 0.28737170905845605,
      "grad_norm": 0.36593905091285706,
      "learning_rate": 2e-05,
      "loss": 0.6957,
      "step": 3220
    },
    {
      "epoch": 0.28826416778224007,
      "grad_norm": 0.3768455684185028,
      "learning_rate": 2e-05,
      "loss": 0.6956,
      "step": 3230
    },
    {
      "epoch": 0.2891566265060241,
      "grad_norm": 0.3834182024002075,
      "learning_rate": 2e-05,
      "loss": 0.6837,
      "step": 3240
    },
    {
      "epoch": 0.2900490852298081,
      "grad_norm": 0.37360700964927673,
      "learning_rate": 2e-05,
      "loss": 0.6778,
      "step": 3250
    },
    {
      "epoch": 0.2909415439535921,
      "grad_norm": 0.3748886287212372,
      "learning_rate": 2e-05,
      "loss": 0.6962,
      "step": 3260
    },
    {
      "epoch": 0.2918340026773762,
      "grad_norm": 0.36913779377937317,
      "learning_rate": 2e-05,
      "loss": 0.7171,
      "step": 3270
    },
    {
      "epoch": 0.2927264614011602,
      "grad_norm": 0.4049180746078491,
      "learning_rate": 2e-05,
      "loss": 0.6959,
      "step": 3280
    },
    {
      "epoch": 0.2936189201249442,
      "grad_norm": 0.4105624258518219,
      "learning_rate": 2e-05,
      "loss": 0.669,
      "step": 3290
    },
    {
      "epoch": 0.29451137884872824,
      "grad_norm": 0.36357712745666504,
      "learning_rate": 2e-05,
      "loss": 0.7103,
      "step": 3300
    },
    {
      "epoch": 0.29540383757251226,
      "grad_norm": 0.3372548520565033,
      "learning_rate": 2e-05,
      "loss": 0.669,
      "step": 3310
    },
    {
      "epoch": 0.2962962962962963,
      "grad_norm": 0.360275000333786,
      "learning_rate": 2e-05,
      "loss": 0.7128,
      "step": 3320
    },
    {
      "epoch": 0.2971887550200803,
      "grad_norm": 0.36750873923301697,
      "learning_rate": 2e-05,
      "loss": 0.7159,
      "step": 3330
    },
    {
      "epoch": 0.29808121374386437,
      "grad_norm": 0.3744310140609741,
      "learning_rate": 2e-05,
      "loss": 0.6991,
      "step": 3340
    },
    {
      "epoch": 0.2989736724676484,
      "grad_norm": 0.36482685804367065,
      "learning_rate": 2e-05,
      "loss": 0.6761,
      "step": 3350
    },
    {
      "epoch": 0.2998661311914324,
      "grad_norm": 0.4055212438106537,
      "learning_rate": 2e-05,
      "loss": 0.6832,
      "step": 3360
    },
    {
      "epoch": 0.3007585899152164,
      "grad_norm": 0.3472900390625,
      "learning_rate": 2e-05,
      "loss": 0.657,
      "step": 3370
    },
    {
      "epoch": 0.30165104863900044,
      "grad_norm": 0.41277948021888733,
      "learning_rate": 2e-05,
      "loss": 0.7004,
      "step": 3380
    },
    {
      "epoch": 0.30254350736278446,
      "grad_norm": 0.5707241892814636,
      "learning_rate": 2e-05,
      "loss": 0.6946,
      "step": 3390
    },
    {
      "epoch": 0.3034359660865685,
      "grad_norm": 0.37944552302360535,
      "learning_rate": 2e-05,
      "loss": 0.7048,
      "step": 3400
    },
    {
      "epoch": 0.3043284248103525,
      "grad_norm": 0.40345481038093567,
      "learning_rate": 2e-05,
      "loss": 0.6784,
      "step": 3410
    },
    {
      "epoch": 0.30522088353413657,
      "grad_norm": 0.36031800508499146,
      "learning_rate": 2e-05,
      "loss": 0.6687,
      "step": 3420
    },
    {
      "epoch": 0.3061133422579206,
      "grad_norm": 0.3758523762226105,
      "learning_rate": 2e-05,
      "loss": 0.713,
      "step": 3430
    },
    {
      "epoch": 0.3070058009817046,
      "grad_norm": 0.5176293849945068,
      "learning_rate": 2e-05,
      "loss": 0.6768,
      "step": 3440
    },
    {
      "epoch": 0.3078982597054886,
      "grad_norm": 0.3904946744441986,
      "learning_rate": 2e-05,
      "loss": 0.7088,
      "step": 3450
    },
    {
      "epoch": 0.30879071842927264,
      "grad_norm": 0.3160179555416107,
      "learning_rate": 2e-05,
      "loss": 0.6631,
      "step": 3460
    },
    {
      "epoch": 0.30968317715305665,
      "grad_norm": 0.45626139640808105,
      "learning_rate": 2e-05,
      "loss": 0.6798,
      "step": 3470
    },
    {
      "epoch": 0.31057563587684067,
      "grad_norm": 0.3389970064163208,
      "learning_rate": 2e-05,
      "loss": 0.685,
      "step": 3480
    },
    {
      "epoch": 0.31146809460062475,
      "grad_norm": 0.37763211131095886,
      "learning_rate": 2e-05,
      "loss": 0.6879,
      "step": 3490
    },
    {
      "epoch": 0.31236055332440876,
      "grad_norm": 0.374933123588562,
      "learning_rate": 2e-05,
      "loss": 0.6715,
      "step": 3500
    },
    {
      "epoch": 0.3132530120481928,
      "grad_norm": 0.37782981991767883,
      "learning_rate": 2e-05,
      "loss": 0.6852,
      "step": 3510
    },
    {
      "epoch": 0.3141454707719768,
      "grad_norm": 0.4102741479873657,
      "learning_rate": 2e-05,
      "loss": 0.7064,
      "step": 3520
