{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 5000,
  "global_step": 75251,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.3288948246191055e-05,
      "grad_norm": 197.0,
      "learning_rate": 0.0,
      "loss": 3.328,
      "num_input_tokens_seen": 453632,
      "step": 1,
      "train_runtime": 20.3737,
      "train_tokens_per_second": 22265.577
    },
    {
      "epoch": 0.00013288948246191056,
      "grad_norm": 70.5,
      "learning_rate": 7.971656333038087e-08,
      "loss": 2.1166,
      "num_input_tokens_seen": 2009088,
      "step": 10,
      "train_runtime": 44.4843,
      "train_tokens_per_second": 45163.957
    },
    {
      "epoch": 0.0002657789649238211,
      "grad_norm": 45.0,
      "learning_rate": 1.682905225863596e-07,
      "loss": 1.6382,
      "num_input_tokens_seen": 2853888,
      "step": 20,
      "train_runtime": 57.039,
      "train_tokens_per_second": 50033.945
    },
    {
      "epoch": 0.0003986684473857317,
      "grad_norm": 43.25,
      "learning_rate": 2.5686448184233836e-07,
      "loss": 1.5045,
      "num_input_tokens_seen": 3339264,
      "step": 30,
      "train_runtime": 63.7964,
      "train_tokens_per_second": 52342.529
    },
    {
      "epoch": 0.0005315579298476422,
      "grad_norm": 74.5,
      "learning_rate": 3.4543844109831714e-07,
      "loss": 1.7177,
      "num_input_tokens_seen": 3567616,
      "step": 40,
      "train_runtime": 67.261,
      "train_tokens_per_second": 53041.402
    },
    {
      "epoch": 0.0006644474123095528,
      "grad_norm": 39.75,
      "learning_rate": 4.3401240035429586e-07,
      "loss": 1.4875,
      "num_input_tokens_seen": 3682816,
      "step": 50,
      "train_runtime": 69.7167,
      "train_tokens_per_second": 52825.421
    },
    {
      "epoch": 0.0007973368947714634,
      "grad_norm": 57.25,
      "learning_rate": 5.225863596102746e-07,
      "loss": 2.2107,
      "num_input_tokens_seen": 5847040,
      "step": 60,
      "train_runtime": 104.3083,
      "train_tokens_per_second": 56055.37
    },
    {
      "epoch": 0.0009302263772333739,
      "grad_norm": 28.0,
      "learning_rate": 6.111603188662533e-07,
      "loss": 1.6215,
      "num_input_tokens_seen": 6743040,
      "step": 70,
      "train_runtime": 117.2388,
      "train_tokens_per_second": 57515.421
    },
    {
      "epoch": 0.0010631158596952845,
      "grad_norm": 61.0,
      "learning_rate": 6.997342781222321e-07,
      "loss": 1.5465,
      "num_input_tokens_seen": 7211008,
      "step": 80,
      "train_runtime": 123.914,
      "train_tokens_per_second": 58193.667
    },
    {
      "epoch": 0.001196005342157195,
      "grad_norm": 78.0,
      "learning_rate": 7.883082373782109e-07,
      "loss": 1.6567,
      "num_input_tokens_seen": 7443456,
      "step": 90,
      "train_runtime": 127.4102,
      "train_tokens_per_second": 58421.181
    },
    {
      "epoch": 0.0013288948246191056,
      "grad_norm": 86.0,
      "learning_rate": 8.768821966341897e-07,
      "loss": 1.4873,
      "num_input_tokens_seen": 7559168,
      "step": 100,
      "train_runtime": 129.9181,
      "train_tokens_per_second": 58184.115
    },
    {
      "epoch": 0.001461784307081016,
      "grad_norm": 70.5,
      "learning_rate": 9.654561558901683e-07,
      "loss": 2.1792,
      "num_input_tokens_seen": 9664256,
      "step": 110,
      "train_runtime": 163.6235,
      "train_tokens_per_second": 59063.981
    },
    {
      "epoch": 0.0015946737895429267,
      "grad_norm": 57.75,
      "learning_rate": 1.0540301151461471e-06,
      "loss": 1.5242,
      "num_input_tokens_seen": 10518272,
      "step": 120,
      "train_runtime": 176.1896,
      "train_tokens_per_second": 59698.592
    },
    {
      "epoch": 0.0017275632720048372,
      "grad_norm": 58.75,
      "learning_rate": 1.142604074402126e-06,
      "loss": 1.5744,
      "num_input_tokens_seen": 11011840,
      "step": 130,
      "train_runtime": 182.9702,
      "train_tokens_per_second": 60183.777
    },
    {
      "epoch": 0.0018604527544667478,
      "grad_norm": 648.0,
      "learning_rate": 1.2311780336581046e-06,
      "loss": 1.6339,
      "num_input_tokens_seen": 11252224,
      "step": 140,
      "train_runtime": 186.5708,
      "train_tokens_per_second": 60310.76
    },
    {
      "epoch": 0.0019933422369286585,
      "grad_norm": 46.0,
      "learning_rate": 1.3197519929140834e-06,
      "loss": 1.4914,
      "num_input_tokens_seen": 11365120,
      "step": 150,
      "train_runtime": 189.1221,
      "train_tokens_per_second": 60094.088
    },
    {
      "epoch": 0.002126231719390569,
      "grad_norm": 55.25,
      "learning_rate": 1.408325952170062e-06,
      "loss": 2.1365,
      "num_input_tokens_seen": 13447680,
      "step": 160,
      "train_runtime": 222.7001,
      "train_tokens_per_second": 60384.707
    },
    {
      "epoch": 0.0022591212018524794,
      "grad_norm": 48.5,
      "learning_rate": 1.4968999114260406e-06,
      "loss": 1.5759,
      "num_input_tokens_seen": 14306304,
      "step": 170,
      "train_runtime": 235.479,
      "train_tokens_per_second": 60754.064
    },
    {
      "epoch": 0.00239201068431439,
      "grad_norm": 70.5,
      "learning_rate": 1.5854738706820197e-06,
      "loss": 1.4918,
      "num_input_tokens_seen": 14775808,
      "step": 180,
      "train_runtime": 242.2312,
      "train_tokens_per_second": 60998.774
    },
    {
      "epoch": 0.0025249001667763003,
      "grad_norm": 78.0,
      "learning_rate": 1.6740478299379983e-06,
      "loss": 1.5815,
      "num_input_tokens_seen": 15016448,
      "step": 190,
      "train_runtime": 245.7813,
      "train_tokens_per_second": 61096.785
    },
    {
      "epoch": 0.002657789649238211,
      "grad_norm": 41.25,
      "learning_rate": 1.7626217891939771e-06,
      "loss": 1.4507,
      "num_input_tokens_seen": 15131648,
      "step": 200,
      "train_runtime": 248.18,
      "train_tokens_per_second": 60970.468
    },
    {
      "epoch": 0.0027906791317001216,
      "grad_norm": 80.5,
      "learning_rate": 1.851195748449956e-06,
      "loss": 2.0751,
      "num_input_tokens_seen": 17174528,
      "step": 210,
      "train_runtime": 282.3621,
      "train_tokens_per_second": 60824.476
    },
    {
      "epoch": 0.002923568614162032,
      "grad_norm": 17.0,
      "learning_rate": 1.9397697077059346e-06,
      "loss": 1.5274,
      "num_input_tokens_seen": 18056192,
      "step": 220,
      "train_runtime": 295.2144,
      "train_tokens_per_second": 61162.988
    },
    {
      "epoch": 0.0030564580966239425,
      "grad_norm": 50.75,
      "learning_rate": 2.0283436669619134e-06,
      "loss": 1.4611,
      "num_input_tokens_seen": 18506240,
      "step": 230,
      "train_runtime": 301.8725,
      "train_tokens_per_second": 61304.832
    },
    {
      "epoch": 0.0031893475790858534,
      "grad_norm": 29.625,
      "learning_rate": 2.116917626217892e-06,
      "loss": 1.5354,
      "num_input_tokens_seen": 18736128,
      "step": 240,
      "train_runtime": 305.399,
      "train_tokens_per_second": 61349.671
    },
    {
      "epoch": 0.003322237061547764,
      "grad_norm": 41.25,
      "learning_rate": 2.205491585473871e-06,
      "loss": 1.4143,
      "num_input_tokens_seen": 18848256,
      "step": 250,
      "train_runtime": 307.9113,
      "train_tokens_per_second": 61213.266
    },
    {
      "epoch": 0.0034551265440096743,
      "grad_norm": 51.0,
      "learning_rate": 2.2940655447298495e-06,
      "loss": 1.975,
      "num_input_tokens_seen": 21017856,
      "step": 260,
      "train_runtime": 342.6869,
      "train_tokens_per_second": 61332.543
    },
    {
      "epoch": 0.0035880160264715848,
      "grad_norm": 44.0,
      "learning_rate": 2.3826395039858283e-06,
      "loss": 1.4139,
      "num_input_tokens_seen": 21857792,
      "step": 270,
      "train_runtime": 355.1559,
      "train_tokens_per_second": 61544.21
    },
    {
      "epoch": 0.0037209055089334957,
      "grad_norm": 32.25,
      "learning_rate": 2.471213463241807e-06,
      "loss": 1.4109,
      "num_input_tokens_seen": 22306816,
      "step": 280,
      "train_runtime": 361.6622,
      "train_tokens_per_second": 61678.586
    },
    {
      "epoch": 0.003853794991395406,
      "grad_norm": 56.75,
      "learning_rate": 2.559787422497786e-06,
      "loss": 1.4777,
      "num_input_tokens_seen": 22536192,
      "step": 290,
      "train_runtime": 365.181,
      "train_tokens_per_second": 61712.384
    },
    {
      "epoch": 0.003986684473857317,
      "grad_norm": 42.5,
      "learning_rate": 2.6483613817537644e-06,
      "loss": 1.4335,
      "num_input_tokens_seen": 22654464,
      "step": 300,
      "train_runtime": 367.5574,
      "train_tokens_per_second": 61635.168
    },
    {
      "epoch": 0.0041195739563192274,
      "grad_norm": 54.75,
      "learning_rate": 2.736935341009743e-06,
      "loss": 1.7973,
      "num_input_tokens_seen": 24707840,
      "step": 310,
      "train_runtime": 400.4527,
      "train_tokens_per_second": 61699.768
    },
    {
      "epoch": 0.004252463438781138,
      "grad_norm": 26.5,
      "learning_rate": 2.825509300265722e-06,
      "loss": 1.4278,
      "num_input_tokens_seen": 25560064,
      "step": 320,
      "train_runtime": 413.3562,
      "train_tokens_per_second": 61835.448
    },
    {
      "epoch": 0.004385352921243048,
      "grad_norm": 31.875,
      "learning_rate": 2.9140832595217013e-06,
      "loss": 1.3771,
      "num_input_tokens_seen": 26013440,
      "step": 330,
      "train_runtime": 420.1048,
      "train_tokens_per_second": 61921.314
    },
    {
      "epoch": 0.004518242403704959,
      "grad_norm": 34.25,
      "learning_rate": 3.0026572187776797e-06,
      "loss": 1.4249,
      "num_input_tokens_seen": 26241024,
      "step": 340,
      "train_runtime": 423.6942,
      "train_tokens_per_second": 61933.877
    },
    {
      "epoch": 0.004651131886166869,
      "grad_norm": 32.5,
      "learning_rate": 3.0912311780336585e-06,
      "loss": 1.3747,
      "num_input_tokens_seen": 26355200,
      "step": 350,
      "train_runtime": 426.1405,
      "train_tokens_per_second": 61846.27
    },
    {
      "epoch": 0.00478402136862878,
      "grad_norm": 32.25,
      "learning_rate": 3.1798051372896373e-06,
      "loss": 1.7941,
      "num_input_tokens_seen": 28461824,
      "step": 360,
      "train_runtime": 460.7859,
      "train_tokens_per_second": 61767.994
    },
    {
      "epoch": 0.00491691085109069,
      "grad_norm": 38.5,
      "learning_rate": 3.2683790965456157e-06,
      "loss": 1.4705,
      "num_input_tokens_seen": 29303552,
      "step": 370,
      "train_runtime": 473.6276,
      "train_tokens_per_second": 61870.453
    },
    {
      "epoch": 0.005049800333552601,
      "grad_norm": 20.875,
      "learning_rate": 3.3569530558015946e-06,
      "loss": 1.4552,
      "num_input_tokens_seen": 29778432,
      "step": 380,
      "train_runtime": 480.4849,
      "train_tokens_per_second": 61975.787
    },
    {
      "epoch": 0.005182689816014512,
      "grad_norm": 31.0,
      "learning_rate": 3.445527015057573e-06,
      "loss": 1.4088,
      "num_input_tokens_seen": 30009600,
      "step": 390,
      "train_runtime": 484.0701,
      "train_tokens_per_second": 61994.333
    },
    {
      "epoch": 0.005315579298476422,
      "grad_norm": 21.0,
      "learning_rate": 3.5341009743135522e-06,
      "loss": 1.3226,
      "num_input_tokens_seen": 30126848,
      "step": 400,
      "train_runtime": 486.5477,
      "train_tokens_per_second": 61919.623
    },
    {
      "epoch": 0.005448468780938333,
      "grad_norm": 23.5,
      "learning_rate": 3.622674933569531e-06,
      "loss": 1.8085,
      "num_input_tokens_seen": 32240384,
      "step": 410,
      "train_runtime": 519.5025,
      "train_tokens_per_second": 62060.111
    },
    {
      "epoch": 0.005581358263400243,
      "grad_norm": 11.375,
      "learning_rate": 3.71124889282551e-06,
      "loss": 1.4839,
      "num_input_tokens_seen": 33106688,
      "step": 420,
      "train_runtime": 532.2777,
      "train_tokens_per_second": 62198.152
    },
    {
      "epoch": 0.005714247745862154,
      "grad_norm": 14.5,
      "learning_rate": 3.7998228520814883e-06,
      "loss": 1.4139,
      "num_input_tokens_seen": 33567232,
      "step": 430,
      "train_runtime": 539.1502,
      "train_tokens_per_second": 62259.516
    },
    {
      "epoch": 0.005847137228324064,
      "grad_norm": 18.5,
      "learning_rate": 3.888396811337467e-06,
      "loss": 1.4351,
      "num_input_tokens_seen": 33799680,
      "step": 440,
      "train_runtime": 543.0852,
      "train_tokens_per_second": 62236.426
    },
    {
      "epoch": 0.005980026710785975,
      "grad_norm": 31.25,
      "learning_rate": 3.9769707705934455e-06,
      "loss": 1.3633,
      "num_input_tokens_seen": 33917184,
      "step": 450,
      "train_runtime": 545.5508,
      "train_tokens_per_second": 62170.531
    },
    {
      "epoch": 0.006112916193247885,
      "grad_norm": 18.0,
      "learning_rate": 4.065544729849424e-06,
      "loss": 1.7128,
      "num_input_tokens_seen": 36020224,
      "step": 460,
      "train_runtime": 580.0334,
      "train_tokens_per_second": 62100.257
    },
    {
      "epoch": 0.0062458056757097955,
      "grad_norm": 51.75,
      "learning_rate": 4.154118689105404e-06,
      "loss": 1.4603,
      "num_input_tokens_seen": 36842752,
      "step": 470,
      "train_runtime": 592.8204,
      "train_tokens_per_second": 62148.254
    },
    {
      "epoch": 0.006378695158171707,
      "grad_norm": 67.5,
      "learning_rate": 4.242692648361382e-06,
      "loss": 1.4163,
      "num_input_tokens_seen": 37329920,
      "step": 480,
      "train_runtime": 599.7066,
      "train_tokens_per_second": 62246.974
    },
    {
      "epoch": 0.006511584640633617,
      "grad_norm": 14.625,
      "learning_rate": 4.331266607617361e-06,
      "loss": 1.3777,
      "num_input_tokens_seen": 37557760,
      "step": 490,
      "train_runtime": 603.2103,
      "train_tokens_per_second": 62263.127
    },
    {
      "epoch": 0.006644474123095528,
      "grad_norm": 15.0625,
      "learning_rate": 4.41984056687334e-06,
      "loss": 1.3613,
      "num_input_tokens_seen": 37674240,
      "step": 500,
      "train_runtime": 606.1433,
      "train_tokens_per_second": 62154.016
    },
    {
      "epoch": 0.006777363605557438,
      "grad_norm": 39.5,
      "learning_rate": 4.5084145261293185e-06,
      "loss": 1.6565,
      "num_input_tokens_seen": 39723776,
      "step": 510,
      "train_runtime": 637.3344,
      "train_tokens_per_second": 62327.995
    },
    {
      "epoch": 0.006910253088019349,
      "grad_norm": 43.0,
      "learning_rate": 4.596988485385297e-06,
      "loss": 1.4623,
      "num_input_tokens_seen": 40543232,
      "step": 520,
      "train_runtime": 649.6162,
      "train_tokens_per_second": 62411.05
    },
    {
      "epoch": 0.007043142570481259,
      "grad_norm": 33.25,
      "learning_rate": 4.685562444641275e-06,
      "loss": 1.3996,
      "num_input_tokens_seen": 41011456,
      "step": 530,
      "train_runtime": 656.2465,
      "train_tokens_per_second": 62493.976
    },
    {
      "epoch": 0.0071760320529431695,
      "grad_norm": 29.25,
      "learning_rate": 4.774136403897255e-06,
      "loss": 1.3533,
      "num_input_tokens_seen": 41236992,
      "step": 540,
      "train_runtime": 659.6891,
      "train_tokens_per_second": 62509.739
    },
    {
      "epoch": 0.00730892153540508,
      "grad_norm": 22.75,
      "learning_rate": 4.862710363153234e-06,
      "loss": 1.3511,
      "num_input_tokens_seen": 41350144,
      "step": 550,
      "train_runtime": 662.6404,
      "train_tokens_per_second": 62402.087
    },
    {
      "epoch": 0.007441811017866991,
      "grad_norm": 22.125,
      "learning_rate": 4.951284322409212e-06,
      "loss": 1.6083,
      "num_input_tokens_seen": 43343104,
      "step": 560,
      "train_runtime": 694.3373,
      "train_tokens_per_second": 62423.7
    },
    {
      "epoch": 0.007574700500328902,
      "grad_norm": 39.75,
      "learning_rate": 5.039858281665191e-06,
      "loss": 1.4212,
      "num_input_tokens_seen": 44132864,
      "step": 570,
      "train_runtime": 706.5236,
      "train_tokens_per_second": 62464.809
    },
    {
      "epoch": 0.007707589982790812,
      "grad_norm": 16.875,
      "learning_rate": 5.1284322409211695e-06,
      "loss": 1.4107,
      "num_input_tokens_seen": 44605440,
      "step": 580,
      "train_runtime": 713.1559,
      "train_tokens_per_second": 62546.551
    },
    {
      "epoch": 0.007840479465252723,
      "grad_norm": 32.0,
      "learning_rate": 5.217006200177148e-06,
      "loss": 1.3539,
      "num_input_tokens_seen": 44835328,
      "step": 590,
      "train_runtime": 717.116,
      "train_tokens_per_second": 62521.722
    },
    {
      "epoch": 0.007973368947714634,
      "grad_norm": 30.375,
      "learning_rate": 5.305580159433127e-06,
      "loss": 1.3582,
      "num_input_tokens_seen": 44949248,
      "step": 600,
      "train_runtime": 719.5594,
      "train_tokens_per_second": 62467.735
    },
    {
      "epoch": 0.008106258430176544,
      "grad_norm": 36.0,
      "learning_rate": 5.394154118689106e-06,
      "loss": 1.6882,
      "num_input_tokens_seen": 46951424,
      "step": 610,
      "train_runtime": 752.5434,
      "train_tokens_per_second": 62390.321
    },
    {
      "epoch": 0.008239147912638455,
      "grad_norm": 30.125,
      "learning_rate": 5.482728077945084e-06,
      "loss": 1.4047,
      "num_input_tokens_seen": 47793664,
      "step": 620,
      "train_runtime": 765.2441,
      "train_tokens_per_second": 62455.453
    },
    {
      "epoch": 0.008372037395100364,
      "grad_norm": 20.625,
      "learning_rate": 5.571302037201063e-06,
      "loss": 1.4241,
      "num_input_tokens_seen": 48285184,
      "step": 630,
      "train_runtime": 772.0558,
      "train_tokens_per_second": 62541.052
    },
    {
      "epoch": 0.008504926877562276,
      "grad_norm": 23.875,
      "learning_rate": 5.659875996457042e-06,
      "loss": 1.3009,
      "num_input_tokens_seen": 48518400,
      "step": 640,
      "train_runtime": 775.6325,
      "train_tokens_per_second": 62553.339
    },
    {
      "epoch": 0.008637816360024185,
      "grad_norm": 29.0,
      "learning_rate": 5.748449955713021e-06,
      "loss": 1.3424,
      "num_input_tokens_seen": 48639744,
      "step": 650,
      "train_runtime": 778.0858,
      "train_tokens_per_second": 62512.059
    },
    {
      "epoch": 0.008770705842486097,
      "grad_norm": 36.75,
      "learning_rate": 5.837023914969e-06,
      "loss": 1.6388,
      "num_input_tokens_seen": 50702080,
      "step": 660,
      "train_runtime": 811.7421,
      "train_tokens_per_second": 62460.826
    },
    {
      "epoch": 0.008903595324948006,
      "grad_norm": 33.25,
      "learning_rate": 5.925597874224979e-06,
      "loss": 1.4774,
      "num_input_tokens_seen": 51555072,
      "step": 670,
      "train_runtime": 824.3387,
      "train_tokens_per_second": 62541.125
    },
    {
      "epoch": 0.009036484807409918,
      "grad_norm": 29.75,
      "learning_rate": 6.014171833480958e-06,
      "loss": 1.3967,
      "num_input_tokens_seen": 52034816,
      "step": 680,
      "train_runtime": 830.9605,
      "train_tokens_per_second": 62620.086
    },
    {
      "epoch": 0.009169374289871829,
      "grad_norm": 36.0,
      "learning_rate": 6.102745792736936e-06,
      "loss": 1.3693,
      "num_input_tokens_seen": 52264704,
      "step": 690,
      "train_runtime": 834.4275,
      "train_tokens_per_second": 62635.402
    },
    {
      "epoch": 0.009302263772333738,
      "grad_norm": 12.4375,
      "learning_rate": 6.1913197519929146e-06,
      "loss": 1.3067,
      "num_input_tokens_seen": 52380416,
      "step": 700,
      "train_runtime": 837.1609,
      "train_tokens_per_second": 62569.118
    },
    {
      "epoch": 0.00943515325479565,
      "grad_norm": 19.375,
      "learning_rate": 6.279893711248893e-06,
      "loss": 1.541,
      "num_input_tokens_seen": 54481920,
      "step": 710,
      "train_runtime": 870.5252,
      "train_tokens_per_second": 62585.115
    },
    {
      "epoch": 0.00956804273725756,
      "grad_norm": 20.375,
      "learning_rate": 6.368467670504872e-06,
      "loss": 1.4069,
      "num_input_tokens_seen": 55332096,
      "step": 720,
      "train_runtime": 883.2312,
      "train_tokens_per_second": 62647.352
    },
    {
      "epoch": 0.00970093221971947,
      "grad_norm": 83.0,
      "learning_rate": 6.457041629760851e-06,
      "loss": 1.3815,
      "num_input_tokens_seen": 55804928,
      "step": 730,
      "train_runtime": 889.8503,
      "train_tokens_per_second": 62712.713
    },
    {
      "epoch": 0.00983382170218138,
      "grad_norm": 13.25,
      "learning_rate": 6.545615589016829e-06,
      "loss": 1.3132,
      "num_input_tokens_seen": 56040448,
      "step": 740,
      "train_runtime": 893.3437,
      "train_tokens_per_second": 62731.113
    },
    {
      "epoch": 0.009966711184643292,
      "grad_norm": 39.0,
      "learning_rate": 6.634189548272808e-06,
      "loss": 1.2984,
      "num_input_tokens_seen": 56155136,
      "step": 750,
      "train_runtime": 895.87,
      "train_tokens_per_second": 62682.235
    },
    {
      "epoch": 0.010099600667105201,
      "grad_norm": 29.125,
      "learning_rate": 6.722763507528787e-06,
      "loss": 1.6157,
      "num_input_tokens_seen": 58154752,
      "step": 760,
      "train_runtime": 928.4596,
      "train_tokens_per_second": 62635.738
    },
    {
      "epoch": 0.010232490149567113,
      "grad_norm": 18.125,
      "learning_rate": 6.8113374667847655e-06,
      "loss": 1.4063,
      "num_input_tokens_seen": 59007744,
      "step": 770,
      "train_runtime": 940.9805,
      "train_tokens_per_second": 62708.785
    },
    {
      "epoch": 0.010365379632029024,
      "grad_norm": 19.375,
      "learning_rate": 6.899911426040744e-06,
      "loss": 1.371,
      "num_input_tokens_seen": 59465472,
      "step": 780,
      "train_runtime": 948.0863,
      "train_tokens_per_second": 62721.584
    },
    {
      "epoch": 0.010498269114490933,
      "grad_norm": 32.0,
      "learning_rate": 6.988485385296724e-06,
      "loss": 1.3112,
      "num_input_tokens_seen": 59692800,
      "step": 790,
      "train_runtime": 951.5276,
      "train_tokens_per_second": 62733.649
    },
    {
      "epoch": 0.010631158596952845,
      "grad_norm": 15.0,
      "learning_rate": 7.077059344552703e-06,
      "loss": 1.2912,
      "num_input_tokens_seen": 59807744,
      "step": 800,
      "train_runtime": 953.9882,
      "train_tokens_per_second": 62692.33
    },
    {
      "epoch": 0.010764048079414754,
      "grad_norm": 17.875,
      "learning_rate": 7.165633303808681e-06,
      "loss": 1.5775,
      "num_input_tokens_seen": 61835520,
      "step": 810,
      "train_runtime": 985.9748,
      "train_tokens_per_second": 62715.111
    },
    {
      "epoch": 0.010896937561876666,
      "grad_norm": 28.625,
      "learning_rate": 7.25420726306466e-06,
      "loss": 1.454,
      "num_input_tokens_seen": 62671616,
      "step": 820,
      "train_runtime": 998.6105,
      "train_tokens_per_second": 62758.818
    },
    {
      "epoch": 0.011029827044338575,
      "grad_norm": 17.125,
      "learning_rate": 7.3427812223206385e-06,
      "loss": 1.3952,
      "num_input_tokens_seen": 63117568,
      "step": 830,
      "train_runtime": 1005.4038,
      "train_tokens_per_second": 62778.328
    },
    {
      "epoch": 0.011162716526800487,
      "grad_norm": 20.125,
      "learning_rate": 7.431355181576617e-06,
      "loss": 1.3163,
      "num_input_tokens_seen": 63351296,
      "step": 840,
      "train_runtime": 1008.9225,
      "train_tokens_per_second": 62791.046
    },
    {
      "epoch": 0.011295606009262396,
      "grad_norm": 34.25,
      "learning_rate": 7.519929140832596e-06,
      "loss": 1.3023,
      "num_input_tokens_seen": 63468032,
      "step": 850,
      "train_runtime": 1011.3522,
      "train_tokens_per_second": 62755.616
    },
    {
      "epoch": 0.011428495491724307,
      "grad_norm": 19.25,
      "learning_rate": 7.608503100088574e-06,
      "loss": 1.5956,
      "num_input_tokens_seen": 65605120,
      "step": 860,
      "train_runtime": 1045.2674,
      "train_tokens_per_second": 62763.958
    },
    {
      "epoch": 0.011561384974186219,
      "grad_norm": 22.375,
      "learning_rate": 7.697077059344553e-06,
      "loss": 1.4019,
      "num_input_tokens_seen": 66430208,
      "step": 870,
      "train_runtime": 1057.9704,
      "train_tokens_per_second": 62790.233
    },
    {
      "epoch": 0.011694274456648128,
      "grad_norm": 12.0625,
      "learning_rate": 7.785651018600532e-06,
      "loss": 1.3324,
      "num_input_tokens_seen": 66872832,
      "step": 880,
      "train_runtime": 1064.5093,
      "train_tokens_per_second": 62820.336
    },
    {
      "epoch": 0.01182716393911004,
      "grad_norm": 11.3125,
      "learning_rate": 7.87422497785651e-06,
      "loss": 1.2835,
      "num_input_tokens_seen": 67100928,
      "step": 890,
      "train_runtime": 1067.9445,
      "train_tokens_per_second": 62831.85
    },
    {
      "epoch": 0.01196005342157195,
      "grad_norm": 37.0,
      "learning_rate": 7.96279893711249e-06,
      "loss": 1.2885,
      "num_input_tokens_seen": 67214592,
      "step": 900,
      "train_runtime": 1070.3256,
      "train_tokens_per_second": 62798.266
    },
    {
      "epoch": 0.01209294290403386,
      "grad_norm": 23.375,
      "learning_rate": 8.051372896368468e-06,
      "loss": 1.5176,
      "num_input_tokens_seen": 69209856,
      "step": 910,
      "train_runtime": 1102.4749,
      "train_tokens_per_second": 62776.807
    },
    {
      "epoch": 0.01222583238649577,
      "grad_norm": 44.5,
      "learning_rate": 8.139946855624447e-06,
      "loss": 1.4248,
      "num_input_tokens_seen": 70027008,
      "step": 920,
      "train_runtime": 1114.6419,
      "train_tokens_per_second": 62824.668
    },
    {
      "epoch": 0.012358721868957681,
      "grad_norm": 22.375,
      "learning_rate": 8.228520814880426e-06,
      "loss": 1.3696,
      "num_input_tokens_seen": 70464512,
      "step": 930,
      "train_runtime": 1121.0643,
      "train_tokens_per_second": 62855.015
    },
    {
      "epoch": 0.012491611351419591,
      "grad_norm": 15.0625,
      "learning_rate": 8.317094774136405e-06,
      "loss": 1.287,
      "num_input_tokens_seen": 70696192,
      "step": 940,
      "train_runtime": 1124.4965,
      "train_tokens_per_second": 62869.199
    },
    {
      "epoch": 0.012624500833881502,
      "grad_norm": 16.875,
      "learning_rate": 8.405668733392384e-06,
      "loss": 1.2639,
      "num_input_tokens_seen": 70810880,
      "step": 950,
      "train_runtime": 1127.3255,
      "train_tokens_per_second": 62813.16
    },
    {
      "epoch": 0.012757390316343414,
      "grad_norm": 30.875,
      "learning_rate": 8.494242692648362e-06,
      "loss": 1.5747,
      "num_input_tokens_seen": 72814848,
      "step": 960,
      "train_runtime": 1159.0328,
      "train_tokens_per_second": 62823.805
    },
    {
      "epoch": 0.012890279798805323,
      "grad_norm": 32.25,
      "learning_rate": 8.582816651904341e-06,
      "loss": 1.3953,
      "num_input_tokens_seen": 73638912,
      "step": 970,
      "train_runtime": 1171.3146,
      "train_tokens_per_second": 62868.6
    },
    {
      "epoch": 0.013023169281267235,
      "grad_norm": 24.625,
      "learning_rate": 8.67139061116032e-06,
      "loss": 1.4119,
      "num_input_tokens_seen": 74121984,
      "step": 980,
      "train_runtime": 1177.9096,
      "train_tokens_per_second": 62926.718
    },
    {
      "epoch": 0.013156058763729144,
      "grad_norm": 17.75,
      "learning_rate": 8.759964570416299e-06,
      "loss": 1.2604,
      "num_input_tokens_seen": 74349568,
      "step": 990,
      "train_runtime": 1181.4729,
      "train_tokens_per_second": 62929.558
    },
    {
      "epoch": 0.013288948246191055,
      "grad_norm": 37.0,
      "learning_rate": 8.848538529672278e-06,
      "loss": 1.2391,
      "num_input_tokens_seen": 74467328,
      "step": 1000,
      "train_runtime": 1184.0359,
      "train_tokens_per_second": 62892.796
    },
    {
      "epoch": 0.013421837728652965,
      "grad_norm": 20.25,
      "learning_rate": 8.937112488928255e-06,
      "loss": 1.4589,
      "num_input_tokens_seen": 76525056,
      "step": 1010,
      "train_runtime": 1217.6597,
      "train_tokens_per_second": 62846.009
    },
    {
      "epoch": 0.013554727211114876,
      "grad_norm": 28.625,
      "learning_rate": 9.025686448184234e-06,
      "loss": 1.3647,
      "num_input_tokens_seen": 77382400,
      "step": 1020,
      "train_runtime": 1230.5693,
      "train_tokens_per_second": 62883.414
    },
    {
      "epoch": 0.013687616693576788,
      "grad_norm": 25.625,
      "learning_rate": 9.114260407440213e-06,
      "loss": 1.3128,
      "num_input_tokens_seen": 77841408,
      "step": 1030,
      "train_runtime": 1237.2769,
      "train_tokens_per_second": 62913.492
    },
    {
      "epoch": 0.013820506176038697,
      "grad_norm": 16.875,
      "learning_rate": 9.202834366696191e-06,
      "loss": 1.3005,
      "num_input_tokens_seen": 78068736,
      "step": 1040,
      "train_runtime": 1240.775,
      "train_tokens_per_second": 62919.335
    },
    {
      "epoch": 0.013953395658500609,
      "grad_norm": 29.125,
      "learning_rate": 9.29140832595217e-06,
      "loss": 1.2492,
      "num_input_tokens_seen": 78181376,
      "step": 1050,
      "train_runtime": 1243.2927,
      "train_tokens_per_second": 62882.518
    },
    {
      "epoch": 0.014086285140962518,
      "grad_norm": 30.125,
      "learning_rate": 9.379982285208149e-06,
      "loss": 1.5995,
      "num_input_tokens_seen": 80182272,
      "step": 1060,
      "train_runtime": 1274.3074,
      "train_tokens_per_second": 62922.237
    },
    {
      "epoch": 0.01421917462342443,
      "grad_norm": 35.0,
      "learning_rate": 9.46855624446413e-06,
      "loss": 1.3966,
      "num_input_tokens_seen": 81054720,
      "step": 1070,
      "train_runtime": 1286.9636,
      "train_tokens_per_second": 62981.36
    },
    {
      "epoch": 0.014352064105886339,
      "grad_norm": 51.75,
      "learning_rate": 9.557130203720107e-06,
      "loss": 1.3466,
      "num_input_tokens_seen": 81539328,
      "step": 1080,
      "train_runtime": 1293.6889,
      "train_tokens_per_second": 63028.542
    },
    {
      "epoch": 0.01448495358834825,
      "grad_norm": 22.75,
      "learning_rate": 9.645704162976086e-06,
      "loss": 1.1802,
      "num_input_tokens_seen": 81774592,
      "step": 1090,
      "train_runtime": 1297.243,
      "train_tokens_per_second": 63037.219
    },
    {
      "epoch": 0.01461784307081016,
      "grad_norm": 23.75,
      "learning_rate": 9.734278122232064e-06,
      "loss": 1.2959,
      "num_input_tokens_seen": 81888256,
      "step": 1100,
      "train_runtime": 1299.822,
      "train_tokens_per_second": 62999.594
    },
    {
      "epoch": 0.014750732553272071,
      "grad_norm": 38.0,
      "learning_rate": 9.822852081488043e-06,
      "loss": 1.4547,
      "num_input_tokens_seen": 83970048,
      "step": 1110,
      "train_runtime": 1333.6241,
      "train_tokens_per_second": 62963.803
    },
    {
      "epoch": 0.014883622035733983,
      "grad_norm": 24.25,
      "learning_rate": 9.911426040744022e-06,
      "loss": 1.4074,
      "num_input_tokens_seen": 84814336,
      "step": 1120,
      "train_runtime": 1346.4659,
      "train_tokens_per_second": 62990.332
    },
    {
      "epoch": 0.015016511518195892,
      "grad_norm": 12.0625,
      "learning_rate": 1e-05,
      "loss": 1.3583,
      "num_input_tokens_seen": 85291520,
      "step": 1130,
      "train_runtime": 1353.2916,
      "train_tokens_per_second": 63025.234
    },
    {
      "epoch": 0.015149401000657804,
      "grad_norm": 44.75,
      "learning_rate": 1.0088573959255981e-05,
      "loss": 1.2414,
      "num_input_tokens_seen": 85526016,
      "step": 1140,
      "train_runtime": 1356.8226,
      "train_tokens_per_second": 63034.047
    },
    {
      "epoch": 0.015282290483119713,
      "grad_norm": 46.5,
      "learning_rate": 1.0177147918511958e-05,
      "loss": 1.2555,
      "num_input_tokens_seen": 85645568,
      "step": 1150,
      "train_runtime": 1359.2726,
      "train_tokens_per_second": 63008.383
    },
    {
      "epoch": 0.015415179965581624,
      "grad_norm": 14.125,
      "learning_rate": 1.0265721877767937e-05,
      "loss": 1.4954,
      "num_input_tokens_seen": 87775232,
      "step": 1160,
      "train_runtime": 1393.3761,
      "train_tokens_per_second": 62994.645
    },
    {
      "epoch": 0.015548069448043534,
      "grad_norm": 36.0,
      "learning_rate": 1.0354295837023916e-05,
      "loss": 1.3423,
      "num_input_tokens_seen": 88617728,
      "step": 1170,
      "train_runtime": 1405.8833,
      "train_tokens_per_second": 63033.489
    },
    {
      "epoch": 0.015680958930505445,
      "grad_norm": 18.625,
      "learning_rate": 1.0442869796279895e-05,
      "loss": 1.2918,
      "num_input_tokens_seen": 89071104,
      "step": 1180,
      "train_runtime": 1412.9838,
      "train_tokens_per_second": 63037.598
    },
    {
      "epoch": 0.015813848412967357,
      "grad_norm": 28.625,
      "learning_rate": 1.0531443755535872e-05,
      "loss": 1.2502,
      "num_input_tokens_seen": 89303296,
      "step": 1190,
      "train_runtime": 1416.552,
      "train_tokens_per_second": 63042.724
    },
    {
      "epoch": 0.015946737895429268,
      "grad_norm": 31.25,
      "learning_rate": 1.0620017714791853e-05,
      "loss": 1.2624,
      "num_input_tokens_seen": 89419520,
      "step": 1200,
      "train_runtime": 1418.9771,
      "train_tokens_per_second": 63016.888
    },
    {
      "epoch": 0.016079627377891176,
      "grad_norm": 17.625,
      "learning_rate": 1.070859167404783e-05,
      "loss": 1.4408,
      "num_input_tokens_seen": 91414528,
      "step": 1210,
      "train_runtime": 1451.302,
      "train_tokens_per_second": 62987.942
    },
    {
      "epoch": 0.016212516860353087,
      "grad_norm": 21.125,
      "learning_rate": 1.079716563330381e-05,
      "loss": 1.346,
      "num_input_tokens_seen": 92259328,
      "step": 1220,
      "train_runtime": 1464.0559,
      "train_tokens_per_second": 63016.259
    },
    {
      "epoch": 0.016345406342815,
      "grad_norm": 22.125,
      "learning_rate": 1.0885739592559787e-05,
      "loss": 1.2991,
      "num_input_tokens_seen": 92753152,
      "step": 1230,
      "train_runtime": 1471.3802,
      "train_tokens_per_second": 63038.194
    },
    {
      "epoch": 0.01647829582527691,
      "grad_norm": 23.375,
      "learning_rate": 1.0974313551815768e-05,
      "loss": 1.2144,
      "num_input_tokens_seen": 92996352,
      "step": 1240,
      "train_runtime": 1475.0256,
      "train_tokens_per_second": 63047.281
    },
    {
      "epoch": 0.016611185307738818,
      "grad_norm": 18.0,
      "learning_rate": 1.1062887511071745e-05,
      "loss": 1.2316,
      "num_input_tokens_seen": 93112832,
      "step": 1250,
      "train_runtime": 1477.4276,
      "train_tokens_per_second": 63023.618
    },
    {
      "epoch": 0.01674407479020073,
      "grad_norm": 28.25,
      "learning_rate": 1.1151461470327724e-05,
      "loss": 1.4076,
      "num_input_tokens_seen": 95092992,
      "step": 1260,
      "train_runtime": 1508.3484,
      "train_tokens_per_second": 63044.446
    },
    {
      "epoch": 0.01687696427266264,
      "grad_norm": 33.5,
      "learning_rate": 1.1240035429583704e-05,
      "loss": 1.3346,
      "num_input_tokens_seen": 95879936,
      "step": 1270,
      "train_runtime": 1520.8151,
      "train_tokens_per_second": 63045.096
    },
    {
      "epoch": 0.01700985375512455,
      "grad_norm": 29.0,
      "learning_rate": 1.1328609388839682e-05,
      "loss": 1.2411,
      "num_input_tokens_seen": 96335616,
      "step": 1280,
      "train_runtime": 1527.2039,
      "train_tokens_per_second": 63079.735
    },
    {
      "epoch": 0.017142743237586463,
      "grad_norm": 20.875,
      "learning_rate": 1.1417183348095662e-05,
      "loss": 1.1968,
      "num_input_tokens_seen": 96561920,
      "step": 1290,
      "train_runtime": 1530.8363,
      "train_tokens_per_second": 63077.886
    },
    {
      "epoch": 0.01727563272004837,
      "grad_norm": 20.25,
      "learning_rate": 1.150575730735164e-05,
      "loss": 1.1966,
      "num_input_tokens_seen": 96675072,
      "step": 1300,
      "train_runtime": 1533.2956,
      "train_tokens_per_second": 63050.51
    },
    {
      "epoch": 0.017408522202510282,
      "grad_norm": 20.5,
      "learning_rate": 1.159433126660762e-05,
      "loss": 1.4283,
      "num_input_tokens_seen": 98643712,
      "step": 1310,
      "train_runtime": 1564.2844,
      "train_tokens_per_second": 63059.961
    },
    {
      "epoch": 0.017541411684972193,
      "grad_norm": 35.75,
      "learning_rate": 1.1682905225863597e-05,
      "loss": 1.3428,
      "num_input_tokens_seen": 99448576,
      "step": 1320,
      "train_runtime": 1576.5695,
      "train_tokens_per_second": 63079.096
    },
    {
      "epoch": 0.017674301167434105,
      "grad_norm": 17.75,
      "learning_rate": 1.1771479185119576e-05,
      "loss": 1.322,
      "num_input_tokens_seen": 99921920,
      "step": 1330,
      "train_runtime": 1583.2329,
      "train_tokens_per_second": 63112.583
    },
    {
      "epoch": 0.017807190649896013,
      "grad_norm": 24.5,
      "learning_rate": 1.1860053144375555e-05,
      "loss": 1.2146,
      "num_input_tokens_seen": 100148480,
      "step": 1340,
      "train_runtime": 1586.6955,
      "train_tokens_per_second": 63117.642
    },
    {
      "epoch": 0.017940080132357924,
      "grad_norm": 24.25,
      "learning_rate": 1.1948627103631533e-05,
      "loss": 1.1609,
      "num_input_tokens_seen": 100265984,
      "step": 1350,
      "train_runtime": 1589.4972,
      "train_tokens_per_second": 63080.315
    },
    {
      "epoch": 0.018072969614819835,
      "grad_norm": 14.375,
      "learning_rate": 1.203720106288751e-05,
      "loss": 1.4457,
      "num_input_tokens_seen": 102322688,
      "step": 1360,
      "train_runtime": 1621.6387,
      "train_tokens_per_second": 63098.325
    },
    {
      "epoch": 0.018205859097281746,
      "grad_norm": 37.0,
      "learning_rate": 1.2125775022143491e-05,
      "loss": 1.3475,
      "num_input_tokens_seen": 103183616,
      "step": 1370,
      "train_runtime": 1634.4913,
      "train_tokens_per_second": 63128.889
    },
    {
      "epoch": 0.018338748579743658,
      "grad_norm": 33.5,
      "learning_rate": 1.2214348981399468e-05,
      "loss": 1.275,
      "num_input_tokens_seen": 103667200,
      "step": 1380,
      "train_runtime": 1641.222,
      "train_tokens_per_second": 63164.642
    },
    {
      "epoch": 0.018471638062205566,
      "grad_norm": 23.625,
      "learning_rate": 1.2302922940655449e-05,
      "loss": 1.1908,
      "num_input_tokens_seen": 103890432,
      "step": 1390,
      "train_runtime": 1644.603,
      "train_tokens_per_second": 63170.523
    },
    {
      "epoch": 0.018604527544667477,
      "grad_norm": 31.5,
      "learning_rate": 1.2391496899911426e-05,
      "loss": 1.1482,
      "num_input_tokens_seen": 104002304,
      "step": 1400,
      "train_runtime": 1647.5544,
      "train_tokens_per_second": 63125.264
    },
    {
      "epoch": 0.01873741702712939,
      "grad_norm": 33.25,
      "learning_rate": 1.2480070859167406e-05,
      "loss": 1.4356,
      "num_input_tokens_seen": 106147840,
      "step": 1410,
      "train_runtime": 1682.4016,
      "train_tokens_per_second": 63093.046
    },
    {
      "epoch": 0.0188703065095913,
      "grad_norm": 26.5,
      "learning_rate": 1.2568644818423385e-05,
      "loss": 1.3505,
      "num_input_tokens_seen": 107051520,
      "step": 1420,
      "train_runtime": 1695.5648,
      "train_tokens_per_second": 63136.201
    },
    {
      "epoch": 0.019003195992053207,
      "grad_norm": 40.25,
      "learning_rate": 1.2657218777679362e-05,
      "loss": 1.2684,
      "num_input_tokens_seen": 107563520,
      "step": 1430,
      "train_runtime": 1702.8794,
      "train_tokens_per_second": 63165.672
    },
    {
      "epoch": 0.01913608547451512,
      "grad_norm": 26.5,
      "learning_rate": 1.2745792736935343e-05,
      "loss": 1.2164,
      "num_input_tokens_seen": 107816704,
      "step": 1440,
      "train_runtime": 1707.0077,
      "train_tokens_per_second": 63161.229
    },
    {
      "epoch": 0.01926897495697703,
      "grad_norm": 31.5,
      "learning_rate": 1.283436669619132e-05,
      "loss": 1.2124,
      "num_input_tokens_seen": 107938304,
      "step": 1450,
      "train_runtime": 1709.4884,
      "train_tokens_per_second": 63140.705
    },
    {
      "epoch": 0.01940186443943894,
      "grad_norm": 21.75,
      "learning_rate": 1.29229406554473e-05,
      "loss": 1.3959,
      "num_input_tokens_seen": 110088448,
      "step": 1460,
      "train_runtime": 1744.4242,
      "train_tokens_per_second": 63108.758
    },
    {
      "epoch": 0.019534753921900853,
      "grad_norm": 18.25,
      "learning_rate": 1.3011514614703278e-05,
      "loss": 1.3615,
      "num_input_tokens_seen": 110943744,
      "step": 1470,
      "train_runtime": 1757.1619,
      "train_tokens_per_second": 63138.034
    },
    {
      "epoch": 0.01966764340436276,
      "grad_norm": 19.875,
      "learning_rate": 1.3100088573959258e-05,
      "loss": 1.2271,
      "num_input_tokens_seen": 111397120,
      "step": 1480,
      "train_runtime": 1763.6174,
      "train_tokens_per_second": 63163.995
    },
    {
      "epoch": 0.019800532886824672,
      "grad_norm": 42.75,
      "learning_rate": 1.3188662533215235e-05,
      "loss": 1.1788,
      "num_input_tokens_seen": 111628288,
      "step": 1490,
      "train_runtime": 1767.105,
      "train_tokens_per_second": 63170.149
    },
    {
      "epoch": 0.019933422369286583,
      "grad_norm": 9.875,
      "learning_rate": 1.3277236492471214e-05,
      "loss": 1.0731,
      "num_input_tokens_seen": 111746304,
      "step": 1500,
      "train_runtime": 1769.5643,
      "train_tokens_per_second": 63149.051
    },
    {
      "epoch": 0.020066311851748495,
      "grad_norm": 44.25,
      "learning_rate": 1.3365810451727193e-05,
      "loss": 1.4135,
      "num_input_tokens_seen": 113723392,
      "step": 1510,
      "train_runtime": 1801.9704,
      "train_tokens_per_second": 63110.577
    },
    {
      "epoch": 0.020199201334210402,
      "grad_norm": 19.375,
      "learning_rate": 1.3454384410983172e-05,
      "loss": 1.3054,
      "num_input_tokens_seen": 114574080,
      "step": 1520,
      "train_runtime": 1814.747,
      "train_tokens_per_second": 63135.017
    },
    {
      "epoch": 0.020332090816672314,
      "grad_norm": 44.75,
      "learning_rate": 1.354295837023915e-05,
      "loss": 1.2262,
      "num_input_tokens_seen": 115035136,
      "step": 1530,
      "train_runtime": 1821.5447,
      "train_tokens_per_second": 63152.518
    },
    {
      "epoch": 0.020464980299134225,
      "grad_norm": 18.0,
      "learning_rate": 1.363153232949513e-05,
      "loss": 1.165,
      "num_input_tokens_seen": 115260928,
      "step": 1540,
      "train_runtime": 1825.0245,
      "train_tokens_per_second": 63155.827
    },
    {
      "epoch": 0.020597869781596136,
      "grad_norm": 26.75,
      "learning_rate": 1.372010628875111e-05,
      "loss": 1.07,
      "num_input_tokens_seen": 115373568,
      "step": 1550,
      "train_runtime": 1827.4308,
      "train_tokens_per_second": 63134.302
    },
    {
      "epoch": 0.020730759264058048,
      "grad_norm": 40.75,
      "learning_rate": 1.3808680248007087e-05,
      "loss": 1.3826,
      "num_input_tokens_seen": 117358848,
      "step": 1560,
      "train_runtime": 1859.2835,
      "train_tokens_per_second": 63120.47
    },
    {
      "epoch": 0.020863648746519955,
      "grad_norm": 40.75,
      "learning_rate": 1.3897254207263066e-05,
      "loss": 1.3172,
      "num_input_tokens_seen": 118170624,
      "step": 1570,
      "train_runtime": 1871.4903,
      "train_tokens_per_second": 63142.527
    },
    {
      "epoch": 0.020996538228981867,
      "grad_norm": 43.0,
      "learning_rate": 1.3985828166519045e-05,
      "loss": 1.2034,
      "num_input_tokens_seen": 118611200,
      "step": 1580,
      "train_runtime": 1877.9561,
      "train_tokens_per_second": 63159.729
    },
    {
      "epoch": 0.021129427711443778,
      "grad_norm": 35.0,
      "learning_rate": 1.4074402125775024e-05,
      "loss": 1.1575,
      "num_input_tokens_seen": 118831360,
      "step": 1590,
      "train_runtime": 1881.4533,
      "train_tokens_per_second": 63159.346
    },
    {
      "epoch": 0.02126231719390569,
      "grad_norm": 17.5,
      "learning_rate": 1.4162976085031002e-05,
      "loss": 1.1425,
      "num_input_tokens_seen": 118945792,
      "step": 1600,
      "train_runtime": 1884.1785,
      "train_tokens_per_second": 63128.729
    },
    {
      "epoch": 0.021395206676367597,
      "grad_norm": 19.0,
      "learning_rate": 1.4251550044286981e-05,
      "loss": 1.5046,
      "num_input_tokens_seen": 121086976,
      "step": 1610,
      "train_runtime": 1919.5615,
      "train_tokens_per_second": 63080.54
    },
    {
      "epoch": 0.02152809615882951,
      "grad_norm": 14.25,
      "learning_rate": 1.4340124003542958e-05,
      "loss": 1.3329,
      "num_input_tokens_seen": 121959168,
      "step": 1620,
      "train_runtime": 1932.2984,
      "train_tokens_per_second": 63116.115
    },
    {
      "epoch": 0.02166098564129142,
      "grad_norm": 10.625,
      "learning_rate": 1.4428697962798939e-05,
      "loss": 1.1529,
      "num_input_tokens_seen": 122425600,
      "step": 1630,
      "train_runtime": 1938.9983,
      "train_tokens_per_second": 63138.579
    },
    {
      "epoch": 0.02179387512375333,
      "grad_norm": 48.0,
      "learning_rate": 1.4517271922054916e-05,
      "loss": 1.1182,
      "num_input_tokens_seen": 122646016,
      "step": 1640,
      "train_runtime": 1942.4835,
      "train_tokens_per_second": 63138.768
    },
    {
      "epoch": 0.021926764606215243,
      "grad_norm": 32.0,
      "learning_rate": 1.4605845881310897e-05,
      "loss": 1.1253,
      "num_input_tokens_seen": 122759424,
      "step": 1650,
      "train_runtime": 1944.9566,
      "train_tokens_per_second": 63116.793
    },
    {
      "epoch": 0.02205965408867715,
      "grad_norm": 19.375,
      "learning_rate": 1.4694419840566874e-05,
      "loss": 1.4639,
      "num_input_tokens_seen": 124898816,
      "step": 1660,
      "train_runtime": 1977.9379,
      "train_tokens_per_second": 63145.974
    },
    {
      "epoch": 0.02219254357113906,
      "grad_norm": 31.0,
      "learning_rate": 1.4782993799822854e-05,
      "loss": 1.3543,
      "num_input_tokens_seen": 125735424,
      "step": 1670,
      "train_runtime": 1990.2631,
      "train_tokens_per_second": 63175.279
    },
    {
      "epoch": 0.022325433053600973,
      "grad_norm": 20.25,
      "learning_rate": 1.4871567759078831e-05,
      "loss": 1.1937,
      "num_input_tokens_seen": 126214144,
      "step": 1680,
      "train_runtime": 1997.3907,
      "train_tokens_per_second": 63189.511
    },
    {
      "epoch": 0.022458322536062884,
      "grad_norm": 58.0,
      "learning_rate": 1.496014171833481e-05,
      "loss": 1.2284,
      "num_input_tokens_seen": 126446080,
      "step": 1690,
      "train_runtime": 2000.8922,
      "train_tokens_per_second": 63194.849
    },
    {
      "epoch": 0.022591212018524792,
      "grad_norm": 14.25,
      "learning_rate": 1.504871567759079e-05,
      "loss": 1.0572,
      "num_input_tokens_seen": 126565888,
      "step": 1700,
      "train_runtime": 2003.3906,
      "train_tokens_per_second": 63175.841
    },
    {
      "epoch": 0.022724101500986704,
      "grad_norm": 19.5,
      "learning_rate": 1.5137289636846768e-05,
      "loss": 1.3749,
      "num_input_tokens_seen": 128610304,
      "step": 1710,
      "train_runtime": 2037.0097,
      "train_tokens_per_second": 63136.815
    },
    {
      "epoch": 0.022856990983448615,
      "grad_norm": 33.0,
      "learning_rate": 1.5225863596102748e-05,
      "loss": 1.3562,
      "num_input_tokens_seen": 129458688,
      "step": 1720,
      "train_runtime": 2049.9346,
      "train_tokens_per_second": 63152.596
    },
    {
      "epoch": 0.022989880465910526,
      "grad_norm": 31.625,
      "learning_rate": 1.5314437555358725e-05,
      "loss": 1.1339,
      "num_input_tokens_seen": 129954560,
      "step": 1730,
      "train_runtime": 2057.2689,
      "train_tokens_per_second": 63168.486
    },
    {
      "epoch": 0.023122769948372437,
      "grad_norm": 64.0,
      "learning_rate": 1.5403011514614706e-05,
      "loss": 1.0876,
      "num_input_tokens_seen": 130190848,
      "step": 1740,
      "train_runtime": 2060.8853,
      "train_tokens_per_second": 63172.293
    },
    {
      "epoch": 0.023255659430834345,
      "grad_norm": 45.25,
      "learning_rate": 1.5491585473870683e-05,
      "loss": 1.1133,
      "num_input_tokens_seen": 130310144,
      "step": 1750,
      "train_runtime": 2063.3889,
      "train_tokens_per_second": 63153.459
    },
    {
      "epoch": 0.023388548913296257,
      "grad_norm": 22.0,
      "learning_rate": 1.5580159433126664e-05,
      "loss": 1.3646,
      "num_input_tokens_seen": 132434944,
      "step": 1760,
      "train_runtime": 2097.5207,
      "train_tokens_per_second": 63138.801
    },
    {
      "epoch": 0.023521438395758168,
      "grad_norm": 29.75,
      "learning_rate": 1.566873339238264e-05,
      "loss": 1.3146,
      "num_input_tokens_seen": 133268224,
      "step": 1770,
      "train_runtime": 2110.4064,
      "train_tokens_per_second": 63148.134
    },
    {
      "epoch": 0.02365432787822008,
      "grad_norm": 14.75,
      "learning_rate": 1.575730735163862e-05,
      "loss": 1.1535,
      "num_input_tokens_seen": 133724416,
      "step": 1780,
      "train_runtime": 2117.0356,
      "train_tokens_per_second": 63165.879
    },
    {
      "epoch": 0.023787217360681987,
      "grad_norm": 39.5,
      "learning_rate": 1.58458813108946e-05,
      "loss": 1.0966,
      "num_input_tokens_seen": 133949184,
      "step": 1790,
      "train_runtime": 2120.487,
      "train_tokens_per_second": 63169.068
    },
    {
      "epoch": 0.0239201068431439,
      "grad_norm": 25.625,
      "learning_rate": 1.5934455270150576e-05,
      "loss": 1.0976,
      "num_input_tokens_seen": 134064640,
      "step": 1800,
      "train_runtime": 2122.863,
      "train_tokens_per_second": 63152.752
    },
    {
      "epoch": 0.02405299632560581,
      "grad_norm": 37.25,
      "learning_rate": 1.6023029229406556e-05,
      "loss": 1.2959,
      "num_input_tokens_seen": 136079872,
      "step": 1810,
      "train_runtime": 2156.3934,
      "train_tokens_per_second": 63105.309
    },
    {
      "epoch": 0.02418588580806772,
      "grad_norm": 10.75,
      "learning_rate": 1.6111603188662533e-05,
      "loss": 1.308,
      "num_input_tokens_seen": 136895488,
      "step": 1820,
      "train_runtime": 2168.7539,
      "train_tokens_per_second": 63121.725
    },
    {
      "epoch": 0.024318775290529632,
      "grad_norm": 56.0,
      "learning_rate": 1.6200177147918514e-05,
      "loss": 1.1154,
      "num_input_tokens_seen": 137371648,
      "step": 1830,
      "train_runtime": 2175.5494,
      "train_tokens_per_second": 63143.428
    },
    {
      "epoch": 0.02445166477299154,
      "grad_norm": 13.0,
      "learning_rate": 1.628875110717449e-05,
      "loss": 1.0808,
      "num_input_tokens_seen": 137607936,
      "step": 1840,
      "train_runtime": 2179.0742,
      "train_tokens_per_second": 63149.725
    },
    {
      "epoch": 0.02458455425545345,
      "grad_norm": 53.75,
      "learning_rate": 1.637732506643047e-05,
      "loss": 1.0572,
      "num_input_tokens_seen": 137725184,
      "step": 1850,
      "train_runtime": 2181.9968,
      "train_tokens_per_second": 63118.876
    },
    {
      "epoch": 0.024717443737915363,
      "grad_norm": 17.375,
      "learning_rate": 1.646589902568645e-05,
      "loss": 1.4381,
      "num_input_tokens_seen": 139770368,
      "step": 1860,
      "train_runtime": 2215.7867,
      "train_tokens_per_second": 63079.342
    },
    {
      "epoch": 0.024850333220377274,
      "grad_norm": 18.375,
      "learning_rate": 1.655447298494243e-05,
      "loss": 1.2652,
      "num_input_tokens_seen": 140603136,
      "step": 1870,
      "train_runtime": 2228.4908,
      "train_tokens_per_second": 63093.434
    },
    {
      "epoch": 0.024983222702839182,
      "grad_norm": 24.0,
      "learning_rate": 1.6643046944198406e-05,
      "loss": 1.0927,
      "num_input_tokens_seen": 141080832,
      "step": 1880,
      "train_runtime": 2235.3432,
      "train_tokens_per_second": 63113.722
    },
    {
      "epoch": 0.025116112185301093,
      "grad_norm": 173.0,
      "learning_rate": 1.6731620903454387e-05,
      "loss": 1.1488,
      "num_input_tokens_seen": 141312768,
      "step": 1890,
      "train_runtime": 2238.7965,
      "train_tokens_per_second": 63119.969
    },
    {
      "epoch": 0.025249001667763005,
      "grad_norm": 36.75,
      "learning_rate": 1.6820194862710364e-05,
      "loss": 1.0095,
      "num_input_tokens_seen": 141428736,
      "step": 1900,
      "train_runtime": 2241.7002,
      "train_tokens_per_second": 63089.942
    },
    {
      "epoch": 0.025381891150224916,
      "grad_norm": 31.625,
      "learning_rate": 1.6908768821966344e-05,
      "loss": 1.3672,
      "num_input_tokens_seen": 143515904,
      "step": 1910,
      "train_runtime": 2275.5611,
      "train_tokens_per_second": 63068.358
    },
    {
      "epoch": 0.025514780632686827,
      "grad_norm": 22.125,
      "learning_rate": 1.699734278122232e-05,
      "loss": 1.2476,
      "num_input_tokens_seen": 144371968,
      "step": 1920,
      "train_runtime": 2288.284,
      "train_tokens_per_second": 63091.806
    },
    {
      "epoch": 0.025647670115148735,
      "grad_norm": 28.5,
      "learning_rate": 1.7085916740478302e-05,
      "loss": 1.1335,
      "num_input_tokens_seen": 144818176,
      "step": 1930,
      "train_runtime": 2294.8298,
      "train_tokens_per_second": 63106.281
    },
    {
      "epoch": 0.025780559597610646,
      "grad_norm": 25.25,
      "learning_rate": 1.717449069973428e-05,
      "loss": 0.9794,
      "num_input_tokens_seen": 145047552,
      "step": 1940,
      "train_runtime": 2298.322,
      "train_tokens_per_second": 63110.195
    },
    {
      "epoch": 0.025913449080072558,
      "grad_norm": 26.75,
      "learning_rate": 1.726306465899026e-05,
      "loss": 0.9558,
      "num_input_tokens_seen": 145163008,
      "step": 1950,
      "train_runtime": 2300.876,
      "train_tokens_per_second": 63090.321
    },
    {
      "epoch": 0.02604633856253447,
      "grad_norm": 21.25,
      "learning_rate": 1.7351638618246237e-05,
      "loss": 1.3762,
      "num_input_tokens_seen": 147281408,
      "step": 1960,
      "train_runtime": 2334.1784,
      "train_tokens_per_second": 63097.75
    },
    {
      "epoch": 0.02617922804499638,
      "grad_norm": 24.5,
      "learning_rate": 1.7440212577502217e-05,
      "loss": 1.2525,
      "num_input_tokens_seen": 148094208,
      "step": 1970,
      "train_runtime": 2346.6612,
      "train_tokens_per_second": 63108.474
    },
    {
      "epoch": 0.02631211752745829,
      "grad_norm": 56.75,
      "learning_rate": 1.7528786536758195e-05,
      "loss": 1.0906,
      "num_input_tokens_seen": 148589568,
      "step": 1980,
      "train_runtime": 2353.7876,
      "train_tokens_per_second": 63127.856
    },
    {
      "epoch": 0.0264450070099202,
      "grad_norm": 19.875,
      "learning_rate": 1.761736049601417e-05,
      "loss": 1.0186,
      "num_input_tokens_seen": 148823040,
      "step": 1990,
      "train_runtime": 2357.2907,
      "train_tokens_per_second": 63133.088
    },
    {
      "epoch": 0.02657789649238211,
      "grad_norm": 59.25,
      "learning_rate": 1.7705934455270152e-05,
      "loss": 0.9606,
      "num_input_tokens_seen": 148938752,
      "step": 2000,
      "train_runtime": 2359.6558,
      "train_tokens_per_second": 63118.845
    },
    {
      "epoch": 0.026710785974844022,
      "grad_norm": 23.75,
      "learning_rate": 1.779450841452613e-05,
      "loss": 1.3333,
      "num_input_tokens_seen": 151013888,
      "step": 2010,
      "train_runtime": 2391.1073,
      "train_tokens_per_second": 63156.467
    },
    {
      "epoch": 0.02684367545730593,
      "grad_norm": 13.375,
      "learning_rate": 1.788308237378211e-05,
      "loss": 1.262,
      "num_input_tokens_seen": 151839744,
      "step": 2020,
      "train_runtime": 2403.6143,
      "train_tokens_per_second": 63171.427
    },
    {
      "epoch": 0.02697656493976784,
      "grad_norm": 24.0,
      "learning_rate": 1.7971656333038087e-05,
      "loss": 1.0712,
      "num_input_tokens_seen": 152288512,
      "step": 2030,
      "train_runtime": 2410.223,
      "train_tokens_per_second": 63184.407
    },
    {
      "epoch": 0.027109454422229753,
      "grad_norm": 26.5,
      "learning_rate": 1.8060230292294067e-05,
      "loss": 0.9946,
      "num_input_tokens_seen": 152520192,
      "step": 2040,
      "train_runtime": 2414.1736,
      "train_tokens_per_second": 63176.977
    },
    {
      "epoch": 0.027242343904691664,
      "grad_norm": 16.375,
      "learning_rate": 1.8148804251550045e-05,
      "loss": 0.9085,
      "num_input_tokens_seen": 152638720,
      "step": 2050,
      "train_runtime": 2416.7044,
      "train_tokens_per_second": 63159.864
    },
    {
      "epoch": 0.027375233387153575,
      "grad_norm": 32.25,
      "learning_rate": 1.8237378210806025e-05,
      "loss": 1.3353,
      "num_input_tokens_seen": 154722304,
      "step": 2060,
      "train_runtime": 2449.8877,
      "train_tokens_per_second": 63154.856
    },
    {
      "epoch": 0.027508122869615483,
      "grad_norm": 14.375,
      "learning_rate": 1.8325952170062002e-05,
      "loss": 1.2653,
      "num_input_tokens_seen": 155593728,
      "step": 2070,
      "train_runtime": 2462.8165,
      "train_tokens_per_second": 63177.149
    },
    {
      "epoch": 0.027641012352077395,
      "grad_norm": 14.9375,
      "learning_rate": 1.8414526129317983e-05,
      "loss": 1.1127,
      "num_input_tokens_seen": 156094720,
      "step": 2080,
      "train_runtime": 2470.1496,
      "train_tokens_per_second": 63192.416
    },
    {
      "epoch": 0.027773901834539306,
      "grad_norm": 31.625,
      "learning_rate": 1.850310008857396e-05,
      "loss": 1.01,
      "num_input_tokens_seen": 156328192,
      "step": 2090,
      "train_runtime": 2473.6882,
      "train_tokens_per_second": 63196.402
    },
    {
      "epoch": 0.027906791317001217,
      "grad_norm": 28.625,
      "learning_rate": 1.859167404782994e-05,
      "loss": 0.9433,
      "num_input_tokens_seen": 156444928,
      "step": 2100,
      "train_runtime": 2476.2515,
      "train_tokens_per_second": 63178.124
    },
    {
      "epoch": 0.028039680799463125,
      "grad_norm": 21.125,
      "learning_rate": 1.868024800708592e-05,
      "loss": 1.3371,
      "num_input_tokens_seen": 158532608,
      "step": 2110,
      "train_runtime": 2510.7415,
      "train_tokens_per_second": 63141.748
    },
    {
      "epoch": 0.028172570281925036,
      "grad_norm": 15.9375,
      "learning_rate": 1.8768821966341898e-05,
      "loss": 1.2479,
      "num_input_tokens_seen": 159388672,
      "step": 2120,
      "train_runtime": 2523.6832,
      "train_tokens_per_second": 63157.162
    },
    {
      "epoch": 0.028305459764386948,
      "grad_norm": 20.25,
      "learning_rate": 1.8857395925597875e-05,
      "loss": 1.1323,
      "num_input_tokens_seen": 159889152,
      "step": 2130,
      "train_runtime": 2530.749,
      "train_tokens_per_second": 63178.59
    },
    {
      "epoch": 0.02843834924684886,
      "grad_norm": 19.625,
      "learning_rate": 1.8945969884853856e-05,
      "loss": 0.9135,
      "num_input_tokens_seen": 160132096,
      "step": 2140,
      "train_runtime": 2534.6122,
      "train_tokens_per_second": 63178.144
    },
    {
      "epoch": 0.02857123872931077,
      "grad_norm": 18.375,
      "learning_rate": 1.9034543844109833e-05,
      "loss": 0.9503,
      "num_input_tokens_seen": 160251904,
      "step": 2150,
      "train_runtime": 2537.0995,
      "train_tokens_per_second": 63163.43
    },
    {
      "epoch": 0.028704128211772678,
      "grad_norm": 35.0,
      "learning_rate": 1.912311780336581e-05,
      "loss": 1.3825,
      "num_input_tokens_seen": 162369280,
      "step": 2160,
      "train_runtime": 2571.8655,
      "train_tokens_per_second": 63132.881
    },
    {
      "epoch": 0.02883701769423459,
      "grad_norm": 13.1875,
      "learning_rate": 1.921169176262179e-05,
      "loss": 1.2424,
      "num_input_tokens_seen": 163239936,
      "step": 2170,
      "train_runtime": 2584.6829,
      "train_tokens_per_second": 63156.659
    },
    {
      "epoch": 0.0289699071766965,
      "grad_norm": 14.25,
      "learning_rate": 1.9300265721877768e-05,
      "loss": 1.0734,
      "num_input_tokens_seen": 163721216,
      "step": 2180,
      "train_runtime": 2591.7375,
      "train_tokens_per_second": 63170.448
    },
    {
      "epoch": 0.029102796659158412,
      "grad_norm": 239.0,
      "learning_rate": 1.9388839681133748e-05,
      "loss": 0.9845,
      "num_input_tokens_seen": 163955200,
      "step": 2190,
      "train_runtime": 2595.3148,
      "train_tokens_per_second": 63173.532
    },
    {
      "epoch": 0.02923568614162032,
      "grad_norm": 27.0,
      "learning_rate": 1.9477413640389725e-05,
      "loss": 0.877,
      "num_input_tokens_seen": 164072192,
      "step": 2200,
      "train_runtime": 2597.8928,
      "train_tokens_per_second": 63155.874
    },
    {
      "epoch": 0.02936857562408223,
      "grad_norm": 20.125,
      "learning_rate": 1.9565987599645706e-05,
      "loss": 1.5061,
      "num_input_tokens_seen": 166159872,
      "step": 2210,
      "train_runtime": 2632.621,
      "train_tokens_per_second": 63115.759
    },
    {
      "epoch": 0.029501465106544143,
      "grad_norm": 9.375,
      "learning_rate": 1.9654561558901683e-05,
      "loss": 1.2596,
      "num_input_tokens_seen": 166991104,
      "step": 2220,
      "train_runtime": 2645.4484,
      "train_tokens_per_second": 63123.932
    },
    {
      "epoch": 0.029634354589006054,
      "grad_norm": 12.9375,
      "learning_rate": 1.9743135518157664e-05,
      "loss": 1.0119,
      "num_input_tokens_seen": 167469824,
      "step": 2230,
      "train_runtime": 2652.2845,
      "train_tokens_per_second": 63141.726
    },
    {
      "epoch": 0.029767244071467965,
      "grad_norm": 20.375,
      "learning_rate": 1.983170947741364e-05,
      "loss": 0.9841,
      "num_input_tokens_seen": 167700480,
      "step": 2240,
      "train_runtime": 2655.7031,
      "train_tokens_per_second": 63147.299
    },
    {
      "epoch": 0.029900133553929873,
      "grad_norm": 18.375,
      "learning_rate": 1.992028343666962e-05,
      "loss": 0.7341,
      "num_input_tokens_seen": 167816192,
      "step": 2250,
      "train_runtime": 2658.2031,
      "train_tokens_per_second": 63131.442
    },
    {
      "epoch": 0.030033023036391784,
      "grad_norm": 65.0,
      "learning_rate": 1.9999999990737947e-05,
      "loss": 1.3953,
      "num_input_tokens_seen": 169875200,
      "step": 2260,
      "train_runtime": 2691.7898,
      "train_tokens_per_second": 63108.643
    },
    {
      "epoch": 0.030165912518853696,
      "grad_norm": 18.625,
      "learning_rate": 1.9999998879291455e-05,
      "loss": 1.2403,
      "num_input_tokens_seen": 170713856,
      "step": 2270,
      "train_runtime": 2704.384,
      "train_tokens_per_second": 63124.859
    },
    {
      "epoch": 0.030298802001315607,
      "grad_norm": 20.125,
      "learning_rate": 1.9999995915434335e-05,
      "loss": 1.0679,
      "num_input_tokens_seen": 171173120,
      "step": 2280,
      "train_runtime": 2710.8708,
      "train_tokens_per_second": 63143.224
    },
    {
      "epoch": 0.030431691483777515,
      "grad_norm": 23.875,
      "learning_rate": 1.9999991099167138e-05,
      "loss": 0.9209,
      "num_input_tokens_seen": 171401984,
      "step": 2290,
      "train_runtime": 2714.779,
      "train_tokens_per_second": 63136.626
    },
    {
      "epoch": 0.030564580966239426,
      "grad_norm": 35.5,
      "learning_rate": 1.999998443049076e-05,
      "loss": 0.8268,
      "num_input_tokens_seen": 171518464,
      "step": 2300,
      "train_runtime": 2717.2153,
      "train_tokens_per_second": 63122.884
    },
    {
      "epoch": 0.030697470448701338,
      "grad_norm": 28.125,
      "learning_rate": 1.9999975909406437e-05,
      "loss": 1.3309,
      "num_input_tokens_seen": 173581568,
      "step": 2310,
      "train_runtime": 2751.4686,
      "train_tokens_per_second": 63086.879
    },
    {
      "epoch": 0.03083035993116325,
      "grad_norm": 14.25,
      "learning_rate": 1.9999965535915743e-05,
      "loss": 1.2739,
      "num_input_tokens_seen": 174425088,
      "step": 2320,
      "train_runtime": 2764.3039,
      "train_tokens_per_second": 63099.101
    },
    {
      "epoch": 0.03096324941362516,
      "grad_norm": 20.375,
      "learning_rate": 1.9999953310020604e-05,
      "loss": 1.024,
      "num_input_tokens_seen": 174917120,
      "step": 2330,
      "train_runtime": 2771.4535,
      "train_tokens_per_second": 63113.856
    },
    {
      "epoch": 0.031096138896087068,
      "grad_norm": 25.75,
      "learning_rate": 1.999993923172328e-05,
      "loss": 0.9345,
      "num_input_tokens_seen": 175159296,
      "step": 2340,
      "train_runtime": 2775.5357,
      "train_tokens_per_second": 63108.283
    },
    {
      "epoch": 0.03122902837854898,
      "grad_norm": 28.875,
      "learning_rate": 1.999992330102638e-05,
      "loss": 0.7217,
      "num_input_tokens_seen": 175279360,
      "step": 2350,
      "train_runtime": 2778.0198,
      "train_tokens_per_second": 63095.072
    },
    {
      "epoch": 0.03136191786101089,
      "grad_norm": 98.5,
      "learning_rate": 1.9999905517932856e-05,
      "loss": 1.3971,
      "num_input_tokens_seen": 177281280,
      "step": 2360,
      "train_runtime": 2810.5696,
      "train_tokens_per_second": 63076.638
    },
    {
      "epoch": 0.0314948073434728,
      "grad_norm": 23.375,
      "learning_rate": 1.9999885882446003e-05,
      "loss": 1.2401,
      "num_input_tokens_seen": 178112256,
      "step": 2370,
      "train_runtime": 2823.1197,
      "train_tokens_per_second": 63090.579
    },
    {
      "epoch": 0.03162769682593471,
      "grad_norm": 30.375,
      "learning_rate": 1.9999864394569458e-05,
      "loss": 1.009,
      "num_input_tokens_seen": 178542848,
      "step": 2380,
      "train_runtime": 2829.5275,
      "train_tokens_per_second": 63099.88
    },
    {
      "epoch": 0.03176058630839662,
      "grad_norm": 34.5,
      "learning_rate": 1.9999841054307202e-05,
      "loss": 0.8648,
      "num_input_tokens_seen": 178762240,
      "step": 2390,
      "train_runtime": 2832.9345,
      "train_tokens_per_second": 63101.437
    },
    {
      "epoch": 0.031893475790858536,
      "grad_norm": 125.5,
      "learning_rate": 1.9999815861663556e-05,
      "loss": 0.8646,
      "num_input_tokens_seen": 178875136,
      "step": 2400,
      "train_runtime": 2835.463,
      "train_tokens_per_second": 63084.983
    },
    {
      "epoch": 0.032026365273320444,
      "grad_norm": 26.75,
      "learning_rate": 1.9999788816643192e-05,
      "loss": 1.379,
      "num_input_tokens_seen": 180836864,
      "step": 2410,
      "train_runtime": 2867.3719,
      "train_tokens_per_second": 63067.112
    },
    {
      "epoch": 0.03215925475578235,
      "grad_norm": 11.6875,
      "learning_rate": 1.999975991925111e-05,
      "loss": 1.2069,
      "num_input_tokens_seen": 181687552,
      "step": 2420,
      "train_runtime": 2880.0826,
      "train_tokens_per_second": 63084.147
    },
    {
      "epoch": 0.032292144238244266,
      "grad_norm": 14.875,
      "learning_rate": 1.9999729169492675e-05,
      "loss": 1.0275,
      "num_input_tokens_seen": 182158848,
      "step": 2430,
      "train_runtime": 2886.9593,
      "train_tokens_per_second": 63097.13
    },
    {
      "epoch": 0.032425033720706174,
      "grad_norm": 111.5,
      "learning_rate": 1.9999696567373578e-05,
      "loss": 1.0141,
      "num_input_tokens_seen": 182392832,
      "step": 2440,
      "train_runtime": 2890.5196,
      "train_tokens_per_second": 63100.361
    },
    {
      "epoch": 0.03255792320316808,
      "grad_norm": 17.875,
      "learning_rate": 1.9999662112899853e-05,
      "loss": 0.7388,
      "num_input_tokens_seen": 182508032,
      "step": 2450,
      "train_runtime": 2893.0732,
      "train_tokens_per_second": 63084.485
    },
    {
      "epoch": 0.03269081268563,
      "grad_norm": 13.375,
      "learning_rate": 1.9999625806077887e-05,
      "loss": 1.3271,
      "num_input_tokens_seen": 184623104,
      "step": 2460,
      "train_runtime": 2927.2173,
      "train_tokens_per_second": 63071.199
    },
    {
      "epoch": 0.032823702168091905,
      "grad_norm": 10.6875,
      "learning_rate": 1.9999587646914405e-05,
      "loss": 1.216,
      "num_input_tokens_seen": 185492992,
      "step": 2470,
      "train_runtime": 2939.7526,
      "train_tokens_per_second": 63098.164
    },
    {
      "epoch": 0.03295659165055382,
      "grad_norm": 41.25,
      "learning_rate": 1.999954763541648e-05,
      "loss": 1.0378,
      "num_input_tokens_seen": 185973248,
      "step": 2480,
      "train_runtime": 2946.872,
      "train_tokens_per_second": 63108.696
    },
    {
      "epoch": 0.03308948113301573,
      "grad_norm": 30.75,
      "learning_rate": 1.9999505771591518e-05,
      "loss": 0.9008,
      "num_input_tokens_seen": 186195712,
      "step": 2490,
      "train_runtime": 2950.3207,
      "train_tokens_per_second": 63110.329
    },
    {
      "epoch": 0.033222370615477635,
      "grad_norm": 28.375,
      "learning_rate": 1.9999462055447275e-05,
      "loss": 0.7729,
      "num_input_tokens_seen": 186308608,
      "step": 2500,
      "train_runtime": 2952.8855,
      "train_tokens_per_second": 63093.747
    },
    {
      "epoch": 0.03335526009793955,
      "grad_norm": 19.5,
      "learning_rate": 1.999941648699185e-05,
      "loss": 1.3041,
      "num_input_tokens_seen": 188347648,
      "step": 2510,
      "train_runtime": 2986.154,
      "train_tokens_per_second": 63073.655
    },
    {
      "epoch": 0.03348814958040146,
      "grad_norm": 13.0625,
      "learning_rate": 1.9999369066233686e-05,
      "loss": 1.1821,
      "num_input_tokens_seen": 189209600,
      "step": 2520,
      "train_runtime": 2999.3393,
      "train_tokens_per_second": 63083.76
    },
    {
      "epoch": 0.03362103906286337,
      "grad_norm": 37.75,
      "learning_rate": 1.9999319793181563e-05,
      "loss": 0.9759,
      "num_input_tokens_seen": 189663744,
      "step": 2530,
      "train_runtime": 3005.8976,
      "train_tokens_per_second": 63097.207
    },
    {
      "epoch": 0.03375392854532528,
      "grad_norm": 28.75,
      "learning_rate": 1.999926866784461e-05,
      "loss": 0.8587,
      "num_input_tokens_seen": 189892096,
      "step": 2540,
      "train_runtime": 3009.302,
      "train_tokens_per_second": 63101.709
    },
    {
      "epoch": 0.03388681802778719,
      "grad_norm": 44.0,
      "learning_rate": 1.99992156902323e-05,
      "loss": 0.7406,
      "num_input_tokens_seen": 190010880,
      "step": 2550,
      "train_runtime": 3011.8288,
      "train_tokens_per_second": 63088.208
    },
    {
      "epoch": 0.0340197075102491,
      "grad_norm": 22.25,
      "learning_rate": 1.9999160860354445e-05,
      "loss": 1.3462,
      "num_input_tokens_seen": 191981568,
      "step": 2560,
      "train_runtime": 3045.203,
      "train_tokens_per_second": 63043.932
    },
    {
      "epoch": 0.03415259699271101,
      "grad_norm": 14.875,
      "learning_rate": 1.99991041782212e-05,
      "loss": 1.1914,
      "num_input_tokens_seen": 192828928,
      "step": 2570,
      "train_runtime": 3058.4806,
      "train_tokens_per_second": 63047.294
    },
    {
      "epoch": 0.034285486475172926,
      "grad_norm": 12.5625,
      "learning_rate": 1.9999045643843065e-05,
      "loss": 0.9691,
      "num_input_tokens_seen": 193296128,
      "step": 2580,
      "train_runtime": 3065.2654,
      "train_tokens_per_second": 63060.16
    },
    {
      "epoch": 0.034418375957634834,
      "grad_norm": 39.5,
      "learning_rate": 1.9998985257230887e-05,
      "loss": 0.9745,
      "num_input_tokens_seen": 193532416,
      "step": 2590,
      "train_runtime": 3068.8109,
      "train_tokens_per_second": 63064.301
    },
    {
      "epoch": 0.03455126544009674,
      "grad_norm": 23.25,
      "learning_rate": 1.9998923018395847e-05,
      "loss": 0.7285,
      "num_input_tokens_seen": 193649664,
      "step": 2600,
      "train_runtime": 3071.2888,
      "train_tokens_per_second": 63051.597
    },
    {
      "epoch": 0.034684154922558656,
      "grad_norm": 20.875,
      "learning_rate": 1.999885892734948e-05,
      "loss": 1.1848,
      "num_input_tokens_seen": 195664640,
      "step": 2610,
      "train_runtime": 3104.2425,
      "train_tokens_per_second": 63031.364
    },
    {
      "epoch": 0.034817044405020564,
      "grad_norm": 21.75,
      "learning_rate": 1.999879298410365e-05,
      "loss": 1.2075,
      "num_input_tokens_seen": 196535552,
      "step": 2620,
      "train_runtime": 3116.8952,
      "train_tokens_per_second": 63054.911
    },
    {
      "epoch": 0.03494993388748247,
      "grad_norm": 26.5,
      "learning_rate": 1.9998725188670582e-05,
      "loss": 0.976,
      "num_input_tokens_seen": 196980480,
      "step": 2630,
      "train_runtime": 3123.4447,
      "train_tokens_per_second": 63065.142
    },
    {
      "epoch": 0.03508282336994439,
      "grad_norm": 27.75,
      "learning_rate": 1.999865554106283e-05,
      "loss": 0.9105,
      "num_input_tokens_seen": 197201152,
      "step": 2640,
      "train_runtime": 3126.8209,
      "train_tokens_per_second": 63067.619
    },
    {
      "epoch": 0.035215712852406295,
      "grad_norm": 14.875,
      "learning_rate": 1.9998584041293295e-05,
      "loss": 0.7597,
      "num_input_tokens_seen": 197311744,
      "step": 2650,
      "train_runtime": 3129.6996,
      "train_tokens_per_second": 63044.947
    },
    {
      "epoch": 0.03534860233486821,
      "grad_norm": 20.375,
      "learning_rate": 1.9998510689375215e-05,
      "loss": 1.2796,
      "num_input_tokens_seen": 199353088,
      "step": 2660,
      "train_runtime": 3164.448,
      "train_tokens_per_second": 62997.745
    },
    {
      "epoch": 0.03548149181733012,
      "grad_norm": 20.25,
      "learning_rate": 1.999843548532219e-05,
      "loss": 1.2433,
      "num_input_tokens_seen": 200207616,
      "step": 2670,
      "train_runtime": 3177.2264,
      "train_tokens_per_second": 63013.329
    },
    {
      "epoch": 0.035614381299792025,
      "grad_norm": 23.25,
      "learning_rate": 1.999835842914815e-05,
      "loss": 1.0337,
      "num_input_tokens_seen": 200667136,
      "step": 2680,
      "train_runtime": 3183.8771,
      "train_tokens_per_second": 63026.03
    },
    {