    },
    {
      "epoch": 0.3150379294957608,
      "grad_norm": 0.36353591084480286,
      "learning_rate": 2e-05,
      "loss": 0.6669,
      "step": 3530
    },
    {
      "epoch": 0.31593038821954483,
      "grad_norm": 0.4682978391647339,
      "learning_rate": 2e-05,
      "loss": 0.6849,
      "step": 3540
    },
    {
      "epoch": 0.31682284694332885,
      "grad_norm": 0.35799211263656616,
      "learning_rate": 2e-05,
      "loss": 0.6638,
      "step": 3550
    },
    {
      "epoch": 0.31771530566711287,
      "grad_norm": 0.35494521260261536,
      "learning_rate": 2e-05,
      "loss": 0.7194,
      "step": 3560
    },
    {
      "epoch": 0.31860776439089694,
      "grad_norm": 0.3495546877384186,
      "learning_rate": 2e-05,
      "loss": 0.7101,
      "step": 3570
    },
    {
      "epoch": 0.31950022311468096,
      "grad_norm": 0.42559295892715454,
      "learning_rate": 2e-05,
      "loss": 0.729,
      "step": 3580
    },
    {
      "epoch": 0.320392681838465,
      "grad_norm": 0.40502825379371643,
      "learning_rate": 2e-05,
      "loss": 0.6711,
      "step": 3590
    },
    {
      "epoch": 0.321285140562249,
      "grad_norm": 0.39933738112449646,
      "learning_rate": 2e-05,
      "loss": 0.6846,
      "step": 3600
    },
    {
      "epoch": 0.322177599286033,
      "grad_norm": 0.39374107122421265,
      "learning_rate": 2e-05,
      "loss": 0.6934,
      "step": 3610
    },
    {
      "epoch": 0.32307005800981703,
      "grad_norm": 0.3833998739719391,
      "learning_rate": 2e-05,
      "loss": 0.7001,
      "step": 3620
    },
    {
      "epoch": 0.32396251673360105,
      "grad_norm": 0.4226708710193634,
      "learning_rate": 2e-05,
      "loss": 0.7118,
      "step": 3630
    },
    {
      "epoch": 0.3248549754573851,
      "grad_norm": 0.3613390028476715,
      "learning_rate": 2e-05,
      "loss": 0.6712,
      "step": 3640
    },
    {
      "epoch": 0.32574743418116914,
      "grad_norm": 0.32615116238594055,
      "learning_rate": 2e-05,
      "loss": 0.6761,
      "step": 3650
    },
    {
      "epoch": 0.32663989290495316,
      "grad_norm": 0.36767154932022095,
      "learning_rate": 2e-05,
      "loss": 0.6738,
      "step": 3660
    },
    {
      "epoch": 0.3275323516287372,
      "grad_norm": 0.3822571635246277,
      "learning_rate": 2e-05,
      "loss": 0.6959,
      "step": 3670
    },
    {
      "epoch": 0.3284248103525212,
      "grad_norm": 0.3931572139263153,
      "learning_rate": 2e-05,
      "loss": 0.7079,
      "step": 3680
    },
    {
      "epoch": 0.3293172690763052,
      "grad_norm": 0.3877595365047455,
      "learning_rate": 2e-05,
      "loss": 0.6933,
      "step": 3690
    },
    {
      "epoch": 0.3302097278000892,
      "grad_norm": 0.3704959750175476,
      "learning_rate": 2e-05,
      "loss": 0.6628,
      "step": 3700
    },
    {
      "epoch": 0.33110218652387324,
      "grad_norm": 0.3556361198425293,
      "learning_rate": 2e-05,
      "loss": 0.6704,
      "step": 3710
    },
    {
      "epoch": 0.3319946452476573,
      "grad_norm": 0.362934410572052,
      "learning_rate": 2e-05,
      "loss": 0.6598,
      "step": 3720
    },
    {
      "epoch": 0.33288710397144133,
      "grad_norm": 0.3868967890739441,
      "learning_rate": 2e-05,
      "loss": 0.7028,
      "step": 3730
    },
    {
      "epoch": 0.33377956269522535,
      "grad_norm": 0.38233718276023865,
      "learning_rate": 2e-05,
      "loss": 0.6698,
      "step": 3740
    },
    {
      "epoch": 0.33467202141900937,
      "grad_norm": 0.3984028398990631,
      "learning_rate": 2e-05,
      "loss": 0.6998,
      "step": 3750
    },
    {
      "epoch": 0.3355644801427934,
      "grad_norm": 0.3509763777256012,
      "learning_rate": 2e-05,
      "loss": 0.6784,
      "step": 3760
    },
    {
      "epoch": 0.3364569388665774,
      "grad_norm": 0.3608149588108063,
      "learning_rate": 2e-05,
      "loss": 0.6694,
      "step": 3770
    },
    {
      "epoch": 0.3373493975903614,
      "grad_norm": 0.3655262291431427,
      "learning_rate": 2e-05,
      "loss": 0.6604,
      "step": 3780
    },
    {
      "epoch": 0.3382418563141455,
      "grad_norm": 0.3790647089481354,
      "learning_rate": 2e-05,
      "loss": 0.6651,
      "step": 3790
    },
    {
      "epoch": 0.3391343150379295,
      "grad_norm": 0.35572123527526855,
      "learning_rate": 2e-05,
      "loss": 0.7299,
      "step": 3800
    },
    {
      "epoch": 0.34002677376171353,
      "grad_norm": 0.3570621609687805,
      "learning_rate": 2e-05,
      "loss": 0.681,
      "step": 3810
    },
    {
      "epoch": 0.34091923248549755,
      "grad_norm": 0.3663609027862549,
      "learning_rate": 2e-05,
      "loss": 0.6557,
      "step": 3820
    },
    {
      "epoch": 0.34181169120928157,
      "grad_norm": 0.37612035870552063,
      "learning_rate": 2e-05,
      "loss": 0.6829,
      "step": 3830
    },
    {
      "epoch": 0.3427041499330656,
      "grad_norm": 0.3695187568664551,
      "learning_rate": 2e-05,
      "loss": 0.6792,
      "step": 3840
    },
    {
      "epoch": 0.3435966086568496,
      "grad_norm": 0.3543972671031952,
      "learning_rate": 2e-05,
      "loss": 0.6603,
      "step": 3850
    },
    {
      "epoch": 0.3444890673806337,
      "grad_norm": 0.3484421372413635,
      "learning_rate": 2e-05,
      "loss": 0.6412,
      "step": 3860
    },
    {
      "epoch": 0.3453815261044177,
      "grad_norm": 0.636989414691925,
      "learning_rate": 2e-05,
      "loss": 0.6685,
      "step": 3870
    },
    {
      "epoch": 0.3462739848282017,
      "grad_norm": 0.353354275226593,
      "learning_rate": 2e-05,
      "loss": 0.6816,
      "step": 3880
    },
    {
      "epoch": 0.3471664435519857,
      "grad_norm": 0.399431049823761,
      "learning_rate": 2e-05,
      "loss": 0.6821,
      "step": 3890
    },
    {
      "epoch": 0.34805890227576974,
      "grad_norm": 0.3725152909755707,
      "learning_rate": 2e-05,
      "loss": 0.6705,
      "step": 3900
    },
    {
      "epoch": 0.34895136099955376,
      "grad_norm": 0.3697719871997833,
      "learning_rate": 2e-05,
      "loss": 0.6736,
      "step": 3910
    },
    {
      "epoch": 0.3498438197233378,
      "grad_norm": 0.37048324942588806,
      "learning_rate": 2e-05,
      "loss": 0.6881,
      "step": 3920
    },
    {
      "epoch": 0.3507362784471218,
      "grad_norm": 0.4095633029937744,
      "learning_rate": 2e-05,
      "loss": 0.6529,
      "step": 3930
    },
    {
      "epoch": 0.35162873717090587,
      "grad_norm": 0.35396143794059753,
      "learning_rate": 2e-05,
      "loss": 0.7209,
      "step": 3940
    },
    {
      "epoch": 0.3525211958946899,
      "grad_norm": 0.3832853436470032,
      "learning_rate": 2e-05,
      "loss": 0.6959,
      "step": 3950
    },
    {
      "epoch": 0.3534136546184739,
      "grad_norm": 0.37400299310684204,
      "learning_rate": 2e-05,
      "loss": 0.6638,
      "step": 3960
    },
    {
      "epoch": 0.3543061133422579,
      "grad_norm": 0.7657487988471985,
      "learning_rate": 2e-05,
      "loss": 0.6751,
      "step": 3970
    },
    {
      "epoch": 0.35519857206604194,
      "grad_norm": 0.3688769042491913,
      "learning_rate": 2e-05,
      "loss": 0.6851,
      "step": 3980
    },
    {
      "epoch": 0.35609103078982596,
      "grad_norm": 0.3446170389652252,
      "learning_rate": 2e-05,
      "loss": 0.6704,
      "step": 3990
    },
    {
      "epoch": 0.35698348951361,
      "grad_norm": 0.38899853825569153,
      "learning_rate": 2e-05,
      "loss": 0.6728,
      "step": 4000
    },
    {
      "epoch": 0.35787594823739405,
      "grad_norm": 0.3742004930973053,
      "learning_rate": 2e-05,
      "loss": 0.6821,
      "step": 4010
    },
    {
      "epoch": 0.35876840696117807,
      "grad_norm": 0.36814481019973755,
      "learning_rate": 2e-05,
      "loss": 0.6905,
      "step": 4020
    },
    {
      "epoch": 0.3596608656849621,
      "grad_norm": 0.3448152542114258,
      "learning_rate": 2e-05,
      "loss": 0.6751,
      "step": 4030
    },
    {
      "epoch": 0.3605533244087461,
      "grad_norm": 0.3532841205596924,
      "learning_rate": 2e-05,
      "loss": 0.6663,
      "step": 4040
    },
    {
      "epoch": 0.3614457831325301,
      "grad_norm": 0.390525758266449,
      "learning_rate": 2e-05,
      "loss": 0.6676,
      "step": 4050
    },
    {
      "epoch": 0.36233824185631414,
      "grad_norm": 0.41043761372566223,
      "learning_rate": 2e-05,
      "loss": 0.6865,
      "step": 4060
    },
    {
      "epoch": 0.36323070058009815,
      "grad_norm": 0.32364386320114136,
      "learning_rate": 2e-05,
      "loss": 0.6736,
      "step": 4070
    },
    {
      "epoch": 0.36412315930388217,
      "grad_norm": 0.39562374353408813,
      "learning_rate": 2e-05,
      "loss": 0.6816,
      "step": 4080
    },
    {
      "epoch": 0.36501561802766624,
      "grad_norm": 0.38002610206604004,
      "learning_rate": 2e-05,
      "loss": 0.6635,
      "step": 4090
    },
    {
      "epoch": 0.36590807675145026,
      "grad_norm": 0.34306979179382324,
      "learning_rate": 2e-05,
      "loss": 0.7087,
      "step": 4100
    },
    {
      "epoch": 0.3668005354752343,
      "grad_norm": 0.36761438846588135,
      "learning_rate": 2e-05,
      "loss": 0.6806,
      "step": 4110
    },
    {
      "epoch": 0.3676929941990183,
      "grad_norm": 0.39585381746292114,
      "learning_rate": 2e-05,
      "loss": 0.6927,
      "step": 4120
    },
    {
      "epoch": 0.3685854529228023,
      "grad_norm": 0.38605520129203796,
      "learning_rate": 2e-05,
      "loss": 0.6897,
      "step": 4130
    },
    {
      "epoch": 0.36947791164658633,
      "grad_norm": 0.3497367799282074,