      "epoch": 0.03574727078225394,
      "grad_norm": 28.875,
      "learning_rate": 1.999827952086736e-05,
      "loss": 0.85,
      "num_input_tokens_seen": 200897792,
      "step": 2690,
      "train_runtime": 3187.3536,
      "train_tokens_per_second": 63029.653
    },
    {
      "epoch": 0.03588016026471585,
      "grad_norm": 31.875,
      "learning_rate": 1.9998198760494443e-05,
      "loss": 0.6787,
      "num_input_tokens_seen": 201014784,
      "step": 2700,
      "train_runtime": 3189.7454,
      "train_tokens_per_second": 63019.069
    },
    {
      "epoch": 0.03601304974717776,
      "grad_norm": 21.125,
      "learning_rate": 1.999811614804436e-05,
      "loss": 1.2232,
      "num_input_tokens_seen": 203031296,
      "step": 2710,
      "train_runtime": 3222.2783,
      "train_tokens_per_second": 63008.615
    },
    {
      "epoch": 0.03614593922963967,
      "grad_norm": 21.5,
      "learning_rate": 1.999803168353241e-05,
      "loss": 1.2528,
      "num_input_tokens_seen": 203845120,
      "step": 2720,
      "train_runtime": 3234.7007,
      "train_tokens_per_second": 63018.231
    },
    {
      "epoch": 0.03627882871210158,
      "grad_norm": 44.75,
      "learning_rate": 1.999794536697424e-05,
      "loss": 1.0291,
      "num_input_tokens_seen": 204287744,
      "step": 2730,
      "train_runtime": 3241.1942,
      "train_tokens_per_second": 63028.542
    },
    {
      "epoch": 0.03641171819456349,
      "grad_norm": 27.75,
      "learning_rate": 1.9997857198385843e-05,
      "loss": 0.8329,
      "num_input_tokens_seen": 204512768,
      "step": 2740,
      "train_runtime": 3244.5968,
      "train_tokens_per_second": 63031.797
    },
    {
      "epoch": 0.0365446076770254,
      "grad_norm": 13.5625,
      "learning_rate": 1.9997767177783545e-05,
      "loss": 0.7376,
      "num_input_tokens_seen": 204627456,
      "step": 2750,
      "train_runtime": 3247.0582,
      "train_tokens_per_second": 63019.337
    },
    {
      "epoch": 0.036677497159487316,
      "grad_norm": 39.5,
      "learning_rate": 1.999767530518403e-05,
      "loss": 1.2632,
      "num_input_tokens_seen": 206701056,
      "step": 2760,
      "train_runtime": 3280.6742,
      "train_tokens_per_second": 63005.664
    },
    {
      "epoch": 0.03681038664194922,
      "grad_norm": 17.0,
      "learning_rate": 1.9997581580604312e-05,
      "loss": 1.1771,
      "num_input_tokens_seen": 207538944,
      "step": 2770,
      "train_runtime": 3293.1365,
      "train_tokens_per_second": 63021.664
    },
    {
      "epoch": 0.03694327612441113,
      "grad_norm": 12.5625,
      "learning_rate": 1.999748600406175e-05,
      "loss": 0.9738,
      "num_input_tokens_seen": 208002816,
      "step": 2780,
      "train_runtime": 3299.7216,
      "train_tokens_per_second": 63036.475
    },
    {
      "epoch": 0.037076165606873046,
      "grad_norm": 14.9375,
      "learning_rate": 1.9997388575574056e-05,
      "loss": 0.8618,
      "num_input_tokens_seen": 208225280,
      "step": 2790,
      "train_runtime": 3303.1779,
      "train_tokens_per_second": 63037.864
    },
    {
      "epoch": 0.037209055089334954,
      "grad_norm": 27.125,
      "learning_rate": 1.999728929515927e-05,
      "loss": 0.7128,
      "num_input_tokens_seen": 208339968,
      "step": 2800,
      "train_runtime": 3305.7177,
      "train_tokens_per_second": 63024.125
    },
    {
      "epoch": 0.03734194457179686,
      "grad_norm": 27.625,
      "learning_rate": 1.999718816283579e-05,
      "loss": 1.2995,
      "num_input_tokens_seen": 210470912,
      "step": 2810,
      "train_runtime": 3341.0925,
      "train_tokens_per_second": 62994.637
    },
    {
      "epoch": 0.03747483405425878,
      "grad_norm": 27.625,
      "learning_rate": 1.9997085178622343e-05,
      "loss": 1.1998,
      "num_input_tokens_seen": 211363840,
      "step": 2820,
      "train_runtime": 3354.278,
      "train_tokens_per_second": 63013.214
    },
    {
      "epoch": 0.037607723536720684,
      "grad_norm": 39.5,
      "learning_rate": 1.9996980342538007e-05,
      "loss": 0.9342,
      "num_input_tokens_seen": 211857408,
      "step": 2830,
      "train_runtime": 3361.2027,
      "train_tokens_per_second": 63030.238
    },
    {
      "epoch": 0.0377406130191826,
      "grad_norm": 32.0,
      "learning_rate": 1.9996873654602208e-05,
      "loss": 0.8424,
      "num_input_tokens_seen": 212081920,
      "step": 2840,
      "train_runtime": 3364.6597,
      "train_tokens_per_second": 63032.206
    },
    {
      "epoch": 0.03787350250164451,
      "grad_norm": 16.0,
      "learning_rate": 1.9996765114834705e-05,
      "loss": 0.7536,
      "num_input_tokens_seen": 212197120,
      "step": 2850,
      "train_runtime": 3367.06,
      "train_tokens_per_second": 63021.485
    },
    {
      "epoch": 0.038006391984106415,
      "grad_norm": 31.5,
      "learning_rate": 1.99966547232556e-05,
      "loss": 1.2091,
      "num_input_tokens_seen": 214354432,
      "step": 2860,
      "train_runtime": 3399.5981,
      "train_tokens_per_second": 63052.874
    },
    {
      "epoch": 0.03813928146656833,
      "grad_norm": 17.375,
      "learning_rate": 1.999654247988535e-05,
      "loss": 1.1594,
      "num_input_tokens_seen": 215195904,
      "step": 2870,
      "train_runtime": 3412.3901,
      "train_tokens_per_second": 63063.102
    },
    {
      "epoch": 0.03827217094903024,
      "grad_norm": 41.25,
      "learning_rate": 1.999642838474474e-05,
      "loss": 0.94,
      "num_input_tokens_seen": 215650304,
      "step": 2880,
      "train_runtime": 3418.9288,
      "train_tokens_per_second": 63075.4
    },
    {
      "epoch": 0.03840506043149215,
      "grad_norm": 18.75,
      "learning_rate": 1.999631243785491e-05,
      "loss": 0.7891,
      "num_input_tokens_seen": 215879680,
      "step": 2890,
      "train_runtime": 3422.4195,
      "train_tokens_per_second": 63078.089
    },
    {
      "epoch": 0.03853794991395406,
      "grad_norm": 15.9375,
      "learning_rate": 1.9996194639237338e-05,
      "loss": 0.6642,
      "num_input_tokens_seen": 215996672,
      "step": 2900,
      "train_runtime": 3424.8156,
      "train_tokens_per_second": 63068.117
    },
    {
      "epoch": 0.03867083939641597,
      "grad_norm": 19.75,
      "learning_rate": 1.999607498891384e-05,
      "loss": 1.1953,
      "num_input_tokens_seen": 218096896,
      "step": 2910,
      "train_runtime": 3459.7026,
      "train_tokens_per_second": 63039.204
    },
    {
      "epoch": 0.03880372887887788,
      "grad_norm": 21.125,
      "learning_rate": 1.999595348690659e-05,
      "loss": 1.1676,
      "num_input_tokens_seen": 218969088,
      "step": 2920,
      "train_runtime": 3473.0089,
      "train_tokens_per_second": 63048.813
    },
    {
      "epoch": 0.03893661836133979,
      "grad_norm": 32.0,
      "learning_rate": 1.9995830133238084e-05,
      "loss": 0.92,
      "num_input_tokens_seen": 219429632,
      "step": 2930,
      "train_runtime": 3479.6056,
      "train_tokens_per_second": 63061.639
    },
    {
      "epoch": 0.039069507843801705,
      "grad_norm": 16.25,
      "learning_rate": 1.9995704927931177e-05,
      "loss": 0.87,
      "num_input_tokens_seen": 219667456,
      "step": 2940,
      "train_runtime": 3483.0792,
      "train_tokens_per_second": 63067.028
    },
    {
      "epoch": 0.03920239732626361,
      "grad_norm": 21.375,
      "learning_rate": 1.999557787100906e-05,
      "loss": 0.7092,
      "num_input_tokens_seen": 219781632,
      "step": 2950,
      "train_runtime": 3485.4996,
      "train_tokens_per_second": 63055.991
    },
    {
      "epoch": 0.03933528680872552,
      "grad_norm": 14.1875,
      "learning_rate": 1.9995448962495277e-05,
      "loss": 1.2319,
      "num_input_tokens_seen": 221901568,
      "step": 2960,
      "train_runtime": 3519.7086,
      "train_tokens_per_second": 63045.438
    },
    {
      "epoch": 0.039468176291187436,
      "grad_norm": 35.75,
      "learning_rate": 1.99953182024137e-05,
      "loss": 1.1557,
      "num_input_tokens_seen": 222764288,
      "step": 2970,
      "train_runtime": 3532.7374,
      "train_tokens_per_second": 63057.131
    },
    {
      "epoch": 0.039601065773649344,
      "grad_norm": 17.0,
      "learning_rate": 1.9995185590788547e-05,
      "loss": 1.0149,
      "num_input_tokens_seen": 223265792,
      "step": 2980,
      "train_runtime": 3539.7124,
      "train_tokens_per_second": 63074.557
    },
    {
      "epoch": 0.03973395525611125,
      "grad_norm": 14.5,
      "learning_rate": 1.9995051127644393e-05,
      "loss": 0.8525,
      "num_input_tokens_seen": 223515648,
      "step": 2990,
      "train_runtime": 3543.6784,
      "train_tokens_per_second": 63074.473
    },
    {
      "epoch": 0.039866844738573166,
      "grad_norm": 23.5,
      "learning_rate": 1.9994914813006142e-05,
      "loss": 0.7031,
      "num_input_tokens_seen": 223632640,
      "step": 3000,
      "train_runtime": 3546.0829,
      "train_tokens_per_second": 63064.695
    },
    {
      "epoch": 0.039999734221035074,
      "grad_norm": 31.375,
      "learning_rate": 1.9994776646899046e-05,
      "loss": 1.2141,
      "num_input_tokens_seen": 225651200,
      "step": 3010,
      "train_runtime": 3578.9871,
      "train_tokens_per_second": 63048.901
    },
    {
      "epoch": 0.04013262370349699,
      "grad_norm": 19.75,
      "learning_rate": 1.9994636629348697e-05,
      "loss": 1.1343,
      "num_input_tokens_seen": 226478592,
      "step": 3020,
      "train_runtime": 3591.3261,
      "train_tokens_per_second": 63062.665
    },
    {
      "epoch": 0.0402655131859589,
      "grad_norm": 41.25,
      "learning_rate": 1.9994494760381033e-05,
      "loss": 0.9701,
      "num_input_tokens_seen": 226948096,
      "step": 3030,
      "train_runtime": 3597.9051,
      "train_tokens_per_second": 63077.843
    },
    {
      "epoch": 0.040398402668420805,
      "grad_norm": 27.125,
      "learning_rate": 1.9994351040022333e-05,
      "loss": 0.8031,
      "num_input_tokens_seen": 227181056,
      "step": 3040,
      "train_runtime": 3601.3599,
      "train_tokens_per_second": 63082.02
    },
    {
      "epoch": 0.04053129215088272,
      "grad_norm": 13.3125,
      "learning_rate": 1.999420546829922e-05,
      "loss": 0.6071,
      "num_input_tokens_seen": 227300096,
      "step": 3050,
      "train_runtime": 3604.1273,
      "train_tokens_per_second": 63066.611
    },
    {
      "epoch": 0.04066418163334463,
      "grad_norm": 43.0,
      "learning_rate": 1.999405804523866e-05,
      "loss": 1.1745,
      "num_input_tokens_seen": 229383424,
      "step": 3060,
      "train_runtime": 3638.4303,
      "train_tokens_per_second": 63044.611
    },
    {
      "epoch": 0.04079707111580654,
      "grad_norm": 14.75,
      "learning_rate": 1.9993908770867965e-05,
      "loss": 1.113,
      "num_input_tokens_seen": 230257920,
      "step": 3070,
      "train_runtime": 3651.2586,
      "train_tokens_per_second": 63062.617
    },
    {
      "epoch": 0.04092996059826845,
      "grad_norm": 29.875,
      "learning_rate": 1.9993757645214785e-05,
      "loss": 1.0378,
      "num_input_tokens_seen": 230706944,
      "step": 3080,
      "train_runtime": 3657.8765,
      "train_tokens_per_second": 63071.278
    },
    {
      "epoch": 0.04106285008073036,
      "grad_norm": 37.25,
      "learning_rate": 1.999360466830711e-05,
      "loss": 0.8463,
      "num_input_tokens_seen": 230931712,
      "step": 3090,
      "train_runtime": 3661.3085,
      "train_tokens_per_second": 63073.546
    },
    {
      "epoch": 0.04119573956319227,
      "grad_norm": 46.0,
      "learning_rate": 1.9993449840173286e-05,
      "loss": 0.7142,
      "num_input_tokens_seen": 231048448,
      "step": 3100,
      "train_runtime": 3664.0975,
      "train_tokens_per_second": 63057.397
    },
    {
      "epoch": 0.04132862904565418,
      "grad_norm": 27.5,
      "learning_rate": 1.9993293160841987e-05,
      "loss": 1.2811,
      "num_input_tokens_seen": 233049856,
      "step": 3110,
      "train_runtime": 3696.8046,
      "train_tokens_per_second": 63040.891
    },
    {
      "epoch": 0.041461518528116095,
      "grad_norm": 24.0,
      "learning_rate": 1.999313463034224e-05,
      "loss": 1.1237,
      "num_input_tokens_seen": 233874176,
      "step": 3120,
      "train_runtime": 3709.2517,
      "train_tokens_per_second": 63051.578
    },
    {
      "epoch": 0.041594408010578,
      "grad_norm": 21.25,
      "learning_rate": 1.9992974248703407e-05,
      "loss": 1.0268,
      "num_input_tokens_seen": 234355968,
      "step": 3130,
      "train_runtime": 3715.9096,
      "train_tokens_per_second": 63068.264
    },
    {
      "epoch": 0.04172729749303991,
      "grad_norm": 15.4375,
      "learning_rate": 1.99928120159552e-05,
      "loss": 0.7893,
      "num_input_tokens_seen": 234587904,
      "step": 3140,
      "train_runtime": 3719.4027,
      "train_tokens_per_second": 63071.391
    },
    {
      "epoch": 0.041860186975501826,
      "grad_norm": 21.25,
      "learning_rate": 1.9992647932127674e-05,
      "loss": 0.5984,
      "num_input_tokens_seen": 234702592,
      "step": 3150,
      "train_runtime": 3721.7923,
      "train_tokens_per_second": 63061.712
    },
    {
      "epoch": 0.041993076457963734,
      "grad_norm": 14.4375,
      "learning_rate": 1.9992481997251222e-05,
      "loss": 1.1461,
      "num_input_tokens_seen": 236851456,
      "step": 3160,
      "train_runtime": 3755.5694,
      "train_tokens_per_second": 63066.723
    },
    {
      "epoch": 0.04212596594042565,
      "grad_norm": 18.25,
      "learning_rate": 1.9992314211356577e-05,
      "loss": 1.0905,
      "num_input_tokens_seen": 237699072,
      "step": 3170,
      "train_runtime": 3768.2705,
      "train_tokens_per_second": 63079.089
    },
    {
      "epoch": 0.042258855422887556,
      "grad_norm": 27.625,
      "learning_rate": 1.9992144574474826e-05,
      "loss": 0.9078,
      "num_input_tokens_seen": 238157056,
      "step": 3180,
      "train_runtime": 3774.8737,
      "train_tokens_per_second": 63090.072
    },
    {
      "epoch": 0.042391744905349464,
      "grad_norm": 15.1875,
      "learning_rate": 1.9991973086637393e-05,
      "loss": 0.8651,
      "num_input_tokens_seen": 238388224,
      "step": 3190,
      "train_runtime": 3778.3629,
      "train_tokens_per_second": 63092.994
    },
    {
      "epoch": 0.04252463438781138,
      "grad_norm": 25.375,
      "learning_rate": 1.999179974787604e-05,
      "loss": 0.6339,
      "num_input_tokens_seen": 238500352,
      "step": 3200,
      "train_runtime": 3780.7132,
      "train_tokens_per_second": 63083.429
    },
    {
      "epoch": 0.04265752387027329,
      "grad_norm": 7.9375,
      "learning_rate": 1.999162455822288e-05,
      "loss": 1.2237,
      "num_input_tokens_seen": 240530688,
      "step": 3210,
      "train_runtime": 3812.6013,
      "train_tokens_per_second": 63088.34
    },
    {
      "epoch": 0.042790413352735195,
      "grad_norm": 36.5,
      "learning_rate": 1.9991447517710364e-05,
      "loss": 1.1473,
      "num_input_tokens_seen": 241363200,
      "step": 3220,
      "train_runtime": 3825.4672,
      "train_tokens_per_second": 63093.784
    },
    {
      "epoch": 0.04292330283519711,
      "grad_norm": 13.9375,
      "learning_rate": 1.9991268626371288e-05,
      "loss": 0.9374,
      "num_input_tokens_seen": 241816832,
      "step": 3230,
      "train_runtime": 3832.028,
      "train_tokens_per_second": 63104.14
    },
    {
      "epoch": 0.04305619231765902,
      "grad_norm": 9.5625,
      "learning_rate": 1.999108788423879e-05,
      "loss": 0.7964,
      "num_input_tokens_seen": 242043392,
      "step": 3240,
      "train_runtime": 3835.4502,
      "train_tokens_per_second": 63106.905
    },
    {
      "epoch": 0.04318908180012093,
      "grad_norm": 24.875,
      "learning_rate": 1.999090529134635e-05,
      "loss": 0.6197,
      "num_input_tokens_seen": 242162944,
      "step": 3250,
      "train_runtime": 3837.8648,
      "train_tokens_per_second": 63098.351
    },
    {
      "epoch": 0.04332197128258284,
      "grad_norm": 13.5625,
      "learning_rate": 1.999072084772779e-05,
      "loss": 1.2144,
      "num_input_tokens_seen": 244173056,
      "step": 3260,
      "train_runtime": 3870.1594,
      "train_tokens_per_second": 63091.214
    },
    {
      "epoch": 0.04345486076504475,
      "grad_norm": 30.75,
      "learning_rate": 1.999053455341728e-05,
      "loss": 1.1001,
      "num_input_tokens_seen": 245011712,
      "step": 3270,
      "train_runtime": 3882.5617,
      "train_tokens_per_second": 63105.684
    },
    {
      "epoch": 0.04358775024750666,
      "grad_norm": 34.5,
      "learning_rate": 1.999034640844933e-05,
      "loss": 0.9705,
      "num_input_tokens_seen": 245479168,
      "step": 3280,
      "train_runtime": 3889.1653,
      "train_tokens_per_second": 63118.728
    },
    {
      "epoch": 0.04372063972996857,
      "grad_norm": 11.8125,
      "learning_rate": 1.9990156412858787e-05,
      "loss": 0.7883,
      "num_input_tokens_seen": 245709312,
      "step": 3290,
      "train_runtime": 3892.6818,
      "train_tokens_per_second": 63120.832
    },
    {
      "epoch": 0.043853529212430485,
      "grad_norm": 24.625,
      "learning_rate": 1.998996456668085e-05,
      "loss": 0.5941,
      "num_input_tokens_seen": 245822976,
      "step": 3300,
      "train_runtime": 3895.0825,
      "train_tokens_per_second": 63111.109
    },
    {
      "epoch": 0.04398641869489239,
      "grad_norm": 21.75,
      "learning_rate": 1.9989770869951054e-05,
      "loss": 1.1874,
      "num_input_tokens_seen": 247926528,
      "step": 3310,
      "train_runtime": 3928.756,
      "train_tokens_per_second": 63105.606
    },
    {
      "epoch": 0.0441193081773543,
      "grad_norm": 29.375,
      "learning_rate": 1.9989575322705286e-05,
      "loss": 1.1652,
      "num_input_tokens_seen": 248780544,
      "step": 3320,
      "train_runtime": 3941.515,
      "train_tokens_per_second": 63118.0
    },
    {
      "epoch": 0.044252197659816216,
      "grad_norm": 38.0,
      "learning_rate": 1.9989377924979764e-05,
      "loss": 0.9127,
      "num_input_tokens_seen": 249231872,
      "step": 3330,
      "train_runtime": 3947.9844,
      "train_tokens_per_second": 63128.89
    },
    {
      "epoch": 0.04438508714227812,
      "grad_norm": 18.625,
      "learning_rate": 1.9989178676811053e-05,
      "loss": 0.766,
      "num_input_tokens_seen": 249458688,
      "step": 3340,
      "train_runtime": 3951.4215,
      "train_tokens_per_second": 63131.378
    },
    {
      "epoch": 0.04451797662474004,
      "grad_norm": 20.125,
      "learning_rate": 1.9988977578236062e-05,
      "loss": 0.6241,
      "num_input_tokens_seen": 249575168,
      "step": 3350,
      "train_runtime": 3953.7991,
      "train_tokens_per_second": 63122.875
    },
    {
      "epoch": 0.044650866107201946,
      "grad_norm": 21.25,
      "learning_rate": 1.998877462929205e-05,
      "loss": 1.0898,
      "num_input_tokens_seen": 251651072,
      "step": 3360,
      "train_runtime": 3987.4608,
      "train_tokens_per_second": 63110.608
    },
    {
      "epoch": 0.044783755589663854,
      "grad_norm": 17.125,
      "learning_rate": 1.9988569830016603e-05,
      "loss": 1.1471,
      "num_input_tokens_seen": 252487936,
      "step": 3370,
      "train_runtime": 4000.2508,
      "train_tokens_per_second": 63118.027
    },
    {
      "epoch": 0.04491664507212577,
      "grad_norm": 13.9375,
      "learning_rate": 1.998836318044766e-05,
      "loss": 0.962,
      "num_input_tokens_seen": 252959488,
      "step": 3380,
      "train_runtime": 4006.9215,
      "train_tokens_per_second": 63130.632
    },
    {
      "epoch": 0.04504953455458768,
      "grad_norm": 23.25,
      "learning_rate": 1.9988154680623503e-05,
      "loss": 0.7574,
      "num_input_tokens_seen": 253188864,
      "step": 3390,
      "train_runtime": 4010.4443,
      "train_tokens_per_second": 63132.372
    },
    {
      "epoch": 0.045182424037049584,
      "grad_norm": 28.0,
      "learning_rate": 1.998794433058276e-05,
      "loss": 0.7401,
      "num_input_tokens_seen": 253308928,
      "step": 3400,
      "train_runtime": 4012.8654,
      "train_tokens_per_second": 63124.202
    },
    {
      "epoch": 0.0453153135195115,
      "grad_norm": 20.125,
      "learning_rate": 1.9987732130364384e-05,
      "loss": 1.1512,
      "num_input_tokens_seen": 255256320,
      "step": 3410,
      "train_runtime": 4043.788,
      "train_tokens_per_second": 63123.071
    },
    {
      "epoch": 0.04544820300197341,
      "grad_norm": 14.8125,
      "learning_rate": 1.998751808000769e-05,
      "loss": 1.1898,
      "num_input_tokens_seen": 256086016,
      "step": 3420,
      "train_runtime": 4056.4336,
      "train_tokens_per_second": 63130.829
    },
    {
      "epoch": 0.04558109248443532,
      "grad_norm": 20.875,
      "learning_rate": 1.9987302179552333e-05,
      "loss": 0.8803,
      "num_input_tokens_seen": 256545024,
      "step": 3430,
      "train_runtime": 4062.9164,
      "train_tokens_per_second": 63143.072
    },
    {
      "epoch": 0.04571398196689723,
      "grad_norm": 40.25,
      "learning_rate": 1.99870844290383e-05,
      "loss": 0.8032,
      "num_input_tokens_seen": 256768768,
      "step": 3440,
      "train_runtime": 4066.3374,
      "train_tokens_per_second": 63144.973
    },
    {
      "epoch": 0.04584687144935914,
      "grad_norm": 9.125,
      "learning_rate": 1.9986864828505926e-05,
      "loss": 0.5853,
      "num_input_tokens_seen": 256883712,
      "step": 3450,
      "train_runtime": 4068.7464,
      "train_tokens_per_second": 63135.838
    },
    {
      "epoch": 0.04597976093182105,
      "grad_norm": 20.75,
      "learning_rate": 1.99866433779959e-05,
      "loss": 1.2125,
      "num_input_tokens_seen": 258943488,
      "step": 3460,
      "train_runtime": 4100.9411,
      "train_tokens_per_second": 63142.455
    },
    {
      "epoch": 0.04611265041428296,
      "grad_norm": 20.0,
      "learning_rate": 1.9986420077549234e-05,
      "loss": 1.1617,
      "num_input_tokens_seen": 259772416,
      "step": 3470,
      "train_runtime": 4113.6097,
      "train_tokens_per_second": 63149.505
    },
    {
      "epoch": 0.046245539896744875,
      "grad_norm": 40.25,
      "learning_rate": 1.9986194927207298e-05,
      "loss": 0.9477,
      "num_input_tokens_seen": 260259584,
      "step": 3480,
      "train_runtime": 4120.8975,
      "train_tokens_per_second": 63156.044
    },
    {
      "epoch": 0.04637842937920678,
      "grad_norm": 32.75,
      "learning_rate": 1.9985967927011795e-05,
      "loss": 0.8452,
      "num_input_tokens_seen": 260490240,
      "step": 3490,
      "train_runtime": 4124.382,
      "train_tokens_per_second": 63158.611
    },
    {
      "epoch": 0.04651131886166869,
      "grad_norm": 30.25,
      "learning_rate": 1.998573907700478e-05,
      "loss": 0.6924,
      "num_input_tokens_seen": 260604928,
      "step": 3500,
      "train_runtime": 4126.7485,
      "train_tokens_per_second": 63150.184
    },
    {
      "epoch": 0.046644208344130605,
      "grad_norm": 41.0,
      "learning_rate": 1.9985508377228636e-05,
      "loss": 1.2229,
      "num_input_tokens_seen": 262786816,
      "step": 3510,
      "train_runtime": 4160.7846,
      "train_tokens_per_second": 63157.996
    },
    {
      "epoch": 0.04677709782659251,
      "grad_norm": 23.75,
      "learning_rate": 1.998527582772611e-05,
      "loss": 1.178,
      "num_input_tokens_seen": 263635456,
      "step": 3520,
      "train_runtime": 4174.0624,
      "train_tokens_per_second": 63160.401
    },
    {
      "epoch": 0.04690998730905443,
      "grad_norm": 16.0,
      "learning_rate": 1.998504142854027e-05,
      "loss": 0.9461,
      "num_input_tokens_seen": 264105216,
      "step": 3530,
      "train_runtime": 4180.915,
      "train_tokens_per_second": 63169.239
    },
    {
      "epoch": 0.047042876791516336,
      "grad_norm": 44.0,
      "learning_rate": 1.9984805179714542e-05,
      "loss": 0.7732,
      "num_input_tokens_seen": 264336128,
      "step": 3540,
      "train_runtime": 4184.3485,
      "train_tokens_per_second": 63172.588
    },
    {
      "epoch": 0.047175766273978244,
      "grad_norm": 29.5,
      "learning_rate": 1.998456708129269e-05,
      "loss": 0.5891,
      "num_input_tokens_seen": 264451584,
      "step": 3550,
      "train_runtime": 4186.7984,
      "train_tokens_per_second": 63163.2
    },
    {
      "epoch": 0.04730865575644016,
      "grad_norm": 7.96875,
      "learning_rate": 1.9984327133318813e-05,
      "loss": 1.1809,
      "num_input_tokens_seen": 266387712,
      "step": 3560,
      "train_runtime": 4218.3159,
      "train_tokens_per_second": 63150.252
    },
    {
      "epoch": 0.047441545238902066,
      "grad_norm": 15.0625,
      "learning_rate": 1.9984085335837365e-05,
      "loss": 1.0923,
      "num_input_tokens_seen": 267193600,
      "step": 3570,
      "train_runtime": 4230.8586,
      "train_tokens_per_second": 63153.517
    },
    {
      "epoch": 0.047574434721363974,
      "grad_norm": 27.5,
      "learning_rate": 1.9983841688893135e-05,
      "loss": 0.91,
      "num_input_tokens_seen": 267663616,
      "step": 3580,
      "train_runtime": 4237.5737,
      "train_tokens_per_second": 63164.357
    },
    {
      "epoch": 0.04770732420382589,
      "grad_norm": 28.625,
      "learning_rate": 1.9983596192531255e-05,
      "loss": 0.8306,
      "num_input_tokens_seen": 267893760,
      "step": 3590,
      "train_runtime": 4241.4742,
      "train_tokens_per_second": 63160.531
    },
    {
      "epoch": 0.0478402136862878,
      "grad_norm": 18.125,
      "learning_rate": 1.9983348846797205e-05,
      "loss": 0.6605,
      "num_input_tokens_seen": 268009216,
      "step": 3600,
      "train_runtime": 4243.8652,
      "train_tokens_per_second": 63152.151
    },
    {
      "epoch": 0.04797310316874971,
      "grad_norm": 19.875,
      "learning_rate": 1.9983099651736798e-05,
      "loss": 1.1349,
      "num_input_tokens_seen": 270150656,
      "step": 3610,
      "train_runtime": 4278.259,
      "train_tokens_per_second": 63144.999
    },
    {
      "epoch": 0.04810599265121162,
      "grad_norm": 17.0,
      "learning_rate": 1.9982848607396202e-05,
      "loss": 1.1417,
      "num_input_tokens_seen": 271004160,
      "step": 3620,
      "train_runtime": 4291.2526,
      "train_tokens_per_second": 63152.694
    },
    {
      "epoch": 0.04823888213367353,
      "grad_norm": 10.1875,
      "learning_rate": 1.9982595713821912e-05,
      "loss": 0.9046,
      "num_input_tokens_seen": 271479552,
      "step": 3630,
      "train_runtime": 4298.4513,
      "train_tokens_per_second": 63157.527
    },
    {
      "epoch": 0.04837177161613544,
      "grad_norm": 18.25,
      "learning_rate": 1.9982340971060783e-05,
      "loss": 0.7164,
      "num_input_tokens_seen": 271710208,
      "step": 3640,
      "train_runtime": 4301.9078,
      "train_tokens_per_second": 63160.398
    },
    {
      "epoch": 0.04850466109859735,
      "grad_norm": 26.5,
      "learning_rate": 1.9982084379159997e-05,
      "loss": 0.6639,
      "num_input_tokens_seen": 271826688,
      "step": 3650,
      "train_runtime": 4304.2909,
      "train_tokens_per_second": 63152.49
    },
    {
      "epoch": 0.048637550581059265,
      "grad_norm": 34.75,
      "learning_rate": 1.9981825938167093e-05,
      "loss": 1.2168,
      "num_input_tokens_seen": 273769216,
      "step": 3660,
      "train_runtime": 4336.8724,
      "train_tokens_per_second": 63125.956
    },
    {
      "epoch": 0.04877044006352117,
      "grad_norm": 37.75,
      "learning_rate": 1.9981565648129936e-05,
      "loss": 1.1719,
      "num_input_tokens_seen": 274616320,
      "step": 3670,
      "train_runtime": 4349.5615,
      "train_tokens_per_second": 63136.553
    },
    {
      "epoch": 0.04890332954598308,
      "grad_norm": 25.875,
      "learning_rate": 1.998130350909675e-05,
      "loss": 0.9483,
      "num_input_tokens_seen": 275107840,
      "step": 3680,
      "train_runtime": 4356.5758,
      "train_tokens_per_second": 63147.723
    },
    {
      "epoch": 0.049036219028444995,
      "grad_norm": 15.25,
      "learning_rate": 1.998103952111609e-05,
      "loss": 0.7942,
      "num_input_tokens_seen": 275342592,
      "step": 3690,
      "train_runtime": 4360.0832,
      "train_tokens_per_second": 63150.765
    },
    {
      "epoch": 0.0491691085109069,
      "grad_norm": 20.75,
      "learning_rate": 1.9980773684236853e-05,
      "loss": 0.548,
      "num_input_tokens_seen": 275457536,
      "step": 3700,
      "train_runtime": 4362.4917,
      "train_tokens_per_second": 63142.248
    },
    {
      "epoch": 0.04930199799336882,
      "grad_norm": 12.125,
      "learning_rate": 1.9980505998508293e-05,
      "loss": 1.2023,
      "num_input_tokens_seen": 277524736,
      "step": 3710,
      "train_runtime": 4394.9979,
      "train_tokens_per_second": 63145.591
    },
    {
      "epoch": 0.049434887475830726,
      "grad_norm": 51.5,
      "learning_rate": 1.9980236463979987e-05,
      "loss": 1.1084,
      "num_input_tokens_seen": 278341120,
      "step": 3720,
      "train_runtime": 4407.4019,
      "train_tokens_per_second": 63153.107
    },
    {
      "epoch": 0.049567776958292634,
      "grad_norm": 25.75,
      "learning_rate": 1.997996508070187e-05,
      "loss": 0.9095,
      "num_input_tokens_seen": 278799104,
      "step": 3730,
      "train_runtime": 4413.8723,
      "train_tokens_per_second": 63164.288
    },
    {
      "epoch": 0.04970066644075455,
      "grad_norm": 23.125,
      "learning_rate": 1.9979691848724214e-05,
      "loss": 0.7301,
      "num_input_tokens_seen": 279035904,
      "step": 3740,
      "train_runtime": 4417.3717,
      "train_tokens_per_second": 63167.857
    },
    {
      "epoch": 0.049833555923216456,
      "grad_norm": 21.25,
      "learning_rate": 1.997941676809763e-05,
      "loss": 0.7436,
      "num_input_tokens_seen": 279152384,
      "step": 3750,
      "train_runtime": 4419.8079,
      "train_tokens_per_second": 63159.392
    },
    {
      "epoch": 0.049966445405678364,
      "grad_norm": 13.5625,
      "learning_rate": 1.9979139838873076e-05,
      "loss": 1.1021,
      "num_input_tokens_seen": 281237760,
      "step": 3760,
      "train_runtime": 4453.6856,
      "train_tokens_per_second": 63147.197
    },
    {
      "epoch": 0.05009933488814028,
      "grad_norm": 16.875,
      "learning_rate": 1.9978861061101843e-05,
      "loss": 1.1423,
      "num_input_tokens_seen": 282082304,
      "step": 3770,
      "train_runtime": 4466.6194,
      "train_tokens_per_second": 63153.423
    },
    {
      "epoch": 0.05023222437060219,
      "grad_norm": 27.375,
      "learning_rate": 1.9978580434835582e-05,
      "loss": 0.9315,
      "num_input_tokens_seen": 282536192,
      "step": 3780,
      "train_runtime": 4473.3398,
      "train_tokens_per_second": 63160.011
    },
    {
      "epoch": 0.0503651138530641,
      "grad_norm": 27.875,
      "learning_rate": 1.9978297960126272e-05,
      "loss": 0.775,
      "num_input_tokens_seen": 282775808,
      "step": 3790,
      "train_runtime": 4476.9194,
      "train_tokens_per_second": 63163.034
    },
    {
      "epoch": 0.05049800333552601,
      "grad_norm": 7.5,
      "learning_rate": 1.9978013637026242e-05,
      "loss": 0.5868,
      "num_input_tokens_seen": 282891776,
      "step": 3800,
      "train_runtime": 4479.3626,
      "train_tokens_per_second": 63154.472
    },
    {
      "epoch": 0.05063089281798792,
      "grad_norm": 20.0,
      "learning_rate": 1.9977727465588157e-05,
      "loss": 1.0946,
      "num_input_tokens_seen": 284926976,
      "step": 3810,
      "train_runtime": 4513.5652,
      "train_tokens_per_second": 63126.811
    },
    {
      "epoch": 0.05076378230044983,
      "grad_norm": 17.875,
      "learning_rate": 1.9977439445865028e-05,
      "loss": 1.1351,
      "num_input_tokens_seen": 285771264,
      "step": 3820,
      "train_runtime": 4526.2666,
      "train_tokens_per_second": 63136.198
    },
    {
      "epoch": 0.05089667178291174,
      "grad_norm": 21.125,
      "learning_rate": 1.9977149577910207e-05,
      "loss": 0.9844,
      "num_input_tokens_seen": 286244352,
      "step": 3830,
      "train_runtime": 4532.9829,
      "train_tokens_per_second": 63147.018
    },
    {
      "epoch": 0.051029561265373655,
      "grad_norm": 22.25,
      "learning_rate": 1.9976857861777393e-05,
      "loss": 0.781,
      "num_input_tokens_seen": 286472192,
      "step": 3840,
      "train_runtime": 4536.4742,
      "train_tokens_per_second": 63148.643
    },
    {
      "epoch": 0.05116245074783556,
      "grad_norm": 43.25,
      "learning_rate": 1.9976564297520624e-05,
      "loss": 0.6008,
      "num_input_tokens_seen": 286586368,
      "step": 3850,
      "train_runtime": 4538.8376,
      "train_tokens_per_second": 63140.917
    },
    {
      "epoch": 0.05129534023029747,
      "grad_norm": 9.625,
      "learning_rate": 1.9976268885194275e-05,
      "loss": 1.0949,
      "num_input_tokens_seen": 288601600,
      "step": 3860,
      "train_runtime": 4571.0591,
      "train_tokens_per_second": 63136.703
    },
    {
      "epoch": 0.051428229712759385,
      "grad_norm": 14.8125,
      "learning_rate": 1.9975971624853075e-05,
      "loss": 1.1495,
      "num_input_tokens_seen": 289456640,
      "step": 3870,
      "train_runtime": 4584.1384,
      "train_tokens_per_second": 63143.085
    },
    {
      "epoch": 0.05156111919522129,
      "grad_norm": 17.375,
      "learning_rate": 1.9975672516552085e-05,
      "loss": 0.8982,
      "num_input_tokens_seen": 289932288,
      "step": 3880,
      "train_runtime": 4590.7528,
      "train_tokens_per_second": 63155.718
    },
    {
      "epoch": 0.05169400867768321,
      "grad_norm": 14.625,
      "learning_rate": 1.997537156034671e-05,
      "loss": 0.7782,
      "num_input_tokens_seen": 290170624,
      "step": 3890,
      "train_runtime": 4594.2907,
      "train_tokens_per_second": 63158.96
    },
    {
      "epoch": 0.051826898160145116,
      "grad_norm": 30.75,
      "learning_rate": 1.9975068756292703e-05,
      "loss": 0.6089,
      "num_input_tokens_seen": 290289152,
      "step": 3900,
      "train_runtime": 4596.6879,
      "train_tokens_per_second": 63151.808
    },
    {
      "epoch": 0.051959787642607023,
      "grad_norm": 11.3125,
      "learning_rate": 1.9974764104446158e-05,
      "loss": 1.0711,
      "num_input_tokens_seen": 292413184,
      "step": 3910,
      "train_runtime": 4630.1971,
      "train_tokens_per_second": 63153.507
    },
    {
      "epoch": 0.05209267712506894,
      "grad_norm": 13.9375,
      "learning_rate": 1.9974457604863503e-05,
      "loss": 1.1551,
      "num_input_tokens_seen": 293258240,
      "step": 3920,
      "train_runtime": 4643.1612,
      "train_tokens_per_second": 63159.177
    },
    {
      "epoch": 0.052225566607530846,
      "grad_norm": 13.875,
      "learning_rate": 1.997414925760152e-05,
      "loss": 0.8023,
      "num_input_tokens_seen": 293750016,
      "step": 3930,
      "train_runtime": 4649.9298,
      "train_tokens_per_second": 63173.0
    },
    {
      "epoch": 0.05235845608999276,
      "grad_norm": 14.375,
      "learning_rate": 1.9973839062717323e-05,
      "loss": 0.7827,
      "num_input_tokens_seen": 293981440,
      "step": 3940,
      "train_runtime": 4653.398,
      "train_tokens_per_second": 63175.649
    },
    {
      "epoch": 0.05249134557245467,
      "grad_norm": 36.5,
      "learning_rate": 1.9973527020268372e-05,
      "loss": 0.5668,
      "num_input_tokens_seen": 294097664,
      "step": 3950,
      "train_runtime": 4655.7661,
      "train_tokens_per_second": 63168.479
    },
    {
      "epoch": 0.05262423505491658,
      "grad_norm": 9.1875,
      "learning_rate": 1.9973213130312478e-05,
      "loss": 1.1252,
      "num_input_tokens_seen": 296060928,
      "step": 3960,
      "train_runtime": 4687.1641,
      "train_tokens_per_second": 63164.191
    },
    {
      "epoch": 0.05275712453737849,
      "grad_norm": 16.25,
      "learning_rate": 1.9972897392907774e-05,
      "loss": 1.1132,
      "num_input_tokens_seen": 296910592,
      "step": 3970,
      "train_runtime": 4700.0866,
      "train_tokens_per_second": 63171.303
    },
    {
      "epoch": 0.0528900140198404,
      "grad_norm": 19.875,
      "learning_rate": 1.997257980811276e-05,
      "loss": 0.9069,
      "num_input_tokens_seen": 297364992,
      "step": 3980,
      "train_runtime": 4706.771,
      "train_tokens_per_second": 63178.131
    },
    {
      "epoch": 0.05302290350230231,
      "grad_norm": 18.5,
      "learning_rate": 1.9972260375986262e-05,
      "loss": 0.7367,
      "num_input_tokens_seen": 297592320,
      "step": 3990,
      "train_runtime": 4710.2149,
      "train_tokens_per_second": 63180.199
    },
    {
      "epoch": 0.05315579298476422,
      "grad_norm": 8.8125,
      "learning_rate": 1.9971939096587446e-05,
      "loss": 0.5034,
      "num_input_tokens_seen": 297705984,
      "step": 4000,
      "train_runtime": 4712.6309,
      "train_tokens_per_second": 63171.929
    },
    {
      "epoch": 0.05328868246722613,
      "grad_norm": 40.25,
      "learning_rate": 1.9971615969975834e-05,
      "loss": 1.1287,
      "num_input_tokens_seen": 299732736,
      "step": 4010,
      "train_runtime": 4744.1766,
      "train_tokens_per_second": 63179.085
    },
    {
      "epoch": 0.053421571949688045,
      "grad_norm": 12.0,
      "learning_rate": 1.9971290996211278e-05,
      "loss": 1.0893,
      "num_input_tokens_seen": 300570368,
      "step": 4020,
      "train_runtime": 4756.5807,
      "train_tokens_per_second": 63190.427
    },
    {
      "epoch": 0.05355446143214995,
      "grad_norm": 12.6875,
      "learning_rate": 1.9970964175353974e-05,
      "loss": 0.8078,
      "num_input_tokens_seen": 301021696,
      "step": 4030,
      "train_runtime": 4763.1708,
      "train_tokens_per_second": 63197.754
    },
    {
      "epoch": 0.05368735091461186,
      "grad_norm": 13.125,
      "learning_rate": 1.997063550746447e-05,
      "loss": 0.7446,
      "num_input_tokens_seen": 301249024,
      "step": 4040,
      "train_runtime": 4766.9928,
      "train_tokens_per_second": 63194.772
    },
    {
      "epoch": 0.053820240397073775,
      "grad_norm": 10.5625,
      "learning_rate": 1.9970304992603645e-05,
      "loss": 0.5615,
      "num_input_tokens_seen": 301363968,
      "step": 4050,
      "train_runtime": 4769.3268,
      "train_tokens_per_second": 63187.947
    },
    {
      "epoch": 0.05395312987953568,
      "grad_norm": 20.5,
      "learning_rate": 1.996997263083272e-05,
      "loss": 1.0982,
      "num_input_tokens_seen": 303423744,
      "step": 4060,
      "train_runtime": 4802.4607,
      "train_tokens_per_second": 63180.891
    },
    {
      "epoch": 0.0540860193619976,
      "grad_norm": 36.75,
      "learning_rate": 1.996963842221327e-05,
      "loss": 1.1163,
      "num_input_tokens_seen": 304276480,
      "step": 4070,
      "train_runtime": 4815.1268,
      "train_tokens_per_second": 63191.79
    },
    {
      "epoch": 0.054218908844459505,
      "grad_norm": 41.0,
      "learning_rate": 1.99693023668072e-05,
      "loss": 0.9143,
      "num_input_tokens_seen": 304761344,
      "step": 4080,
      "train_runtime": 4821.8248,
      "train_tokens_per_second": 63204.566
    },
    {
      "epoch": 0.05435179832692141,
      "grad_norm": 15.25,
      "learning_rate": 1.9968964464676755e-05,
      "loss": 0.7318,
      "num_input_tokens_seen": 304995328,
      "step": 4090,
      "train_runtime": 4825.5891,
      "train_tokens_per_second": 63203.75
    },
    {
      "epoch": 0.05448468780938333,
      "grad_norm": 19.875,
      "learning_rate": 1.996862471588454e-05,
      "loss": 0.5998,
      "num_input_tokens_seen": 305110272,
      "step": 4100,
      "train_runtime": 4827.9429,
      "train_tokens_per_second": 63196.745
    },
    {
      "epoch": 0.054617577291845236,
      "grad_norm": 26.375,
      "learning_rate": 1.9968283120493483e-05,
      "loss": 1.0788,
      "num_input_tokens_seen": 307244032,
      "step": 4110,
      "train_runtime": 4863.5523,
      "train_tokens_per_second": 63172.762
    },
    {
      "epoch": 0.05475046677430715,
      "grad_norm": 13.625,
      "learning_rate": 1.9967939678566865e-05,
      "loss": 1.084,
      "num_input_tokens_seen": 308141056,
      "step": 4120,
      "train_runtime": 4876.8561,
      "train_tokens_per_second": 63184.365
    },
    {
      "epoch": 0.05488335625676906,
      "grad_norm": 18.75,
      "learning_rate": 1.99675943901683e-05,
      "loss": 0.8535,
      "num_input_tokens_seen": 308638464,
      "step": 4130,
      "train_runtime": 4884.3221,
      "train_tokens_per_second": 63189.621
    },
    {
      "epoch": 0.055016245739230966,
      "grad_norm": 26.25,
      "learning_rate": 1.9967247255361758e-05,
      "loss": 0.7361,
      "num_input_tokens_seen": 308881920,
      "step": 4140,
      "train_runtime": 4887.8568,
      "train_tokens_per_second": 63193.733
    },
    {
      "epoch": 0.05514913522169288,
      "grad_norm": 13.5625,
      "learning_rate": 1.9966898274211536e-05,
      "loss": 0.6539,
      "num_input_tokens_seen": 308998912,
      "step": 4150,
      "train_runtime": 4890.2744,
      "train_tokens_per_second": 63186.415
    },
    {
      "epoch": 0.05528202470415479,
      "grad_norm": 21.625,
      "learning_rate": 1.9966547446782285e-05,
      "loss": 1.0917,
      "num_input_tokens_seen": 311012096,
      "step": 4160,
      "train_runtime": 4922.7216,
      "train_tokens_per_second": 63178.892
    },
    {
      "epoch": 0.0554149141866167,
      "grad_norm": 28.5,
      "learning_rate": 1.9966194773138988e-05,
      "loss": 1.1026,
      "num_input_tokens_seen": 311835648,
      "step": 4170,
      "train_runtime": 4935.1275,
      "train_tokens_per_second": 63186.948
    },
    {
      "epoch": 0.05554780366907861,
      "grad_norm": 34.0,
      "learning_rate": 1.9965840253346972e-05,
      "loss": 0.8978,
      "num_input_tokens_seen": 312274432,
      "step": 4180,
      "train_runtime": 4941.5713,
      "train_tokens_per_second": 63193.347
    },
    {
      "epoch": 0.05568069315154052,
      "grad_norm": 29.375,
      "learning_rate": 1.9965483887471917e-05,
      "loss": 0.7272,
      "num_input_tokens_seen": 312493312,
      "step": 4190,
      "train_runtime": 4944.9189,
      "train_tokens_per_second": 63194.831
    },
    {
      "epoch": 0.055813582634002434,
      "grad_norm": 9.4375,
      "learning_rate": 1.9965125675579833e-05,
      "loss": 0.596,
      "num_input_tokens_seen": 312607232,
      "step": 4200,
      "train_runtime": 4947.28,
      "train_tokens_per_second": 63187.697
    },
    {
      "epoch": 0.05594647211646434,
      "grad_norm": 23.625,
      "learning_rate": 1.9964765617737073e-05,
      "loss": 1.1209,
      "num_input_tokens_seen": 314681088,
      "step": 4210,
      "train_runtime": 4981.4393,
      "train_tokens_per_second": 63170.716
    },
    {
      "epoch": 0.05607936159892625,
      "grad_norm": 26.0,
      "learning_rate": 1.996440371401034e-05,
      "loss": 1.0652,
      "num_input_tokens_seen": 315543040,
      "step": 4220,
      "train_runtime": 4994.3703,
      "train_tokens_per_second": 63179.745
    },
    {
      "epoch": 0.056212251081388165,
      "grad_norm": 14.4375,
      "learning_rate": 1.9964039964466666e-05,
      "loss": 0.8338,
      "num_input_tokens_seen": 316023296,
      "step": 4230,
      "train_runtime": 5001.2534,
      "train_tokens_per_second": 63188.819
    },
    {
      "epoch": 0.05634514056385007,
      "grad_norm": 36.0,
      "learning_rate": 1.9963674369173437e-05,
      "loss": 0.7857,
      "num_input_tokens_seen": 316259840,
      "step": 4240,
      "train_runtime": 5004.7974,
      "train_tokens_per_second": 63191.337
    },
    {
      "epoch": 0.05647803004631199,
      "grad_norm": 31.0,
      "learning_rate": 1.9963306928198377e-05,
      "loss": 0.7141,
      "num_input_tokens_seen": 316375040,
      "step": 4250,
      "train_runtime": 5007.2266,
      "train_tokens_per_second": 63183.688
    },
    {
      "epoch": 0.056610919528773895,
      "grad_norm": 22.125,
      "learning_rate": 1.9962937641609547e-05,
      "loss": 1.0242,
      "num_input_tokens_seen": 318492416,
      "step": 4260,
      "train_runtime": 5040.8569,
      "train_tokens_per_second": 63182.197
    },
    {
      "epoch": 0.0567438090112358,
      "grad_norm": 19.375,
      "learning_rate": 1.9962566509475357e-05,
      "loss": 1.0332,
      "num_input_tokens_seen": 319312640,
      "step": 4270,
      "train_runtime": 5053.3961,
      "train_tokens_per_second": 63187.733
    },
    {
      "epoch": 0.05687669849369772,
      "grad_norm": 18.5,
      "learning_rate": 1.996219353186456e-05,
      "loss": 0.837,
      "num_input_tokens_seen": 319744000,
      "step": 4280,
      "train_runtime": 5059.8199,
      "train_tokens_per_second": 63192.763
    },
    {
      "epoch": 0.057009587976159626,
      "grad_norm": 15.125,
      "learning_rate": 1.9961818708846236e-05,
      "loss": 0.7652,
      "num_input_tokens_seen": 319970048,
      "step": 4290,
      "train_runtime": 5063.2379,
      "train_tokens_per_second": 63194.749
    },
    {
      "epoch": 0.05714247745862154,
      "grad_norm": 24.125,
      "learning_rate": 1.9961442040489828e-05,
      "loss": 0.5759,
      "num_input_tokens_seen": 320080896,
      "step": 4300,
      "train_runtime": 5065.6162,
      "train_tokens_per_second": 63186.962
    },
    {
      "epoch": 0.05727536694108345,
      "grad_norm": 18.25,
      "learning_rate": 1.9961063526865108e-05,
      "loss": 1.1627,
      "num_input_tokens_seen": 322080512,
      "step": 4310,
      "train_runtime": 5098.8365,
      "train_tokens_per_second": 63167.452
    },
    {
      "epoch": 0.057408256423545356,
      "grad_norm": 15.9375,
      "learning_rate": 1.996068316804219e-05,
      "loss": 1.0959,
      "num_input_tokens_seen": 322931712,
      "step": 4320,
      "train_runtime": 5111.4616,
      "train_tokens_per_second": 63177.959
    },
    {
      "epoch": 0.05754114590600727,
      "grad_norm": 17.625,
      "learning_rate": 1.996030096409153e-05,
      "loss": 0.8543,
      "num_input_tokens_seen": 323399168,
      "step": 4330,
      "train_runtime": 5118.0013,
      "train_tokens_per_second": 63188.567
    },
    {
      "epoch": 0.05767403538846918,
      "grad_norm": 29.75,
      "learning_rate": 1.9959916915083936e-05,
      "loss": 0.7748,
      "num_input_tokens_seen": 323617536,
      "step": 4340,
      "train_runtime": 5121.365,
      "train_tokens_per_second": 63189.704
    },
    {
      "epoch": 0.05780692487093109,
      "grad_norm": 10.6875,
      "learning_rate": 1.995953102109054e-05,
      "loss": 0.6315,
      "num_input_tokens_seen": 323730432,
      "step": 4350,
      "train_runtime": 5123.7441,
      "train_tokens_per_second": 63182.397
    },
    {
      "epoch": 0.057939814353393,
      "grad_norm": 28.25,
      "learning_rate": 1.995914328218283e-05,
      "loss": 1.0128,
      "num_input_tokens_seen": 325821952,
      "step": 4360,
      "train_runtime": 5159.1624,
      "train_tokens_per_second": 63154.041
    },
    {
      "epoch": 0.05807270383585491,
      "grad_norm": 40.25,
      "learning_rate": 1.9958753698432632e-05,
      "loss": 1.0874,
      "num_input_tokens_seen": 326684672,
      "step": 4370,
      "train_runtime": 5172.2373,
      "train_tokens_per_second": 63161.192
    },
    {
      "epoch": 0.058205593318316824,
      "grad_norm": 17.875,
      "learning_rate": 1.9958362269912112e-05,
      "loss": 0.885,
      "num_input_tokens_seen": 327186688,
      "step": 4380,
      "train_runtime": 5179.3143,
      "train_tokens_per_second": 63171.815
    },
    {
      "epoch": 0.05833848280077873,
      "grad_norm": 30.5,
      "learning_rate": 1.9957968996693776e-05,
      "loss": 0.7853,
      "num_input_tokens_seen": 327432192,
      "step": 4390,
      "train_runtime": 5183.8041,
      "train_tokens_per_second": 63164.461
    },
    {
      "epoch": 0.05847137228324064,
      "grad_norm": 26.75,
      "learning_rate": 1.995757387885048e-05,
      "loss": 0.6725,
      "num_input_tokens_seen": 327551488,
      "step": 4400,
      "train_runtime": 5186.3112,
      "train_tokens_per_second": 63156.929
    },
    {
      "epoch": 0.058604261765702555,
      "grad_norm": 52.0,
      "learning_rate": 1.9957176916455416e-05,
      "loss": 1.0709,
      "num_input_tokens_seen": 329540352,
      "step": 4410,
      "train_runtime": 5218.7383,
      "train_tokens_per_second": 63145.598
    },
    {
      "epoch": 0.05873715124816446,
      "grad_norm": 54.0,
      "learning_rate": 1.9956778109582107e-05,
      "loss": 1.1354,
      "num_input_tokens_seen": 330387456,
      "step": 4420,
      "train_runtime": 5231.5763,
      "train_tokens_per_second": 63152.564
    },
    {
      "epoch": 0.05887004073062638,
      "grad_norm": 19.75,
      "learning_rate": 1.995637745830444e-05,
      "loss": 0.9354,
      "num_input_tokens_seen": 330875648,
      "step": 4430,
      "train_runtime": 5238.5654,
      "train_tokens_per_second": 63161.5
    },
    {
      "epoch": 0.059002930213088285,
      "grad_norm": 23.5,
      "learning_rate": 1.995597496269663e-05,
      "loss": 0.7565,
      "num_input_tokens_seen": 331115008,
      "step": 4440,
      "train_runtime": 5242.1468,
      "train_tokens_per_second": 63164.009
    },
    {
      "epoch": 0.05913581969555019,
      "grad_norm": 22.375,
      "learning_rate": 1.9955570622833233e-05,
      "loss": 0.656,
      "num_input_tokens_seen": 331229952,
      "step": 4450,
      "train_runtime": 5245.1078,
      "train_tokens_per_second": 63150.265
    },
    {
      "epoch": 0.05926870917801211,
      "grad_norm": 27.75,
      "learning_rate": 1.995516443878915e-05,
      "loss": 1.0452,
      "num_input_tokens_seen": 333408768,
      "step": 4460,
      "train_runtime": 5280.1939,
      "train_tokens_per_second": 63143.281
    },
    {
      "epoch": 0.059401598660474016,
      "grad_norm": 14.75,
      "learning_rate": 1.995475641063962e-05,
      "loss": 1.0292,
      "num_input_tokens_seen": 334265344,
      "step": 4470,
      "train_runtime": 5293.2238,
      "train_tokens_per_second": 63149.671
    },
    {
      "epoch": 0.05953448814293593,
      "grad_norm": 18.75,
      "learning_rate": 1.9954346538460236e-05,
      "loss": 0.8874,
      "num_input_tokens_seen": 334754816,
      "step": 4480,
      "train_runtime": 5300.075,
      "train_tokens_per_second": 63160.392
    },
    {
      "epoch": 0.05966737762539784,
      "grad_norm": 13.4375,
      "learning_rate": 1.9953934822326912e-05,
      "loss": 0.6601,
      "num_input_tokens_seen": 334985728,
      "step": 4490,
      "train_runtime": 5303.5626,
      "train_tokens_per_second": 63162.397
    },
    {
      "epoch": 0.059800267107859746,
      "grad_norm": 16.625,
      "learning_rate": 1.995352126231592e-05,
      "loss": 0.6116,
      "num_input_tokens_seen": 335098880,
      "step": 4500,
      "train_runtime": 5306.5463,
      "train_tokens_per_second": 63148.206
    },
    {
      "epoch": 0.05993315659032166,
      "grad_norm": 21.75,
      "learning_rate": 1.9953105858503875e-05,
      "loss": 1.0241,
      "num_input_tokens_seen": 337079296,
      "step": 4510,
      "train_runtime": 5337.9443,
      "train_tokens_per_second": 63147.773
    },
    {
      "epoch": 0.06006604607278357,
      "grad_norm": 16.625,
      "learning_rate": 1.995268861096771e-05,
      "loss": 1.0091,
      "num_input_tokens_seen": 337891584,
      "step": 4520,
      "train_runtime": 5350.3257,
      "train_tokens_per_second": 63153.461
    },
    {
      "epoch": 0.060198935555245484,
      "grad_norm": 48.5,
      "learning_rate": 1.9952269519784733e-05,
      "loss": 0.829,
      "num_input_tokens_seen": 338355200,
      "step": 4530,
      "train_runtime": 5356.9968,
      "train_tokens_per_second": 63161.359
    },
    {
      "epoch": 0.06033182503770739,
      "grad_norm": 27.25,
      "learning_rate": 1.9951848585032567e-05,
      "loss": 0.7776,
      "num_input_tokens_seen": 338588928,
      "step": 4540,
      "train_runtime": 5360.5333,
      "train_tokens_per_second": 63163.292
    },
    {
      "epoch": 0.0604647145201693,
      "grad_norm": 9.8125,
      "learning_rate": 1.9951425806789194e-05,
      "loss": 0.5886,
      "num_input_tokens_seen": 338709760,
      "step": 4550,
      "train_runtime": 5363.0394,
      "train_tokens_per_second": 63156.307
    },
    {
      "epoch": 0.060597604002631214,
      "grad_norm": 44.0,
      "learning_rate": 1.995100118513292e-05,
      "loss": 1.0651,
      "num_input_tokens_seen": 340796160,
      "step": 4560,
      "train_runtime": 5397.9443,
      "train_tokens_per_second": 63134.435
    },
    {
      "epoch": 0.06073049348509312,
      "grad_norm": 36.5,
      "learning_rate": 1.995057472014241e-05,
      "loss": 1.0532,
      "num_input_tokens_seen": 341639168,
      "step": 4570,
      "train_runtime": 5410.8621,
      "train_tokens_per_second": 63139.508
    },
    {
      "epoch": 0.06086338296755503,
      "grad_norm": 24.75,
      "learning_rate": 1.9950146411896662e-05,
      "loss": 0.8628,
      "num_input_tokens_seen": 342116352,
      "step": 4580,
      "train_runtime": 5417.4714,
      "train_tokens_per_second": 63150.56
    },
    {
      "epoch": 0.060996272450016945,
      "grad_norm": 16.875,
      "learning_rate": 1.9949716260475015e-05,
      "loss": 0.7917,
      "num_input_tokens_seen": 342342144,
      "step": 4590,
      "train_runtime": 5420.9254,
      "train_tokens_per_second": 63151.975
    },
    {
      "epoch": 0.06112916193247885,
      "grad_norm": 26.25,
      "learning_rate": 1.9949284265957146e-05,
      "loss": 0.6031,
      "num_input_tokens_seen": 342457344,
      "step": 4600,
      "train_runtime": 5423.3516,
      "train_tokens_per_second": 63144.965
    },
    {
      "epoch": 0.06126205141494077,
      "grad_norm": 39.25,
      "learning_rate": 1.994885042842309e-05,
      "loss": 1.0729,
      "num_input_tokens_seen": 344426496,
      "step": 4610,
      "train_runtime": 5456.5999,
      "train_tokens_per_second": 63121.084
    },
    {
      "epoch": 0.061394940897402675,
      "grad_norm": 15.25,
      "learning_rate": 1.99484147479532e-05,
      "loss": 1.0639,
      "num_input_tokens_seen": 345278208,
      "step": 4620,
      "train_runtime": 5469.3217,
      "train_tokens_per_second": 63129.987
    },
    {
      "epoch": 0.06152783037986458,
      "grad_norm": 48.25,
      "learning_rate": 1.9947977224628187e-05,
      "loss": 0.9273,
      "num_input_tokens_seen": 345742592,
      "step": 4630,
      "train_runtime": 5476.2985,
      "train_tokens_per_second": 63134.358
    },
    {
      "epoch": 0.0616607198623265,
      "grad_norm": 26.25,
      "learning_rate": 1.9947537858529098e-05,
      "loss": 0.7801,
      "num_input_tokens_seen": 345978880,
      "step": 4640,
      "train_runtime": 5479.756,
      "train_tokens_per_second": 63137.643
    },
    {
      "epoch": 0.061793609344788406,
      "grad_norm": 20.625,
      "learning_rate": 1.994709664973732e-05,
      "loss": 0.601,
      "num_input_tokens_seen": 346095360,
      "step": 4650,
      "train_runtime": 5482.2424,
      "train_tokens_per_second": 63130.255
    },
    {
      "epoch": 0.06192649882725032,
      "grad_norm": 34.25,
      "learning_rate": 1.9946653598334585e-05,
      "loss": 1.0417,
      "num_input_tokens_seen": 348218624,
      "step": 4660,
      "train_runtime": 5517.0868,
      "train_tokens_per_second": 63116.394
    },
    {
      "epoch": 0.06205938830971223,
      "grad_norm": 12.375,
      "learning_rate": 1.9946208704402964e-05,
      "loss": 1.1296,
      "num_input_tokens_seen": 349096448,
      "step": 4670,
      "train_runtime": 5530.9133,
      "train_tokens_per_second": 63117.324
    },
    {
      "epoch": 0.062192277792174136,
      "grad_norm": 14.625,
      "learning_rate": 1.994576196802487e-05,
      "loss": 0.9982,
      "num_input_tokens_seen": 349565696,
      "step": 4680,
      "train_runtime": 5537.6144,
      "train_tokens_per_second": 63125.684
    },
    {
      "epoch": 0.06232516727463605,
      "grad_norm": 19.0,
      "learning_rate": 1.9945313389283053e-05,
      "loss": 0.7032,
      "num_input_tokens_seen": 349801728,
      "step": 4690,
      "train_runtime": 5541.1728,
      "train_tokens_per_second": 63127.742
    },
    {
      "epoch": 0.06245805675709796,
      "grad_norm": 16.125,
      "learning_rate": 1.9944862968260617e-05,
      "loss": 0.6008,
      "num_input_tokens_seen": 349921536,
      "step": 4700,
      "train_runtime": 5543.5605,
      "train_tokens_per_second": 63122.164
    },
    {
      "epoch": 0.06259094623955987,
      "grad_norm": 11.9375,
      "learning_rate": 1.9944410705040988e-05,
      "loss": 1.0516,
      "num_input_tokens_seen": 351840000,
      "step": 4710,
      "train_runtime": 5574.1817,
      "train_tokens_per_second": 63119.578
    },
    {
      "epoch": 0.06272383572202178,
      "grad_norm": 22.0,
      "learning_rate": 1.994395659970795e-05,
      "loss": 1.0795,
      "num_input_tokens_seen": 352700160,
      "step": 4720,
      "train_runtime": 5586.7099,
      "train_tokens_per_second": 63131.998
    },
    {
      "epoch": 0.0628567252044837,
      "grad_norm": 14.125,
      "learning_rate": 1.9943500652345622e-05,
      "loss": 0.8563,
      "num_input_tokens_seen": 353164032,
      "step": 4730,
      "train_runtime": 5593.1615,
      "train_tokens_per_second": 63142.112
    },
    {
      "epoch": 0.0629896146869456,
      "grad_norm": 14.375,
      "learning_rate": 1.9943042863038462e-05,
      "loss": 0.6708,
      "num_input_tokens_seen": 353388032,
      "step": 4740,
      "train_runtime": 5596.5634,
      "train_tokens_per_second": 63143.755
    },
    {
      "epoch": 0.06312250416940751,
      "grad_norm": 11.0625,
      "learning_rate": 1.994258323187127e-05,
      "loss": 0.6112,
      "num_input_tokens_seen": 353499648,
      "step": 4750,
      "train_runtime": 5599.4648,
      "train_tokens_per_second": 63130.971
    },
    {
      "epoch": 0.06325539365186943,
      "grad_norm": 15.625,
      "learning_rate": 1.9942121758929192e-05,
      "loss": 1.0647,
      "num_input_tokens_seen": 355536128,
      "step": 4760,
      "train_runtime": 5631.2734,
      "train_tokens_per_second": 63136.009
    },
    {
      "epoch": 0.06338828313433133,
      "grad_norm": 14.125,
      "learning_rate": 1.994165844429771e-05,
      "loss": 1.0213,
      "num_input_tokens_seen": 356375552,
      "step": 4770,
      "train_runtime": 5644.0199,
      "train_tokens_per_second": 63142.15
    },
    {
      "epoch": 0.06352117261679324,
      "grad_norm": 19.5,
      "learning_rate": 1.994119328806265e-05,
      "loss": 0.8583,
      "num_input_tokens_seen": 356868352,
      "step": 4780,
      "train_runtime": 5650.7776,
      "train_tokens_per_second": 63153.848
    },
    {
      "epoch": 0.06365406209925516,
      "grad_norm": 25.25,
      "learning_rate": 1.9940726290310176e-05,
      "loss": 0.7085,
      "num_input_tokens_seen": 357094144,
      "step": 4790,
      "train_runtime": 5654.3514,
      "train_tokens_per_second": 63153.865
    },
    {
      "epoch": 0.06378695158171707,
      "grad_norm": 16.375,
      "learning_rate": 1.99402574511268e-05,
      "loss": 0.6494,
      "num_input_tokens_seen": 357211648,
      "step": 4800,
      "train_runtime": 5656.7491,
      "train_tokens_per_second": 63147.868
    },
    {
      "epoch": 0.06391984106417897,
      "grad_norm": 13.5625,
      "learning_rate": 1.9939786770599364e-05,
      "loss": 1.0087,
      "num_input_tokens_seen": 359361280,
      "step": 4810,
      "train_runtime": 5691.3596,
      "train_tokens_per_second": 63141.552
    },
    {
      "epoch": 0.06405273054664089,
      "grad_norm": 33.0,
      "learning_rate": 1.993931424881506e-05,
      "loss": 1.0794,
      "num_input_tokens_seen": 360216832,
      "step": 4820,
      "train_runtime": 5704.2067,
      "train_tokens_per_second": 63149.33
    },
    {
      "epoch": 0.0641856200291028,
      "grad_norm": 30.75,
      "learning_rate": 1.993883988586142e-05,
      "loss": 0.9117,
      "num_input_tokens_seen": 360686848,
      "step": 4830,
      "train_runtime": 5711.0569,
      "train_tokens_per_second": 63155.884
    },
    {
      "epoch": 0.0643185095115647,
      "grad_norm": 37.75,
      "learning_rate": 1.9938363681826317e-05,
      "loss": 0.8634,
      "num_input_tokens_seen": 360915200,
      "step": 4840,
      "train_runtime": 5714.5313,
      "train_tokens_per_second": 63157.446
    },
    {
      "epoch": 0.06445139899402662,
      "grad_norm": 15.5625,
      "learning_rate": 1.9937885636797953e-05,
      "loss": 0.5454,
      "num_input_tokens_seen": 361030144,
      "step": 4850,
      "train_runtime": 5716.9725,
      "train_tokens_per_second": 63150.583
    },
    {
      "epoch": 0.06458428847648853,
      "grad_norm": 16.375,
      "learning_rate": 1.9937405750864896e-05,
      "loss": 1.011,
      "num_input_tokens_seen": 363065600,
      "step": 4860,
      "train_runtime": 5751.7343,
      "train_tokens_per_second": 63122.805
    },
    {
      "epoch": 0.06471717795895043,
      "grad_norm": 46.25,
      "learning_rate": 1.9936924024116032e-05,
      "loss": 1.0818,
      "num_input_tokens_seen": 363919104,
      "step": 4870,
      "train_runtime": 5764.6606,
      "train_tokens_per_second": 63129.32
    },
    {
      "epoch": 0.06485006744141235,
      "grad_norm": 15.5625,
      "learning_rate": 1.9936440456640595e-05,
      "loss": 0.8356,
      "num_input_tokens_seen": 364423680,
      "step": 4880,
      "train_runtime": 5771.5818,
      "train_tokens_per_second": 63141.041
    },
    {
      "epoch": 0.06498295692387426,
      "grad_norm": 15.8125,
      "learning_rate": 1.993595504852817e-05,
      "loss": 0.7981,
      "num_input_tokens_seen": 364665344,
      "step": 4890,
      "train_runtime": 5775.2309,
      "train_tokens_per_second": 63142.989
    },
    {
      "epoch": 0.06511584640633616,
      "grad_norm": 14.0,
      "learning_rate": 1.9935467799868668e-05,
      "loss": 0.6504,
      "num_input_tokens_seen": 364784128,
      "step": 4900,
      "train_runtime": 5777.7654,
      "train_tokens_per_second": 63135.85
    },
    {
      "epoch": 0.06524873588879808,
      "grad_norm": 19.0,
      "learning_rate": 1.9934978710752347e-05,
      "loss": 1.0381,
      "num_input_tokens_seen": 366792960,
      "step": 4910,
      "train_runtime": 5812.0217,
      "train_tokens_per_second": 63109.358
    },
    {
      "epoch": 0.06538162537126,
      "grad_norm": 22.125,
      "learning_rate": 1.9934487781269812e-05,
      "loss": 1.0763,
      "num_input_tokens_seen": 367683584,
      "step": 4920,
      "train_runtime": 5825.1865,
      "train_tokens_per_second": 63119.624
    },
    {
      "epoch": 0.06551451485372191,
      "grad_norm": 16.5,
      "learning_rate": 1.9933995011511997e-05,
      "loss": 0.8631,
      "num_input_tokens_seen": 368176384,
      "step": 4930,
      "train_runtime": 5832.6184,
      "train_tokens_per_second": 63123.688
    },
    {
      "epoch": 0.06564740433618381,
      "grad_norm": 29.625,
      "learning_rate": 1.9933500401570184e-05,
      "loss": 0.7677,
      "num_input_tokens_seen": 368409600,
      "step": 4940,
      "train_runtime": 5836.1514,
      "train_tokens_per_second": 63125.436
    },
    {
      "epoch": 0.06578029381864572,
      "grad_norm": 9.75,
      "learning_rate": 1.9933003951536e-05,
      "loss": 0.5892,
      "num_input_tokens_seen": 368526592,
      "step": 4950,
      "train_runtime": 5838.5484,
      "train_tokens_per_second": 63119.557
    },
    {
      "epoch": 0.06591318330110764,
      "grad_norm": 44.75,
      "learning_rate": 1.9932505661501405e-05,
      "loss": 1.072,
      "num_input_tokens_seen": 370614016,
      "step": 4960,
      "train_runtime": 5871.8381,
      "train_tokens_per_second": 63117.206
    },
    {
      "epoch": 0.06604607278356954,
      "grad_norm": 24.375,
      "learning_rate": 1.9932005531558703e-05,
      "loss": 1.0137,
      "num_input_tokens_seen": 371447552,
      "step": 4970,
      "train_runtime": 5884.141,
      "train_tokens_per_second": 63126.895
    },
    {
      "epoch": 0.06617896226603145,
      "grad_norm": 63.5,
      "learning_rate": 1.9931503561800537e-05,
      "loss": 0.8674,
      "num_input_tokens_seen": 371899904,
      "step": 4980,
      "train_runtime": 5891.086,
      "train_tokens_per_second": 63129.261
    },
    {
      "epoch": 0.06631185174849337,
      "grad_norm": 18.125,
      "learning_rate": 1.9930999752319895e-05,
      "loss": 0.7761,
      "num_input_tokens_seen": 372124928,
      "step": 4990,
      "train_runtime": 5894.5056,
      "train_tokens_per_second": 63130.812
    },
    {
      "epoch": 0.06644474123095527,
      "grad_norm": 11.4375,
      "learning_rate": 1.99304941032101e-05,
      "loss": 0.6221,
      "num_input_tokens_seen": 372242688,
      "step": 5000,
      "train_runtime": 5896.9359,
      "train_tokens_per_second": 63124.764
    },
    {
      "epoch": 0.06644474123095527,
      "eval_accuracy": 0.7847,
      "eval_balanced_accuracy": 0.785803704949524,
      "eval_invalid_f1": 0.800629687934068,
      "eval_invalid_precision": 0.8267355134825014,
      "eval_invalid_recall": 0.7761220825852783,
      "eval_loss": 0.42647016048431396,
      "eval_macro_f1": 0.7833166915922921,
      "eval_mcc": 0.5684775016844967,
      "eval_runtime": 78.5512,
      "eval_samples_per_second": 127.305,
      "eval_steps_per_second": 31.826,
      "eval_valid_f1": 0.7660036952505163,
      "eval_valid_precision": 0.7386292181932509,
      "eval_valid_recall": 0.7954853273137698,
      "num_input_tokens_seen": 372242688,
      "step": 5000
    },
    {
      "epoch": 0.06657763071341719,
      "grad_norm": 46.75,
      "learning_rate": 1.992998661456482e-05,
      "loss": 1.026,
      "num_input_tokens_seen": 374180096,
      "step": 5010,
      "train_runtime": 6011.3427,
      "train_tokens_per_second": 62245.677
    },
    {
      "epoch": 0.0667105201958791,
      "grad_norm": 13.9375,
      "learning_rate": 1.9929477286478066e-05,
      "loss": 1.06,
      "num_input_tokens_seen": 375010816,
      "step": 5020,
      "train_runtime": 6024.0934,
      "train_tokens_per_second": 62251.826
    },
    {
      "epoch": 0.066843409678341,
      "grad_norm": 20.75,
      "learning_rate": 1.9928966119044188e-05,
      "loss": 0.9005,
      "num_input_tokens_seen": 375461888,
      "step": 5030,
      "train_runtime": 6030.6102,
      "train_tokens_per_second": 62259.353
    },
    {
      "epoch": 0.06697629916080292,
      "grad_norm": 79.0,
      "learning_rate": 1.9928453112357866e-05,
      "loss": 0.706,
      "num_input_tokens_seen": 375687424,
      "step": 5040,
      "train_runtime": 6033.9724,
      "train_tokens_per_second": 62262.039
    },
    {
      "epoch": 0.06710918864326483,
      "grad_norm": 17.75,
      "learning_rate": 1.9927938266514136e-05,
      "loss": 0.5898,
      "num_input_tokens_seen": 375803904,
      "step": 5050,
      "train_runtime": 6036.3897,
      "train_tokens_per_second": 62256.403
    },
    {
      "epoch": 0.06724207812572675,
      "grad_norm": 18.125,
      "learning_rate": 1.992742158160837e-05,
      "loss": 1.0316,
      "num_input_tokens_seen": 377862656,
      "step": 5060,
      "train_runtime": 6070.5792,
      "train_tokens_per_second": 62244.91
    },
    {
      "epoch": 0.06737496760818865,
      "grad_norm": 26.75,
      "learning_rate": 1.9926903057736275e-05,
      "loss": 1.012,
      "num_input_tokens_seen": 378685184,
      "step": 5070,
      "train_runtime": 6083.1775,
      "train_tokens_per_second": 62251.214
    },
    {
      "epoch": 0.06750785709065056,
      "grad_norm": 24.5,
      "learning_rate": 1.9926382694993908e-05,
      "loss": 0.8224,
      "num_input_tokens_seen": 379137536,
      "step": 5080,
      "train_runtime": 6089.9999,
      "train_tokens_per_second": 62255.754
    },
    {
      "epoch": 0.06764074657311248,
      "grad_norm": 146.0,
      "learning_rate": 1.9925860493477654e-05,
      "loss": 0.7206,
      "num_input_tokens_seen": 379362816,
      "step": 5090,
      "train_runtime": 6093.4237,
      "train_tokens_per_second": 62257.745
    },
    {
      "epoch": 0.06777363605557438,
      "grad_norm": 16.125,
      "learning_rate": 1.9925336453284253e-05,
      "loss": 0.598,
      "num_input_tokens_seen": 379476736,
      "step": 5100,
      "train_runtime": 6095.8537,
      "train_tokens_per_second": 62251.615
    },
    {
      "epoch": 0.06790652553803629,
      "grad_norm": 44.0,
      "learning_rate": 1.992481057451078e-05,
      "loss": 1.0367,
      "num_input_tokens_seen": 381477376,
      "step": 5110,
      "train_runtime": 6128.3317,
      "train_tokens_per_second": 62248.161
    },
    {
      "epoch": 0.0680394150204982,
      "grad_norm": 21.0,
      "learning_rate": 1.9924282857254642e-05,
      "loss": 1.0966,
      "num_input_tokens_seen": 382301184,
      "step": 5120,
      "train_runtime": 6140.6364,
      "train_tokens_per_second": 62257.583
    },
    {
      "epoch": 0.06817230450296011,
      "grad_norm": 34.25,
      "learning_rate": 1.9923753301613598e-05,
      "loss": 0.8748,
      "num_input_tokens_seen": 382769152,
      "step": 5130,
      "train_runtime": 6147.3939,
      "train_tokens_per_second": 62265.272
    },
    {
      "epoch": 0.06830519398542202,
      "grad_norm": 12.125,
      "learning_rate": 1.9923221907685744e-05,
      "loss": 0.7043,
      "num_input_tokens_seen": 382988800,
      "step": 5140,
      "train_runtime": 6150.7631,
      "train_tokens_per_second": 62266.875
    },
    {
      "epoch": 0.06843808346788394,
      "grad_norm": 10.1875,
      "learning_rate": 1.9922688675569517e-05,
      "loss": 0.5457,
      "num_input_tokens_seen": 383103488,
      "step": 5150,
      "train_runtime": 6153.1953,
      "train_tokens_per_second": 62260.902
    },
    {
      "epoch": 0.06857097295034585,
      "grad_norm": 26.875,
      "learning_rate": 1.992215360536369e-05,
      "loss": 1.057,
      "num_input_tokens_seen": 385143040,
      "step": 5160,
      "train_runtime": 6185.9865,
      "train_tokens_per_second": 62260.569
    },
    {
      "epoch": 0.06870386243280775,
      "grad_norm": 37.5,
      "learning_rate": 1.9921616697167384e-05,
      "loss": 1.0723,
      "num_input_tokens_seen": 385991936,
      "step": 5170,
      "train_runtime": 6198.7424,
      "train_tokens_per_second": 62269.395
    },
    {
      "epoch": 0.06883675191526967,
      "grad_norm": 24.75,
      "learning_rate": 1.9921077951080054e-05,
      "loss": 0.8419,
      "num_input_tokens_seen": 386475264,
      "step": 5180,
      "train_runtime": 6205.5773,
      "train_tokens_per_second": 62278.696
    },
    {
      "epoch": 0.06896964139773158,
      "grad_norm": 20.625,
      "learning_rate": 1.9920537367201493e-05,
      "loss": 0.7108,
      "num_input_tokens_seen": 386711296,
      "step": 5190,
      "train_runtime": 6209.0955,
      "train_tokens_per_second": 62281.422
    },
    {
      "epoch": 0.06910253088019348,
      "grad_norm": 16.125,
      "learning_rate": 1.991999494563185e-05,
      "loss": 0.5635,
      "num_input_tokens_seen": 386827776,
      "step": 5200,
      "train_runtime": 6211.4902,
      "train_tokens_per_second": 62276.163
    },
    {
      "epoch": 0.0692354203626554,
      "grad_norm": 14.5,
      "learning_rate": 1.9919450686471595e-05,
      "loss": 1.0191,
      "num_input_tokens_seen": 388896768,
      "step": 5210,
      "train_runtime": 6244.4195,
      "train_tokens_per_second": 62279.091
    },
    {
      "epoch": 0.06936830984511731,
      "grad_norm": 12.375,
      "learning_rate": 1.991890458982155e-05,
      "loss": 1.0371,
      "num_input_tokens_seen": 389735680,
      "step": 5220,
      "train_runtime": 6257.263,
      "train_tokens_per_second": 62285.329
    },
    {
      "epoch": 0.06950119932757921,
      "grad_norm": 22.875,
      "learning_rate": 1.991835665578288e-05,
      "loss": 0.925,
      "num_input_tokens_seen": 390198528,
      "step": 5230,
      "train_runtime": 6264.1584,
      "train_tokens_per_second": 62290.655
    },
    {
      "epoch": 0.06963408881004113,
      "grad_norm": 21.375,
      "learning_rate": 1.9917806884457076e-05,
      "loss": 0.7298,
      "num_input_tokens_seen": 390437632,
      "step": 5240,
      "train_runtime": 6267.647,
      "train_tokens_per_second": 62294.133
    },
    {
      "epoch": 0.06976697829250304,
      "grad_norm": 14.875,
      "learning_rate": 1.9917255275945984e-05,
      "loss": 0.5878,
      "num_input_tokens_seen": 390556928,
      "step": 5250,
      "train_runtime": 6270.0323,
      "train_tokens_per_second": 62289.46
    },
    {
      "epoch": 0.06989986777496494,
      "grad_norm": 51.5,
      "learning_rate": 1.9916701830351783e-05,
      "loss": 1.0406,
      "num_input_tokens_seen": 392572416,
      "step": 5260,
      "train_runtime": 6302.4354,
      "train_tokens_per_second": 62289.003
    },
    {
      "epoch": 0.07003275725742686,
      "grad_norm": 18.375,
      "learning_rate": 1.991614654777699e-05,
      "loss": 1.0655,
      "num_input_tokens_seen": 393385472,
      "step": 5270,
      "train_runtime": 6315.1611,
      "train_tokens_per_second": 62292.231
    },
    {
      "epoch": 0.07016564673988877,
      "grad_norm": 17.5,
      "learning_rate": 1.9915589428324473e-05,
      "loss": 0.8268,
      "num_input_tokens_seen": 393860608,
      "step": 5280,
      "train_runtime": 6321.9348,
      "train_tokens_per_second": 62300.644
    },
    {
      "epoch": 0.07029853622235069,
      "grad_norm": 25.625,
      "learning_rate": 1.991503047209743e-05,
      "loss": 0.7841,
      "num_input_tokens_seen": 394099456,
      "step": 5290,
      "train_runtime": 6325.4246,
      "train_tokens_per_second": 62304.032
    },
    {
      "epoch": 0.07043142570481259,
      "grad_norm": 36.0,
      "learning_rate": 1.9914469679199402e-05,
      "loss": 0.6713,
      "num_input_tokens_seen": 394211072,
      "step": 5300,
      "train_runtime": 6327.748,
      "train_tokens_per_second": 62298.795
    },
    {
      "epoch": 0.0705643151872745,
      "grad_norm": 14.1875,
      "learning_rate": 1.991390704973427e-05,
      "loss": 1.0649,
      "num_input_tokens_seen": 396218112,
      "step": 5310,
      "train_runtime": 6361.3397,
      "train_tokens_per_second": 62285.325
    },
    {
      "epoch": 0.07069720466973642,
      "grad_norm": 26.125,
      "learning_rate": 1.9913342583806257e-05,
      "loss": 1.0075,
      "num_input_tokens_seen": 397040384,
      "step": 5320,
      "train_runtime": 6374.0931,
      "train_tokens_per_second": 62289.706
    },
    {
      "epoch": 0.07083009415219832,
      "grad_norm": 88.5,
      "learning_rate": 1.991277628151993e-05,
      "loss": 0.8463,
      "num_input_tokens_seen": 397506560,
      "step": 5330,
      "train_runtime": 6380.649,
      "train_tokens_per_second": 62298.766
    },
    {
      "epoch": 0.07096298363466023,
      "grad_norm": 19.75,
      "learning_rate": 1.991220814298019e-05,
      "loss": 0.7151,
      "num_input_tokens_seen": 397731584,
      "step": 5340,
      "train_runtime": 6384.0418,