      "learning_rate": 2e-05,
      "loss": 0.6812,
      "step": 4140
    },
    {
      "epoch": 0.37037037037037035,
      "grad_norm": 4.066261291503906,
      "learning_rate": 2e-05,
      "loss": 0.7264,
      "step": 4150
    },
    {
      "epoch": 0.3712628290941544,
      "grad_norm": 0.40141093730926514,
      "learning_rate": 2e-05,
      "loss": 0.693,
      "step": 4160
    },
    {
      "epoch": 0.37215528781793844,
      "grad_norm": 0.3579643666744232,
      "learning_rate": 2e-05,
      "loss": 0.7018,
      "step": 4170
    },
    {
      "epoch": 0.37304774654172246,
      "grad_norm": 0.3658234775066376,
      "learning_rate": 2e-05,
      "loss": 0.6767,
      "step": 4180
    },
    {
      "epoch": 0.3739402052655065,
      "grad_norm": 0.4451167583465576,
      "learning_rate": 2e-05,
      "loss": 0.7064,
      "step": 4190
    },
    {
      "epoch": 0.3748326639892905,
      "grad_norm": 0.3791666328907013,
      "learning_rate": 2e-05,
      "loss": 0.6824,
      "step": 4200
    },
    {
      "epoch": 0.3757251227130745,
      "grad_norm": 0.4027309715747833,
      "learning_rate": 2e-05,
      "loss": 0.6778,
      "step": 4210
    },
    {
      "epoch": 0.37661758143685853,
      "grad_norm": 0.3656576871871948,
      "learning_rate": 2e-05,
      "loss": 0.6654,
      "step": 4220
    },
    {
      "epoch": 0.37751004016064255,
      "grad_norm": 0.3933909237384796,
      "learning_rate": 2e-05,
      "loss": 0.6497,
      "step": 4230
    },
    {
      "epoch": 0.3784024988844266,
      "grad_norm": 0.36298003792762756,
      "learning_rate": 2e-05,
      "loss": 0.6643,
      "step": 4240
    },
    {
      "epoch": 0.37929495760821064,
      "grad_norm": 0.41185012459754944,
      "learning_rate": 2e-05,
      "loss": 0.6787,
      "step": 4250
    },
    {
      "epoch": 0.38018741633199465,
      "grad_norm": 0.40756914019584656,
      "learning_rate": 2e-05,
      "loss": 0.716,
      "step": 4260
    },
    {
      "epoch": 0.38107987505577867,
      "grad_norm": 0.4002636671066284,
      "learning_rate": 2e-05,
      "loss": 0.6634,
      "step": 4270
    },
    {
      "epoch": 0.3819723337795627,
      "grad_norm": 0.35606589913368225,
      "learning_rate": 2e-05,
      "loss": 0.6638,
      "step": 4280
    },
    {
      "epoch": 0.3828647925033467,
      "grad_norm": 0.36240053176879883,
      "learning_rate": 2e-05,
      "loss": 0.6598,
      "step": 4290
    },
    {
      "epoch": 0.3837572512271307,
      "grad_norm": 0.3612985610961914,
      "learning_rate": 2e-05,
      "loss": 0.6844,
      "step": 4300
    },
    {
      "epoch": 0.3846497099509148,
      "grad_norm": 0.873512864112854,
      "learning_rate": 2e-05,
      "loss": 0.6547,
      "step": 4310
    },
    {
      "epoch": 0.3855421686746988,
      "grad_norm": 0.3159959614276886,
      "learning_rate": 2e-05,
      "loss": 0.6505,
      "step": 4320
    },
    {
      "epoch": 0.38643462739848283,
      "grad_norm": 0.3532978892326355,
      "learning_rate": 2e-05,
      "loss": 0.6909,
      "step": 4330
    },
    {
      "epoch": 0.38732708612226685,
      "grad_norm": 0.3583095669746399,
      "learning_rate": 2e-05,
      "loss": 0.6779,
      "step": 4340
    },
    {
      "epoch": 0.38821954484605087,
      "grad_norm": 0.407150000333786,
      "learning_rate": 2e-05,
      "loss": 0.6786,
      "step": 4350
    },
    {
      "epoch": 0.3891120035698349,
      "grad_norm": 0.3659987449645996,
      "learning_rate": 2e-05,
      "loss": 0.6788,
      "step": 4360
    },
    {
      "epoch": 0.3900044622936189,
      "grad_norm": 0.3716491460800171,
      "learning_rate": 2e-05,
      "loss": 0.668,
      "step": 4370
    },
    {
      "epoch": 0.3908969210174029,
      "grad_norm": 0.38405466079711914,
      "learning_rate": 2e-05,
      "loss": 0.6684,
      "step": 4380
    },
    {
      "epoch": 0.391789379741187,
      "grad_norm": 0.3628016710281372,
      "learning_rate": 2e-05,
      "loss": 0.6983,
      "step": 4390
    },
    {
      "epoch": 0.392681838464971,
      "grad_norm": 0.39058199524879456,
      "learning_rate": 2e-05,
      "loss": 0.6593,
      "step": 4400
    },
    {
      "epoch": 0.39357429718875503,
      "grad_norm": 0.3607003390789032,
      "learning_rate": 2e-05,
      "loss": 0.6737,
      "step": 4410
    },
    {
      "epoch": 0.39446675591253905,
      "grad_norm": 0.49269410967826843,
      "learning_rate": 2e-05,
      "loss": 0.6588,
      "step": 4420
    },
    {
      "epoch": 0.39535921463632306,
      "grad_norm": 0.37630751729011536,
      "learning_rate": 2e-05,
      "loss": 0.6647,
      "step": 4430
    },
    {
      "epoch": 0.3962516733601071,
      "grad_norm": 0.38567686080932617,
      "learning_rate": 2e-05,
      "loss": 0.6516,
      "step": 4440
    },
    {
      "epoch": 0.3971441320838911,