      "train_tokens_per_second": 62300.906
    },
    {
      "epoch": 0.07109587311712215,
      "grad_norm": 41.5,
      "learning_rate": 1.991163816829227e-05,
      "loss": 0.5532,
      "num_input_tokens_seen": 397843712,
      "step": 5350,
      "train_runtime": 6386.4674,
      "train_tokens_per_second": 62294.801
    },
    {
      "epoch": 0.07122876259958405,
      "grad_norm": 17.75,
      "learning_rate": 1.9911066357561762e-05,
      "loss": 0.9858,
      "num_input_tokens_seen": 399900160,
      "step": 5360,
      "train_runtime": 6419.5868,
      "train_tokens_per_second": 62293.754
    },
    {
      "epoch": 0.07136165208204596,
      "grad_norm": 18.25,
      "learning_rate": 1.9910492710894583e-05,
      "loss": 0.972,
      "num_input_tokens_seen": 400736512,
      "step": 5370,
      "train_runtime": 6432.4168,
      "train_tokens_per_second": 62299.525
    },
    {
      "epoch": 0.07149454156450788,
      "grad_norm": 33.25,
      "learning_rate": 1.9909917228397003e-05,
      "loss": 0.8387,
      "num_input_tokens_seen": 401197824,
      "step": 5380,
      "train_runtime": 6439.0427,
      "train_tokens_per_second": 62307.061
    },
    {
      "epoch": 0.0716274310469698,
      "grad_norm": 14.0625,
      "learning_rate": 1.9909339910175624e-05,
      "loss": 0.6574,
      "num_input_tokens_seen": 401429504,
      "step": 5390,
      "train_runtime": 6442.5235,
      "train_tokens_per_second": 62309.358
    },
    {
      "epoch": 0.0717603205294317,
      "grad_norm": 13.4375,
      "learning_rate": 1.9908760756337388e-05,
      "loss": 0.5894,
      "num_input_tokens_seen": 401539584,
      "step": 5400,
      "train_runtime": 6444.8579,
      "train_tokens_per_second": 62303.869
    },
    {
      "epoch": 0.07189321001189361,
      "grad_norm": 22.625,
      "learning_rate": 1.990817976698957e-05,
      "loss": 1.0205,
      "num_input_tokens_seen": 403558656,
      "step": 5410,
      "train_runtime": 6478.2847,
      "train_tokens_per_second": 62294.06
    },
    {
      "epoch": 0.07202609949435552,
      "grad_norm": 19.5,
      "learning_rate": 1.9907596942239804e-05,
      "loss": 1.1322,
      "num_input_tokens_seen": 404391424,
      "step": 5420,
      "train_runtime": 6490.9323,
      "train_tokens_per_second": 62300.977
    },
    {
      "epoch": 0.07215898897681743,
      "grad_norm": 19.75,
      "learning_rate": 1.9907012282196047e-05,
      "loss": 0.7928,
      "num_input_tokens_seen": 404863744,
      "step": 5430,
      "train_runtime": 6497.6506,
      "train_tokens_per_second": 62309.251
    },
    {
      "epoch": 0.07229187845927934,
      "grad_norm": 17.75,
      "learning_rate": 1.9906425786966607e-05,
      "loss": 0.6782,
      "num_input_tokens_seen": 405086208,
      "step": 5440,
      "train_runtime": 6501.4144,
      "train_tokens_per_second": 62307.397
    },
    {
      "epoch": 0.07242476794174126,
      "grad_norm": 17.0,
      "learning_rate": 1.990583745666012e-05,
      "loss": 0.6016,
      "num_input_tokens_seen": 405199616,
      "step": 5450,
      "train_runtime": 6503.7745,
      "train_tokens_per_second": 62302.224
    },
    {
      "epoch": 0.07255765742420316,
      "grad_norm": 27.375,
      "learning_rate": 1.9905247291385574e-05,
      "loss": 1.0265,
      "num_input_tokens_seen": 407165952,
      "step": 5460,
      "train_runtime": 6535.1119,
      "train_tokens_per_second": 62304.358
    },
    {
      "epoch": 0.07269054690666507,
      "grad_norm": 23.125,
      "learning_rate": 1.9904655291252292e-05,
      "loss": 1.1288,
      "num_input_tokens_seen": 408012288,
      "step": 5470,
      "train_runtime": 6547.9554,
      "train_tokens_per_second": 62311.403
    },
    {
      "epoch": 0.07282343638912699,
      "grad_norm": 33.5,
      "learning_rate": 1.9904061456369935e-05,
      "loss": 0.8106,
      "num_input_tokens_seen": 408487936,
      "step": 5480,
      "train_runtime": 6554.8725,
      "train_tokens_per_second": 62318.212
    },
    {
      "epoch": 0.07295632587158889,
      "grad_norm": 16.125,
      "learning_rate": 1.9903465786848506e-05,
      "loss": 0.7227,
      "num_input_tokens_seen": 408724736,
      "step": 5490,
      "train_runtime": 6558.7858,
      "train_tokens_per_second": 62317.135
    },
    {
      "epoch": 0.0730892153540508,
      "grad_norm": 15.75,
      "learning_rate": 1.990286828279835e-05,
      "loss": 0.5736,
      "num_input_tokens_seen": 408839680,
      "step": 5500,
      "train_runtime": 6561.177,
      "train_tokens_per_second": 62311.942
    },
    {
      "epoch": 0.07322210483651272,
      "grad_norm": 31.25,
      "learning_rate": 1.9902268944330145e-05,
      "loss": 1.032,
      "num_input_tokens_seen": 410918400,
      "step": 5510,
      "train_runtime": 6593.7925,
      "train_tokens_per_second": 62318.976
    },
    {
      "epoch": 0.07335499431897463,
      "grad_norm": 23.625,
      "learning_rate": 1.990166777155491e-05,
      "loss": 1.0889,
      "num_input_tokens_seen": 411761920,
      "step": 5520,
      "train_runtime": 6606.6529,
      "train_tokens_per_second": 62325.345
    },
    {
      "epoch": 0.07348788380143653,
      "grad_norm": 43.25,
      "learning_rate": 1.9901064764584017e-05,
      "loss": 0.8717,
      "num_input_tokens_seen": 412259584,
      "step": 5530,
      "train_runtime": 6613.5665,
      "train_tokens_per_second": 62335.441
    },
    {
      "epoch": 0.07362077328389845,
      "grad_norm": 12.875,
      "learning_rate": 1.990045992352916e-05,
      "loss": 0.7543,
      "num_input_tokens_seen": 412492288,
      "step": 5540,
      "train_runtime": 6617.4128,
      "train_tokens_per_second": 62334.374
    },
    {
      "epoch": 0.07375366276636036,
      "grad_norm": 270.0,
      "learning_rate": 1.9899853248502388e-05,
      "loss": 0.5828,
      "num_input_tokens_seen": 412610048,
      "step": 5550,
      "train_runtime": 6619.8119,
      "train_tokens_per_second": 62329.573
    },
    {
      "epoch": 0.07388655224882226,
      "grad_norm": 15.5625,
      "learning_rate": 1.9899244739616072e-05,
      "loss": 1.0501,
      "num_input_tokens_seen": 414838016,
      "step": 5560,
      "train_runtime": 6654.2198,
      "train_tokens_per_second": 62342.097
    },
    {
      "epoch": 0.07401944173128418,
      "grad_norm": 11.75,
      "learning_rate": 1.9898634396982942e-05,
      "loss": 1.0414,
      "num_input_tokens_seen": 415681792,
      "step": 5570,
      "train_runtime": 6667.1907,
      "train_tokens_per_second": 62347.368
    },
    {
      "epoch": 0.07415233121374609,
      "grad_norm": 20.125,
      "learning_rate": 1.989802222071605e-05,
      "loss": 0.872,
      "num_input_tokens_seen": 416151552,
      "step": 5580,
      "train_runtime": 6674.0457,
      "train_tokens_per_second": 62353.717
    },
    {
      "epoch": 0.074285220696208,
      "grad_norm": 29.875,
      "learning_rate": 1.9897408210928804e-05,
      "loss": 0.6899,
      "num_input_tokens_seen": 416381440,
      "step": 5590,
      "train_runtime": 6677.4886,
      "train_tokens_per_second": 62355.994
    },
    {
      "epoch": 0.07441811017866991,
      "grad_norm": 13.4375,
      "learning_rate": 1.989679236773494e-05,
      "loss": 0.5555,
      "num_input_tokens_seen": 416495616,
      "step": 5600,
      "train_runtime": 6680.2289,
      "train_tokens_per_second": 62347.506
    },
    {
      "epoch": 0.07455099966113182,
      "grad_norm": 59.5,
      "learning_rate": 1.9896174691248535e-05,
      "loss": 1.0667,
      "num_input_tokens_seen": 418483712,
      "step": 5610,
      "train_runtime": 6711.5472,
      "train_tokens_per_second": 62352.793
    },
    {
      "epoch": 0.07468388914359372,
      "grad_norm": 40.75,
      "learning_rate": 1.9895555181584013e-05,
      "loss": 1.035,
      "num_input_tokens_seen": 419323136,
      "step": 5620,
      "train_runtime": 6723.8954,
      "train_tokens_per_second": 62363.126
    },
    {
      "epoch": 0.07481677862605564,
      "grad_norm": 16.875,
      "learning_rate": 1.9894933838856133e-05,
      "loss": 0.8781,
      "num_input_tokens_seen": 419766784,
      "step": 5630,
      "train_runtime": 6730.4827,
      "train_tokens_per_second": 62368.006
    },
    {
      "epoch": 0.07494966810851755,
      "grad_norm": 18.75,
      "learning_rate": 1.9894310663179985e-05,
      "loss": 0.7743,
      "num_input_tokens_seen": 419995136,
      "step": 5640,
      "train_runtime": 6733.9399,
      "train_tokens_per_second": 62369.897
    },
    {
      "epoch": 0.07508255759097947,
      "grad_norm": 25.5,
      "learning_rate": 1.9893685654671017e-05,
      "loss": 0.575,
      "num_input_tokens_seen": 420110336,
      "step": 5650,
      "train_runtime": 6736.6841,
      "train_tokens_per_second": 62361.591
    },
    {
      "epoch": 0.07521544707344137,
      "grad_norm": 24.875,
      "learning_rate": 1.9893058813445e-05,
      "loss": 0.973,
      "num_input_tokens_seen": 422163200,
      "step": 5660,
      "train_runtime": 6770.7293,
      "train_tokens_per_second": 62351.215
    },
    {
      "epoch": 0.07534833655590328,
      "grad_norm": 33.25,
      "learning_rate": 1.989243013961806e-05,
      "loss": 1.0905,
      "num_input_tokens_seen": 423004416,
      "step": 5670,
      "train_runtime": 6783.2206,
      "train_tokens_per_second": 62360.41
    },
    {
      "epoch": 0.0754812260383652,
      "grad_norm": 20.625,
      "learning_rate": 1.9891799633306635e-05,
      "loss": 0.909,
      "num_input_tokens_seen": 423476992,
      "step": 5680,
      "train_runtime": 6789.6766,
      "train_tokens_per_second": 62370.716
    },
    {
      "epoch": 0.0756141155208271,
      "grad_norm": 53.5,
      "learning_rate": 1.9891167294627536e-05,
      "loss": 0.6382,
      "num_input_tokens_seen": 423703040,
      "step": 5690,
      "train_runtime": 6793.058,
      "train_tokens_per_second": 62372.946
    },
    {
      "epoch": 0.07574700500328901,
      "grad_norm": 15.5,
      "learning_rate": 1.9890533123697895e-05,
      "loss": 0.5938,
      "num_input_tokens_seen": 423816448,
      "step": 5700,
      "train_runtime": 6795.9322,
      "train_tokens_per_second": 62363.254
    },
    {
      "epoch": 0.07587989448575093,
      "grad_norm": 18.75,
      "learning_rate": 1.9889897120635184e-05,
      "loss": 1.0045,
      "num_input_tokens_seen": 425806592,
      "step": 5710,
      "train_runtime": 6828.4867,
      "train_tokens_per_second": 62357.388
    },
    {
      "epoch": 0.07601278396821283,
      "grad_norm": 20.0,
      "learning_rate": 1.9889259285557216e-05,
      "loss": 0.9829,
      "num_input_tokens_seen": 426649088,
      "step": 5720,
      "train_runtime": 6840.8915,
      "train_tokens_per_second": 62367.468
    },
    {
      "epoch": 0.07614567345067474,
      "grad_norm": 28.5,
      "learning_rate": 1.9888619618582147e-05,
      "loss": 0.8188,
      "num_input_tokens_seen": 427098880,
      "step": 5730,
      "train_runtime": 6847.4371,
      "train_tokens_per_second": 62373.538
    },
    {
      "epoch": 0.07627856293313666,
      "grad_norm": 14.875,
      "learning_rate": 1.9887978119828474e-05,
      "loss": 0.7501,
      "num_input_tokens_seen": 427323392,
      "step": 5740,
      "train_runtime": 6850.8299,
      "train_tokens_per_second": 62375.42
    },
    {
      "epoch": 0.07641145241559857,
      "grad_norm": 11.25,
      "learning_rate": 1.9887334789415017e-05,
      "loss": 0.5408,
      "num_input_tokens_seen": 427433984,
      "step": 5750,
      "train_runtime": 6853.6659,
      "train_tokens_per_second": 62365.746
    },
    {
      "epoch": 0.07654434189806048,
      "grad_norm": 31.375,
      "learning_rate": 1.9886689627460954e-05,
      "loss": 0.9848,
      "num_input_tokens_seen": 429381632,
      "step": 5760,
      "train_runtime": 6885.7031,
      "train_tokens_per_second": 62358.43
    },
    {
      "epoch": 0.07667723138052239,
      "grad_norm": 13.9375,
      "learning_rate": 1.9886042634085802e-05,
      "loss": 1.0351,
      "num_input_tokens_seen": 430211328,
      "step": 5770,
      "train_runtime": 6898.2431,
      "train_tokens_per_second": 62365.347
    },
    {
      "epoch": 0.0768101208629843,
      "grad_norm": 17.625,
      "learning_rate": 1.98853938094094e-05,
      "loss": 0.8198,
      "num_input_tokens_seen": 430667008,
      "step": 5780,
      "train_runtime": 6904.8945,
      "train_tokens_per_second": 62371.266
    },
    {
      "epoch": 0.0769430103454462,
      "grad_norm": 29.625,
      "learning_rate": 1.988474315355194e-05,
      "loss": 0.7534,
      "num_input_tokens_seen": 430893824,
      "step": 5790,
      "train_runtime": 6908.3102,
      "train_tokens_per_second": 62373.259
    },
    {
      "epoch": 0.07707589982790812,
      "grad_norm": 17.125,
      "learning_rate": 1.988409066663395e-05,
      "loss": 0.5895,
      "num_input_tokens_seen": 431007744,
      "step": 5800,
      "train_runtime": 6910.6224,
      "train_tokens_per_second": 62368.875
    },
    {
      "epoch": 0.07720878931037004,
      "grad_norm": 18.25,
      "learning_rate": 1.9883436348776308e-05,
      "loss": 1.0226,
      "num_input_tokens_seen": 432979200,
      "step": 5810,
      "train_runtime": 6942.0668,
      "train_tokens_per_second": 62370.359
    },
    {
      "epoch": 0.07734167879283194,
      "grad_norm": 41.75,
      "learning_rate": 1.98827802001002e-05,
      "loss": 0.9895,
      "num_input_tokens_seen": 433842944,
      "step": 5820,
      "train_runtime": 6954.9693,
      "train_tokens_per_second": 62378.843
    },
    {
      "epoch": 0.07747456827529385,
      "grad_norm": 19.25,
      "learning_rate": 1.9882122220727188e-05,
      "loss": 0.8948,
      "num_input_tokens_seen": 434303744,
      "step": 5830,
      "train_runtime": 6962.0314,
      "train_tokens_per_second": 62381.756
    },
    {
      "epoch": 0.07760745775775577,
      "grad_norm": 12.125,
      "learning_rate": 1.9881462410779157e-05,
      "loss": 0.6796,
      "num_input_tokens_seen": 434529024,
      "step": 5840,
      "train_runtime": 6965.4428,
      "train_tokens_per_second": 62383.546
    },
    {
      "epoch": 0.07774034724021767,
      "grad_norm": 27.125,
      "learning_rate": 1.988080077037832e-05,
      "loss": 0.6324,
      "num_input_tokens_seen": 434644480,
      "step": 5850,
      "train_runtime": 6967.8278,
      "train_tokens_per_second": 62378.763
    },
    {
      "epoch": 0.07787323672267958,
      "grad_norm": 11.125,
      "learning_rate": 1.9880137299647244e-05,
      "loss": 0.9845,
      "num_input_tokens_seen": 436735488,
      "step": 5860,
      "train_runtime": 7001.4025,
      "train_tokens_per_second": 62378.286
    },
    {
      "epoch": 0.0780061262051415,
      "grad_norm": 11.0625,
      "learning_rate": 1.9879471998708836e-05,
      "loss": 1.0031,
      "num_input_tokens_seen": 437580032,
      "step": 5870,
      "train_runtime": 7014.2099,
      "train_tokens_per_second": 62384.793
    },
    {
      "epoch": 0.07813901568760341,
      "grad_norm": 17.0,
      "learning_rate": 1.9878804867686332e-05,
      "loss": 0.8011,
      "num_input_tokens_seen": 438055168,
      "step": 5880,
      "train_runtime": 7021.3304,
      "train_tokens_per_second": 62389.197
    },
    {
      "epoch": 0.07827190517006531,
      "grad_norm": 82.5,
      "learning_rate": 1.9878135906703316e-05,
      "loss": 0.6829,
      "num_input_tokens_seen": 438283776,
      "step": 5890,
      "train_runtime": 7024.7484,
      "train_tokens_per_second": 62391.384
    },
    {
      "epoch": 0.07840479465252723,
      "grad_norm": 23.75,
      "learning_rate": 1.9877465115883704e-05,
      "loss": 0.5364,
      "num_input_tokens_seen": 438398464,
      "step": 5900,
      "train_runtime": 7027.1195,
      "train_tokens_per_second": 62386.653
    },
    {
      "epoch": 0.07853768413498914,
      "grad_norm": 49.75,
      "learning_rate": 1.9876792495351756e-05,
      "loss": 0.9651,
      "num_input_tokens_seen": 440485632,
      "step": 5910,
      "train_runtime": 7060.227,
      "train_tokens_per_second": 62389.726
    },
    {
      "epoch": 0.07867057361745104,
      "grad_norm": 15.625,
      "learning_rate": 1.9876118045232068e-05,
      "loss": 0.9765,
      "num_input_tokens_seen": 441307904,
      "step": 5920,
      "train_runtime": 7072.7502,
      "train_tokens_per_second": 62395.517
    },
    {
      "epoch": 0.07880346309991296,
      "grad_norm": 17.25,
      "learning_rate": 1.9875441765649573e-05,
      "loss": 0.7611,
      "num_input_tokens_seen": 441756160,
      "step": 5930,
      "train_runtime": 7079.269,
      "train_tokens_per_second": 62401.381
    },
    {
      "epoch": 0.07893635258237487,
      "grad_norm": 23.375,
      "learning_rate": 1.987476365672955e-05,
      "loss": 0.7862,
      "num_input_tokens_seen": 441975296,
      "step": 5940,
      "train_runtime": 7082.6737,
      "train_tokens_per_second": 62402.324
    },
    {
      "epoch": 0.07906924206483677,
      "grad_norm": 11.875,
      "learning_rate": 1.9874083718597613e-05,
      "loss": 0.5785,
      "num_input_tokens_seen": 442085632,
      "step": 5950,
      "train_runtime": 7084.965,
      "train_tokens_per_second": 62397.715
    },
    {
      "epoch": 0.07920213154729869,
      "grad_norm": 16.125,
      "learning_rate": 1.9873401951379713e-05,
      "loss": 0.9984,
      "num_input_tokens_seen": 444084992,
      "step": 5960,
      "train_runtime": 7116.8094,
      "train_tokens_per_second": 62399.45
    },
    {
      "epoch": 0.0793350210297606,
      "grad_norm": 30.5,
      "learning_rate": 1.9872718355202138e-05,
      "loss": 0.9376,
      "num_input_tokens_seen": 444929024,
      "step": 5970,
      "train_runtime": 7129.0667,
      "train_tokens_per_second": 62410.557
    },
    {
      "epoch": 0.0794679105122225,
      "grad_norm": 24.125,
      "learning_rate": 1.987203293019152e-05,
      "loss": 0.8116,
      "num_input_tokens_seen": 445356288,
      "step": 5980,
      "train_runtime": 7135.5047,
      "train_tokens_per_second": 62414.126
    },
    {
      "epoch": 0.07960079999468442,
      "grad_norm": 14.9375,
      "learning_rate": 1.9871345676474837e-05,
      "loss": 0.6912,
      "num_input_tokens_seen": 445578752,
      "step": 5990,
      "train_runtime": 7138.927,
      "train_tokens_per_second": 62415.367
    },
    {
      "epoch": 0.07973368947714633,
      "grad_norm": 20.125,
      "learning_rate": 1.987065659417938e-05,
      "loss": 0.5349,
      "num_input_tokens_seen": 445694720,
      "step": 6000,
      "train_runtime": 7141.6404,
      "train_tokens_per_second": 62407.891
    },
    {
      "epoch": 0.07986657895960825,
      "grad_norm": 15.875,
      "learning_rate": 1.9869965683432808e-05,
      "loss": 0.91,
      "num_input_tokens_seen": 447619584,
      "step": 6010,
      "train_runtime": 7172.8457,
      "train_tokens_per_second": 62404.741
    },
    {
      "epoch": 0.07999946844207015,
      "grad_norm": 17.875,
      "learning_rate": 1.9869272944363104e-05,
      "loss": 1.0066,
      "num_input_tokens_seen": 448489216,
      "step": 6020,
      "train_runtime": 7185.561,
      "train_tokens_per_second": 62415.338
    },
    {
      "epoch": 0.08013235792453206,
      "grad_norm": 135.0,
      "learning_rate": 1.9868578377098585e-05,
      "loss": 0.8456,
      "num_input_tokens_seen": 448957184,
      "step": 6030,
      "train_runtime": 7192.3189,
      "train_tokens_per_second": 62421.757
    },
    {
      "epoch": 0.08026524740699398,
      "grad_norm": 11.3125,
      "learning_rate": 1.9867881981767918e-05,
      "loss": 0.7499,
      "num_input_tokens_seen": 449186304,
      "step": 6040,
      "train_runtime": 7195.7688,
      "train_tokens_per_second": 62423.671
    },
    {
      "epoch": 0.08039813688945588,
      "grad_norm": 10.625,
      "learning_rate": 1.9867183758500107e-05,
      "loss": 0.54,
      "num_input_tokens_seen": 449297920,
      "step": 6050,
      "train_runtime": 7198.1762,
      "train_tokens_per_second": 62418.3
    },
    {
      "epoch": 0.0805310263719178,
      "grad_norm": 43.5,
      "learning_rate": 1.986648370742449e-05,
      "loss": 1.0614,
      "num_input_tokens_seen": 451322368,
      "step": 6060,
      "train_runtime": 7231.3773,
      "train_tokens_per_second": 62411.675
    },
    {
      "epoch": 0.08066391585437971,
      "grad_norm": 23.375,
      "learning_rate": 1.986578182867074e-05,
      "loss": 0.9198,
      "num_input_tokens_seen": 452167424,
      "step": 6070,
      "train_runtime": 7244.5796,
      "train_tokens_per_second": 62414.584
    },
    {
      "epoch": 0.08079680533684161,
      "grad_norm": 26.125,
      "learning_rate": 1.9865078122368882e-05,
      "loss": 0.8427,
      "num_input_tokens_seen": 452673024,
      "step": 6080,
      "train_runtime": 7251.6781,
      "train_tokens_per_second": 62423.21
    },
    {
      "epoch": 0.08092969481930352,
      "grad_norm": 36.5,
      "learning_rate": 1.9864372588649263e-05,
      "loss": 0.7262,
      "num_input_tokens_seen": 452913664,
      "step": 6090,
      "train_runtime": 7255.2843,
      "train_tokens_per_second": 62425.35
    },
    {
      "epoch": 0.08106258430176544,
      "grad_norm": 46.0,
      "learning_rate": 1.9863665227642583e-05,
      "loss": 0.5965,
      "num_input_tokens_seen": 453032704,
      "step": 6100,
      "train_runtime": 7258.0532,
      "train_tokens_per_second": 62417.936
    },
    {
      "epoch": 0.08119547378422735,
      "grad_norm": 18.0,
      "learning_rate": 1.9862956039479873e-05,
      "loss": 0.9547,
      "num_input_tokens_seen": 455006208,
      "step": 6110,
      "train_runtime": 7290.3786,
      "train_tokens_per_second": 62411.876
    },
    {
      "epoch": 0.08132836326668925,
      "grad_norm": 24.125,
      "learning_rate": 1.9862245024292503e-05,
      "loss": 0.9986,
      "num_input_tokens_seen": 455842816,
      "step": 6120,
      "train_runtime": 7303.1038,
      "train_tokens_per_second": 62417.682
    },
    {
      "epoch": 0.08146125274915117,
      "grad_norm": 28.75,
      "learning_rate": 1.986153218221218e-05,
      "loss": 0.831,
      "num_input_tokens_seen": 456331520,
      "step": 6130,
      "train_runtime": 7309.8519,
      "train_tokens_per_second": 62426.917
    },
    {
      "epoch": 0.08159414223161308,
      "grad_norm": 23.875,
      "learning_rate": 1.9860817513370956e-05,
      "loss": 0.7876,
      "num_input_tokens_seen": 456561152,
      "step": 6140,
      "train_runtime": 7313.3553,
      "train_tokens_per_second": 62428.411
    },
    {
      "epoch": 0.08172703171407499,
      "grad_norm": 20.625,
      "learning_rate": 1.9860101017901214e-05,
      "loss": 0.5887,
      "num_input_tokens_seen": 456675328,
      "step": 6150,
      "train_runtime": 7315.7306,
      "train_tokens_per_second": 62423.749
    },
    {
      "epoch": 0.0818599211965369,
      "grad_norm": 24.125,
      "learning_rate": 1.9859382695935682e-05,
      "loss": 0.9385,
      "num_input_tokens_seen": 458755328,
      "step": 6160,
      "train_runtime": 7348.7072,
      "train_tokens_per_second": 62426.671
    },
    {
      "epoch": 0.08199281067899881,
      "grad_norm": 35.0,
      "learning_rate": 1.9858662547607417e-05,
      "loss": 0.9994,
      "num_input_tokens_seen": 459591168,
      "step": 6170,
      "train_runtime": 7361.861,
      "train_tokens_per_second": 62428.667
    },
    {
      "epoch": 0.08212570016146072,
      "grad_norm": 21.75,
      "learning_rate": 1.985794057304982e-05,
      "loss": 0.8146,
      "num_input_tokens_seen": 460079360,
      "step": 6180,
      "train_runtime": 7368.6654,
      "train_tokens_per_second": 62437.271
    },
    {
      "epoch": 0.08225858964392263,
      "grad_norm": 41.0,
      "learning_rate": 1.985721677239664e-05,
      "loss": 0.6709,
      "num_input_tokens_seen": 460303616,
      "step": 6190,
      "train_runtime": 7372.1177,
      "train_tokens_per_second": 62438.452
    },
    {
      "epoch": 0.08239147912638455,
      "grad_norm": 11.875,
      "learning_rate": 1.9856491145781947e-05,
      "loss": 0.5793,
      "num_input_tokens_seen": 460419584,
      "step": 6200,
      "train_runtime": 7374.4593,
      "train_tokens_per_second": 62434.352
    },
    {
      "epoch": 0.08252436860884645,
      "grad_norm": 17.625,
      "learning_rate": 1.9855763693340157e-05,
      "loss": 0.8778,
      "num_input_tokens_seen": 462407936,
      "step": 6210,
      "train_runtime": 7407.6588,
      "train_tokens_per_second": 62422.953
    },
    {
      "epoch": 0.08265725809130836,
      "grad_norm": 17.375,
      "learning_rate": 1.9855034415206022e-05,
      "loss": 1.0376,
      "num_input_tokens_seen": 463245056,
      "step": 6220,
      "train_runtime": 7420.2841,
      "train_tokens_per_second": 62429.558
    },
    {
      "epoch": 0.08279014757377028,
      "grad_norm": 20.25,
      "learning_rate": 1.9854303311514637e-05,
      "loss": 0.8505,
      "num_input_tokens_seen": 463691264,
      "step": 6230,
      "train_runtime": 7426.6723,
      "train_tokens_per_second": 62435.94
    },
    {
      "epoch": 0.08292303705623219,
      "grad_norm": 15.625,
      "learning_rate": 1.9853570382401437e-05,
      "loss": 0.7238,
      "num_input_tokens_seen": 463921664,
      "step": 6240,
      "train_runtime": 7430.1498,
      "train_tokens_per_second": 62437.727
    },
    {
      "epoch": 0.08305592653869409,
      "grad_norm": 11.8125,
      "learning_rate": 1.9852835628002184e-05,
      "loss": 0.5377,
      "num_input_tokens_seen": 464037632,
      "step": 6250,
      "train_runtime": 7432.5518,
      "train_tokens_per_second": 62433.151
    },
    {
      "epoch": 0.083188816021156,
      "grad_norm": 51.0,
      "learning_rate": 1.985209904845299e-05,
      "loss": 0.9297,
      "num_input_tokens_seen": 466160640,
      "step": 6260,
      "train_runtime": 7466.0509,
      "train_tokens_per_second": 62437.378
    },
    {
      "epoch": 0.08332170550361792,
      "grad_norm": 25.625,
      "learning_rate": 1.9851360643890294e-05,
      "loss": 1.0334,
      "num_input_tokens_seen": 467012096,
      "step": 6270,
      "train_runtime": 7478.6377,
      "train_tokens_per_second": 62446.146
    },
    {
      "epoch": 0.08345459498607982,
      "grad_norm": 28.25,
      "learning_rate": 1.985062041445088e-05,
      "loss": 0.8241,
      "num_input_tokens_seen": 467482624,
      "step": 6280,
      "train_runtime": 7485.1914,
      "train_tokens_per_second": 62454.331
    },
    {
      "epoch": 0.08358748446854174,
      "grad_norm": 24.625,
      "learning_rate": 1.9849878360271875e-05,
      "loss": 0.7332,
      "num_input_tokens_seen": 467705856,
      "step": 6290,
      "train_runtime": 7488.6207,
      "train_tokens_per_second": 62455.541
    },
    {
      "epoch": 0.08372037395100365,
      "grad_norm": 22.5,
      "learning_rate": 1.984913448149073e-05,
      "loss": 0.6613,
      "num_input_tokens_seen": 467818496,
      "step": 6300,
      "train_runtime": 7490.9969,
      "train_tokens_per_second": 62450.766
    },
    {
      "epoch": 0.08385326343346555,
      "grad_norm": 14.125,
      "learning_rate": 1.9848388778245248e-05,
      "loss": 1.1361,
      "num_input_tokens_seen": 469896448,
      "step": 6310,
      "train_runtime": 7524.1324,
      "train_tokens_per_second": 62451.911
    },
    {
      "epoch": 0.08398615291592747,
      "grad_norm": 13.5,
      "learning_rate": 1.9847641250673558e-05,
      "loss": 1.0573,
      "num_input_tokens_seen": 470726400,
      "step": 6320,
      "train_runtime": 7537.0118,
      "train_tokens_per_second": 62455.309
    },
    {
      "epoch": 0.08411904239838938,
      "grad_norm": 18.375,
      "learning_rate": 1.984689189891414e-05,
      "loss": 0.8339,
      "num_input_tokens_seen": 471228160,
      "step": 6330,
      "train_runtime": 7543.9291,
      "train_tokens_per_second": 62464.553
    },
    {
      "epoch": 0.0842519318808513,
      "grad_norm": 29.5,
      "learning_rate": 1.98461407231058e-05,
      "loss": 0.6899,
      "num_input_tokens_seen": 471457536,
      "step": 6340,
      "train_runtime": 7547.8211,
      "train_tokens_per_second": 62462.733
    },
    {
      "epoch": 0.0843848213633132,
      "grad_norm": 18.25,
      "learning_rate": 1.9845387723387685e-05,
      "loss": 0.5742,
      "num_input_tokens_seen": 471576576,
      "step": 6350,
      "train_runtime": 7550.195,
      "train_tokens_per_second": 62458.86
    },
    {
      "epoch": 0.08451771084577511,
      "grad_norm": 25.875,
      "learning_rate": 1.9844632899899286e-05,
      "loss": 1.0427,
      "num_input_tokens_seen": 473564928,
      "step": 6360,
      "train_runtime": 7582.4538,
      "train_tokens_per_second": 62455.366
    },
    {
      "epoch": 0.08465060032823703,
      "grad_norm": 25.625,
      "learning_rate": 1.9843876252780425e-05,
      "loss": 1.0455,
      "num_input_tokens_seen": 474434560,
      "step": 6370,
      "train_runtime": 7595.288,
      "train_tokens_per_second": 62464.328
    },
    {
      "epoch": 0.08478348981069893,
      "grad_norm": 19.375,
      "learning_rate": 1.9843117782171263e-05,
      "loss": 0.8259,
      "num_input_tokens_seen": 474940416,
      "step": 6380,
      "train_runtime": 7602.3063,
      "train_tokens_per_second": 62473.202
    },
    {
      "epoch": 0.08491637929316084,
      "grad_norm": 22.875,
      "learning_rate": 1.9842357488212305e-05,
      "loss": 0.669,
      "num_input_tokens_seen": 475177472,
      "step": 6390,
      "train_runtime": 7605.8121,
      "train_tokens_per_second": 62475.573
    },
    {
      "epoch": 0.08504926877562276,
      "grad_norm": 19.0,
      "learning_rate": 1.984159537104438e-05,
      "loss": 0.5603,
      "num_input_tokens_seen": 475294464,
      "step": 6400,
      "train_runtime": 7608.5896,
      "train_tokens_per_second": 62468.143
    },
    {
      "epoch": 0.08518215825808466,
      "grad_norm": 19.0,
      "learning_rate": 1.9840831430808675e-05,
      "loss": 1.0339,
      "num_input_tokens_seen": 477445120,
      "step": 6410,
      "train_runtime": 7642.0657,
      "train_tokens_per_second": 62475.925
    },
    {
      "epoch": 0.08531504774054657,
      "grad_norm": 21.875,
      "learning_rate": 1.984006566764669e-05,
      "loss": 0.9238,
      "num_input_tokens_seen": 478259968,
      "step": 6420,
      "train_runtime": 7654.4596,
      "train_tokens_per_second": 62481.219
    },
    {
      "epoch": 0.08544793722300849,
      "grad_norm": 32.5,
      "learning_rate": 1.9839298081700285e-05,
      "loss": 0.8216,
      "num_input_tokens_seen": 478708480,
      "step": 6430,
      "train_runtime": 7660.8998,
      "train_tokens_per_second": 62487.239
    },
    {
      "epoch": 0.08558082670547039,
      "grad_norm": 24.0,
      "learning_rate": 1.9838528673111645e-05,
      "loss": 0.6932,
      "num_input_tokens_seen": 478936576,
      "step": 6440,
      "train_runtime": 7664.2996,
      "train_tokens_per_second": 62489.282
    },
    {
      "epoch": 0.0857137161879323,
      "grad_norm": 16.25,
      "learning_rate": 1.9837757442023297e-05,
      "loss": 0.6279,
      "num_input_tokens_seen": 479049472,
      "step": 6450,
      "train_runtime": 7667.0595,
      "train_tokens_per_second": 62481.513
    },
    {
      "epoch": 0.08584660567039422,
      "grad_norm": 29.0,
      "learning_rate": 1.9836984388578106e-05,
      "loss": 0.9398,
      "num_input_tokens_seen": 481093376,
      "step": 6460,
      "train_runtime": 7699.5299,
      "train_tokens_per_second": 62483.474
    },
    {
      "epoch": 0.08597949515285613,
      "grad_norm": 33.0,
      "learning_rate": 1.983620951291927e-05,
      "loss": 1.0055,
      "num_input_tokens_seen": 481903360,
      "step": 6470,
      "train_runtime": 7711.813,
      "train_tokens_per_second": 62488.984
    },
    {
      "epoch": 0.08611238463531803,
      "grad_norm": 15.3125,
      "learning_rate": 1.983543281519033e-05,
      "loss": 0.8809,
      "num_input_tokens_seen": 482367232,
      "step": 6480,
      "train_runtime": 7718.2817,
      "train_tokens_per_second": 62496.712
    },
    {
      "epoch": 0.08624527411777995,
      "grad_norm": 15.1875,
      "learning_rate": 1.983465429553516e-05,
      "loss": 0.7495,
      "num_input_tokens_seen": 482585600,
      "step": 6490,
      "train_runtime": 7721.6714,
      "train_tokens_per_second": 62497.557
    },
    {
      "epoch": 0.08637816360024186,
      "grad_norm": 25.0,
      "learning_rate": 1.983387395409798e-05,
      "loss": 0.6022,
      "num_input_tokens_seen": 482701568,
      "step": 6500,
      "train_runtime": 7724.0234,
      "train_tokens_per_second": 62493.54
    },
    {
      "epoch": 0.08651105308270376,
      "grad_norm": 15.6875,
      "learning_rate": 1.9833091791023334e-05,
      "loss": 0.9308,
      "num_input_tokens_seen": 484816640,
      "step": 6510,
      "train_runtime": 7758.28,
      "train_tokens_per_second": 62490.222
    },
    {
      "epoch": 0.08664394256516568,
      "grad_norm": 20.5,
      "learning_rate": 1.9832307806456114e-05,
      "loss": 0.9972,
      "num_input_tokens_seen": 485633536,
      "step": 6520,
      "train_runtime": 7770.9702,
      "train_tokens_per_second": 62493.295
    },
    {
      "epoch": 0.0867768320476276,
      "grad_norm": 15.625,
      "learning_rate": 1.9831522000541547e-05,
      "loss": 0.8389,
      "num_input_tokens_seen": 486121472,
      "step": 6530,
      "train_runtime": 7777.9016,
      "train_tokens_per_second": 62500.337
    },
    {
      "epoch": 0.0869097215300895,
      "grad_norm": 14.3125,
      "learning_rate": 1.9830734373425194e-05,
      "loss": 0.6504,
      "num_input_tokens_seen": 486364672,
      "step": 6540,
      "train_runtime": 7781.4953,
      "train_tokens_per_second": 62502.727
    },
    {
      "epoch": 0.08704261101255141,
      "grad_norm": 18.875,
      "learning_rate": 1.982994492525296e-05,
      "loss": 0.5654,
      "num_input_tokens_seen": 486480896,
      "step": 6550,
      "train_runtime": 7784.2843,
      "train_tokens_per_second": 62495.263
    },
    {
      "epoch": 0.08717550049501332,
      "grad_norm": 19.375,
      "learning_rate": 1.982915365617108e-05,
      "loss": 0.8627,
      "num_input_tokens_seen": 488662016,
      "step": 6560,
      "train_runtime": 7819.4232,
      "train_tokens_per_second": 62493.358
    },
    {
      "epoch": 0.08730838997747523,
      "grad_norm": 25.125,
      "learning_rate": 1.982836056632613e-05,
      "loss": 0.9646,
      "num_input_tokens_seen": 489537792,
      "step": 6570,
      "train_runtime": 7832.3836,
      "train_tokens_per_second": 62501.764
    },
    {
      "epoch": 0.08744127945993714,
      "grad_norm": 21.125,
      "learning_rate": 1.9827565655865018e-05,
      "loss": 0.7727,
      "num_input_tokens_seen": 490019840,
      "step": 6580,
      "train_runtime": 7839.134,
      "train_tokens_per_second": 62509.436
    },
    {
      "epoch": 0.08757416894239906,
      "grad_norm": 16.375,
      "learning_rate": 1.9826768924935005e-05,
      "loss": 0.6526,
      "num_input_tokens_seen": 490241280,
      "step": 6590,
      "train_runtime": 7842.487,
      "train_tokens_per_second": 62510.946
    },
    {
      "epoch": 0.08770705842486097,
      "grad_norm": 14.9375,
      "learning_rate": 1.982597037368367e-05,
      "loss": 0.5827,
      "num_input_tokens_seen": 490357248,
      "step": 6600,
      "train_runtime": 7844.8388,
      "train_tokens_per_second": 62506.988
    },
    {
      "epoch": 0.08783994790732287,
      "grad_norm": 44.0,
      "learning_rate": 1.982517000225894e-05,
      "loss": 0.9842,
      "num_input_tokens_seen": 492423680,
      "step": 6610,
      "train_runtime": 7878.7771,
      "train_tokens_per_second": 62500.014
    },
    {
      "epoch": 0.08797283738978479,
      "grad_norm": 17.5,
      "learning_rate": 1.9824367810809078e-05,
      "loss": 0.9207,
      "num_input_tokens_seen": 493251840,
      "step": 6620,
      "train_runtime": 7891.0442,
      "train_tokens_per_second": 62507.803
    },
    {
      "epoch": 0.0881057268722467,
      "grad_norm": 109.5,
      "learning_rate": 1.982356379948268e-05,
      "loss": 0.7463,
      "num_input_tokens_seen": 493696000,
      "step": 6630,
      "train_runtime": 7897.8487,
      "train_tokens_per_second": 62510.187
    },
    {
      "epoch": 0.0882386163547086,
      "grad_norm": 14.8125,
      "learning_rate": 1.9822757968428684e-05,
      "loss": 0.7224,
      "num_input_tokens_seen": 493918976,
      "step": 6640,
      "train_runtime": 7901.2238,
      "train_tokens_per_second": 62511.706
    },
    {
      "epoch": 0.08837150583717052,
      "grad_norm": 15.875,
      "learning_rate": 1.982195031779636e-05,
      "loss": 0.4623,
      "num_input_tokens_seen": 494029312,
      "step": 6650,
      "train_runtime": 7903.5374,
      "train_tokens_per_second": 62507.367
    },
    {
      "epoch": 0.08850439531963243,
      "grad_norm": 33.25,
      "learning_rate": 1.9821140847735322e-05,
      "loss": 0.9769,
      "num_input_tokens_seen": 496122112,
      "step": 6660,
      "train_runtime": 7936.5762,
      "train_tokens_per_second": 62510.848
    },
    {
      "epoch": 0.08863728480209433,
      "grad_norm": 12.4375,
      "learning_rate": 1.9820329558395517e-05,
      "loss": 1.0077,
      "num_input_tokens_seen": 496926208,
      "step": 6670,
      "train_runtime": 7948.7966,
      "train_tokens_per_second": 62515.904
    },
    {
      "epoch": 0.08877017428455625,
      "grad_norm": 15.375,
      "learning_rate": 1.9819516449927225e-05,
      "loss": 0.795,
      "num_input_tokens_seen": 497384192,
      "step": 6680,
      "train_runtime": 7955.3943,
      "train_tokens_per_second": 62521.626
    },
    {
      "epoch": 0.08890306376701816,
      "grad_norm": 15.3125,
      "learning_rate": 1.981870152248107e-05,
      "loss": 0.6702,
      "num_input_tokens_seen": 497612032,
      "step": 6690,
      "train_runtime": 7959.1867,
      "train_tokens_per_second": 62520.462
    },
    {
      "epoch": 0.08903595324948008,
      "grad_norm": 30.875,
      "learning_rate": 1.981788477620801e-05,
      "loss": 0.6239,
      "num_input_tokens_seen": 497728768,
      "step": 6700,
      "train_runtime": 7961.5489,
      "train_tokens_per_second": 62516.575
    },
    {
      "epoch": 0.08916884273194198,
      "grad_norm": 23.625,
      "learning_rate": 1.9817066211259342e-05,
      "loss": 0.8908,
      "num_input_tokens_seen": 499755264,
      "step": 6710,
      "train_runtime": 7993.9679,
      "train_tokens_per_second": 62516.546
    },
    {
      "epoch": 0.08930173221440389,
      "grad_norm": 13.8125,
      "learning_rate": 1.9816245827786693e-05,
      "loss": 1.0238,
      "num_input_tokens_seen": 500548608,
      "step": 6720,
      "train_runtime": 8006.1395,
      "train_tokens_per_second": 62520.596
    },
    {
      "epoch": 0.08943462169686581,
      "grad_norm": 15.5625,
      "learning_rate": 1.9815423625942033e-05,
      "loss": 0.7912,
      "num_input_tokens_seen": 500996096,
      "step": 6730,
      "train_runtime": 8012.4437,
      "train_tokens_per_second": 62527.253
    },
    {
      "epoch": 0.08956751117932771,
      "grad_norm": 31.0,
      "learning_rate": 1.9814599605877672e-05,
      "loss": 0.6922,
      "num_input_tokens_seen": 501218560,
      "step": 6740,
      "train_runtime": 8016.1695,
      "train_tokens_per_second": 62525.943
    },
    {
      "epoch": 0.08970040066178962,
      "grad_norm": 17.125,
      "learning_rate": 1.9813773767746248e-05,
      "loss": 0.5685,
      "num_input_tokens_seen": 501334016,
      "step": 6750,
      "train_runtime": 8018.5462,
      "train_tokens_per_second": 62521.809
    },
    {
      "epoch": 0.08983329014425154,
      "grad_norm": 55.25,
      "learning_rate": 1.9812946111700742e-05,
      "loss": 0.9827,
      "num_input_tokens_seen": 503241472,
      "step": 6760,
      "train_runtime": 8049.3714,
      "train_tokens_per_second": 62519.351
    },
    {
      "epoch": 0.08996617962671344,
      "grad_norm": 33.75,
      "learning_rate": 1.9812116637894468e-05,
      "loss": 1.0154,
      "num_input_tokens_seen": 504050432,
      "step": 6770,
      "train_runtime": 8061.4431,
      "train_tokens_per_second": 62526.08
    },
    {
      "epoch": 0.09009906910917535,
      "grad_norm": 35.0,
      "learning_rate": 1.981128534648108e-05,
      "loss": 0.8499,
      "num_input_tokens_seen": 504487168,
      "step": 6780,
      "train_runtime": 8067.7106,
      "train_tokens_per_second": 62531.639
    },
    {
      "epoch": 0.09023195859163727,
      "grad_norm": 21.0,
      "learning_rate": 1.981045223761457e-05,
      "loss": 0.7514,
      "num_input_tokens_seen": 504705280,
      "step": 6790,
      "train_runtime": 8071.0718,
      "train_tokens_per_second": 62532.622
    },
    {
      "epoch": 0.09036484807409917,
      "grad_norm": 32.5,
      "learning_rate": 1.9809617311449255e-05,
      "loss": 0.4578,
      "num_input_tokens_seen": 504818688,
      "step": 6800,
      "train_runtime": 8073.432,
      "train_tokens_per_second": 62528.388
    },
    {
      "epoch": 0.09049773755656108,
      "grad_norm": 28.25,
      "learning_rate": 1.980878056813981e-05,
      "loss": 0.8997,
      "num_input_tokens_seen": 506949632,
      "step": 6810,
      "train_runtime": 8107.6443,
      "train_tokens_per_second": 62527.365
    },
    {
      "epoch": 0.090630627039023,
      "grad_norm": 26.25,
      "learning_rate": 1.9807942007841223e-05,
      "loss": 0.9957,
      "num_input_tokens_seen": 507791872,
      "step": 6820,
      "train_runtime": 8120.5164,
      "train_tokens_per_second": 62531.968
    },
    {
      "epoch": 0.09076351652148491,
      "grad_norm": 20.0,
      "learning_rate": 1.980710163070884e-05,
      "loss": 0.8364,
      "num_input_tokens_seen": 508250624,
      "step": 6830,
      "train_runtime": 8127.0553,
      "train_tokens_per_second": 62538.103
    },
    {
      "epoch": 0.09089640600394681,
      "grad_norm": 20.0,
      "learning_rate": 1.9806259436898325e-05,
      "loss": 0.6699,
      "num_input_tokens_seen": 508478720,
      "step": 6840,
      "train_runtime": 8130.492,
      "train_tokens_per_second": 62539.724
    },
    {
      "epoch": 0.09102929548640873,
      "grad_norm": 61.75,
      "learning_rate": 1.9805415426565687e-05,
      "loss": 0.5515,
      "num_input_tokens_seen": 508594688,
      "step": 6850,
      "train_runtime": 8132.8496,
      "train_tokens_per_second": 62535.853
    },
    {
      "epoch": 0.09116218496887064,
      "grad_norm": 13.0,
      "learning_rate": 1.9804569599867282e-05,
      "loss": 0.9681,
      "num_input_tokens_seen": 510549760,
      "step": 6860,
      "train_runtime": 8165.8008,
      "train_tokens_per_second": 62522.926
    },
    {
      "epoch": 0.09129507445133254,
      "grad_norm": 32.25,
      "learning_rate": 1.9803721956959778e-05,
      "loss": 0.9946,
      "num_input_tokens_seen": 511396864,
      "step": 6870,
      "train_runtime": 8178.6989,
      "train_tokens_per_second": 62527.9
    },
    {
      "epoch": 0.09142796393379446,
      "grad_norm": 27.0,
      "learning_rate": 1.9802872498000204e-05,
      "loss": 0.8519,
      "num_input_tokens_seen": 511873536,
      "step": 6880,
      "train_runtime": 8185.6614,
      "train_tokens_per_second": 62532.948
    },
    {
      "epoch": 0.09156085341625637,
      "grad_norm": 11.1875,
      "learning_rate": 1.980202122314591e-05,
      "loss": 0.7244,
      "num_input_tokens_seen": 512107008,
      "step": 6890,
      "train_runtime": 8189.1902,
      "train_tokens_per_second": 62534.511
    },
    {
      "epoch": 0.09169374289871828,
      "grad_norm": 34.5,
      "learning_rate": 1.9801168132554583e-05,
      "loss": 0.561,
      "num_input_tokens_seen": 512225536,
      "step": 6900,
      "train_runtime": 8191.5967,
      "train_tokens_per_second": 62530.61
    },
    {
      "epoch": 0.09182663238118019,
      "grad_norm": 32.5,
      "learning_rate": 1.980031322638426e-05,
      "loss": 0.9555,
      "num_input_tokens_seen": 514425856,
      "step": 6910,
      "train_runtime": 8226.0372,
      "train_tokens_per_second": 62536.291
    },
    {
      "epoch": 0.0919595218636421,
      "grad_norm": 19.25,
      "learning_rate": 1.9799456504793295e-05,
      "loss": 1.0459,
      "num_input_tokens_seen": 515253760,
      "step": 6920,
      "train_runtime": 8239.143,
      "train_tokens_per_second": 62537.301
    },
    {
      "epoch": 0.09209241134610402,
      "grad_norm": 15.0625,
      "learning_rate": 1.9798597967940398e-05,
      "loss": 0.7747,
      "num_input_tokens_seen": 515706112,
      "step": 6930,
      "train_runtime": 8245.5578,
      "train_tokens_per_second": 62543.508
    },
    {
      "epoch": 0.09222530082856592,
      "grad_norm": 13.875,
      "learning_rate": 1.9797737615984595e-05,
      "loss": 0.71,
      "num_input_tokens_seen": 515928576,
      "step": 6940,
      "train_runtime": 8248.9622,
      "train_tokens_per_second": 62544.665
    },
    {
      "epoch": 0.09235819031102784,
      "grad_norm": 33.25,
      "learning_rate": 1.9796875449085266e-05,
      "loss": 0.5755,
      "num_input_tokens_seen": 516042496,
      "step": 6950,
      "train_runtime": 8251.4188,
      "train_tokens_per_second": 62539.85
    },
    {
      "epoch": 0.09249107979348975,
      "grad_norm": 34.5,
      "learning_rate": 1.9796011467402116e-05,
      "loss": 0.9424,
      "num_input_tokens_seen": 518059008,
      "step": 6960,
      "train_runtime": 8284.1502,
      "train_tokens_per_second": 62536.168
    },
    {
      "epoch": 0.09262396927595165,
      "grad_norm": 40.75,
      "learning_rate": 1.979514567109519e-05,
      "loss": 0.9916,
      "num_input_tokens_seen": 518913792,
      "step": 6970,
      "train_runtime": 8296.7837,
      "train_tokens_per_second": 62543.97
    },
    {
      "epoch": 0.09275685875841357,
      "grad_norm": 12.4375,
      "learning_rate": 1.9794278060324874e-05,
      "loss": 0.799,
      "num_input_tokens_seen": 519361280,
      "step": 6980,
      "train_runtime": 8303.6941,
      "train_tokens_per_second": 62545.81
    },
    {
      "epoch": 0.09288974824087548,
      "grad_norm": 24.0,
      "learning_rate": 1.979340863525188e-05,
      "loss": 0.7001,
      "num_input_tokens_seen": 519587328,
      "step": 6990,
      "train_runtime": 8307.1369,
      "train_tokens_per_second": 62547.101
    },
    {
      "epoch": 0.09302263772333738,
      "grad_norm": 18.875,
      "learning_rate": 1.979253739603726e-05,
      "loss": 0.6221,
      "num_input_tokens_seen": 519700736,
      "step": 7000,
      "train_runtime": 8309.5252,
      "train_tokens_per_second": 62542.771
    },
    {
      "epoch": 0.0931555272057993,
      "grad_norm": 13.875,
      "learning_rate": 1.9791664342842407e-05,
      "loss": 0.8478,
      "num_input_tokens_seen": 521778688,
      "step": 7010,
      "train_runtime": 8343.6218,
      "train_tokens_per_second": 62536.235
    },
    {
      "epoch": 0.09328841668826121,
      "grad_norm": 28.25,
      "learning_rate": 1.9790789475829046e-05,
      "loss": 0.9862,
      "num_input_tokens_seen": 522670080,
      "step": 7020,
      "train_runtime": 8356.4115,
      "train_tokens_per_second": 62547.193
    },
    {
      "epoch": 0.09342130617072311,
      "grad_norm": 49.75,
      "learning_rate": 1.9789912795159238e-05,
      "loss": 0.8466,
      "num_input_tokens_seen": 523149312,
      "step": 7030,
      "train_runtime": 8363.5768,
      "train_tokens_per_second": 62550.907
    },
    {
      "epoch": 0.09355419565318503,
      "grad_norm": 17.0,
      "learning_rate": 1.9789034300995376e-05,
      "loss": 0.7243,
      "num_input_tokens_seen": 523385600,
      "step": 7040,
      "train_runtime": 8367.037,
      "train_tokens_per_second": 62553.279
    },
    {
      "epoch": 0.09368708513564694,
      "grad_norm": 27.125,
      "learning_rate": 1.9788153993500203e-05,
      "loss": 0.5715,
      "num_input_tokens_seen": 523499776,
      "step": 7050,
      "train_runtime": 8369.4114,
      "train_tokens_per_second": 62549.175
    },
    {
      "epoch": 0.09381997461810886,
      "grad_norm": 19.0,
      "learning_rate": 1.9787271872836776e-05,
      "loss": 0.9616,
      "num_input_tokens_seen": 525537280,
      "step": 7060,
      "train_runtime": 8402.301,
      "train_tokens_per_second": 62546.829
    },
    {
      "epoch": 0.09395286410057076,
      "grad_norm": 19.75,
      "learning_rate": 1.978638793916851e-05,
      "loss": 0.9477,
      "num_input_tokens_seen": 526391808,
      "step": 7070,
      "train_runtime": 8415.2915,
      "train_tokens_per_second": 62551.821
    },
    {
      "epoch": 0.09408575358303267,
      "grad_norm": 17.75,
      "learning_rate": 1.9785502192659142e-05,
      "loss": 0.7297,
      "num_input_tokens_seen": 526881792,
      "step": 7080,
      "train_runtime": 8422.1499,
      "train_tokens_per_second": 62559.061
    },
    {
      "epoch": 0.09421864306549459,
      "grad_norm": 15.0,
      "learning_rate": 1.9784614633472748e-05,
      "loss": 0.6741,
      "num_input_tokens_seen": 527118848,
      "step": 7090,
      "train_runtime": 8425.6999,
      "train_tokens_per_second": 62560.838
    },
    {
      "epoch": 0.09435153254795649,
      "grad_norm": 18.125,
      "learning_rate": 1.978372526177374e-05,
      "loss": 0.6125,
      "num_input_tokens_seen": 527235072,
      "step": 7100,
      "train_runtime": 8428.0982,
      "train_tokens_per_second": 62556.826
    },
    {
      "epoch": 0.0944844220304184,
      "grad_norm": 15.4375,
      "learning_rate": 1.9782834077726866e-05,
      "loss": 0.9502,
      "num_input_tokens_seen": 529362432,
      "step": 7110,
      "train_runtime": 8461.8957,
      "train_tokens_per_second": 62558.374
    },
    {
      "epoch": 0.09461731151288032,
      "grad_norm": 30.0,
      "learning_rate": 1.978194108149721e-05,
      "loss": 1.0131,
      "num_input_tokens_seen": 530224128,
      "step": 7120,
      "train_runtime": 8474.5633,
      "train_tokens_per_second": 62566.543
    },
    {
      "epoch": 0.09475020099534222,
      "grad_norm": 17.625,
      "learning_rate": 1.9781046273250198e-05,
      "loss": 0.8293,
      "num_input_tokens_seen": 530700032,
      "step": 7130,
      "train_runtime": 8481.3255,
      "train_tokens_per_second": 62572.77
    },
    {
      "epoch": 0.09488309047780413,
      "grad_norm": 22.5,
      "learning_rate": 1.9780149653151576e-05,
      "loss": 0.7142,
      "num_input_tokens_seen": 530931712,
      "step": 7140,
      "train_runtime": 8484.7697,
      "train_tokens_per_second": 62574.676
    },
    {
      "epoch": 0.09501597996026605,
      "grad_norm": 28.125,
      "learning_rate": 1.977925122136744e-05,
      "loss": 0.5549,
      "num_input_tokens_seen": 531044352,
      "step": 7150,
      "train_runtime": 8487.5476,
      "train_tokens_per_second": 62567.467
    },
    {
      "epoch": 0.09514886944272795,
      "grad_norm": 23.75,
      "learning_rate": 1.9778350978064214e-05,
      "loss": 0.9601,
      "num_input_tokens_seen": 532999680,
      "step": 7160,
      "train_runtime": 8519.7834,
      "train_tokens_per_second": 62560.239
    },
    {
      "epoch": 0.09528175892518986,
      "grad_norm": 27.0,
      "learning_rate": 1.977744892340866e-05,
      "loss": 0.9632,
      "num_input_tokens_seen": 533820160,
      "step": 7170,
      "train_runtime": 8532.3922,
      "train_tokens_per_second": 62563.95
    },
    {
      "epoch": 0.09541464840765178,
      "grad_norm": 15.25,
      "learning_rate": 1.9776545057567882e-05,
      "loss": 0.7594,
      "num_input_tokens_seen": 534300160,
      "step": 7180,
      "train_runtime": 8539.1128,
      "train_tokens_per_second": 62570.922
    },
    {
      "epoch": 0.09554753789011369,
      "grad_norm": 16.375,
      "learning_rate": 1.9775639380709303e-05,
      "loss": 0.7724,
      "num_input_tokens_seen": 534524416,
      "step": 7190,
      "train_runtime": 8542.5034,
      "train_tokens_per_second": 62572.339
    },
    {
      "epoch": 0.0956804273725756,
      "grad_norm": 46.0,
      "learning_rate": 1.9774731893000697e-05,
      "loss": 0.6721,
      "num_input_tokens_seen": 534635520,
      "step": 7200,
      "train_runtime": 8544.8203,
      "train_tokens_per_second": 62568.375
    },
    {
      "epoch": 0.09581331685503751,
      "grad_norm": 24.625,
      "learning_rate": 1.9773822594610174e-05,
      "loss": 0.9281,
      "num_input_tokens_seen": 536709120,
      "step": 7210,
      "train_runtime": 8578.596,
      "train_tokens_per_second": 62563.748
    },
    {
      "epoch": 0.09594620633749942,
      "grad_norm": 14.625,
      "learning_rate": 1.977291148570616e-05,
      "loss": 0.9604,
      "num_input_tokens_seen": 537565696,
      "step": 7220,
      "train_runtime": 8591.4912,
      "train_tokens_per_second": 62569.545
    },
    {
      "epoch": 0.09607909581996132,
      "grad_norm": 21.375,
      "learning_rate": 1.9771998566457438e-05,
      "loss": 0.7884,
      "num_input_tokens_seen": 538090752,
      "step": 7230,
      "train_runtime": 8598.7701,
      "train_tokens_per_second": 62577.641
    },
    {
      "epoch": 0.09621198530242324,
      "grad_norm": 23.75,
      "learning_rate": 1.9771083837033118e-05,
      "loss": 0.7573,
      "num_input_tokens_seen": 538328064,
      "step": 7240,
      "train_runtime": 8602.3464,
      "train_tokens_per_second": 62579.212
    },
    {
      "epoch": 0.09634487478488515,
      "grad_norm": 14.3125,
      "learning_rate": 1.9770167297602643e-05,
      "loss": 0.6033,
      "num_input_tokens_seen": 538440448,
      "step": 7250,
      "train_runtime": 8604.7126,
      "train_tokens_per_second": 62575.065
    },
    {
      "epoch": 0.09647776426734705,
      "grad_norm": 41.25,
      "learning_rate": 1.9769248948335796e-05,
      "loss": 0.9566,
      "num_input_tokens_seen": 540566272,
      "step": 7260,
      "train_runtime": 8640.1673,
      "train_tokens_per_second": 62564.329
    },
    {
      "epoch": 0.09661065374980897,
      "grad_norm": 50.5,
      "learning_rate": 1.976832878940269e-05,
      "loss": 1.0353,
      "num_input_tokens_seen": 541376256,
      "step": 7270,
      "train_runtime": 8652.6843,
      "train_tokens_per_second": 62567.434
    },
    {
      "epoch": 0.09674354323227088,
      "grad_norm": 27.75,
      "learning_rate": 1.976740682097378e-05,
      "loss": 0.852,
      "num_input_tokens_seen": 541879552,
      "step": 7280,
      "train_runtime": 8659.4956,
      "train_tokens_per_second": 62576.341
    },
    {
      "epoch": 0.0968764327147328,
      "grad_norm": 19.875,
      "learning_rate": 1.976648304321985e-05,
      "loss": 0.7117,
      "num_input_tokens_seen": 542115584,
      "step": 7290,
      "train_runtime": 8663.0867,
      "train_tokens_per_second": 62577.647
    },
    {
      "epoch": 0.0970093221971947,
      "grad_norm": 16.625,
      "learning_rate": 1.9765557456312026e-05,
      "loss": 0.5831,
      "num_input_tokens_seen": 542233856,
      "step": 7300,
      "train_runtime": 8665.9381,
      "train_tokens_per_second": 62570.705
    },
    {
      "epoch": 0.09714221167965661,
      "grad_norm": 17.25,
      "learning_rate": 1.9764630060421756e-05,
      "loss": 0.9105,
      "num_input_tokens_seen": 544344576,
      "step": 7310,
      "train_runtime": 8700.2568,
      "train_tokens_per_second": 62566.496
    },
    {
      "epoch": 0.09727510116211853,
      "grad_norm": 13.5,
      "learning_rate": 1.976370085572084e-05,
      "loss": 1.0111,
      "num_input_tokens_seen": 545187328,
      "step": 7320,
      "train_runtime": 8712.7743,
      "train_tokens_per_second": 62573.333
    },
    {
      "epoch": 0.09740799064458043,
      "grad_norm": 17.25,
      "learning_rate": 1.9762769842381403e-05,
      "loss": 0.7762,
      "num_input_tokens_seen": 545624320,
      "step": 7330,
      "train_runtime": 8719.1063,
      "train_tokens_per_second": 62578.01
    },
    {
      "epoch": 0.09754088012704235,
      "grad_norm": 23.875,
      "learning_rate": 1.9761837020575903e-05,
      "loss": 0.691,
      "num_input_tokens_seen": 545843968,
      "step": 7340,
      "train_runtime": 8722.5005,
      "train_tokens_per_second": 62578.841
    },
    {
      "epoch": 0.09767376960950426,
      "grad_norm": 19.375,
      "learning_rate": 1.9760902390477138e-05,
      "loss": 0.5564,
      "num_input_tokens_seen": 545956352,
      "step": 7350,
      "train_runtime": 8724.8542,
      "train_tokens_per_second": 62574.84
    },
    {
      "epoch": 0.09780665909196616,
      "grad_norm": 23.125,
      "learning_rate": 1.9759965952258246e-05,
      "loss": 0.907,
      "num_input_tokens_seen": 547983616,
      "step": 7360,
      "train_runtime": 8759.0526,
      "train_tokens_per_second": 62561.974
    },
    {
      "epoch": 0.09793954857442808,
      "grad_norm": 24.25,
      "learning_rate": 1.9759027706092684e-05,
      "loss": 0.9505,
      "num_input_tokens_seen": 548815872,
      "step": 7370,
      "train_runtime": 8771.5679,
      "train_tokens_per_second": 62567.591
    },
    {
      "epoch": 0.09807243805688999,
      "grad_norm": 13.5625,
      "learning_rate": 1.9758087652154265e-05,
      "loss": 0.755,
      "num_input_tokens_seen": 549274368,
      "step": 7380,
      "train_runtime": 8778.3962,
      "train_tokens_per_second": 62571.152
    },
    {
      "epoch": 0.09820532753935189,
      "grad_norm": 18.875,
      "learning_rate": 1.9757145790617118e-05,
      "loss": 0.6664,
      "num_input_tokens_seen": 549500416,
      "step": 7390,
      "train_runtime": 8781.8444,
      "train_tokens_per_second": 62572.324
    },
    {
      "epoch": 0.0983382170218138,
      "grad_norm": 13.0625,
      "learning_rate": 1.9756202121655718e-05,
      "loss": 0.5911,
      "num_input_tokens_seen": 549616896,
      "step": 7400,
      "train_runtime": 8784.2098,
      "train_tokens_per_second": 62568.735
    },
    {
      "epoch": 0.09847110650427572,
      "grad_norm": 19.5,
      "learning_rate": 1.9755256645444866e-05,
      "loss": 0.9433,
      "num_input_tokens_seen": 551630592,
      "step": 7410,
      "train_runtime": 8815.3609,
      "train_tokens_per_second": 62576.065
    },
    {
      "epoch": 0.09860399598673764,
      "grad_norm": 45.0,
      "learning_rate": 1.9754309362159704e-05,
      "loss": 0.9477,
      "num_input_tokens_seen": 552454656,
      "step": 7420,
      "train_runtime": 8827.7215,
      "train_tokens_per_second": 62581.795
    },
    {
      "epoch": 0.09873688546919954,
      "grad_norm": 14.75,
      "learning_rate": 1.9753360271975713e-05,
      "loss": 0.7143,
      "num_input_tokens_seen": 552941312,
      "step": 7430,
      "train_runtime": 8834.8395,
      "train_tokens_per_second": 62586.458
    },
    {
      "epoch": 0.09886977495166145,
      "grad_norm": 8.5,
      "learning_rate": 1.9752409375068705e-05,
      "loss": 0.6294,
      "num_input_tokens_seen": 553179392,
      "step": 7440,
      "train_runtime": 8838.3948,
      "train_tokens_per_second": 62588.219
    },
    {
      "epoch": 0.09900266443412337,
      "grad_norm": 12.0625,
      "learning_rate": 1.9751456671614814e-05,
      "loss": 0.5821,
      "num_input_tokens_seen": 553297920,
      "step": 7450,
      "train_runtime": 8840.7674,
      "train_tokens_per_second": 62584.83
    },
    {
      "epoch": 0.09913555391658527,
      "grad_norm": 23.75,
      "learning_rate": 1.975050216179053e-05,
      "loss": 0.9294,
      "num_input_tokens_seen": 555384064,
      "step": 7460,
      "train_runtime": 8874.0206,
      "train_tokens_per_second": 62585.393
    },
    {
      "epoch": 0.09926844339904718,
      "grad_norm": 21.125,
      "learning_rate": 1.9749545845772665e-05,
      "loss": 0.9708,
      "num_input_tokens_seen": 556246528,
      "step": 7470,
      "train_runtime": 8886.5336,
      "train_tokens_per_second": 62594.32
    },
    {
      "epoch": 0.0994013328815091,
      "grad_norm": 11.25,
      "learning_rate": 1.9748587723738364e-05,
      "loss": 0.7876,
      "num_input_tokens_seen": 556710656,
      "step": 7480,
      "train_runtime": 8893.5268,
      "train_tokens_per_second": 62597.288
    },
    {
      "epoch": 0.099534222363971,
      "grad_norm": 19.0,
      "learning_rate": 1.9747627795865118e-05,
      "loss": 0.7325,
      "num_input_tokens_seen": 556942336,
      "step": 7490,
      "train_runtime": 8897.0034,
      "train_tokens_per_second": 62598.867
    },
    {
      "epoch": 0.09966711184643291,
      "grad_norm": 23.5,
      "learning_rate": 1.9746666062330738e-05,
      "loss": 0.5797,
      "num_input_tokens_seen": 557058816,
      "step": 7500,
      "train_runtime": 8899.4102,
      "train_tokens_per_second": 62595.027
    },
    {
      "epoch": 0.09980000132889483,
      "grad_norm": 15.0625,
      "learning_rate": 1.974570252331338e-05,
      "loss": 0.8873,
      "num_input_tokens_seen": 559164928,
      "step": 7510,
      "train_runtime": 8933.1306,
      "train_tokens_per_second": 62594.51
    },
    {
      "epoch": 0.09993289081135673,
      "grad_norm": 21.375,
      "learning_rate": 1.9744737178991528e-05,
      "loss": 0.9981,
      "num_input_tokens_seen": 559999488,
      "step": 7520,
      "train_runtime": 8945.686,
      "train_tokens_per_second": 62599.949
    },
    {
      "epoch": 0.10006578029381864,
      "grad_norm": 19.375,
      "learning_rate": 1.9743770029544008e-05,
      "loss": 0.8842,
      "num_input_tokens_seen": 560475904,
      "step": 7530,
      "train_runtime": 8952.4247,
      "train_tokens_per_second": 62606.045
    },
    {
      "epoch": 0.10019866977628056,
      "grad_norm": 12.6875,
      "learning_rate": 1.9742801075149976e-05,
      "loss": 0.6283,
      "num_input_tokens_seen": 560708864,
      "step": 7540,
      "train_runtime": 8955.9092,
      "train_tokens_per_second": 62607.699
    },
    {
      "epoch": 0.10033155925874247,
      "grad_norm": 21.125,
      "learning_rate": 1.9741830315988917e-05,
      "loss": 0.6238,
      "num_input_tokens_seen": 560826368,
      "step": 7550,
      "train_runtime": 8958.6648,
      "train_tokens_per_second": 62601.558
    },
    {
      "epoch": 0.10046444874120437,
      "grad_norm": 56.5,
      "learning_rate": 1.9740857752240662e-05,
      "loss": 0.9282,
      "num_input_tokens_seen": 562946048,
      "step": 7560,
      "train_runtime": 8992.7532,
      "train_tokens_per_second": 62599.966
    },
    {
      "epoch": 0.10059733822366629,
      "grad_norm": 97.0,
      "learning_rate": 1.973988338408536e-05,
      "loss": 0.9497,
      "num_input_tokens_seen": 563788288,
      "step": 7570,
      "train_runtime": 9005.3911,
      "train_tokens_per_second": 62605.642
    },
    {
      "epoch": 0.1007302277061282,
      "grad_norm": 13.875,
      "learning_rate": 1.9738907211703513e-05,
      "loss": 0.8166,
      "num_input_tokens_seen": 564264704,
      "step": 7580,
      "train_runtime": 9012.3175,
      "train_tokens_per_second": 62610.389
    },
    {
      "epoch": 0.1008631171885901,
      "grad_norm": 10.625,
      "learning_rate": 1.9737929235275946e-05,
      "loss": 0.618,
      "num_input_tokens_seen": 564503296,
      "step": 7590,
      "train_runtime": 9015.9403,
      "train_tokens_per_second": 62611.694
    },
    {
      "epoch": 0.10099600667105202,
      "grad_norm": 14.875,
      "learning_rate": 1.9736949454983817e-05,
      "loss": 0.5786,
      "num_input_tokens_seen": 564618496,
      "step": 7600,
      "train_runtime": 9018.7472,
      "train_tokens_per_second": 62604.981
    },
    {
      "epoch": 0.10112889615351393,
      "grad_norm": 38.0,
      "learning_rate": 1.9735967871008627e-05,
      "loss": 0.9289,
      "num_input_tokens_seen": 566719232,
      "step": 7610,
      "train_runtime": 9052.878,
      "train_tokens_per_second": 62601.002
    },
    {
      "epoch": 0.10126178563597583,
      "grad_norm": 32.25,
      "learning_rate": 1.9734984483532203e-05,
      "loss": 0.9816,
      "num_input_tokens_seen": 567534592,
      "step": 7620,
      "train_runtime": 9065.1901,
      "train_tokens_per_second": 62605.923
    },
    {
      "epoch": 0.10139467511843775,
      "grad_norm": 20.875,
      "learning_rate": 1.9733999292736706e-05,
      "loss": 0.8187,
      "num_input_tokens_seen": 567971584,
      "step": 7630,
      "train_runtime": 9071.6531,
      "train_tokens_per_second": 62609.491
    },
    {
      "epoch": 0.10152756460089966,
      "grad_norm": 23.75,
      "learning_rate": 1.973301229880464e-05,
      "loss": 0.6486,
      "num_input_tokens_seen": 568204288,
      "step": 7640,
      "train_runtime": 9075.1366,
      "train_tokens_per_second": 62611.101
    },
    {
      "epoch": 0.10166045408336158,
      "grad_norm": 156.0,
      "learning_rate": 1.973202350191883e-05,
      "loss": 0.5539,
      "num_input_tokens_seen": 568318720,
      "step": 7650,
      "train_runtime": 9077.5462,
      "train_tokens_per_second": 62607.086
    },
    {
      "epoch": 0.10179334356582348,
      "grad_norm": 12.1875,
      "learning_rate": 1.9731032902262445e-05,
      "loss": 0.9174,
      "num_input_tokens_seen": 570390272,
      "step": 7660,
      "train_runtime": 9110.6991,
      "train_tokens_per_second": 62606.641
    },
    {
      "epoch": 0.1019262330482854,
      "grad_norm": 13.875,
      "learning_rate": 1.9730040500018987e-05,
      "loss": 0.8941,
      "num_input_tokens_seen": 571275520,
      "step": 7670,
      "train_runtime": 9123.4519,
      "train_tokens_per_second": 62616.16
    },
    {
      "epoch": 0.10205912253074731,
      "grad_norm": 25.375,
      "learning_rate": 1.972904629537229e-05,
      "loss": 0.7636,
      "num_input_tokens_seen": 571750912,
      "step": 7680,
      "train_runtime": 9130.2086,
      "train_tokens_per_second": 62621.89
    },
    {
      "epoch": 0.10219201201320921,
      "grad_norm": 11.8125,
      "learning_rate": 1.9728050288506515e-05,
      "loss": 0.7492,
      "num_input_tokens_seen": 571979520,
      "step": 7690,
      "train_runtime": 9133.687,
      "train_tokens_per_second": 62623.07
    },
    {
      "epoch": 0.10232490149567113,
      "grad_norm": 8.9375,
      "learning_rate": 1.9727052479606167e-05,
      "loss": 0.5437,
      "num_input_tokens_seen": 572096512,
      "step": 7700,
      "train_runtime": 9136.0598,
      "train_tokens_per_second": 62619.611
    },
    {
      "epoch": 0.10245779097813304,
      "grad_norm": 19.75,
      "learning_rate": 1.9726052868856086e-05,
      "loss": 0.9527,
      "num_input_tokens_seen": 574067200,
      "step": 7710,
      "train_runtime": 9168.0084,
      "train_tokens_per_second": 62616.347
    },
    {
      "epoch": 0.10259068046059494,
      "grad_norm": 18.375,
      "learning_rate": 1.972505145644143e-05,
      "loss": 0.9774,
      "num_input_tokens_seen": 574895360,
      "step": 7720,
      "train_runtime": 9180.6064,
      "train_tokens_per_second": 62620.63
    },
    {
      "epoch": 0.10272356994305686,
      "grad_norm": 13.0,
      "learning_rate": 1.9724048242547713e-05,
      "loss": 0.7344,
      "num_input_tokens_seen": 575361792,
      "step": 7730,
      "train_runtime": 9187.2268,
      "train_tokens_per_second": 62626.275
    },
    {
      "epoch": 0.10285645942551877,
      "grad_norm": 32.5,
      "learning_rate": 1.972304322736077e-05,
      "loss": 0.6627,
      "num_input_tokens_seen": 575599872,
      "step": 7740,
      "train_runtime": 9190.7409,
      "train_tokens_per_second": 62628.234
    },
    {
      "epoch": 0.10298934890798067,
      "grad_norm": 11.0,
      "learning_rate": 1.9722036411066763e-05,
      "loss": 0.4876,
      "num_input_tokens_seen": 575718912,
      "step": 7750,
      "train_runtime": 9193.145,
      "train_tokens_per_second": 62624.805
    },
    {
      "epoch": 0.10312223839044259,
      "grad_norm": 24.625,
      "learning_rate": 1.97210277938522e-05,
      "loss": 0.8401,
      "num_input_tokens_seen": 577797888,
      "step": 7760,
      "train_runtime": 9226.1866,
      "train_tokens_per_second": 62625.862
    },
    {
      "epoch": 0.1032551278729045,
      "grad_norm": 42.75,
      "learning_rate": 1.9720017375903922e-05,
      "loss": 1.0436,
      "num_input_tokens_seen": 578644992,
      "step": 7770,
      "train_runtime": 9239.1341,
      "train_tokens_per_second": 62629.786
    },
    {
      "epoch": 0.10338801735536642,
      "grad_norm": 24.5,
      "learning_rate": 1.9719005157409095e-05,
      "loss": 0.8324,
      "num_input_tokens_seen": 579134464,
      "step": 7780,
      "train_runtime": 9245.983,
      "train_tokens_per_second": 62636.332
    },
    {
      "epoch": 0.10352090683782832,
      "grad_norm": 26.375,
      "learning_rate": 1.9717991138555222e-05,
      "loss": 0.7241,
      "num_input_tokens_seen": 579369472,
      "step": 7790,
      "train_runtime": 9249.5211,
      "train_tokens_per_second": 62637.78
    },
    {
      "epoch": 0.10365379632029023,
      "grad_norm": 8.75,
      "learning_rate": 1.971697531953015e-05,
      "loss": 0.5348,
      "num_input_tokens_seen": 579485184,
      "step": 7800,
      "train_runtime": 9251.9083,
      "train_tokens_per_second": 62634.125
    },
    {
      "epoch": 0.10378668580275215,
      "grad_norm": 10.0,
      "learning_rate": 1.9715957700522043e-05,
      "loss": 0.8659,
      "num_input_tokens_seen": 581534208,
      "step": 7810,
      "train_runtime": 9283.429,
      "train_tokens_per_second": 62642.178
    },
    {
      "epoch": 0.10391957528521405,
      "grad_norm": 30.25,
      "learning_rate": 1.9714938281719408e-05,
      "loss": 1.0023,
      "num_input_tokens_seen": 582358784,
      "step": 7820,
      "train_runtime": 9295.6716,
      "train_tokens_per_second": 62648.382
    },
    {
      "epoch": 0.10405246476767596,
      "grad_norm": 14.6875,
      "learning_rate": 1.9713917063311082e-05,
      "loss": 0.7453,
      "num_input_tokens_seen": 582828288,
      "step": 7830,
      "train_runtime": 9302.533,
      "train_tokens_per_second": 62652.644
    },
    {
      "epoch": 0.10418535425013788,
      "grad_norm": 13.625,
      "learning_rate": 1.9712894045486238e-05,
      "loss": 0.7184,
      "num_input_tokens_seen": 583060480,
      "step": 7840,
      "train_runtime": 9306.0304,
      "train_tokens_per_second": 62654.049
    },
    {
      "epoch": 0.10431824373259978,
      "grad_norm": 12.0,
      "learning_rate": 1.9711869228434383e-05,
      "loss": 0.5504,
      "num_input_tokens_seen": 583176704,
      "step": 7850,
      "train_runtime": 9308.4489,
      "train_tokens_per_second": 62650.256
    },
    {
      "epoch": 0.10445113321506169,
      "grad_norm": 22.5,
      "learning_rate": 1.9710842612345348e-05,
      "loss": 0.8583,
      "num_input_tokens_seen": 585216000,
      "step": 7860,
      "train_runtime": 9340.1817,
      "train_tokens_per_second": 62655.74
    },
    {
      "epoch": 0.10458402269752361,
      "grad_norm": 21.25,
      "learning_rate": 1.9709814197409313e-05,
      "loss": 0.9322,
      "num_input_tokens_seen": 586034688,
      "step": 7870,
      "train_runtime": 9352.8177,
      "train_tokens_per_second": 62658.624
    },
    {
      "epoch": 0.10471691217998552,
      "grad_norm": 25.125,
      "learning_rate": 1.9708783983816775e-05,
      "loss": 0.7289,
      "num_input_tokens_seen": 586507776,
      "step": 7880,
      "train_runtime": 9359.9682,
      "train_tokens_per_second": 62661.3
    },
    {
      "epoch": 0.10484980166244742,
      "grad_norm": 20.25,
      "learning_rate": 1.9707751971758576e-05,
      "loss": 0.6971,
      "num_input_tokens_seen": 586745344,
      "step": 7890,
      "train_runtime": 9363.4572,
      "train_tokens_per_second": 62663.323
    },
    {
      "epoch": 0.10498269114490934,
      "grad_norm": 109.0,
      "learning_rate": 1.9706718161425887e-05,
      "loss": 0.6632,
      "num_input_tokens_seen": 586863872,
      "step": 7900,
      "train_runtime": 9365.8441,
      "train_tokens_per_second": 62660.009
    },
    {
      "epoch": 0.10511558062737125,
      "grad_norm": 14.375,
      "learning_rate": 1.9705682553010214e-05,
      "loss": 0.9535,
      "num_input_tokens_seen": 588866816,
      "step": 7910,
      "train_runtime": 9399.0667,
      "train_tokens_per_second": 62651.626
    },
    {
      "epoch": 0.10524847010983315,
      "grad_norm": 31.75,
      "learning_rate": 1.970464514670339e-05,
      "loss": 1.0211,
      "num_input_tokens_seen": 589706240,
      "step": 7920,
      "train_runtime": 9411.648,
      "train_tokens_per_second": 62657.065
    },
    {
      "epoch": 0.10538135959229507,
      "grad_norm": 14.0,
      "learning_rate": 1.970360594269759e-05,
      "loss": 0.7945,
      "num_input_tokens_seen": 590158592,
      "step": 7930,
      "train_runtime": 9417.9962,
      "train_tokens_per_second": 62662.862
    },
    {
      "epoch": 0.10551424907475698,
      "grad_norm": 171.0,
      "learning_rate": 1.9702564941185313e-05,
      "loss": 0.6787,
      "num_input_tokens_seen": 590388736,
      "step": 7940,
      "train_runtime": 9421.4683,
      "train_tokens_per_second": 62664.196
    },
    {
      "epoch": 0.10564713855721888,
      "grad_norm": 34.25,
      "learning_rate": 1.9701522142359395e-05,
      "loss": 0.5692,
      "num_input_tokens_seen": 590505472,
      "step": 7950,
      "train_runtime": 9423.8496,
      "train_tokens_per_second": 62660.749
    },
    {
      "epoch": 0.1057800280396808,
      "grad_norm": 23.25,
      "learning_rate": 1.970047754641301e-05,
      "loss": 0.9458,
      "num_input_tokens_seen": 592671488,
      "step": 7960,
      "train_runtime": 9458.7256,
      "train_tokens_per_second": 62658.704
    },
    {
      "epoch": 0.10591291752214271,
      "grad_norm": 18.875,
      "learning_rate": 1.969943115353965e-05,
      "loss": 1.0127,
      "num_input_tokens_seen": 593521920,
      "step": 7970,
      "train_runtime": 9471.5001,
      "train_tokens_per_second": 62663.983
    },
    {
      "epoch": 0.10604580700460461,
      "grad_norm": 13.5,
      "learning_rate": 1.9698382963933165e-05,
      "loss": 0.8048,
      "num_input_tokens_seen": 593986304,
      "step": 7980,
      "train_runtime": 9478.0094,
      "train_tokens_per_second": 62669.943
    },
    {
      "epoch": 0.10617869648706653,
      "grad_norm": 26.625,
      "learning_rate": 1.9697332977787712e-05,
      "loss": 0.663,
      "num_input_tokens_seen": 594209536,
      "step": 7990,
      "train_runtime": 9481.6844,
      "train_tokens_per_second": 62669.196
    },
    {
      "epoch": 0.10631158596952844,
      "grad_norm": 16.375,
      "learning_rate": 1.9696281195297793e-05,
      "loss": 0.5984,
      "num_input_tokens_seen": 594322432,
      "step": 8000,
      "train_runtime": 9484.0269,
      "train_tokens_per_second": 62665.621
    },
    {
      "epoch": 0.10644447545199036,
      "grad_norm": 31.5,
      "learning_rate": 1.9695227616658246e-05,
      "loss": 0.8075,
      "num_input_tokens_seen": 596317440,
      "step": 8010,
      "train_runtime": 9517.027,
      "train_tokens_per_second": 62657.954
    },
    {
      "epoch": 0.10657736493445226,
      "grad_norm": 12.125,
      "learning_rate": 1.9694172242064234e-05,
      "loss": 0.9543,
      "num_input_tokens_seen": 597150976,
      "step": 8020,
      "train_runtime": 9529.4621,
      "train_tokens_per_second": 62663.661
    },
    {
      "epoch": 0.10671025441691417,
      "grad_norm": 29.75,
      "learning_rate": 1.969311507171125e-05,