      "grad_norm": 0.3609790802001953,
      "learning_rate": 2e-05,
      "loss": 0.6845,
      "step": 4450
    },
    {
      "epoch": 0.3980365908076752,
      "grad_norm": 0.3757692277431488,
      "learning_rate": 2e-05,
      "loss": 0.6646,
      "step": 4460
    },
    {
      "epoch": 0.3989290495314592,
      "grad_norm": 0.35269299149513245,
      "learning_rate": 2e-05,
      "loss": 0.6779,
      "step": 4470
    },
    {
      "epoch": 0.3998215082552432,
      "grad_norm": 0.3557558059692383,
      "learning_rate": 2e-05,
      "loss": 0.6973,
      "step": 4480
    },
    {
      "epoch": 0.4007139669790272,
      "grad_norm": 0.3537740707397461,
      "learning_rate": 2e-05,
      "loss": 0.6511,
      "step": 4490
    },
    {
      "epoch": 0.40160642570281124,
      "grad_norm": 0.3563501834869385,
      "learning_rate": 2e-05,
      "loss": 0.6701,
      "step": 4500
    },
    {
      "epoch": 0.40249888442659526,
      "grad_norm": 0.3577269911766052,
      "learning_rate": 2e-05,
      "loss": 0.6839,
      "step": 4510
    },
    {
      "epoch": 0.4033913431503793,
      "grad_norm": 0.34676194190979004,
      "learning_rate": 2e-05,
      "loss": 0.6549,
      "step": 4520
    },
    {
      "epoch": 0.4042838018741633,
      "grad_norm": 0.40689483284950256,
      "learning_rate": 2e-05,
      "loss": 0.6689,
      "step": 4530
    },
    {
      "epoch": 0.40517626059794737,
      "grad_norm": 0.35380804538726807,
      "learning_rate": 2e-05,
      "loss": 0.6631,
      "step": 4540
    },
    {
      "epoch": 0.4060687193217314,
      "grad_norm": 0.4036043882369995,
      "learning_rate": 2e-05,
      "loss": 0.6916,
      "step": 4550
    },
    {
      "epoch": 0.4069611780455154,
      "grad_norm": 0.39215508103370667,
      "learning_rate": 2e-05,
      "loss": 0.6759,
      "step": 4560
    },
    {
      "epoch": 0.4078536367692994,
      "grad_norm": 0.33775123953819275,
      "learning_rate": 2e-05,
      "loss": 0.6527,
      "step": 4570
    },
    {
      "epoch": 0.40874609549308344,
      "grad_norm": 0.35432910919189453,
      "learning_rate": 2e-05,
      "loss": 0.6624,
      "step": 4580
    },
    {
      "epoch": 0.40963855421686746,
      "grad_norm": 0.47398561239242554,
      "learning_rate": 2e-05,
      "loss": 0.669,
      "step": 4590
    },
    {
      "epoch": 0.4105310129406515,
      "grad_norm": 0.3641202449798584,
      "learning_rate": 2e-05,
      "loss": 0.6768,
      "step": 4600
    },
    {
      "epoch": 0.41142347166443555,
      "grad_norm": 0.3806946277618408,
      "learning_rate": 2e-05,
      "loss": 0.6915,
      "step": 4610
    },
    {
      "epoch": 0.41231593038821956,
      "grad_norm": 0.39436906576156616,
      "learning_rate": 2e-05,
      "loss": 0.6989,
      "step": 4620
    },
    {
      "epoch": 0.4132083891120036,
      "grad_norm": 0.39246389269828796,
      "learning_rate": 2e-05,
      "loss": 0.6684,
      "step": 4630
    },
    {
      "epoch": 0.4141008478357876,
      "grad_norm": 0.36079880595207214,
      "learning_rate": 2e-05,
      "loss": 0.6804,
      "step": 4640
    },
    {
      "epoch": 0.4149933065595716,
      "grad_norm": 0.3715951144695282,
      "learning_rate": 2e-05,
      "loss": 0.6908,
      "step": 4650
    },
    {
      "epoch": 0.41588576528335564,
      "grad_norm": 0.35639339685440063,
      "learning_rate": 2e-05,
      "loss": 0.6645,
      "step": 4660
    },
    {
      "epoch": 0.41677822400713965,
      "grad_norm": 0.37775149941444397,
      "learning_rate": 2e-05,
      "loss": 0.6581,
      "step": 4670
    },
    {
      "epoch": 0.41767068273092367,
      "grad_norm": 0.3476824462413788,
      "learning_rate": 2e-05,
      "loss": 0.6469,
      "step": 4680
    },
    {
      "epoch": 0.41856314145470774,
      "grad_norm": 0.33308857679367065,
      "learning_rate": 2e-05,
      "loss": 0.6528,
      "step": 4690
    },
    {
      "epoch": 0.41945560017849176,
      "grad_norm": 0.3618086278438568,
      "learning_rate": 2e-05,
      "loss": 0.6824,
      "step": 4700
    },
    {
      "epoch": 0.4203480589022758,
      "grad_norm": 0.36656132340431213,
      "learning_rate": 2e-05,
      "loss": 0.6754,
      "step": 4710
    },
    {
      "epoch": 0.4212405176260598,
      "grad_norm": 0.36235707998275757,
      "learning_rate": 2e-05,
      "loss": 0.6905,
      "step": 4720
    },
    {
      "epoch": 0.4221329763498438,
      "grad_norm": 0.35832539200782776,
      "learning_rate": 2e-05,
      "loss": 0.6703,
      "step": 4730
    },
    {
      "epoch": 0.42302543507362783,
      "grad_norm": 0.4274637699127197,
      "learning_rate": 2e-05,
      "loss": 0.6881,
      "step": 4740
    },
    {
      "epoch": 0.42391789379741185,
      "grad_norm": 0.3612465560436249,
      "learning_rate": 2e-05,
      "loss": 0.6648,
      "step": 4750
    },
    {