      "loss": 0.7848,
      "num_input_tokens_seen": 597607168,
      "step": 8030,
      "train_runtime": 9536.2195,
      "train_tokens_per_second": 62667.094
    },
    {
      "epoch": 0.10684314389937609,
      "grad_norm": 25.5,
      "learning_rate": 1.9692056105795134e-05,
      "loss": 0.6793,
      "num_input_tokens_seen": 597832192,
      "step": 8040,
      "train_runtime": 9539.6037,
      "train_tokens_per_second": 62668.451
    },
    {
      "epoch": 0.10697603338183799,
      "grad_norm": 94.0,
      "learning_rate": 1.969099534451205e-05,
      "loss": 0.5655,
      "num_input_tokens_seen": 597945088,
      "step": 8050,
      "train_runtime": 9542.0624,
      "train_tokens_per_second": 62664.135
    },
    {
      "epoch": 0.1071089228642999,
      "grad_norm": 16.0,
      "learning_rate": 1.968993278805849e-05,
      "loss": 0.8555,
      "num_input_tokens_seen": 600084480,
      "step": 8060,
      "train_runtime": 9577.759,
      "train_tokens_per_second": 62653.955
    },
    {
      "epoch": 0.10724181234676182,
      "grad_norm": 29.25,
      "learning_rate": 1.968886843663128e-05,
      "loss": 1.0326,
      "num_input_tokens_seen": 600957440,
      "step": 8070,
      "train_runtime": 9590.5996,
      "train_tokens_per_second": 62661.092
    },
    {
      "epoch": 0.10737470182922372,
      "grad_norm": 9.75,
      "learning_rate": 1.968780229042759e-05,
      "loss": 0.7585,
      "num_input_tokens_seen": 601411584,
      "step": 8080,
      "train_runtime": 9597.179,
      "train_tokens_per_second": 62665.454
    },
    {
      "epoch": 0.10750759131168564,
      "grad_norm": 36.75,
      "learning_rate": 1.9686734349644913e-05,
      "loss": 0.6684,
      "num_input_tokens_seen": 601633024,
      "step": 8090,
      "train_runtime": 9600.9936,
      "train_tokens_per_second": 62663.621
    },
    {
      "epoch": 0.10764048079414755,
      "grad_norm": 35.5,
      "learning_rate": 1.9685664614481066e-05,
      "loss": 0.567,
      "num_input_tokens_seen": 601747456,
      "step": 8100,
      "train_runtime": 9603.3962,
      "train_tokens_per_second": 62659.859
    },
    {
      "epoch": 0.10777337027660945,
      "grad_norm": 21.125,
      "learning_rate": 1.968459308513422e-05,
      "loss": 0.9149,
      "num_input_tokens_seen": 603776256,
      "step": 8110,
      "train_runtime": 9636.9165,
      "train_tokens_per_second": 62652.432
    },
    {
      "epoch": 0.10790625975907137,
      "grad_norm": 24.75,
      "learning_rate": 1.968351976180286e-05,
      "loss": 0.8837,
      "num_input_tokens_seen": 604628736,
      "step": 8120,
      "train_runtime": 9649.763,
      "train_tokens_per_second": 62657.367
    },
    {
      "epoch": 0.10803914924153328,
      "grad_norm": 13.4375,
      "learning_rate": 1.968244464468581e-05,
      "loss": 0.8211,
      "num_input_tokens_seen": 605100288,
      "step": 8130,
      "train_runtime": 9656.4995,
      "train_tokens_per_second": 62662.489
    },
    {
      "epoch": 0.1081720387239952,
      "grad_norm": 13.4375,
      "learning_rate": 1.9681367733982225e-05,
      "loss": 0.6736,
      "num_input_tokens_seen": 605331712,
      "step": 8140,
      "train_runtime": 9660.3967,
      "train_tokens_per_second": 62661.165
    },
    {
      "epoch": 0.1083049282064571,
      "grad_norm": 9.9375,
      "learning_rate": 1.9680289029891595e-05,
      "loss": 0.506,
      "num_input_tokens_seen": 605446656,
      "step": 8150,
      "train_runtime": 9662.7353,
      "train_tokens_per_second": 62657.895
    },
    {
      "epoch": 0.10843781768891901,
      "grad_norm": 282.0,
      "learning_rate": 1.967920853261374e-05,
      "loss": 0.9159,
      "num_input_tokens_seen": 607405056,
      "step": 8160,
      "train_runtime": 9694.0619,
      "train_tokens_per_second": 62657.435
    },
    {
      "epoch": 0.10857070717138093,
      "grad_norm": 36.5,
      "learning_rate": 1.9678126242348813e-05,
      "loss": 0.9592,
      "num_input_tokens_seen": 608231424,
      "step": 8170,
      "train_runtime": 9706.1588,
      "train_tokens_per_second": 62664.483
    },
    {
      "epoch": 0.10870359665384283,
      "grad_norm": 20.375,
      "learning_rate": 1.96770421592973e-05,
      "loss": 0.7623,
      "num_input_tokens_seen": 608670976,
      "step": 8180,
      "train_runtime": 9712.5191,
      "train_tokens_per_second": 62668.703
    },
    {
      "epoch": 0.10883648613630474,
      "grad_norm": 18.25,
      "learning_rate": 1.967595628366001e-05,
      "loss": 0.6484,
      "num_input_tokens_seen": 608904192,
      "step": 8190,
      "train_runtime": 9715.9875,
      "train_tokens_per_second": 62670.335
    },
    {
      "epoch": 0.10896937561876666,
      "grad_norm": 16.25,
      "learning_rate": 1.9674868615638098e-05,
      "loss": 0.54,
      "num_input_tokens_seen": 609020416,
      "step": 8200,
      "train_runtime": 9718.4182,
      "train_tokens_per_second": 62666.62
    },
    {
      "epoch": 0.10910226510122856,
      "grad_norm": 26.375,
      "learning_rate": 1.9673779155433047e-05,
      "loss": 0.8638,
      "num_input_tokens_seen": 611111168,
      "step": 8210,
      "train_runtime": 9753.2889,
      "train_tokens_per_second": 62656.933
    },
    {
      "epoch": 0.10923515458369047,
      "grad_norm": 18.375,
      "learning_rate": 1.9672687903246665e-05,
      "loss": 0.9825,
      "num_input_tokens_seen": 611947264,
      "step": 8220,
      "train_runtime": 9766.1358,
      "train_tokens_per_second": 62660.123
    },
    {
      "epoch": 0.10936804406615239,
      "grad_norm": 23.25,
      "learning_rate": 1.96715948592811e-05,
      "loss": 0.7651,
      "num_input_tokens_seen": 612407552,
      "step": 8230,
      "train_runtime": 9772.6643,
      "train_tokens_per_second": 62665.363
    },
    {
      "epoch": 0.1095009335486143,
      "grad_norm": 17.375,
      "learning_rate": 1.9670500023738828e-05,
      "loss": 0.6811,
      "num_input_tokens_seen": 612645376,
      "step": 8240,
      "train_runtime": 9776.1273,
      "train_tokens_per_second": 62667.492
    },
    {
      "epoch": 0.1096338230310762,
      "grad_norm": 8.5,
      "learning_rate": 1.9669403396822657e-05,
      "loss": 0.5559,
      "num_input_tokens_seen": 612761600,
      "step": 8250,
      "train_runtime": 9778.4797,
      "train_tokens_per_second": 62664.302
    },
    {
      "epoch": 0.10976671251353812,
      "grad_norm": 16.5,
      "learning_rate": 1.9668304978735725e-05,
      "loss": 0.8653,
      "num_input_tokens_seen": 614893824,
      "step": 8260,
      "train_runtime": 9813.3356,
      "train_tokens_per_second": 62659.003
    },
    {
      "epoch": 0.10989960199600003,
      "grad_norm": 21.75,
      "learning_rate": 1.966720476968151e-05,
      "loss": 0.9126,
      "num_input_tokens_seen": 615738112,
      "step": 8270,
      "train_runtime": 9825.7769,
      "train_tokens_per_second": 62665.591
    },
    {
      "epoch": 0.11003249147846193,
      "grad_norm": 11.625,
      "learning_rate": 1.9666102769863807e-05,
      "loss": 0.763,
      "num_input_tokens_seen": 616219136,
      "step": 8280,
      "train_runtime": 9832.8359,
      "train_tokens_per_second": 62669.523
    },
    {
      "epoch": 0.11016538096092385,
      "grad_norm": 15.4375,
      "learning_rate": 1.9664998979486762e-05,
      "loss": 0.675,
      "num_input_tokens_seen": 616446720,
      "step": 8290,
      "train_runtime": 9836.323,
      "train_tokens_per_second": 62670.443
    },
    {
      "epoch": 0.11029827044338576,
      "grad_norm": 11.9375,
      "learning_rate": 1.9663893398754835e-05,
      "loss": 0.6056,
      "num_input_tokens_seen": 616563200,
      "step": 8300,
      "train_runtime": 9838.737,
      "train_tokens_per_second": 62666.905
    },
    {
      "epoch": 0.11043115992584766,
      "grad_norm": 19.5,
      "learning_rate": 1.9662786027872828e-05,
      "loss": 1.0071,
      "num_input_tokens_seen": 618700288,
      "step": 8310,
      "train_runtime": 9873.5532,
      "train_tokens_per_second": 62662.374
    },
    {
      "epoch": 0.11056404940830958,
      "grad_norm": 14.0625,
      "learning_rate": 1.966167686704587e-05,
      "loss": 0.9133,
      "num_input_tokens_seen": 619576064,
      "step": 8320,
      "train_runtime": 9886.3937,
      "train_tokens_per_second": 62669.572
    },
    {
      "epoch": 0.11069693889077149,
      "grad_norm": 21.875,
      "learning_rate": 1.9660565916479424e-05,
      "loss": 0.835,
      "num_input_tokens_seen": 620052736,
      "step": 8330,
      "train_runtime": 9893.6197,
      "train_tokens_per_second": 62671.98
    },
    {
      "epoch": 0.1108298283732334,
      "grad_norm": 12.125,
      "learning_rate": 1.9659453176379282e-05,
      "loss": 0.6261,
      "num_input_tokens_seen": 620285952,
      "step": 8340,
      "train_runtime": 9897.0912,
      "train_tokens_per_second": 62673.561
    },
    {
      "epoch": 0.11096271785569531,
      "grad_norm": 16.25,
      "learning_rate": 1.9658338646951573e-05,
      "loss": 0.5688,
      "num_input_tokens_seen": 620398080,
      "step": 8350,
      "train_runtime": 9899.4646,
      "train_tokens_per_second": 62669.862
    },
    {
      "epoch": 0.11109560733815722,
      "grad_norm": 19.75,
      "learning_rate": 1.9657222328402753e-05,
      "loss": 0.8939,
      "num_input_tokens_seen": 622377472,
      "step": 8360,
      "train_runtime": 9931.7836,
      "train_tokens_per_second": 62665.227
    },
    {
      "epoch": 0.11122849682061914,
      "grad_norm": 30.375,
      "learning_rate": 1.9656104220939604e-05,
      "loss": 0.9725,
      "num_input_tokens_seen": 623209728,
      "step": 8370,
      "train_runtime": 9944.5017,
      "train_tokens_per_second": 62668.774
    },
    {
      "epoch": 0.11136138630308104,
      "grad_norm": 14.875,
      "learning_rate": 1.9654984324769254e-05,
      "loss": 0.761,
      "num_input_tokens_seen": 623686400,
      "step": 8380,
      "train_runtime": 9951.3199,
      "train_tokens_per_second": 62673.736
    },
    {
      "epoch": 0.11149427578554295,
      "grad_norm": 19.0,
      "learning_rate": 1.9653862640099147e-05,
      "loss": 0.679,
      "num_input_tokens_seen": 623926016,
      "step": 8390,
      "train_runtime": 9954.8922,
      "train_tokens_per_second": 62675.316
    },
    {
      "epoch": 0.11162716526800487,
      "grad_norm": 17.375,
      "learning_rate": 1.9652739167137072e-05,
      "loss": 0.5867,
      "num_input_tokens_seen": 624044544,
      "step": 8400,
      "train_runtime": 9957.3022,
      "train_tokens_per_second": 62672.05
    },
    {
      "epoch": 0.11176005475046677,
      "grad_norm": 41.0,
      "learning_rate": 1.9651613906091133e-05,
      "loss": 0.8508,
      "num_input_tokens_seen": 626138880,
      "step": 8410,
      "train_runtime": 9990.96,
      "train_tokens_per_second": 62670.542
    },
    {
      "epoch": 0.11189294423292868,
      "grad_norm": 17.75,
      "learning_rate": 1.9650486857169783e-05,
      "loss": 0.9379,
      "num_input_tokens_seen": 626969088,
      "step": 8420,
      "train_runtime": 10003.4706,
      "train_tokens_per_second": 62675.157
    },
    {
      "epoch": 0.1120258337153906,
      "grad_norm": 20.125,
      "learning_rate": 1.9649358020581793e-05,
      "loss": 0.813,
      "num_input_tokens_seen": 627458304,
      "step": 8430,
      "train_runtime": 10010.1571,
      "train_tokens_per_second": 62682.163
    },
    {
      "epoch": 0.1121587231978525,
      "grad_norm": 17.5,
      "learning_rate": 1.9648227396536272e-05,
      "loss": 0.6592,
      "num_input_tokens_seen": 627680512,
      "step": 8440,
      "train_runtime": 10013.6204,
      "train_tokens_per_second": 62682.675
    },
    {
      "epoch": 0.11229161268031441,
      "grad_norm": 22.75,
      "learning_rate": 1.964709498524266e-05,
      "loss": 0.5535,
      "num_input_tokens_seen": 627794176,
      "step": 8450,
      "train_runtime": 10016.3561,
      "train_tokens_per_second": 62676.902
    },
    {
      "epoch": 0.11242450216277633,
      "grad_norm": 14.9375,
      "learning_rate": 1.964596078691072e-05,
      "loss": 0.8525,
      "num_input_tokens_seen": 629887744,
      "step": 8460,
      "train_runtime": 10050.3721,
      "train_tokens_per_second": 62673.077
    },
    {
      "epoch": 0.11255739164523824,
      "grad_norm": 34.75,
      "learning_rate": 1.9644824801750557e-05,
      "loss": 0.948,
      "num_input_tokens_seen": 630760448,
      "step": 8470,
      "train_runtime": 10063.3332,
      "train_tokens_per_second": 62679.078
    },
    {
      "epoch": 0.11269028112770015,
      "grad_norm": 17.25,
      "learning_rate": 1.96436870299726e-05,
      "loss": 0.756,
      "num_input_tokens_seen": 631240448,
      "step": 8480,
      "train_runtime": 10070.2173,
      "train_tokens_per_second": 62683.895
    },
    {
      "epoch": 0.11282317061016206,
      "grad_norm": 11.75,
      "learning_rate": 1.9642547471787613e-05,
      "loss": 0.6957,
      "num_input_tokens_seen": 631468544,
      "step": 8490,
      "train_runtime": 10073.6198,
      "train_tokens_per_second": 62685.366
    },
    {
      "epoch": 0.11295606009262397,
      "grad_norm": 31.25,
      "learning_rate": 1.9641406127406685e-05,
      "loss": 0.5839,
      "num_input_tokens_seen": 631583744,
      "step": 8500,
      "train_runtime": 10075.9496,
      "train_tokens_per_second": 62682.305
    },
    {
      "epoch": 0.11308894957508588,
      "grad_norm": 17.25,
      "learning_rate": 1.9640262997041247e-05,
      "loss": 0.8174,
      "num_input_tokens_seen": 633622016,
      "step": 8510,
      "train_runtime": 10110.3094,
      "train_tokens_per_second": 62670.883
    },
    {
      "epoch": 0.11322183905754779,
      "grad_norm": 13.75,
      "learning_rate": 1.9639118080903046e-05,
      "loss": 0.9377,
      "num_input_tokens_seen": 634469888,
      "step": 8520,
      "train_runtime": 10122.9006,
      "train_tokens_per_second": 62676.689
    },
    {
      "epoch": 0.1133547285400097,
      "grad_norm": 12.875,
      "learning_rate": 1.9637971379204173e-05,
      "loss": 0.7953,
      "num_input_tokens_seen": 634908160,
      "step": 8530,
      "train_runtime": 10129.4716,
      "train_tokens_per_second": 62679.297
    },
    {
      "epoch": 0.1134876180224716,
      "grad_norm": 24.875,
      "learning_rate": 1.963682289215704e-05,
      "loss": 0.6844,
      "num_input_tokens_seen": 635130112,
      "step": 8540,
      "train_runtime": 10132.8479,
      "train_tokens_per_second": 62680.317
    },
    {
      "epoch": 0.11362050750493352,
      "grad_norm": 24.375,
      "learning_rate": 1.9635672619974398e-05,
      "loss": 0.5467,
      "num_input_tokens_seen": 635243776,
      "step": 8550,
      "train_runtime": 10135.2298,
      "train_tokens_per_second": 62676.8
    },
    {
      "epoch": 0.11375339698739544,
      "grad_norm": 12.3125,
      "learning_rate": 1.9634520562869322e-05,
      "loss": 0.8735,
      "num_input_tokens_seen": 637249792,
      "step": 8560,
      "train_runtime": 10168.0993,
      "train_tokens_per_second": 62671.476
    },
    {
      "epoch": 0.11388628646985734,
      "grad_norm": 33.75,
      "learning_rate": 1.9633366721055226e-05,
      "loss": 0.9405,
      "num_input_tokens_seen": 638059776,
      "step": 8570,
      "train_runtime": 10180.4141,
      "train_tokens_per_second": 62675.228
    },
    {
      "epoch": 0.11401917595231925,
      "grad_norm": 31.75,
      "learning_rate": 1.9632211094745838e-05,
      "loss": 0.8249,
      "num_input_tokens_seen": 638489856,
      "step": 8580,
      "train_runtime": 10186.6021,
      "train_tokens_per_second": 62679.376
    },
    {
      "epoch": 0.11415206543478117,
      "grad_norm": 24.75,
      "learning_rate": 1.9631053684155234e-05,
      "loss": 0.6775,
      "num_input_tokens_seen": 638699264,
      "step": 8590,
      "train_runtime": 10189.8533,
      "train_tokens_per_second": 62679.927
    },
    {
      "epoch": 0.11428495491724308,
      "grad_norm": 18.25,
      "learning_rate": 1.9629894489497813e-05,
      "loss": 0.5298,
      "num_input_tokens_seen": 638809344,
      "step": 8600,
      "train_runtime": 10192.2252,
      "train_tokens_per_second": 62676.141
    },
    {
      "epoch": 0.11441784439970498,
      "grad_norm": 23.625,
      "learning_rate": 1.962873351098831e-05,
      "loss": 0.8963,
      "num_input_tokens_seen": 640791296,
      "step": 8610,
      "train_runtime": 10224.6185,
      "train_tokens_per_second": 62671.414
    },
    {
      "epoch": 0.1145507338821669,
      "grad_norm": 23.375,
      "learning_rate": 1.962757074884178e-05,
      "loss": 0.9289,
      "num_input_tokens_seen": 641607168,
      "step": 8620,
      "train_runtime": 10236.7748,
      "train_tokens_per_second": 62676.691
    },
    {
      "epoch": 0.11468362336462881,
      "grad_norm": 22.25,
      "learning_rate": 1.9626406203273615e-05,
      "loss": 0.7651,
      "num_input_tokens_seen": 642074368,
      "step": 8630,
      "train_runtime": 10243.5702,
      "train_tokens_per_second": 62680.721
    },
    {
      "epoch": 0.11481651284709071,
      "grad_norm": 20.125,
      "learning_rate": 1.962523987449954e-05,
      "loss": 0.6611,
      "num_input_tokens_seen": 642309888,
      "step": 8640,
      "train_runtime": 10247.0293,
      "train_tokens_per_second": 62682.546
    },
    {
      "epoch": 0.11494940232955263,
      "grad_norm": 13.8125,
      "learning_rate": 1.9624071762735603e-05,
      "loss": 0.5594,
      "num_input_tokens_seen": 642419456,
      "step": 8650,
      "train_runtime": 10249.4222,
      "train_tokens_per_second": 62678.602
    },
    {
      "epoch": 0.11508229181201454,
      "grad_norm": 17.5,
      "learning_rate": 1.962290186819819e-05,
      "loss": 0.9047,
      "num_input_tokens_seen": 644443648,
      "step": 8660,
      "train_runtime": 10282.5659,
      "train_tokens_per_second": 62673.427
    },
    {
      "epoch": 0.11521518129447644,
      "grad_norm": 18.125,
      "learning_rate": 1.962173019110401e-05,
      "loss": 0.9904,
      "num_input_tokens_seen": 645286912,
      "step": 8670,
      "train_runtime": 10295.4469,
      "train_tokens_per_second": 62676.921
    },
    {
      "epoch": 0.11534807077693836,
      "grad_norm": 18.625,
      "learning_rate": 1.9620556731670112e-05,
      "loss": 0.7439,
      "num_input_tokens_seen": 645794048,
      "step": 8680,
      "train_runtime": 10303.0145,
      "train_tokens_per_second": 62680.107
    },
    {
      "epoch": 0.11548096025940027,
      "grad_norm": 23.625,
      "learning_rate": 1.9619381490113858e-05,
      "loss": 0.6812,
      "num_input_tokens_seen": 646034944,
      "step": 8690,
      "train_runtime": 10306.622,
      "train_tokens_per_second": 62681.541
    },
    {
      "epoch": 0.11561384974186217,
      "grad_norm": 10.5,
      "learning_rate": 1.961820446665296e-05,
      "loss": 0.5114,
      "num_input_tokens_seen": 646154240,
      "step": 8700,
      "train_runtime": 10309.0158,
      "train_tokens_per_second": 62678.558
    },
    {
      "epoch": 0.11574673922432409,
      "grad_norm": 48.5,
      "learning_rate": 1.9617025661505448e-05,
      "loss": 0.9214,
      "num_input_tokens_seen": 648285184,
      "step": 8710,
      "train_runtime": 10343.9021,
      "train_tokens_per_second": 62673.175
    },
    {
      "epoch": 0.115879628706786,
      "grad_norm": 25.125,
      "learning_rate": 1.961584507488969e-05,
      "loss": 0.9997,
      "num_input_tokens_seen": 649170176,
      "step": 8720,
      "train_runtime": 10356.9926,
      "train_tokens_per_second": 62679.409
    },
    {
      "epoch": 0.11601251818924792,
      "grad_norm": 28.25,
      "learning_rate": 1.9614662707024367e-05,
      "loss": 0.783,
      "num_input_tokens_seen": 649627392,
      "step": 8730,
      "train_runtime": 10364.1651,
      "train_tokens_per_second": 62680.147
    },
    {
      "epoch": 0.11614540767170982,
      "grad_norm": 15.4375,
      "learning_rate": 1.9613478558128516e-05,
      "loss": 0.712,
      "num_input_tokens_seen": 649864960,
      "step": 8740,
      "train_runtime": 10367.7212,
      "train_tokens_per_second": 62681.562
    },
    {
      "epoch": 0.11627829715417173,
      "grad_norm": 29.875,
      "learning_rate": 1.9612292628421482e-05,
      "loss": 0.4981,
      "num_input_tokens_seen": 649982208,
      "step": 8750,
      "train_runtime": 10370.1351,
      "train_tokens_per_second": 62678.278
    },
    {
      "epoch": 0.11641118663663365,
      "grad_norm": 16.75,
      "learning_rate": 1.961110491812295e-05,
      "loss": 0.88,
      "num_input_tokens_seen": 651924992,
      "step": 8760,
      "train_runtime": 10400.6454,
      "train_tokens_per_second": 62681.206
    },
    {
      "epoch": 0.11654407611909555,
      "grad_norm": 20.5,
      "learning_rate": 1.960991542745293e-05,
      "loss": 0.9031,
      "num_input_tokens_seen": 652768512,
      "step": 8770,
      "train_runtime": 10413.2447,
      "train_tokens_per_second": 62686.37
    },
    {
      "epoch": 0.11667696560155746,
      "grad_norm": 16.125,
      "learning_rate": 1.960872415663177e-05,
      "loss": 0.9127,
      "num_input_tokens_seen": 653250304,
      "step": 8780,
      "train_runtime": 10420.0428,
      "train_tokens_per_second": 62691.71
    },
    {
      "epoch": 0.11680985508401938,
      "grad_norm": 14.875,
      "learning_rate": 1.960753110588014e-05,
      "loss": 0.6785,
      "num_input_tokens_seen": 653488640,
      "step": 8790,
      "train_runtime": 10423.5537,
      "train_tokens_per_second": 62693.459
    },
    {
      "epoch": 0.11694274456648128,
      "grad_norm": 11.0625,
      "learning_rate": 1.960633627541904e-05,
      "loss": 0.6141,
      "num_input_tokens_seen": 653606144,
      "step": 8800,
      "train_runtime": 10426.3716,
      "train_tokens_per_second": 62687.785
    },
    {
      "epoch": 0.1170756340489432,
      "grad_norm": 22.0,
      "learning_rate": 1.9605139665469805e-05,
      "loss": 0.8824,
      "num_input_tokens_seen": 655663360,
      "step": 8810,
      "train_runtime": 10458.9897,
      "train_tokens_per_second": 62688.977
    },
    {
      "epoch": 0.11720852353140511,
      "grad_norm": 46.25,
      "learning_rate": 1.960394127625409e-05,
      "loss": 0.9146,
      "num_input_tokens_seen": 656500480,
      "step": 8820,
      "train_runtime": 10471.6772,
      "train_tokens_per_second": 62692.964
    },
    {
      "epoch": 0.11734141301386702,
      "grad_norm": 10.5,
      "learning_rate": 1.9602741107993895e-05,
      "loss": 0.7749,
      "num_input_tokens_seen": 656956160,
      "step": 8830,
      "train_runtime": 10478.2282,
      "train_tokens_per_second": 62697.256
    },
    {
      "epoch": 0.11747430249632893,
      "grad_norm": 12.4375,
      "learning_rate": 1.9601539160911532e-05,
      "loss": 0.7225,
      "num_input_tokens_seen": 657181184,
      "step": 8840,
      "train_runtime": 10481.6322,
      "train_tokens_per_second": 62698.363
    },
    {
      "epoch": 0.11760719197879084,
      "grad_norm": 15.6875,
      "learning_rate": 1.9600335435229657e-05,
      "loss": 0.5683,
      "num_input_tokens_seen": 657294336,
      "step": 8850,
      "train_runtime": 10483.967,
      "train_tokens_per_second": 62695.193
    },
    {
      "epoch": 0.11774008146125275,
      "grad_norm": 27.375,
      "learning_rate": 1.9599129931171247e-05,
      "loss": 0.8738,
      "num_input_tokens_seen": 659400192,
      "step": 8860,
      "train_runtime": 10518.6427,
      "train_tokens_per_second": 62688.715
    },
    {
      "epoch": 0.11787297094371466,
      "grad_norm": 19.875,
      "learning_rate": 1.959792264895961e-05,
      "loss": 0.9838,
      "num_input_tokens_seen": 660221696,
      "step": 8870,
      "train_runtime": 10531.4312,
      "train_tokens_per_second": 62690.596
    },
    {
      "epoch": 0.11800586042617657,
      "grad_norm": 18.25,
      "learning_rate": 1.959671358881839e-05,
      "loss": 0.7875,
      "num_input_tokens_seen": 660739584,
      "step": 8880,
      "train_runtime": 10538.3,
      "train_tokens_per_second": 62698.878
    },
    {
      "epoch": 0.11813874990863849,
      "grad_norm": 18.625,
      "learning_rate": 1.959550275097154e-05,
      "loss": 0.7014,
      "num_input_tokens_seen": 660975872,
      "step": 8890,
      "train_runtime": 10541.9037,
      "train_tokens_per_second": 62699.859
    },
    {
      "epoch": 0.11827163939110039,
      "grad_norm": 10.8125,
      "learning_rate": 1.9594290135643377e-05,
      "loss": 0.639,
      "num_input_tokens_seen": 661095680,
      "step": 8900,
      "train_runtime": 10544.4151,
      "train_tokens_per_second": 62696.288
    },
    {
      "epoch": 0.1184045288735623,
      "grad_norm": 21.25,
      "learning_rate": 1.9593075743058513e-05,
      "loss": 0.8287,
      "num_input_tokens_seen": 663194624,
      "step": 8910,
      "train_runtime": 10579.0263,
      "train_tokens_per_second": 62689.571
    },
    {
      "epoch": 0.11853741835602422,
      "grad_norm": 17.625,
      "learning_rate": 1.959185957344191e-05,
      "loss": 1.0096,
      "num_input_tokens_seen": 664041984,
      "step": 8920,
      "train_runtime": 10591.7256,
      "train_tokens_per_second": 62694.41
    },
    {
      "epoch": 0.11867030783848612,
      "grad_norm": 24.25,
      "learning_rate": 1.9590641627018847e-05,
      "loss": 0.7721,
      "num_input_tokens_seen": 664515584,
      "step": 8930,
      "train_runtime": 10598.3912,
      "train_tokens_per_second": 62699.666
    },
    {
      "epoch": 0.11880319732094803,
      "grad_norm": 45.25,
      "learning_rate": 1.958942190401494e-05,
      "loss": 0.7089,
      "num_input_tokens_seen": 664738048,
      "step": 8940,
      "train_runtime": 10601.795,
      "train_tokens_per_second": 62700.519
    },
    {
      "epoch": 0.11893608680340995,
      "grad_norm": 13.75,
      "learning_rate": 1.9588200404656133e-05,
      "loss": 0.576,
      "num_input_tokens_seen": 664852480,
      "step": 8950,
      "train_runtime": 10604.1673,
      "train_tokens_per_second": 62697.283
    },
    {
      "epoch": 0.11906897628587186,
      "grad_norm": 14.4375,
      "learning_rate": 1.95869771291687e-05,
      "loss": 0.8293,
      "num_input_tokens_seen": 667034368,
      "step": 8960,
      "train_runtime": 10638.4335,
      "train_tokens_per_second": 62700.431
    },
    {
      "epoch": 0.11920186576833376,
      "grad_norm": 20.375,
      "learning_rate": 1.9585752077779238e-05,
      "loss": 0.8949,
      "num_input_tokens_seen": 667905024,
      "step": 8970,
      "train_runtime": 10651.5789,
      "train_tokens_per_second": 62704.79
    },
    {
      "epoch": 0.11933475525079568,
      "grad_norm": 23.5,
      "learning_rate": 1.9584525250714678e-05,
      "loss": 0.7774,
      "num_input_tokens_seen": 668403200,
      "step": 8980,
      "train_runtime": 10658.5241,
      "train_tokens_per_second": 62710.671
    },
    {
      "epoch": 0.11946764473325759,
      "grad_norm": 21.75,
      "learning_rate": 1.9583296648202278e-05,
      "loss": 0.7399,
      "num_input_tokens_seen": 668644096,
      "step": 8990,
      "train_runtime": 10662.5554,
      "train_tokens_per_second": 62709.554
    },
    {
      "epoch": 0.11960053421571949,
      "grad_norm": 18.75,
      "learning_rate": 1.958206627046963e-05,
      "loss": 0.5036,
      "num_input_tokens_seen": 668765184,
      "step": 9000,
      "train_runtime": 10665.0266,
      "train_tokens_per_second": 62706.377
    },
    {
      "epoch": 0.11973342369818141,
      "grad_norm": 45.25,
      "learning_rate": 1.9580834117744644e-05,
      "loss": 0.9217,
      "num_input_tokens_seen": 670715392,
      "step": 9010,
      "train_runtime": 10696.6525,
      "train_tokens_per_second": 62703.298
    },
    {
      "epoch": 0.11986631318064332,
      "grad_norm": 17.375,
      "learning_rate": 1.9579600190255568e-05,
      "loss": 1.0201,
      "num_input_tokens_seen": 671546112,
      "step": 9020,
      "train_runtime": 10709.5515,
      "train_tokens_per_second": 62705.344
    },
    {
      "epoch": 0.11999920266310522,
      "grad_norm": 37.75,
      "learning_rate": 1.957836448823098e-05,
      "loss": 0.8274,
      "num_input_tokens_seen": 672025088,
      "step": 9030,
      "train_runtime": 10716.4326,
      "train_tokens_per_second": 62709.776
    },
    {
      "epoch": 0.12013209214556714,
      "grad_norm": 24.875,
      "learning_rate": 1.9577127011899776e-05,
      "loss": 0.6304,
      "num_input_tokens_seen": 672261120,
      "step": 9040,
      "train_runtime": 10719.988,
      "train_tokens_per_second": 62710.996
    },
    {
      "epoch": 0.12026498162802905,
      "grad_norm": 16.125,
      "learning_rate": 1.9575887761491193e-05,
      "loss": 0.5774,
      "num_input_tokens_seen": 672380672,
      "step": 9050,
      "train_runtime": 10722.8082,
      "train_tokens_per_second": 62705.651
    },
    {
      "epoch": 0.12039787111049097,
      "grad_norm": 24.5,
      "learning_rate": 1.9574646737234787e-05,
      "loss": 0.8595,
      "num_input_tokens_seen": 674354944,
      "step": 9060,
      "train_runtime": 10754.3765,
      "train_tokens_per_second": 62705.164
    },
    {
      "epoch": 0.12053076059295287,
      "grad_norm": 20.75,
      "learning_rate": 1.9573403939360446e-05,
      "loss": 0.8897,
      "num_input_tokens_seen": 675191552,
      "step": 9070,
      "train_runtime": 10766.7697,
      "train_tokens_per_second": 62710.69
    },
    {
      "epoch": 0.12066365007541478,
      "grad_norm": 19.0,
      "learning_rate": 1.9572159368098394e-05,
      "loss": 0.7964,
      "num_input_tokens_seen": 675624448,
      "step": 9080,
      "train_runtime": 10773.0785,
      "train_tokens_per_second": 62714.149
    },
    {
      "epoch": 0.1207965395578767,
      "grad_norm": 14.6875,
      "learning_rate": 1.9570913023679173e-05,
      "loss": 0.6528,
      "num_input_tokens_seen": 675855360,
      "step": 9090,
      "train_runtime": 10776.521,
      "train_tokens_per_second": 62715.542
    },
    {
      "epoch": 0.1209294290403386,
      "grad_norm": 15.875,
      "learning_rate": 1.956966490633365e-05,
      "loss": 0.547,
      "num_input_tokens_seen": 675970048,
      "step": 9100,
      "train_runtime": 10779.3149,
      "train_tokens_per_second": 62709.927
    },
    {
      "epoch": 0.12106231852280051,
      "grad_norm": 50.0,
      "learning_rate": 1.9568415016293043e-05,
      "loss": 0.8638,
      "num_input_tokens_seen": 677980160,
      "step": 9110,
      "train_runtime": 10811.8384,
      "train_tokens_per_second": 62707.204
    },
    {
      "epoch": 0.12119520800526243,
      "grad_norm": 42.75,
      "learning_rate": 1.956716335378887e-05,
      "loss": 0.9574,
      "num_input_tokens_seen": 678822912,
      "step": 9120,
      "train_runtime": 10824.5195,
      "train_tokens_per_second": 62711.598
    },
    {
      "epoch": 0.12132809748772433,
      "grad_norm": 42.25,
      "learning_rate": 1.956590991905299e-05,
      "loss": 0.7679,
      "num_input_tokens_seen": 679303680,
      "step": 9130,
      "train_runtime": 10831.2787,
      "train_tokens_per_second": 62716.85
    },
    {
      "epoch": 0.12146098697018624,
      "grad_norm": 21.25,
      "learning_rate": 1.9564654712317602e-05,
      "loss": 0.7695,
      "num_input_tokens_seen": 679535104,
      "step": 9140,
      "train_runtime": 10834.7722,
      "train_tokens_per_second": 62717.987
    },
    {
      "epoch": 0.12159387645264816,
      "grad_norm": 42.0,
      "learning_rate": 1.9563397733815215e-05,
      "loss": 0.6132,
      "num_input_tokens_seen": 679652096,
      "step": 9150,
      "train_runtime": 10837.521,
      "train_tokens_per_second": 62712.875
    },
    {
      "epoch": 0.12172676593511006,
      "grad_norm": 13.5625,
      "learning_rate": 1.956213898377867e-05,
      "loss": 0.8546,
      "num_input_tokens_seen": 681811456,
      "step": 9160,
      "train_runtime": 10872.6815,
      "train_tokens_per_second": 62708.675
    },
    {
      "epoch": 0.12185965541757197,
      "grad_norm": 37.0,
      "learning_rate": 1.9560878462441142e-05,
      "loss": 0.9137,
      "num_input_tokens_seen": 682659072,
      "step": 9170,
      "train_runtime": 10885.6768,
      "train_tokens_per_second": 62711.679
    },
    {
      "epoch": 0.12199254490003389,
      "grad_norm": 13.8125,
      "learning_rate": 1.9559616170036137e-05,
      "loss": 0.7957,
      "num_input_tokens_seen": 683153664,
      "step": 9180,
      "train_runtime": 10892.6942,
      "train_tokens_per_second": 62716.684
    },
    {
      "epoch": 0.1221254343824958,
      "grad_norm": 22.75,
      "learning_rate": 1.9558352106797473e-05,
      "loss": 0.6352,
      "num_input_tokens_seen": 683393280,
      "step": 9190,
      "train_runtime": 10896.2393,
      "train_tokens_per_second": 62718.27
    },
    {
      "epoch": 0.1222583238649577,
      "grad_norm": 7.53125,
      "learning_rate": 1.955708627295931e-05,
      "loss": 0.5296,
      "num_input_tokens_seen": 683508736,
      "step": 9200,
      "train_runtime": 10898.637,
      "train_tokens_per_second": 62715.066
    },
    {
      "epoch": 0.12239121334741962,
      "grad_norm": 16.875,
      "learning_rate": 1.9555818668756137e-05,
      "loss": 0.9008,
      "num_input_tokens_seen": 685558016,
      "step": 9210,
      "train_runtime": 10931.4898,
      "train_tokens_per_second": 62714.052
    },
    {
      "epoch": 0.12252410282988153,
      "grad_norm": 28.375,
      "learning_rate": 1.9554549294422762e-05,
      "loss": 0.9381,
      "num_input_tokens_seen": 686409728,
      "step": 9220,
      "train_runtime": 10944.5772,
      "train_tokens_per_second": 62716.879
    },
    {
      "epoch": 0.12265699231234344,
      "grad_norm": 19.875,
      "learning_rate": 1.9553278150194325e-05,
      "loss": 0.7776,
      "num_input_tokens_seen": 686903040,
      "step": 9230,
      "train_runtime": 10951.5133,
      "train_tokens_per_second": 62722.203
    },
    {
      "epoch": 0.12278988179480535,
      "grad_norm": 45.5,
      "learning_rate": 1.9552005236306297e-05,
      "loss": 0.7175,
      "num_input_tokens_seen": 687143168,
      "step": 9240,
      "train_runtime": 10955.1086,
      "train_tokens_per_second": 62723.538
    },
    {
      "epoch": 0.12292277127726726,
      "grad_norm": 9.8125,
      "learning_rate": 1.9550730552994473e-05,
      "loss": 0.5202,
      "num_input_tokens_seen": 687257856,
      "step": 9250,
      "train_runtime": 10957.5125,
      "train_tokens_per_second": 62720.244
    },
    {
      "epoch": 0.12305566075972917,
      "grad_norm": 20.0,
      "learning_rate": 1.954945410049498e-05,
      "loss": 0.9312,
      "num_input_tokens_seen": 689345792,
      "step": 9260,
      "train_runtime": 10990.9828,
      "train_tokens_per_second": 62719.213
    },
    {
      "epoch": 0.12318855024219108,
      "grad_norm": 13.375,
      "learning_rate": 1.954817587904426e-05,
      "loss": 1.0518,
      "num_input_tokens_seen": 690198528,
      "step": 9270,
      "train_runtime": 11004.2601,
      "train_tokens_per_second": 62721.03
    },
    {
      "epoch": 0.123321439724653,
      "grad_norm": 11.9375,
      "learning_rate": 1.95468958888791e-05,
      "loss": 0.7447,
      "num_input_tokens_seen": 690690560,
      "step": 9280,
      "train_runtime": 11011.1509,
      "train_tokens_per_second": 62726.464
    },
    {
      "epoch": 0.1234543292071149,
      "grad_norm": 20.25,
      "learning_rate": 1.9545614130236608e-05,
      "loss": 0.6584,
      "num_input_tokens_seen": 690933760,
      "step": 9290,
      "train_runtime": 11014.7428,
      "train_tokens_per_second": 62728.089
    },
    {
      "epoch": 0.12358721868957681,
      "grad_norm": 19.5,
      "learning_rate": 1.954433060335421e-05,
      "loss": 0.6001,
      "num_input_tokens_seen": 691048192,
      "step": 9300,
      "train_runtime": 11017.1077,
      "train_tokens_per_second": 62725.01
    },
    {
      "epoch": 0.12372010817203873,
      "grad_norm": 16.375,
      "learning_rate": 1.9543045308469676e-05,
      "loss": 0.8207,
      "num_input_tokens_seen": 693099776,
      "step": 9310,
      "train_runtime": 11050.6209,
      "train_tokens_per_second": 62720.437
    },
    {
      "epoch": 0.12385299765450064,
      "grad_norm": 30.875,
      "learning_rate": 1.9541758245821092e-05,
      "loss": 0.9585,
      "num_input_tokens_seen": 693966592,
      "step": 9320,
      "train_runtime": 11063.5256,
      "train_tokens_per_second": 62725.628
    },
    {
      "epoch": 0.12398588713696254,
      "grad_norm": 19.0,
      "learning_rate": 1.9540469415646872e-05,
      "loss": 0.804,
      "num_input_tokens_seen": 694454784,
      "step": 9330,
      "train_runtime": 11070.647,
      "train_tokens_per_second": 62729.376
    },
    {
      "epoch": 0.12411877661942446,
      "grad_norm": 12.4375,
      "learning_rate": 1.9539178818185767e-05,
      "loss": 0.62,
      "num_input_tokens_seen": 694691328,
      "step": 9340,
      "train_runtime": 11074.1191,
      "train_tokens_per_second": 62731.069
    },
    {
      "epoch": 0.12425166610188637,
      "grad_norm": 14.3125,
      "learning_rate": 1.9537886453676847e-05,
      "loss": 0.5342,
      "num_input_tokens_seen": 694808320,
      "step": 9350,
      "train_runtime": 11076.4771,
      "train_tokens_per_second": 62728.277
    },
    {
      "epoch": 0.12438455558434827,
      "grad_norm": 25.5,
      "learning_rate": 1.9536592322359503e-05,
      "loss": 0.8555,
      "num_input_tokens_seen": 696814336,
      "step": 9360,
      "train_runtime": 11108.7272,
      "train_tokens_per_second": 62726.748
    },
    {
      "epoch": 0.12451744506681019,
      "grad_norm": 11.875,
      "learning_rate": 1.953529642447347e-05,
      "loss": 0.9794,
      "num_input_tokens_seen": 697649664,
      "step": 9370,
      "train_runtime": 11121.1759,
      "train_tokens_per_second": 62731.646
    },
    {
      "epoch": 0.1246503345492721,
      "grad_norm": 12.625,
      "learning_rate": 1.95339987602588e-05,
      "loss": 0.8081,
      "num_input_tokens_seen": 698095872,
      "step": 9380,
      "train_runtime": 11127.571,
      "train_tokens_per_second": 62735.693
    },
    {
      "epoch": 0.124783224031734,
      "grad_norm": 12.3125,
      "learning_rate": 1.9532699329955875e-05,
      "loss": 0.6882,
      "num_input_tokens_seen": 698329856,
      "step": 9390,
      "train_runtime": 11131.4717,
      "train_tokens_per_second": 62734.729
    },
    {
      "epoch": 0.12491611351419592,
      "grad_norm": 11.3125,
      "learning_rate": 1.95313981338054e-05,
      "loss": 0.5753,
      "num_input_tokens_seen": 698446592,
      "step": 9400,
      "train_runtime": 11133.9002,
      "train_tokens_per_second": 62731.53
    },
    {
      "epoch": 0.12504900299665783,
      "grad_norm": 10.125,
      "learning_rate": 1.9530095172048408e-05,
      "loss": 0.9417,
      "num_input_tokens_seen": 700490240,
      "step": 9410,
      "train_runtime": 11165.605,
      "train_tokens_per_second": 62736.434
    },
    {
      "epoch": 0.12518189247911973,
      "grad_norm": 11.125,
      "learning_rate": 1.9528790444926266e-05,
      "loss": 0.9126,
      "num_input_tokens_seen": 701321728,
      "step": 9420,
      "train_runtime": 11178.3303,
      "train_tokens_per_second": 62739.399
    },
    {
      "epoch": 0.12531478196158166,
      "grad_norm": 35.25,
      "learning_rate": 1.952748395268066e-05,
      "loss": 0.8143,
      "num_input_tokens_seen": 701793280,
      "step": 9430,
      "train_runtime": 11185.236,
      "train_tokens_per_second": 62742.823
    },
    {
      "epoch": 0.12544767144404356,
      "grad_norm": 13.875,
      "learning_rate": 1.9526175695553605e-05,
      "loss": 0.7119,
      "num_input_tokens_seen": 702030336,
      "step": 9440,
      "train_runtime": 11189.1804,
      "train_tokens_per_second": 62741.891
    },
    {
      "epoch": 0.12558056092650546,
      "grad_norm": 17.375,
      "learning_rate": 1.9524865673787446e-05,
      "loss": 0.6341,
      "num_input_tokens_seen": 702144768,
      "step": 9450,
      "train_runtime": 11191.6219,
      "train_tokens_per_second": 62738.428
    },
    {
      "epoch": 0.1257134504089674,
      "grad_norm": 18.0,
      "learning_rate": 1.9523553887624852e-05,
      "loss": 0.9143,
      "num_input_tokens_seen": 704129792,
      "step": 9460,
      "train_runtime": 11223.8069,
      "train_tokens_per_second": 62735.38
    },
    {
      "epoch": 0.1258463398914293,
      "grad_norm": 45.0,
      "learning_rate": 1.9522240337308825e-05,
      "loss": 0.9719,
      "num_input_tokens_seen": 704967680,
      "step": 9470,
      "train_runtime": 11236.5835,
      "train_tokens_per_second": 62738.614
    },
    {
      "epoch": 0.1259792293738912,
      "grad_norm": 22.875,
      "learning_rate": 1.952092502308268e-05,
      "loss": 0.8838,
      "num_input_tokens_seen": 705469184,
      "step": 9480,
      "train_runtime": 11243.5259,
      "train_tokens_per_second": 62744.48
    },
    {
      "epoch": 0.12611211885635312,
      "grad_norm": 19.5,
      "learning_rate": 1.9519607945190074e-05,
      "loss": 0.7802,
      "num_input_tokens_seen": 705709312,
      "step": 9490,
      "train_runtime": 11247.1553,
      "train_tokens_per_second": 62745.582
    },
    {
      "epoch": 0.12624500833881502,
      "grad_norm": 12.625,
      "learning_rate": 1.9518289103874976e-05,
      "loss": 0.6065,
      "num_input_tokens_seen": 705828352,
      "step": 9500,
      "train_runtime": 11249.6194,
      "train_tokens_per_second": 62742.421
    },
    {
      "epoch": 0.12637789782127692,
      "grad_norm": 19.375,
      "learning_rate": 1.95169684993817e-05,
      "loss": 0.861,
      "num_input_tokens_seen": 707869184,
      "step": 9510,
      "train_runtime": 11282.4519,
      "train_tokens_per_second": 62740.723
    },
    {
      "epoch": 0.12651078730373885,
      "grad_norm": 26.5,
      "learning_rate": 1.9515646131954868e-05,
      "loss": 0.9794,
      "num_input_tokens_seen": 708725504,
      "step": 9520,
      "train_runtime": 11295.3091,
      "train_tokens_per_second": 62745.118
    },
    {
      "epoch": 0.12664367678620075,
      "grad_norm": 24.5,
      "learning_rate": 1.9514322001839438e-05,
      "loss": 0.7254,
      "num_input_tokens_seen": 709181184,
      "step": 9530,
      "train_runtime": 11302.0619,
      "train_tokens_per_second": 62747.947
    },
    {
      "epoch": 0.12677656626866265,
      "grad_norm": 21.0,
      "learning_rate": 1.9512996109280694e-05,
      "loss": 0.6762,
      "num_input_tokens_seen": 709415168,
      "step": 9540,
      "train_runtime": 11305.5667,
      "train_tokens_per_second": 62749.191
    },
    {
      "epoch": 0.12690945575112458,
      "grad_norm": 12.0625,
      "learning_rate": 1.951166845452425e-05,
      "loss": 0.5797,
      "num_input_tokens_seen": 709531136,
      "step": 9550,
      "train_runtime": 11308.0063,
      "train_tokens_per_second": 62745.909
    },
    {
      "epoch": 0.12704234523358648,
      "grad_norm": 18.5,
      "learning_rate": 1.951033903781604e-05,
      "loss": 0.8872,
      "num_input_tokens_seen": 711647232,
      "step": 9560,
      "train_runtime": 11340.5688,
      "train_tokens_per_second": 62752.34
    },
    {
      "epoch": 0.12717523471604839,
      "grad_norm": 17.375,
      "learning_rate": 1.950900785940232e-05,
      "loss": 0.8853,
      "num_input_tokens_seen": 712504064,
      "step": 9570,
      "train_runtime": 11353.2152,
      "train_tokens_per_second": 62757.91
    },
    {
      "epoch": 0.12730812419851031,
      "grad_norm": 12.625,
      "learning_rate": 1.9507674919529686e-05,
      "loss": 0.7205,
      "num_input_tokens_seen": 712970496,
      "step": 9580,
      "train_runtime": 11359.861,
      "train_tokens_per_second": 62762.255
    },
    {
      "epoch": 0.12744101368097221,
      "grad_norm": 14.3125,
      "learning_rate": 1.9506340218445053e-05,
      "loss": 0.7109,
      "num_input_tokens_seen": 713203200,
      "step": 9590,
      "train_runtime": 11363.3514,
      "train_tokens_per_second": 62763.456
    },
    {
      "epoch": 0.12757390316343414,
      "grad_norm": 9.3125,
      "learning_rate": 1.950500375639566e-05,
      "loss": 0.5399,
      "num_input_tokens_seen": 713323008,
      "step": 9600,
      "train_runtime": 11365.7985,
      "train_tokens_per_second": 62760.483
    },
    {
      "epoch": 0.12770679264589604,
      "grad_norm": 19.375,
      "learning_rate": 1.9503665533629078e-05,
      "loss": 0.9031,
      "num_input_tokens_seen": 715407360,
      "step": 9610,
      "train_runtime": 11399.0826,
      "train_tokens_per_second": 62760.082
    },
    {
      "epoch": 0.12783968212835795,
      "grad_norm": 24.75,
      "learning_rate": 1.9502325550393196e-05,
      "loss": 0.9802,
      "num_input_tokens_seen": 716231936,
      "step": 9620,
      "train_runtime": 11412.067,
      "train_tokens_per_second": 62760.93
    },
    {
      "epoch": 0.12797257161081987,
      "grad_norm": 30.5,
      "learning_rate": 1.950098380693624e-05,
      "loss": 0.7687,
      "num_input_tokens_seen": 716713216,
      "step": 9630,
      "train_runtime": 11418.7913,
      "train_tokens_per_second": 62766.119
    },
    {
      "epoch": 0.12810546109328177,
      "grad_norm": 11.125,
      "learning_rate": 1.949964030350675e-05,
      "loss": 0.6073,
      "num_input_tokens_seen": 716942592,
      "step": 9640,
      "train_runtime": 11422.269,
      "train_tokens_per_second": 62767.091
    },
    {
      "epoch": 0.12823835057574368,
      "grad_norm": 31.75,
      "learning_rate": 1.94982950403536e-05,
      "loss": 0.5238,
      "num_input_tokens_seen": 717055744,
      "step": 9650,
      "train_runtime": 11424.6175,
      "train_tokens_per_second": 62764.092
    },
    {
      "epoch": 0.1283712400582056,
      "grad_norm": 18.25,
      "learning_rate": 1.9496948017725988e-05,
      "loss": 0.843,
      "num_input_tokens_seen": 719076352,
      "step": 9660,
      "train_runtime": 11457.4368,
      "train_tokens_per_second": 62760.665
    },
    {
      "epoch": 0.1285041295406675,
      "grad_norm": 20.25,
      "learning_rate": 1.949559923587344e-05,
      "loss": 0.9096,
      "num_input_tokens_seen": 719905280,
      "step": 9670,
      "train_runtime": 11469.9302,
      "train_tokens_per_second": 62764.574
    },
    {
      "epoch": 0.1286370190231294,
      "grad_norm": 21.875,
      "learning_rate": 1.9494248695045806e-05,
      "loss": 0.7922,
      "num_input_tokens_seen": 720369408,
      "step": 9680,
      "train_runtime": 11476.8529,
      "train_tokens_per_second": 62767.155
    },
    {
      "epoch": 0.12876990850559134,
      "grad_norm": 107.0,
      "learning_rate": 1.9492896395493256e-05,
      "loss": 0.7067,
      "num_input_tokens_seen": 720593920,
      "step": 9690,
      "train_runtime": 11480.2558,
      "train_tokens_per_second": 62768.107
    },
    {
      "epoch": 0.12890279798805324,
      "grad_norm": 17.25,
      "learning_rate": 1.9491542337466294e-05,
      "loss": 0.5386,
      "num_input_tokens_seen": 720708096,
      "step": 9700,
      "train_runtime": 11482.629,
      "train_tokens_per_second": 62765.077
    },
    {
      "epoch": 0.12903568747051514,
      "grad_norm": 23.0,
      "learning_rate": 1.949018652121575e-05,
      "loss": 0.8934,
      "num_input_tokens_seen": 722748672,
      "step": 9710,
      "train_runtime": 11515.4596,
      "train_tokens_per_second": 62763.337
    },
    {
      "epoch": 0.12916857695297707,
      "grad_norm": 19.75,
      "learning_rate": 1.948882894699278e-05,
      "loss": 0.9583,
      "num_input_tokens_seen": 723599360,
      "step": 9720,
      "train_runtime": 11528.2253,
      "train_tokens_per_second": 62767.628
    },
    {
      "epoch": 0.12930146643543897,
      "grad_norm": 23.125,
      "learning_rate": 1.9487469615048856e-05,
      "loss": 0.7361,
      "num_input_tokens_seen": 724048640,
      "step": 9730,
      "train_runtime": 11535.2112,
      "train_tokens_per_second": 62768.564
    },
    {
      "epoch": 0.12943435591790087,
      "grad_norm": 17.75,
      "learning_rate": 1.948610852563578e-05,
      "loss": 0.6935,
      "num_input_tokens_seen": 724280320,
      "step": 9740,
      "train_runtime": 11538.6461,
      "train_tokens_per_second": 62769.957
    },
    {
      "epoch": 0.1295672454003628,
      "grad_norm": 16.375,
      "learning_rate": 1.948474567900569e-05,
      "loss": 0.5303,
      "num_input_tokens_seen": 724396544,
      "step": 9750,
      "train_runtime": 11541.0222,
      "train_tokens_per_second": 62767.104
    },
    {
      "epoch": 0.1297001348828247,
      "grad_norm": 20.125,
      "learning_rate": 1.9483381075411033e-05,
      "loss": 0.9079,
      "num_input_tokens_seen": 726440704,
      "step": 9760,
      "train_runtime": 11573.0045,
      "train_tokens_per_second": 62770.278
    },
    {
      "epoch": 0.1298330243652866,
      "grad_norm": 29.875,
      "learning_rate": 1.9482014715104597e-05,
      "loss": 0.927,
      "num_input_tokens_seen": 727253760,
      "step": 9770,
      "train_runtime": 11585.5773,
      "train_tokens_per_second": 62772.337
    },
    {
      "epoch": 0.12996591384774853,
      "grad_norm": 14.4375,
      "learning_rate": 1.9480646598339487e-05,
      "loss": 0.7856,
      "num_input_tokens_seen": 727698688,
      "step": 9780,
      "train_runtime": 11591.9751,
      "train_tokens_per_second": 62776.074
    },
    {
      "epoch": 0.13009880333021043,
      "grad_norm": 16.125,
      "learning_rate": 1.9479276725369125e-05,
      "loss": 0.7155,
      "num_input_tokens_seen": 727921664,
      "step": 9790,
      "train_runtime": 11595.3393,
      "train_tokens_per_second": 62777.091
    },
    {
      "epoch": 0.13023169281267233,
      "grad_norm": 38.5,
      "learning_rate": 1.9477905096447277e-05,
      "loss": 0.5025,
      "num_input_tokens_seen": 728035840,
      "step": 9800,
      "train_runtime": 11597.7185,
      "train_tokens_per_second": 62774.057
    },
    {
      "epoch": 0.13036458229513426,
      "grad_norm": 22.5,
      "learning_rate": 1.9476531711828027e-05,
      "loss": 0.8346,
      "num_input_tokens_seen": 730029312,
      "step": 9810,
      "train_runtime": 11631.0301,
      "train_tokens_per_second": 62765.663
    },
    {
      "epoch": 0.13049747177759616,
      "grad_norm": 63.5,
      "learning_rate": 1.9475156571765775e-05,
      "loss": 1.0274,
      "num_input_tokens_seen": 730862848,
      "step": 9820,
      "train_runtime": 11643.5555,
      "train_tokens_per_second": 62769.731
    },
    {
      "epoch": 0.1306303612600581,
      "grad_norm": 19.375,
      "learning_rate": 1.9473779676515256e-05,
      "loss": 0.7766,
      "num_input_tokens_seen": 731333120,
      "step": 9830,
      "train_runtime": 11650.1903,
      "train_tokens_per_second": 62774.35
    },
    {
      "epoch": 0.13076325074252,
      "grad_norm": 12.125,
      "learning_rate": 1.9472401026331526e-05,
      "loss": 0.6069,
      "num_input_tokens_seen": 731553280,
      "step": 9840,
      "train_runtime": 11653.5578,
      "train_tokens_per_second": 62775.102
    },
    {
      "epoch": 0.1308961402249819,
      "grad_norm": 9.5625,
      "learning_rate": 1.9471020621469976e-05,
      "loss": 0.5247,
      "num_input_tokens_seen": 731668992,
      "step": 9850,
      "train_runtime": 11656.4015,
      "train_tokens_per_second": 62769.714
    },
    {
      "epoch": 0.13102902970744382,
      "grad_norm": 23.125,
      "learning_rate": 1.94696384621863e-05,
      "loss": 0.8182,
      "num_input_tokens_seen": 733771520,
      "step": 9860,
      "train_runtime": 11689.5464,
      "train_tokens_per_second": 62771.599
    },
    {
      "epoch": 0.13116191918990572,
      "grad_norm": 17.75,
      "learning_rate": 1.9468254548736545e-05,
      "loss": 0.9192,
      "num_input_tokens_seen": 734619392,
      "step": 9870,
      "train_runtime": 11702.0965,
      "train_tokens_per_second": 62776.733
    },
    {
      "epoch": 0.13129480867236762,
      "grad_norm": 19.75,
      "learning_rate": 1.9466868881377058e-05,
      "loss": 0.8046,
      "num_input_tokens_seen": 735065856,
      "step": 9880,
      "train_runtime": 11708.4744,
      "train_tokens_per_second": 62780.669
    },
    {
      "epoch": 0.13142769815482955,
      "grad_norm": 9.625,
      "learning_rate": 1.9465481460364524e-05,
      "loss": 0.7156,
      "num_input_tokens_seen": 735290112,
      "step": 9890,
      "train_runtime": 11711.8978,
      "train_tokens_per_second": 62781.466
    },
    {
      "epoch": 0.13156058763729145,
      "grad_norm": 13.375,
      "learning_rate": 1.9464092285955955e-05,
      "loss": 0.6007,
      "num_input_tokens_seen": 735402752,
      "step": 9900,
      "train_runtime": 11714.2992,
      "train_tokens_per_second": 62778.211
    },
    {
      "epoch": 0.13169347711975335,
      "grad_norm": 11.9375,
      "learning_rate": 1.9462701358408678e-05,
      "loss": 0.898,
      "num_input_tokens_seen": 737432064,
      "step": 9910,
      "train_runtime": 11747.095,
      "train_tokens_per_second": 62775.696
    },
    {
      "epoch": 0.13182636660221528,
      "grad_norm": 33.75,
      "learning_rate": 1.946130867798035e-05,
      "loss": 0.8978,
      "num_input_tokens_seen": 738268928,
      "step": 9920,
      "train_runtime": 11759.111,
      "train_tokens_per_second": 62782.716
    },
    {
      "epoch": 0.13195925608467718,
      "grad_norm": 25.0,
      "learning_rate": 1.945991424492896e-05,
      "loss": 0.7341,
      "num_input_tokens_seen": 738699520,
      "step": 9930,
      "train_runtime": 11765.3418,
      "train_tokens_per_second": 62786.065
    },
    {
      "epoch": 0.13209214556713908,
      "grad_norm": 65.5,
      "learning_rate": 1.9458518059512802e-05,
      "loss": 0.6678,
      "num_input_tokens_seen": 738920448,
      "step": 9940,
      "train_runtime": 11768.6728,
      "train_tokens_per_second": 62787.067
    },
    {
      "epoch": 0.132225035049601,
      "grad_norm": 17.25,
      "learning_rate": 1.9457120121990517e-05,
      "loss": 0.4801,
      "num_input_tokens_seen": 739035136,
      "step": 9950,
      "train_runtime": 11771.4281,
      "train_tokens_per_second": 62782.113
    },
    {
      "epoch": 0.1323579245320629,
      "grad_norm": 14.25,
      "learning_rate": 1.9455720432621053e-05,
      "loss": 0.8531,
      "num_input_tokens_seen": 741247744,
      "step": 9960,
      "train_runtime": 11805.1956,
      "train_tokens_per_second": 62789.959
    },
    {
      "epoch": 0.1324908140145248,
      "grad_norm": 25.125,
      "learning_rate": 1.94543189916637e-05,
      "loss": 0.8738,
      "num_input_tokens_seen": 742113280,
      "step": 9970,
      "train_runtime": 11818.2192,
      "train_tokens_per_second": 62794.002
    },
    {
      "epoch": 0.13262370349698674,
      "grad_norm": 45.25,
      "learning_rate": 1.9452915799378053e-05,
      "loss": 0.7807,
      "num_input_tokens_seen": 742587648,
      "step": 9980,
      "train_runtime": 11825.0575,
      "train_tokens_per_second": 62797.804
    },
    {
      "epoch": 0.13275659297944864,
      "grad_norm": 16.125,
      "learning_rate": 1.9451510856024045e-05,
      "loss": 0.7007,
      "num_input_tokens_seen": 742820352,
      "step": 9990,
      "train_runtime": 11828.5953,
      "train_tokens_per_second": 62798.695
    },
    {
      "epoch": 0.13288948246191054,
      "grad_norm": 24.625,
      "learning_rate": 1.9450104161861928e-05,
      "loss": 0.5709,
      "num_input_tokens_seen": 742934528,
      "step": 10000,
      "train_runtime": 11830.9699,
      "train_tokens_per_second": 62795.742
    },
    {
      "epoch": 0.13288948246191054,
      "eval_accuracy": 0.8089,
      "eval_balanced_accuracy": 0.8057947080254995,
      "eval_invalid_f1": 0.8292377803592171,
      "eval_invalid_precision": 0.8254758939690446,
      "eval_invalid_recall": 0.8330341113105925,
      "eval_loss": 0.38208621740341187,
      "eval_macro_f1": 0.8061502785324296,
      "eval_mcc": 0.61234356250324,
      "eval_runtime": 78.1249,
      "eval_samples_per_second": 128.0,
      "eval_steps_per_second": 32.0,
      "eval_valid_f1": 0.7830627767056421,
      "eval_valid_precision": 0.7876227449189313,
      "eval_valid_recall": 0.7785553047404064,
      "num_input_tokens_seen": 742934528,
      "step": 10000
    },
    {
      "epoch": 0.13302237194437247,
      "grad_norm": 21.75,
      "learning_rate": 1.944869571715228e-05,
      "loss": 0.8852,
      "num_input_tokens_seen": 744894208,
      "step": 10010,
      "train_runtime": 11944.8059,
      "train_tokens_per_second": 62361.349
    },
    {
      "epoch": 0.13315526142683437,
      "grad_norm": 20.375,
      "learning_rate": 1.9447285522156006e-05,
      "loss": 0.9781,
      "num_input_tokens_seen": 745745408,
      "step": 10020,
      "train_runtime": 11957.7803,
      "train_tokens_per_second": 62364.87
    },
    {
      "epoch": 0.13328815090929627,
      "grad_norm": 20.0,
      "learning_rate": 1.9445873577134327e-05,
      "loss": 0.8226,
      "num_input_tokens_seen": 746194176,
      "step": 10030,
      "train_runtime": 11964.2132,
      "train_tokens_per_second": 62368.846
    },
    {
      "epoch": 0.1334210403917582,
      "grad_norm": 35.25,
      "learning_rate": 1.9444459882348794e-05,
      "loss": 0.7063,
      "num_input_tokens_seen": 746420736,
      "step": 10040,
      "train_runtime": 11967.668,
      "train_tokens_per_second": 62369.773
    },
    {
      "epoch": 0.1335539298742201,
      "grad_norm": 14.375,
      "learning_rate": 1.9443044438061283e-05,
      "loss": 0.5708,
      "num_input_tokens_seen": 746535424,
      "step": 10050,
      "train_runtime": 11970.092,
      "train_tokens_per_second": 62366.724
    },
    {
      "epoch": 0.133686819356682,
      "grad_norm": 23.5,
      "learning_rate": 1.944162724453399e-05,
      "loss": 0.8429,
      "num_input_tokens_seen": 748652032,
      "step": 10060,
      "train_runtime": 12003.7589,
      "train_tokens_per_second": 62368.133
    },
    {
      "epoch": 0.13381970883914393,
      "grad_norm": 16.5,
      "learning_rate": 1.944020830202944e-05,
      "loss": 0.9416,
      "num_input_tokens_seen": 749489664,
      "step": 10070,
      "train_runtime": 12016.5233,
      "train_tokens_per_second": 62371.59
    },
    {
      "epoch": 0.13395259832160583,
      "grad_norm": 15.0625,
      "learning_rate": 1.943878761081048e-05,
      "loss": 0.7071,
      "num_input_tokens_seen": 749955584,
      "step": 10080,
      "train_runtime": 12023.1523,
      "train_tokens_per_second": 62375.953
    },
    {
      "epoch": 0.13408548780406776,
      "grad_norm": 11.625,
      "learning_rate": 1.9437365171140282e-05,
      "loss": 0.6696,
      "num_input_tokens_seen": 750180096,
      "step": 10090,
      "train_runtime": 12026.5379,
      "train_tokens_per_second": 62377.062
    },
    {
      "epoch": 0.13421837728652966,
      "grad_norm": 18.625,
      "learning_rate": 1.9435940983282335e-05,
      "loss": 0.5382,
      "num_input_tokens_seen": 750294528,
      "step": 10100,
      "train_runtime": 12028.8737,
      "train_tokens_per_second": 62374.462
    },
    {
      "epoch": 0.13435126676899156,
      "grad_norm": 14.1875,
      "learning_rate": 1.943451504750046e-05,
      "loss": 0.8654,
      "num_input_tokens_seen": 752395264,
      "step": 10110,
      "train_runtime": 12061.988,
      "train_tokens_per_second": 62377.384
    },
    {
      "epoch": 0.1344841562514535,
      "grad_norm": 47.5,
      "learning_rate": 1.9433087364058796e-05,
      "loss": 0.9679,
      "num_input_tokens_seen": 753265920,
      "step": 10120,
      "train_runtime": 12074.7232,
      "train_tokens_per_second": 62383.701
    },
    {
      "epoch": 0.1346170457339154,
      "grad_norm": 16.5,
      "learning_rate": 1.9431657933221814e-05,
      "loss": 0.7649,
      "num_input_tokens_seen": 753725696,
      "step": 10130,
      "train_runtime": 12081.2934,
      "train_tokens_per_second": 62387.831
    },
    {
      "epoch": 0.1347499352163773,
      "grad_norm": 18.125,
      "learning_rate": 1.94302267552543e-05,
      "loss": 0.7051,
      "num_input_tokens_seen": 753956096,
      "step": 10140,
      "train_runtime": 12085.1128,
      "train_tokens_per_second": 62387.179
    },
    {
      "epoch": 0.13488282469883922,
      "grad_norm": 11.0625,
      "learning_rate": 1.9428793830421364e-05,
      "loss": 0.5212,
      "num_input_tokens_seen": 754069760,
      "step": 10150,
      "train_runtime": 12087.5033,
      "train_tokens_per_second": 62384.245
    },
    {
      "epoch": 0.13501571418130112,
      "grad_norm": 28.125,
      "learning_rate": 1.9427359158988448e-05,
      "loss": 0.8147,
      "num_input_tokens_seen": 756066816,
      "step": 10160,
      "train_runtime": 12118.8717,
      "train_tokens_per_second": 62387.558
    },
    {
      "epoch": 0.13514860366376302,
      "grad_norm": 27.375,
      "learning_rate": 1.9425922741221314e-05,
      "loss": 0.9286,
      "num_input_tokens_seen": 756934400,
      "step": 10170,
      "train_runtime": 12131.7644,
      "train_tokens_per_second": 62392.771
    },
    {
      "epoch": 0.13528149314622495,
      "grad_norm": 19.75,
      "learning_rate": 1.9424484577386036e-05,
      "loss": 0.7491,
      "num_input_tokens_seen": 757434880,
      "step": 10180,
      "train_runtime": 12138.6926,
      "train_tokens_per_second": 62398.39
    },
    {
      "epoch": 0.13541438262868685,
      "grad_norm": 28.125,
      "learning_rate": 1.942304466774903e-05,
      "loss": 0.6189,
      "num_input_tokens_seen": 757666816,
      "step": 10190,
      "train_runtime": 12142.2394,
      "train_tokens_per_second": 62399.265
    },
    {
      "epoch": 0.13554727211114875,
      "grad_norm": 20.375,
      "learning_rate": 1.942160301257702e-05,
      "loss": 0.5828,
      "num_input_tokens_seen": 757783552,
      "step": 10200,
      "train_runtime": 12145.0545,
      "train_tokens_per_second": 62394.413
    },
    {
      "epoch": 0.13568016159361068,
      "grad_norm": 27.125,
      "learning_rate": 1.9420159612137062e-05,
      "loss": 0.933,
      "num_input_tokens_seen": 759945984,
      "step": 10210,
      "train_runtime": 12179.0256,
      "train_tokens_per_second": 62397.93
    },
    {
      "epoch": 0.13581305107607258,
      "grad_norm": 91.0,
      "learning_rate": 1.941871446669654e-05,
      "loss": 0.9402,
      "num_input_tokens_seen": 760794880,
      "step": 10220,
      "train_runtime": 12191.5453,
      "train_tokens_per_second": 62403.482
    },
    {
      "epoch": 0.13594594055853448,
      "grad_norm": 26.375,
      "learning_rate": 1.9417267576523144e-05,
      "loss": 0.7804,
      "num_input_tokens_seen": 761251328,
      "step": 10230,
      "train_runtime": 12198.161,
      "train_tokens_per_second": 62407.057
    },
    {
      "epoch": 0.1360788300409964,
      "grad_norm": 22.25,
      "learning_rate": 1.9415818941884903e-05,
      "loss": 0.5941,
      "num_input_tokens_seen": 761476096,
      "step": 10240,
      "train_runtime": 12201.5958,
      "train_tokens_per_second": 62407.91
    },
    {
      "epoch": 0.1362117195234583,
      "grad_norm": 44.25,
      "learning_rate": 1.941436856305016e-05,
      "loss": 0.6032,
      "num_input_tokens_seen": 761593088,
      "step": 10250,
      "train_runtime": 12204.3993,
      "train_tokens_per_second": 62403.16
    },
    {
      "epoch": 0.13634460900592021,
      "grad_norm": 17.75,
      "learning_rate": 1.941291644028759e-05,
      "loss": 0.8441,
      "num_input_tokens_seen": 763606528,
      "step": 10260,
      "train_runtime": 12236.9138,
      "train_tokens_per_second": 62401.888
    },
    {
      "epoch": 0.13647749848838214,
      "grad_norm": 11.3125,
      "learning_rate": 1.9411462573866182e-05,
      "loss": 0.8732,
      "num_input_tokens_seen": 764432384,
      "step": 10270,
      "train_runtime": 12249.2693,
      "train_tokens_per_second": 62406.366
    },
    {
      "epoch": 0.13661038797084404,
      "grad_norm": 15.1875,
      "learning_rate": 1.9410006964055254e-05,
      "loss": 0.6385,
      "num_input_tokens_seen": 764869888,
      "step": 10280,
      "train_runtime": 12255.6425,
      "train_tokens_per_second": 62409.611
    },
    {
      "epoch": 0.13674327745330594,
      "grad_norm": 13.125,
      "learning_rate": 1.940854961112444e-05,
      "loss": 0.6577,
      "num_input_tokens_seen": 765087488,
      "step": 10290,
      "train_runtime": 12258.9949,
      "train_tokens_per_second": 62410.295
    },
    {
      "epoch": 0.13687616693576787,
      "grad_norm": 27.25,
      "learning_rate": 1.9407090515343704e-05,
      "loss": 0.5017,
      "num_input_tokens_seen": 765200384,
      "step": 10300,
      "train_runtime": 12261.776,
      "train_tokens_per_second": 62405.347
    },
    {
      "epoch": 0.13700905641822977,
      "grad_norm": 27.75,
      "learning_rate": 1.9405629676983336e-05,
      "loss": 0.8677,
      "num_input_tokens_seen": 767270656,
      "step": 10310,
      "train_runtime": 12296.6173,
      "train_tokens_per_second": 62396.888
    },
    {
      "epoch": 0.1371419459006917,
      "grad_norm": 16.0,
      "learning_rate": 1.9404167096313937e-05,
      "loss": 0.9487,
      "num_input_tokens_seen": 768144384,
      "step": 10320,
      "train_runtime": 12309.7581,
      "train_tokens_per_second": 62401.257
    },
    {
      "epoch": 0.1372748353831536,
      "grad_norm": 15.375,
      "learning_rate": 1.9402702773606438e-05,
      "loss": 0.8131,
      "num_input_tokens_seen": 768625408,
      "step": 10330,
      "train_runtime": 12316.603,
      "train_tokens_per_second": 62405.633
    },
    {
      "epoch": 0.1374077248656155,
      "grad_norm": 12.875,
      "learning_rate": 1.9401236709132096e-05,
      "loss": 0.6817,
      "num_input_tokens_seen": 768855808,
      "step": 10340,
      "train_runtime": 12320.0652,
      "train_tokens_per_second": 62406.797
    },
    {
      "epoch": 0.13754061434807743,
      "grad_norm": 23.0,
      "learning_rate": 1.9399768903162476e-05,
      "loss": 0.5304,
      "num_input_tokens_seen": 768972032,
      "step": 10350,
      "train_runtime": 12322.5022,
      "train_tokens_per_second": 62403.887
    },
    {
      "epoch": 0.13767350383053933,
      "grad_norm": 22.0,
      "learning_rate": 1.939829935596949e-05,
      "loss": 0.9633,
      "num_input_tokens_seen": 771006464,
      "step": 10360,
      "train_runtime": 12354.6708,
      "train_tokens_per_second": 62406.071
    },
    {
      "epoch": 0.13780639331300124,
      "grad_norm": 16.375,
      "learning_rate": 1.9396828067825347e-05,
      "loss": 0.9297,
      "num_input_tokens_seen": 771863296,
      "step": 10370,
      "train_runtime": 12367.0568,
      "train_tokens_per_second": 62412.853
    },
    {
      "epoch": 0.13793928279546316,
      "grad_norm": 13.4375,
      "learning_rate": 1.9395355039002595e-05,
      "loss": 0.7157,
      "num_input_tokens_seen": 772335360,
      "step": 10380,
      "train_runtime": 12373.7006,
      "train_tokens_per_second": 62417.492
    },
    {
      "epoch": 0.13807217227792506,
      "grad_norm": 16.625,
      "learning_rate": 1.93938802697741e-05,
      "loss": 0.7255,
      "num_input_tokens_seen": 772568832,
      "step": 10390,
      "train_runtime": 12377.1364,
      "train_tokens_per_second": 62419.029
    },
    {
      "epoch": 0.13820506176038697,
      "grad_norm": 9.625,
      "learning_rate": 1.939240376041305e-05,
      "loss": 0.5858,
      "num_input_tokens_seen": 772686080,
      "step": 10400,
      "train_runtime": 12379.5052,
      "train_tokens_per_second": 62416.556
    },
    {
      "epoch": 0.1383379512428489,
      "grad_norm": 12.625,
      "learning_rate": 1.939092551119295e-05,
      "loss": 0.9063,
      "num_input_tokens_seen": 774755840,
      "step": 10410,
      "train_runtime": 12412.8204,
      "train_tokens_per_second": 62415.778
    },
    {
      "epoch": 0.1384708407253108,
      "grad_norm": 15.0,
      "learning_rate": 1.938944552238764e-05,
      "loss": 0.9775,
      "num_input_tokens_seen": 775614976,
      "step": 10420,
      "train_runtime": 12425.4868,
      "train_tokens_per_second": 62421.295
    },
    {
      "epoch": 0.1386037302077727,
      "grad_norm": 16.375,
      "learning_rate": 1.9387963794271272e-05,
      "loss": 0.8151,
      "num_input_tokens_seen": 776104960,
      "step": 10430,
      "train_runtime": 12432.2042,
      "train_tokens_per_second": 62426.98
    },
    {
      "epoch": 0.13873661969023462,
      "grad_norm": 17.125,
      "learning_rate": 1.9386480327118317e-05,
      "loss": 0.6828,
      "num_input_tokens_seen": 776329216,
      "step": 10440,
      "train_runtime": 12435.659,
      "train_tokens_per_second": 62427.67
    },
    {
      "epoch": 0.13886950917269653,
      "grad_norm": 6.4375,
      "learning_rate": 1.9384995121203586e-05,
      "loss": 0.5084,
      "num_input_tokens_seen": 776443904,
      "step": 10450,
      "train_runtime": 12438.0301,
      "train_tokens_per_second": 62424.99
    },
    {
      "epoch": 0.13900239865515843,
      "grad_norm": 19.875,
      "learning_rate": 1.9383508176802187e-05,
      "loss": 0.8796,
      "num_input_tokens_seen": 778575360,
      "step": 10460,
      "train_runtime": 12471.9243,
      "train_tokens_per_second": 62426.242
    },
    {
      "epoch": 0.13913528813762036,
      "grad_norm": 14.0,
      "learning_rate": 1.9382019494189573e-05,
      "loss": 0.9041,
      "num_input_tokens_seen": 779432192,
      "step": 10470,
      "train_runtime": 12484.996,
      "train_tokens_per_second": 62429.511
    },
    {
      "epoch": 0.13926817762008226,
      "grad_norm": 17.5,
      "learning_rate": 1.9380529073641503e-05,
      "loss": 0.7681,
      "num_input_tokens_seen": 779922688,
      "step": 10480,
      "train_runtime": 12492.0081,
      "train_tokens_per_second": 62433.732
    },
    {
      "epoch": 0.13940106710254416,
      "grad_norm": 28.875,
      "learning_rate": 1.9379036915434067e-05,
      "loss": 0.6902,
      "num_input_tokens_seen": 780161792,
      "step": 10490,
      "train_runtime": 12495.506,
      "train_tokens_per_second": 62435.39
    },
    {
      "epoch": 0.13953395658500609,
      "grad_norm": 14.5,
      "learning_rate": 1.9377543019843674e-05,
      "loss": 0.5422,
      "num_input_tokens_seen": 780276224,
      "step": 10500,
      "train_runtime": 12497.8774,
      "train_tokens_per_second": 62432.699
    },
    {
      "epoch": 0.139666846067468,
      "grad_norm": 16.75,
      "learning_rate": 1.9376047387147058e-05,
      "loss": 0.8499,
      "num_input_tokens_seen": 782362368,
      "step": 10510,
      "train_runtime": 12532.0466,
      "train_tokens_per_second": 62428.939
    },
    {
      "epoch": 0.1397997355499299,
      "grad_norm": 25.875,
      "learning_rate": 1.9374550017621266e-05,
      "loss": 0.9269,
      "num_input_tokens_seen": 783193344,
      "step": 10520,
      "train_runtime": 12544.5658,
      "train_tokens_per_second": 62432.878
    },
    {
      "epoch": 0.13993262503239182,
      "grad_norm": 28.0,
      "learning_rate": 1.937305091154367e-05,
      "loss": 0.7386,
      "num_input_tokens_seen": 783667456,
      "step": 10530,
      "train_runtime": 12551.5221,
      "train_tokens_per_second": 62436.05
    },
    {
      "epoch": 0.14006551451485372,
      "grad_norm": 15.875,
      "learning_rate": 1.937155006919198e-05,
      "loss": 0.7223,
      "num_input_tokens_seen": 783901440,
      "step": 10540,
      "train_runtime": 12555.0201,
      "train_tokens_per_second": 62437.291
    },
    {
      "epoch": 0.14019840399731565,
      "grad_norm": 13.9375,
      "learning_rate": 1.9370047490844196e-05,
      "loss": 0.5486,
      "num_input_tokens_seen": 784016896,
      "step": 10550,
      "train_runtime": 12557.3915,
      "train_tokens_per_second": 62434.694
    },
    {
      "epoch": 0.14033129347977755,
      "grad_norm": 23.625,
      "learning_rate": 1.936854317677867e-05,
      "loss": 0.8942,
      "num_input_tokens_seen": 786093056,
      "step": 10560,
      "train_runtime": 12590.7948,
      "train_tokens_per_second": 62433.95
    },
    {
      "epoch": 0.14046418296223945,
      "grad_norm": 35.5,
      "learning_rate": 1.9367037127274057e-05,
      "loss": 0.9459,
      "num_input_tokens_seen": 786915840,
      "step": 10570,
      "train_runtime": 12603.3074,
      "train_tokens_per_second": 62437.249
    },
    {
      "epoch": 0.14059707244470138,
      "grad_norm": 14.25,
      "learning_rate": 1.936552934260934e-05,
      "loss": 0.713,
      "num_input_tokens_seen": 787397120,
      "step": 10580,
      "train_runtime": 12610.3984,
      "train_tokens_per_second": 62440.305
    },
    {
      "epoch": 0.14072996192716328,
      "grad_norm": 33.25,
      "learning_rate": 1.936401982306382e-05,
      "loss": 0.6441,
      "num_input_tokens_seen": 787621376,
      "step": 10590,
      "train_runtime": 12613.8018,
      "train_tokens_per_second": 62441.236
    },
    {
      "epoch": 0.14086285140962518,
      "grad_norm": 10.3125,
      "learning_rate": 1.936250856891713e-05,
      "loss": 0.4885,
      "num_input_tokens_seen": 787731200,
      "step": 10600,
      "train_runtime": 12616.1702,
      "train_tokens_per_second": 62438.219
    },
    {
      "epoch": 0.1409957408920871,
      "grad_norm": 21.625,
      "learning_rate": 1.936099558044921e-05,
      "loss": 0.8629,
      "num_input_tokens_seen": 789730816,
      "step": 10610,
      "train_runtime": 12648.5766,
      "train_tokens_per_second": 62436.339
    },
    {
      "epoch": 0.141128630374549,
      "grad_norm": 34.75,
      "learning_rate": 1.935948085794033e-05,
      "loss": 0.8972,
      "num_input_tokens_seen": 790548992,
      "step": 10620,
      "train_runtime": 12661.4462,
      "train_tokens_per_second": 62437.496
    },
    {
      "epoch": 0.1412615198570109,
      "grad_norm": 25.5,
      "learning_rate": 1.935796440167108e-05,
      "loss": 0.7315,
      "num_input_tokens_seen": 791016704,
      "step": 10630,
      "train_runtime": 12667.9948,
      "train_tokens_per_second": 62442.14
    },
    {
      "epoch": 0.14139440933947284,
      "grad_norm": 20.125,
      "learning_rate": 1.9356446211922364e-05,
      "loss": 0.6801,
      "num_input_tokens_seen": 791251200,
      "step": 10640,
      "train_runtime": 12671.5048,
      "train_tokens_per_second": 62443.349
    },
    {
      "epoch": 0.14152729882193474,
      "grad_norm": 11.9375,
      "learning_rate": 1.9354926288975418e-05,
      "loss": 0.5466,
      "num_input_tokens_seen": 791366400,
      "step": 10650,
      "train_runtime": 12673.8932,
      "train_tokens_per_second": 62440.671
    },
    {
      "epoch": 0.14166018830439664,
      "grad_norm": 26.875,
      "learning_rate": 1.9353404633111793e-05,
      "loss": 0.8077,
      "num_input_tokens_seen": 793402880,
      "step": 10660,
      "train_runtime": 12706.4948,
      "train_tokens_per_second": 62440.736
    },
    {
      "epoch": 0.14179307778685857,
      "grad_norm": 33.75,
      "learning_rate": 1.9351881244613366e-05,
      "loss": 0.8588,
      "num_input_tokens_seen": 794239744,
      "step": 10670,
      "train_runtime": 12719.269,
      "train_tokens_per_second": 62443.82
    },
    {
      "epoch": 0.14192596726932047,
      "grad_norm": 12.8125,
      "learning_rate": 1.9350356123762325e-05,
      "loss": 0.7553,
      "num_input_tokens_seen": 794704128,
      "step": 10680,
      "train_runtime": 12726.0971,
      "train_tokens_per_second": 62446.807
    },
    {
      "epoch": 0.14205885675178237,
      "grad_norm": 17.75,
      "learning_rate": 1.9348829270841187e-05,
      "loss": 0.6514,
      "num_input_tokens_seen": 794937344,
      "step": 10690,
      "train_runtime": 12730.0287,
      "train_tokens_per_second": 62445.841
    },
    {
      "epoch": 0.1421917462342443,
      "grad_norm": 9.25,
      "learning_rate": 1.9347300686132788e-05,
      "loss": 0.5453,
      "num_input_tokens_seen": 795058688,
      "step": 10700,
      "train_runtime": 12732.447,
      "train_tokens_per_second": 62443.511
    },
    {
      "epoch": 0.1423246357167062,
      "grad_norm": 24.375,
      "learning_rate": 1.9345770369920286e-05,
      "loss": 0.8878,
      "num_input_tokens_seen": 797082624,
      "step": 10710,
      "train_runtime": 12764.1758,
      "train_tokens_per_second": 62446.854
    },
    {
      "epoch": 0.1424575251991681,
      "grad_norm": 14.125,
      "learning_rate": 1.934423832248716e-05,
      "loss": 0.9932,
      "num_input_tokens_seen": 797892864,
      "step": 10720,
      "train_runtime": 12776.4365,
      "train_tokens_per_second": 62450.345
    },
    {
      "epoch": 0.14259041468163003,
      "grad_norm": 19.25,
      "learning_rate": 1.9342704544117202e-05,
      "loss": 0.745,
      "num_input_tokens_seen": 798328320,
      "step": 10730,
      "train_runtime": 12782.7797,
      "train_tokens_per_second": 62453.421
    },
    {
      "epoch": 0.14272330416409193,
      "grad_norm": 24.625,
      "learning_rate": 1.9341169035094536e-05,
      "loss": 0.6444,
      "num_input_tokens_seen": 798558976,
      "step": 10740,
      "train_runtime": 12786.6123,
      "train_tokens_per_second": 62452.74
    },
    {
      "epoch": 0.14285619364655383,
      "grad_norm": 13.375,
      "learning_rate": 1.9339631795703603e-05,
      "loss": 0.6286,
      "num_input_tokens_seen": 798673920,
      "step": 10750,
      "train_runtime": 12789.0162,
      "train_tokens_per_second": 62449.989
    },
    {
      "epoch": 0.14298908312901576,
      "grad_norm": 13.1875,
      "learning_rate": 1.9338092826229156e-05,
      "loss": 0.9633,
      "num_input_tokens_seen": 800640000,
      "step": 10760,
      "train_runtime": 12821.5046,
      "train_tokens_per_second": 62445.09
    },
    {
      "epoch": 0.14312197261147766,
      "grad_norm": 18.25,
      "learning_rate": 1.9336552126956282e-05,
      "loss": 1.0358,
      "num_input_tokens_seen": 801474048,
      "step": 10770,
      "train_runtime": 12834.1975,
      "train_tokens_per_second": 62448.318
    },
    {
      "epoch": 0.1432548620939396,
      "grad_norm": 13.5,
      "learning_rate": 1.9335009698170375e-05,
      "loss": 0.7384,
      "num_input_tokens_seen": 801978368,
      "step": 10780,
      "train_runtime": 12841.463,
      "train_tokens_per_second": 62452.258
    },
    {
      "epoch": 0.1433877515764015,
      "grad_norm": 19.625,
      "learning_rate": 1.9333465540157164e-05,
      "loss": 0.6713,
      "num_input_tokens_seen": 802206976,
      "step": 10790,
      "train_runtime": 12844.9292,
      "train_tokens_per_second": 62453.204
    },
    {
      "epoch": 0.1435206410588634,
      "grad_norm": 16.375,
      "learning_rate": 1.9331919653202684e-05,
      "loss": 0.6054,
      "num_input_tokens_seen": 802326528,
      "step": 10800,
      "train_runtime": 12847.307,
      "train_tokens_per_second": 62450.95
    },
    {
      "epoch": 0.14365353054132532,
      "grad_norm": 28.25,
      "learning_rate": 1.93303720375933e-05,
      "loss": 0.8033,
      "num_input_tokens_seen": 804488704,
      "step": 10810,
      "train_runtime": 12881.3643,
      "train_tokens_per_second": 62453.688
    },
    {
      "epoch": 0.14378642002378722,
      "grad_norm": 46.5,
      "learning_rate": 1.9328822693615694e-05,
      "loss": 0.9577,
      "num_input_tokens_seen": 805357824,
      "step": 10820,
      "train_runtime": 12894.0573,
      "train_tokens_per_second": 62459.613
    },
    {
      "epoch": 0.14391930950624912,
      "grad_norm": 25.0,
      "learning_rate": 1.9327271621556867e-05,
      "loss": 0.7385,
      "num_input_tokens_seen": 805853440,
      "step": 10830,
      "train_runtime": 12900.9458,
      "train_tokens_per_second": 62464.679
    },
    {
      "epoch": 0.14405219898871105,
      "grad_norm": 15.8125,
      "learning_rate": 1.9325718821704144e-05,
      "loss": 0.6885,
      "num_input_tokens_seen": 806091264,
      "step": 10840,
      "train_runtime": 12904.4842,
      "train_tokens_per_second": 62465.981
    },
    {
      "epoch": 0.14418508847117295,
      "grad_norm": 57.5,
      "learning_rate": 1.9324164294345164e-05,
      "loss": 0.5105,
      "num_input_tokens_seen": 806205952,
      "step": 10850,
      "train_runtime": 12907.2282,
      "train_tokens_per_second": 62461.586
    },
    {
      "epoch": 0.14431797795363485,
      "grad_norm": 62.0,
      "learning_rate": 1.932260803976789e-05,
      "loss": 0.8992,
      "num_input_tokens_seen": 808271872,
      "step": 10860,
      "train_runtime": 12940.2466,
      "train_tokens_per_second": 62461.86
    },
    {
      "epoch": 0.14445086743609678,
      "grad_norm": 18.875,
      "learning_rate": 1.9321050058260604e-05,
      "loss": 0.8978,
      "num_input_tokens_seen": 809067264,
      "step": 10870,
      "train_runtime": 12952.536,
      "train_tokens_per_second": 62464.004
    },
    {
      "epoch": 0.14458375691855868,
      "grad_norm": 39.0,
      "learning_rate": 1.931949035011191e-05,
      "loss": 0.7672,
      "num_input_tokens_seen": 809521152,
      "step": 10880,
      "train_runtime": 12959.1058,
      "train_tokens_per_second": 62467.362
    },
    {
      "epoch": 0.14471664640102058,
      "grad_norm": 11.0,
      "learning_rate": 1.9317928915610727e-05,
      "loss": 0.6766,
      "num_input_tokens_seen": 809748224,
      "step": 10890,
      "train_runtime": 12962.4999,
      "train_tokens_per_second": 62468.523
    },
    {
      "epoch": 0.1448495358834825,
      "grad_norm": 58.75,
      "learning_rate": 1.9316365755046305e-05,
      "loss": 0.544,
      "num_input_tokens_seen": 809862400,
      "step": 10900,
      "train_runtime": 12965.2695,
      "train_tokens_per_second": 62463.985
    },
    {
      "epoch": 0.1449824253659444,
      "grad_norm": 37.75,
      "learning_rate": 1.9314800868708192e-05,
      "loss": 0.8928,
      "num_input_tokens_seen": 812061696,
      "step": 10910,
      "train_runtime": 12999.637,
      "train_tokens_per_second": 62468.028
    },
    {
      "epoch": 0.1451153148484063,
      "grad_norm": 38.5,
      "learning_rate": 1.9313234256886278e-05,
      "loss": 0.9903,
      "num_input_tokens_seen": 812886016,
      "step": 10920,
      "train_runtime": 13012.24,
      "train_tokens_per_second": 62470.875
    },
    {
      "epoch": 0.14524820433086824,
      "grad_norm": 24.375,
      "learning_rate": 1.9311665919870765e-05,
      "loss": 0.7862,
      "num_input_tokens_seen": 813336064,
      "step": 10930,
      "train_runtime": 13018.9236,
      "train_tokens_per_second": 62473.373
    },
    {
      "epoch": 0.14538109381333014,
      "grad_norm": 10.8125,
      "learning_rate": 1.9310095857952174e-05,
      "loss": 0.7402,
      "num_input_tokens_seen": 813566976,
      "step": 10940,
      "train_runtime": 13022.3745,
      "train_tokens_per_second": 62474.549
    },
    {
      "epoch": 0.14551398329579204,
      "grad_norm": 8.625,
      "learning_rate": 1.9308524071421336e-05,
      "loss": 0.5768,
      "num_input_tokens_seen": 813678592,
      "step": 10950,
      "train_runtime": 13024.7674,
      "train_tokens_per_second": 62471.641
    },
    {
      "epoch": 0.14564687277825397,
      "grad_norm": 17.375,
      "learning_rate": 1.930695056056942e-05,
      "loss": 0.9316,
      "num_input_tokens_seen": 815757312,
      "step": 10960,
      "train_runtime": 13056.0485,
      "train_tokens_per_second": 62481.18
    },
    {
      "epoch": 0.14577976226071587,
      "grad_norm": 26.0,
      "learning_rate": 1.93053753256879e-05,
      "loss": 0.9812,
      "num_input_tokens_seen": 816635648,
      "step": 10970,
      "train_runtime": 13068.3199,
      "train_tokens_per_second": 62489.72
    },
    {
      "epoch": 0.14591265174317777,
      "grad_norm": 15.375,
      "learning_rate": 1.9303798367068573e-05,
      "loss": 0.7656,
      "num_input_tokens_seen": 817080832,
      "step": 10980,
      "train_runtime": 13075.0274,
      "train_tokens_per_second": 62491.711
    },
    {
      "epoch": 0.1460455412256397,
      "grad_norm": 14.6875,
      "learning_rate": 1.930221968500356e-05,
      "loss": 0.6224,
      "num_input_tokens_seen": 817301760,
      "step": 10990,
      "train_runtime": 13078.4428,
      "train_tokens_per_second": 62492.284
    },
    {
      "epoch": 0.1461784307081016,
      "grad_norm": 7.875,
      "learning_rate": 1.93006392797853e-05,
      "loss": 0.5432,
      "num_input_tokens_seen": 817417728,
      "step": 11000,
      "train_runtime": 13080.8706,
      "train_tokens_per_second": 62489.551
    },
    {
      "epoch": 0.1463113201905635,
      "grad_norm": 25.375,
      "learning_rate": 1.929905715170654e-05,
      "loss": 0.8421,
      "num_input_tokens_seen": 819434752,
      "step": 11010,
      "train_runtime": 13112.6567,
      "train_tokens_per_second": 62491.894
    },
    {
      "epoch": 0.14644420967302543,
      "grad_norm": 234.0,
      "learning_rate": 1.9297473301060367e-05,
      "loss": 0.9489,
      "num_input_tokens_seen": 820264704,
      "step": 11020,
      "train_runtime": 13124.9775,
      "train_tokens_per_second": 62496.466
    },
    {
      "epoch": 0.14657709915548733,
      "grad_norm": 16.375,
      "learning_rate": 1.9295887728140164e-05,
      "loss": 0.7852,
      "num_input_tokens_seen": 820729088,
      "step": 11030,
      "train_runtime": 13132.0143,
      "train_tokens_per_second": 62498.339
    },
    {
      "epoch": 0.14670998863794926,
      "grad_norm": 19.75,
      "learning_rate": 1.9294300433239656e-05,
      "loss": 0.6238,
      "num_input_tokens_seen": 820962816,
      "step": 11040,
      "train_runtime": 13135.5018,
      "train_tokens_per_second": 62499.54
    },
    {
      "epoch": 0.14684287812041116,
      "grad_norm": 36.25,
      "learning_rate": 1.9292711416652863e-05,
      "loss": 0.6062,
      "num_input_tokens_seen": 821081856,
      "step": 11050,
      "train_runtime": 13137.8869,
      "train_tokens_per_second": 62497.254
    },
    {
      "epoch": 0.14697576760287306,
      "grad_norm": 11.4375,
      "learning_rate": 1.929112067867414e-05,
      "loss": 0.904,
      "num_input_tokens_seen": 823091200,
      "step": 11060,
      "train_runtime": 13169.579,
      "train_tokens_per_second": 62499.431
    },
    {
      "epoch": 0.147108657085335,
      "grad_norm": 17.0,
      "learning_rate": 1.9289528219598163e-05,
      "loss": 0.9772,
      "num_input_tokens_seen": 823936512,
      "step": 11070,
      "train_runtime": 13182.334,
      "train_tokens_per_second": 62503.083
    },
    {
      "epoch": 0.1472415465677969,
      "grad_norm": 17.125,
      "learning_rate": 1.9287934039719914e-05,
      "loss": 0.8357,
      "num_input_tokens_seen": 824392448,
      "step": 11080,
      "train_runtime": 13189.0207,
      "train_tokens_per_second": 62505.964
    },
    {
      "epoch": 0.1473744360502588,
      "grad_norm": 12.5,
      "learning_rate": 1.9286338139334703e-05,
      "loss": 0.6757,
      "num_input_tokens_seen": 824615680,
      "step": 11090,
      "train_runtime": 13192.4262,
      "train_tokens_per_second": 62506.749
    },
    {
      "epoch": 0.14750732553272072,
      "grad_norm": 32.75,
      "learning_rate": 1.9284740518738157e-05,
      "loss": 0.5385,
      "num_input_tokens_seen": 824734464,
      "step": 11100,
      "train_runtime": 13194.8034,
      "train_tokens_per_second": 62504.49
    },
    {
      "epoch": 0.14764021501518262,
      "grad_norm": 10.625,
      "learning_rate": 1.9283141178226222e-05,
      "loss": 0.809,
      "num_input_tokens_seen": 826822912,
      "step": 11110,
      "train_runtime": 13227.9068,
      "train_tokens_per_second": 62505.952
    },
    {
      "epoch": 0.14777310449764453,
      "grad_norm": 14.5,
      "learning_rate": 1.9281540118095157e-05,
      "loss": 0.906,
      "num_input_tokens_seen": 827653120,
      "step": 11120,
      "train_runtime": 13240.6725,
      "train_tokens_per_second": 62508.39
    },
    {
      "epoch": 0.14790599398010645,
      "grad_norm": 30.25,
      "learning_rate": 1.9279937338641552e-05,
      "loss": 0.7537,
      "num_input_tokens_seen": 828130048,
      "step": 11130,
      "train_runtime": 13247.3519,
      "train_tokens_per_second": 62512.875
    },
    {
      "epoch": 0.14803888346256835,
      "grad_norm": 28.5,
      "learning_rate": 1.92783328401623e-05,
      "loss": 0.7161,
      "num_input_tokens_seen": 828360448,
      "step": 11140,
      "train_runtime": 13250.8182,
      "train_tokens_per_second": 62513.91
    },
    {
      "epoch": 0.14817177294503026,
      "grad_norm": 39.5,
      "learning_rate": 1.927672662295462e-05,
      "loss": 0.5367,
      "num_input_tokens_seen": 828476416,
      "step": 11150,
      "train_runtime": 13253.6157,
      "train_tokens_per_second": 62509.464
    },
    {
      "epoch": 0.14830466242749218,
      "grad_norm": 15.6875,
      "learning_rate": 1.9275118687316056e-05,
      "loss": 0.8713,
      "num_input_tokens_seen": 830596352,
      "step": 11160,
      "train_runtime": 13286.9928,
      "train_tokens_per_second": 62511.989
    },
    {
      "epoch": 0.14843755190995409,
      "grad_norm": 36.25,
      "learning_rate": 1.927350903354446e-05,
      "loss": 0.9449,
      "num_input_tokens_seen": 831472896,
      "step": 11170,
      "train_runtime": 13299.8744,
      "train_tokens_per_second": 62517.35
    },
    {
      "epoch": 0.148570441392416,
      "grad_norm": 53.25,
      "learning_rate": 1.9271897661938003e-05,
      "loss": 0.7351,
      "num_input_tokens_seen": 831931136,
      "step": 11180,
      "train_runtime": 13306.6339,
      "train_tokens_per_second": 62520.029
    },
    {
      "epoch": 0.14870333087487791,
      "grad_norm": 17.375,
      "learning_rate": 1.927028457279518e-05,
      "loss": 0.6436,
      "num_input_tokens_seen": 832158720,
      "step": 11190,
      "train_runtime": 13310.0707,
      "train_tokens_per_second": 62520.984
    },
    {
      "epoch": 0.14883622035733982,
      "grad_norm": 15.6875,
      "learning_rate": 1.9268669766414804e-05,
      "loss": 0.5154,
      "num_input_tokens_seen": 832276480,
      "step": 11200,
      "train_runtime": 13312.4615,
      "train_tokens_per_second": 62518.602
    },
    {
      "epoch": 0.14896910983980172,
      "grad_norm": 34.25,
      "learning_rate": 1.9267053243095998e-05,
      "loss": 0.8183,
      "num_input_tokens_seen": 834426368,
      "step": 11210,
      "train_runtime": 13347.3585,
      "train_tokens_per_second": 62516.218
    },
    {
      "epoch": 0.14910199932226365,
      "grad_norm": 34.0,
      "learning_rate": 1.9265435003138212e-05,
      "loss": 0.8523,
      "num_input_tokens_seen": 835268608,
      "step": 11220,
      "train_runtime": 13360.028,
      "train_tokens_per_second": 62519.975
    },
    {
      "epoch": 0.14923488880472555,
      "grad_norm": 26.375,
      "learning_rate": 1.926381504684121e-05,
      "loss": 0.8054,
      "num_input_tokens_seen": 835730176,
      "step": 11230,
      "train_runtime": 13366.694,
      "train_tokens_per_second": 62523.327
    },
    {
      "epoch": 0.14936777828718745,
      "grad_norm": 17.5,
      "learning_rate": 1.9262193374505078e-05,
      "loss": 0.6724,
      "num_input_tokens_seen": 835960576,
      "step": 11240,
      "train_runtime": 13370.1356,
      "train_tokens_per_second": 62524.465
    },
    {
      "epoch": 0.14950066776964938,
      "grad_norm": 24.0,
      "learning_rate": 1.926056998643021e-05,
      "loss": 0.529,
      "num_input_tokens_seen": 836073984,
      "step": 11250,
      "train_runtime": 13372.5383,
      "train_tokens_per_second": 62521.712
    },
    {
      "epoch": 0.14963355725211128,
      "grad_norm": 11.75,
      "learning_rate": 1.925894488291733e-05,
      "loss": 0.814,
      "num_input_tokens_seen": 838114816,
      "step": 11260,
      "train_runtime": 13406.0137,
      "train_tokens_per_second": 62517.825
    },
    {
      "epoch": 0.1497664467345732,
      "grad_norm": 20.25,
      "learning_rate": 1.9257318064267468e-05,
      "loss": 0.9093,
      "num_input_tokens_seen": 838970624,
      "step": 11270,
      "train_runtime": 13419.1272,
      "train_tokens_per_second": 62520.506
    },
    {
      "epoch": 0.1498993362170351,
      "grad_norm": 14.4375,
      "learning_rate": 1.925568953078198e-05,
      "loss": 0.8004,
      "num_input_tokens_seen": 839462656,
      "step": 11280,
      "train_runtime": 13425.8205,
      "train_tokens_per_second": 62525.985
    },
    {
      "epoch": 0.150032225699497,
      "grad_norm": 14.4375,
      "learning_rate": 1.925405928276254e-05,
      "loss": 0.589,
      "num_input_tokens_seen": 839692288,
      "step": 11290,
      "train_runtime": 13429.3329,
      "train_tokens_per_second": 62526.731
    },
    {
      "epoch": 0.15016511518195894,
      "grad_norm": 40.5,
      "learning_rate": 1.925242732051113e-05,
      "loss": 0.6135,
      "num_input_tokens_seen": 839808768,
      "step": 11300,
      "train_runtime": 13431.7736,
      "train_tokens_per_second": 62524.041
    },
    {
      "epoch": 0.15029800466442084,
      "grad_norm": 26.0,
      "learning_rate": 1.9250793644330068e-05,
      "loss": 0.8346,
      "num_input_tokens_seen": 841843712,
      "step": 11310,
      "train_runtime": 13463.5587,
      "train_tokens_per_second": 62527.578
    },
    {
      "epoch": 0.15043089414688274,
      "grad_norm": 25.0,
      "learning_rate": 1.924915825452197e-05,
      "loss": 0.9224,
      "num_input_tokens_seen": 842693888,
      "step": 11320,
      "train_runtime": 13476.3789,
      "train_tokens_per_second": 62531.181
    },
    {
      "epoch": 0.15056378362934467,
      "grad_norm": 18.125,
      "learning_rate": 1.9247521151389774e-05,
      "loss": 0.8066,
      "num_input_tokens_seen": 843159040,
      "step": 11330,
      "train_runtime": 13483.09,
      "train_tokens_per_second": 62534.555
    },
    {
      "epoch": 0.15069667311180657,
      "grad_norm": 26.625,
      "learning_rate": 1.9245882335236747e-05,
      "loss": 0.6994,
      "num_input_tokens_seen": 843396608,
      "step": 11340,
      "train_runtime": 13486.6453,
      "train_tokens_per_second": 62535.685
    },
    {
      "epoch": 0.15082956259426847,
      "grad_norm": 8.6875,
      "learning_rate": 1.924424180636646e-05,
      "loss": 0.5756,
      "num_input_tokens_seen": 843513856,
      "step": 11350,
      "train_runtime": 13489.0146,
      "train_tokens_per_second": 62533.393
    },
    {
      "epoch": 0.1509624520767304,
      "grad_norm": 21.375,
      "learning_rate": 1.924259956508281e-05,
      "loss": 0.9574,
      "num_input_tokens_seen": 845579264,
      "step": 11360,
      "train_runtime": 13522.3722,
      "train_tokens_per_second": 62531.873
    },
    {
      "epoch": 0.1510953415591923,
      "grad_norm": 27.625,
      "learning_rate": 1.924095561169e-05,
      "loss": 0.9373,
      "num_input_tokens_seen": 846445312,
      "step": 11370,
      "train_runtime": 13535.5131,
      "train_tokens_per_second": 62535.148
    },
    {
      "epoch": 0.1512282310416542,
      "grad_norm": 13.9375,
      "learning_rate": 1.9239309946492567e-05,
      "loss": 0.7692,
      "num_input_tokens_seen": 846919424,
      "step": 11380,
      "train_runtime": 13542.3586,
      "train_tokens_per_second": 62538.547
    },
    {
      "epoch": 0.15136112052411613,
      "grad_norm": 18.75,
      "learning_rate": 1.9237662569795353e-05,
      "loss": 0.6665,
      "num_input_tokens_seen": 847156224,
      "step": 11390,
      "train_runtime": 13545.849,
      "train_tokens_per_second": 62539.913
    },
    {
      "epoch": 0.15149401000657803,
      "grad_norm": 10.5625,
      "learning_rate": 1.9236013481903518e-05,
      "loss": 0.5105,
      "num_input_tokens_seen": 847273728,
      "step": 11400,
      "train_runtime": 13548.2593,
      "train_tokens_per_second": 62537.46
    },
    {
      "epoch": 0.15162689948903993,
      "grad_norm": 30.75,
      "learning_rate": 1.923436268312254e-05,
      "loss": 0.8419,
      "num_input_tokens_seen": 849258496,
      "step": 11410,
      "train_runtime": 13580.7834,
      "train_tokens_per_second": 62533.837
    },
    {
      "epoch": 0.15175978897150186,
      "grad_norm": 28.25,
      "learning_rate": 1.9232710173758215e-05,
      "loss": 0.8288,
      "num_input_tokens_seen": 850080256,
      "step": 11420,
      "train_runtime": 13593.237,
      "train_tokens_per_second": 62537.0
    },
    {
      "epoch": 0.15189267845396376,
      "grad_norm": 50.0,
      "learning_rate": 1.9231055954116658e-05,
      "loss": 0.8569,
      "num_input_tokens_seen": 850539776,
      "step": 11430,
      "train_runtime": 13599.7429,
      "train_tokens_per_second": 62540.871
    },
    {
      "epoch": 0.15202556793642566,
      "grad_norm": 11.5625,
      "learning_rate": 1.92294000245043e-05,
      "loss": 0.57,
      "num_input_tokens_seen": 850770944,
      "step": 11440,
      "train_runtime": 13603.2034,
      "train_tokens_per_second": 62541.956
    },
    {
      "epoch": 0.1521584574188876,
      "grad_norm": 18.125,
      "learning_rate": 1.9227742385227885e-05,
      "loss": 0.541,
      "num_input_tokens_seen": 850886400,
      "step": 11450,
      "train_runtime": 13605.9929,
      "train_tokens_per_second": 62537.619
    },
    {
      "epoch": 0.1522913469013495,
      "grad_norm": 15.1875,
      "learning_rate": 1.922608303659447e-05,
      "loss": 0.826,
      "num_input_tokens_seen": 852846080,
      "step": 11460,
      "train_runtime": 13636.8826,
      "train_tokens_per_second": 62539.666
    },
    {
      "epoch": 0.1524242363838114,
      "grad_norm": 31.0,
      "learning_rate": 1.9224421978911443e-05,
      "loss": 0.9282,
      "num_input_tokens_seen": 853680640,
      "step": 11470,
      "train_runtime": 13649.0907,
      "train_tokens_per_second": 62544.873
    },
    {
      "epoch": 0.15255712586627332,
      "grad_norm": 21.25,
      "learning_rate": 1.9222759212486498e-05,
      "loss": 0.7923,
      "num_input_tokens_seen": 854135040,
      "step": 11480,
      "train_runtime": 13655.5384,
      "train_tokens_per_second": 62548.617
    },
    {
      "epoch": 0.15269001534873522,
      "grad_norm": 14.375,
      "learning_rate": 1.9221094737627646e-05,
      "loss": 0.6583,
      "num_input_tokens_seen": 854357760,
      "step": 11490,
      "train_runtime": 13658.9387,
      "train_tokens_per_second": 62549.352
    },
    {
      "epoch": 0.15282290483119715,
      "grad_norm": 16.75,
      "learning_rate": 1.9219428554643216e-05,
      "loss": 0.5009,
      "num_input_tokens_seen": 854473472,
      "step": 11500,
      "train_runtime": 13661.365,
      "train_tokens_per_second": 62546.713
    },
    {
      "epoch": 0.15295579431365905,
      "grad_norm": 14.875,
      "learning_rate": 1.9217760663841854e-05,
      "loss": 0.8659,
      "num_input_tokens_seen": 856557056,
      "step": 11510,
      "train_runtime": 13694.825,
      "train_tokens_per_second": 62546.039
    },
    {
      "epoch": 0.15308868379612095,
      "grad_norm": 15.625,
      "learning_rate": 1.9216091065532523e-05,
      "loss": 0.8931,
      "num_input_tokens_seen": 857419264,
      "step": 11520,
      "train_runtime": 13707.6348,
      "train_tokens_per_second": 62550.489
    },
    {
      "epoch": 0.15322157327858288,
      "grad_norm": 18.75,
      "learning_rate": 1.9214419760024502e-05,
      "loss": 0.7517,
      "num_input_tokens_seen": 857894400,
      "step": 11530,
      "train_runtime": 13714.4751,
      "train_tokens_per_second": 62553.936
    },
    {
      "epoch": 0.15335446276104478,
      "grad_norm": 13.6875,
      "learning_rate": 1.9212746747627382e-05,
      "loss": 0.6396,
      "num_input_tokens_seen": 858122752,
      "step": 11540,
      "train_runtime": 13717.97,
      "train_tokens_per_second": 62554.645
    },
    {
      "epoch": 0.15348735224350668,
      "grad_norm": 11.6875,
      "learning_rate": 1.9211072028651076e-05,
      "loss": 0.5804,
      "num_input_tokens_seen": 858239488,
      "step": 11550,
      "train_runtime": 13720.3714,
      "train_tokens_per_second": 62552.205
    },
    {
      "epoch": 0.1536202417259686,
      "grad_norm": 34.25,
      "learning_rate": 1.920939560340581e-05,
      "loss": 0.8515,
      "num_input_tokens_seen": 860405248,
      "step": 11560,
      "train_runtime": 13755.1587,
      "train_tokens_per_second": 62551.459
    },
    {
      "epoch": 0.1537531312084305,
      "grad_norm": 21.5,
      "learning_rate": 1.9207717472202127e-05,
      "loss": 0.9323,
      "num_input_tokens_seen": 861244160,
      "step": 11570,
      "train_runtime": 13767.8921,
      "train_tokens_per_second": 62554.54
    },
    {
      "epoch": 0.1538860206908924,
      "grad_norm": 16.125,
      "learning_rate": 1.9206037635350886e-05,
      "loss": 0.7688,
      "num_input_tokens_seen": 861719296,
      "step": 11580,
      "train_runtime": 13774.6011,
      "train_tokens_per_second": 62558.566
    },
    {
      "epoch": 0.15401891017335434,
      "grad_norm": 12.625,
      "learning_rate": 1.9204356093163265e-05,
      "loss": 0.7479,
      "num_input_tokens_seen": 861949696,
      "step": 11590,
      "train_runtime": 13778.0481,
      "train_tokens_per_second": 62559.638
    },
    {
      "epoch": 0.15415179965581624,
      "grad_norm": 12.75,
      "learning_rate": 1.9202672845950746e-05,
      "loss": 0.6161,
      "num_input_tokens_seen": 862062336,
      "step": 11600,
      "train_runtime": 13780.3966,
      "train_tokens_per_second": 62557.15
    },
    {
      "epoch": 0.15428468913827814,
      "grad_norm": 23.875,
      "learning_rate": 1.9200987894025145e-05,
      "loss": 0.8413,
      "num_input_tokens_seen": 864196096,
      "step": 11610,
      "train_runtime": 13814.0848,
      "train_tokens_per_second": 62559.055
    },
    {
      "epoch": 0.15441757862074007,
      "grad_norm": 27.25,
      "learning_rate": 1.919930123769858e-05,
      "loss": 0.9049,
      "num_input_tokens_seen": 865031936,
      "step": 11620,
      "train_runtime": 13826.5701,
      "train_tokens_per_second": 62563.017
    },
    {
      "epoch": 0.15455046810320197,
      "grad_norm": 15.625,
      "learning_rate": 1.9197612877283486e-05,
      "loss": 0.7729,
      "num_input_tokens_seen": 865505536,
      "step": 11630,
      "train_runtime": 13833.1632,
      "train_tokens_per_second": 62567.435
    },
    {
      "epoch": 0.15468335758566387,
      "grad_norm": 15.9375,
      "learning_rate": 1.9195922813092626e-05,
      "loss": 0.6464,
      "num_input_tokens_seen": 865730560,
      "step": 11640,
      "train_runtime": 13836.5933,
      "train_tokens_per_second": 62568.187
    },
    {
      "epoch": 0.1548162470681258,
      "grad_norm": 20.75,
      "learning_rate": 1.9194231045439057e-05,
      "loss": 0.4994,
      "num_input_tokens_seen": 865844224,
      "step": 11650,
      "train_runtime": 13839.0203,
      "train_tokens_per_second": 62565.428
    },
    {
      "epoch": 0.1549491365505877,
      "grad_norm": 47.0,
      "learning_rate": 1.9192537574636175e-05,
      "loss": 0.9979,
      "num_input_tokens_seen": 867750912,
      "step": 11660,
      "train_runtime": 13869.5618,
      "train_tokens_per_second": 62565.128
    },
    {
      "epoch": 0.1550820260330496,
      "grad_norm": 18.0,
      "learning_rate": 1.919084240099767e-05,
      "loss": 0.9592,
      "num_input_tokens_seen": 868577024,
      "step": 11670,
      "train_runtime": 13882.4005,
      "train_tokens_per_second": 62566.775
    },
    {
      "epoch": 0.15521491551551153,
      "grad_norm": 13.1875,
      "learning_rate": 1.9189145524837568e-05,
      "loss": 0.7509,
      "num_input_tokens_seen": 869044992,
      "step": 11680,
      "train_runtime": 13889.0186,
      "train_tokens_per_second": 62570.655
    },
    {
      "epoch": 0.15534780499797343,
      "grad_norm": 20.75,
      "learning_rate": 1.9187446946470195e-05,
      "loss": 0.6606,
      "num_input_tokens_seen": 869270528,
      "step": 11690,
      "train_runtime": 13892.5007,
      "train_tokens_per_second": 62571.206
    },
    {
      "epoch": 0.15548069448043533,
      "grad_norm": 18.0,
      "learning_rate": 1.9185746666210198e-05,
      "loss": 0.5498,
      "num_input_tokens_seen": 869384192,
      "step": 11700,
      "train_runtime": 13894.904,
      "train_tokens_per_second": 62568.564
    },
    {
      "epoch": 0.15561358396289726,
      "grad_norm": 16.875,
      "learning_rate": 1.918404468437254e-05,
      "loss": 0.8482,
      "num_input_tokens_seen": 871352576,
      "step": 11710,
      "train_runtime": 13926.7129,
      "train_tokens_per_second": 62566.995
    },
    {
      "epoch": 0.15574647344535916,
      "grad_norm": 17.875,
      "learning_rate": 1.9182341001272497e-05,
      "loss": 0.9086,
      "num_input_tokens_seen": 872191232,
      "step": 11720,
      "train_runtime": 13939.876,
      "train_tokens_per_second": 62568.077
    },
    {
      "epoch": 0.1558793629278211,
      "grad_norm": 27.75,
      "learning_rate": 1.918063561722566e-05,
      "loss": 0.7081,
      "num_input_tokens_seen": 872655616,
      "step": 11730,
      "train_runtime": 13946.6296,
      "train_tokens_per_second": 62571.075
    },
    {
      "epoch": 0.156012252410283,
      "grad_norm": 11.3125,
      "learning_rate": 1.9178928532547937e-05,
      "loss": 0.6443,
      "num_input_tokens_seen": 872889344,
      "step": 11740,
      "train_runtime": 13950.1246,
      "train_tokens_per_second": 62572.154
    },
    {
      "epoch": 0.1561451418927449,
      "grad_norm": 19.5,
      "learning_rate": 1.917721974755555e-05,
      "loss": 0.5627,
      "num_input_tokens_seen": 873006080,
      "step": 11750,
      "train_runtime": 13952.5292,
      "train_tokens_per_second": 62569.737
    },
    {
      "epoch": 0.15627803137520682,
      "grad_norm": 27.875,
      "learning_rate": 1.9175509262565036e-05,
      "loss": 0.8365,
      "num_input_tokens_seen": 875067136,
      "step": 11760,
      "train_runtime": 13985.5149,
      "train_tokens_per_second": 62569.533
    },
    {
      "epoch": 0.15641092085766872,
      "grad_norm": 12.6875,
      "learning_rate": 1.917379707789325e-05,
      "loss": 0.9359,
      "num_input_tokens_seen": 875928320,
      "step": 11770,
      "train_runtime": 13998.129,
      "train_tokens_per_second": 62574.671
    },
    {
      "epoch": 0.15654381034013062,
      "grad_norm": 15.75,
      "learning_rate": 1.9172083193857353e-05,
      "loss": 0.7583,
      "num_input_tokens_seen": 876416768,
      "step": 11780,
      "train_runtime": 14004.8384,
      "train_tokens_per_second": 62579.57
    },
    {
      "epoch": 0.15667669982259255,
      "grad_norm": 14.4375,
      "learning_rate": 1.9170367610774833e-05,
      "loss": 0.6196,
      "num_input_tokens_seen": 876656128,
      "step": 11790,
      "train_runtime": 14008.8123,
      "train_tokens_per_second": 62578.905
    },
    {
      "epoch": 0.15680958930505445,
      "grad_norm": 10.625,
      "learning_rate": 1.916865032896348e-05,
      "loss": 0.5853,
      "num_input_tokens_seen": 876774912,
      "step": 11800,
      "train_runtime": 14011.2154,
      "train_tokens_per_second": 62576.649
    },
    {
      "epoch": 0.15694247878751635,
      "grad_norm": 21.5,
      "learning_rate": 1.9166931348741414e-05,
      "loss": 0.8752,
      "num_input_tokens_seen": 878870272,
      "step": 11810,
      "train_runtime": 14045.5603,
      "train_tokens_per_second": 62572.817
    },
    {
      "epoch": 0.15707536826997828,
      "grad_norm": 21.5,
      "learning_rate": 1.9165210670427054e-05,
      "loss": 0.9186,
      "num_input_tokens_seen": 879746304,
      "step": 11820,
      "train_runtime": 14058.254,
      "train_tokens_per_second": 62578.632
    },
    {
      "epoch": 0.15720825775244018,
      "grad_norm": 20.375,
      "learning_rate": 1.916348829433914e-05,
      "loss": 0.7385,
      "num_input_tokens_seen": 880200960,
      "step": 11830,
      "train_runtime": 14064.7875,
      "train_tokens_per_second": 62581.888
    },
    {
      "epoch": 0.15734114723490208,
      "grad_norm": 23.125,
      "learning_rate": 1.916176422079673e-05,
      "loss": 0.724,
      "num_input_tokens_seen": 880431616,
      "step": 11840,
      "train_runtime": 14068.293,
      "train_tokens_per_second": 62582.69
    },
    {
      "epoch": 0.157474036717364,
      "grad_norm": 14.0,
      "learning_rate": 1.916003845011919e-05,
      "loss": 0.5541,
      "num_input_tokens_seen": 880548864,
      "step": 11850,
      "train_runtime": 14071.6816,
      "train_tokens_per_second": 62575.951
    },
    {
      "epoch": 0.15760692619982591,
      "grad_norm": 17.125,
      "learning_rate": 1.9158310982626212e-05,
      "loss": 0.8209,
      "num_input_tokens_seen": 882600960,
      "step": 11860,
      "train_runtime": 14105.0678,
      "train_tokens_per_second": 62573.323
    },
    {
      "epoch": 0.15773981568228782,
      "grad_norm": 13.1875,
      "learning_rate": 1.9156581818637784e-05,
      "loss": 0.9542,
      "num_input_tokens_seen": 883449856,
      "step": 11870,
      "train_runtime": 14117.6438,
      "train_tokens_per_second": 62577.713
    },
    {
      "epoch": 0.15787270516474974,
      "grad_norm": 21.25,
      "learning_rate": 1.915485095847422e-05,
      "loss": 0.7521,
      "num_input_tokens_seen": 883902720,
      "step": 11880,
      "train_runtime": 14123.9878,
      "train_tokens_per_second": 62581.668
    },
    {
      "epoch": 0.15800559464721164,
      "grad_norm": 15.3125,
      "learning_rate": 1.915311840245615e-05,
      "loss": 0.6749,
      "num_input_tokens_seen": 884132352,
      "step": 11890,
      "train_runtime": 14127.4117,
      "train_tokens_per_second": 62582.755
    },
    {
      "epoch": 0.15813848412967355,
      "grad_norm": 32.75,
      "learning_rate": 1.9151384150904514e-05,
      "loss": 0.6068,
      "num_input_tokens_seen": 884249344,
      "step": 11900,
      "train_runtime": 14129.8639,
      "train_tokens_per_second": 62580.174
    },
    {
      "epoch": 0.15827137361213547,
      "grad_norm": 22.875,
      "learning_rate": 1.9149648204140563e-05,
      "loss": 0.896,
      "num_input_tokens_seen": 886262528,
      "step": 11910,
      "train_runtime": 14163.3582,
      "train_tokens_per_second": 62574.321
    },
    {
      "epoch": 0.15840426309459738,
      "grad_norm": 22.375,
      "learning_rate": 1.9147910562485867e-05,
      "loss": 0.9891,
      "num_input_tokens_seen": 887138304,
      "step": 11920,
      "train_runtime": 14176.5005,
      "train_tokens_per_second": 62578.089
    },
    {
      "epoch": 0.15853715257705928,
      "grad_norm": 10.375,
      "learning_rate": 1.9146171226262314e-05,
      "loss": 0.6703,
      "num_input_tokens_seen": 887644416,
      "step": 11930,
      "train_runtime": 14183.5464,
      "train_tokens_per_second": 62582.685
    },
    {
      "epoch": 0.1586700420595212,
      "grad_norm": 14.9375,
      "learning_rate": 1.914443019579209e-05,
      "loss": 0.6938,
      "num_input_tokens_seen": 887877120,
      "step": 11940,
      "train_runtime": 14187.1771,
      "train_tokens_per_second": 62583.072
    },
    {
      "epoch": 0.1588029315419831,
      "grad_norm": 15.125,
      "learning_rate": 1.9142687471397718e-05,
      "loss": 0.5667,
      "num_input_tokens_seen": 887995648,
      "step": 11950,
      "train_runtime": 14189.7203,
      "train_tokens_per_second": 62580.208
    },
    {
      "epoch": 0.158935821024445,
      "grad_norm": 27.25,
      "learning_rate": 1.914094305340201e-05,
      "loss": 0.8508,
      "num_input_tokens_seen": 889987840,
      "step": 11960,
      "train_runtime": 14222.1318,
      "train_tokens_per_second": 62577.668
    },
    {
      "epoch": 0.15906871050690694,
      "grad_norm": 17.25,
      "learning_rate": 1.9139196942128117e-05,
      "loss": 0.9703,
      "num_input_tokens_seen": 890825472,
      "step": 11970,
      "train_runtime": 14235.2841,
      "train_tokens_per_second": 62578.693
    },
    {
      "epoch": 0.15920159998936884,
      "grad_norm": 18.375,
      "learning_rate": 1.9137449137899476e-05,
      "loss": 0.7217,
      "num_input_tokens_seen": 891300096,
      "step": 11980,
      "train_runtime": 14242.0153,
      "train_tokens_per_second": 62582.442
    },
    {
      "epoch": 0.15933448947183076,
      "grad_norm": 11.75,
      "learning_rate": 1.9135699641039865e-05,
      "loss": 0.5935,
      "num_input_tokens_seen": 891530752,
      "step": 11990,
      "train_runtime": 14245.488,
      "train_tokens_per_second": 62583.378
    },
    {
      "epoch": 0.15946737895429267,
      "grad_norm": 17.875,
      "learning_rate": 1.9133948451873356e-05,
      "loss": 0.5179,
      "num_input_tokens_seen": 891644928,
      "step": 12000,
      "train_runtime": 14247.8878,
      "train_tokens_per_second": 62580.85
    },
    {
      "epoch": 0.15960026843675457,
      "grad_norm": 23.25,
      "learning_rate": 1.9132195570724338e-05,
      "loss": 0.8588,
      "num_input_tokens_seen": 893685504,
      "step": 12010,
      "train_runtime": 14280.5525,
      "train_tokens_per_second": 62580.597
    },
    {
      "epoch": 0.1597331579192165,
      "grad_norm": 14.25,
      "learning_rate": 1.9130440997917527e-05,
      "loss": 0.8593,
      "num_input_tokens_seen": 894469632,
      "step": 12020,
      "train_runtime": 14292.3075,
      "train_tokens_per_second": 62583.99
    },
    {
      "epoch": 0.1598660474016784,
      "grad_norm": 23.25,
      "learning_rate": 1.912868473377793e-05,
      "loss": 0.8031,
      "num_input_tokens_seen": 894895104,
      "step": 12030,
      "train_runtime": 14298.4481,
      "train_tokens_per_second": 62586.87
    },
    {
      "epoch": 0.1599989368841403,
      "grad_norm": 21.375,
      "learning_rate": 1.912692677863089e-05,
      "loss": 0.7038,
      "num_input_tokens_seen": 895109120,
      "step": 12040,
      "train_runtime": 14301.7669,
      "train_tokens_per_second": 62587.31
    },
    {
      "epoch": 0.16013182636660223,
      "grad_norm": 16.125,
      "learning_rate": 1.912516713280204e-05,
      "loss": 0.5609,
      "num_input_tokens_seen": 895224064,
      "step": 12050,
      "train_runtime": 14304.1444,
      "train_tokens_per_second": 62584.943
    },
    {
      "epoch": 0.16026471584906413,
      "grad_norm": 67.5,
      "learning_rate": 1.912340579661735e-05,
      "loss": 0.795,
      "num_input_tokens_seen": 897222400,
      "step": 12060,
      "train_runtime": 14334.4453,
      "train_tokens_per_second": 62592.056
    },
    {
      "epoch": 0.16039760533152603,
      "grad_norm": 21.375,
      "learning_rate": 1.912164277040309e-05,
      "loss": 0.9355,
      "num_input_tokens_seen": 898042112,
      "step": 12070,
      "train_runtime": 14346.7118,
      "train_tokens_per_second": 62595.675
    },
    {
      "epoch": 0.16053049481398796,
      "grad_norm": 11.6875,
      "learning_rate": 1.9119878054485844e-05,
      "loss": 0.7786,
      "num_input_tokens_seen": 898498304,
      "step": 12080,
      "train_runtime": 14353.7406,
      "train_tokens_per_second": 62596.805
    },
    {
      "epoch": 0.16066338429644986,
      "grad_norm": 26.5,
      "learning_rate": 1.9118111649192505e-05,
      "loss": 0.6747,
      "num_input_tokens_seen": 898719744,
      "step": 12090,
      "train_runtime": 14357.0575,
      "train_tokens_per_second": 62597.767
    },
    {
      "epoch": 0.16079627377891176,
      "grad_norm": 13.3125,
      "learning_rate": 1.911634355485029e-05,
      "loss": 0.5323,
      "num_input_tokens_seen": 898831360,
      "step": 12100,
      "train_runtime": 14359.4566,
      "train_tokens_per_second": 62595.082
    },
    {
      "epoch": 0.1609291632613737,
      "grad_norm": 11.75,
      "learning_rate": 1.9114573771786722e-05,
      "loss": 0.8179,
      "num_input_tokens_seen": 900886272,
      "step": 12110,
      "train_runtime": 14394.4022,
      "train_tokens_per_second": 62585.876
    },
    {
      "epoch": 0.1610620527438356,
      "grad_norm": 21.125,
      "learning_rate": 1.9112802300329633e-05,
      "loss": 0.9093,
      "num_input_tokens_seen": 901770752,
      "step": 12120,
      "train_runtime": 14407.2896,
      "train_tokens_per_second": 62591.284
    },
    {
      "epoch": 0.1611949422262975,
      "grad_norm": 16.375,
      "learning_rate": 1.9111029140807176e-05,
      "loss": 0.7901,
      "num_input_tokens_seen": 902268160,
      "step": 12130,
      "train_runtime": 14414.5935,
      "train_tokens_per_second": 62594.076
    },
    {
      "epoch": 0.16132783170875942,
      "grad_norm": 8.6875,
      "learning_rate": 1.9109254293547814e-05,
      "loss": 0.7297,
      "num_input_tokens_seen": 902503168,
      "step": 12140,
      "train_runtime": 14418.0613,
      "train_tokens_per_second": 62595.321
    },
    {
      "epoch": 0.16146072119122132,
      "grad_norm": 32.25,
      "learning_rate": 1.910747775888032e-05,
      "loss": 0.5681,
      "num_input_tokens_seen": 902616832,
      "step": 12150,
      "train_runtime": 14420.463,
      "train_tokens_per_second": 62592.778
    },
    {
      "epoch": 0.16159361067368322,
      "grad_norm": 32.5,
      "learning_rate": 1.9105699537133776e-05,
      "loss": 0.9075,
      "num_input_tokens_seen": 904657920,
      "step": 12160,
      "train_runtime": 14453.0426,
      "train_tokens_per_second": 62592.905
    },
    {
      "epoch": 0.16172650015614515,
      "grad_norm": 25.875,
      "learning_rate": 1.910391962863759e-05,
      "loss": 0.9276,
      "num_input_tokens_seen": 905482496,
      "step": 12170,
      "train_runtime": 14466.0893,
      "train_tokens_per_second": 62593.454
    },
    {
      "epoch": 0.16185938963860705,
      "grad_norm": 29.125,
      "learning_rate": 1.9102138033721475e-05,
      "loss": 0.7657,
      "num_input_tokens_seen": 905952256,
      "step": 12180,
      "train_runtime": 14472.8002,
      "train_tokens_per_second": 62596.888
    },
    {
      "epoch": 0.16199227912106895,
      "grad_norm": 11.5,
      "learning_rate": 1.9100354752715448e-05,
      "loss": 0.5962,
      "num_input_tokens_seen": 906177280,
      "step": 12190,
      "train_runtime": 14476.151,
      "train_tokens_per_second": 62597.943
    },
    {
      "epoch": 0.16212516860353088,
      "grad_norm": 15.1875,
      "learning_rate": 1.9098569785949846e-05,
      "loss": 0.5365,
      "num_input_tokens_seen": 906291712,
      "step": 12200,
      "train_runtime": 14478.5021,
      "train_tokens_per_second": 62595.682
    },
    {
      "epoch": 0.16225805808599278,
      "grad_norm": 18.75,
      "learning_rate": 1.9096783133755325e-05,
      "loss": 0.8383,
      "num_input_tokens_seen": 908271360,
      "step": 12210,
      "train_runtime": 14510.9328,
      "train_tokens_per_second": 62592.21
    },
    {
      "epoch": 0.1623909475684547,
      "grad_norm": 19.875,
      "learning_rate": 1.9094994796462844e-05,
      "loss": 0.8985,
      "num_input_tokens_seen": 909159936,
      "step": 12220,
      "train_runtime": 14523.7173,
      "train_tokens_per_second": 62598.295
    },
    {
      "epoch": 0.1625238370509166,
      "grad_norm": 35.25,
      "learning_rate": 1.9093204774403673e-05,
      "loss": 0.7353,
      "num_input_tokens_seen": 909638912,
      "step": 12230,
      "train_runtime": 14530.6345,
      "train_tokens_per_second": 62601.458
    },
    {
      "epoch": 0.1626567265333785,
      "grad_norm": 72.0,
      "learning_rate": 1.90914130679094e-05,
      "loss": 0.7277,
      "num_input_tokens_seen": 909876224,
      "step": 12240,
      "train_runtime": 14534.2076,
      "train_tokens_per_second": 62602.396
    },
    {
      "epoch": 0.16278961601584044,
      "grad_norm": 28.375,
      "learning_rate": 1.9089619677311925e-05,
      "loss": 0.5488,
      "num_input_tokens_seen": 909992448,
      "step": 12250,
      "train_runtime": 14537.0647,
      "train_tokens_per_second": 62598.088
    },
    {
      "epoch": 0.16292250549830234,
      "grad_norm": 12.1875,
      "learning_rate": 1.9087824602943455e-05,
      "loss": 0.8474,
      "num_input_tokens_seen": 912046080,
      "step": 12260,
      "train_runtime": 14569.8823,
      "train_tokens_per_second": 62598.04
    },
    {
      "epoch": 0.16305539498076424,
      "grad_norm": 14.8125,
      "learning_rate": 1.9086027845136512e-05,
      "loss": 0.9138,
      "num_input_tokens_seen": 912882176,
      "step": 12270,
      "train_runtime": 14582.6891,
      "train_tokens_per_second": 62600.4
    },
    {
      "epoch": 0.16318828446322617,
      "grad_norm": 16.75,
      "learning_rate": 1.9084229404223928e-05,
      "loss": 0.8269,
      "num_input_tokens_seen": 913380096,
      "step": 12280,
      "train_runtime": 14589.557,
      "train_tokens_per_second": 62605.06
    },
    {
      "epoch": 0.16332117394568807,
      "grad_norm": 21.0,
      "learning_rate": 1.908242928053885e-05,
      "loss": 0.7151,
      "num_input_tokens_seen": 913610240,
      "step": 12290,
      "train_runtime": 14593.0619,
      "train_tokens_per_second": 62605.795
    },
    {
      "epoch": 0.16345406342814997,
      "grad_norm": 22.125,
      "learning_rate": 1.9080627474414733e-05,
      "loss": 0.5015,
      "num_input_tokens_seen": 913726208,
      "step": 12300,
      "train_runtime": 14595.463,
      "train_tokens_per_second": 62603.441
    },
    {
      "epoch": 0.1635869529106119,
      "grad_norm": 18.0,
      "learning_rate": 1.907882398618535e-05,
      "loss": 0.8327,
      "num_input_tokens_seen": 915773696,
      "step": 12310,
      "train_runtime": 14628.3729,
      "train_tokens_per_second": 62602.567
    },
    {
      "epoch": 0.1637198423930738,
      "grad_norm": 10.5625,
      "learning_rate": 1.907701881618478e-05,
      "loss": 0.885,
      "num_input_tokens_seen": 916656640,
      "step": 12320,
      "train_runtime": 14641.4674,
      "train_tokens_per_second": 62606.883
    },
    {
      "epoch": 0.1638527318755357,
      "grad_norm": 26.5,
      "learning_rate": 1.9075211964747406e-05,
      "loss": 0.7576,
      "num_input_tokens_seen": 917139456,
      "step": 12330,
      "train_runtime": 14648.3803,
      "train_tokens_per_second": 62610.298
    },
    {
      "epoch": 0.16398562135799763,
      "grad_norm": 11.1875,
      "learning_rate": 1.907340343220794e-05,
      "loss": 0.5742,
      "num_input_tokens_seen": 917379840,
      "step": 12340,
      "train_runtime": 14651.9205,
      "train_tokens_per_second": 62611.576
    },
    {
      "epoch": 0.16411851084045953,
      "grad_norm": 17.875,
      "learning_rate": 1.9071593218901397e-05,
      "loss": 0.5791,
      "num_input_tokens_seen": 917493760,
      "step": 12350,
      "train_runtime": 14654.339,
      "train_tokens_per_second": 62609.017
    },
    {
      "epoch": 0.16425140032292143,
      "grad_norm": 16.625,
      "learning_rate": 1.9069781325163097e-05,
      "loss": 0.9439,
      "num_input_tokens_seen": 919573760,
      "step": 12360,
      "train_runtime": 14686.6094,
      "train_tokens_per_second": 62613.074
    },
    {
      "epoch": 0.16438428980538336,
      "grad_norm": 13.5625,
      "learning_rate": 1.906796775132868e-05,
      "loss": 0.8931,
      "num_input_tokens_seen": 920427008,
      "step": 12370,
      "train_runtime": 14699.444,
      "train_tokens_per_second": 62616.451
    },
    {
      "epoch": 0.16451717928784526,
      "grad_norm": 19.0,
      "learning_rate": 1.9066152497734096e-05,
      "loss": 0.7925,
      "num_input_tokens_seen": 920945664,
      "step": 12380,
      "train_runtime": 14706.397,
      "train_tokens_per_second": 62622.114
    },
    {
      "epoch": 0.16465006877030716,
      "grad_norm": 18.125,
      "learning_rate": 1.9064335564715604e-05,
      "loss": 0.673,
      "num_input_tokens_seen": 921190912,
      "step": 12390,
      "train_runtime": 14709.9501,
      "train_tokens_per_second": 62623.66
    },
    {
      "epoch": 0.1647829582527691,
      "grad_norm": 11.5625,
      "learning_rate": 1.906251695260977e-05,
      "loss": 0.5421,
      "num_input_tokens_seen": 921308160,
      "step": 12400,
      "train_runtime": 14712.3344,
      "train_tokens_per_second": 62621.48
    },
    {
      "epoch": 0.164915847735231,
      "grad_norm": 15.3125,
      "learning_rate": 1.9060696661753483e-05,
      "loss": 0.7888,
      "num_input_tokens_seen": 923384576,
      "step": 12410,
      "train_runtime": 14745.9355,
      "train_tokens_per_second": 62619.6
    },
    {
      "epoch": 0.1650487372176929,
      "grad_norm": 28.25,
      "learning_rate": 1.9058874692483933e-05,
      "loss": 1.0371,
      "num_input_tokens_seen": 924240896,
      "step": 12420,
      "train_runtime": 14759.1792,
      "train_tokens_per_second": 62621.429
    },
    {
      "epoch": 0.16518162670015482,
      "grad_norm": 19.875,
      "learning_rate": 1.9057051045138623e-05,
      "loss": 0.7583,
      "num_input_tokens_seen": 924705536,
      "step": 12430,
      "train_runtime": 14765.7933,
      "train_tokens_per_second": 62624.846
    },
    {
      "epoch": 0.16531451618261672,
      "grad_norm": 9.375,
      "learning_rate": 1.9055225720055365e-05,
      "loss": 0.6506,
      "num_input_tokens_seen": 924932608,
      "step": 12440,
      "train_runtime": 14769.2434,
      "train_tokens_per_second": 62625.592
    },
    {
      "epoch": 0.16544740566507865,
      "grad_norm": 12.3125,
      "learning_rate": 1.9053398717572283e-05,
      "loss": 0.6478,
      "num_input_tokens_seen": 925046784,
      "step": 12450,
      "train_runtime": 14771.6443,
      "train_tokens_per_second": 62623.142
    },
    {
      "epoch": 0.16558029514754055,
      "grad_norm": 25.75,
      "learning_rate": 1.9051570038027822e-05,
      "loss": 0.8376,
      "num_input_tokens_seen": 927028480,
      "step": 12460,
      "train_runtime": 14804.5903,
      "train_tokens_per_second": 62617.638
    },
    {
      "epoch": 0.16571318463000245,
      "grad_norm": 16.25,
      "learning_rate": 1.904973968176072e-05,
      "loss": 0.9488,
      "num_input_tokens_seen": 927884288,
      "step": 12470,
      "train_runtime": 14817.5036,
      "train_tokens_per_second": 62620.824
    },
    {
      "epoch": 0.16584607411246438,
      "grad_norm": 17.0,
      "learning_rate": 1.904790764911004e-05,
      "loss": 0.6978,
      "num_input_tokens_seen": 928340480,
      "step": 12480,
      "train_runtime": 14824.0456,
      "train_tokens_per_second": 62623.963
    },
    {
      "epoch": 0.16597896359492628,
      "grad_norm": 15.8125,
      "learning_rate": 1.904607394041514e-05,
      "loss": 0.7201,
      "num_input_tokens_seen": 928568320,
      "step": 12490,
      "train_runtime": 14827.4273,
      "train_tokens_per_second": 62625.046
    },
    {
      "epoch": 0.16611185307738818,
      "grad_norm": 18.625,
      "learning_rate": 1.9044238556015714e-05,
      "loss": 0.6383,
      "num_input_tokens_seen": 928682240,
      "step": 12500,
      "train_runtime": 14829.7507,
      "train_tokens_per_second": 62622.917
    },
    {
      "epoch": 0.1662447425598501,
      "grad_norm": 32.25,
      "learning_rate": 1.9042401496251733e-05,
      "loss": 0.8688,
      "num_input_tokens_seen": 930766080,
      "step": 12510,
      "train_runtime": 14863.3711,
      "train_tokens_per_second": 62621.466
    },
    {
      "epoch": 0.166377632042312,
      "grad_norm": 27.25,
      "learning_rate": 1.9040562761463512e-05,
      "loss": 0.9444,
      "num_input_tokens_seen": 931634176,
      "step": 12520,
      "train_runtime": 14876.0108,
      "train_tokens_per_second": 62626.613
    },
    {
      "epoch": 0.1665105215247739,
      "grad_norm": 18.875,
      "learning_rate": 1.9038722351991648e-05,
      "loss": 0.7299,
      "num_input_tokens_seen": 932090880,
      "step": 12530,
      "train_runtime": 14882.6972,
      "train_tokens_per_second": 62629.164
    },
    {
      "epoch": 0.16664341100723584,
      "grad_norm": 14.0625,
      "learning_rate": 1.9036880268177064e-05,
      "loss": 0.7341,
      "num_input_tokens_seen": 932323072,
      "step": 12540,
      "train_runtime": 14886.1824,
      "train_tokens_per_second": 62630.098
    },
    {
      "epoch": 0.16677630048969774,
      "grad_norm": 6.09375,
      "learning_rate": 1.9035036510360995e-05,
      "loss": 0.4826,
      "num_input_tokens_seen": 932440320,
      "step": 12550,
      "train_runtime": 14888.5736,
      "train_tokens_per_second": 62627.915
    },
    {
      "epoch": 0.16690918997215964,
      "grad_norm": 19.125,
      "learning_rate": 1.9033191078884973e-05,
      "loss": 0.7924,
      "num_input_tokens_seen": 934563584,
      "step": 12560,
      "train_runtime": 14923.7621,
      "train_tokens_per_second": 62622.52
    },
    {
      "epoch": 0.16704207945462157,
      "grad_norm": 13.25,
      "learning_rate": 1.9031343974090853e-05,
      "loss": 0.8741,
      "num_input_tokens_seen": 935420416,
      "step": 12570,
      "train_runtime": 14936.9296,
      "train_tokens_per_second": 62624.679
    },
    {
      "epoch": 0.16717496893708347,
      "grad_norm": 27.75,
      "learning_rate": 1.902949519632079e-05,
      "loss": 0.8034,
      "num_input_tokens_seen": 935902464,
      "step": 12580,
      "train_runtime": 14943.7831,
      "train_tokens_per_second": 62628.215
    },
    {
      "epoch": 0.16730785841954537,
      "grad_norm": 17.875,
      "learning_rate": 1.902764474591726e-05,
      "loss": 0.6652,
      "num_input_tokens_seen": 936150784,
      "step": 12590,
      "train_runtime": 14947.8363,
      "train_tokens_per_second": 62627.846
    },
    {
      "epoch": 0.1674407479020073,
      "grad_norm": 35.0,
      "learning_rate": 1.902579262322304e-05,
      "loss": 0.5696,
      "num_input_tokens_seen": 936273152,
      "step": 12600,
      "train_runtime": 14950.2331,
      "train_tokens_per_second": 62625.99
    },
    {
      "epoch": 0.1675736373844692,
      "grad_norm": 40.0,
      "learning_rate": 1.9023938828581212e-05,
      "loss": 0.8418,
      "num_input_tokens_seen": 938296320,
      "step": 12610,
      "train_runtime": 14982.9674,
      "train_tokens_per_second": 62624.198
    },
    {
      "epoch": 0.1677065268669311,
      "grad_norm": 10.8125,
      "learning_rate": 1.9022083362335184e-05,
      "loss": 0.9522,
      "num_input_tokens_seen": 939153664,
      "step": 12620,
      "train_runtime": 14995.6376,
      "train_tokens_per_second": 62628.458
    },
    {
      "epoch": 0.16783941634939303,
      "grad_norm": 17.25,
      "learning_rate": 1.9020226224828662e-05,
      "loss": 0.762,
      "num_input_tokens_seen": 939612928,
      "step": 12630,
      "train_runtime": 15002.1308,
      "train_tokens_per_second": 62631.965
    },
    {
      "epoch": 0.16797230583185493,
      "grad_norm": 12.3125,
      "learning_rate": 1.9018367416405663e-05,
      "loss": 0.6402,
      "num_input_tokens_seen": 939843840,
      "step": 12640,
      "train_runtime": 15005.559,
      "train_tokens_per_second": 62633.044
    },
    {
      "epoch": 0.16810519531431684,
      "grad_norm": 14.4375,
      "learning_rate": 1.901650693741052e-05,
      "loss": 0.5337,
      "num_input_tokens_seen": 939957248,
      "step": 12650,
      "train_runtime": 15008.3479,
      "train_tokens_per_second": 62628.962
    },
    {
      "epoch": 0.16823808479677876,
      "grad_norm": 31.625,
      "learning_rate": 1.901464478818786e-05,
      "loss": 0.821,
      "num_input_tokens_seen": 942049792,
      "step": 12660,
      "train_runtime": 15040.5056,
      "train_tokens_per_second": 62634.184
    },
    {
      "epoch": 0.16837097427924066,
      "grad_norm": 14.9375,
      "learning_rate": 1.9012780969082633e-05,
      "loss": 0.9857,
      "num_input_tokens_seen": 942932736,
      "step": 12670,
      "train_runtime": 15053.3037,
      "train_tokens_per_second": 62639.588
    },
    {
      "epoch": 0.1685038637617026,
      "grad_norm": 23.25,
      "learning_rate": 1.9010915480440105e-05,
      "loss": 0.7711,
      "num_input_tokens_seen": 943399168,
      "step": 12680,
      "train_runtime": 15059.9795,
      "train_tokens_per_second": 62642.792
    },
    {
      "epoch": 0.1686367532441645,
      "grad_norm": 13.8125,
      "learning_rate": 1.9009048322605825e-05,
      "loss": 0.7064,
      "num_input_tokens_seen": 943626752,
      "step": 12690,
      "train_runtime": 15063.837,
      "train_tokens_per_second": 62641.859
    },
    {
      "epoch": 0.1687696427266264,
      "grad_norm": 11.625,
      "learning_rate": 1.900717949592568e-05,
      "loss": 0.5712,
      "num_input_tokens_seen": 943736576,
      "step": 12700,
      "train_runtime": 15066.1835,
      "train_tokens_per_second": 62639.392
    },
    {
      "epoch": 0.16890253220908832,
      "grad_norm": 10.75,
      "learning_rate": 1.9005309000745847e-05,
      "loss": 0.848,
      "num_input_tokens_seen": 945923584,
      "step": 12710,
      "train_runtime": 15100.7065,
      "train_tokens_per_second": 62641.015
    },
    {
      "epoch": 0.16903542169155023,
      "grad_norm": 63.5,
      "learning_rate": 1.900343683741282e-05,
      "loss": 0.9575,
      "num_input_tokens_seen": 946784256,
      "step": 12720,
      "train_runtime": 15113.6411,
      "train_tokens_per_second": 62644.352
    },
    {
      "epoch": 0.16916831117401213,
      "grad_norm": 110.5,
      "learning_rate": 1.9001563006273404e-05,
      "loss": 0.7759,
      "num_input_tokens_seen": 947261952,
      "step": 12730,
      "train_runtime": 15120.4869,
      "train_tokens_per_second": 62647.583
    },
    {
      "epoch": 0.16930120065647405,
      "grad_norm": 14.9375,
      "learning_rate": 1.8999687507674708e-05,
      "loss": 0.657,
      "num_input_tokens_seen": 947491584,
      "step": 12740,
      "train_runtime": 15123.9526,
      "train_tokens_per_second": 62648.41
    },
    {
      "epoch": 0.16943409013893596,
      "grad_norm": 11.1875,
      "learning_rate": 1.899781034196415e-05,
      "loss": 0.5648,
      "num_input_tokens_seen": 947606784,
      "step": 12750,
      "train_runtime": 15126.318,
      "train_tokens_per_second": 62646.229
    },
    {
      "epoch": 0.16956697962139786,
      "grad_norm": 16.5,
      "learning_rate": 1.899593150948945e-05,
      "loss": 0.8739,
      "num_input_tokens_seen": 949609216,
      "step": 12760,
      "train_runtime": 15159.7311,
      "train_tokens_per_second": 62640.242
    },
    {
      "epoch": 0.16969986910385979,
      "grad_norm": 22.75,
      "learning_rate": 1.8994051010598657e-05,
      "loss": 0.9043,
      "num_input_tokens_seen": 950462208,
      "step": 12770,
      "train_runtime": 15173.2996,
      "train_tokens_per_second": 62640.443
    },
    {
      "epoch": 0.16983275858632169,
      "grad_norm": 14.875,
      "learning_rate": 1.8992168845640117e-05,
      "loss": 0.7641,
      "num_input_tokens_seen": 950987008,
      "step": 12780,
      "train_runtime": 15180.4827,
      "train_tokens_per_second": 62645.373
    },
    {
      "epoch": 0.1699656480687836,
      "grad_norm": 16.125,
      "learning_rate": 1.8990285014962477e-05,
      "loss": 0.6875,
      "num_input_tokens_seen": 951227904,
      "step": 12790,
      "train_runtime": 15184.0843,
      "train_tokens_per_second": 62646.379
    },
    {
      "epoch": 0.17009853755124552,
      "grad_norm": 15.625,
      "learning_rate": 1.89883995189147e-05,
      "loss": 0.5122,
      "num_input_tokens_seen": 951344640,
      "step": 12800,
      "train_runtime": 15186.5974,
      "train_tokens_per_second": 62643.699
    },
    {
      "epoch": 0.17023142703370742,
      "grad_norm": 65.0,
      "learning_rate": 1.8986512357846065e-05,
      "loss": 0.8812,
      "num_input_tokens_seen": 953462016,
      "step": 12810,
      "train_runtime": 15220.3183,
      "train_tokens_per_second": 62644.026
    },
    {
      "epoch": 0.17036431651616932,
      "grad_norm": 19.375,
      "learning_rate": 1.8984623532106146e-05,
      "loss": 0.9469,
      "num_input_tokens_seen": 954312960,
      "step": 12820,
      "train_runtime": 15233.1873,
      "train_tokens_per_second": 62646.966
    },
    {
      "epoch": 0.17049720599863125,
      "grad_norm": 8.9375,
      "learning_rate": 1.8982733042044832e-05,
      "loss": 0.7476,
      "num_input_tokens_seen": 954812672,
      "step": 12830,
      "train_runtime": 15240.2927,
      "train_tokens_per_second": 62650.547
    },
    {
      "epoch": 0.17063009548109315,
      "grad_norm": 16.125,
      "learning_rate": 1.898084088801232e-05,
      "loss": 0.6574,
      "num_input_tokens_seen": 955046400,
      "step": 12840,
      "train_runtime": 15243.7811,
      "train_tokens_per_second": 62651.543
    },
    {
      "epoch": 0.17076298496355505,
      "grad_norm": 16.5,
      "learning_rate": 1.8978947070359108e-05,
      "loss": 0.5864,
      "num_input_tokens_seen": 955162112,
      "step": 12850,
      "train_runtime": 15246.1476,
      "train_tokens_per_second": 62649.407
    },
    {
      "epoch": 0.17089587444601698,
      "grad_norm": 38.75,
      "learning_rate": 1.8977051589436022e-05,
      "loss": 0.8527,
      "num_input_tokens_seen": 957145856,
      "step": 12860,
      "train_runtime": 15279.016,
      "train_tokens_per_second": 62644.47
    },
    {
      "epoch": 0.17102876392847888,
      "grad_norm": 64.5,
      "learning_rate": 1.8975154445594172e-05,
      "loss": 0.9522,
      "num_input_tokens_seen": 958011648,
      "step": 12870,
      "train_runtime": 15291.8527,
      "train_tokens_per_second": 62648.501
    },
    {
      "epoch": 0.17116165341094078,
      "grad_norm": 45.0,
      "learning_rate": 1.8973255639184994e-05,
      "loss": 0.7058,
      "num_input_tokens_seen": 958468608,
      "step": 12880,
      "train_runtime": 15298.8105,
      "train_tokens_per_second": 62649.878
    },
    {
      "epoch": 0.1712945428934027,
      "grad_norm": 13.0625,
      "learning_rate": 1.897135517056022e-05,
      "loss": 0.6141,
      "num_input_tokens_seen": 958700288,
      "step": 12890,
      "train_runtime": 15302.2686,
      "train_tokens_per_second": 62650.86
    },
    {
      "epoch": 0.1714274323758646,
      "grad_norm": 10.75,
      "learning_rate": 1.89694530400719e-05,
      "loss": 0.599,
      "num_input_tokens_seen": 958812672,
      "step": 12900,
      "train_runtime": 15304.6113,
      "train_tokens_per_second": 62648.613
    },
    {
      "epoch": 0.17156032185832654,
      "grad_norm": 12.3125,
      "learning_rate": 1.896754924807238e-05,
      "loss": 0.8417,
      "num_input_tokens_seen": 960793344,
      "step": 12910,
      "train_runtime": 15336.235,
      "train_tokens_per_second": 62648.58
    },
    {
      "epoch": 0.17169321134078844,
      "grad_norm": 30.5,
      "learning_rate": 1.8965643794914323e-05,
      "loss": 0.9784,
      "num_input_tokens_seen": 961618944,
      "step": 12920,
      "train_runtime": 15348.6763,
      "train_tokens_per_second": 62651.588
    },
    {
      "epoch": 0.17182610082325034,
      "grad_norm": 10.8125,
      "learning_rate": 1.8963736680950702e-05,
      "loss": 0.775,
      "num_input_tokens_seen": 962089984,
      "step": 12930,
      "train_runtime": 15355.7132,
      "train_tokens_per_second": 62653.553
    },
    {
      "epoch": 0.17195899030571227,
      "grad_norm": 22.125,
      "learning_rate": 1.8961827906534785e-05,
      "loss": 0.5982,
      "num_input_tokens_seen": 962314240,
      "step": 12940,
      "train_runtime": 15359.0638,
      "train_tokens_per_second": 62654.485
    },
    {
      "epoch": 0.17209187978817417,
      "grad_norm": 8.625,
      "learning_rate": 1.8959917472020163e-05,
      "loss": 0.4978,
      "num_input_tokens_seen": 962425600,
      "step": 12950,
      "train_runtime": 15361.458,
      "train_tokens_per_second": 62651.969
    },
    {
      "epoch": 0.17222476927063607,
      "grad_norm": 23.75,
      "learning_rate": 1.8958005377760724e-05,
      "loss": 0.8084,
      "num_input_tokens_seen": 964530688,
      "step": 12960,
      "train_runtime": 15395.394,
      "train_tokens_per_second": 62650.601
    },
    {
      "epoch": 0.172357658753098,
      "grad_norm": 13.125,
      "learning_rate": 1.895609162411066e-05,
      "loss": 0.8522,
      "num_input_tokens_seen": 965390336,
      "step": 12970,
      "train_runtime": 15408.222,
      "train_tokens_per_second": 62654.233
    },
    {
      "epoch": 0.1724905482355599,
      "grad_norm": 52.25,
      "learning_rate": 1.895417621142449e-05,
      "loss": 0.7704,
      "num_input_tokens_seen": 965850880,
      "step": 12980,
      "train_runtime": 15414.922,
      "train_tokens_per_second": 62656.877
    },
    {
      "epoch": 0.1726234377180218,
      "grad_norm": 14.8125,
      "learning_rate": 1.8952259140057016e-05,
      "loss": 0.6281,
      "num_input_tokens_seen": 966085376,
      "step": 12990,
      "train_runtime": 15418.3663,
      "train_tokens_per_second": 62658.09
    },
    {
      "epoch": 0.17275632720048373,
      "grad_norm": 21.25,
      "learning_rate": 1.8950340410363363e-05,
      "loss": 0.5998,
      "num_input_tokens_seen": 966200576,
      "step": 13000,
      "train_runtime": 15421.2164,
      "train_tokens_per_second": 62653.979
    },
    {
      "epoch": 0.17288921668294563,
      "grad_norm": 20.0,
      "learning_rate": 1.8948420022698953e-05,
      "loss": 0.8847,
      "num_input_tokens_seen": 968271872,
      "step": 13010,
      "train_runtime": 15454.0733,
      "train_tokens_per_second": 62654.8
    },
    {
      "epoch": 0.17302210616540753,
      "grad_norm": 11.75,
      "learning_rate": 1.8946497977419528e-05,
      "loss": 0.9326,
      "num_input_tokens_seen": 969142528,
      "step": 13020,
      "train_runtime": 15466.8651,
      "train_tokens_per_second": 62659.273
    },
    {
      "epoch": 0.17315499564786946,
      "grad_norm": 10.4375,
      "learning_rate": 1.894457427488113e-05,
      "loss": 0.7888,
      "num_input_tokens_seen": 969645824,
      "step": 13030,
      "train_runtime": 15473.557,
      "train_tokens_per_second": 62664.701
    },
    {
      "epoch": 0.17328788513033136,
      "grad_norm": 13.625,
      "learning_rate": 1.8942648915440106e-05,
      "loss": 0.6911,
      "num_input_tokens_seen": 969880320,
      "step": 13040,
      "train_runtime": 15477.0718,
      "train_tokens_per_second": 62665.621
    },
    {
      "epoch": 0.17342077461279326,
      "grad_norm": 5.8125,
      "learning_rate": 1.8940721899453106e-05,
      "loss": 0.5982,
      "num_input_tokens_seen": 969994752,
      "step": 13050,
      "train_runtime": 15479.9113,
      "train_tokens_per_second": 62661.519
    },
    {
      "epoch": 0.1735536640952552,
      "grad_norm": 26.875,
      "learning_rate": 1.8938793227277097e-05,
      "loss": 0.8806,
      "num_input_tokens_seen": 972084480,
      "step": 13060,
      "train_runtime": 15513.6591,
      "train_tokens_per_second": 62659.91
    },
    {
      "epoch": 0.1736865535777171,
      "grad_norm": 14.1875,
      "learning_rate": 1.893686289926935e-05,
      "loss": 0.952,
      "num_input_tokens_seen": 972939264,
      "step": 13070,
      "train_runtime": 15526.5703,
      "train_tokens_per_second": 62662.858
    },
    {
      "epoch": 0.173819443060179,
      "grad_norm": 22.875,
      "learning_rate": 1.893493091578744e-05,
      "loss": 0.7197,
      "num_input_tokens_seen": 973419264,
      "step": 13080,
      "train_runtime": 15533.3438,
      "train_tokens_per_second": 62666.434
    },
    {
      "epoch": 0.17395233254264092,
      "grad_norm": 18.125,
      "learning_rate": 1.893299727718925e-05,
      "loss": 0.6433,
      "num_input_tokens_seen": 973649408,
      "step": 13090,
      "train_runtime": 15536.8808,
      "train_tokens_per_second": 62666.981
    },
    {
      "epoch": 0.17408522202510282,
      "grad_norm": 18.375,
      "learning_rate": 1.893106198383297e-05,
      "loss": 0.514,
      "num_input_tokens_seen": 973767680,
      "step": 13100,
      "train_runtime": 15539.319,
      "train_tokens_per_second": 62664.759
    },
    {
      "epoch": 0.17421811150756472,
      "grad_norm": 19.25,
      "learning_rate": 1.8929125036077088e-05,
      "loss": 0.8149,
      "num_input_tokens_seen": 975944192,
      "step": 13110,
      "train_runtime": 15574.106,
      "train_tokens_per_second": 62664.54
    },
    {
      "epoch": 0.17435100099002665,
      "grad_norm": 23.375,
      "learning_rate": 1.8927186434280418e-05,
      "loss": 0.9289,
      "num_input_tokens_seen": 976776704,
      "step": 13120,
      "train_runtime": 15586.7015,
      "train_tokens_per_second": 62667.313
    },
    {
      "epoch": 0.17448389047248855,
      "grad_norm": 30.75,
      "learning_rate": 1.892524617880206e-05,
      "loss": 0.8493,
      "num_input_tokens_seen": 977258240,
      "step": 13130,
      "train_runtime": 15593.5787,
      "train_tokens_per_second": 62670.556
    },
    {
      "epoch": 0.17461677995495045,
      "grad_norm": 11.75,
      "learning_rate": 1.892330427000144e-05,
      "loss": 0.641,
      "num_input_tokens_seen": 977494016,
      "step": 13140,
      "train_runtime": 15597.1924,
      "train_tokens_per_second": 62671.152
    },
    {
      "epoch": 0.17474966943741238,
      "grad_norm": 37.75,
      "learning_rate": 1.8921360708238263e-05,
      "loss": 0.6151,
      "num_input_tokens_seen": 977610752,
      "step": 13150,
      "train_runtime": 15599.6317,
      "train_tokens_per_second": 62668.835
    },
    {
      "epoch": 0.17488255891987428,
      "grad_norm": 26.875,
      "learning_rate": 1.8919415493872567e-05,
      "loss": 0.8554,
      "num_input_tokens_seen": 979741184,
      "step": 13160,
      "train_runtime": 15633.8443,
      "train_tokens_per_second": 62667.963
    },
    {
      "epoch": 0.1750154484023362,
      "grad_norm": 40.0,
      "learning_rate": 1.8917468627264685e-05,
      "loss": 0.9375,
      "num_input_tokens_seen": 980581120,
      "step": 13170,
      "train_runtime": 15646.5354,
      "train_tokens_per_second": 62670.815
    },
    {
      "epoch": 0.1751483378847981,
      "grad_norm": 13.5,
      "learning_rate": 1.8915520108775253e-05,
      "loss": 0.7456,
      "num_input_tokens_seen": 981062656,
      "step": 13180,
      "train_runtime": 15653.4261,
      "train_tokens_per_second": 62673.989
    },
    {
      "epoch": 0.17528122736726,
      "grad_norm": 11.3125,
      "learning_rate": 1.8913569938765223e-05,
      "loss": 0.6259,
      "num_input_tokens_seen": 981298688,
      "step": 13190,
      "train_runtime": 15656.9786,
      "train_tokens_per_second": 62674.844
    },
    {
      "epoch": 0.17541411684972194,
      "grad_norm": 22.5,
      "learning_rate": 1.8911618117595837e-05,
      "loss": 0.5843,
      "num_input_tokens_seen": 981415168,
      "step": 13200,
      "train_runtime": 15659.3281,
      "train_tokens_per_second": 62672.879
    },
    {
      "epoch": 0.17554700633218384,
      "grad_norm": 19.125,
      "learning_rate": 1.8909664645628663e-05,
      "loss": 0.835,
      "num_input_tokens_seen": 983600896,
      "step": 13210,
      "train_runtime": 15693.9822,
      "train_tokens_per_second": 62673.762
    },
    {
      "epoch": 0.17567989581464574,
      "grad_norm": 25.625,
      "learning_rate": 1.890770952322556e-05,
      "loss": 0.9047,
      "num_input_tokens_seen": 984475904,
      "step": 13220,
      "train_runtime": 15707.396,
      "train_tokens_per_second": 62675.946
    },
    {
      "epoch": 0.17581278529710767,
      "grad_norm": 19.125,
      "learning_rate": 1.8905752750748693e-05,
      "loss": 0.7204,
      "num_input_tokens_seen": 984955136,
      "step": 13230,
      "train_runtime": 15714.0886,
      "train_tokens_per_second": 62679.75
    },
    {
      "epoch": 0.17594567477956957,
      "grad_norm": 14.0625,
      "learning_rate": 1.890379432856054e-05,
      "loss": 0.6328,
      "num_input_tokens_seen": 985186560,
      "step": 13240,
      "train_runtime": 15717.5285,
      "train_tokens_per_second": 62680.756
    },
    {
      "epoch": 0.17607856426203147,
      "grad_norm": 19.375,
      "learning_rate": 1.8901834257023883e-05,
      "loss": 0.5692,
      "num_input_tokens_seen": 985300224,
      "step": 13250,
      "train_runtime": 15719.8733,
      "train_tokens_per_second": 62678.636
    },
    {
      "epoch": 0.1762114537444934,
      "grad_norm": 57.0,
      "learning_rate": 1.8899872536501807e-05,
      "loss": 0.8632,
      "num_input_tokens_seen": 987280128,
      "step": 13260,
      "train_runtime": 15752.0415,
      "train_tokens_per_second": 62676.328
    },
    {
      "epoch": 0.1763443432269553,
      "grad_norm": 42.75,
      "learning_rate": 1.8897909167357704e-05,
      "loss": 0.8931,
      "num_input_tokens_seen": 988101120,
      "step": 13270,
      "train_runtime": 15764.9016,
      "train_tokens_per_second": 62677.278
    },
    {
      "epoch": 0.1764772327094172,
      "grad_norm": 147.0,
      "learning_rate": 1.8895944149955264e-05,
      "loss": 0.7219,
      "num_input_tokens_seen": 988559872,
      "step": 13280,
      "train_runtime": 15771.3051,
      "train_tokens_per_second": 62680.917
    },
    {
      "epoch": 0.17661012219187913,
      "grad_norm": 15.4375,
      "learning_rate": 1.8893977484658496e-05,
      "loss": 0.6306,
      "num_input_tokens_seen": 988781056,
      "step": 13290,
      "train_runtime": 15774.6013,
      "train_tokens_per_second": 62681.841
    },
    {
      "epoch": 0.17674301167434103,
      "grad_norm": 8.125,
      "learning_rate": 1.8892009171831705e-05,
      "loss": 0.5264,
      "num_input_tokens_seen": 988894720,
      "step": 13300,
      "train_runtime": 15776.9405,
      "train_tokens_per_second": 62679.752
    },
    {
      "epoch": 0.17687590115680293,
      "grad_norm": 15.0625,
      "learning_rate": 1.8890039211839505e-05,
      "loss": 0.941,
      "num_input_tokens_seen": 990969344,
      "step": 13310,
      "train_runtime": 15810.0154,
      "train_tokens_per_second": 62679.847
    },
    {
      "epoch": 0.17700879063926486,
      "grad_norm": 12.0625,
      "learning_rate": 1.8888067605046808e-05,
      "loss": 0.9084,
      "num_input_tokens_seen": 991817984,
      "step": 13320,
      "train_runtime": 15822.7761,
      "train_tokens_per_second": 62682.931
    },
    {
      "epoch": 0.17714168012172676,
      "grad_norm": 21.75,
      "learning_rate": 1.8886094351818844e-05,
      "loss": 0.765,
      "num_input_tokens_seen": 992292096,
      "step": 13330,
      "train_runtime": 15829.3977,
      "train_tokens_per_second": 62686.662
    },
    {
      "epoch": 0.17727456960418866,
      "grad_norm": 12.9375,
      "learning_rate": 1.8884119452521138e-05,
      "loss": 0.6821,
      "num_input_tokens_seen": 992527616,
      "step": 13340,
      "train_runtime": 15832.91,
      "train_tokens_per_second": 62687.631
    },
    {
      "epoch": 0.1774074590866506,
      "grad_norm": 9.125,