      "epoch": 0.4248103525211959,
      "grad_norm": 0.3838100731372833,
      "learning_rate": 2e-05,
      "loss": 0.6885,
      "step": 4760
    },
    {
      "epoch": 0.42570281124497994,
      "grad_norm": 0.3913135230541229,
      "learning_rate": 2e-05,
      "loss": 0.678,
      "step": 4770
    },
    {
      "epoch": 0.42659526996876396,
      "grad_norm": 0.3443652391433716,
      "learning_rate": 2e-05,
      "loss": 0.6725,
      "step": 4780
    },
    {
      "epoch": 0.427487728692548,
      "grad_norm": 0.3562873899936676,
      "learning_rate": 2e-05,
      "loss": 0.6989,
      "step": 4790
    },
    {
      "epoch": 0.428380187416332,
      "grad_norm": 0.5659394264221191,
      "learning_rate": 2e-05,
      "loss": 0.6778,
      "step": 4800
    },
    {
      "epoch": 0.429272646140116,
      "grad_norm": 0.350859135389328,
      "learning_rate": 2e-05,
      "loss": 0.6428,
      "step": 4810
    },
    {
      "epoch": 0.4301651048639,
      "grad_norm": 0.4072927236557007,
      "learning_rate": 2e-05,
      "loss": 0.6619,
      "step": 4820
    },
    {
      "epoch": 0.43105756358768405,
      "grad_norm": 0.39504697918891907,
      "learning_rate": 2e-05,
      "loss": 0.6648,
      "step": 4830
    },
    {
      "epoch": 0.4319500223114681,
      "grad_norm": 0.3588836193084717,
      "learning_rate": 2e-05,
      "loss": 0.6775,
      "step": 4840
    },
    {
      "epoch": 0.43284248103525214,
      "grad_norm": 0.34355223178863525,
      "learning_rate": 2e-05,
      "loss": 0.6492,
      "step": 4850
    },
    {
      "epoch": 0.43373493975903615,
      "grad_norm": 0.3690549433231354,
      "learning_rate": 2e-05,
      "loss": 0.7112,
      "step": 4860
    },
    {
      "epoch": 0.43462739848282017,
      "grad_norm": 0.40262696146965027,
      "learning_rate": 2e-05,
      "loss": 0.6977,
      "step": 4870
    },
    {
      "epoch": 0.4355198572066042,
      "grad_norm": 0.3697252571582794,
      "learning_rate": 2e-05,
      "loss": 0.6673,
      "step": 4880
    },
    {
      "epoch": 0.4364123159303882,
      "grad_norm": 0.3683331310749054,
      "learning_rate": 2e-05,
      "loss": 0.6588,
      "step": 4890
    },
    {
      "epoch": 0.4373047746541722,
      "grad_norm": 0.35332024097442627,
      "learning_rate": 2e-05,
      "loss": 0.6573,
      "step": 4900
    },
    {
      "epoch": 0.4381972333779563,
      "grad_norm": 0.33469337224960327,
      "learning_rate": 2e-05,
      "loss": 0.6429,
      "step": 4910
    },
    {
      "epoch": 0.4390896921017403,
      "grad_norm": 0.34541958570480347,
      "learning_rate": 2e-05,
      "loss": 0.6413,
      "step": 4920
    },
    {
      "epoch": 0.43998215082552433,
      "grad_norm": 0.39203688502311707,
      "learning_rate": 2e-05,
      "loss": 0.6642,
      "step": 4930
    },
    {
      "epoch": 0.44087460954930835,
      "grad_norm": 0.3529221713542938,
      "learning_rate": 2e-05,
      "loss": 0.6416,
      "step": 4940
    },
    {
      "epoch": 0.44176706827309237,
      "grad_norm": 0.35118046402931213,
      "learning_rate": 2e-05,
      "loss": 0.6432,
      "step": 4950
    },
    {
      "epoch": 0.4426595269968764,
      "grad_norm": 0.3619842231273651,
      "learning_rate": 2e-05,
      "loss": 0.6691,
      "step": 4960
    },
    {
      "epoch": 0.4435519857206604,
      "grad_norm": 0.3873862028121948,
      "learning_rate": 2e-05,
      "loss": 0.6572,
      "step": 4970
    },
    {
      "epoch": 0.4444444444444444,
      "grad_norm": 0.37688902020454407,
      "learning_rate": 2e-05,
      "loss": 0.6723,
      "step": 4980
    },
    {
      "epoch": 0.4453369031682285,
      "grad_norm": 0.353566437959671,
      "learning_rate": 2e-05,
      "loss": 0.6676,
      "step": 4990
    },
    {
      "epoch": 0.4462293618920125,
      "grad_norm": 0.35387760400772095,
      "learning_rate": 2e-05,
      "loss": 0.6553,
      "step": 5000
    },
    {
      "epoch": 0.44712182061579653,
      "grad_norm": 0.41578125953674316,
      "learning_rate": 2e-05,
      "loss": 0.679,
      "step": 5010
    },
    {
      "epoch": 0.44801427933958055,
      "grad_norm": 0.389415979385376,
      "learning_rate": 2e-05,
      "loss": 0.6602,
      "step": 5020
    },
    {
      "epoch": 0.44890673806336456,
      "grad_norm": 0.40956026315689087,
      "learning_rate": 2e-05,
      "loss": 0.6733,
      "step": 5030
    },
    {
      "epoch": 0.4497991967871486,
      "grad_norm": 0.38854891061782837,
      "learning_rate": 2e-05,
      "loss": 0.6591,
      "step": 5040
    },
    {
      "epoch": 0.4506916555109326,
      "grad_norm": 0.35887807607650757,
      "learning_rate": 2e-05,
      "loss": 0.6565,
      "step": 5050
    },
    {
      "epoch": 0.45158411423471667,
      "grad_norm": 0.3465573489665985,
      "learning_rate": 2e-05,
      "loss": 0.695,
      "step": 5060