      "learning_rate": 1.888214290751952e-05,
      "loss": 0.5758,
      "num_input_tokens_seen": 992648448,
      "step": 13350,
      "train_runtime": 15835.282,
      "train_tokens_per_second": 62685.871
    },
    {
      "epoch": 0.1775403485691125,
      "grad_norm": 13.125,
      "learning_rate": 1.8880164717180127e-05,
      "loss": 0.8504,
      "num_input_tokens_seen": 994724352,
      "step": 13360,
      "train_runtime": 15869.81,
      "train_tokens_per_second": 62680.294
    },
    {
      "epoch": 0.1776732380515744,
      "grad_norm": 11.625,
      "learning_rate": 1.8878184881869404e-05,
      "loss": 0.86,
      "num_input_tokens_seen": 995551232,
      "step": 13370,
      "train_runtime": 15882.011,
      "train_tokens_per_second": 62684.205
    },
    {
      "epoch": 0.17780612753403632,
      "grad_norm": 16.875,
      "learning_rate": 1.88762034019541e-05,
      "loss": 0.7597,
      "num_input_tokens_seen": 995991552,
      "step": 13380,
      "train_runtime": 15888.375,
      "train_tokens_per_second": 62686.811
    },
    {
      "epoch": 0.17793901701649822,
      "grad_norm": 25.75,
      "learning_rate": 1.887422027780126e-05,
      "loss": 0.7499,
      "num_input_tokens_seen": 996213248,
      "step": 13390,
      "train_runtime": 15892.2525,
      "train_tokens_per_second": 62685.465
    },
    {
      "epoch": 0.17807190649896015,
      "grad_norm": 8.5,
      "learning_rate": 1.8872235509778246e-05,
      "loss": 0.4972,
      "num_input_tokens_seen": 996328960,
      "step": 13400,
      "train_runtime": 15894.6046,
      "train_tokens_per_second": 62683.469
    },
    {
      "epoch": 0.17820479598142205,
      "grad_norm": 38.75,
      "learning_rate": 1.8870249098252717e-05,
      "loss": 0.8083,
      "num_input_tokens_seen": 998266112,
      "step": 13410,
      "train_runtime": 15926.5068,
      "train_tokens_per_second": 62679.539
    },
    {
      "epoch": 0.17833768546388395,
      "grad_norm": 42.0,
      "learning_rate": 1.8868261043592632e-05,
      "loss": 0.9089,
      "num_input_tokens_seen": 999124224,
      "step": 13420,
      "train_runtime": 15939.2835,
      "train_tokens_per_second": 62683.133
    },
    {
      "epoch": 0.17847057494634588,
      "grad_norm": 41.75,
      "learning_rate": 1.8866271346166272e-05,
      "loss": 0.7927,
      "num_input_tokens_seen": 999584256,
      "step": 13430,
      "train_runtime": 15945.9819,
      "train_tokens_per_second": 62685.651
    },
    {
      "epoch": 0.17860346442880778,
      "grad_norm": 27.25,
      "learning_rate": 1.88642800063422e-05,
      "loss": 0.718,
      "num_input_tokens_seen": 999817728,
      "step": 13440,
      "train_runtime": 15949.8746,
      "train_tokens_per_second": 62684.99
    },
    {
      "epoch": 0.17873635391126969,
      "grad_norm": 7.40625,
      "learning_rate": 1.88622870244893e-05,
      "loss": 0.4691,
      "num_input_tokens_seen": 999931136,
      "step": 13450,
      "train_runtime": 15952.2326,
      "train_tokens_per_second": 62682.833
    },
    {
      "epoch": 0.17886924339373161,
      "grad_norm": 17.125,
      "learning_rate": 1.886029240097675e-05,
      "loss": 0.8513,
      "num_input_tokens_seen": 1002062592,
      "step": 13460,
      "train_runtime": 15985.6926,
      "train_tokens_per_second": 62684.966
    },
    {
      "epoch": 0.17900213287619351,
      "grad_norm": 17.125,
      "learning_rate": 1.8858296136174042e-05,
      "loss": 0.9633,
      "num_input_tokens_seen": 1002914048,
      "step": 13470,
      "train_runtime": 15998.4877,
      "train_tokens_per_second": 62688.053
    },
    {
      "epoch": 0.17913502235865542,
      "grad_norm": 22.25,
      "learning_rate": 1.885629823045096e-05,
      "loss": 0.7858,
      "num_input_tokens_seen": 1003398400,
      "step": 13480,
      "train_runtime": 16005.8167,
      "train_tokens_per_second": 62689.61
    },
    {
      "epoch": 0.17926791184111734,
      "grad_norm": 21.75,
      "learning_rate": 1.88542986841776e-05,
      "loss": 0.6954,
      "num_input_tokens_seen": 1003634176,
      "step": 13490,
      "train_runtime": 16009.373,
      "train_tokens_per_second": 62690.411
    },
    {
      "epoch": 0.17940080132357925,
      "grad_norm": 8.625,
      "learning_rate": 1.8852297497724364e-05,
      "loss": 0.5742,
      "num_input_tokens_seen": 1003750912,
      "step": 13500,
      "train_runtime": 16011.7725,
      "train_tokens_per_second": 62688.307
    },
    {
      "epoch": 0.17953369080604115,
      "grad_norm": 28.125,
      "learning_rate": 1.8850294671461952e-05,
      "loss": 0.8224,
      "num_input_tokens_seen": 1005832192,
      "step": 13510,
      "train_runtime": 16046.5945,
      "train_tokens_per_second": 62681.972
    },
    {
      "epoch": 0.17966658028850307,
      "grad_norm": 12.1875,
      "learning_rate": 1.8848290205761365e-05,
      "loss": 0.9297,
      "num_input_tokens_seen": 1006656768,
      "step": 13520,
      "train_runtime": 16059.0617,
      "train_tokens_per_second": 62684.656
    },
    {
      "epoch": 0.17979946977096498,
      "grad_norm": 23.375,
      "learning_rate": 1.8846284100993916e-05,
      "loss": 0.787,
      "num_input_tokens_seen": 1007113216,
      "step": 13530,
      "train_runtime": 16065.6805,
      "train_tokens_per_second": 62687.243
    },
    {
      "epoch": 0.17993235925342688,
      "grad_norm": 29.0,
      "learning_rate": 1.884427635753122e-05,
      "loss": 0.6903,
      "num_input_tokens_seen": 1007337984,
      "step": 13540,
      "train_runtime": 16069.0978,
      "train_tokens_per_second": 62687.899
    },
    {
      "epoch": 0.1800652487358888,
      "grad_norm": 9.0625,
      "learning_rate": 1.8842266975745192e-05,
      "loss": 0.5345,
      "num_input_tokens_seen": 1007452416,
      "step": 13550,
      "train_runtime": 16071.5236,
      "train_tokens_per_second": 62685.558
    },
    {
      "epoch": 0.1801981382183507,
      "grad_norm": 18.75,
      "learning_rate": 1.8840255956008052e-05,
      "loss": 0.8245,
      "num_input_tokens_seen": 1009447936,
      "step": 13560,
      "train_runtime": 16103.5402,
      "train_tokens_per_second": 62684.846
    },
    {
      "epoch": 0.1803310277008126,
      "grad_norm": 54.75,
      "learning_rate": 1.883824329869232e-05,
      "loss": 0.9445,
      "num_input_tokens_seen": 1010264320,
      "step": 13570,
      "train_runtime": 16115.667,
      "train_tokens_per_second": 62688.334
    },
    {
      "epoch": 0.18046391718327454,
      "grad_norm": 32.0,
      "learning_rate": 1.883622900417083e-05,
      "loss": 0.7512,
      "num_input_tokens_seen": 1010704128,
      "step": 13580,
      "train_runtime": 16122.077,
      "train_tokens_per_second": 62690.69
    },
    {
      "epoch": 0.18059680666573644,
      "grad_norm": 16.125,
      "learning_rate": 1.8834213072816707e-05,
      "loss": 0.653,
      "num_input_tokens_seen": 1010938112,
      "step": 13590,
      "train_runtime": 16125.5798,
      "train_tokens_per_second": 62691.582
    },
    {
      "epoch": 0.18072969614819834,
      "grad_norm": 14.125,
      "learning_rate": 1.8832195505003383e-05,
      "loss": 0.504,
      "num_input_tokens_seen": 1011050496,
      "step": 13600,
      "train_runtime": 16128.4591,
      "train_tokens_per_second": 62687.358
    },
    {
      "epoch": 0.18086258563066027,
      "grad_norm": 10.1875,
      "learning_rate": 1.8830176301104603e-05,
      "loss": 0.8826,
      "num_input_tokens_seen": 1013114112,
      "step": 13610,
      "train_runtime": 16162.4895,
      "train_tokens_per_second": 62683.049
    },
    {
      "epoch": 0.18099547511312217,
      "grad_norm": 28.875,
      "learning_rate": 1.8828155461494396e-05,
      "loss": 0.9102,
      "num_input_tokens_seen": 1013951232,
      "step": 13620,
      "train_runtime": 16175.5129,
      "train_tokens_per_second": 62684.333
    },
    {
      "epoch": 0.1811283645955841,
      "grad_norm": 14.25,
      "learning_rate": 1.882613298654711e-05,
      "loss": 0.7405,
      "num_input_tokens_seen": 1014442752,
      "step": 13630,
      "train_runtime": 16182.5432,
      "train_tokens_per_second": 62687.474
    },
    {
      "epoch": 0.181261254078046,
      "grad_norm": 10.6875,
      "learning_rate": 1.882410887663739e-05,
      "loss": 0.6215,
      "num_input_tokens_seen": 1014675200,
      "step": 13640,
      "train_runtime": 16186.0547,
      "train_tokens_per_second": 62688.235
    },
    {
      "epoch": 0.1813941435605079,
      "grad_norm": 12.0625,
      "learning_rate": 1.882208313214018e-05,
      "loss": 0.5334,
      "num_input_tokens_seen": 1014791168,
      "step": 13650,
      "train_runtime": 16188.4335,
      "train_tokens_per_second": 62686.187
    },
    {
      "epoch": 0.18152703304296983,
      "grad_norm": 22.375,
      "learning_rate": 1.8820055753430738e-05,
      "loss": 0.7733,
      "num_input_tokens_seen": 1016730112,
      "step": 13660,
      "train_runtime": 16220.4049,
      "train_tokens_per_second": 62682.166
    },
    {
      "epoch": 0.18165992252543173,
      "grad_norm": 27.0,
      "learning_rate": 1.8818026740884614e-05,
      "loss": 0.9409,
      "num_input_tokens_seen": 1017578496,
      "step": 13670,
      "train_runtime": 16233.1452,
      "train_tokens_per_second": 62685.233
    },
    {
      "epoch": 0.18179281200789363,
      "grad_norm": 21.0,
      "learning_rate": 1.8815996094877668e-05,
      "loss": 0.752,
      "num_input_tokens_seen": 1018052864,
      "step": 13680,
      "train_runtime": 16240.439,
      "train_tokens_per_second": 62686.29
    },
    {
      "epoch": 0.18192570149035556,
      "grad_norm": 11.3125,
      "learning_rate": 1.881396381578605e-05,
      "loss": 0.753,
      "num_input_tokens_seen": 1018286592,
      "step": 13690,
      "train_runtime": 16243.934,
      "train_tokens_per_second": 62687.191
    },
    {
      "epoch": 0.18205859097281746,
      "grad_norm": 12.9375,
      "learning_rate": 1.881192990398623e-05,
      "loss": 0.5753,
      "num_input_tokens_seen": 1018405376,
      "step": 13700,
      "train_runtime": 16246.3216,
      "train_tokens_per_second": 62685.29
    },
    {
      "epoch": 0.18219148045527936,
      "grad_norm": 27.375,
      "learning_rate": 1.8809894359854973e-05,
      "loss": 0.7955,
      "num_input_tokens_seen": 1020413952,
      "step": 13710,
      "train_runtime": 16278.4872,
      "train_tokens_per_second": 62684.815
    },
    {
      "epoch": 0.1823243699377413,
      "grad_norm": 30.75,
      "learning_rate": 1.880785718376934e-05,
      "loss": 0.975,
      "num_input_tokens_seen": 1021233920,
      "step": 13720,
      "train_runtime": 16291.0221,
      "train_tokens_per_second": 62686.915
    },
    {
      "epoch": 0.1824572594202032,
      "grad_norm": 14.8125,
      "learning_rate": 1.8805818376106703e-05,
      "loss": 0.7307,
      "num_input_tokens_seen": 1021727232,
      "step": 13730,
      "train_runtime": 16297.6897,
      "train_tokens_per_second": 62691.538
    },
    {
      "epoch": 0.1825901489026651,
      "grad_norm": 15.5625,
      "learning_rate": 1.8803777937244733e-05,
      "loss": 0.6379,
      "num_input_tokens_seen": 1021960704,
      "step": 13740,
      "train_runtime": 16301.5699,
      "train_tokens_per_second": 62690.938
    },
    {
      "epoch": 0.18272303838512702,
      "grad_norm": 7.75,
      "learning_rate": 1.88017358675614e-05,
      "loss": 0.4433,
      "num_input_tokens_seen": 1022074624,
      "step": 13750,
      "train_runtime": 16303.938,
      "train_tokens_per_second": 62688.819
    },
    {
      "epoch": 0.18285592786758892,
      "grad_norm": 17.5,
      "learning_rate": 1.879969216743498e-05,
      "loss": 0.9267,
      "num_input_tokens_seen": 1024114688,
      "step": 13760,
      "train_runtime": 16336.9911,
      "train_tokens_per_second": 62686.861
    },
    {
      "epoch": 0.18298881735005082,
      "grad_norm": 13.375,
      "learning_rate": 1.8797646837244052e-05,
      "loss": 0.9576,
      "num_input_tokens_seen": 1024943872,
      "step": 13770,
      "train_runtime": 16349.5336,
      "train_tokens_per_second": 62689.487
    },
    {
      "epoch": 0.18312170683251275,
      "grad_norm": 18.5,
      "learning_rate": 1.8795599877367495e-05,
      "loss": 0.7701,
      "num_input_tokens_seen": 1025418496,
      "step": 13780,
      "train_runtime": 16356.7153,
      "train_tokens_per_second": 62690.979
    },
    {
      "epoch": 0.18325459631497465,
      "grad_norm": 10.3125,
      "learning_rate": 1.879355128818449e-05,
      "loss": 0.6273,
      "num_input_tokens_seen": 1025649664,
      "step": 13790,
      "train_runtime": 16360.1626,
      "train_tokens_per_second": 62691.899
    },
    {
      "epoch": 0.18338748579743655,
      "grad_norm": 20.875,
      "learning_rate": 1.879150107007452e-05,
      "loss": 0.5297,
      "num_input_tokens_seen": 1025762304,
      "step": 13800,
      "train_runtime": 16362.5418,
      "train_tokens_per_second": 62689.667
    },
    {
      "epoch": 0.18352037527989848,
      "grad_norm": 35.25,
      "learning_rate": 1.8789449223417368e-05,
      "loss": 0.826,
      "num_input_tokens_seen": 1027857664,
      "step": 13810,
      "train_runtime": 16396.4534,
      "train_tokens_per_second": 62687.804
    },
    {
      "epoch": 0.18365326476236038,
      "grad_norm": 14.3125,
      "learning_rate": 1.878739574859312e-05,
      "loss": 0.8449,
      "num_input_tokens_seen": 1028715008,
      "step": 13820,
      "train_runtime": 16409.0687,
      "train_tokens_per_second": 62691.858
    },
    {
      "epoch": 0.18378615424482228,
      "grad_norm": 41.75,
      "learning_rate": 1.878534064598217e-05,
      "loss": 0.7759,
      "num_input_tokens_seen": 1029164800,
      "step": 13830,
      "train_runtime": 16415.5345,
      "train_tokens_per_second": 62694.565
    },
    {
      "epoch": 0.1839190437272842,
      "grad_norm": 15.4375,
      "learning_rate": 1.87832839159652e-05,
      "loss": 0.6605,
      "num_input_tokens_seen": 1029403136,
      "step": 13840,
      "train_runtime": 16419.0379,
      "train_tokens_per_second": 62695.704
    },
    {
      "epoch": 0.1840519332097461,
      "grad_norm": 13.9375,
      "learning_rate": 1.8781225558923205e-05,
      "loss": 0.5051,
      "num_input_tokens_seen": 1029518848,
      "step": 13850,
      "train_runtime": 16421.4234,
      "train_tokens_per_second": 62693.643
    },
    {
      "epoch": 0.18418482269220804,
      "grad_norm": 17.875,
      "learning_rate": 1.877916557523747e-05,
      "loss": 0.7865,
      "num_input_tokens_seen": 1031586048,
      "step": 13860,
      "train_runtime": 16455.2722,
      "train_tokens_per_second": 62690.306
    },
    {
      "epoch": 0.18431771217466994,
      "grad_norm": 10.8125,
      "learning_rate": 1.8777103965289604e-05,
      "loss": 0.9771,
      "num_input_tokens_seen": 1032431616,
      "step": 13870,
      "train_runtime": 16468.1105,
      "train_tokens_per_second": 62692.779
    },
    {
      "epoch": 0.18445060165713184,
      "grad_norm": 17.75,
      "learning_rate": 1.8775040729461486e-05,
      "loss": 0.7593,
      "num_input_tokens_seen": 1032905728,
      "step": 13880,
      "train_runtime": 16474.6792,
      "train_tokens_per_second": 62696.561
    },
    {
      "epoch": 0.18458349113959377,
      "grad_norm": 25.125,
      "learning_rate": 1.8772975868135323e-05,
      "loss": 0.7233,
      "num_input_tokens_seen": 1033128192,
      "step": 13890,
      "train_runtime": 16478.027,
      "train_tokens_per_second": 62697.324
    },
    {
      "epoch": 0.18471638062205567,
      "grad_norm": 20.0,
      "learning_rate": 1.87709093816936e-05,
      "loss": 0.6996,
      "num_input_tokens_seen": 1033241344,
      "step": 13900,
      "train_runtime": 16480.885,
      "train_tokens_per_second": 62693.317
    },
    {
      "epoch": 0.18484927010451757,
      "grad_norm": 24.375,
      "learning_rate": 1.8768841270519133e-05,
      "loss": 0.8396,
      "num_input_tokens_seen": 1035425024,
      "step": 13910,
      "train_runtime": 16516.3876,
      "train_tokens_per_second": 62690.768
    },
    {
      "epoch": 0.1849821595869795,
      "grad_norm": 17.875,
      "learning_rate": 1.8766771534995004e-05,
      "loss": 0.8842,
      "num_input_tokens_seen": 1036289024,
      "step": 13920,
      "train_runtime": 16529.2693,
      "train_tokens_per_second": 62694.182
    },
    {
      "epoch": 0.1851150490694414,
      "grad_norm": 23.0,
      "learning_rate": 1.876470017550462e-05,
      "loss": 0.7151,
      "num_input_tokens_seen": 1036769536,
      "step": 13930,
      "train_runtime": 16536.0038,
      "train_tokens_per_second": 62697.708
    },
    {
      "epoch": 0.1852479385519033,
      "grad_norm": 16.875,
      "learning_rate": 1.8762627192431685e-05,
      "loss": 0.6708,
      "num_input_tokens_seen": 1036993536,
      "step": 13940,
      "train_runtime": 16539.4407,
      "train_tokens_per_second": 62698.223
    },
    {
      "epoch": 0.18538082803436523,
      "grad_norm": 9.875,
      "learning_rate": 1.8760552586160196e-05,
      "loss": 0.5245,
      "num_input_tokens_seen": 1037107456,
      "step": 13950,
      "train_runtime": 16541.8025,
      "train_tokens_per_second": 62696.158
    },
    {
      "epoch": 0.18551371751682713,
      "grad_norm": 13.25,
      "learning_rate": 1.8758476357074455e-05,
      "loss": 0.8828,
      "num_input_tokens_seen": 1039206144,
      "step": 13960,
      "train_runtime": 16573.8331,
      "train_tokens_per_second": 62701.618
    },
    {
      "epoch": 0.18564660699928903,
      "grad_norm": 26.0,
      "learning_rate": 1.875639850555907e-05,
      "loss": 0.9003,
      "num_input_tokens_seen": 1040043776,
      "step": 13970,
      "train_runtime": 16586.8276,
      "train_tokens_per_second": 62702.995
    },
    {
      "epoch": 0.18577949648175096,
      "grad_norm": 15.3125,
      "learning_rate": 1.875431903199894e-05,
      "loss": 0.7454,
      "num_input_tokens_seen": 1040512000,
      "step": 13980,
      "train_runtime": 16593.489,
      "train_tokens_per_second": 62706.041
    },
    {
      "epoch": 0.18591238596421286,
      "grad_norm": 15.0,
      "learning_rate": 1.8752237936779267e-05,
      "loss": 0.6649,
      "num_input_tokens_seen": 1040749824,
      "step": 13990,
      "train_runtime": 16596.9878,
      "train_tokens_per_second": 62707.151
    },
    {
      "epoch": 0.18604527544667476,
      "grad_norm": 15.8125,
      "learning_rate": 1.8750155220285564e-05,
      "loss": 0.5484,
      "num_input_tokens_seen": 1040866816,
      "step": 14000,
      "train_runtime": 16599.3542,
      "train_tokens_per_second": 62705.26
    },
    {
      "epoch": 0.1861781649291367,
      "grad_norm": 14.4375,
      "learning_rate": 1.8748070882903626e-05,
      "loss": 0.8189,
      "num_input_tokens_seen": 1042835456,
      "step": 14010,
      "train_runtime": 16631.1699,
      "train_tokens_per_second": 62703.674
    },
    {
      "epoch": 0.1863110544115986,
      "grad_norm": 22.625,
      "learning_rate": 1.8745984925019565e-05,
      "loss": 0.907,
      "num_input_tokens_seen": 1043652096,
      "step": 14020,
      "train_runtime": 16643.6475,
      "train_tokens_per_second": 62705.732
    },
    {
      "epoch": 0.1864439438940605,
      "grad_norm": 21.25,
      "learning_rate": 1.874389734701978e-05,
      "loss": 0.714,
      "num_input_tokens_seen": 1044119296,
      "step": 14030,
      "train_runtime": 16650.0643,
      "train_tokens_per_second": 62709.625
    },
    {
      "epoch": 0.18657683337652242,
      "grad_norm": 34.25,
      "learning_rate": 1.8741808149290983e-05,
      "loss": 0.6025,
      "num_input_tokens_seen": 1044346368,
      "step": 14040,
      "train_runtime": 16653.4675,
      "train_tokens_per_second": 62710.445
    },
    {
      "epoch": 0.18670972285898432,
      "grad_norm": 20.5,
      "learning_rate": 1.8739717332220172e-05,
      "loss": 0.6056,
      "num_input_tokens_seen": 1044464128,
      "step": 14050,
      "train_runtime": 16655.8317,
      "train_tokens_per_second": 62708.614
    },
    {
      "epoch": 0.18684261234144622,
      "grad_norm": 13.25,
      "learning_rate": 1.873762489619466e-05,
      "loss": 0.8372,
      "num_input_tokens_seen": 1046448896,
      "step": 14060,
      "train_runtime": 16688.7515,
      "train_tokens_per_second": 62703.846
    },
    {
      "epoch": 0.18697550182390815,
      "grad_norm": 25.0,
      "learning_rate": 1.8735530841602042e-05,
      "loss": 0.9922,
      "num_input_tokens_seen": 1047238400,
      "step": 14070,
      "train_runtime": 16700.8586,
      "train_tokens_per_second": 62705.662
    },
    {
      "epoch": 0.18710839130637005,
      "grad_norm": 12.6875,
      "learning_rate": 1.8733435168830233e-05,
      "loss": 0.7068,
      "num_input_tokens_seen": 1047670784,
      "step": 14080,
      "train_runtime": 16707.587,
      "train_tokens_per_second": 62706.289
    },
    {
      "epoch": 0.18724128078883195,
      "grad_norm": 30.25,
      "learning_rate": 1.8731337878267432e-05,
      "loss": 0.6014,
      "num_input_tokens_seen": 1047882752,
      "step": 14090,
      "train_runtime": 16710.8506,
      "train_tokens_per_second": 62706.727
    },
    {
      "epoch": 0.18737417027129388,
      "grad_norm": 13.6875,
      "learning_rate": 1.8729238970302148e-05,
      "loss": 0.4753,
      "num_input_tokens_seen": 1047995136,
      "step": 14100,
      "train_runtime": 16713.2128,
      "train_tokens_per_second": 62704.589
    },
    {
      "epoch": 0.18750705975375578,
      "grad_norm": 21.25,
      "learning_rate": 1.8727138445323177e-05,
      "loss": 0.9076,
      "num_input_tokens_seen": 1050014464,
      "step": 14110,
      "train_runtime": 16745.8839,
      "train_tokens_per_second": 62702.839
    },
    {
      "epoch": 0.1876399492362177,
      "grad_norm": 27.25,
      "learning_rate": 1.872503630371963e-05,
      "loss": 0.9088,
      "num_input_tokens_seen": 1050860800,
      "step": 14120,
      "train_runtime": 16758.5256,
      "train_tokens_per_second": 62706.041
    },
    {
      "epoch": 0.1877728387186796,
      "grad_norm": 26.75,
      "learning_rate": 1.8722932545880907e-05,
      "loss": 0.8225,
      "num_input_tokens_seen": 1051320576,
      "step": 14130,
      "train_runtime": 16765.2277,
      "train_tokens_per_second": 62708.398
    },
    {
      "epoch": 0.18790572820114151,
      "grad_norm": 27.25,
      "learning_rate": 1.8720827172196715e-05,
      "loss": 0.6363,
      "num_input_tokens_seen": 1051562496,
      "step": 14140,
      "train_runtime": 16769.2524,
      "train_tokens_per_second": 62707.774
    },
    {
      "epoch": 0.18803861768360344,
      "grad_norm": 11.75,
      "learning_rate": 1.8718720183057046e-05,
      "loss": 0.5348,
      "num_input_tokens_seen": 1051679744,
      "step": 14150,
      "train_runtime": 16771.6518,
      "train_tokens_per_second": 62705.794
    },
    {
      "epoch": 0.18817150716606534,
      "grad_norm": 25.125,
      "learning_rate": 1.871661157885221e-05,
      "loss": 0.8511,
      "num_input_tokens_seen": 1053711872,
      "step": 14160,
      "train_runtime": 16804.6883,
      "train_tokens_per_second": 62703.446
    },
    {
      "epoch": 0.18830439664852724,
      "grad_norm": 11.375,
      "learning_rate": 1.8714501359972802e-05,
      "loss": 0.8836,
      "num_input_tokens_seen": 1054510080,
      "step": 14170,
      "train_runtime": 16816.89,
      "train_tokens_per_second": 62705.416
    },
    {
      "epoch": 0.18843728613098917,
      "grad_norm": 23.625,
      "learning_rate": 1.8712389526809723e-05,
      "loss": 0.7204,
      "num_input_tokens_seen": 1054959104,
      "step": 14180,
      "train_runtime": 16823.2115,
      "train_tokens_per_second": 62708.544
    },
    {
      "epoch": 0.18857017561345107,
      "grad_norm": 13.9375,
      "learning_rate": 1.871027607975417e-05,
      "loss": 0.6624,
      "num_input_tokens_seen": 1055191552,
      "step": 14190,
      "train_runtime": 16827.0148,
      "train_tokens_per_second": 62708.185
    },
    {
      "epoch": 0.18870306509591298,
      "grad_norm": 13.625,
      "learning_rate": 1.8708161019197644e-05,
      "loss": 0.5258,
      "num_input_tokens_seen": 1055309056,
      "step": 14200,
      "train_runtime": 16829.3964,
      "train_tokens_per_second": 62706.293
    },
    {
      "epoch": 0.1888359545783749,
      "grad_norm": 16.375,
      "learning_rate": 1.8706044345531936e-05,
      "loss": 0.8782,
      "num_input_tokens_seen": 1057303040,
      "step": 14210,
      "train_runtime": 16862.7739,
      "train_tokens_per_second": 62700.422
    },
    {
      "epoch": 0.1889688440608368,
      "grad_norm": 26.25,
      "learning_rate": 1.870392605914914e-05,
      "loss": 0.9986,
      "num_input_tokens_seen": 1058159616,
      "step": 14220,
      "train_runtime": 16875.5268,
      "train_tokens_per_second": 62703.797
    },
    {
      "epoch": 0.1891017335432987,
      "grad_norm": 15.5625,
      "learning_rate": 1.870180616044166e-05,
      "loss": 0.7967,
      "num_input_tokens_seen": 1058629120,
      "step": 14230,
      "train_runtime": 16882.1835,
      "train_tokens_per_second": 62706.884
    },
    {
      "epoch": 0.18923462302576063,
      "grad_norm": 9.125,
      "learning_rate": 1.869968464980218e-05,
      "loss": 0.6143,
      "num_input_tokens_seen": 1058855168,
      "step": 14240,
      "train_runtime": 16885.9841,
      "train_tokens_per_second": 62706.157
    },
    {
      "epoch": 0.18936751250822254,
      "grad_norm": 6.8125,
      "learning_rate": 1.869756152762369e-05,
      "loss": 0.5541,
      "num_input_tokens_seen": 1058971904,
      "step": 14250,
      "train_runtime": 16888.393,
      "train_tokens_per_second": 62704.125
    },
    {
      "epoch": 0.18950040199068444,
      "grad_norm": 11.625,
      "learning_rate": 1.869543679429948e-05,
      "loss": 0.8289,
      "num_input_tokens_seen": 1061100544,
      "step": 14260,
      "train_runtime": 16922.6243,
      "train_tokens_per_second": 62703.073
    },
    {
      "epoch": 0.18963329147314636,
      "grad_norm": 16.75,
      "learning_rate": 1.8693310450223145e-05,
      "loss": 0.922,
      "num_input_tokens_seen": 1061948160,
      "step": 14270,
      "train_runtime": 16934.9955,
      "train_tokens_per_second": 62707.319
    },
    {
      "epoch": 0.18976618095560827,
      "grad_norm": 22.75,
      "learning_rate": 1.8691182495788563e-05,
      "loss": 0.767,
      "num_input_tokens_seen": 1062390272,
      "step": 14280,
      "train_runtime": 16941.3203,
      "train_tokens_per_second": 62710.005
    },
    {
      "epoch": 0.18989907043807017,
      "grad_norm": 35.5,
      "learning_rate": 1.8689052931389923e-05,
      "loss": 0.6993,
      "num_input_tokens_seen": 1062612992,
      "step": 14290,
      "train_runtime": 16944.6722,
      "train_tokens_per_second": 62710.743
    },
    {
      "epoch": 0.1900319599205321,
      "grad_norm": 5.09375,
      "learning_rate": 1.8686921757421703e-05,
      "loss": 0.4445,
      "num_input_tokens_seen": 1062729216,
      "step": 14300,
      "train_runtime": 16947.4189,
      "train_tokens_per_second": 62707.438
    },
    {
      "epoch": 0.190164849402994,
      "grad_norm": 31.0,
      "learning_rate": 1.868478897427869e-05,
      "loss": 0.9167,
      "num_input_tokens_seen": 1064731392,
      "step": 14310,
      "train_runtime": 16977.6794,
      "train_tokens_per_second": 62713.6
    },
    {
      "epoch": 0.1902977388854559,
      "grad_norm": 15.5625,
      "learning_rate": 1.868265458235596e-05,
      "loss": 0.9252,
      "num_input_tokens_seen": 1065523712,
      "step": 14320,
      "train_runtime": 16989.5782,
      "train_tokens_per_second": 62716.313
    },
    {
      "epoch": 0.19043062836791783,
      "grad_norm": 16.5,
      "learning_rate": 1.868051858204889e-05,
      "loss": 0.7458,
      "num_input_tokens_seen": 1065947648,
      "step": 14330,
      "train_runtime": 16995.7018,
      "train_tokens_per_second": 62718.66
    },
    {
      "epoch": 0.19056351785037973,
      "grad_norm": 12.3125,
      "learning_rate": 1.8678380973753155e-05,
      "loss": 0.6654,
      "num_input_tokens_seen": 1066167040,
      "step": 14340,
      "train_runtime": 16999.0499,
      "train_tokens_per_second": 62719.213
    },
    {
      "epoch": 0.19069640733284166,
      "grad_norm": 32.25,
      "learning_rate": 1.867624175786473e-05,
      "loss": 0.5194,
      "num_input_tokens_seen": 1066278656,
      "step": 14350,
      "train_runtime": 17001.7858,
      "train_tokens_per_second": 62715.686
    },
    {
      "epoch": 0.19082929681530356,
      "grad_norm": 20.875,
      "learning_rate": 1.8674100934779883e-05,
      "loss": 0.8352,
      "num_input_tokens_seen": 1068331264,
      "step": 14360,
      "train_runtime": 17034.0645,
      "train_tokens_per_second": 62717.343
    },
    {
      "epoch": 0.19096218629776546,
      "grad_norm": 16.0,
      "learning_rate": 1.8671958504895183e-05,
      "loss": 0.885,
      "num_input_tokens_seen": 1069160960,
      "step": 14370,
      "train_runtime": 17046.4149,
      "train_tokens_per_second": 62720.576
    },
    {
      "epoch": 0.19109507578022739,
      "grad_norm": 14.0,
      "learning_rate": 1.8669814468607494e-05,
      "loss": 0.7484,
      "num_input_tokens_seen": 1069614848,
      "step": 14380,
      "train_runtime": 17052.9454,
      "train_tokens_per_second": 62723.173
    },
    {
      "epoch": 0.1912279652626893,
      "grad_norm": 10.375,
      "learning_rate": 1.8667668826313986e-05,
      "loss": 0.7054,
      "num_input_tokens_seen": 1069838848,
      "step": 14390,
      "train_runtime": 17056.349,
      "train_tokens_per_second": 62723.79
    },
    {
      "epoch": 0.1913608547451512,
      "grad_norm": 12.25,
      "learning_rate": 1.8665521578412115e-05,
      "loss": 0.5277,
      "num_input_tokens_seen": 1069953024,
      "step": 14400,
      "train_runtime": 17058.7504,
      "train_tokens_per_second": 62721.653
    },
    {
      "epoch": 0.19149374422761312,
      "grad_norm": 43.0,
      "learning_rate": 1.8663372725299637e-05,
      "loss": 0.8222,
      "num_input_tokens_seen": 1072045824,
      "step": 14410,
      "train_runtime": 17091.8388,
      "train_tokens_per_second": 62722.674
    },
    {
      "epoch": 0.19162663371007502,
      "grad_norm": 17.625,
      "learning_rate": 1.8661222267374617e-05,
      "loss": 0.9223,
      "num_input_tokens_seen": 1072875520,
      "step": 14420,
      "train_runtime": 17104.3781,
      "train_tokens_per_second": 62725.199
    },
    {
      "epoch": 0.19175952319253692,
      "grad_norm": 76.0,
      "learning_rate": 1.8659070205035398e-05,
      "loss": 0.8337,
      "num_input_tokens_seen": 1073338880,
      "step": 14430,
      "train_runtime": 17111.0803,
      "train_tokens_per_second": 62727.71
    },
    {
      "epoch": 0.19189241267499885,
      "grad_norm": 38.5,
      "learning_rate": 1.865691653868064e-05,
      "loss": 0.6336,
      "num_input_tokens_seen": 1073564928,
      "step": 14440,
      "train_runtime": 17114.4691,
      "train_tokens_per_second": 62728.497
    },
    {
      "epoch": 0.19202530215746075,
      "grad_norm": 16.0,
      "learning_rate": 1.8654761268709283e-05,
      "loss": 0.4918,
      "num_input_tokens_seen": 1073680128,
      "step": 14450,
      "train_runtime": 17116.8146,
      "train_tokens_per_second": 62726.632
    },
    {
      "epoch": 0.19215819163992265,
      "grad_norm": 38.0,
      "learning_rate": 1.8652604395520574e-05,
      "loss": 0.8325,
      "num_input_tokens_seen": 1075758848,
      "step": 14460,
      "train_runtime": 17149.8954,
      "train_tokens_per_second": 62726.846
    },
    {
      "epoch": 0.19229108112238458,
      "grad_norm": 26.625,
      "learning_rate": 1.8650445919514055e-05,
      "loss": 0.9452,
      "num_input_tokens_seen": 1076583168,
      "step": 14470,
      "train_runtime": 17162.4151,
      "train_tokens_per_second": 62729.118
    },
    {
      "epoch": 0.19242397060484648,
      "grad_norm": 14.5,
      "learning_rate": 1.8648285841089562e-05,
      "loss": 0.7292,
      "num_input_tokens_seen": 1077069568,
      "step": 14480,
      "train_runtime": 17168.9547,
      "train_tokens_per_second": 62733.555
    },
    {
      "epoch": 0.19255686008730838,
      "grad_norm": 18.125,
      "learning_rate": 1.8646124160647235e-05,
      "loss": 0.603,
      "num_input_tokens_seen": 1077295616,
      "step": 14490,
      "train_runtime": 17172.3523,
      "train_tokens_per_second": 62734.307
    },
    {
      "epoch": 0.1926897495697703,
      "grad_norm": 13.875,
      "learning_rate": 1.86439608785875e-05,
      "loss": 0.5208,
      "num_input_tokens_seen": 1077410560,
      "step": 14500,
      "train_runtime": 17174.6885,
      "train_tokens_per_second": 62732.466
    },
    {
      "epoch": 0.1928226390522322,
      "grad_norm": 31.375,
      "learning_rate": 1.864179599531109e-05,
      "loss": 0.8694,
      "num_input_tokens_seen": 1079382272,
      "step": 14510,
      "train_runtime": 17205.6724,
      "train_tokens_per_second": 62734.094
    },
    {
      "epoch": 0.1929555285346941,
      "grad_norm": 21.875,
      "learning_rate": 1.8639629511219034e-05,
      "loss": 0.8971,
      "num_input_tokens_seen": 1080225280,
      "step": 14520,
      "train_runtime": 17217.8609,
      "train_tokens_per_second": 62738.646
    },
    {
      "epoch": 0.19308841801715604,
      "grad_norm": 13.75,
      "learning_rate": 1.8637461426712643e-05,
      "loss": 0.7713,
      "num_input_tokens_seen": 1080678656,
      "step": 14530,
      "train_runtime": 17224.7648,
      "train_tokens_per_second": 62739.821
    },
    {
      "epoch": 0.19322130749961794,
      "grad_norm": 18.75,
      "learning_rate": 1.8635291742193545e-05,
      "loss": 0.6477,
      "num_input_tokens_seen": 1080908800,
      "step": 14540,
      "train_runtime": 17228.2007,
      "train_tokens_per_second": 62740.667
    },
    {
      "epoch": 0.19335419698207984,
      "grad_norm": 11.5,
      "learning_rate": 1.8633120458063652e-05,
      "loss": 0.4972,
      "num_input_tokens_seen": 1081020928,
      "step": 14550,
      "train_runtime": 17230.5828,
      "train_tokens_per_second": 62738.501
    },
    {
      "epoch": 0.19348708646454177,
      "grad_norm": 28.0,
      "learning_rate": 1.863094757472517e-05,
      "loss": 0.8795,
      "num_input_tokens_seen": 1083060736,
      "step": 14560,
      "train_runtime": 17264.0437,
      "train_tokens_per_second": 62735.055
    },
    {
      "epoch": 0.19361997594700367,
      "grad_norm": 37.0,
      "learning_rate": 1.8628773092580616e-05,
      "loss": 0.9652,
      "num_input_tokens_seen": 1083903488,
      "step": 14570,
      "train_runtime": 17276.6525,
      "train_tokens_per_second": 62738.05
    },
    {
      "epoch": 0.1937528654294656,
      "grad_norm": 14.8125,
      "learning_rate": 1.8626597012032785e-05,
      "loss": 0.7457,
      "num_input_tokens_seen": 1084365056,
      "step": 14580,
      "train_runtime": 17283.2438,
      "train_tokens_per_second": 62740.83
    },
    {
      "epoch": 0.1938857549119275,
      "grad_norm": 22.0,
      "learning_rate": 1.862441933348478e-05,
      "loss": 0.6566,
      "num_input_tokens_seen": 1084592896,
      "step": 14590,
      "train_runtime": 17286.7025,
      "train_tokens_per_second": 62741.457
    },
    {
      "epoch": 0.1940186443943894,
      "grad_norm": 13.75,
      "learning_rate": 1.8622240057339996e-05,
      "loss": 0.5404,
      "num_input_tokens_seen": 1084707584,
      "step": 14600,
      "train_runtime": 17289.5869,
      "train_tokens_per_second": 62737.623
    },
    {
      "epoch": 0.19415153387685133,
      "grad_norm": 520.0,
      "learning_rate": 1.8620059184002125e-05,
      "loss": 0.8625,
      "num_input_tokens_seen": 1086840832,
      "step": 14610,
      "train_runtime": 17322.0435,
      "train_tokens_per_second": 62743.223
    },
    {
      "epoch": 0.19428442335931323,
      "grad_norm": 19.75,
      "learning_rate": 1.861787671387515e-05,
      "loss": 0.8953,
      "num_input_tokens_seen": 1087764480,
      "step": 14620,
      "train_runtime": 17335.4578,
      "train_tokens_per_second": 62747.952
    },
    {
      "epoch": 0.19441731284177513,
      "grad_norm": 19.625,
      "learning_rate": 1.861569264736336e-05,
      "loss": 0.757,
      "num_input_tokens_seen": 1088268288,
      "step": 14630,
      "train_runtime": 17342.71,
      "train_tokens_per_second": 62750.763
    },
    {
      "epoch": 0.19455020232423706,
      "grad_norm": 21.875,
      "learning_rate": 1.8613506984871334e-05,
      "loss": 0.6897,
      "num_input_tokens_seen": 1088505856,
      "step": 14640,
      "train_runtime": 17346.7778,
      "train_tokens_per_second": 62749.743
    },
    {
      "epoch": 0.19468309180669896,
      "grad_norm": 9.5,
      "learning_rate": 1.8611319726803943e-05,
      "loss": 0.5212,
      "num_input_tokens_seen": 1088622336,
      "step": 14650,
      "train_runtime": 17349.1618,
      "train_tokens_per_second": 62747.835
    },
    {
      "epoch": 0.19481598128916086,
      "grad_norm": 29.5,
      "learning_rate": 1.860913087356636e-05,
      "loss": 0.9261,
      "num_input_tokens_seen": 1090790912,
      "step": 14660,
      "train_runtime": 17384.3295,
      "train_tokens_per_second": 62745.642
    },
    {
      "epoch": 0.1949488707716228,
      "grad_norm": 15.25,
      "learning_rate": 1.8606940425564043e-05,
      "loss": 0.944,
      "num_input_tokens_seen": 1091670528,
      "step": 14670,
      "train_runtime": 17397.1798,
      "train_tokens_per_second": 62749.856
    },
    {
      "epoch": 0.1950817602540847,
      "grad_norm": 18.0,
      "learning_rate": 1.8604748383202763e-05,
      "loss": 0.7583,
      "num_input_tokens_seen": 1092142848,
      "step": 14680,
      "train_runtime": 17404.0578,
      "train_tokens_per_second": 62752.196
    },
    {
      "epoch": 0.1952146497365466,
      "grad_norm": 16.5,
      "learning_rate": 1.8602554746888572e-05,
      "loss": 0.6515,
      "num_input_tokens_seen": 1092387584,
      "step": 14690,
      "train_runtime": 17407.6154,
      "train_tokens_per_second": 62753.431
    },
    {
      "epoch": 0.19534753921900852,
      "grad_norm": 15.1875,
      "learning_rate": 1.8600359517027824e-05,
      "loss": 0.4832,
      "num_input_tokens_seen": 1092506880,
      "step": 14700,
      "train_runtime": 17410.0351,
      "train_tokens_per_second": 62751.561
    },
    {
      "epoch": 0.19548042870147042,
      "grad_norm": 29.625,
      "learning_rate": 1.8598162694027162e-05,
      "loss": 0.8773,
      "num_input_tokens_seen": 1094519552,
      "step": 14710,
      "train_runtime": 17443.2242,
      "train_tokens_per_second": 62747.548
    },
    {
      "epoch": 0.19561331818393232,
      "grad_norm": 17.25,
      "learning_rate": 1.8595964278293526e-05,
      "loss": 0.8546,
      "num_input_tokens_seen": 1095402752,
      "step": 14720,
      "train_runtime": 17456.0501,
      "train_tokens_per_second": 62752.04
    },
    {
      "epoch": 0.19574620766639425,
      "grad_norm": 20.125,
      "learning_rate": 1.859376427023416e-05,
      "loss": 0.8005,
      "num_input_tokens_seen": 1095886592,
      "step": 14730,
      "train_runtime": 17462.9903,
      "train_tokens_per_second": 62754.807
    },
    {
      "epoch": 0.19587909714885615,
      "grad_norm": 14.5,
      "learning_rate": 1.859156267025659e-05,
      "loss": 0.6733,
      "num_input_tokens_seen": 1096129280,
      "step": 14740,
      "train_runtime": 17466.5897,
      "train_tokens_per_second": 62755.77
    },
    {
      "epoch": 0.19601198663131805,
      "grad_norm": 5.96875,
      "learning_rate": 1.858935947876865e-05,
      "loss": 0.5323,
      "num_input_tokens_seen": 1096243712,
      "step": 14750,
      "train_runtime": 17469.031,
      "train_tokens_per_second": 62753.55
    },
    {
      "epoch": 0.19614487611377998,
      "grad_norm": 23.375,
      "learning_rate": 1.8587154696178447e-05,
      "loss": 0.9012,
      "num_input_tokens_seen": 1098329600,
      "step": 14760,
      "train_runtime": 17502.6778,
      "train_tokens_per_second": 62752.089
    },
    {
      "epoch": 0.19627776559624188,
      "grad_norm": 64.5,
      "learning_rate": 1.8584948322894415e-05,
      "loss": 0.9294,
      "num_input_tokens_seen": 1099204352,
      "step": 14770,
      "train_runtime": 17515.5762,
      "train_tokens_per_second": 62755.82
    },
    {
      "epoch": 0.19641065507870378,
      "grad_norm": 22.5,
      "learning_rate": 1.8582740359325254e-05,
      "loss": 0.7742,
      "num_input_tokens_seen": 1099709952,
      "step": 14780,
      "train_runtime": 17522.4092,
      "train_tokens_per_second": 62760.202
    },
    {
      "epoch": 0.1965435445611657,
      "grad_norm": 19.25,
      "learning_rate": 1.858053080587997e-05,
      "loss": 0.6861,
      "num_input_tokens_seen": 1099941376,
      "step": 14790,
      "train_runtime": 17525.8887,
      "train_tokens_per_second": 62760.947
    },
    {
      "epoch": 0.1966764340436276,
      "grad_norm": 13.5625,
      "learning_rate": 1.8578319662967873e-05,
      "loss": 0.5469,
      "num_input_tokens_seen": 1100054784,
      "step": 14800,
      "train_runtime": 17528.2831,
      "train_tokens_per_second": 62758.844
    },
    {
      "epoch": 0.19680932352608954,
      "grad_norm": 18.875,
      "learning_rate": 1.857610693099854e-05,
      "loss": 0.8387,
      "num_input_tokens_seen": 1102089728,
      "step": 14810,
      "train_runtime": 17562.1672,
      "train_tokens_per_second": 62753.629
    },
    {
      "epoch": 0.19694221300855144,
      "grad_norm": 14.125,
      "learning_rate": 1.8573892610381876e-05,
      "loss": 0.9458,
      "num_input_tokens_seen": 1102930432,
      "step": 14820,
      "train_runtime": 17575.1919,
      "train_tokens_per_second": 62754.958
    },
    {
      "epoch": 0.19707510249101334,
      "grad_norm": 20.375,
      "learning_rate": 1.8571676701528054e-05,
      "loss": 0.7279,
      "num_input_tokens_seen": 1103364608,
      "step": 14830,
      "train_runtime": 17581.5944,
      "train_tokens_per_second": 62756.8
    },
    {
      "epoch": 0.19720799197347527,
      "grad_norm": 12.125,
      "learning_rate": 1.856945920484756e-05,
      "loss": 0.6617,
      "num_input_tokens_seen": 1103583488,
      "step": 14840,
      "train_runtime": 17584.976,
      "train_tokens_per_second": 62757.179
    },
    {
      "epoch": 0.19734088145593717,
      "grad_norm": 11.5625,
      "learning_rate": 1.8567240120751157e-05,
      "loss": 0.5933,
      "num_input_tokens_seen": 1103700224,
      "step": 14850,
      "train_runtime": 17587.3155,
      "train_tokens_per_second": 62755.469
    },
    {
      "epoch": 0.19747377093839907,
      "grad_norm": 30.5,
      "learning_rate": 1.8565019449649915e-05,
      "loss": 0.8906,
      "num_input_tokens_seen": 1105846016,
      "step": 14860,
      "train_runtime": 17621.2495,
      "train_tokens_per_second": 62756.391
    },
    {
      "epoch": 0.197606660420861,
      "grad_norm": 11.25,
      "learning_rate": 1.8562797191955196e-05,
      "loss": 0.9393,
      "num_input_tokens_seen": 1106674432,
      "step": 14870,
      "train_runtime": 17633.9626,
      "train_tokens_per_second": 62758.125
    },
    {
      "epoch": 0.1977395499033229,
      "grad_norm": 13.4375,
      "learning_rate": 1.8560573348078645e-05,
      "loss": 0.7019,
      "num_input_tokens_seen": 1107169024,
      "step": 14880,
      "train_runtime": 17640.7901,
      "train_tokens_per_second": 62761.873
    },
    {
      "epoch": 0.1978724393857848,
      "grad_norm": 16.0,
      "learning_rate": 1.8558347918432217e-05,
      "loss": 0.7058,
      "num_input_tokens_seen": 1107399168,
      "step": 14890,
      "train_runtime": 17644.2423,
      "train_tokens_per_second": 62762.637
    },
    {
      "epoch": 0.19800532886824673,
      "grad_norm": 16.125,
      "learning_rate": 1.8556120903428148e-05,
      "loss": 0.5519,
      "num_input_tokens_seen": 1107511808,
      "step": 14900,
      "train_runtime": 17646.6124,
      "train_tokens_per_second": 62760.59
    },
    {
      "epoch": 0.19813821835070863,
      "grad_norm": 10.5625,
      "learning_rate": 1.855389230347898e-05,
      "loss": 0.814,
      "num_input_tokens_seen": 1109571072,
      "step": 14910,
      "train_runtime": 17678.8568,
      "train_tokens_per_second": 62762.603
    },
    {
      "epoch": 0.19827110783317053,
      "grad_norm": 14.0625,
      "learning_rate": 1.8551662118997534e-05,
      "loss": 0.9031,
      "num_input_tokens_seen": 1110426112,
      "step": 14920,
      "train_runtime": 17691.3843,
      "train_tokens_per_second": 62766.491
    },
    {
      "epoch": 0.19840399731563246,
      "grad_norm": 31.625,
      "learning_rate": 1.8549430350396935e-05,
      "loss": 0.7626,
      "num_input_tokens_seen": 1110896128,
      "step": 14930,
      "train_runtime": 17698.2149,
      "train_tokens_per_second": 62768.824
    },
    {
      "epoch": 0.19853688679809436,
      "grad_norm": 16.25,
      "learning_rate": 1.8547196998090597e-05,
      "loss": 0.654,
      "num_input_tokens_seen": 1111128064,
      "step": 14940,
      "train_runtime": 17702.2517,
      "train_tokens_per_second": 62767.612
    },
    {
      "epoch": 0.19866977628055627,
      "grad_norm": 12.0,
      "learning_rate": 1.854496206249223e-05,
      "loss": 0.5054,
      "num_input_tokens_seen": 1111244544,
      "step": 14950,
      "train_runtime": 17704.6112,
      "train_tokens_per_second": 62765.826
    },
    {
      "epoch": 0.1988026657630182,
      "grad_norm": 22.125,
      "learning_rate": 1.8542725544015834e-05,
      "loss": 0.8494,
      "num_input_tokens_seen": 1113282816,
      "step": 14960,
      "train_runtime": 17738.0966,
      "train_tokens_per_second": 62762.248
    },
    {
      "epoch": 0.1989355552454801,
      "grad_norm": 18.25,
      "learning_rate": 1.8540487443075705e-05,
      "loss": 0.958,
      "num_input_tokens_seen": 1114131456,
      "step": 14970,
      "train_runtime": 17750.7594,
      "train_tokens_per_second": 62765.284
    },
    {
      "epoch": 0.199068444727942,
      "grad_norm": 29.5,
      "learning_rate": 1.8538247760086428e-05,
      "loss": 0.8281,
      "num_input_tokens_seen": 1114613760,
      "step": 14980,
      "train_runtime": 17757.6424,
      "train_tokens_per_second": 62768.116
    },
    {
      "epoch": 0.19920133421040392,
      "grad_norm": 12.9375,
      "learning_rate": 1.8536006495462893e-05,
      "loss": 0.6361,
      "num_input_tokens_seen": 1114840832,
      "step": 14990,
      "train_runtime": 17761.1102,
      "train_tokens_per_second": 62768.645
    },
    {
      "epoch": 0.19933422369286583,
      "grad_norm": 10.125,
      "learning_rate": 1.8533763649620268e-05,
      "loss": 0.4846,
      "num_input_tokens_seen": 1114956800,
      "step": 15000,
      "train_runtime": 17763.9119,
      "train_tokens_per_second": 62765.274
    },
    {
      "epoch": 0.19933422369286583,
      "eval_accuracy": 0.8134,
      "eval_balanced_accuracy": 0.8090025977604953,
      "eval_invalid_f1": 0.8349840820657941,
      "eval_invalid_precision": 0.8227605437434646,
      "eval_invalid_recall": 0.8475763016157989,
      "eval_loss": 0.37445446848869324,
      "eval_macro_f1": 0.8101519581981065,
      "eval_mcc": 0.6207755340871028,
      "eval_runtime": 77.1003,
      "eval_samples_per_second": 129.701,
      "eval_steps_per_second": 32.425,
      "eval_valid_f1": 0.7853198343304187,
      "eval_valid_precision": 0.8007977475363679,
      "eval_valid_recall": 0.7704288939051919,
      "num_input_tokens_seen": 1114956800,
      "step": 15000
    },
    {
      "epoch": 0.19946711317532773,
      "grad_norm": 18.875,
      "learning_rate": 1.8531519222974016e-05,
      "loss": 0.8308,
      "num_input_tokens_seen": 1116992768,
      "step": 15010,
      "train_runtime": 17877.9508,
      "train_tokens_per_second": 62478.792
    },
    {
      "epoch": 0.19960000265778965,
      "grad_norm": 30.25,
      "learning_rate": 1.8529273215939908e-05,
      "loss": 0.9405,
      "num_input_tokens_seen": 1117816576,
      "step": 15020,
      "train_runtime": 17890.7106,
      "train_tokens_per_second": 62480.278
    },
    {
      "epoch": 0.19973289214025156,
      "grad_norm": 21.75,
      "learning_rate": 1.8527025628933987e-05,
      "loss": 0.6653,
      "num_input_tokens_seen": 1118305792,
      "step": 15030,
      "train_runtime": 17897.6168,
      "train_tokens_per_second": 62483.503
    },
    {
      "epoch": 0.19986578162271346,
      "grad_norm": 21.25,
      "learning_rate": 1.8524776462372606e-05,
      "loss": 0.7303,
      "num_input_tokens_seen": 1118550528,
      "step": 15040,
      "train_runtime": 17901.2024,
      "train_tokens_per_second": 62484.659
    },
    {
      "epoch": 0.19999867110517539,
      "grad_norm": 35.25,
      "learning_rate": 1.8522525716672397e-05,
      "loss": 0.6052,
      "num_input_tokens_seen": 1118669824,
      "step": 15050,
      "train_runtime": 17903.6062,
      "train_tokens_per_second": 62482.933
    },
    {
      "epoch": 0.20013156058763729,
      "grad_norm": 22.75,
      "learning_rate": 1.8520273392250294e-05,
      "loss": 0.8395,
      "num_input_tokens_seen": 1120675584,
      "step": 15060,
      "train_runtime": 17936.3611,
      "train_tokens_per_second": 62480.655
    },
    {
      "epoch": 0.20026445007009921,
      "grad_norm": 54.25,
      "learning_rate": 1.8518019489523518e-05,
      "loss": 0.9646,
      "num_input_tokens_seen": 1121530624,
      "step": 15070,
      "train_runtime": 17949.6296,
      "train_tokens_per_second": 62482.104
    },
    {
      "epoch": 0.20039733955256112,
      "grad_norm": 13.1875,
      "learning_rate": 1.8515764008909585e-05,
      "loss": 0.7605,
      "num_input_tokens_seen": 1121997056,
      "step": 15080,
      "train_runtime": 17956.213,
      "train_tokens_per_second": 62485.172
    },
    {
      "epoch": 0.20053022903502302,
      "grad_norm": 20.625,
      "learning_rate": 1.8513506950826303e-05,
      "loss": 0.6734,
      "num_input_tokens_seen": 1122216448,
      "step": 15090,
      "train_runtime": 17959.5648,
      "train_tokens_per_second": 62485.726
    },
    {
      "epoch": 0.20066311851748495,
      "grad_norm": 18.625,
      "learning_rate": 1.8511248315691773e-05,
      "loss": 0.6131,
      "num_input_tokens_seen": 1122331392,
      "step": 15100,
      "train_runtime": 17961.9307,
      "train_tokens_per_second": 62483.895
    },
    {
      "epoch": 0.20079600799994685,
      "grad_norm": 10.625,
      "learning_rate": 1.8508988103924386e-05,
      "loss": 0.8883,
      "num_input_tokens_seen": 1124390656,
      "step": 15110,
      "train_runtime": 17994.8104,
      "train_tokens_per_second": 62484.162
    },
    {
      "epoch": 0.20092889748240875,
      "grad_norm": 14.0625,
      "learning_rate": 1.8506726315942826e-05,
      "loss": 0.9294,
      "num_input_tokens_seen": 1125232128,
      "step": 15120,
      "train_runtime": 18007.1185,
      "train_tokens_per_second": 62488.184
    },
    {
      "epoch": 0.20106178696487068,
      "grad_norm": 13.3125,
      "learning_rate": 1.850446295216607e-05,
      "loss": 0.7329,
      "num_input_tokens_seen": 1125681408,
      "step": 15130,
      "train_runtime": 18014.0773,
      "train_tokens_per_second": 62488.985
    },
    {
      "epoch": 0.20119467644733258,
      "grad_norm": 19.25,
      "learning_rate": 1.8502198013013382e-05,
      "loss": 0.7019,
      "num_input_tokens_seen": 1125919488,
      "step": 15140,
      "train_runtime": 18017.6017,
      "train_tokens_per_second": 62489.975
    },
    {
      "epoch": 0.20132756592979448,
      "grad_norm": 14.5625,
      "learning_rate": 1.8499931498904327e-05,
      "loss": 0.634,
      "num_input_tokens_seen": 1126038784,
      "step": 15150,
      "train_runtime": 18019.985,
      "train_tokens_per_second": 62488.331
    },
    {
      "epoch": 0.2014604554122564,
      "grad_norm": 18.875,
      "learning_rate": 1.8497663410258754e-05,
      "loss": 0.868,
      "num_input_tokens_seen": 1128134144,
      "step": 15160,
      "train_runtime": 18052.7354,
      "train_tokens_per_second": 62491.036
    },
    {
      "epoch": 0.2015933448947183,
      "grad_norm": 18.375,
      "learning_rate": 1.8495393747496805e-05,
      "loss": 0.8811,
      "num_input_tokens_seen": 1128963840,
      "step": 15170,
      "train_runtime": 18065.6738,
      "train_tokens_per_second": 62492.208
    },
    {
      "epoch": 0.2017262343771802,
      "grad_norm": 38.25,
      "learning_rate": 1.8493122511038915e-05,
      "loss": 0.7997,
      "num_input_tokens_seen": 1129413120,
      "step": 15180,
      "train_runtime": 18072.191,
      "train_tokens_per_second": 62494.532
    },
    {
      "epoch": 0.20185912385964214,
      "grad_norm": 18.125,
      "learning_rate": 1.8490849701305818e-05,
      "loss": 0.6862,
      "num_input_tokens_seen": 1129637120,
      "step": 15190,
      "train_runtime": 18075.6135,
      "train_tokens_per_second": 62495.092
    },
    {
      "epoch": 0.20199201334210404,
      "grad_norm": 15.0,
      "learning_rate": 1.848857531871852e-05,
      "loss": 0.4663,
      "num_input_tokens_seen": 1129753088,
      "step": 15200,
      "train_runtime": 18078.0034,
      "train_tokens_per_second": 62493.245
    },
    {
      "epoch": 0.20212490282456594,
      "grad_norm": 21.25,
      "learning_rate": 1.8486299363698336e-05,
      "loss": 0.8681,
      "num_input_tokens_seen": 1131806464,
      "step": 15210,
      "train_runtime": 18110.5554,
      "train_tokens_per_second": 62494.299
    },
    {
      "epoch": 0.20225779230702787,
      "grad_norm": 20.125,
      "learning_rate": 1.8484021836666865e-05,
      "loss": 0.9586,
      "num_input_tokens_seen": 1132638208,
      "step": 15220,
      "train_runtime": 18123.3258,
      "train_tokens_per_second": 62496.157
    },
    {
      "epoch": 0.20239068178948977,
      "grad_norm": 19.375,
      "learning_rate": 1.8481742738046002e-05,
      "loss": 0.7131,
      "num_input_tokens_seen": 1133138176,
      "step": 15230,
      "train_runtime": 18130.1654,
      "train_tokens_per_second": 62500.156
    },
    {
      "epoch": 0.20252357127195167,
      "grad_norm": 11.5,
      "learning_rate": 1.8479462068257927e-05,
      "loss": 0.6621,
      "num_input_tokens_seen": 1133366016,
      "step": 15240,
      "train_runtime": 18133.6174,
      "train_tokens_per_second": 62500.823
    },
    {
      "epoch": 0.2026564607544136,
      "grad_norm": 12.0625,
      "learning_rate": 1.8477179827725115e-05,
      "loss": 0.563,
      "num_input_tokens_seen": 1133483264,
      "step": 15250,
      "train_runtime": 18136.3673,
      "train_tokens_per_second": 62497.812
    },
    {
      "epoch": 0.2027893502368755,
      "grad_norm": 14.125,
      "learning_rate": 1.8474896016870325e-05,
      "loss": 0.7973,
      "num_input_tokens_seen": 1135448832,
      "step": 15260,
      "train_runtime": 18167.8354,
      "train_tokens_per_second": 62497.75
    },
    {
      "epoch": 0.2029222397193374,
      "grad_norm": 33.0,
      "learning_rate": 1.8472610636116618e-05,
      "loss": 0.8801,
      "num_input_tokens_seen": 1136278784,
      "step": 15270,
      "train_runtime": 18180.1714,
      "train_tokens_per_second": 62500.994
    },
    {
      "epoch": 0.20305512920179933,
      "grad_norm": 11.625,
      "learning_rate": 1.8470323685887343e-05,
      "loss": 0.7519,
      "num_input_tokens_seen": 1136755200,
      "step": 15280,
      "train_runtime": 18186.9137,
      "train_tokens_per_second": 62504.019
    },
    {
      "epoch": 0.20318801868426123,
      "grad_norm": 14.0625,
      "learning_rate": 1.846803516660613e-05,
      "loss": 0.6588,
      "num_input_tokens_seen": 1136982528,
      "step": 15290,
      "train_runtime": 18190.3696,
      "train_tokens_per_second": 62504.641
    },
    {
      "epoch": 0.20332090816672316,
      "grad_norm": 10.0,
      "learning_rate": 1.8465745078696914e-05,
      "loss": 0.4717,
      "num_input_tokens_seen": 1137094656,
      "step": 15300,
      "train_runtime": 18193.4492,
      "train_tokens_per_second": 62500.224
    },
    {
      "epoch": 0.20345379764918506,
      "grad_norm": 28.625,
      "learning_rate": 1.846345342258391e-05,
      "loss": 0.8055,
      "num_input_tokens_seen": 1139154944,
      "step": 15310,
      "train_runtime": 18227.6697,
      "train_tokens_per_second": 62495.918
    },
    {
      "epoch": 0.20358668713164696,
      "grad_norm": 19.0,
      "learning_rate": 1.8461160198691625e-05,
      "loss": 0.9065,
      "num_input_tokens_seen": 1140041984,
      "step": 15320,
      "train_runtime": 18240.7748,
      "train_tokens_per_second": 62499.647
    },
    {
      "epoch": 0.2037195766141089,
      "grad_norm": 14.25,
      "learning_rate": 1.8458865407444858e-05,
      "loss": 0.8035,
      "num_input_tokens_seen": 1140532224,
      "step": 15330,
      "train_runtime": 18247.5403,
      "train_tokens_per_second": 62503.34
    },
    {
      "epoch": 0.2038524660965708,
      "grad_norm": 30.75,
      "learning_rate": 1.8456569049268702e-05,
      "loss": 0.6899,
      "num_input_tokens_seen": 1140754688,
      "step": 15340,
      "train_runtime": 18250.9102,
      "train_tokens_per_second": 62503.989
    },
    {
      "epoch": 0.2039853555790327,
      "grad_norm": 25.75,
      "learning_rate": 1.8454271124588536e-05,
      "loss": 0.5151,
      "num_input_tokens_seen": 1140867328,
      "step": 15350,
      "train_runtime": 18253.3097,
      "train_tokens_per_second": 62501.943
    },
    {
      "epoch": 0.20411824506149462,
      "grad_norm": 15.25,
      "learning_rate": 1.845197163383003e-05,
      "loss": 0.9001,
      "num_input_tokens_seen": 1142898944,
      "step": 15360,
      "train_runtime": 18286.7249,
      "train_tokens_per_second": 62498.832
    },
    {
      "epoch": 0.20425113454395652,
      "grad_norm": 26.625,
      "learning_rate": 1.844967057741914e-05,
      "loss": 0.9014,
      "num_input_tokens_seen": 1143769856,
      "step": 15370,
      "train_runtime": 18299.6744,
      "train_tokens_per_second": 62502.197
    },
    {
      "epoch": 0.20438402402641842,
      "grad_norm": 12.375,
      "learning_rate": 1.8447367955782124e-05,
      "loss": 0.7789,
      "num_input_tokens_seen": 1144264960,
      "step": 15380,
      "train_runtime": 18307.3196,
      "train_tokens_per_second": 62503.14
    },
    {
      "epoch": 0.20451691350888035,
      "grad_norm": 10.1875,
      "learning_rate": 1.8445063769345517e-05,
      "loss": 0.5761,
      "num_input_tokens_seen": 1144493568,
      "step": 15390,
      "train_runtime": 18310.8363,
      "train_tokens_per_second": 62503.621
    },
    {
      "epoch": 0.20464980299134225,
      "grad_norm": 19.0,
      "learning_rate": 1.844275801853615e-05,
      "loss": 0.4585,
      "num_input_tokens_seen": 1144614400,
      "step": 15400,
      "train_runtime": 18313.3909,
      "train_tokens_per_second": 62501.5
    },
    {
      "epoch": 0.20478269247380415,
      "grad_norm": 29.875,
      "learning_rate": 1.8440450703781143e-05,
      "loss": 0.8939,
      "num_input_tokens_seen": 1146718720,
      "step": 15410,
      "train_runtime": 18347.2977,
      "train_tokens_per_second": 62500.688
    },
    {
      "epoch": 0.20491558195626608,
      "grad_norm": 27.75,
      "learning_rate": 1.8438141825507903e-05,
      "loss": 0.9812,
      "num_input_tokens_seen": 1147549952,
      "step": 15420,
      "train_runtime": 18359.8887,
      "train_tokens_per_second": 62503.099
    },
    {
      "epoch": 0.20504847143872798,
      "grad_norm": 14.1875,
      "learning_rate": 1.8435831384144135e-05,
      "loss": 0.7799,
      "num_input_tokens_seen": 1148025088,
      "step": 15430,
      "train_runtime": 18366.5004,
      "train_tokens_per_second": 62506.469
    },
    {
      "epoch": 0.20518136092118988,
      "grad_norm": 21.25,
      "learning_rate": 1.8433519380117823e-05,
      "loss": 0.6672,
      "num_input_tokens_seen": 1148249344,
      "step": 15440,
      "train_runtime": 18370.4143,
      "train_tokens_per_second": 62505.359
    },
    {
      "epoch": 0.2053142504036518,
      "grad_norm": 8.875,
      "learning_rate": 1.8431205813857246e-05,
      "loss": 0.5205,
      "num_input_tokens_seen": 1148360960,
      "step": 15450,
      "train_runtime": 18372.749,
      "train_tokens_per_second": 62503.492
    },
    {
      "epoch": 0.2054471398861137,
      "grad_norm": 72.5,
      "learning_rate": 1.842889068579097e-05,
      "loss": 0.8664,
      "num_input_tokens_seen": 1150494720,
      "step": 15460,
      "train_runtime": 18406.6595,
      "train_tokens_per_second": 62504.265
    },
    {
      "epoch": 0.2055800293685756,
      "grad_norm": 29.0,
      "learning_rate": 1.8426573996347854e-05,
      "loss": 0.9109,
      "num_input_tokens_seen": 1151329280,
      "step": 15470,
      "train_runtime": 18419.3753,
      "train_tokens_per_second": 62506.424
    },
    {
      "epoch": 0.20571291885103754,
      "grad_norm": 50.75,
      "learning_rate": 1.8424255745957044e-05,
      "loss": 0.8232,
      "num_input_tokens_seen": 1151785472,
      "step": 15480,
      "train_runtime": 18425.8981,
      "train_tokens_per_second": 62509.055
    },
    {
      "epoch": 0.20584580833349944,
      "grad_norm": 34.25,
      "learning_rate": 1.8421935935047974e-05,
      "loss": 0.682,
      "num_input_tokens_seen": 1152019200,
      "step": 15490,
      "train_runtime": 18429.8009,
      "train_tokens_per_second": 62508.5
    },
    {
      "epoch": 0.20597869781596134,
      "grad_norm": 17.5,
      "learning_rate": 1.841961456405037e-05,
      "loss": 0.5698,
      "num_input_tokens_seen": 1152136960,
      "step": 15500,
      "train_runtime": 18432.1933,
      "train_tokens_per_second": 62506.775
    },
    {
      "epoch": 0.20611158729842327,
      "grad_norm": 20.25,
      "learning_rate": 1.8417291633394242e-05,
      "loss": 0.909,
      "num_input_tokens_seen": 1154129920,
      "step": 15510,
      "train_runtime": 18464.1722,
      "train_tokens_per_second": 62506.453
    },
    {
      "epoch": 0.20624447678088517,
      "grad_norm": 12.75,
      "learning_rate": 1.8414967143509898e-05,
      "loss": 0.8215,
      "num_input_tokens_seen": 1154977280,
      "step": 15520,
      "train_runtime": 18476.6966,
      "train_tokens_per_second": 62509.945
    },
    {
      "epoch": 0.2063773662633471,
      "grad_norm": 11.1875,
      "learning_rate": 1.8412641094827926e-05,
      "loss": 0.7349,
      "num_input_tokens_seen": 1155419136,
      "step": 15530,
      "train_runtime": 18483.1094,
      "train_tokens_per_second": 62512.162
    },
    {
      "epoch": 0.206510255745809,
      "grad_norm": 12.5,
      "learning_rate": 1.8410313487779204e-05,
      "loss": 0.7169,
      "num_input_tokens_seen": 1155641344,
      "step": 15540,
      "train_runtime": 18486.4689,
      "train_tokens_per_second": 62512.822
    },
    {
      "epoch": 0.2066431452282709,
      "grad_norm": 15.75,
      "learning_rate": 1.8407984322794902e-05,
      "loss": 0.477,
      "num_input_tokens_seen": 1155753984,
      "step": 15550,
      "train_runtime": 18489.2922,
      "train_tokens_per_second": 62509.369
    },
    {
      "epoch": 0.20677603471073283,
      "grad_norm": 22.25,
      "learning_rate": 1.8405653600306474e-05,
      "loss": 0.835,
      "num_input_tokens_seen": 1157767936,
      "step": 15560,
      "train_runtime": 18521.9469,
      "train_tokens_per_second": 62507.896
    },
    {
      "epoch": 0.20690892419319473,
      "grad_norm": 16.5,
      "learning_rate": 1.840332132074567e-05,
      "loss": 0.8793,
      "num_input_tokens_seen": 1158582272,
      "step": 15570,
      "train_runtime": 18534.4602,
      "train_tokens_per_second": 62509.631
    },
    {
      "epoch": 0.20704181367565663,
      "grad_norm": 17.75,
      "learning_rate": 1.8400987484544523e-05,
      "loss": 0.7407,
      "num_input_tokens_seen": 1159044352,
      "step": 15580,
      "train_runtime": 18541.02,
      "train_tokens_per_second": 62512.437
    },
    {
      "epoch": 0.20717470315811856,
      "grad_norm": 13.3125,
      "learning_rate": 1.8398652092135358e-05,
      "loss": 0.606,
      "num_input_tokens_seen": 1159276800,
      "step": 15590,
      "train_runtime": 18544.4966,
      "train_tokens_per_second": 62513.253
    },
    {
      "epoch": 0.20730759264058046,
      "grad_norm": 5.59375,
      "learning_rate": 1.8396315143950776e-05,
      "loss": 0.5333,
      "num_input_tokens_seen": 1159390464,
      "step": 15600,
      "train_runtime": 18547.4219,
      "train_tokens_per_second": 62509.521
    },
    {
      "epoch": 0.20744048212304236,
      "grad_norm": 25.375,
      "learning_rate": 1.8393976640423683e-05,
      "loss": 0.9165,
      "num_input_tokens_seen": 1161401344,
      "step": 15610,
      "train_runtime": 18580.0479,
      "train_tokens_per_second": 62507.985
    },
    {
      "epoch": 0.2075733716055043,
      "grad_norm": 35.75,
      "learning_rate": 1.839163658198727e-05,
      "loss": 0.9197,
      "num_input_tokens_seen": 1162214912,
      "step": 15620,
      "train_runtime": 18592.2465,
      "train_tokens_per_second": 62510.73
    },
    {
      "epoch": 0.2077062610879662,
      "grad_norm": 22.75,
      "learning_rate": 1.8389294969075003e-05,
      "loss": 0.7975,
      "num_input_tokens_seen": 1162671616,
      "step": 15630,
      "train_runtime": 18598.6078,
      "train_tokens_per_second": 62513.906
    },
    {
      "epoch": 0.2078391505704281,
      "grad_norm": 12.75,
      "learning_rate": 1.838695180212065e-05,
      "loss": 0.6428,
      "num_input_tokens_seen": 1162894080,
      "step": 15640,
      "train_runtime": 18602.0082,
      "train_tokens_per_second": 62514.438
    },
    {
      "epoch": 0.20797204005289002,
      "grad_norm": 17.875,
      "learning_rate": 1.838460708155826e-05,
      "loss": 0.5194,
      "num_input_tokens_seen": 1163006976,
      "step": 15650,
      "train_runtime": 18604.8231,
      "train_tokens_per_second": 62511.047
    },
    {
      "epoch": 0.20810492953535192,
      "grad_norm": 16.875,
      "learning_rate": 1.838226080782217e-05,
      "loss": 0.8425,
      "num_input_tokens_seen": 1165012480,
      "step": 15660,
      "train_runtime": 18637.0704,
      "train_tokens_per_second": 62510.494
    },
    {
      "epoch": 0.20823781901781382,
      "grad_norm": 33.25,
      "learning_rate": 1.8379912981347015e-05,
      "loss": 0.8999,
      "num_input_tokens_seen": 1165847552,
      "step": 15670,
      "train_runtime": 18649.5269,
      "train_tokens_per_second": 62513.519
    },
    {
      "epoch": 0.20837070850027575,
      "grad_norm": 17.875,
      "learning_rate": 1.83775636025677e-05,
      "loss": 0.7306,
      "num_input_tokens_seen": 1166303488,
      "step": 15680,
      "train_runtime": 18656.1739,
      "train_tokens_per_second": 62515.685
    },
    {
      "epoch": 0.20850359798273765,
      "grad_norm": 11.4375,
      "learning_rate": 1.8375212671919425e-05,
      "loss": 0.6412,
      "num_input_tokens_seen": 1166529024,
      "step": 15690,
      "train_runtime": 18659.596,
      "train_tokens_per_second": 62516.307
    },
    {
      "epoch": 0.20863648746519955,
      "grad_norm": 18.75,
      "learning_rate": 1.837286018983769e-05,
      "loss": 0.5835,
      "num_input_tokens_seen": 1166645504,
      "step": 15700,
      "train_runtime": 18662.4371,
      "train_tokens_per_second": 62513.031
    },
    {
      "epoch": 0.20876937694766148,
      "grad_norm": 15.875,
      "learning_rate": 1.837050615675826e-05,
      "loss": 0.859,
      "num_input_tokens_seen": 1168580352,
      "step": 15710,
      "train_runtime": 18693.839,
      "train_tokens_per_second": 62511.523
    },
    {
      "epoch": 0.20890226643012338,
      "grad_norm": 23.875,
      "learning_rate": 1.83681505731172e-05,
      "loss": 0.9571,
      "num_input_tokens_seen": 1169373952,
      "step": 15720,
      "train_runtime": 18705.9689,
      "train_tokens_per_second": 62513.413
    },
    {
      "epoch": 0.20903515591258529,
      "grad_norm": 17.75,
      "learning_rate": 1.836579343935087e-05,
      "loss": 0.7093,
      "num_input_tokens_seen": 1169835008,
      "step": 15730,
      "train_runtime": 18712.4436,
      "train_tokens_per_second": 62516.421
    },
    {
      "epoch": 0.20916804539504721,
      "grad_norm": 17.375,
      "learning_rate": 1.83634347558959e-05,
      "loss": 0.6229,
      "num_input_tokens_seen": 1170071808,
      "step": 15740,
      "train_runtime": 18716.3342,
      "train_tokens_per_second": 62516.078
    },
    {
      "epoch": 0.20930093487750911,
      "grad_norm": 13.625,
      "learning_rate": 1.8361074523189214e-05,
      "loss": 0.6045,
      "num_input_tokens_seen": 1170186752,
      "step": 15750,
      "train_runtime": 18718.7142,
      "train_tokens_per_second": 62514.27
    },
    {
      "epoch": 0.20943382435997104,
      "grad_norm": 38.25,
      "learning_rate": 1.835871274166803e-05,
      "loss": 0.8365,
      "num_input_tokens_seen": 1172257024,
      "step": 15760,
      "train_runtime": 18751.8455,
      "train_tokens_per_second": 62514.221
    },
    {
      "epoch": 0.20956671384243294,
      "grad_norm": 27.25,
      "learning_rate": 1.835634941176984e-05,
      "loss": 0.9416,
      "num_input_tokens_seen": 1173132288,
      "step": 15770,
      "train_runtime": 18764.6852,
      "train_tokens_per_second": 62518.09
    },
    {
      "epoch": 0.20969960332489485,
      "grad_norm": 17.5,
      "learning_rate": 1.8353984533932436e-05,
      "loss": 0.7816,
      "num_input_tokens_seen": 1173627392,
      "step": 15780,
      "train_runtime": 18771.4651,
      "train_tokens_per_second": 62521.885
    },
    {
      "epoch": 0.20983249280735677,
      "grad_norm": 16.625,
      "learning_rate": 1.835161810859389e-05,
      "loss": 0.6468,
      "num_input_tokens_seen": 1173861888,
      "step": 15790,
      "train_runtime": 18774.9968,
      "train_tokens_per_second": 62522.614
    },
    {
      "epoch": 0.20996538228981868,
      "grad_norm": 16.5,
      "learning_rate": 1.8349250136192558e-05,
      "loss": 0.5603,
      "num_input_tokens_seen": 1173979648,
      "step": 15800,
      "train_runtime": 18777.4031,
      "train_tokens_per_second": 62520.874
    },
    {
      "epoch": 0.21009827177228058,
      "grad_norm": 24.375,
      "learning_rate": 1.8346880617167088e-05,
      "loss": 0.8478,
      "num_input_tokens_seen": 1176024832,
      "step": 15810,
      "train_runtime": 18809.6762,
      "train_tokens_per_second": 62522.333
    },
    {
      "epoch": 0.2102311612547425,
      "grad_norm": 20.75,
      "learning_rate": 1.8344509551956408e-05,
      "loss": 0.9081,
      "num_input_tokens_seen": 1176869376,
      "step": 15820,
      "train_runtime": 18822.7094,
      "train_tokens_per_second": 62523.909
    },
    {
      "epoch": 0.2103640507372044,
      "grad_norm": 18.125,
      "learning_rate": 1.8342136940999744e-05,
      "loss": 0.7439,
      "num_input_tokens_seen": 1177349632,
      "step": 15830,
      "train_runtime": 18829.4736,
      "train_tokens_per_second": 62526.954
    },
    {
      "epoch": 0.2104969402196663,
      "grad_norm": 10.375,
      "learning_rate": 1.833976278473659e-05,
      "loss": 0.6933,
      "num_input_tokens_seen": 1177577984,
      "step": 15840,
      "train_runtime": 18832.9232,
      "train_tokens_per_second": 62527.626
    },
    {
      "epoch": 0.21062982970212824,
      "grad_norm": 10.5625,
      "learning_rate": 1.8337387083606752e-05,
      "loss": 0.4896,
      "num_input_tokens_seen": 1177691648,
      "step": 15850,
      "train_runtime": 18835.2996,
      "train_tokens_per_second": 62525.772
    },
    {
      "epoch": 0.21076271918459014,
      "grad_norm": 33.25,
      "learning_rate": 1.83350098380503e-05,
      "loss": 0.8999,
      "num_input_tokens_seen": 1179804160,
      "step": 15860,
      "train_runtime": 18869.5681,
      "train_tokens_per_second": 62524.174
    },
    {
      "epoch": 0.21089560866705204,
      "grad_norm": 33.75,
      "learning_rate": 1.8332631048507594e-05,
      "loss": 0.9657,
      "num_input_tokens_seen": 1180654080,
      "step": 15870,
      "train_runtime": 18882.0391,
      "train_tokens_per_second": 62527.891
    },
    {
      "epoch": 0.21102849814951397,
      "grad_norm": 25.5,
      "learning_rate": 1.833025071541929e-05,
      "loss": 0.8455,
      "num_input_tokens_seen": 1181114880,
      "step": 15880,
      "train_runtime": 18888.5528,
      "train_tokens_per_second": 62530.724
    },
    {
      "epoch": 0.21116138763197587,
      "grad_norm": 11.5625,
      "learning_rate": 1.8327868839226316e-05,
      "loss": 0.61,
      "num_input_tokens_seen": 1181344000,
      "step": 15890,
      "train_runtime": 18891.9544,
      "train_tokens_per_second": 62531.593
    },
    {
      "epoch": 0.21129427711443777,
      "grad_norm": 20.875,
      "learning_rate": 1.83254854203699e-05,
      "loss": 0.5078,
      "num_input_tokens_seen": 1181456384,
      "step": 15900,
      "train_runtime": 18894.3303,
      "train_tokens_per_second": 62529.678
    },
    {
      "epoch": 0.2114271665968997,
      "grad_norm": 17.625,
      "learning_rate": 1.8323100459291548e-05,
      "loss": 0.8301,
      "num_input_tokens_seen": 1183476224,
      "step": 15910,
      "train_runtime": 18925.8406,
      "train_tokens_per_second": 62532.294
    },
    {
      "epoch": 0.2115600560793616,
      "grad_norm": 50.75,
      "learning_rate": 1.832071395643305e-05,
      "loss": 0.9061,
      "num_input_tokens_seen": 1184311552,
      "step": 15920,
      "train_runtime": 18938.3853,
      "train_tokens_per_second": 62534.98
    },
    {
      "epoch": 0.2116929455618235,
      "grad_norm": 68.5,
      "learning_rate": 1.831832591223649e-05,
      "loss": 0.6834,
      "num_input_tokens_seen": 1184774656,
      "step": 15930,
      "train_runtime": 18945.3698,
      "train_tokens_per_second": 62536.37
    },
    {
      "epoch": 0.21182583504428543,
      "grad_norm": 60.5,
      "learning_rate": 1.8315936327144223e-05,
      "loss": 0.7037,
      "num_input_tokens_seen": 1184992768,
      "step": 15940,
      "train_runtime": 18948.6777,
      "train_tokens_per_second": 62536.964
    },
    {
      "epoch": 0.21195872452674733,
      "grad_norm": 20.875,
      "learning_rate": 1.8313545201598903e-05,
      "loss": 0.6446,
      "num_input_tokens_seen": 1185105664,
      "step": 15950,
      "train_runtime": 18951.0877,
      "train_tokens_per_second": 62534.968
    },
    {
      "epoch": 0.21209161400920923,
      "grad_norm": 20.75,
      "learning_rate": 1.831115253604347e-05,
      "loss": 0.9073,
      "num_input_tokens_seen": 1187170816,
      "step": 15960,
      "train_runtime": 18984.8902,
      "train_tokens_per_second": 62532.404
    },
    {
      "epoch": 0.21222450349167116,
      "grad_norm": 21.375,
      "learning_rate": 1.8308758330921134e-05,
      "loss": 0.9141,
      "num_input_tokens_seen": 1188054528,
      "step": 15970,
      "train_runtime": 18997.6685,
      "train_tokens_per_second": 62536.86
    },
    {
      "epoch": 0.21235739297413306,
      "grad_norm": 22.875,
      "learning_rate": 1.8306362586675406e-05,
      "loss": 0.7737,
      "num_input_tokens_seen": 1188534528,
      "step": 15980,
      "train_runtime": 19004.912,
      "train_tokens_per_second": 62538.281
    },
    {
      "epoch": 0.212490282456595,
      "grad_norm": 12.125,
      "learning_rate": 1.8303965303750074e-05,
      "loss": 0.7106,
      "num_input_tokens_seen": 1188762880,
      "step": 15990,
      "train_runtime": 19008.3897,
      "train_tokens_per_second": 62538.853
    },
    {
      "epoch": 0.2126231719390569,
      "grad_norm": 38.0,
      "learning_rate": 1.830156648258922e-05,
      "loss": 0.5566,
      "num_input_tokens_seen": 1188876544,
      "step": 16000,
      "train_runtime": 19010.7768,
      "train_tokens_per_second": 62536.979
    },
    {
      "epoch": 0.2127560614215188,
      "grad_norm": 22.125,
      "learning_rate": 1.8299166123637194e-05,
      "loss": 0.8262,
      "num_input_tokens_seen": 1190925312,
      "step": 16010,
      "train_runtime": 19044.0774,
      "train_tokens_per_second": 62535.207
    },
    {
      "epoch": 0.21288895090398072,
      "grad_norm": 12.5625,
      "learning_rate": 1.8296764227338648e-05,
      "loss": 0.8969,
      "num_input_tokens_seen": 1191751936,
      "step": 16020,
      "train_runtime": 19056.5631,
      "train_tokens_per_second": 62537.611
    },
    {
      "epoch": 0.21302184038644262,
      "grad_norm": 17.0,
      "learning_rate": 1.829436079413851e-05,
      "loss": 0.7748,
      "num_input_tokens_seen": 1192220416,
      "step": 16030,
      "train_runtime": 19063.5632,
      "train_tokens_per_second": 62539.222
    },
    {
      "epoch": 0.21315472986890452,
      "grad_norm": 14.75,
      "learning_rate": 1.8291955824481993e-05,
      "loss": 0.6895,
      "num_input_tokens_seen": 1192443136,
      "step": 16040,
      "train_runtime": 19066.9933,
      "train_tokens_per_second": 62539.653
    },
    {
      "epoch": 0.21328761935136645,
      "grad_norm": 17.375,
      "learning_rate": 1.8289549318814598e-05,
      "loss": 0.499,
      "num_input_tokens_seen": 1192557056,
      "step": 16050,
      "train_runtime": 19069.3905,
      "train_tokens_per_second": 62537.765
    },
    {
      "epoch": 0.21342050883382835,
      "grad_norm": 29.25,
      "learning_rate": 1.8287141277582108e-05,
      "loss": 0.8283,
      "num_input_tokens_seen": 1194695168,
      "step": 16060,
      "train_runtime": 19102.8774,
      "train_tokens_per_second": 62540.064
    },
    {
      "epoch": 0.21355339831629025,
      "grad_norm": 19.875,
      "learning_rate": 1.828473170123059e-05,
      "loss": 0.9726,
      "num_input_tokens_seen": 1195546112,
      "step": 16070,
      "train_runtime": 19115.7993,
      "train_tokens_per_second": 62542.303
    },
    {
      "epoch": 0.21368628779875218,
      "grad_norm": 13.5625,
      "learning_rate": 1.8282320590206402e-05,
      "loss": 0.7558,
      "num_input_tokens_seen": 1196008448,
      "step": 16080,
      "train_runtime": 19122.5266,
      "train_tokens_per_second": 62544.478
    },
    {
      "epoch": 0.21381917728121408,
      "grad_norm": 24.375,
      "learning_rate": 1.8279907944956173e-05,
      "loss": 0.7081,
      "num_input_tokens_seen": 1196234496,
      "step": 16090,
      "train_runtime": 19125.9153,
      "train_tokens_per_second": 62545.215
    },
    {
      "epoch": 0.21395206676367598,
      "grad_norm": 7.21875,
      "learning_rate": 1.827749376592683e-05,
      "loss": 0.5433,
      "num_input_tokens_seen": 1196349696,
      "step": 16100,
      "train_runtime": 19128.252,
      "train_tokens_per_second": 62543.597
    },
    {
      "epoch": 0.2140849562461379,
      "grad_norm": 34.0,
      "learning_rate": 1.8275078053565575e-05,
      "loss": 0.8228,
      "num_input_tokens_seen": 1198508800,
      "step": 16110,
      "train_runtime": 19163.0965,
      "train_tokens_per_second": 62542.544
    },
    {
      "epoch": 0.2142178457285998,
      "grad_norm": 26.625,
      "learning_rate": 1.82726608083199e-05,
      "loss": 0.9478,
      "num_input_tokens_seen": 1199339776,
      "step": 16120,
      "train_runtime": 19175.489,
      "train_tokens_per_second": 62545.46
    },
    {
      "epoch": 0.2143507352110617,
      "grad_norm": 17.125,
      "learning_rate": 1.8270242030637573e-05,
      "loss": 0.7258,
      "num_input_tokens_seen": 1199794432,
      "step": 16130,
      "train_runtime": 19182.0391,
      "train_tokens_per_second": 62547.805
    },
    {
      "epoch": 0.21448362469352364,
      "grad_norm": 14.1875,
      "learning_rate": 1.8267821720966654e-05,
      "loss": 0.6806,
      "num_input_tokens_seen": 1200024064,
      "step": 16140,
      "train_runtime": 19185.4645,
      "train_tokens_per_second": 62548.606
    },
    {
      "epoch": 0.21461651417598554,
      "grad_norm": 16.125,
      "learning_rate": 1.8265399879755487e-05,
      "loss": 0.5184,
      "num_input_tokens_seen": 1200140800,
      "step": 16150,
      "train_runtime": 19187.8163,
      "train_tokens_per_second": 62547.024
    },
    {
      "epoch": 0.21474940365844744,
      "grad_norm": 23.0,
      "learning_rate": 1.826297650745269e-05,
      "loss": 0.8604,
      "num_input_tokens_seen": 1202269440,
      "step": 16160,
      "train_runtime": 19222.1137,
      "train_tokens_per_second": 62546.162
    },
    {
      "epoch": 0.21488229314090937,
      "grad_norm": 32.25,
      "learning_rate": 1.8260551604507176e-05,
      "loss": 0.9023,
      "num_input_tokens_seen": 1203113984,
      "step": 16170,
      "train_runtime": 19235.482,
      "train_tokens_per_second": 62546.599
    },
    {
      "epoch": 0.21501518262337127,
      "grad_norm": 22.375,
      "learning_rate": 1.8258125171368137e-05,
      "loss": 0.8274,
      "num_input_tokens_seen": 1203621376,
      "step": 16180,
      "train_runtime": 19242.456,
      "train_tokens_per_second": 62550.299
    },
    {
      "epoch": 0.21514807210583317,
      "grad_norm": 25.625,
      "learning_rate": 1.8255697208485045e-05,
      "loss": 0.6651,
      "num_input_tokens_seen": 1203857152,
      "step": 16190,
      "train_runtime": 19245.9104,
      "train_tokens_per_second": 62551.323
    },
    {
      "epoch": 0.2152809615882951,
      "grad_norm": 14.125,
      "learning_rate": 1.8253267716307662e-05,
      "loss": 0.5169,
      "num_input_tokens_seen": 1203971584,
      "step": 16200,
      "train_runtime": 19248.2809,
      "train_tokens_per_second": 62549.564
    },
    {
      "epoch": 0.215413851070757,
      "grad_norm": 12.0625,
      "learning_rate": 1.8250836695286022e-05,
      "loss": 0.8467,
      "num_input_tokens_seen": 1205976576,
      "step": 16210,
      "train_runtime": 19281.9933,
      "train_tokens_per_second": 62544.186
    },
    {
      "epoch": 0.2155467405532189,
      "grad_norm": 29.5,
      "learning_rate": 1.824840414587046e-05,
      "loss": 0.8994,
      "num_input_tokens_seen": 1206774528,
      "step": 16220,
      "train_runtime": 19294.5503,
      "train_tokens_per_second": 62544.838
    },
    {
      "epoch": 0.21567963003568083,
      "grad_norm": 16.75,
      "learning_rate": 1.8245970068511578e-05,
      "loss": 0.7835,
      "num_input_tokens_seen": 1207193344,
      "step": 16230,
      "train_runtime": 19300.7441,
      "train_tokens_per_second": 62546.467
    },
    {
      "epoch": 0.21581251951814273,
      "grad_norm": 13.125,
      "learning_rate": 1.8243534463660274e-05,
      "loss": 0.6532,
      "num_input_tokens_seen": 1207411200,
      "step": 16240,
      "train_runtime": 19304.0849,
      "train_tokens_per_second": 62546.928
    },
    {
      "epoch": 0.21594540900060466,
      "grad_norm": 25.125,
      "learning_rate": 1.824109733176771e-05,
      "loss": 0.5889,
      "num_input_tokens_seen": 1207523072,
      "step": 16250,
      "train_runtime": 19306.4451,
      "train_tokens_per_second": 62545.076
    },
    {
      "epoch": 0.21607829848306656,
      "grad_norm": 15.125,
      "learning_rate": 1.8238658673285352e-05,
      "loss": 0.8184,
      "num_input_tokens_seen": 1209527040,
      "step": 16260,
      "train_runtime": 19339.2435,
      "train_tokens_per_second": 62542.624
    },
    {
      "epoch": 0.21621118796552846,
      "grad_norm": 19.125,
      "learning_rate": 1.8236218488664938e-05,
      "loss": 0.9233,
      "num_input_tokens_seen": 1210390784,
      "step": 16270,
      "train_runtime": 19351.7597,
      "train_tokens_per_second": 62546.807
    },
    {
      "epoch": 0.2163440774479904,
      "grad_norm": 22.75,
      "learning_rate": 1.823377677835849e-05,
      "loss": 0.7442,
      "num_input_tokens_seen": 1210853888,
      "step": 16280,
      "train_runtime": 19358.3102,
      "train_tokens_per_second": 62549.565
    },
    {
      "epoch": 0.2164769669304523,
      "grad_norm": 23.75,
      "learning_rate": 1.8231333542818313e-05,
      "loss": 0.6956,
      "num_input_tokens_seen": 1211086080,
      "step": 16290,
      "train_runtime": 19361.7889,
      "train_tokens_per_second": 62550.319
    },
    {
      "epoch": 0.2166098564129142,
      "grad_norm": 8.375,
      "learning_rate": 1.8228888782496993e-05,
      "loss": 0.5722,
      "num_input_tokens_seen": 1211200256,
      "step": 16300,
      "train_runtime": 19364.2108,
      "train_tokens_per_second": 62548.392
    },
    {
      "epoch": 0.21674274589537612,
      "grad_norm": 15.4375,
      "learning_rate": 1.8226442497847403e-05,
      "loss": 0.8896,
      "num_input_tokens_seen": 1213191424,
      "step": 16310,
      "train_runtime": 19395.7694,
      "train_tokens_per_second": 62549.281
    },
    {
      "epoch": 0.21687563537783802,
      "grad_norm": 14.375,
      "learning_rate": 1.8223994689322694e-05,
      "loss": 0.995,
      "num_input_tokens_seen": 1214046464,
      "step": 16320,
      "train_runtime": 19408.3507,
      "train_tokens_per_second": 62552.789
    },
    {
      "epoch": 0.21700852486029992,
      "grad_norm": 171.0,
      "learning_rate": 1.8221545357376304e-05,
      "loss": 0.7792,
      "num_input_tokens_seen": 1214507008,
      "step": 16330,
      "train_runtime": 19415.1351,
      "train_tokens_per_second": 62554.651
    },
    {
      "epoch": 0.21714141434276185,
      "grad_norm": 61.25,
      "learning_rate": 1.821909450246194e-05,
      "loss": 0.8058,
      "num_input_tokens_seen": 1214748672,
      "step": 16340,
      "train_runtime": 19419.1121,
      "train_tokens_per_second": 62554.285
    },
    {
      "epoch": 0.21727430382522375,
      "grad_norm": 14.5625,
      "learning_rate": 1.8216642125033616e-05,
      "loss": 0.505,
      "num_input_tokens_seen": 1214866944,
      "step": 16350,
      "train_runtime": 19421.5025,
      "train_tokens_per_second": 62552.676
    },
    {
      "epoch": 0.21740719330768565,
      "grad_norm": 12.375,
      "learning_rate": 1.82141882255456e-05,
      "loss": 0.8572,
      "num_input_tokens_seen": 1216826624,
      "step": 16360,
      "train_runtime": 19453.2707,
      "train_tokens_per_second": 62551.262
    },
    {
      "epoch": 0.21754008279014758,
      "grad_norm": 13.1875,
      "learning_rate": 1.8211732804452457e-05,
      "loss": 0.958,
      "num_input_tokens_seen": 1217660160,
      "step": 16370,
      "train_runtime": 19465.8149,
      "train_tokens_per_second": 62553.773
    },
    {
      "epoch": 0.21767297227260948,
      "grad_norm": 12.875,
      "learning_rate": 1.8209275862209037e-05,
      "loss": 0.741,
      "num_input_tokens_seen": 1218124544,
      "step": 16380,
      "train_runtime": 19472.4916,
      "train_tokens_per_second": 62556.172
    },
    {
      "epoch": 0.21780586175507138,
      "grad_norm": 23.75,
      "learning_rate": 1.8206817399270464e-05,
      "loss": 0.7018,
      "num_input_tokens_seen": 1218358784,
      "step": 16390,
      "train_runtime": 19475.9695,
      "train_tokens_per_second": 62557.029
    },
    {
      "epoch": 0.2179387512375333,
      "grad_norm": 21.0,
      "learning_rate": 1.820435741609215e-05,
      "loss": 0.5979,
      "num_input_tokens_seen": 1218477568,
      "step": 16400,
      "train_runtime": 19478.8619,
      "train_tokens_per_second": 62553.838
    },
    {
      "epoch": 0.2180716407199952,
      "grad_norm": 20.375,
      "learning_rate": 1.820189591312978e-05,
      "loss": 0.8366,
      "num_input_tokens_seen": 1220607744,
      "step": 16410,
      "train_runtime": 19513.1572,
      "train_tokens_per_second": 62553.063
    },
    {
      "epoch": 0.21820453020245711,
      "grad_norm": 16.375,
      "learning_rate": 1.8199432890839324e-05,
      "loss": 0.9393,
      "num_input_tokens_seen": 1221439744,
      "step": 16420,
      "train_runtime": 19525.6717,
      "train_tokens_per_second": 62555.581
    },
    {
      "epoch": 0.21833741968491904,
      "grad_norm": 13.9375,
      "learning_rate": 1.819696834967704e-05,
      "loss": 0.8113,
      "num_input_tokens_seen": 1221908992,
      "step": 16430,
      "train_runtime": 19532.2048,
      "train_tokens_per_second": 62558.682
    },
    {
      "epoch": 0.21847030916738094,
      "grad_norm": 17.25,
      "learning_rate": 1.819450229009946e-05,
      "loss": 0.6428,
      "num_input_tokens_seen": 1222127104,
      "step": 16440,
      "train_runtime": 19535.525,
      "train_tokens_per_second": 62559.215
    },
    {
      "epoch": 0.21860319864984284,
      "grad_norm": 39.75,
      "learning_rate": 1.81920347125634e-05,
      "loss": 0.5125,
      "num_input_tokens_seen": 1222239488,
      "step": 16450,
      "train_runtime": 19538.3678,
      "train_tokens_per_second": 62555.864
    },
    {
      "epoch": 0.21873608813230477,
      "grad_norm": 13.6875,
      "learning_rate": 1.818956561752596e-05,
      "loss": 0.8589,
      "num_input_tokens_seen": 1224331776,
      "step": 16460,
      "train_runtime": 19572.505,
      "train_tokens_per_second": 62553.658
    },
    {
      "epoch": 0.21886897761476667,
      "grad_norm": 16.125,
      "learning_rate": 1.818709500544451e-05,
      "loss": 0.902,
      "num_input_tokens_seen": 1225159680,
      "step": 16470,
      "train_runtime": 19584.8582,
      "train_tokens_per_second": 62556.474
    },
    {
      "epoch": 0.2190018670972286,
      "grad_norm": 42.75,
      "learning_rate": 1.8184622876776715e-05,
      "loss": 0.7551,
      "num_input_tokens_seen": 1225632256,
      "step": 16480,
      "train_runtime": 19591.4601,
      "train_tokens_per_second": 62559.516
    },
    {
      "epoch": 0.2191347565796905,
      "grad_norm": 14.5,
      "learning_rate": 1.8182149231980512e-05,
      "loss": 0.6298,
      "num_input_tokens_seen": 1225860096,
      "step": 16490,
      "train_runtime": 19594.8578,
      "train_tokens_per_second": 62560.296
    },
    {
      "epoch": 0.2192676460621524,
      "grad_norm": 14.625,
      "learning_rate": 1.817967407151412e-05,
      "loss": 0.5386,
      "num_input_tokens_seen": 1225977344,
      "step": 16500,
      "train_runtime": 19597.7488,
      "train_tokens_per_second": 62557.05
    },
    {
      "epoch": 0.21940053554461433,
      "grad_norm": 15.0625,
      "learning_rate": 1.8177197395836048e-05,
      "loss": 0.833,
      "num_input_tokens_seen": 1228113152,
      "step": 16510,
      "train_runtime": 19631.5592,
      "train_tokens_per_second": 62558.105
    },
    {
      "epoch": 0.21953342502707623,
      "grad_norm": 21.25,
      "learning_rate": 1.8174719205405072e-05,
      "loss": 0.9863,
      "num_input_tokens_seen": 1228960768,
      "step": 16520,
      "train_runtime": 19644.2357,
      "train_tokens_per_second": 62560.885
    },
    {
      "epoch": 0.21966631450953814,
      "grad_norm": 17.25,
      "learning_rate": 1.8172239500680256e-05,
      "loss": 0.7749,
      "num_input_tokens_seen": 1229420544,
      "step": 16530,
      "train_runtime": 19650.8678,
      "train_tokens_per_second": 62563.168
    },
    {
      "epoch": 0.21979920399200006,
      "grad_norm": 19.625,
      "learning_rate": 1.816975828212094e-05,
      "loss": 0.6511,
      "num_input_tokens_seen": 1229641472,
      "step": 16540,
      "train_runtime": 19654.1959,
      "train_tokens_per_second": 62563.815
    },
    {
      "epoch": 0.21993209347446196,
      "grad_norm": 21.5,
      "learning_rate": 1.8167275550186754e-05,
      "loss": 0.5542,
      "num_input_tokens_seen": 1229753856,
      "step": 16550,
      "train_runtime": 19656.5494,
      "train_tokens_per_second": 62562.041
    },
    {
      "epoch": 0.22006498295692387,
      "grad_norm": 13.4375,
      "learning_rate": 1.8164791305337598e-05,
      "loss": 0.8837,
      "num_input_tokens_seen": 1231857408,
      "step": 16560,
      "train_runtime": 19690.4736,
      "train_tokens_per_second": 62561.086
    },
    {
      "epoch": 0.2201978724393858,
      "grad_norm": 13.625,
      "learning_rate": 1.8162305548033655e-05,
      "loss": 0.903,
      "num_input_tokens_seen": 1232691200,
      "step": 16570,
      "train_runtime": 19703.2026,
      "train_tokens_per_second": 62562.987
    },
    {
      "epoch": 0.2203307619218477,
      "grad_norm": 38.75,
      "learning_rate": 1.815981827873539e-05,
      "loss": 0.8407,
      "num_input_tokens_seen": 1233182208,
      "step": 16580,
      "train_runtime": 19710.0232,
      "train_tokens_per_second": 62566.248
    },
    {
      "epoch": 0.2204636514043096,
      "grad_norm": 11.0,
      "learning_rate": 1.815732949790355e-05,
      "loss": 0.6748,
      "num_input_tokens_seen": 1233416448,
      "step": 16590,
      "train_runtime": 19713.548,
      "train_tokens_per_second": 62566.944
    },
    {
      "epoch": 0.22059654088677152,
      "grad_norm": 13.625,
      "learning_rate": 1.815483920599916e-05,
      "loss": 0.5517,
      "num_input_tokens_seen": 1233534720,
      "step": 16600,
      "train_runtime": 19715.949,
      "train_tokens_per_second": 62565.323
    },
    {
      "epoch": 0.22072943036923343,
      "grad_norm": 29.625,
      "learning_rate": 1.8152347403483522e-05,
      "loss": 0.8807,
      "num_input_tokens_seen": 1235481088,
      "step": 16610,
      "train_runtime": 19747.3959,
      "train_tokens_per_second": 62564.254
    },
    {
      "epoch": 0.22086231985169533,
      "grad_norm": 16.375,
      "learning_rate": 1.814985409081822e-05,
      "loss": 0.8782,
      "num_input_tokens_seen": 1236323328,
      "step": 16620,
      "train_runtime": 19759.8092,
      "train_tokens_per_second": 62567.574
    },
    {
      "epoch": 0.22099520933415726,
      "grad_norm": 36.25,
      "learning_rate": 1.814735926846512e-05,
      "loss": 0.8086,
      "num_input_tokens_seen": 1236790528,
      "step": 16630,
      "train_runtime": 19766.2536,
      "train_tokens_per_second": 62570.811
    },
    {
      "epoch": 0.22112809881661916,
      "grad_norm": 9.875,
      "learning_rate": 1.8144862936886364e-05,
      "loss": 0.6732,
      "num_input_tokens_seen": 1237017856,
      "step": 16640,
      "train_runtime": 19769.7511,
      "train_tokens_per_second": 62571.241
    },
    {
      "epoch": 0.22126098829908106,
      "grad_norm": 16.25,
      "learning_rate": 1.8142365096544373e-05,
      "loss": 0.5653,
      "num_input_tokens_seen": 1237135872,
      "step": 16650,
      "train_runtime": 19772.1482,
      "train_tokens_per_second": 62569.624
    },
    {
      "epoch": 0.22139387778154299,
      "grad_norm": 33.0,
      "learning_rate": 1.8139865747901852e-05,
      "loss": 0.8339,
      "num_input_tokens_seen": 1239290880,
      "step": 16660,
      "train_runtime": 19806.7466,
      "train_tokens_per_second": 62569.129
    },
    {
      "epoch": 0.2215267672640049,
      "grad_norm": 12.125,
      "learning_rate": 1.8137364891421785e-05,
      "loss": 0.8536,
      "num_input_tokens_seen": 1240139008,
      "step": 16670,
      "train_runtime": 19819.2175,
      "train_tokens_per_second": 62572.552
    },
    {
      "epoch": 0.2216596567464668,
      "grad_norm": 15.375,
      "learning_rate": 1.8134862527567428e-05,
      "loss": 0.6955,
      "num_input_tokens_seen": 1240582656,
      "step": 16680,
      "train_runtime": 19825.6142,
      "train_tokens_per_second": 62574.74
    },
    {
      "epoch": 0.22179254622892872,
      "grad_norm": 14.375,
      "learning_rate": 1.813235865680233e-05,
      "loss": 0.6422,
      "num_input_tokens_seen": 1240794880,
      "step": 16690,
      "train_runtime": 19828.9204,
      "train_tokens_per_second": 62575.009
    },
    {
      "epoch": 0.22192543571139062,
      "grad_norm": 56.75,
      "learning_rate": 1.81298532795903e-05,
      "loss": 0.4882,
      "num_input_tokens_seen": 1240905984,
      "step": 16700,
      "train_runtime": 19831.2671,
      "train_tokens_per_second": 62573.207
    },
    {
      "epoch": 0.22205832519385255,
      "grad_norm": 104.5,
      "learning_rate": 1.812734639639545e-05,
      "loss": 0.7981,
      "num_input_tokens_seen": 1243031296,
      "step": 16710,
      "train_runtime": 19865.3235,
      "train_tokens_per_second": 62572.92
    },
    {
      "epoch": 0.22219121467631445,
      "grad_norm": 17.0,
      "learning_rate": 1.8124838007682145e-05,
      "loss": 0.9257,
      "num_input_tokens_seen": 1243862528,
      "step": 16720,
      "train_runtime": 19878.3537,
      "train_tokens_per_second": 62573.72
    },
    {
      "epoch": 0.22232410415877635,
      "grad_norm": 18.625,
      "learning_rate": 1.8122328113915046e-05,
      "loss": 0.7415,
      "num_input_tokens_seen": 1244326912,
      "step": 16730,
      "train_runtime": 19885.1743,
      "train_tokens_per_second": 62575.61
    },
    {
      "epoch": 0.22245699364123828,
      "grad_norm": 17.375,
      "learning_rate": 1.811981671555909e-05,
      "loss": 0.6161,
      "num_input_tokens_seen": 1244559104,
      "step": 16740,
      "train_runtime": 19888.6735,
      "train_tokens_per_second": 62576.275
    },
    {
      "epoch": 0.22258988312370018,
      "grad_norm": 11.0625,
      "learning_rate": 1.811730381307949e-05,
      "loss": 0.5149,
      "num_input_tokens_seen": 1244674560,
      "step": 16750,
      "train_runtime": 19891.0819,
      "train_tokens_per_second": 62574.503
    },
    {
      "epoch": 0.22272277260616208,
      "grad_norm": 24.625,
      "learning_rate": 1.8114789406941744e-05,
      "loss": 0.8453,
      "num_input_tokens_seen": 1246705408,
      "step": 16760,
      "train_runtime": 19925.0831,
      "train_tokens_per_second": 62569.647
    },
    {
      "epoch": 0.222855662088624,
      "grad_norm": 12.875,
      "learning_rate": 1.8112273497611618e-05,
      "loss": 0.8627,
      "num_input_tokens_seen": 1247586048,
      "step": 16770,
      "train_runtime": 19938.1989,
      "train_tokens_per_second": 62572.655
    },
    {
      "epoch": 0.2229885515710859,
      "grad_norm": 12.1875,
      "learning_rate": 1.810975608555516e-05,
      "loss": 0.6886,
      "num_input_tokens_seen": 1248082688,
      "step": 16780,
      "train_runtime": 19945.0009,
      "train_tokens_per_second": 62576.216
    },
    {
      "epoch": 0.2231214410535478,
      "grad_norm": 22.625,
      "learning_rate": 1.8107237171238698e-05,
      "loss": 0.633,
      "num_input_tokens_seen": 1248314624,
      "step": 16790,
      "train_runtime": 19948.5124,
      "train_tokens_per_second": 62576.828
    },
    {
      "epoch": 0.22325433053600974,
      "grad_norm": 19.75,
      "learning_rate": 1.8104716755128846e-05,
      "loss": 0.5663,
      "num_input_tokens_seen": 1248433664,
      "step": 16800,
      "train_runtime": 19950.8823,
      "train_tokens_per_second": 62575.361
    },
    {
      "epoch": 0.22338722001847164,
      "grad_norm": 43.0,
      "learning_rate": 1.810219483769248e-05,
      "loss": 0.8845,
      "num_input_tokens_seen": 1250378752,
      "step": 16810,
      "train_runtime": 19983.05,
      "train_tokens_per_second": 62571.967
    },
    {
      "epoch": 0.22352010950093354,
      "grad_norm": 25.375,
      "learning_rate": 1.8099671419396766e-05,
      "loss": 0.9087,
      "num_input_tokens_seen": 1251163904,
      "step": 16820,
      "train_runtime": 19994.9999,
      "train_tokens_per_second": 62573.839
    },
    {
      "epoch": 0.22365299898339547,
      "grad_norm": 16.25,
      "learning_rate": 1.8097146500709147e-05,
      "loss": 0.8263,
      "num_input_tokens_seen": 1251598080,
      "step": 16830,
      "train_runtime": 20001.736,
      "train_tokens_per_second": 62574.472
    },
    {
      "epoch": 0.22378588846585737,
      "grad_norm": 27.0,
      "learning_rate": 1.809462008209734e-05,
      "loss": 0.6517,
      "num_input_tokens_seen": 1251835136,
      "step": 16840,
      "train_runtime": 20005.2338,
      "train_tokens_per_second": 62575.381
    },
    {
      "epoch": 0.22391877794831927,
      "grad_norm": 15.25,
      "learning_rate": 1.809209216402934e-05,
      "loss": 0.5471,
      "num_input_tokens_seen": 1251949568,
      "step": 16850,
      "train_runtime": 20007.6939,
      "train_tokens_per_second": 62573.407
    },
    {
      "epoch": 0.2240516674307812,
      "grad_norm": 19.0,
      "learning_rate": 1.8089562746973423e-05,
      "loss": 0.8276,
      "num_input_tokens_seen": 1254003200,
      "step": 16860,
      "train_runtime": 20041.4213,
      "train_tokens_per_second": 62570.572
    },
    {
      "epoch": 0.2241845569132431,
      "grad_norm": 16.625,
      "learning_rate": 1.808703183139814e-05,
      "loss": 0.9617,
      "num_input_tokens_seen": 1254870528,
      "step": 16870,
      "train_runtime": 20054.9451,
      "train_tokens_per_second": 62571.626
    },
    {
      "epoch": 0.224317446395705,
      "grad_norm": 19.25,
      "learning_rate": 1.808449941777232e-05,
      "loss": 0.7416,
      "num_input_tokens_seen": 1255357952,
      "step": 16880,
      "train_runtime": 20061.821,
      "train_tokens_per_second": 62574.477
    },
    {
      "epoch": 0.22445033587816693,
      "grad_norm": 13.8125,
      "learning_rate": 1.8081965506565074e-05,
      "loss": 0.6305,
      "num_input_tokens_seen": 1255589888,
      "step": 16890,
      "train_runtime": 20065.3032,
      "train_tokens_per_second": 62575.177
    },
    {
      "epoch": 0.22458322536062883,
      "grad_norm": 10.875,
      "learning_rate": 1.807943009824578e-05,
      "loss": 0.5646,
      "num_input_tokens_seen": 1255706368,
      "step": 16900,
      "train_runtime": 20067.6945,
      "train_tokens_per_second": 62573.524
    },
    {
      "epoch": 0.22471611484309073,
      "grad_norm": 21.75,
      "learning_rate": 1.8076893193284103e-05,
      "loss": 0.8652,
      "num_input_tokens_seen": 1257915648,
      "step": 16910,
      "train_runtime": 20102.5199,
      "train_tokens_per_second": 62575.023
    },
    {
      "epoch": 0.22484900432555266,
      "grad_norm": 15.125,
      "learning_rate": 1.8074354792149985e-05,
      "loss": 0.9848,
      "num_input_tokens_seen": 1258808320,
      "step": 16920,
      "train_runtime": 20115.4099,
      "train_tokens_per_second": 62579.302
    },
    {
      "epoch": 0.22498189380801456,
      "grad_norm": 15.3125,
      "learning_rate": 1.807181489531364e-05,
      "loss": 0.7788,
      "num_input_tokens_seen": 1259285504,
      "step": 16930,
      "train_runtime": 20122.1969,
      "train_tokens_per_second": 62581.91
    },
    {
      "epoch": 0.2251147832904765,
      "grad_norm": 14.75,
      "learning_rate": 1.806927350324556e-05,
      "loss": 0.6931,
      "num_input_tokens_seen": 1259517696,
      "step": 16940,
      "train_runtime": 20126.2259,
      "train_tokens_per_second": 62580.918
    },
    {
      "epoch": 0.2252476727729384,
      "grad_norm": 19.25,
      "learning_rate": 1.8066730616416514e-05,
      "loss": 0.5201,
      "num_input_tokens_seen": 1259637248,
      "step": 16950,
      "train_runtime": 20128.5854,
      "train_tokens_per_second": 62579.522
    },
    {
      "epoch": 0.2253805622554003,
      "grad_norm": 23.5,
      "learning_rate": 1.8064186235297554e-05,
      "loss": 0.8274,
      "num_input_tokens_seen": 1261667072,
      "step": 16960,
      "train_runtime": 20161.8826,
      "train_tokens_per_second": 62576.848
    },
    {
      "epoch": 0.22551345173786222,
      "grad_norm": 13.0625,
      "learning_rate": 1.8061640360359996e-05,
      "loss": 0.8904,
      "num_input_tokens_seen": 1262490880,
      "step": 16970,
      "train_runtime": 20174.4249,
      "train_tokens_per_second": 62578.779
    },
    {
      "epoch": 0.22564634122032412,
      "grad_norm": 13.0625,
      "learning_rate": 1.805909299207545e-05,
      "loss": 0.759,
      "num_input_tokens_seen": 1262932992,
      "step": 16980,
      "train_runtime": 20180.8522,
      "train_tokens_per_second": 62580.756
    },
    {
      "epoch": 0.22577923070278602,
      "grad_norm": 11.0,
      "learning_rate": 1.8056544130915785e-05,
      "loss": 0.6832,
      "num_input_tokens_seen": 1263152384,
      "step": 16990,
      "train_runtime": 20184.6879,
      "train_tokens_per_second": 62579.733
    },
    {
      "epoch": 0.22591212018524795,
      "grad_norm": 14.8125,
      "learning_rate": 1.8053993777353162e-05,
      "loss": 0.4727,
      "num_input_tokens_seen": 1263271680,
      "step": 17000,
      "train_runtime": 20187.0205,
      "train_tokens_per_second": 62578.412
    },
    {
      "epoch": 0.22604500966770985,
      "grad_norm": 16.125,
      "learning_rate": 1.805144193186001e-05,
      "loss": 0.8426,
      "num_input_tokens_seen": 1265326336,
      "step": 17010,
      "train_runtime": 20219.4803,
      "train_tokens_per_second": 62579.568
    },
    {
      "epoch": 0.22617789915017175,
      "grad_norm": 26.25,
      "learning_rate": 1.804888859490903e-05,
      "loss": 0.8752,
      "num_input_tokens_seen": 1266163456,
      "step": 17020,
      "train_runtime": 20231.7645,
      "train_tokens_per_second": 62582.947
    },
    {
      "epoch": 0.22631078863263368,
      "grad_norm": 21.75,
      "learning_rate": 1.8046333766973206e-05,
      "loss": 0.7641,
      "num_input_tokens_seen": 1266629632,
      "step": 17030,
      "train_runtime": 20238.2774,
      "train_tokens_per_second": 62585.842
    },
    {
      "epoch": 0.22644367811509558,
      "grad_norm": 42.5,
      "learning_rate": 1.8043777448525804e-05,
      "loss": 0.6306,
      "num_input_tokens_seen": 1266855680,
      "step": 17040,
      "train_runtime": 20242.3202,
      "train_tokens_per_second": 62584.509
    },
    {
      "epoch": 0.22657656759755748,
      "grad_norm": 15.25,
      "learning_rate": 1.8041219640040353e-05,
      "loss": 0.5154,
      "num_input_tokens_seen": 1266969600,
      "step": 17050,
      "train_runtime": 20244.7101,
      "train_tokens_per_second": 62582.749
    },
    {
      "epoch": 0.2267094570800194,
      "grad_norm": 26.75,
      "learning_rate": 1.8038660341990665e-05,
      "loss": 0.8703,
      "num_input_tokens_seen": 1268977664,
      "step": 17060,
      "train_runtime": 20277.1406,
      "train_tokens_per_second": 62581.687
    },
    {
      "epoch": 0.2268423465624813,
      "grad_norm": 13.5625,
      "learning_rate": 1.803609955485083e-05,
      "loss": 0.8939,
      "num_input_tokens_seen": 1269823232,
      "step": 17070,
      "train_runtime": 20289.6617,
      "train_tokens_per_second": 62584.741
    },
    {
      "epoch": 0.2269752360449432,
      "grad_norm": 21.0,
      "learning_rate": 1.803353727909521e-05,
      "loss": 0.7847,
      "num_input_tokens_seen": 1270300416,
      "step": 17080,
      "train_runtime": 20296.3487,
      "train_tokens_per_second": 62587.633
    },
    {
      "epoch": 0.22710812552740514,
      "grad_norm": 28.0,
      "learning_rate": 1.803097351519844e-05,
      "loss": 0.6559,
      "num_input_tokens_seen": 1270529280,
      "step": 17090,
      "train_runtime": 20299.8093,
      "train_tokens_per_second": 62588.237
    },
    {
      "epoch": 0.22724101500986704,
      "grad_norm": 7.25,
      "learning_rate": 1.8028408263635437e-05,
      "loss": 0.503,
      "num_input_tokens_seen": 1270642944,
      "step": 17100,
      "train_runtime": 20302.151,
      "train_tokens_per_second": 62586.617
    },
    {
      "epoch": 0.22737390449232894,
      "grad_norm": 15.0625,
      "learning_rate": 1.8025841524881393e-05,
      "loss": 0.7862,
      "num_input_tokens_seen": 1272587264,
      "step": 17110,
      "train_runtime": 20333.426,
      "train_tokens_per_second": 62585.974
    },
    {
      "epoch": 0.22750679397479087,
      "grad_norm": 18.75,
      "learning_rate": 1.802327329941177e-05,
      "loss": 0.9942,
      "num_input_tokens_seen": 1273417216,
      "step": 17120,
      "train_runtime": 20345.9317,
      "train_tokens_per_second": 62588.297
    },
    {
      "epoch": 0.22763968345725277,
      "grad_norm": 20.375,
      "learning_rate": 1.802070358770231e-05,
      "loss": 0.816,
      "num_input_tokens_seen": 1273858048,
      "step": 17130,
      "train_runtime": 20352.2816,
      "train_tokens_per_second": 62590.43
    },
    {
      "epoch": 0.22777257293971467,
      "grad_norm": 10.9375,
      "learning_rate": 1.801813239022903e-05,
      "loss": 0.5996,
      "num_input_tokens_seen": 1274072832,
      "step": 17140,
      "train_runtime": 20355.66,
      "train_tokens_per_second": 62590.593
    },
    {
      "epoch": 0.2279054624221766,
      "grad_norm": 9.0625,
      "learning_rate": 1.8015559707468225e-05,
      "loss": 0.565,
      "num_input_tokens_seen": 1274189568,
      "step": 17150,
      "train_runtime": 20358.603,
      "train_tokens_per_second": 62587.279
    },
    {
      "epoch": 0.2280383519046385,
      "grad_norm": 13.8125,
      "learning_rate": 1.8012985539896453e-05,
      "loss": 0.9368,
      "num_input_tokens_seen": 1276263424,
      "step": 17160,
      "train_runtime": 20392.0241,
      "train_tokens_per_second": 62586.402
    },
    {
      "epoch": 0.2281712413871004,
      "grad_norm": 43.75,
      "learning_rate": 1.8010409887990562e-05,
      "loss": 0.9043,
      "num_input_tokens_seen": 1277089792,
      "step": 17170,
      "train_runtime": 20404.483,
      "train_tokens_per_second": 62588.687
    },
    {
      "epoch": 0.22830413086956233,
      "grad_norm": 10.75,
      "learning_rate": 1.8007832752227668e-05,
      "loss": 0.7476,
      "num_input_tokens_seen": 1277571840,
      "step": 17180,
      "train_runtime": 20411.0361,
      "train_tokens_per_second": 62592.209
    },
    {
      "epoch": 0.22843702035202423,
      "grad_norm": 17.5,
      "learning_rate": 1.800525413308516e-05,
      "loss": 0.6982,
      "num_input_tokens_seen": 1277795584,
      "step": 17190,
      "train_runtime": 20414.4769,
      "train_tokens_per_second": 62592.62
    },
    {
      "epoch": 0.22856990983448616,
      "grad_norm": 14.6875,
      "learning_rate": 1.8002674031040707e-05,
      "loss": 0.4837,
      "num_input_tokens_seen": 1277912320,
      "step": 17200,
      "train_runtime": 20416.8806,
      "train_tokens_per_second": 62590.968
    },
    {
      "epoch": 0.22870279931694806,
      "grad_norm": 21.5,
      "learning_rate": 1.8000092446572245e-05,
      "loss": 0.7857,
      "num_input_tokens_seen": 1279842560,
      "step": 17210,
      "train_runtime": 20448.2991,
      "train_tokens_per_second": 62589.194
    },
    {
      "epoch": 0.22883568879940996,
      "grad_norm": 19.25,
      "learning_rate": 1.7997509380157995e-05,
      "loss": 0.9324,
      "num_input_tokens_seen": 1280636416,
      "step": 17220,
      "train_runtime": 20460.2184,
      "train_tokens_per_second": 62591.532
    },
    {
      "epoch": 0.2289685782818719,
      "grad_norm": 15.375,
      "learning_rate": 1.7994924832276442e-05,
      "loss": 0.7429,
      "num_input_tokens_seen": 1281100800,
      "step": 17230,
      "train_runtime": 20466.9911,
      "train_tokens_per_second": 62593.51
    },
    {
      "epoch": 0.2291014677643338,
      "grad_norm": 15.375,
      "learning_rate": 1.7992338803406354e-05,
      "loss": 0.7037,
      "num_input_tokens_seen": 1281329152,
      "step": 17240,
      "train_runtime": 20470.371,
      "train_tokens_per_second": 62594.33
    },
    {
      "epoch": 0.2292343572467957,
      "grad_norm": 5.6875,
      "learning_rate": 1.798975129402677e-05,
      "loss": 0.5349,
      "num_input_tokens_seen": 1281444352,
      "step": 17250,
      "train_runtime": 20472.7573,
      "train_tokens_per_second": 62592.661
    },
    {
      "epoch": 0.22936724672925762,
      "grad_norm": 13.1875,
      "learning_rate": 1.7987162304617004e-05,
      "loss": 0.8135,
      "num_input_tokens_seen": 1283522560,
      "step": 17260,
      "train_runtime": 20506.9982,
      "train_tokens_per_second": 62589.49
    },
    {
      "epoch": 0.22950013621171952,
      "grad_norm": 22.375,
      "learning_rate": 1.7984571835656642e-05,
      "loss": 0.9689,
      "num_input_tokens_seen": 1284353792,
      "step": 17270,
      "train_runtime": 20519.4624,
      "train_tokens_per_second": 62591.981
    },
    {
      "epoch": 0.22963302569418143,
      "grad_norm": 13.0,
      "learning_rate": 1.7981979887625542e-05,
      "loss": 0.7954,
      "num_input_tokens_seen": 1284838144,
      "step": 17280,
      "train_runtime": 20526.2649,
      "train_tokens_per_second": 62594.834
    },
    {
      "epoch": 0.22976591517664335,
      "grad_norm": 25.0,
      "learning_rate": 1.797938646100384e-05,
      "loss": 0.6279,
      "num_input_tokens_seen": 1285071872,
      "step": 17290,
      "train_runtime": 20530.2626,
      "train_tokens_per_second": 62594.03
    },
    {
      "epoch": 0.22989880465910525,
      "grad_norm": 17.75,
      "learning_rate": 1.797679155627195e-05,
      "loss": 0.5101,
      "num_input_tokens_seen": 1285190656,
      "step": 17300,
      "train_runtime": 20532.6172,
      "train_tokens_per_second": 62592.637
    },
    {
      "epoch": 0.23003169414156716,
      "grad_norm": 27.25,
      "learning_rate": 1.7974195173910553e-05,
      "loss": 0.8567,
      "num_input_tokens_seen": 1287160576,
      "step": 17310,
      "train_runtime": 20563.562,
      "train_tokens_per_second": 62594.242
    },
    {
      "epoch": 0.23016458362402908,
      "grad_norm": 43.75,
      "learning_rate": 1.79715973144006e-05,
      "loss": 0.9421,
      "num_input_tokens_seen": 1287960576,
      "step": 17320,
      "train_runtime": 20575.6419,
      "train_tokens_per_second": 62596.374
    },
    {
      "epoch": 0.23029747310649099,
      "grad_norm": 15.8125,
      "learning_rate": 1.7968997978223326e-05,
      "loss": 0.7244,
      "num_input_tokens_seen": 1288384000,
      "step": 17330,
      "train_runtime": 20582.3283,
      "train_tokens_per_second": 62596.611
    },
    {
      "epoch": 0.2304303625889529,
      "grad_norm": 10.6875,
      "learning_rate": 1.796639716586024e-05,
      "loss": 0.6561,
      "num_input_tokens_seen": 1288601600,
      "step": 17340,
      "train_runtime": 20585.6429,
      "train_tokens_per_second": 62597.102
    },
    {
      "epoch": 0.23056325207141481,
      "grad_norm": 5.84375,
      "learning_rate": 1.7963794877793114e-05,
      "loss": 0.5765,
      "num_input_tokens_seen": 1288711680,
      "step": 17350,
      "train_runtime": 20588.0366,
      "train_tokens_per_second": 62595.171
    },
    {
      "epoch": 0.23069614155387672,
      "grad_norm": 13.1875,
      "learning_rate": 1.7961191114503993e-05,
      "loss": 0.7959,
      "num_input_tokens_seen": 1290922496,
      "step": 17360,
      "train_runtime": 20623.0759,
      "train_tokens_per_second": 62596.021
    },
    {
      "epoch": 0.23082903103633862,
      "grad_norm": 342.0,
      "learning_rate": 1.795858587647521e-05,
      "loss": 0.9182,
      "num_input_tokens_seen": 1291749888,
      "step": 17370,
      "train_runtime": 20635.704,
      "train_tokens_per_second": 62597.811
    },
    {
      "epoch": 0.23096192051880055,
      "grad_norm": 17.0,
      "learning_rate": 1.795597916418936e-05,
      "loss": 0.7797,
      "num_input_tokens_seen": 1292205312,
      "step": 17380,
      "train_runtime": 20642.3398,
      "train_tokens_per_second": 62599.75
    },
    {
      "epoch": 0.23109481000126245,
      "grad_norm": 16.875,
      "learning_rate": 1.795337097812931e-05,
      "loss": 0.6154,
      "num_input_tokens_seen": 1292441088,
      "step": 17390,
      "train_runtime": 20646.2874,
      "train_tokens_per_second": 62599.201
    },
    {
      "epoch": 0.23122769948372435,
      "grad_norm": 19.375,
      "learning_rate": 1.7950761318778205e-05,
      "loss": 0.505,
      "num_input_tokens_seen": 1292557824,
      "step": 17400,
      "train_runtime": 20648.6434,
      "train_tokens_per_second": 62597.711
    },
    {
      "epoch": 0.23136058896618628,
      "grad_norm": 34.0,
      "learning_rate": 1.7948150186619463e-05,
      "loss": 0.8271,
      "num_input_tokens_seen": 1294567168,
      "step": 17410,
      "train_runtime": 20681.6601,
      "train_tokens_per_second": 62594.935
    },
    {
      "epoch": 0.23149347844864818,
      "grad_norm": 20.625,
      "learning_rate": 1.794553758213677e-05,
      "loss": 0.8983,
      "num_input_tokens_seen": 1295403776,
      "step": 17420,
      "train_runtime": 20693.8209,
      "train_tokens_per_second": 62598.579
    },
    {
      "epoch": 0.2316263679311101,
      "grad_norm": 30.5,
      "learning_rate": 1.794292350581409e-05,
      "loss": 0.7794,
      "num_input_tokens_seen": 1295862784,
      "step": 17430,
      "train_runtime": 20700.7749,
      "train_tokens_per_second": 62599.723
    },
    {
      "epoch": 0.231759257413572,
      "grad_norm": 12.375,
      "learning_rate": 1.7940307958135654e-05,
      "loss": 0.6597,
      "num_input_tokens_seen": 1296088576,
      "step": 17440,
      "train_runtime": 20704.2008,
      "train_tokens_per_second": 62600.271
    },
    {
      "epoch": 0.2318921468960339,
      "grad_norm": 9.8125,
      "learning_rate": 1.7937690939585973e-05,
      "loss": 0.4805,
      "num_input_tokens_seen": 1296206336,
      "step": 17450,
      "train_runtime": 20706.5762,
      "train_tokens_per_second": 62598.777
    },
    {
      "epoch": 0.23202503637849584,
      "grad_norm": 25.875,
      "learning_rate": 1.7935072450649826e-05,
      "loss": 0.8398,
      "num_input_tokens_seen": 1298278144,
      "step": 17460,
      "train_runtime": 20740.6704,
      "train_tokens_per_second": 62595.766
    },
    {
      "epoch": 0.23215792586095774,
      "grad_norm": 19.125,
      "learning_rate": 1.793245249181226e-05,
      "loss": 0.9332,
      "num_input_tokens_seen": 1299136000,
      "step": 17470,
      "train_runtime": 20753.5567,
      "train_tokens_per_second": 62598.234
    },
    {
      "epoch": 0.23229081534341964,
      "grad_norm": 33.5,
      "learning_rate": 1.79298310635586e-05,
      "loss": 0.7584,
      "num_input_tokens_seen": 1299597312,
      "step": 17480,
      "train_runtime": 20760.2357,
      "train_tokens_per_second": 62600.316
    },
    {
      "epoch": 0.23242370482588157,
      "grad_norm": 20.875,
      "learning_rate": 1.7927208166374454e-05,
      "loss": 0.634,
      "num_input_tokens_seen": 1299821824,
      "step": 17490,
      "train_runtime": 20763.6448,
      "train_tokens_per_second": 62600.851
    },
    {
      "epoch": 0.23255659430834347,
      "grad_norm": 22.875,
      "learning_rate": 1.7924583800745674e-05,
      "loss": 0.6037,
      "num_input_tokens_seen": 1299936256,
      "step": 17500,
      "train_runtime": 20766.4562,
      "train_tokens_per_second": 62597.886
    },
    {
      "epoch": 0.23268948379080537,
      "grad_norm": 66.5,
      "learning_rate": 1.792195796715841e-05,
      "loss": 0.829,
      "num_input_tokens_seen": 1301920000,
      "step": 17510,
      "train_runtime": 20799.1805,
      "train_tokens_per_second": 62594.774
    },
    {
      "epoch": 0.2328223732732673,
      "grad_norm": 18.625,
      "learning_rate": 1.7919330666099068e-05,
      "loss": 0.8481,
      "num_input_tokens_seen": 1302740992,
      "step": 17520,
      "train_runtime": 20811.399,
      "train_tokens_per_second": 62597.473
    },
    {
      "epoch": 0.2329552627557292,
      "grad_norm": 16.5,
      "learning_rate": 1.7916701898054336e-05,
      "loss": 0.8345,
      "num_input_tokens_seen": 1303177728,
      "step": 17530,
      "train_runtime": 20817.6079,
      "train_tokens_per_second": 62599.783
    },
    {
      "epoch": 0.2330881522381911,
      "grad_norm": 11.625,
      "learning_rate": 1.791407166351117e-05,
      "loss": 0.6508,
      "num_input_tokens_seen": 1303392512,
      "step": 17540,
      "train_runtime": 20820.8866,
      "train_tokens_per_second": 62600.241
    },
    {
      "epoch": 0.23322104172065303,
      "grad_norm": 11.5,
      "learning_rate": 1.79114399629568e-05,
      "loss": 0.4318,
      "num_input_tokens_seen": 1303503104,
      "step": 17550,
      "train_runtime": 20824.0676,
      "train_tokens_per_second": 62595.989
    },
    {
      "epoch": 0.23335393120311493,
      "grad_norm": 16.25,
      "learning_rate": 1.7908806796878717e-05,
      "loss": 0.9238,
      "num_input_tokens_seen": 1305458688,
      "step": 17560,
      "train_runtime": 20857.1866,
      "train_tokens_per_second": 62590.354
    },
    {
      "epoch": 0.23348682068557683,
      "grad_norm": 17.375,
      "learning_rate": 1.7906172165764698e-05,
      "loss": 0.9793,
      "num_input_tokens_seen": 1306289920,
      "step": 17570,
      "train_runtime": 20869.6512,
      "train_tokens_per_second": 62592.801
    },
    {
      "epoch": 0.23361971016803876,
      "grad_norm": 11.5625,
      "learning_rate": 1.7903536070102785e-05,
      "loss": 0.6904,
      "num_input_tokens_seen": 1306729728,
      "step": 17580,
      "train_runtime": 20876.1358,
      "train_tokens_per_second": 62594.425
    },
    {
      "epoch": 0.23375259965050066,
      "grad_norm": 21.125,
      "learning_rate": 1.790089851038129e-05,
      "loss": 0.6611,
      "num_input_tokens_seen": 1306949376,
      "step": 17590,
      "train_runtime": 20879.5081,
      "train_tokens_per_second": 62594.836
    },
    {
      "epoch": 0.23388548913296256,
      "grad_norm": 22.0,
      "learning_rate": 1.7898259487088792e-05,
      "loss": 0.5374,
      "num_input_tokens_seen": 1307059712,
      "step": 17600,
      "train_runtime": 20881.8291,
      "train_tokens_per_second": 62593.162
    },
    {
      "epoch": 0.2340183786154245,
      "grad_norm": 22.0,
      "learning_rate": 1.7895619000714158e-05,
      "loss": 0.7769,
      "num_input_tokens_seen": 1309045504,
      "step": 17610,
      "train_runtime": 20915.2258,
      "train_tokens_per_second": 62588.16
    },
    {
      "epoch": 0.2341512680978864,
      "grad_norm": 20.0,
      "learning_rate": 1.7892977051746503e-05,
      "loss": 0.9031,
      "num_input_tokens_seen": 1309859584,
      "step": 17620,
      "train_runtime": 20927.4869,
      "train_tokens_per_second": 62590.391
    },
    {
      "epoch": 0.2342841575803483,
      "grad_norm": 22.625,
      "learning_rate": 1.7890333640675234e-05,
      "loss": 0.8226,
      "num_input_tokens_seen": 1310339328,
      "step": 17630,
      "train_runtime": 20934.0474,
      "train_tokens_per_second": 62593.693
    },
    {
      "epoch": 0.23441704706281022,
      "grad_norm": 16.75,
      "learning_rate": 1.788768876799001e-05,
      "loss": 0.7046,
      "num_input_tokens_seen": 1310566912,
      "step": 17640,
      "train_runtime": 20937.4345,
      "train_tokens_per_second": 62594.437
    },
    {
      "epoch": 0.23454993654527212,
      "grad_norm": 7.0625,
      "learning_rate": 1.7885042434180778e-05,
      "loss": 0.5393,
      "num_input_tokens_seen": 1310683392,
      "step": 17650,
      "train_runtime": 20939.8024,
      "train_tokens_per_second": 62592.921
    },
    {
      "epoch": 0.23468282602773405,
      "grad_norm": 69.0,
      "learning_rate": 1.788239463973774e-05,
      "loss": 0.8289,
      "num_input_tokens_seen": 1312760320,
      "step": 17660,
      "train_runtime": 20973.6209,
      "train_tokens_per_second": 62591.02
    },
    {
      "epoch": 0.23481571551019595,
      "grad_norm": 32.25,
      "learning_rate": 1.7879745385151388e-05,
      "loss": 0.9076,
      "num_input_tokens_seen": 1313600512,
      "step": 17670,
      "train_runtime": 20987.1053,
      "train_tokens_per_second": 62590.838
    },
    {
      "epoch": 0.23494860499265785,
      "grad_norm": 16.0,
      "learning_rate": 1.7877094670912462e-05,
      "loss": 0.7369,
      "num_input_tokens_seen": 1314089216,
      "step": 17680,
      "train_runtime": 20994.1729,
      "train_tokens_per_second": 62593.045
    },
    {
      "epoch": 0.23508149447511978,
      "grad_norm": 22.875,
      "learning_rate": 1.787444249751199e-05,
      "loss": 0.724,
      "num_input_tokens_seen": 1314334976,
      "step": 17690,
      "train_runtime": 20997.8168,
      "train_tokens_per_second": 62593.887
    },
    {
      "epoch": 0.23521438395758168,
      "grad_norm": 8.9375,
      "learning_rate": 1.7871788865441258e-05,
      "loss": 0.5458,
      "num_input_tokens_seen": 1314455296,
      "step": 17700,
      "train_runtime": 21000.1819,
      "train_tokens_per_second": 62592.567
    },
    {
      "epoch": 0.23534727344004358,
      "grad_norm": 26.5,
      "learning_rate": 1.786913377519183e-05,
      "loss": 0.8464,
      "num_input_tokens_seen": 1316624128,
      "step": 17710,
      "train_runtime": 21033.7388,
      "train_tokens_per_second": 62595.82
    },
    {
      "epoch": 0.2354801629225055,
      "grad_norm": 21.0,
      "learning_rate": 1.7866477227255538e-05,
      "loss": 0.975,
      "num_input_tokens_seen": 1317481216,
      "step": 17720,
      "train_runtime": 21047.0235,
      "train_tokens_per_second": 62597.033
    },
    {
      "epoch": 0.2356130524049674,
      "grad_norm": 66.0,
      "learning_rate": 1.7863819222124482e-05,
      "loss": 0.6959,
      "num_input_tokens_seen": 1317974528,
      "step": 17730,
      "train_runtime": 21054.0989,
      "train_tokens_per_second": 62599.427
    },
    {
      "epoch": 0.2357459418874293,
      "grad_norm": 13.625,
      "learning_rate": 1.786115976029104e-05,
      "loss": 0.673,
      "num_input_tokens_seen": 1318211328,
      "step": 17740,
      "train_runtime": 21057.6599,
      "train_tokens_per_second": 62600.086
    },
    {
      "epoch": 0.23587883136989124,
      "grad_norm": 29.5,
      "learning_rate": 1.7858498842247844e-05,
      "loss": 0.5471,
      "num_input_tokens_seen": 1318326528,
      "step": 17750,
      "train_runtime": 21060.0193,
      "train_tokens_per_second": 62598.543
    },
    {
      "epoch": 0.23601172085235314,
      "grad_norm": 19.125,
      "learning_rate": 1.7855836468487812e-05,
      "loss": 0.9139,
      "num_input_tokens_seen": 1320381184,
      "step": 17760,
      "train_runtime": 21091.6244,
      "train_tokens_per_second": 62602.157
    },
    {
      "epoch": 0.23614461033481504,
      "grad_norm": 17.875,
      "learning_rate": 1.7853172639504126e-05,
      "loss": 0.9061,
      "num_input_tokens_seen": 1321209856,
      "step": 17770,
      "train_runtime": 21103.9251,
      "train_tokens_per_second": 62604.935
    },
    {
      "epoch": 0.23627749981727697,
      "grad_norm": 20.875,
      "learning_rate": 1.7850507355790234e-05,
      "loss": 0.7782,
      "num_input_tokens_seen": 1321657344,
      "step": 17780,
      "train_runtime": 21110.7589,
      "train_tokens_per_second": 62605.866
    },
    {
      "epoch": 0.23641038929973887,
      "grad_norm": 12.5625,
      "learning_rate": 1.7847840617839853e-05,
      "loss": 0.6461,
      "num_input_tokens_seen": 1321892096,
      "step": 17790,
      "train_runtime": 21114.1861,
      "train_tokens_per_second": 62606.822
    },
    {
      "epoch": 0.23654327878220077,
      "grad_norm": 11.0,
      "learning_rate": 1.7845172426146975e-05,
      "loss": 0.5699,
      "num_input_tokens_seen": 1322012160,
      "step": 17800,
      "train_runtime": 21116.5451,
      "train_tokens_per_second": 62605.514
    },
    {
      "epoch": 0.2366761682646627,
      "grad_norm": 63.5,
      "learning_rate": 1.7842502781205863e-05,
      "loss": 0.7458,
      "num_input_tokens_seen": 1324142848,
      "step": 17810,
      "train_runtime": 21151.0525,
      "train_tokens_per_second": 62604.111
    },
    {
      "epoch": 0.2368090577471246,
      "grad_norm": 11.25,
      "learning_rate": 1.783983168351104e-05,
      "loss": 0.9535,
      "num_input_tokens_seen": 1324976384,
      "step": 17820,
      "train_runtime": 21163.5164,
      "train_tokens_per_second": 62606.627
    },
    {
      "epoch": 0.2369419472295865,
      "grad_norm": 31.875,
      "learning_rate": 1.7837159133557303e-05,
      "loss": 0.7777,
      "num_input_tokens_seen": 1325430528,
      "step": 17830,
      "train_runtime": 21170.1688,
      "train_tokens_per_second": 62608.406
    },
    {
      "epoch": 0.23707483671204843,
      "grad_norm": 17.25,
      "learning_rate": 1.783448513183972e-05,
      "loss": 0.6918,
      "num_input_tokens_seen": 1325661952,
      "step": 17840,
      "train_runtime": 21174.0412,
      "train_tokens_per_second": 62607.886
    },
    {
      "epoch": 0.23720772619451033,
      "grad_norm": 21.75,
      "learning_rate": 1.7831809678853623e-05,
      "loss": 0.5696,
      "num_input_tokens_seen": 1325778944,
      "step": 17850,
      "train_runtime": 21176.4743,
      "train_tokens_per_second": 62606.217
    },
    {
      "epoch": 0.23734061567697223,
      "grad_norm": 148.0,
      "learning_rate": 1.782913277509462e-05,
      "loss": 0.7541,
      "num_input_tokens_seen": 1327861504,
      "step": 17860,
      "train_runtime": 21210.2492,
      "train_tokens_per_second": 62604.71
    },
    {
      "epoch": 0.23747350515943416,
      "grad_norm": 19.875,
      "learning_rate": 1.7826454421058577e-05,
      "loss": 0.9008,
      "num_input_tokens_seen": 1328699904,
      "step": 17870,
      "train_runtime": 21222.6223,
      "train_tokens_per_second": 62607.716
    },
    {
      "epoch": 0.23760639464189606,
      "grad_norm": 17.25,
      "learning_rate": 1.7823774617241646e-05,
      "loss": 0.8167,
      "num_input_tokens_seen": 1329159424,
      "step": 17880,
      "train_runtime": 21229.2201,
      "train_tokens_per_second": 62609.904
    },
    {
      "epoch": 0.237739284124358,
      "grad_norm": 10.0,
      "learning_rate": 1.782109336414023e-05,
      "loss": 0.6556,
      "num_input_tokens_seen": 1329399040,
      "step": 17890,
      "train_runtime": 21233.2053,
      "train_tokens_per_second": 62609.437
    },
    {
      "epoch": 0.2378721736068199,
      "grad_norm": 8.0625,
      "learning_rate": 1.7818410662251005e-05,
      "loss": 0.4969,
      "num_input_tokens_seen": 1329515776,
      "step": 17900,
      "train_runtime": 21235.6291,
      "train_tokens_per_second": 62607.789
    },
    {
      "epoch": 0.2380050630892818,
      "grad_norm": 15.9375,
      "learning_rate": 1.781572651207092e-05,
      "loss": 0.8557,
      "num_input_tokens_seen": 1331563264,
      "step": 17910,
      "train_runtime": 21268.6851,
      "train_tokens_per_second": 62606.751
    },
    {
      "epoch": 0.23813795257174372,
      "grad_norm": 12.5625,
      "learning_rate": 1.781304091409719e-05,
      "loss": 0.8606,
      "num_input_tokens_seen": 1332426240,
      "step": 17920,
      "train_runtime": 21281.2774,
      "train_tokens_per_second": 62610.257
    },
    {
      "epoch": 0.23827084205420562,
      "grad_norm": 13.125,
      "learning_rate": 1.78103538688273e-05,
      "loss": 0.7069,
      "num_input_tokens_seen": 1332889088,
      "step": 17930,
      "train_runtime": 21288.4216,
      "train_tokens_per_second": 62610.987
    },
    {
      "epoch": 0.23840373153666752,
      "grad_norm": 13.0,
      "learning_rate": 1.7807665376759e-05,
      "loss": 0.6802,
      "num_input_tokens_seen": 1333118464,
      "step": 17940,
      "train_runtime": 21291.9193,
      "train_tokens_per_second": 62611.475
    },
    {
      "epoch": 0.23853662101912945,
      "grad_norm": 41.5,
      "learning_rate": 1.780497543839031e-05,
      "loss": 0.4205,
      "num_input_tokens_seen": 1333233408,
      "step": 17950,
      "train_runtime": 21294.3588,
      "train_tokens_per_second": 62609.7
    },
    {
      "epoch": 0.23866951050159135,
      "grad_norm": 21.0,
      "learning_rate": 1.780228405421951e-05,
      "loss": 0.8504,
      "num_input_tokens_seen": 1335311360,
      "step": 17960,
      "train_runtime": 21326.7446,
      "train_tokens_per_second": 62612.058
    },
    {
      "epoch": 0.23880239998405325,
      "grad_norm": 17.25,
      "learning_rate": 1.7799591224745166e-05,
      "loss": 0.8859,
      "num_input_tokens_seen": 1336169216,
      "step": 17970,
      "train_runtime": 21339.4128,
      "train_tokens_per_second": 62615.088
    },
    {
      "epoch": 0.23893528946651518,
      "grad_norm": 24.625,
      "learning_rate": 1.779689695046609e-05,
      "loss": 0.7554,
      "num_input_tokens_seen": 1336634112,
      "step": 17980,
      "train_runtime": 21346.0379,
      "train_tokens_per_second": 62617.434
    },
    {
      "epoch": 0.23906817894897708,
      "grad_norm": 16.25,
      "learning_rate": 1.7794201231881384e-05,
      "loss": 0.625,
      "num_input_tokens_seen": 1336858624,
      "step": 17990,
      "train_runtime": 21349.5056,
      "train_tokens_per_second": 62617.779
    },
    {
      "epoch": 0.23920106843143898,
      "grad_norm": 17.0,
      "learning_rate": 1.7791504069490395e-05,
      "loss": 0.5503,
      "num_input_tokens_seen": 1336979712,
      "step": 18000,
      "train_runtime": 21352.3549,
      "train_tokens_per_second": 62615.094
    },
    {
      "epoch": 0.2393339579139009,
      "grad_norm": 16.5,
      "learning_rate": 1.7788805463792757e-05,
      "loss": 0.849,
      "num_input_tokens_seen": 1338933760,
      "step": 18010,
      "train_runtime": 21383.4911,
      "train_tokens_per_second": 62615.302
    },
    {
      "epoch": 0.23946684739636281,
      "grad_norm": 14.9375,
      "learning_rate": 1.7786105415288354e-05,
      "loss": 0.9663,
      "num_input_tokens_seen": 1339720448,
      "step": 18020,
      "train_runtime": 21395.629,
      "train_tokens_per_second": 62616.549
    },
    {
      "epoch": 0.23959973687882472,
      "grad_norm": 14.375,
      "learning_rate": 1.778340392447735e-05,
      "loss": 0.751,
      "num_input_tokens_seen": 1340183296,
      "step": 18030,
      "train_runtime": 21402.1216,
      "train_tokens_per_second": 62619.18
    },
    {
      "epoch": 0.23973262636128664,
      "grad_norm": 23.5,
      "learning_rate": 1.7780700991860173e-05,
      "loss": 0.6795,
      "num_input_tokens_seen": 1340417536,
      "step": 18040,
      "train_runtime": 21405.6013,
      "train_tokens_per_second": 62619.943
    },
    {
      "epoch": 0.23986551584374854,
      "grad_norm": 18.375,
      "learning_rate": 1.777799661793752e-05,
      "loss": 0.529,
      "num_input_tokens_seen": 1340531968,
      "step": 18050,
      "train_runtime": 21408.4532,
      "train_tokens_per_second": 62616.947
    },
    {
      "epoch": 0.23999840532621045,
      "grad_norm": 64.0,
      "learning_rate": 1.7775290803210344e-05,
      "loss": 0.8633,
      "num_input_tokens_seen": 1342511104,
      "step": 18060,
      "train_runtime": 21440.2523,
      "train_tokens_per_second": 62616.386
    },
    {
      "epoch": 0.24013129480867237,
      "grad_norm": 16.375,
      "learning_rate": 1.777258354817988e-05,
      "loss": 0.8814,
      "num_input_tokens_seen": 1343341312,
      "step": 18070,
      "train_runtime": 21452.855,
      "train_tokens_per_second": 62618.3
    },
    {
      "epoch": 0.24026418429113428,
      "grad_norm": 14.25,
      "learning_rate": 1.7769874853347618e-05,
      "loss": 0.8061,
      "num_input_tokens_seen": 1343822336,
      "step": 18080,
      "train_runtime": 21459.5857,
      "train_tokens_per_second": 62621.076
    },
    {
      "epoch": 0.24039707377359618,
      "grad_norm": 10.9375,
      "learning_rate": 1.7767164719215326e-05,
      "loss": 0.6344,
      "num_input_tokens_seen": 1344053504,
      "step": 18090,
      "train_runtime": 21463.05,
      "train_tokens_per_second": 62621.739
    },
    {
      "epoch": 0.2405299632560581,
      "grad_norm": 11.25,
      "learning_rate": 1.7764453146285027e-05,
      "loss": 0.5716,
      "num_input_tokens_seen": 1344169216,
      "step": 18100,
      "train_runtime": 21465.4424,
      "train_tokens_per_second": 62620.15
    },
    {
      "epoch": 0.24066285273852,
      "grad_norm": 18.125,
      "learning_rate": 1.7761740135059016e-05,
      "loss": 0.8754,
      "num_input_tokens_seen": 1346169088,
      "step": 18110,
      "train_runtime": 21498.6576,
      "train_tokens_per_second": 62616.425
    },
    {
      "epoch": 0.24079574222098193,
      "grad_norm": 26.625,
      "learning_rate": 1.7759025686039857e-05,
      "loss": 0.8608,
      "num_input_tokens_seen": 1346989312,
      "step": 18120,
      "train_runtime": 21511.4084,
      "train_tokens_per_second": 62617.44
    },
    {
      "epoch": 0.24092863170344384,
      "grad_norm": 24.25,
      "learning_rate": 1.7756309799730375e-05,
      "loss": 0.695,
      "num_input_tokens_seen": 1347453184,
      "step": 18130,
      "train_runtime": 21517.9665,
      "train_tokens_per_second": 62619.913
    },
    {
      "epoch": 0.24106152118590574,
      "grad_norm": 27.0,
      "learning_rate": 1.7753592476633663e-05,
      "loss": 0.7018,
      "num_input_tokens_seen": 1347682048,
      "step": 18140,
      "train_runtime": 21521.4886,
      "train_tokens_per_second": 62620.299
    },
    {
      "epoch": 0.24119441066836766,
      "grad_norm": 7.71875,
      "learning_rate": 1.7750873717253082e-05,
      "loss": 0.5093,
      "num_input_tokens_seen": 1347797504,
      "step": 18150,
      "train_runtime": 21523.9158,
      "train_tokens_per_second": 62618.601
    },
    {
      "epoch": 0.24132730015082957,
      "grad_norm": 17.625,
      "learning_rate": 1.7748153522092256e-05,
      "loss": 0.8655,
      "num_input_tokens_seen": 1349837824,
      "step": 18160,
      "train_runtime": 21556.8643,
      "train_tokens_per_second": 62617.54
    },
    {
      "epoch": 0.24146018963329147,
      "grad_norm": 21.5,
      "learning_rate": 1.7745431891655086e-05,
      "loss": 0.9716,
      "num_input_tokens_seen": 1350652160,
      "step": 18170,
      "train_runtime": 21569.5832,
      "train_tokens_per_second": 62618.371
    },
    {
      "epoch": 0.2415930791157534,
      "grad_norm": 23.75,
      "learning_rate": 1.7742708826445717e-05,
      "loss": 0.7587,
      "num_input_tokens_seen": 1351090176,
      "step": 18180,
      "train_runtime": 21576.12,
      "train_tokens_per_second": 62619.701
    },
    {
      "epoch": 0.2417259685982153,
      "grad_norm": 10.8125,
      "learning_rate": 1.773998432696858e-05,
      "loss": 0.6392,
      "num_input_tokens_seen": 1351316736,
      "step": 18190,
      "train_runtime": 21579.5586,
      "train_tokens_per_second": 62620.221
    },
    {
      "epoch": 0.2418588580806772,
      "grad_norm": 10.8125,
      "learning_rate": 1.7737258393728363e-05,
      "loss": 0.5814,
      "num_input_tokens_seen": 1351434752,
      "step": 18200,
      "train_runtime": 21582.0083,
      "train_tokens_per_second": 62618.582
    },
    {
      "epoch": 0.24199174756313913,
      "grad_norm": 13.0625,
      "learning_rate": 1.773453102723002e-05,
      "loss": 0.7925,
      "num_input_tokens_seen": 1353440256,
      "step": 18210,
      "train_runtime": 21614.9993,
      "train_tokens_per_second": 62615.79
    },
    {
      "epoch": 0.24212463704560103,
      "grad_norm": 27.125,
      "learning_rate": 1.7731802227978773e-05,
      "loss": 0.9163,
      "num_input_tokens_seen": 1354281216,
      "step": 18220,
      "train_runtime": 21627.9738,
      "train_tokens_per_second": 62617.11
    },
    {
      "epoch": 0.24225752652806293,
      "grad_norm": 12.75,
      "learning_rate": 1.7729071996480107e-05,
      "loss": 0.7488,
      "num_input_tokens_seen": 1354769408,
      "step": 18230,
      "train_runtime": 21635.0169,
      "train_tokens_per_second": 62619.291
    },
    {
      "epoch": 0.24239041601052486,
      "grad_norm": 28.125,
      "learning_rate": 1.7726340333239773e-05,
      "loss": 0.6252,
      "num_input_tokens_seen": 1355007744,
      "step": 18240,
      "train_runtime": 21638.6187,
      "train_tokens_per_second": 62619.882
    },
    {
      "epoch": 0.24252330549298676,
      "grad_norm": 19.625,
      "learning_rate": 1.7723607238763782e-05,
      "loss": 0.5516,
      "num_input_tokens_seen": 1355128320,
      "step": 18250,
      "train_runtime": 21641.0983,
      "train_tokens_per_second": 62618.278
    },
    {
      "epoch": 0.24265619497544866,
      "grad_norm": 19.0,
      "learning_rate": 1.7720872713558422e-05,
      "loss": 0.8304,
      "num_input_tokens_seen": 1357142272,
      "step": 18260,
      "train_runtime": 21674.0371,
      "train_tokens_per_second": 62616.035
    },
    {
      "epoch": 0.2427890844579106,
      "grad_norm": 34.5,
      "learning_rate": 1.771813675813024e-05,
      "loss": 0.9623,
      "num_input_tokens_seen": 1357988608,
      "step": 18270,
      "train_runtime": 21686.6847,
      "train_tokens_per_second": 62618.544
    },
    {
      "epoch": 0.2429219739403725,
      "grad_norm": 13.9375,
      "learning_rate": 1.7715399372986043e-05,
      "loss": 0.7174,
      "num_input_tokens_seen": 1358444288,
      "step": 18280,
      "train_runtime": 21693.394,
      "train_tokens_per_second": 62620.182
    },
    {
      "epoch": 0.2430548634228344,
      "grad_norm": 10.75,
      "learning_rate": 1.7712660558632907e-05,
      "loss": 0.6895,
      "num_input_tokens_seen": 1358673664,
      "step": 18290,
      "train_runtime": 21697.3884,
      "train_tokens_per_second": 62619.226
    },
    {
      "epoch": 0.24318775290529632,
      "grad_norm": 17.625,
      "learning_rate": 1.770992031557818e-05,
      "loss": 0.5243,
      "num_input_tokens_seen": 1358787584,
      "step": 18300,
      "train_runtime": 21699.7975,
      "train_tokens_per_second": 62617.524
    },
    {
      "epoch": 0.24332064238775822,
      "grad_norm": 13.5,
      "learning_rate": 1.770717864432946e-05,
      "loss": 0.8989,
      "num_input_tokens_seen": 1360732416,
      "step": 18310,
      "train_runtime": 21731.1374,
      "train_tokens_per_second": 62616.714
    },
    {
      "epoch": 0.24345353187022012,
      "grad_norm": 35.5,
      "learning_rate": 1.770443554539462e-05,
      "loss": 1.0343,
      "num_input_tokens_seen": 1361561344,
      "step": 18320,
      "train_runtime": 21743.6077,
      "train_tokens_per_second": 62618.925
    },
    {
      "epoch": 0.24358642135268205,
      "grad_norm": 26.375,
      "learning_rate": 1.7701691019281796e-05,
      "loss": 0.743,
      "num_input_tokens_seen": 1362043904,
      "step": 18330,
      "train_runtime": 21750.2479,
      "train_tokens_per_second": 62621.994
    },
    {
      "epoch": 0.24371931083514395,
      "grad_norm": 25.125,
      "learning_rate": 1.769894506649938e-05,
      "loss": 0.6336,
      "num_input_tokens_seen": 1362272768,
      "step": 18340,
      "train_runtime": 21754.1942,
      "train_tokens_per_second": 62621.155
    },
    {
      "epoch": 0.24385220031760585,
      "grad_norm": 9.0,
      "learning_rate": 1.7696197687556047e-05,
      "loss": 0.4698,
      "num_input_tokens_seen": 1362385408,
      "step": 18350,
      "train_runtime": 21756.5737,
      "train_tokens_per_second": 62619.484
    },
    {
      "epoch": 0.24398508980006778,
      "grad_norm": 17.25,
      "learning_rate": 1.7693448882960714e-05,
      "loss": 0.8778,
      "num_input_tokens_seen": 1364539904,
      "step": 18360,
      "train_runtime": 21791.1265,
      "train_tokens_per_second": 62619.062
    },
    {
      "epoch": 0.24411797928252968,
      "grad_norm": 65.0,
      "learning_rate": 1.769069865322258e-05,
      "loss": 0.8618,
      "num_input_tokens_seen": 1365381376,
      "step": 18370,
      "train_runtime": 21804.0215,
      "train_tokens_per_second": 62620.622
    },
    {
      "epoch": 0.2442508687649916,
      "grad_norm": 18.5,
      "learning_rate": 1.7687946998851097e-05,
      "loss": 0.7902,
      "num_input_tokens_seen": 1365861120,
      "step": 18380,
      "train_runtime": 21810.8286,
      "train_tokens_per_second": 62623.073
    },
    {
      "epoch": 0.2443837582474535,
      "grad_norm": 15.4375,
      "learning_rate": 1.768519392035598e-05,
      "loss": 0.6462,
      "num_input_tokens_seen": 1366085376,
      "step": 18390,
      "train_runtime": 21814.235,
      "train_tokens_per_second": 62623.575
    },
    {
      "epoch": 0.2445166477299154,
      "grad_norm": 6.78125,
      "learning_rate": 1.768243941824722e-05,
      "loss": 0.5277,
      "num_input_tokens_seen": 1366201088,
      "step": 18400,
      "train_runtime": 21816.5822,
      "train_tokens_per_second": 62622.141
    },
    {
      "epoch": 0.24464953721237734,
      "grad_norm": 19.25,
      "learning_rate": 1.767968349303506e-05,
      "loss": 0.8533,
      "num_input_tokens_seen": 1368215296,
      "step": 18410,
      "train_runtime": 21849.0337,
      "train_tokens_per_second": 62621.318
    },
    {
      "epoch": 0.24478242669483924,
      "grad_norm": 21.125,
      "learning_rate": 1.7676926145230014e-05,
      "loss": 0.9415,
      "num_input_tokens_seen": 1369014784,
      "step": 18420,
      "train_runtime": 21861.3462,
      "train_tokens_per_second": 62622.62
    },
    {
      "epoch": 0.24491531617730114,
      "grad_norm": 28.875,
      "learning_rate": 1.767416737534285e-05,
      "loss": 0.7064,
      "num_input_tokens_seen": 1369457664,
      "step": 18430,
      "train_runtime": 21867.7011,
      "train_tokens_per_second": 62624.674
    },
    {
      "epoch": 0.24504820565976307,
      "grad_norm": 22.125,
      "learning_rate": 1.7671407183884613e-05,
      "loss": 0.6162,
      "num_input_tokens_seen": 1369684736,
      "step": 18440,
      "train_runtime": 21871.0836,
      "train_tokens_per_second": 62625.371
    },
    {
      "epoch": 0.24518109514222497,
      "grad_norm": 86.5,
      "learning_rate": 1.7668645571366595e-05,
      "loss": 0.6542,
      "num_input_tokens_seen": 1369797120,
      "step": 18450,
      "train_runtime": 21873.4629,
      "train_tokens_per_second": 62623.697
    },
    {
      "epoch": 0.24531398462468687,
      "grad_norm": 27.0,
      "learning_rate": 1.766588253830037e-05,
      "loss": 0.7945,
      "num_input_tokens_seen": 1371860224,
      "step": 18460,
      "train_runtime": 21907.9168,
      "train_tokens_per_second": 62619.383
    },
    {
      "epoch": 0.2454468741071488,
      "grad_norm": 34.5,
      "learning_rate": 1.7663118085197758e-05,
      "loss": 0.8168,
      "num_input_tokens_seen": 1372711424,
      "step": 18470,
      "train_runtime": 21920.9099,
      "train_tokens_per_second": 62621.097
    },
    {
      "epoch": 0.2455797635896107,
      "grad_norm": 27.0,
      "learning_rate": 1.766035221257085e-05,
      "loss": 0.8292,
      "num_input_tokens_seen": 1373208832,
      "step": 18480,
      "train_runtime": 21927.8996,
      "train_tokens_per_second": 62623.82
    },
    {
      "epoch": 0.2457126530720726,
      "grad_norm": 18.0,
      "learning_rate": 1.7657584920932e-05,
      "loss": 0.6389,
      "num_input_tokens_seen": 1373443584,
      "step": 18490,
      "train_runtime": 21931.4391,
      "train_tokens_per_second": 62624.417
    },
    {
      "epoch": 0.24584554255453453,
      "grad_norm": 11.5625,
      "learning_rate": 1.7654816210793825e-05,
      "loss": 0.4921,
      "num_input_tokens_seen": 1373560064,
      "step": 18500,
      "train_runtime": 21933.8139,
      "train_tokens_per_second": 62622.947
    },
    {
      "epoch": 0.24597843203699643,
      "grad_norm": 24.625,
      "learning_rate": 1.765204608266921e-05,
      "loss": 0.9047,
      "num_input_tokens_seen": 1375638528,
      "step": 18510,
      "train_runtime": 21967.778,
      "train_tokens_per_second": 62620.741
    },
    {
      "epoch": 0.24611132151945833,
      "grad_norm": 26.25,
      "learning_rate": 1.7649274537071282e-05,
      "loss": 0.9344,
      "num_input_tokens_seen": 1376502272,
      "step": 18520,
      "train_runtime": 21980.6659,
      "train_tokens_per_second": 62623.32
    },
    {
      "epoch": 0.24624421100192026,
      "grad_norm": 11.875,
      "learning_rate": 1.7646501574513456e-05,
      "loss": 0.7735,
      "num_input_tokens_seen": 1376994560,
      "step": 18530,
      "train_runtime": 21987.5859,
      "train_tokens_per_second": 62626.0
    },
    {
      "epoch": 0.24637710048438216,
      "grad_norm": 61.5,
      "learning_rate": 1.76437271955094e-05,
      "loss": 0.7045,
      "num_input_tokens_seen": 1377228544,
      "step": 18540,
      "train_runtime": 21991.6309,
      "train_tokens_per_second": 62625.121
    },
    {
      "epoch": 0.24650998996684406,
      "grad_norm": 11.5625,
      "learning_rate": 1.7640951400573033e-05,
      "loss": 0.5756,
      "num_input_tokens_seen": 1377346816,
      "step": 18550,
      "train_runtime": 21994.06,
      "train_tokens_per_second": 62623.582
    },
    {
      "epoch": 0.246642879449306,
      "grad_norm": 22.875,
      "learning_rate": 1.7638174190218557e-05,
      "loss": 0.8247,
      "num_input_tokens_seen": 1379389440,
      "step": 18560,
      "train_runtime": 22025.1543,
      "train_tokens_per_second": 62627.913
    },
    {
      "epoch": 0.2467757689317679,
      "grad_norm": 24.5,
      "learning_rate": 1.763539556496042e-05,
      "loss": 0.8902,
      "num_input_tokens_seen": 1380217344,
      "step": 18570,
      "train_runtime": 22037.6399,
      "train_tokens_per_second": 62629.998
    },
    {
      "epoch": 0.2469086584142298,
      "grad_norm": 16.625,
      "learning_rate": 1.7632615525313333e-05,
      "loss": 0.6775,
      "num_input_tokens_seen": 1380674304,
      "step": 18580,
      "train_runtime": 22044.2048,
      "train_tokens_per_second": 62632.076
    },
    {
      "epoch": 0.24704154789669172,
      "grad_norm": 14.125,
      "learning_rate": 1.7629834071792285e-05,
      "loss": 0.6101,
      "num_input_tokens_seen": 1380901632,
      "step": 18590,
      "train_runtime": 22047.6191,
      "train_tokens_per_second": 62632.687
    },
    {
      "epoch": 0.24717443737915362,
      "grad_norm": 7.21875,
      "learning_rate": 1.7627051204912505e-05,
      "loss": 0.5412,
      "num_input_tokens_seen": 1381016832,
      "step": 18600,
      "train_runtime": 22050.4903,
      "train_tokens_per_second": 62629.756
    },
    {
      "epoch": 0.24730732686161555,
      "grad_norm": 30.0,
      "learning_rate": 1.76242669251895e-05,
      "loss": 0.8594,
      "num_input_tokens_seen": 1382939136,
      "step": 18610,
      "train_runtime": 22081.0482,
      "train_tokens_per_second": 62630.14
    },
    {
      "epoch": 0.24744021634407745,
      "grad_norm": 20.125,
      "learning_rate": 1.762148123313903e-05,
      "loss": 0.8925,
      "num_input_tokens_seen": 1383775744,
      "step": 18620,
      "train_runtime": 22093.4605,
      "train_tokens_per_second": 62632.821
    },
    {
      "epoch": 0.24757310582653935,
      "grad_norm": 37.5,
      "learning_rate": 1.7618694129277124e-05,
      "loss": 0.7506,
      "num_input_tokens_seen": 1384224256,
      "step": 18630,
      "train_runtime": 22099.824,
      "train_tokens_per_second": 62635.081
    },
    {
      "epoch": 0.24770599530900128,
      "grad_norm": 16.0,
      "learning_rate": 1.7615905614120065e-05,
      "loss": 0.6315,