    },
    {
      "epoch": 0.4524765729585007,
      "grad_norm": 0.37553462386131287,
      "learning_rate": 2e-05,
      "loss": 0.6736,
      "step": 5070
    },
    {
      "epoch": 0.4533690316822847,
      "grad_norm": 0.38836538791656494,
      "learning_rate": 2e-05,
      "loss": 0.653,
      "step": 5080
    },
    {
      "epoch": 0.4542614904060687,
      "grad_norm": 0.7184606194496155,
      "learning_rate": 1.9999810479633703e-05,
      "loss": 0.6423,
      "step": 5090
    },
    {
      "epoch": 0.45515394912985274,
      "grad_norm": 0.39216458797454834,
      "learning_rate": 1.997707677020786e-05,
      "loss": 0.6697,
      "step": 5100
    },
    {
      "epoch": 0.45604640785363676,
      "grad_norm": 0.3205508291721344,
      "learning_rate": 1.9916538196124637e-05,
      "loss": 0.6656,
      "step": 5110
    },
    {
      "epoch": 0.4569388665774208,
      "grad_norm": 0.3439967632293701,
      "learning_rate": 1.9818425303863402e-05,
      "loss": 0.6725,
      "step": 5120
    },
    {
      "epoch": 0.4578313253012048,
      "grad_norm": 0.35886335372924805,
      "learning_rate": 1.9683111732582887e-05,
      "loss": 0.658,
      "step": 5130
    },
    {
      "epoch": 0.45872378402498887,
      "grad_norm": 0.34354665875434875,
      "learning_rate": 1.951111279120706e-05,
      "loss": 0.6601,
      "step": 5140
    },
    {
      "epoch": 0.4596162427487729,
      "grad_norm": 0.36273109912872314,
      "learning_rate": 1.9303083495996028e-05,
      "loss": 0.7008,
      "step": 5150
    },
    {
      "epoch": 0.4605087014725569,
      "grad_norm": 0.4169177711009979,
      "learning_rate": 1.9059816076075482e-05,
      "loss": 0.6676,
      "step": 5160
    },
    {
      "epoch": 0.4614011601963409,
      "grad_norm": 0.3571546673774719,
      "learning_rate": 1.8782236956424142e-05,
      "loss": 0.6293,
      "step": 5170
    },
    {
      "epoch": 0.46229361892012494,
      "grad_norm": 0.4092195928096771,
      "learning_rate": 1.8471403229808897e-05,
      "loss": 0.6755,
      "step": 5180
    },
    {
      "epoch": 0.46318607764390896,
      "grad_norm": 0.3595106303691864,
      "learning_rate": 1.8128498631103333e-05,
      "loss": 0.6831,
      "step": 5190
    },
    {
      "epoch": 0.464078536367693,
      "grad_norm": 0.3474644720554352,
      "learning_rate": 1.7754829029320507e-05,
      "loss": 0.6655,
      "step": 5200
    },
    {
      "epoch": 0.46497099509147705,
      "grad_norm": 0.4022049307823181,
      "learning_rate": 1.7351817454527473e-05,
      "loss": 0.6788,
      "step": 5210
    },
    {
      "epoch": 0.46586345381526106,
      "grad_norm": 0.4258089065551758,
      "learning_rate": 1.6920998678580345e-05,
      "loss": 0.6453,
      "step": 5220
    },
    {
      "epoch": 0.4667559125390451,
      "grad_norm": 0.3806667625904083,
      "learning_rate": 1.6464013370317854e-05,
      "loss": 0.6607,
      "step": 5230
    },
    {
      "epoch": 0.4676483712628291,
      "grad_norm": 0.3446228802204132,
      "learning_rate": 1.5982601847471915e-05,
      "loss": 0.6651,
      "step": 5240
    },
    {
      "epoch": 0.4685408299866131,
      "grad_norm": 0.36848434805870056,
      "learning_rate": 1.5478597449089553e-05,
      "loss": 0.6633,
      "step": 5250
    },
    {
      "epoch": 0.46943328871039713,
      "grad_norm": 0.3431642949581146,
      "learning_rate": 1.4953919553705725e-05,
      "loss": 0.6598,
      "step": 5260
    },
    {
      "epoch": 0.47032574743418115,
      "grad_norm": 0.3706854283809662,
      "learning_rate": 1.441056626985559e-05,
      "loss": 0.682,
      "step": 5270
    },
    {
      "epoch": 0.47121820615796517,
      "grad_norm": 0.4674109220504761,
      "learning_rate": 1.3850606826762668e-05,
      "loss": 0.6803,
      "step": 5280
    },
    {
      "epoch": 0.47211066488174924,
      "grad_norm": 0.366987019777298,
      "learning_rate": 1.3276173694181128e-05,
      "loss": 0.6437,
      "step": 5290
    },
    {
      "epoch": 0.47300312360553326,
      "grad_norm": 0.3717084228992462,
      "learning_rate": 1.2689454461401822e-05,
      "loss": 0.6591,
      "step": 5300
    },
    {
      "epoch": 0.4738955823293173,
      "grad_norm": 0.36789411306381226,
      "learning_rate": 1.2092683506348939e-05,
      "loss": 0.6955,
      "step": 5310
    },
    {
      "epoch": 0.4747880410531013,
      "grad_norm": 0.3508588671684265,
      "learning_rate": 1.14881334864935e-05,
      "loss": 0.6521,
      "step": 5320
    },
    {
      "epoch": 0.4756804997768853,
      "grad_norm": 0.3830310106277466,
      "learning_rate": 1.0878106683988481e-05,
      "loss": 0.6584,
      "step": 5330
    },
    {
      "epoch": 0.47657295850066933,
      "grad_norm": 0.39885595440864563,
      "learning_rate": 1.026492623798551e-05,
      "loss": 0.6615,
      "step": 5340
    },
    {
      "epoch": 0.47746541722445335,
      "grad_norm": 0.35537248849868774,
      "learning_rate": 9.650927297522743e-06,
      "loss": 0.6811,
      "step": 5350
    },
    {
      "epoch": 0.4783578759482374,
      "grad_norm": 0.3321537673473358,
      "learning_rate": 9.038448128676012e-06,
      "loss": 0.6245,
      "step": 5360
    },
    {
      "epoch": 0.47925033467202144,
      "grad_norm": 0.3931525945663452,
      "learning_rate": 8.42982120983949e-06,
      "loss": 0.6772,
      "step": 5370
    },
    {
      "epoch": 0.48014279339580546,
      "grad_norm": 0.36104288697242737,
      "learning_rate": 7.827364349047303e-06,
      "loss": 0.6505,
      "step": 5380
    },
    {
      "epoch": 0.4810352521195895,
      "grad_norm": 0.34397444128990173,
      "learning_rate": 7.233371857163706e-06,
      "loss": 0.6357,
      "step": 5390
    },
    {
      "epoch": 0.4819277108433735,
      "grad_norm": 0.37029504776000977,
      "learning_rate": 6.650105810556492e-06,
      "loss": 0.6555,
      "step": 5400
    },
    {
      "epoch": 0.4828201695671575,
      "grad_norm": 0.345136821269989,
      "learning_rate": 6.079787436527681e-06,
      "loss": 0.6405,
      "step": 5410
    },
    {
      "epoch": 0.4837126282909415,
      "grad_norm": 0.3560040593147278,
      "learning_rate": 5.524588654307961e-06,
      "loss": 0.6489,
      "step": 5420
    },
    {
      "epoch": 0.48460508701472554,
      "grad_norm": 0.32775047421455383,
      "learning_rate": 4.986623803828985e-06,
      "loss": 0.6758,
      "step": 5430
    },
    {
      "epoch": 0.4854975457385096,
      "grad_norm": 0.3954639434814453,
      "learning_rate": 4.467941593772423e-06,
      "loss": 0.6532,
      "step": 5440
    },
    {
      "epoch": 0.48639000446229363,
      "grad_norm": 0.4013110399246216,
      "learning_rate": 3.9705172995597606e-06,
      "loss": 0.6778,
      "step": 5450
    },
    {
      "epoch": 0.48728246318607765,
      "grad_norm": 0.3553372919559479,
      "learning_rate": 3.4962452409948346e-06,
      "loss": 0.644,
      "step": 5460
    },
    {
      "epoch": 0.48817492190986167,
      "grad_norm": 0.3670588433742523,
      "learning_rate": 3.046931568206211e-06,
      "loss": 0.6542,
      "step": 5470
    },
    {
      "epoch": 0.4890673806336457,
      "grad_norm": 0.3435574471950531,
      "learning_rate": 2.624287383362404e-06,
      "loss": 0.6439,
      "step": 5480
    },
    {
      "epoch": 0.4899598393574297,
      "grad_norm": 0.33368292450904846,
      "learning_rate": 2.2299222243541845e-06,
      "loss": 0.6476,
      "step": 5490
    },
    {
      "epoch": 0.4908522980812137,
      "grad_norm": 0.33577534556388855,
      "learning_rate": 1.8653379352598246e-06,
      "loss": 0.6352,
      "step": 5500
    },
    {
      "epoch": 0.4917447568049978,
      "grad_norm": 0.3742958903312683,
      "learning_rate": 1.5319229469361075e-06,
      "loss": 0.6678,
      "step": 5510
    },
    {
      "epoch": 0.4926372155287818,
      "grad_norm": 0.3951159715652466,
      "learning_rate": 1.2309469895160917e-06,
      "loss": 0.6721,
      "step": 5520
    },
    {
      "epoch": 0.49352967425256583,
      "grad_norm": 0.31695225834846497,
      "learning_rate": 9.635562569497974e-07,
      "loss": 0.6362,
      "step": 5530
    },
    {
      "epoch": 0.49442213297634985,
      "grad_norm": 0.3108005225658417,
      "learning_rate": 7.307690420024867e-07,
      "loss": 0.6771,
      "step": 5540
    },
    {
      "epoch": 0.49531459170013387,
      "grad_norm": 0.37071311473846436,
      "learning_rate": 5.334718583335942e-07,
      "loss": 0.6253,
      "step": 5550
    },
    {
      "epoch": 0.4962070504239179,
      "grad_norm": 0.34029334783554077,
      "learning_rate": 3.7241606442442763e-07,
      "loss": 0.6142,
      "step": 5560
    },
    {
      "epoch": 0.4970995091477019,
      "grad_norm": 0.33255934715270996,
      "learning_rate": 2.482150022115879e-07,
      "loss": 0.6465,
      "step": 5570
    },
    {
      "epoch": 0.4979919678714859,
      "grad_norm": 0.3747519552707672,
      "learning_rate": 1.6134166132293174e-07,
      "loss": 0.6516,
      "step": 5580
    },
    {
      "epoch": 0.49888442659527,
      "grad_norm": 0.3537963926792145,
      "learning_rate": 1.1212687781127357e-07,
      "loss": 0.6359,
      "step": 5590
    },
    {
      "epoch": 0.4995091477019188,
      "step": 5597,
      "total_flos": 2.2737574718397743e+19,
      "train_loss": 0.022180273806769783,
      "train_runtime": 1803.6018,
      "train_samples_per_second": 37.239,
      "train_steps_per_second": 3.103
    }
  ],
  "logging_steps": 10,
  "max_steps": 5597,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 159,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 2.2737574718397743e+19,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}