      "num_input_tokens_seen": 1384445184,
      "step": 18640,
      "train_runtime": 22103.2272,
      "train_tokens_per_second": 62635.432
    },
    {
      "epoch": 0.24783888479146318,
      "grad_norm": 11.1875,
      "learning_rate": 1.7613115688184397e-05,
      "loss": 0.4859,
      "num_input_tokens_seen": 1384560128,
      "step": 18650,
      "train_runtime": 22106.0513,
      "train_tokens_per_second": 62632.63
    },
    {
      "epoch": 0.24797177427392508,
      "grad_norm": 18.25,
      "learning_rate": 1.7610324351986935e-05,
      "loss": 0.8009,
      "num_input_tokens_seen": 1386617856,
      "step": 18660,
      "train_runtime": 22138.0843,
      "train_tokens_per_second": 62634.952
    },
    {
      "epoch": 0.248104663756387,
      "grad_norm": 11.3125,
      "learning_rate": 1.7607531606044744e-05,
      "loss": 0.9352,
      "num_input_tokens_seen": 1387445248,
      "step": 18670,
      "train_runtime": 22150.3914,
      "train_tokens_per_second": 62637.505
    },
    {
      "epoch": 0.2482375532388489,
      "grad_norm": 23.375,
      "learning_rate": 1.7604737450875166e-05,
      "loss": 0.758,
      "num_input_tokens_seen": 1387891456,
      "step": 18680,
      "train_runtime": 22156.7981,
      "train_tokens_per_second": 62639.532
    },
    {
      "epoch": 0.2483704427213108,
      "grad_norm": 13.5625,
      "learning_rate": 1.7601941886995776e-05,
      "loss": 0.6312,
      "num_input_tokens_seen": 1388117248,
      "step": 18690,
      "train_runtime": 22160.1993,
      "train_tokens_per_second": 62640.107
    },
    {
      "epoch": 0.24850333220377274,
      "grad_norm": 14.75,
      "learning_rate": 1.7599144914924443e-05,
      "loss": 0.5273,
      "num_input_tokens_seen": 1388233472,
      "step": 18700,
      "train_runtime": 22162.588,
      "train_tokens_per_second": 62638.599
    },
    {
      "epoch": 0.24863622168623464,
      "grad_norm": 28.625,
      "learning_rate": 1.759634653517927e-05,
      "loss": 0.867,
      "num_input_tokens_seen": 1390377984,
      "step": 18710,
      "train_runtime": 22194.8333,
      "train_tokens_per_second": 62644.218
    },
    {
      "epoch": 0.24876911116869654,
      "grad_norm": 15.3125,
      "learning_rate": 1.7593546748278642e-05,
      "loss": 0.8556,
      "num_input_tokens_seen": 1391224320,
      "step": 18720,
      "train_runtime": 22207.4201,
      "train_tokens_per_second": 62646.823
    },
    {
      "epoch": 0.24890200065115847,
      "grad_norm": 24.125,
      "learning_rate": 1.759074555474119e-05,
      "loss": 0.7507,
      "num_input_tokens_seen": 1391678464,
      "step": 18730,
      "train_runtime": 22214.4455,
      "train_tokens_per_second": 62647.455
    },
    {
      "epoch": 0.24903489013362037,
      "grad_norm": 51.0,
      "learning_rate": 1.7587942955085803e-05,
      "loss": 0.6241,
      "num_input_tokens_seen": 1391913984,
      "step": 18740,
      "train_runtime": 22217.9357,
      "train_tokens_per_second": 62648.214
    },
    {
      "epoch": 0.24916777961608227,
      "grad_norm": 14.6875,
      "learning_rate": 1.7585138949831647e-05,
      "loss": 0.5487,
      "num_input_tokens_seen": 1392028672,
      "step": 18750,
      "train_runtime": 22220.283,
      "train_tokens_per_second": 62646.757
    },
    {
      "epoch": 0.2493006690985442,
      "grad_norm": 24.0,
      "learning_rate": 1.7582333539498137e-05,
      "loss": 0.9015,
      "num_input_tokens_seen": 1394138624,
      "step": 18760,
      "train_runtime": 22253.8724,
      "train_tokens_per_second": 62647.012
    },
    {
      "epoch": 0.2494335585810061,
      "grad_norm": 33.0,
      "learning_rate": 1.757952672460495e-05,
      "loss": 0.993,
      "num_input_tokens_seen": 1394996224,
      "step": 18770,
      "train_runtime": 22266.9206,
      "train_tokens_per_second": 62648.817
    },
    {
      "epoch": 0.249566448063468,
      "grad_norm": 13.8125,
      "learning_rate": 1.757671850567202e-05,
      "loss": 0.7815,
      "num_input_tokens_seen": 1395487744,
      "step": 18780,
      "train_runtime": 22273.9254,
      "train_tokens_per_second": 62651.181
    },
    {
      "epoch": 0.24969933754592993,
      "grad_norm": 10.5,
      "learning_rate": 1.757390888321955e-05,
      "loss": 0.6753,
      "num_input_tokens_seen": 1395714816,
      "step": 18790,
      "train_runtime": 22277.8659,
      "train_tokens_per_second": 62650.293
    },
    {
      "epoch": 0.24983222702839183,
      "grad_norm": 15.9375,
      "learning_rate": 1.7571097857767992e-05,
      "loss": 0.5379,
      "num_input_tokens_seen": 1395829760,
      "step": 18800,
      "train_runtime": 22280.3002,
      "train_tokens_per_second": 62648.606
    },
    {
      "epoch": 0.24996511651085374,
      "grad_norm": 34.0,
      "learning_rate": 1.7568285429838065e-05,
      "loss": 0.8344,
      "num_input_tokens_seen": 1397766400,
      "step": 18810,
      "train_runtime": 22311.7988,
      "train_tokens_per_second": 62646.961
    },
    {
      "epoch": 0.25009800599331566,
      "grad_norm": 11.5625,
      "learning_rate": 1.756547159995075e-05,
      "loss": 0.9732,
      "num_input_tokens_seen": 1398598912,
      "step": 18820,
      "train_runtime": 22324.1789,
      "train_tokens_per_second": 62649.512
    },
    {
      "epoch": 0.2502308954757776,
      "grad_norm": 19.625,
      "learning_rate": 1.756265636862728e-05,
      "loss": 0.7499,
      "num_input_tokens_seen": 1399066880,
      "step": 18830,
      "train_runtime": 22330.8485,
      "train_tokens_per_second": 62651.756
    },
    {
      "epoch": 0.25036378495823947,
      "grad_norm": 41.75,
      "learning_rate": 1.7559839736389154e-05,
      "loss": 0.6777,
      "num_input_tokens_seen": 1399289088,
      "step": 18840,
      "train_runtime": 22334.6966,
      "train_tokens_per_second": 62650.911
    },
    {
      "epoch": 0.2504966744407014,
      "grad_norm": 7.78125,
      "learning_rate": 1.7557021703758126e-05,
      "loss": 0.561,
      "num_input_tokens_seen": 1399405056,
      "step": 18850,
      "train_runtime": 22337.0585,
      "train_tokens_per_second": 62649.478
    },
    {
      "epoch": 0.2506295639231633,
      "grad_norm": 21.125,
      "learning_rate": 1.755420227125621e-05,
      "loss": 0.864,
      "num_input_tokens_seen": 1401480192,
      "step": 18860,
      "train_runtime": 22370.1807,
      "train_tokens_per_second": 62649.48
    },
    {
      "epoch": 0.2507624534056252,
      "grad_norm": 12.4375,
      "learning_rate": 1.7551381439405684e-05,
      "loss": 0.9452,
      "num_input_tokens_seen": 1402333696,
      "step": 18870,
      "train_runtime": 22382.6717,
      "train_tokens_per_second": 62652.65
    },
    {
      "epoch": 0.2508953428880871,
      "grad_norm": 17.625,
      "learning_rate": 1.7548559208729082e-05,
      "loss": 0.668,
      "num_input_tokens_seen": 1402787328,
      "step": 18880,
      "train_runtime": 22389.6718,
      "train_tokens_per_second": 62653.323
    },
    {
      "epoch": 0.25102823237054905,
      "grad_norm": 11.9375,
      "learning_rate": 1.75457355797492e-05,
      "loss": 0.6419,
      "num_input_tokens_seen": 1403006208,
      "step": 18890,
      "train_runtime": 22393.0404,
      "train_tokens_per_second": 62653.672
    },
    {
      "epoch": 0.2511611218530109,
      "grad_norm": 26.5,
      "learning_rate": 1.754291055298908e-05,
      "loss": 0.593,
      "num_input_tokens_seen": 1403122176,
      "step": 18900,
      "train_runtime": 22395.41,
      "train_tokens_per_second": 62652.221
    },
    {
      "epoch": 0.25129401133547286,
      "grad_norm": 14.625,
      "learning_rate": 1.754008412897204e-05,
      "loss": 0.902,
      "num_input_tokens_seen": 1405151744,
      "step": 18910,
      "train_runtime": 22428.7057,
      "train_tokens_per_second": 62649.703
    },
    {
      "epoch": 0.2514269008179348,
      "grad_norm": 35.0,
      "learning_rate": 1.7537256308221647e-05,
      "loss": 0.9951,
      "num_input_tokens_seen": 1405994240,
      "step": 18920,
      "train_runtime": 22441.3071,
      "train_tokens_per_second": 62652.065
    },
    {
      "epoch": 0.25155979030039666,
      "grad_norm": 13.625,
      "learning_rate": 1.7534427091261735e-05,
      "loss": 0.6479,
      "num_input_tokens_seen": 1406472960,
      "step": 18930,
      "train_runtime": 22447.7981,
      "train_tokens_per_second": 62655.275
    },
    {
      "epoch": 0.2516926797828586,
      "grad_norm": 17.0,
      "learning_rate": 1.753159647861639e-05,
      "loss": 0.6927,
      "num_input_tokens_seen": 1406706432,
      "step": 18940,
      "train_runtime": 22451.1978,
      "train_tokens_per_second": 62656.186
    },
    {
      "epoch": 0.2518255692653205,
      "grad_norm": 12.4375,
      "learning_rate": 1.7528764470809947e-05,
      "loss": 0.4877,
      "num_input_tokens_seen": 1406818816,
      "step": 18950,
      "train_runtime": 22454.0102,
      "train_tokens_per_second": 62653.344
    },
    {
      "epoch": 0.2519584587477824,
      "grad_norm": 13.125,
      "learning_rate": 1.7525931068367023e-05,
      "loss": 0.8272,
      "num_input_tokens_seen": 1408876288,
      "step": 18960,
      "train_runtime": 22486.2182,
      "train_tokens_per_second": 62655.102
    },
    {
      "epoch": 0.2520913482302443,
      "grad_norm": 15.125,
      "learning_rate": 1.7523096271812477e-05,
      "loss": 0.9478,
      "num_input_tokens_seen": 1409730048,
      "step": 18970,
      "train_runtime": 22498.9203,
      "train_tokens_per_second": 62657.676
    },
    {
      "epoch": 0.25222423771270625,
      "grad_norm": 160.0,
      "learning_rate": 1.752026008167143e-05,
      "loss": 0.6876,
      "num_input_tokens_seen": 1410202112,
      "step": 18980,
      "train_runtime": 22505.5728,
      "train_tokens_per_second": 62660.13
    },
    {
      "epoch": 0.2523571271951681,
      "grad_norm": 10.4375,
      "learning_rate": 1.7517422498469257e-05,
      "loss": 0.7337,
      "num_input_tokens_seen": 1410429952,
      "step": 18990,
      "train_runtime": 22509.5504,
      "train_tokens_per_second": 62659.179
    },
    {
      "epoch": 0.25249001667763005,
      "grad_norm": 13.5625,
      "learning_rate": 1.7514583522731598e-05,
      "loss": 0.4881,
      "num_input_tokens_seen": 1410544896,
      "step": 19000,
      "train_runtime": 22511.9398,
      "train_tokens_per_second": 62657.634
    },
    {
      "epoch": 0.252622906160092,
      "grad_norm": 18.625,
      "learning_rate": 1.7511743154984347e-05,
      "loss": 0.855,
      "num_input_tokens_seen": 1412594944,
      "step": 19010,
      "train_runtime": 22543.7261,
      "train_tokens_per_second": 62660.225
    },
    {
      "epoch": 0.25275579564255385,
      "grad_norm": 39.0,
      "learning_rate": 1.7508901395753658e-05,
      "loss": 0.9236,
      "num_input_tokens_seen": 1413455360,
      "step": 19020,
      "train_runtime": 22556.4629,
      "train_tokens_per_second": 62662.988
    },
    {
      "epoch": 0.2528886851250158,
      "grad_norm": 14.0625,
      "learning_rate": 1.750605824556594e-05,
      "loss": 0.6974,
      "num_input_tokens_seen": 1413921792,
      "step": 19030,
      "train_runtime": 22563.0338,
      "train_tokens_per_second": 62665.411
    },
    {
      "epoch": 0.2530215746074777,
      "grad_norm": 14.8125,
      "learning_rate": 1.7503213704947867e-05,
      "loss": 0.6345,
      "num_input_tokens_seen": 1414149120,
      "step": 19040,
      "train_runtime": 22566.4529,
      "train_tokens_per_second": 62665.99
    },
    {
      "epoch": 0.2531544640899396,
      "grad_norm": 10.875,
      "learning_rate": 1.750036777442636e-05,
      "loss": 0.5468,
      "num_input_tokens_seen": 1414266112,
      "step": 19050,
      "train_runtime": 22568.8208,
      "train_tokens_per_second": 62664.599
    },
    {
      "epoch": 0.2532873535724015,
      "grad_norm": 25.0,
      "learning_rate": 1.7497520454528597e-05,
      "loss": 0.8136,
      "num_input_tokens_seen": 1416276736,
      "step": 19060,
      "train_runtime": 22602.7124,
      "train_tokens_per_second": 62659.592
    },
    {
      "epoch": 0.25342024305486344,
      "grad_norm": 12.875,
      "learning_rate": 1.7494671745782025e-05,
      "loss": 0.912,
      "num_input_tokens_seen": 1417117952,
      "step": 19070,
      "train_runtime": 22615.0359,
      "train_tokens_per_second": 62662.644
    },
    {
      "epoch": 0.2535531325373253,
      "grad_norm": 26.75,
      "learning_rate": 1.7491821648714343e-05,
      "loss": 0.7835,
      "num_input_tokens_seen": 1417552640,
      "step": 19080,
      "train_runtime": 22621.3594,
      "train_tokens_per_second": 62664.344
    },
    {
      "epoch": 0.25368602201978724,
      "grad_norm": 14.25,
      "learning_rate": 1.7488970163853505e-05,
      "loss": 0.6757,
      "num_input_tokens_seen": 1417766912,
      "step": 19090,
      "train_runtime": 22624.6864,
      "train_tokens_per_second": 62664.599
    },
    {
      "epoch": 0.25381891150224917,
      "grad_norm": 14.1875,
      "learning_rate": 1.748611729172772e-05,
      "loss": 0.6321,
      "num_input_tokens_seen": 1417881344,
      "step": 19100,
      "train_runtime": 22627.1064,
      "train_tokens_per_second": 62662.955
    },
    {
      "epoch": 0.25395180098471104,
      "grad_norm": 21.875,
      "learning_rate": 1.748326303286546e-05,
      "loss": 0.7708,
      "num_input_tokens_seen": 1419873792,
      "step": 19110,
      "train_runtime": 22660.1649,
      "train_tokens_per_second": 62659.464
    },
    {
      "epoch": 0.25408469046717297,
      "grad_norm": 57.5,
      "learning_rate": 1.748040738779545e-05,
      "loss": 0.9073,
      "num_input_tokens_seen": 1420720128,
      "step": 19120,
      "train_runtime": 22672.821,
      "train_tokens_per_second": 62661.816
    },
    {
      "epoch": 0.2542175799496349,
      "grad_norm": 27.125,
      "learning_rate": 1.747755035704667e-05,
      "loss": 0.7399,
      "num_input_tokens_seen": 1421200896,
      "step": 19130,
      "train_runtime": 22679.7615,
      "train_tokens_per_second": 62663.838
    },
    {
      "epoch": 0.25435046943209677,
      "grad_norm": 19.75,
      "learning_rate": 1.7474691941148366e-05,
      "loss": 0.6643,
      "num_input_tokens_seen": 1421437952,
      "step": 19140,
      "train_runtime": 22683.357,
      "train_tokens_per_second": 62664.356
    },
    {
      "epoch": 0.2544833589145587,
      "grad_norm": 73.5,
      "learning_rate": 1.747183214063003e-05,
      "loss": 0.591,
      "num_input_tokens_seen": 1421558528,
      "step": 19150,
      "train_runtime": 22685.74,
      "train_tokens_per_second": 62663.088
    },
    {
      "epoch": 0.25461624839702063,
      "grad_norm": 184.0,
      "learning_rate": 1.7468970956021415e-05,
      "loss": 0.8091,
      "num_input_tokens_seen": 1423581952,
      "step": 19160,
      "train_runtime": 22718.399,
      "train_tokens_per_second": 62662.072
    },
    {
      "epoch": 0.2547491378794825,
      "grad_norm": 28.0,
      "learning_rate": 1.7466108387852528e-05,
      "loss": 0.905,
      "num_input_tokens_seen": 1424393728,
      "step": 19170,
      "train_runtime": 22730.7296,
      "train_tokens_per_second": 62663.793
    },
    {
      "epoch": 0.25488202736194443,
      "grad_norm": 12.8125,
      "learning_rate": 1.7463244436653636e-05,
      "loss": 0.7426,
      "num_input_tokens_seen": 1424854016,
      "step": 19180,
      "train_runtime": 22737.4481,
      "train_tokens_per_second": 62665.52
    },
    {
      "epoch": 0.25501491684440636,
      "grad_norm": 14.8125,
      "learning_rate": 1.7460379102955265e-05,
      "loss": 0.595,
      "num_input_tokens_seen": 1425077504,
      "step": 19190,
      "train_runtime": 22741.3981,
      "train_tokens_per_second": 62664.463
    },
    {
      "epoch": 0.2551478063268683,
      "grad_norm": 13.6875,
      "learning_rate": 1.7457512387288186e-05,
      "loss": 0.4628,
      "num_input_tokens_seen": 1425196032,
      "step": 19200,
      "train_runtime": 22743.8009,
      "train_tokens_per_second": 62663.054
    },
    {
      "epoch": 0.25528069580933016,
      "grad_norm": 38.25,
      "learning_rate": 1.7454644290183433e-05,
      "loss": 0.8633,
      "num_input_tokens_seen": 1427298560,
      "step": 19210,
      "train_runtime": 22775.6566,
      "train_tokens_per_second": 62667.724
    },
    {
      "epoch": 0.2554135852917921,
      "grad_norm": 23.125,
      "learning_rate": 1.74517748121723e-05,
      "loss": 0.9268,
      "num_input_tokens_seen": 1428119552,
      "step": 19220,
      "train_runtime": 22788.1639,
      "train_tokens_per_second": 62669.356
    },
    {
      "epoch": 0.255546474774254,
      "grad_norm": 13.75,
      "learning_rate": 1.7448903953786325e-05,
      "loss": 0.7179,
      "num_input_tokens_seen": 1428580608,
      "step": 19230,
      "train_runtime": 22794.6237,
      "train_tokens_per_second": 62671.822
    },
    {
      "epoch": 0.2556793642567159,
      "grad_norm": 8.9375,
      "learning_rate": 1.7446031715557317e-05,
      "loss": 0.6381,
      "num_input_tokens_seen": 1428806144,
      "step": 19240,
      "train_runtime": 22798.0494,
      "train_tokens_per_second": 62672.298
    },
    {
      "epoch": 0.2558122537391778,
      "grad_norm": 6.78125,
      "learning_rate": 1.744315809801733e-05,
      "loss": 0.5185,
      "num_input_tokens_seen": 1428919552,
      "step": 19250,
      "train_runtime": 22800.9022,
      "train_tokens_per_second": 62669.43
    },
    {
      "epoch": 0.25594514322163975,
      "grad_norm": 50.25,
      "learning_rate": 1.744028310169867e-05,
      "loss": 0.8059,
      "num_input_tokens_seen": 1430871296,
      "step": 19260,
      "train_runtime": 22831.7392,
      "train_tokens_per_second": 62670.272
    },
    {
      "epoch": 0.2560780327041016,
      "grad_norm": 45.75,
      "learning_rate": 1.7437406727133912e-05,
      "loss": 0.9608,
      "num_input_tokens_seen": 1431685888,
      "step": 19270,
      "train_runtime": 22844.1951,
      "train_tokens_per_second": 62671.759
    },
    {
      "epoch": 0.25621092218656355,
      "grad_norm": 41.75,
      "learning_rate": 1.7434528974855874e-05,
      "loss": 0.7312,
      "num_input_tokens_seen": 1432164864,
      "step": 19280,
      "train_runtime": 22851.0288,
      "train_tokens_per_second": 62673.978
    },
    {
      "epoch": 0.2563438116690255,
      "grad_norm": 28.375,
      "learning_rate": 1.7431649845397637e-05,
      "loss": 0.6605,
      "num_input_tokens_seen": 1432394496,
      "step": 19290,
      "train_runtime": 22854.4791,
      "train_tokens_per_second": 62674.563
    },
    {
      "epoch": 0.25647670115148735,
      "grad_norm": 10.5,
      "learning_rate": 1.7428769339292535e-05,
      "loss": 0.4726,
      "num_input_tokens_seen": 1432507648,
      "step": 19300,
      "train_runtime": 22857.265,
      "train_tokens_per_second": 62671.875
    },
    {
      "epoch": 0.2566095906339493,
      "grad_norm": 12.625,
      "learning_rate": 1.742588745707415e-05,
      "loss": 0.7895,
      "num_input_tokens_seen": 1434490368,
      "step": 19310,
      "train_runtime": 22889.2738,
      "train_tokens_per_second": 62670.855
    },
    {
      "epoch": 0.2567424801164112,
      "grad_norm": 22.75,
      "learning_rate": 1.7423004199276336e-05,
      "loss": 0.9151,
      "num_input_tokens_seen": 1435340544,
      "step": 19320,
      "train_runtime": 22901.7289,
      "train_tokens_per_second": 62673.895
    },
    {
      "epoch": 0.2568753695988731,
      "grad_norm": 14.3125,
      "learning_rate": 1.7420119566433176e-05,
      "loss": 0.7147,
      "num_input_tokens_seen": 1435785472,
      "step": 19330,
      "train_runtime": 22908.1367,
      "train_tokens_per_second": 62675.786
    },
    {
      "epoch": 0.257008259081335,
      "grad_norm": 8.75,
      "learning_rate": 1.7417233559079035e-05,
      "loss": 0.6183,
      "num_input_tokens_seen": 1436004096,
      "step": 19340,
      "train_runtime": 22911.5266,
      "train_tokens_per_second": 62676.055
    },
    {
      "epoch": 0.25714114856379694,
      "grad_norm": 13.5625,
      "learning_rate": 1.7414346177748515e-05,
      "loss": 0.5158,
      "num_input_tokens_seen": 1436119040,
      "step": 19350,
      "train_runtime": 22913.9062,
      "train_tokens_per_second": 62674.562
    },
    {
      "epoch": 0.2572740380462588,
      "grad_norm": 17.0,
      "learning_rate": 1.7411457422976473e-05,
      "loss": 0.8044,
      "num_input_tokens_seen": 1438196992,
      "step": 19360,
      "train_runtime": 22947.7423,
      "train_tokens_per_second": 62672.701
    },
    {
      "epoch": 0.25740692752872074,
      "grad_norm": 54.5,
      "learning_rate": 1.7408567295298035e-05,
      "loss": 0.9397,
      "num_input_tokens_seen": 1439038464,
      "step": 19370,
      "train_runtime": 22960.2002,
      "train_tokens_per_second": 62675.345
    },
    {
      "epoch": 0.25753981701118267,
      "grad_norm": 23.125,
      "learning_rate": 1.7405675795248568e-05,
      "loss": 0.7486,
      "num_input_tokens_seen": 1439517440,
      "step": 19380,
      "train_runtime": 22967.3009,
      "train_tokens_per_second": 62676.822
    },
    {
      "epoch": 0.25767270649364454,
      "grad_norm": 62.75,
      "learning_rate": 1.740278292336369e-05,
      "loss": 0.6582,
      "num_input_tokens_seen": 1439754240,
      "step": 19390,
      "train_runtime": 22970.8124,
      "train_tokens_per_second": 62677.55
    },
    {
      "epoch": 0.25780559597610647,
      "grad_norm": 28.0,
      "learning_rate": 1.7399888680179288e-05,
      "loss": 0.5096,
      "num_input_tokens_seen": 1439870976,
      "step": 19400,
      "train_runtime": 22973.2434,
      "train_tokens_per_second": 62675.999
    },
    {
      "epoch": 0.2579384854585684,
      "grad_norm": 35.5,
      "learning_rate": 1.7396993066231492e-05,
      "loss": 0.8532,
      "num_input_tokens_seen": 1441987328,
      "step": 19410,
      "train_runtime": 23006.4957,
      "train_tokens_per_second": 62677.4
    },
    {
      "epoch": 0.2580713749410303,
      "grad_norm": 21.625,
      "learning_rate": 1.739409608205669e-05,
      "loss": 0.8792,
      "num_input_tokens_seen": 1442848512,
      "step": 19420,
      "train_runtime": 23019.1619,
      "train_tokens_per_second": 62680.323
    },
    {
      "epoch": 0.2582042644234922,
      "grad_norm": 23.5,
      "learning_rate": 1.7391197728191517e-05,
      "loss": 0.8032,
      "num_input_tokens_seen": 1443345408,
      "step": 19430,
      "train_runtime": 23025.933,
      "train_tokens_per_second": 62683.471
    },
    {
      "epoch": 0.25833715390595413,
      "grad_norm": 29.25,
      "learning_rate": 1.738829800517287e-05,
      "loss": 0.6754,
      "num_input_tokens_seen": 1443574784,
      "step": 19440,
      "train_runtime": 23029.8907,
      "train_tokens_per_second": 62682.659
    },
    {
      "epoch": 0.258470043388416,
      "grad_norm": 12.9375,
      "learning_rate": 1.7385396913537902e-05,
      "loss": 0.496,
      "num_input_tokens_seen": 1443690496,
      "step": 19450,
      "train_runtime": 23032.2194,
      "train_tokens_per_second": 62681.345
    },
    {
      "epoch": 0.25860293287087793,
      "grad_norm": 31.75,
      "learning_rate": 1.7382494453824006e-05,
      "loss": 0.789,
      "num_input_tokens_seen": 1445710848,
      "step": 19460,
      "train_runtime": 23064.445,
      "train_tokens_per_second": 62681.363
    },
    {
      "epoch": 0.25873582235333986,
      "grad_norm": 34.5,
      "learning_rate": 1.7379590626568843e-05,
      "loss": 0.9539,
      "num_input_tokens_seen": 1446540800,
      "step": 19470,
      "train_runtime": 23077.2443,
      "train_tokens_per_second": 62682.562
    },
    {
      "epoch": 0.25886871183580173,
      "grad_norm": 24.75,
      "learning_rate": 1.737668543231032e-05,
      "loss": 0.713,
      "num_input_tokens_seen": 1446994944,
      "step": 19480,
      "train_runtime": 23083.7714,
      "train_tokens_per_second": 62684.512
    },
    {
      "epoch": 0.25900160131826366,
      "grad_norm": 26.125,
      "learning_rate": 1.7373778871586595e-05,
      "loss": 0.6834,
      "num_input_tokens_seen": 1447214848,
      "step": 19490,
      "train_runtime": 23087.5316,
      "train_tokens_per_second": 62683.828
    },
    {
      "epoch": 0.2591344908007256,
      "grad_norm": 7.25,
      "learning_rate": 1.7370870944936086e-05,
      "loss": 0.5524,
      "num_input_tokens_seen": 1447329024,
      "step": 19500,
      "train_runtime": 23089.8497,
      "train_tokens_per_second": 62682.479
    },
    {
      "epoch": 0.25926738028318747,
      "grad_norm": 19.0,
      "learning_rate": 1.736796165289746e-05,
      "loss": 0.7161,
      "num_input_tokens_seen": 1449437696,
      "step": 19510,
      "train_runtime": 23124.2859,
      "train_tokens_per_second": 62680.322
    },
    {
      "epoch": 0.2594002697656494,
      "grad_norm": 19.5,
      "learning_rate": 1.7365050996009635e-05,
      "loss": 0.9021,
      "num_input_tokens_seen": 1450281216,
      "step": 19520,
      "train_runtime": 23136.9259,
      "train_tokens_per_second": 62682.537
    },
    {
      "epoch": 0.2595331592481113,
      "grad_norm": 29.75,
      "learning_rate": 1.7362138974811786e-05,
      "loss": 0.7634,
      "num_input_tokens_seen": 1450733824,
      "step": 19530,
      "train_runtime": 23144.0263,
      "train_tokens_per_second": 62682.863
    },
    {
      "epoch": 0.2596660487305732,
      "grad_norm": 23.625,
      "learning_rate": 1.735922558984334e-05,
      "loss": 0.6141,
      "num_input_tokens_seen": 1450964480,
      "step": 19540,
      "train_runtime": 23147.4446,
      "train_tokens_per_second": 62683.571
    },
    {
      "epoch": 0.2597989382130351,
      "grad_norm": 11.3125,
      "learning_rate": 1.7356310841643974e-05,
      "loss": 0.5634,
      "num_input_tokens_seen": 1451080192,
      "step": 19550,
      "train_runtime": 23149.8008,
      "train_tokens_per_second": 62682.189
    },
    {
      "epoch": 0.25993182769549705,
      "grad_norm": 16.875,
      "learning_rate": 1.7353394730753622e-05,
      "loss": 0.8692,
      "num_input_tokens_seen": 1453141760,
      "step": 19560,
      "train_runtime": 23182.7431,
      "train_tokens_per_second": 62682.046
    },
    {
      "epoch": 0.2600647171779589,
      "grad_norm": 35.5,
      "learning_rate": 1.7350477257712464e-05,
      "loss": 0.9364,
      "num_input_tokens_seen": 1453984512,
      "step": 19570,
      "train_runtime": 23195.2138,
      "train_tokens_per_second": 62684.678
    },
    {
      "epoch": 0.26019760666042085,
      "grad_norm": 18.0,
      "learning_rate": 1.7347558423060934e-05,
      "loss": 0.7602,
      "num_input_tokens_seen": 1454456064,
      "step": 19580,
      "train_runtime": 23201.894,
      "train_tokens_per_second": 62686.954
    },
    {
      "epoch": 0.2603304961428828,
      "grad_norm": 20.75,
      "learning_rate": 1.7344638227339724e-05,
      "loss": 0.6672,
      "num_input_tokens_seen": 1454685184,
      "step": 19590,
      "train_runtime": 23205.352,
      "train_tokens_per_second": 62687.486
    },
    {
      "epoch": 0.26046338562534466,
      "grad_norm": 12.8125,
      "learning_rate": 1.7341716671089778e-05,
      "loss": 0.5694,
      "num_input_tokens_seen": 1454803968,
      "step": 19600,
      "train_runtime": 23207.7708,
      "train_tokens_per_second": 62686.071
    },
    {
      "epoch": 0.2605962751078066,
      "grad_norm": 59.25,
      "learning_rate": 1.7338793754852278e-05,
      "loss": 0.8411,
      "num_input_tokens_seen": 1456804864,
      "step": 19610,
      "train_runtime": 23240.02,
      "train_tokens_per_second": 62685.181
    },
    {
      "epoch": 0.2607291645902685,
      "grad_norm": 12.25,
      "learning_rate": 1.7335869479168676e-05,
      "loss": 0.8978,
      "num_input_tokens_seen": 1457675264,
      "step": 19620,
      "train_runtime": 23252.7744,
      "train_tokens_per_second": 62688.23
    },
    {
      "epoch": 0.2608620540727304,
      "grad_norm": 36.5,
      "learning_rate": 1.7332943844580667e-05,
      "loss": 0.8424,
      "num_input_tokens_seen": 1458155520,
      "step": 19630,
      "train_runtime": 23259.4423,
      "train_tokens_per_second": 62690.906
    },
    {
      "epoch": 0.2609949435551923,
      "grad_norm": 12.3125,
      "learning_rate": 1.7330016851630197e-05,
      "loss": 0.6433,
      "num_input_tokens_seen": 1458377472,
      "step": 19640,
      "train_runtime": 23262.8671,
      "train_tokens_per_second": 62691.218
    },
    {
      "epoch": 0.26112783303765424,
      "grad_norm": 13.875,
      "learning_rate": 1.7327088500859465e-05,
      "loss": 0.5221,
      "num_input_tokens_seen": 1458492672,
      "step": 19650,
      "train_runtime": 23265.2544,
      "train_tokens_per_second": 62689.737
    },
    {
      "epoch": 0.2612607225201162,
      "grad_norm": 16.375,
      "learning_rate": 1.732415879281092e-05,
      "loss": 0.8132,
      "num_input_tokens_seen": 1460570368,
      "step": 19660,
      "train_runtime": 23298.5875,
      "train_tokens_per_second": 62689.224
    },
    {
      "epoch": 0.26139361200257805,
      "grad_norm": 19.125,
      "learning_rate": 1.732122772802727e-05,
      "loss": 0.919,
      "num_input_tokens_seen": 1461401600,
      "step": 19670,
      "train_runtime": 23311.4118,
      "train_tokens_per_second": 62690.394
    },
    {
      "epoch": 0.26152650148504,
      "grad_norm": 44.75,
      "learning_rate": 1.7318295307051466e-05,
      "loss": 0.7557,
      "num_input_tokens_seen": 1461880576,
      "step": 19680,
      "train_runtime": 23318.2179,
      "train_tokens_per_second": 62692.637
    },
    {
      "epoch": 0.2616593909675019,
      "grad_norm": 12.625,
      "learning_rate": 1.731536153042671e-05,
      "loss": 0.6515,
      "num_input_tokens_seen": 1462105088,
      "step": 19690,
      "train_runtime": 23321.6474,
      "train_tokens_per_second": 62693.045
    },
    {
      "epoch": 0.2617922804499638,
      "grad_norm": 41.75,
      "learning_rate": 1.7312426398696462e-05,
      "loss": 0.5304,
      "num_input_tokens_seen": 1462221568,
      "step": 19700,
      "train_runtime": 23324.0195,
      "train_tokens_per_second": 62691.663
    },
    {
      "epoch": 0.2619251699324257,
      "grad_norm": 16.5,
      "learning_rate": 1.7309489912404427e-05,
      "loss": 0.8117,
      "num_input_tokens_seen": 1464247552,
      "step": 19710,
      "train_runtime": 23357.4165,
      "train_tokens_per_second": 62688.763
    },
    {
      "epoch": 0.26205805941488763,
      "grad_norm": 20.5,
      "learning_rate": 1.7306552072094566e-05,
      "loss": 0.9742,
      "num_input_tokens_seen": 1465087744,
      "step": 19720,
      "train_runtime": 23369.9828,
      "train_tokens_per_second": 62691.007
    },
    {
      "epoch": 0.2621909488973495,
      "grad_norm": 12.625,
      "learning_rate": 1.730361287831108e-05,
      "loss": 0.7342,
      "num_input_tokens_seen": 1465555968,
      "step": 19730,
      "train_runtime": 23376.6461,
      "train_tokens_per_second": 62693.167
    },
    {
      "epoch": 0.26232383837981144,
      "grad_norm": 23.375,
      "learning_rate": 1.7300672331598435e-05,
      "loss": 0.5994,
      "num_input_tokens_seen": 1465789952,
      "step": 19740,
      "train_runtime": 23380.1781,
      "train_tokens_per_second": 62693.704
    },
    {
      "epoch": 0.26245672786227336,
      "grad_norm": 15.0,
      "learning_rate": 1.729773043250134e-05,
      "loss": 0.5013,
      "num_input_tokens_seen": 1465909504,
      "step": 19750,
      "train_runtime": 23382.5917,
      "train_tokens_per_second": 62692.345
    },
    {
      "epoch": 0.26258961734473524,
      "grad_norm": 23.375,
      "learning_rate": 1.7294787181564755e-05,
      "loss": 0.8568,
      "num_input_tokens_seen": 1467948544,
      "step": 19760,
      "train_runtime": 23414.9214,
      "train_tokens_per_second": 62692.867
    },
    {
      "epoch": 0.26272250682719717,
      "grad_norm": 13.625,
      "learning_rate": 1.7291842579333894e-05,
      "loss": 0.9205,
      "num_input_tokens_seen": 1468789248,
      "step": 19770,
      "train_runtime": 23427.4772,
      "train_tokens_per_second": 62695.152
    },
    {
      "epoch": 0.2628553963096591,
      "grad_norm": 14.0625,
      "learning_rate": 1.728889662635421e-05,
      "loss": 0.8162,
      "num_input_tokens_seen": 1469264640,
      "step": 19780,
      "train_runtime": 23434.2762,
      "train_tokens_per_second": 62697.249
    },
    {
      "epoch": 0.26298828579212097,
      "grad_norm": 21.125,
      "learning_rate": 1.7285949323171425e-05,
      "loss": 0.7401,
      "num_input_tokens_seen": 1469493760,
      "step": 19790,
      "train_runtime": 23438.1538,
      "train_tokens_per_second": 62696.652
    },
    {
      "epoch": 0.2631211752745829,
      "grad_norm": 21.5,
      "learning_rate": 1.728300067033149e-05,
      "loss": 0.5233,
      "num_input_tokens_seen": 1469611264,
      "step": 19800,
      "train_runtime": 23440.5269,
      "train_tokens_per_second": 62695.317
    },
    {
      "epoch": 0.2632540647570448,
      "grad_norm": 42.75,
      "learning_rate": 1.7280050668380625e-05,
      "loss": 0.7711,
      "num_input_tokens_seen": 1471723008,
      "step": 19810,
      "train_runtime": 23473.4353,
      "train_tokens_per_second": 62697.385
    },
    {
      "epoch": 0.2633869542395067,
      "grad_norm": 26.625,
      "learning_rate": 1.727709931786529e-05,
      "loss": 0.8713,
      "num_input_tokens_seen": 1472552960,
      "step": 19820,
      "train_runtime": 23485.8158,
      "train_tokens_per_second": 62699.673
    },
    {
      "epoch": 0.2635198437219686,
      "grad_norm": 32.25,
      "learning_rate": 1.7274146619332195e-05,
      "loss": 0.7551,
      "num_input_tokens_seen": 1473022208,
      "step": 19830,
      "train_runtime": 23492.5936,
      "train_tokens_per_second": 62701.557
    },
    {
      "epoch": 0.26365273320443056,
      "grad_norm": 15.125,
      "learning_rate": 1.72711925733283e-05,
      "loss": 0.6609,
      "num_input_tokens_seen": 1473255680,
      "step": 19840,
      "train_runtime": 23496.1339,
      "train_tokens_per_second": 62702.047
    },
    {
      "epoch": 0.26378562268689243,
      "grad_norm": 14.125,
      "learning_rate": 1.7268237180400816e-05,
      "loss": 0.5072,
      "num_input_tokens_seen": 1473373440,
      "step": 19850,
      "train_runtime": 23499.0026,
      "train_tokens_per_second": 62699.403
    },
    {
      "epoch": 0.26391851216935436,
      "grad_norm": 25.625,
      "learning_rate": 1.7265280441097206e-05,
      "loss": 0.8058,
      "num_input_tokens_seen": 1475460608,
      "step": 19860,
      "train_runtime": 23531.6748,
      "train_tokens_per_second": 62701.045
    },
    {
      "epoch": 0.2640514016518163,
      "grad_norm": 15.0,
      "learning_rate": 1.7262322355965175e-05,
      "loss": 0.9302,
      "num_input_tokens_seen": 1476270336,
      "step": 19870,
      "train_runtime": 23543.9961,
      "train_tokens_per_second": 62702.624
    },
    {
      "epoch": 0.26418429113427816,
      "grad_norm": 19.125,
      "learning_rate": 1.7259362925552683e-05,
      "loss": 0.7837,
      "num_input_tokens_seen": 1476730368,
      "step": 19880,
      "train_runtime": 23550.6372,
      "train_tokens_per_second": 62704.476
    },
    {
      "epoch": 0.2643171806167401,
      "grad_norm": 14.3125,
      "learning_rate": 1.7256402150407944e-05,
      "loss": 0.6813,
      "num_input_tokens_seen": 1476957696,
      "step": 19890,
      "train_runtime": 23554.1188,
      "train_tokens_per_second": 62704.859
    },
    {
      "epoch": 0.264450070099202,
      "grad_norm": 20.375,
      "learning_rate": 1.725344003107941e-05,
      "loss": 0.5928,
      "num_input_tokens_seen": 1477072640,
      "step": 19900,
      "train_runtime": 23556.9437,
      "train_tokens_per_second": 62702.219
    },
    {
      "epoch": 0.2645829595816639,
      "grad_norm": 22.625,
      "learning_rate": 1.7250476568115787e-05,
      "loss": 0.9015,
      "num_input_tokens_seen": 1479104512,
      "step": 19910,
      "train_runtime": 23589.2747,
      "train_tokens_per_second": 62702.416
    },
    {
      "epoch": 0.2647158490641258,
      "grad_norm": 24.125,
      "learning_rate": 1.7247511762066033e-05,
      "loss": 0.948,
      "num_input_tokens_seen": 1479975168,
      "step": 19920,
      "train_runtime": 23601.9544,
      "train_tokens_per_second": 62705.619
    },
    {
      "epoch": 0.26484873854658775,
      "grad_norm": 13.0,
      "learning_rate": 1.7244545613479347e-05,
      "loss": 0.7142,
      "num_input_tokens_seen": 1480443392,
      "step": 19930,
      "train_runtime": 23608.5354,
      "train_tokens_per_second": 62707.973
    },
    {
      "epoch": 0.2649816280290496,
      "grad_norm": 23.25,
      "learning_rate": 1.7241578122905186e-05,
      "loss": 0.597,
      "num_input_tokens_seen": 1480674560,
      "step": 19940,
      "train_runtime": 23612.4586,
      "train_tokens_per_second": 62707.344
    },
    {
      "epoch": 0.26511451751151155,
      "grad_norm": 10.0,
      "learning_rate": 1.7238609290893248e-05,
      "loss": 0.5715,
      "num_input_tokens_seen": 1480789248,
      "step": 19950,
      "train_runtime": 23614.904,
      "train_tokens_per_second": 62705.707
    },
    {
      "epoch": 0.2652474069939735,
      "grad_norm": 12.5,
      "learning_rate": 1.7235639117993487e-05,
      "loss": 0.8624,
      "num_input_tokens_seen": 1482764288,
      "step": 19960,
      "train_runtime": 23646.9892,
      "train_tokens_per_second": 62704.147
    },
    {
      "epoch": 0.26538029647643535,
      "grad_norm": 29.625,
      "learning_rate": 1.7232667604756098e-05,
      "loss": 0.9454,
      "num_input_tokens_seen": 1483611136,
      "step": 19970,
      "train_runtime": 23659.4606,
      "train_tokens_per_second": 62706.888
    },
    {
      "epoch": 0.2655131859588973,
      "grad_norm": 10.625,
      "learning_rate": 1.7229694751731528e-05,
      "loss": 0.7145,
      "num_input_tokens_seen": 1484097024,
      "step": 19980,
      "train_runtime": 23666.2427,
      "train_tokens_per_second": 62709.448
    },
    {
      "epoch": 0.2656460754413592,
      "grad_norm": 16.875,
      "learning_rate": 1.722672055947047e-05,
      "loss": 0.6094,
      "num_input_tokens_seen": 1484329472,
      "step": 19990,
      "train_runtime": 23669.7898,
      "train_tokens_per_second": 62709.871
    },
    {
      "epoch": 0.2657789649238211,
      "grad_norm": 6.75,
      "learning_rate": 1.7223745028523867e-05,
      "loss": 0.5845,
      "num_input_tokens_seen": 1484447232,
      "step": 20000,
      "train_runtime": 23672.1642,
      "train_tokens_per_second": 62708.556
    },
    {
      "epoch": 0.2657789649238211,
      "eval_accuracy": 0.8144,
      "eval_balanced_accuracy": 0.8176388342904386,
      "eval_invalid_f1": 0.825694966190834,
      "eval_invalid_precision": 0.8656951555730603,
      "eval_invalid_recall": 0.7892280071813286,
      "eval_loss": 0.3702210783958435,
      "eval_macro_f1": 0.8136173718892579,
      "eval_mcc": 0.6312129451061786,
      "eval_runtime": 77.8954,
      "eval_samples_per_second": 128.377,
      "eval_steps_per_second": 32.094,
      "eval_valid_f1": 0.8015397775876818,
      "eval_valid_precision": 0.7614790735473385,
      "eval_valid_recall": 0.8460496613995485,
      "num_input_tokens_seen": 1484447232,
      "step": 20000
    },
    {
      "epoch": 0.265911854406283,
      "grad_norm": 38.75,
      "learning_rate": 1.7220768159442912e-05,
      "loss": 0.8531,
      "num_input_tokens_seen": 1486541056,
      "step": 20010,
      "train_runtime": 23786.8315,
      "train_tokens_per_second": 62494.286
    },
    {
      "epoch": 0.26604474388874494,
      "grad_norm": 11.625,
      "learning_rate": 1.721778995277904e-05,
      "loss": 0.9209,
      "num_input_tokens_seen": 1487352832,
      "step": 20020,
      "train_runtime": 23799.1895,
      "train_tokens_per_second": 62495.945
    },
    {
      "epoch": 0.2661776333712068,
      "grad_norm": 25.25,
      "learning_rate": 1.721481040908394e-05,
      "loss": 0.8222,
      "num_input_tokens_seen": 1487843584,
      "step": 20030,
      "train_runtime": 23806.0179,
      "train_tokens_per_second": 62498.633
    },
    {
      "epoch": 0.26631052285366874,
      "grad_norm": 10.8125,
      "learning_rate": 1.7211829528909544e-05,
      "loss": 0.6505,
      "num_input_tokens_seen": 1488075776,
      "step": 20040,
      "train_runtime": 23809.5184,
      "train_tokens_per_second": 62499.197
    },
    {
      "epoch": 0.26644341233613067,
      "grad_norm": 8.6875,
      "learning_rate": 1.720884731280804e-05,
      "loss": 0.5052,
      "num_input_tokens_seen": 1488189184,
      "step": 20050,
      "train_runtime": 23811.8891,
      "train_tokens_per_second": 62497.737
    },
    {
      "epoch": 0.26657630181859254,
      "grad_norm": 22.375,
      "learning_rate": 1.7205863761331847e-05,
      "loss": 0.8509,
      "num_input_tokens_seen": 1490292736,
      "step": 20060,
      "train_runtime": 23845.5349,
      "train_tokens_per_second": 62497.769
    },
    {
      "epoch": 0.26670919130105447,
      "grad_norm": 23.5,
      "learning_rate": 1.7202878875033645e-05,
      "loss": 0.9074,
      "num_input_tokens_seen": 1491139328,
      "step": 20070,
      "train_runtime": 23858.2088,
      "train_tokens_per_second": 62500.054
    },
    {
      "epoch": 0.2668420807835164,
      "grad_norm": 11.75,
      "learning_rate": 1.719989265446636e-05,
      "loss": 0.7702,
      "num_input_tokens_seen": 1491588096,
      "step": 20080,
      "train_runtime": 23864.5998,
      "train_tokens_per_second": 62502.121
    },
    {
      "epoch": 0.2669749702659783,
      "grad_norm": 12.625,
      "learning_rate": 1.719690510018316e-05,
      "loss": 0.6703,
      "num_input_tokens_seen": 1491816448,
      "step": 20090,
      "train_runtime": 23868.045,
      "train_tokens_per_second": 62502.666
    },
    {
      "epoch": 0.2671078597484402,
      "grad_norm": 17.625,
      "learning_rate": 1.7193916212737463e-05,
      "loss": 0.5673,
      "num_input_tokens_seen": 1491932160,
      "step": 20100,
      "train_runtime": 23870.4158,
      "train_tokens_per_second": 62501.306
    },
    {
      "epoch": 0.26724074923090213,
      "grad_norm": 14.875,
      "learning_rate": 1.7190925992682936e-05,
      "loss": 0.8794,
      "num_input_tokens_seen": 1494065920,
      "step": 20110,
      "train_runtime": 23904.1475,
      "train_tokens_per_second": 62502.372
    },
    {
      "epoch": 0.267373638713364,
      "grad_norm": 14.4375,
      "learning_rate": 1.718793444057349e-05,
      "loss": 0.8977,
      "num_input_tokens_seen": 1494937600,
      "step": 20120,
      "train_runtime": 23916.9679,
      "train_tokens_per_second": 62505.314
    },
    {
      "epoch": 0.26750652819582593,
      "grad_norm": 10.9375,
      "learning_rate": 1.718494155696328e-05,
      "loss": 0.8162,
      "num_input_tokens_seen": 1495393280,
      "step": 20130,
      "train_runtime": 23924.2902,
      "train_tokens_per_second": 62505.231
    },
    {
      "epoch": 0.26763941767828786,
      "grad_norm": 15.25,
      "learning_rate": 1.7181947342406714e-05,
      "loss": 0.5803,
      "num_input_tokens_seen": 1495625216,
      "step": 20140,
      "train_runtime": 23927.7724,
      "train_tokens_per_second": 62505.828
    },
    {
      "epoch": 0.2677723071607498,
      "grad_norm": 23.375,
      "learning_rate": 1.7178951797458445e-05,
      "loss": 0.5542,
      "num_input_tokens_seen": 1495739904,
      "step": 20150,
      "train_runtime": 23930.357,
      "train_tokens_per_second": 62503.869
    },
    {
      "epoch": 0.26790519664321166,
      "grad_norm": 26.625,
      "learning_rate": 1.7175954922673366e-05,
      "loss": 0.7848,
      "num_input_tokens_seen": 1497753088,
      "step": 20160,
      "train_runtime": 23963.3104,
      "train_tokens_per_second": 62501.927
    },
    {
      "epoch": 0.2680380861256736,
      "grad_norm": 22.0,
      "learning_rate": 1.7172956718606628e-05,
      "loss": 0.9183,
      "num_input_tokens_seen": 1498590976,
      "step": 20170,
      "train_runtime": 23975.8394,
      "train_tokens_per_second": 62504.213
    },
    {
      "epoch": 0.2681709756081355,
      "grad_norm": 26.5,
      "learning_rate": 1.7169957185813617e-05,
      "loss": 0.732,
      "num_input_tokens_seen": 1499065600,
      "step": 20180,
      "train_runtime": 23983.0427,
      "train_tokens_per_second": 62505.23
    },
    {
      "epoch": 0.2683038650905974,
      "grad_norm": 12.5625,
      "learning_rate": 1.716695632484997e-05,
      "loss": 0.6505,
      "num_input_tokens_seen": 1499299840,
      "step": 20190,
      "train_runtime": 23986.6044,
      "train_tokens_per_second": 62505.714
    },
    {
      "epoch": 0.2684367545730593,
      "grad_norm": 33.25,
      "learning_rate": 1.716395413627157e-05,
      "loss": 0.5092,
      "num_input_tokens_seen": 1499417856,
      "step": 20200,
      "train_runtime": 23989.0206,
      "train_tokens_per_second": 62504.338
    },
    {
      "epoch": 0.26856964405552125,
      "grad_norm": 14.5625,
      "learning_rate": 1.7160950620634546e-05,
      "loss": 0.8175,
      "num_input_tokens_seen": 1501445888,
      "step": 20210,
      "train_runtime": 24022.0104,
      "train_tokens_per_second": 62502.924
    },
    {
      "epoch": 0.2687025335379831,
      "grad_norm": 18.125,
      "learning_rate": 1.7157945778495273e-05,
      "loss": 0.8853,
      "num_input_tokens_seen": 1502264064,
      "step": 20220,
      "train_runtime": 24034.8523,
      "train_tokens_per_second": 62503.57
    },
    {
      "epoch": 0.26883542302044505,
      "grad_norm": 15.8125,
      "learning_rate": 1.715493961041037e-05,
      "loss": 0.7053,
      "num_input_tokens_seen": 1502707968,
      "step": 20230,
      "train_runtime": 24041.2985,
      "train_tokens_per_second": 62505.275
    },
    {
      "epoch": 0.268968312502907,
      "grad_norm": 14.5625,
      "learning_rate": 1.7151932116936705e-05,
      "loss": 0.7149,
      "num_input_tokens_seen": 1502932736,
      "step": 20240,
      "train_runtime": 24044.6831,
      "train_tokens_per_second": 62505.824
    },
    {
      "epoch": 0.26910120198536885,
      "grad_norm": 32.25,
      "learning_rate": 1.7148923298631387e-05,
      "loss": 0.619,
      "num_input_tokens_seen": 1503049216,
      "step": 20250,
      "train_runtime": 24047.0622,
      "train_tokens_per_second": 62504.484
    },
    {
      "epoch": 0.2692340914678308,
      "grad_norm": 19.125,
      "learning_rate": 1.7145913156051774e-05,
      "loss": 0.787,
      "num_input_tokens_seen": 1505134592,
      "step": 20260,
      "train_runtime": 24080.6842,
      "train_tokens_per_second": 62503.813
    },
    {
      "epoch": 0.2693669809502927,
      "grad_norm": 14.75,
      "learning_rate": 1.7142901689755467e-05,
      "loss": 0.8607,
      "num_input_tokens_seen": 1505981440,
      "step": 20270,
      "train_runtime": 24093.0453,
      "train_tokens_per_second": 62506.894
    },
    {
      "epoch": 0.2694998704327546,
      "grad_norm": 21.75,
      "learning_rate": 1.7139888900300314e-05,
      "loss": 0.7748,
      "num_input_tokens_seen": 1506431232,
      "step": 20280,
      "train_runtime": 24099.6023,
      "train_tokens_per_second": 62508.552
    },
    {
      "epoch": 0.2696327599152165,
      "grad_norm": 14.8125,
      "learning_rate": 1.713687478824441e-05,
      "loss": 0.6223,
      "num_input_tokens_seen": 1506661376,
      "step": 20290,
      "train_runtime": 24103.0167,
      "train_tokens_per_second": 62509.245
    },
    {
      "epoch": 0.26976564939767844,
      "grad_norm": 9.0625,
      "learning_rate": 1.7133859354146088e-05,
      "loss": 0.4828,
      "num_input_tokens_seen": 1506774784,
      "step": 20300,
      "train_runtime": 24105.9165,
      "train_tokens_per_second": 62506.43
    },
    {
      "epoch": 0.2698985388801403,
      "grad_norm": 20.125,
      "learning_rate": 1.7130842598563933e-05,
      "loss": 0.8422,
      "num_input_tokens_seen": 1508752384,
      "step": 20310,
      "train_runtime": 24137.4238,
      "train_tokens_per_second": 62506.769
    },
    {
      "epoch": 0.27003142836260224,
      "grad_norm": 18.625,
      "learning_rate": 1.7127824522056768e-05,
      "loss": 0.8952,
      "num_input_tokens_seen": 1509610752,
      "step": 20320,
      "train_runtime": 24150.033,
      "train_tokens_per_second": 62509.677
    },
    {
      "epoch": 0.2701643178450642,
      "grad_norm": 17.0,
      "learning_rate": 1.712480512518367e-05,
      "loss": 0.7904,
      "num_input_tokens_seen": 1510099712,
      "step": 20330,
      "train_runtime": 24156.7192,
      "train_tokens_per_second": 62512.616
    },
    {
      "epoch": 0.27029720732752605,
      "grad_norm": 8.8125,
      "learning_rate": 1.7121784408503955e-05,
      "loss": 0.6595,
      "num_input_tokens_seen": 1510340352,
      "step": 20340,
      "train_runtime": 24160.2659,
      "train_tokens_per_second": 62513.4
    },
    {
      "epoch": 0.270430096809988,
      "grad_norm": 39.25,
      "learning_rate": 1.711876237257718e-05,
      "loss": 0.5352,
      "num_input_tokens_seen": 1510458112,
      "step": 20350,
      "train_runtime": 24162.6839,
      "train_tokens_per_second": 62512.017
    },
    {
      "epoch": 0.2705629862924499,
      "grad_norm": 20.5,
      "learning_rate": 1.7115739017963153e-05,
      "loss": 0.8388,
      "num_input_tokens_seen": 1512418048,
      "step": 20360,
      "train_runtime": 24194.1551,
      "train_tokens_per_second": 62511.712
    },
    {
      "epoch": 0.2706958757749118,
      "grad_norm": 27.25,
      "learning_rate": 1.7112714345221923e-05,
      "loss": 0.8907,
      "num_input_tokens_seen": 1513266688,
      "step": 20370,
      "train_runtime": 24206.6806,
      "train_tokens_per_second": 62514.424
    },
    {
      "epoch": 0.2708287652573737,
      "grad_norm": 13.25,
      "learning_rate": 1.710968835491378e-05,
      "loss": 0.719,
      "num_input_tokens_seen": 1513747712,
      "step": 20380,
      "train_runtime": 24213.5747,
      "train_tokens_per_second": 62516.491
    },
    {
      "epoch": 0.27096165473983563,
      "grad_norm": 35.25,
      "learning_rate": 1.710666104759927e-05,
      "loss": 0.713,
      "num_input_tokens_seen": 1513986304,
      "step": 20390,
      "train_runtime": 24217.1593,
      "train_tokens_per_second": 62517.089
    },
    {
      "epoch": 0.2710945442222975,
      "grad_norm": 22.125,
      "learning_rate": 1.710363242383917e-05,
      "loss": 0.5171,
      "num_input_tokens_seen": 1514102016,
      "step": 20400,
      "train_runtime": 24220.0564,
      "train_tokens_per_second": 62514.389
    },
    {
      "epoch": 0.27122743370475944,
      "grad_norm": 33.5,
      "learning_rate": 1.71006024841945e-05,
      "loss": 0.8911,
      "num_input_tokens_seen": 1516160256,
      "step": 20410,
      "train_runtime": 24252.0153,
      "train_tokens_per_second": 62516.877
    },
    {
      "epoch": 0.27136032318722136,
      "grad_norm": 20.25,
      "learning_rate": 1.709757122922654e-05,
      "loss": 0.8966,
      "num_input_tokens_seen": 1516973824,
      "step": 20420,
      "train_runtime": 24264.3731,
      "train_tokens_per_second": 62518.567
    },
    {
      "epoch": 0.27149321266968324,
      "grad_norm": 13.5625,
      "learning_rate": 1.7094538659496794e-05,
      "loss": 0.7309,
      "num_input_tokens_seen": 1517454592,
      "step": 20430,
      "train_runtime": 24271.0519,
      "train_tokens_per_second": 62521.171
    },
    {
      "epoch": 0.27162610215214517,
      "grad_norm": 9.1875,
      "learning_rate": 1.7091504775567024e-05,
      "loss": 0.6562,
      "num_input_tokens_seen": 1517681152,
      "step": 20440,
      "train_runtime": 24274.5046,
      "train_tokens_per_second": 62521.612
    },
    {
      "epoch": 0.2717589916346071,
      "grad_norm": 19.75,
      "learning_rate": 1.7088469577999233e-05,
      "loss": 0.5935,
      "num_input_tokens_seen": 1517795328,
      "step": 20450,
      "train_runtime": 24276.8822,
      "train_tokens_per_second": 62520.192
    },
    {
      "epoch": 0.27189188111706897,
      "grad_norm": 16.375,
      "learning_rate": 1.708543306735565e-05,
      "loss": 0.8502,
      "num_input_tokens_seen": 1519836672,
      "step": 20460,
      "train_runtime": 24308.6531,
      "train_tokens_per_second": 62522.455
    },
    {
      "epoch": 0.2720247705995309,
      "grad_norm": 31.125,
      "learning_rate": 1.7082395244198776e-05,
      "loss": 0.9431,
      "num_input_tokens_seen": 1520681216,
      "step": 20470,
      "train_runtime": 24321.7404,
      "train_tokens_per_second": 62523.536
    },
    {
      "epoch": 0.2721576600819928,
      "grad_norm": 14.4375,
      "learning_rate": 1.7079356109091336e-05,
      "loss": 0.7314,
      "num_input_tokens_seen": 1521145088,
      "step": 20480,
      "train_runtime": 24328.4392,
      "train_tokens_per_second": 62525.388
    },
    {
      "epoch": 0.2722905495644547,
      "grad_norm": 12.125,
      "learning_rate": 1.7076315662596302e-05,
      "loss": 0.6305,
      "num_input_tokens_seen": 1521375744,
      "step": 20490,
      "train_runtime": 24331.932,
      "train_tokens_per_second": 62525.892
    },
    {
      "epoch": 0.2724234390469166,
      "grad_norm": 23.375,
      "learning_rate": 1.707327390527689e-05,
      "loss": 0.5379,
      "num_input_tokens_seen": 1521492736,
      "step": 20500,
      "train_runtime": 24334.2924,
      "train_tokens_per_second": 62524.634
    },
    {
      "epoch": 0.27255632852937856,
      "grad_norm": 21.375,
      "learning_rate": 1.7070230837696557e-05,
      "loss": 0.8929,
      "num_input_tokens_seen": 1523563264,
      "step": 20510,
      "train_runtime": 24368.6996,
      "train_tokens_per_second": 62521.32
    },
    {
      "epoch": 0.27268921801184043,
      "grad_norm": 30.0,
      "learning_rate": 1.7067186460419005e-05,
      "loss": 0.9244,
      "num_input_tokens_seen": 1524401664,
      "step": 20520,
      "train_runtime": 24381.4754,
      "train_tokens_per_second": 62522.946
    },
    {
      "epoch": 0.27282210749430236,
      "grad_norm": 19.875,
      "learning_rate": 1.706414077400818e-05,
      "loss": 0.7795,
      "num_input_tokens_seen": 1524891136,
      "step": 20530,
      "train_runtime": 24388.2236,
      "train_tokens_per_second": 62525.716
    },
    {
      "epoch": 0.2729549969767643,
      "grad_norm": 25.5,
      "learning_rate": 1.7061093779028266e-05,
      "loss": 0.6529,
      "num_input_tokens_seen": 1525117184,
      "step": 20540,
      "train_runtime": 24391.7398,
      "train_tokens_per_second": 62525.97
    },
    {
      "epoch": 0.27308788645922616,
      "grad_norm": 11.5,
      "learning_rate": 1.705804547604369e-05,
      "loss": 0.5385,
      "num_input_tokens_seen": 1525232640,
      "step": 20550,
      "train_runtime": 24394.1392,
      "train_tokens_per_second": 62524.553
    },
    {
      "epoch": 0.2732207759416881,
      "grad_norm": 13.0625,
      "learning_rate": 1.7054995865619125e-05,
      "loss": 0.8391,
      "num_input_tokens_seen": 1527290880,
      "step": 20560,
      "train_runtime": 24429.1806,
      "train_tokens_per_second": 62519.12
    },
    {
      "epoch": 0.27335366542415,
      "grad_norm": 16.5,
      "learning_rate": 1.705194494831949e-05,
      "loss": 0.8725,
      "num_input_tokens_seen": 1528124160,
      "step": 20570,
      "train_runtime": 24442.0048,
      "train_tokens_per_second": 62520.41
    },
    {
      "epoch": 0.2734865549066119,
      "grad_norm": 11.375,
      "learning_rate": 1.7048892724709928e-05,
      "loss": 0.6849,
      "num_input_tokens_seen": 1528632576,
      "step": 20580,
      "train_runtime": 24449.0755,
      "train_tokens_per_second": 62523.124
    },
    {
      "epoch": 0.2736194443890738,
      "grad_norm": 18.125,
      "learning_rate": 1.7045839195355846e-05,
      "loss": 0.7197,
      "num_input_tokens_seen": 1528865536,
      "step": 20590,
      "train_runtime": 24452.6929,
      "train_tokens_per_second": 62523.402
    },
    {
      "epoch": 0.27375233387153575,
      "grad_norm": 10.5,
      "learning_rate": 1.7042784360822878e-05,
      "loss": 0.5403,
      "num_input_tokens_seen": 1528985088,
      "step": 20600,
      "train_runtime": 24455.3153,
      "train_tokens_per_second": 62521.586
    },
    {
      "epoch": 0.2738852233539977,
      "grad_norm": 25.625,
      "learning_rate": 1.7039728221676903e-05,
      "loss": 0.8597,
      "num_input_tokens_seen": 1530918912,
      "step": 20610,
      "train_runtime": 24486.2986,
      "train_tokens_per_second": 62521.451
    },
    {
      "epoch": 0.27401811283645955,
      "grad_norm": 95.0,
      "learning_rate": 1.7036670778484052e-05,
      "loss": 0.9554,
      "num_input_tokens_seen": 1531760384,
      "step": 20620,
      "train_runtime": 24498.8415,
      "train_tokens_per_second": 62523.788
    },
    {
      "epoch": 0.2741510023189215,
      "grad_norm": 18.125,
      "learning_rate": 1.7033612031810683e-05,
      "loss": 0.7516,
      "num_input_tokens_seen": 1532229376,
      "step": 20630,
      "train_runtime": 24505.4951,
      "train_tokens_per_second": 62525.951
    },
    {
      "epoch": 0.2742838918013834,
      "grad_norm": 20.5,
      "learning_rate": 1.7030551982223406e-05,
      "loss": 0.6792,
      "num_input_tokens_seen": 1532467456,
      "step": 20640,
      "train_runtime": 24509.4344,
      "train_tokens_per_second": 62525.615
    },
    {
      "epoch": 0.2744167812838453,
      "grad_norm": 20.75,
      "learning_rate": 1.702749063028906e-05,
      "loss": 0.5018,
      "num_input_tokens_seen": 1532582912,
      "step": 20650,
      "train_runtime": 24511.8394,
      "train_tokens_per_second": 62524.19
    },
    {
      "epoch": 0.2745496707663072,
      "grad_norm": 14.0625,
      "learning_rate": 1.7024427976574737e-05,
      "loss": 0.7919,
      "num_input_tokens_seen": 1534592512,
      "step": 20660,
      "train_runtime": 24545.4188,
      "train_tokens_per_second": 62520.527
    },
    {
      "epoch": 0.27468256024876914,
      "grad_norm": 26.125,
      "learning_rate": 1.7021364021647768e-05,
      "loss": 0.8506,
      "num_input_tokens_seen": 1535440896,
      "step": 20670,
      "train_runtime": 24557.9374,
      "train_tokens_per_second": 62523.203
    },
    {
      "epoch": 0.274815449731231,
      "grad_norm": 18.125,
      "learning_rate": 1.7018298766075724e-05,
      "loss": 0.7724,
      "num_input_tokens_seen": 1535909632,
      "step": 20680,
      "train_runtime": 24564.5873,
      "train_tokens_per_second": 62525.359
    },
    {
      "epoch": 0.27494833921369294,
      "grad_norm": 12.625,
      "learning_rate": 1.7015232210426413e-05,
      "loss": 0.6188,
      "num_input_tokens_seen": 1536138240,
      "step": 20690,
      "train_runtime": 24568.5725,
      "train_tokens_per_second": 62524.522
    },
    {
      "epoch": 0.27508122869615487,
      "grad_norm": 31.625,
      "learning_rate": 1.701216435526789e-05,
      "loss": 0.4968,
      "num_input_tokens_seen": 1536255488,
      "step": 20700,
      "train_runtime": 24571.0737,
      "train_tokens_per_second": 62522.929
    },
    {
      "epoch": 0.27521411817861674,
      "grad_norm": 19.5,
      "learning_rate": 1.7009095201168447e-05,
      "loss": 0.8138,
      "num_input_tokens_seen": 1538329600,
      "step": 20710,
      "train_runtime": 24604.3486,
      "train_tokens_per_second": 62522.671
    },
    {
      "epoch": 0.27534700766107867,
      "grad_norm": 18.75,
      "learning_rate": 1.7006024748696615e-05,
      "loss": 0.8194,
      "num_input_tokens_seen": 1539153408,
      "step": 20720,
      "train_runtime": 24616.6489,
      "train_tokens_per_second": 62524.896
    },
    {
      "epoch": 0.2754798971435406,
      "grad_norm": 14.0,
      "learning_rate": 1.7002952998421172e-05,
      "loss": 0.8137,
      "num_input_tokens_seen": 1539636736,
      "step": 20730,
      "train_runtime": 24623.284,
      "train_tokens_per_second": 62527.676
    },
    {
      "epoch": 0.27561278662600247,
      "grad_norm": 12.5625,
      "learning_rate": 1.6999879950911125e-05,
      "loss": 0.6119,
      "num_input_tokens_seen": 1539872512,
      "step": 20740,
      "train_runtime": 24626.7635,
      "train_tokens_per_second": 62528.416
    },
    {
      "epoch": 0.2757456761084644,
      "grad_norm": 18.875,
      "learning_rate": 1.6996805606735738e-05,
      "loss": 0.5029,
      "num_input_tokens_seen": 1539989248,
      "step": 20750,
      "train_runtime": 24629.1697,
      "train_tokens_per_second": 62527.047
    },
    {
      "epoch": 0.27587856559092633,
      "grad_norm": 20.375,
      "learning_rate": 1.6993729966464504e-05,
      "loss": 0.9039,
      "num_input_tokens_seen": 1541999104,
      "step": 20760,
      "train_runtime": 24660.0824,
      "train_tokens_per_second": 62530.168
    },
    {
      "epoch": 0.2760114550733882,
      "grad_norm": 25.25,
      "learning_rate": 1.699065303066715e-05,
      "loss": 0.8748,
      "num_input_tokens_seen": 1542806016,
      "step": 20770,
      "train_runtime": 24672.6659,
      "train_tokens_per_second": 62530.982
    },
    {
      "epoch": 0.27614434455585013,
      "grad_norm": 12.25,
      "learning_rate": 1.698757479991366e-05,
      "loss": 0.722,
      "num_input_tokens_seen": 1543259136,
      "step": 20780,
      "train_runtime": 24679.1105,
      "train_tokens_per_second": 62533.013
    },
    {
      "epoch": 0.27627723403831206,
      "grad_norm": 13.6875,
      "learning_rate": 1.6984495274774245e-05,
      "loss": 0.6298,
      "num_input_tokens_seen": 1543485440,
      "step": 20790,
      "train_runtime": 24682.508,
      "train_tokens_per_second": 62533.574
    },
    {
      "epoch": 0.27641012352077393,
      "grad_norm": 37.0,
      "learning_rate": 1.6981414455819364e-05,
      "loss": 0.5139,
      "num_input_tokens_seen": 1543599104,
      "step": 20800,
      "train_runtime": 24684.8968,
      "train_tokens_per_second": 62532.127
    },
    {
      "epoch": 0.27654301300323586,
      "grad_norm": 27.75,
      "learning_rate": 1.6978332343619702e-05,
      "loss": 0.7702,
      "num_input_tokens_seen": 1545613312,
      "step": 20810,
      "train_runtime": 24717.0389,
      "train_tokens_per_second": 62532.301
    },
    {
      "epoch": 0.2766759024856978,
      "grad_norm": 18.625,
      "learning_rate": 1.69752489387462e-05,
      "loss": 0.8596,
      "num_input_tokens_seen": 1546432768,
      "step": 20820,
      "train_runtime": 24729.3023,
      "train_tokens_per_second": 62534.428
    },
    {
      "epoch": 0.27680879196815966,
      "grad_norm": 25.375,
      "learning_rate": 1.6972164241770025e-05,
      "loss": 0.7933,
      "num_input_tokens_seen": 1546895104,
      "step": 20830,
      "train_runtime": 24736.2646,
      "train_tokens_per_second": 62535.517
    },
    {
      "epoch": 0.2769416814506216,
      "grad_norm": 24.375,
      "learning_rate": 1.69690782532626e-05,
      "loss": 0.651,
      "num_input_tokens_seen": 1547121920,
      "step": 20840,
      "train_runtime": 24739.706,
      "train_tokens_per_second": 62535.986
    },
    {
      "epoch": 0.2770745709330835,
      "grad_norm": 16.625,
      "learning_rate": 1.6965990973795564e-05,
      "loss": 0.5081,
      "num_input_tokens_seen": 1547240192,
      "step": 20850,
      "train_runtime": 24742.1137,
      "train_tokens_per_second": 62534.681
    },
    {
      "epoch": 0.2772074604155454,
      "grad_norm": 18.0,
      "learning_rate": 1.696290240394082e-05,
      "loss": 0.8174,
      "num_input_tokens_seen": 1549321216,
      "step": 20860,
      "train_runtime": 24775.3736,
      "train_tokens_per_second": 62534.727
    },
    {
      "epoch": 0.2773403498980073,
      "grad_norm": 10.5625,
      "learning_rate": 1.6959812544270493e-05,
      "loss": 0.8363,
      "num_input_tokens_seen": 1550184192,
      "step": 20870,
      "train_runtime": 24788.532,
      "train_tokens_per_second": 62536.345
    },
    {
      "epoch": 0.27747323938046925,
      "grad_norm": 35.25,
      "learning_rate": 1.6956721395356952e-05,
      "loss": 0.7161,
      "num_input_tokens_seen": 1550681344,
      "step": 20880,
      "train_runtime": 24795.2555,
      "train_tokens_per_second": 62539.438
    },
    {
      "epoch": 0.2776061288629311,
      "grad_norm": 23.625,
      "learning_rate": 1.69536289577728e-05,
      "loss": 0.5571,
      "num_input_tokens_seen": 1550911744,
      "step": 20890,
      "train_runtime": 24798.763,
      "train_tokens_per_second": 62539.883
    },
    {
      "epoch": 0.27773901834539305,
      "grad_norm": 8.75,
      "learning_rate": 1.6950535232090892e-05,
      "loss": 0.6117,
      "num_input_tokens_seen": 1551030784,
      "step": 20900,
      "train_runtime": 24801.1864,
      "train_tokens_per_second": 62538.572
    },
    {
      "epoch": 0.277871907827855,
      "grad_norm": 30.75,
      "learning_rate": 1.6947440218884307e-05,
      "loss": 0.8375,
      "num_input_tokens_seen": 1553060352,
      "step": 20910,
      "train_runtime": 24834.3636,
      "train_tokens_per_second": 62536.749
    },
    {
      "epoch": 0.27800479731031685,
      "grad_norm": 49.0,
      "learning_rate": 1.694434391872637e-05,
      "loss": 0.8807,
      "num_input_tokens_seen": 1553893376,
      "step": 20920,
      "train_runtime": 24847.2217,
      "train_tokens_per_second": 62537.913
    },
    {
      "epoch": 0.2781376867927788,
      "grad_norm": 12.4375,
      "learning_rate": 1.6941246332190644e-05,
      "loss": 0.6903,
      "num_input_tokens_seen": 1554393088,
      "step": 20930,
      "train_runtime": 24854.0193,
      "train_tokens_per_second": 62540.914
    },
    {
      "epoch": 0.2782705762752407,
      "grad_norm": 19.25,
      "learning_rate": 1.6938147459850933e-05,
      "loss": 0.7119,
      "num_input_tokens_seen": 1554613504,
      "step": 20940,
      "train_runtime": 24857.4373,
      "train_tokens_per_second": 62541.182
    },
    {
      "epoch": 0.2784034657577026,
      "grad_norm": 14.25,
      "learning_rate": 1.6935047302281266e-05,
      "loss": 0.5582,
      "num_input_tokens_seen": 1554728448,
      "step": 20950,
      "train_runtime": 24860.3286,
      "train_tokens_per_second": 62538.532
    },
    {
      "epoch": 0.2785363552401645,
      "grad_norm": 29.625,
      "learning_rate": 1.6931945860055926e-05,
      "loss": 0.8211,
      "num_input_tokens_seen": 1556623616,
      "step": 20960,
      "train_runtime": 24891.5029,
      "train_tokens_per_second": 62536.345
    },
    {
      "epoch": 0.27866924472262644,
      "grad_norm": 15.6875,
      "learning_rate": 1.6928843133749427e-05,
      "loss": 0.9361,
      "num_input_tokens_seen": 1557494528,
      "step": 20970,
      "train_runtime": 24904.232,
      "train_tokens_per_second": 62539.352
    },
    {
      "epoch": 0.2788021342050883,
      "grad_norm": 18.375,
      "learning_rate": 1.6925739123936526e-05,
      "loss": 0.827,
      "num_input_tokens_seen": 1557962240,
      "step": 20980,
      "train_runtime": 24910.8974,
      "train_tokens_per_second": 62541.394
    },
    {
      "epoch": 0.27893502368755024,
      "grad_norm": 13.3125,
      "learning_rate": 1.69226338311922e-05,
      "loss": 0.6477,
      "num_input_tokens_seen": 1558196480,
      "step": 20990,
      "train_runtime": 24914.4274,
      "train_tokens_per_second": 62541.934
    },
    {
      "epoch": 0.27906791317001217,
      "grad_norm": 7.71875,
      "learning_rate": 1.6919527256091687e-05,
      "loss": 0.5057,
      "num_input_tokens_seen": 1558313728,
      "step": 21000,
      "train_runtime": 24916.8569,
      "train_tokens_per_second": 62540.542
    },
    {
      "epoch": 0.27920080265247404,
      "grad_norm": 66.0,
      "learning_rate": 1.691641939921045e-05,
      "loss": 0.8394,
      "num_input_tokens_seen": 1560412928,
      "step": 21010,
      "train_runtime": 24950.1775,
      "train_tokens_per_second": 62541.155
    },
    {
      "epoch": 0.279333692134936,
      "grad_norm": 47.5,
      "learning_rate": 1.6913310261124195e-05,
      "loss": 0.9356,
      "num_input_tokens_seen": 1561249024,
      "step": 21020,
      "train_runtime": 24962.748,
      "train_tokens_per_second": 62543.155
    },
    {
      "epoch": 0.2794665816173979,
      "grad_norm": 16.25,
      "learning_rate": 1.6910199842408856e-05,
      "loss": 0.7292,
      "num_input_tokens_seen": 1561707264,
      "step": 21030,
      "train_runtime": 24969.3639,
      "train_tokens_per_second": 62544.936
    },
    {
      "epoch": 0.2795994710998598,
      "grad_norm": 16.25,
      "learning_rate": 1.6907088143640616e-05,
      "loss": 0.6314,
      "num_input_tokens_seen": 1561944576,
      "step": 21040,
      "train_runtime": 24972.8746,
      "train_tokens_per_second": 62545.646
    },
    {
      "epoch": 0.2797323605823217,
      "grad_norm": 12.0625,
      "learning_rate": 1.6903975165395886e-05,
      "loss": 0.521,
      "num_input_tokens_seen": 1562062336,
      "step": 21050,
      "train_runtime": 24975.8133,
      "train_tokens_per_second": 62543.002
    },
    {
      "epoch": 0.27986525006478363,
      "grad_norm": 29.5,
      "learning_rate": 1.6900860908251315e-05,
      "loss": 0.8545,
      "num_input_tokens_seen": 1564044800,
      "step": 21060,
      "train_runtime": 25007.4337,
      "train_tokens_per_second": 62543.195
    },
    {
      "epoch": 0.2799981395472455,
      "grad_norm": 14.1875,
      "learning_rate": 1.6897745372783797e-05,
      "loss": 0.9361,
      "num_input_tokens_seen": 1564877312,
      "step": 21070,
      "train_runtime": 25020.0119,
      "train_tokens_per_second": 62545.027
    },
    {
      "epoch": 0.28013102902970743,
      "grad_norm": 22.25,
      "learning_rate": 1.6894628559570455e-05,
      "loss": 0.7487,
      "num_input_tokens_seen": 1565327616,
      "step": 21080,
      "train_runtime": 25026.4841,
      "train_tokens_per_second": 62546.845
    },
    {
      "epoch": 0.28026391851216936,
      "grad_norm": 18.125,
      "learning_rate": 1.689151046918865e-05,
      "loss": 0.5864,
      "num_input_tokens_seen": 1565553408,
      "step": 21090,
      "train_runtime": 25029.8642,
      "train_tokens_per_second": 62547.419
    },
    {
      "epoch": 0.2803968079946313,
      "grad_norm": 16.875,
      "learning_rate": 1.688839110221598e-05,
      "loss": 0.523,
      "num_input_tokens_seen": 1565665792,
      "step": 21100,
      "train_runtime": 25032.2407,
      "train_tokens_per_second": 62545.971
    },
    {
      "epoch": 0.28052969747709317,
      "grad_norm": 23.5,
      "learning_rate": 1.6885270459230283e-05,
      "loss": 0.8648,
      "num_input_tokens_seen": 1567878400,
      "step": 21110,
      "train_runtime": 25066.4495,
      "train_tokens_per_second": 62548.882
    },
    {
      "epoch": 0.2806625869595551,
      "grad_norm": 19.875,
      "learning_rate": 1.688214854080963e-05,
      "loss": 0.926,
      "num_input_tokens_seen": 1568744960,
      "step": 21120,
      "train_runtime": 25079.7565,
      "train_tokens_per_second": 62550.247
    },
    {
      "epoch": 0.280795476442017,
      "grad_norm": 14.4375,
      "learning_rate": 1.6879025347532323e-05,
      "loss": 0.7687,
      "num_input_tokens_seen": 1569226240,
      "step": 21130,
      "train_runtime": 25086.6228,
      "train_tokens_per_second": 62552.311
    },
    {
      "epoch": 0.2809283659244789,
      "grad_norm": 13.75,
      "learning_rate": 1.6875900879976914e-05,
      "loss": 0.6296,
      "num_input_tokens_seen": 1569470976,
      "step": 21140,
      "train_runtime": 25090.1975,
      "train_tokens_per_second": 62553.153
    },
    {
      "epoch": 0.2810612554069408,
      "grad_norm": 8.1875,
      "learning_rate": 1.6872775138722174e-05,
      "loss": 0.5198,
      "num_input_tokens_seen": 1569591040,
      "step": 21150,
      "train_runtime": 25092.6353,
      "train_tokens_per_second": 62551.861
    },
    {
      "epoch": 0.28119414488940275,
      "grad_norm": 25.875,
      "learning_rate": 1.6869648124347123e-05,
      "loss": 0.8199,
      "num_input_tokens_seen": 1571650560,
      "step": 21160,
      "train_runtime": 25126.6958,
      "train_tokens_per_second": 62549.034
    },
    {
      "epoch": 0.2813270343718646,
      "grad_norm": 23.25,
      "learning_rate": 1.6866519837431013e-05,
      "loss": 0.9368,
      "num_input_tokens_seen": 1572472576,
      "step": 21170,
      "train_runtime": 25139.7515,
      "train_tokens_per_second": 62549.249
    },
    {
      "epoch": 0.28145992385432655,
      "grad_norm": 15.3125,
      "learning_rate": 1.6863390278553332e-05,
      "loss": 0.7764,
      "num_input_tokens_seen": 1572957184,
      "step": 21180,
      "train_runtime": 25146.4675,
      "train_tokens_per_second": 62551.815
    },
    {
      "epoch": 0.2815928133367885,
      "grad_norm": 10.125,
      "learning_rate": 1.68602594482938e-05,
      "loss": 0.5955,
      "num_input_tokens_seen": 1573188864,
      "step": 21190,
      "train_runtime": 25149.9537,
      "train_tokens_per_second": 62552.356
    },
    {
      "epoch": 0.28172570281925036,
      "grad_norm": 29.125,
      "learning_rate": 1.6857127347232377e-05,
      "loss": 0.5672,
      "num_input_tokens_seen": 1573305088,
      "step": 21200,
      "train_runtime": 25152.3092,
      "train_tokens_per_second": 62551.119
    },
    {
      "epoch": 0.2818585923017123,
      "grad_norm": 27.0,
      "learning_rate": 1.685399397594926e-05,
      "loss": 0.8346,
      "num_input_tokens_seen": 1575355136,
      "step": 21210,
      "train_runtime": 25185.7345,
      "train_tokens_per_second": 62549.501
    },
    {
      "epoch": 0.2819914817841742,
      "grad_norm": 23.0,
      "learning_rate": 1.685085933502487e-05,
      "loss": 0.9237,
      "num_input_tokens_seen": 1576216576,
      "step": 21220,
      "train_runtime": 25198.529,
      "train_tokens_per_second": 62551.928
    },
    {
      "epoch": 0.2821243712666361,
      "grad_norm": 17.625,
      "learning_rate": 1.684772342503988e-05,
      "loss": 0.7173,
      "num_input_tokens_seen": 1576693504,
      "step": 21230,
      "train_runtime": 25205.2429,
      "train_tokens_per_second": 62554.188
    },
    {
      "epoch": 0.282257260749098,
      "grad_norm": 52.25,
      "learning_rate": 1.6844586246575184e-05,
      "loss": 0.6676,
      "num_input_tokens_seen": 1576918272,
      "step": 21240,
      "train_runtime": 25208.6917,
      "train_tokens_per_second": 62554.546
    },
    {
      "epoch": 0.28239015023155994,
      "grad_norm": 12.625,
      "learning_rate": 1.6841447800211917e-05,
      "loss": 0.5527,
      "num_input_tokens_seen": 1577033984,
      "step": 21250,
      "train_runtime": 25211.1369,
      "train_tokens_per_second": 62553.069
    },
    {
      "epoch": 0.2825230397140218,
      "grad_norm": 27.0,
      "learning_rate": 1.683830808653145e-05,
      "loss": 0.8298,
      "num_input_tokens_seen": 1579149824,
      "step": 21260,
      "train_runtime": 25244.8495,
      "train_tokens_per_second": 62553.347
    },
    {
      "epoch": 0.28265592919648375,
      "grad_norm": 17.375,
      "learning_rate": 1.6835167106115387e-05,
      "loss": 0.9426,
      "num_input_tokens_seen": 1579987968,
      "step": 21270,
      "train_runtime": 25257.6667,
      "train_tokens_per_second": 62554.787
    },
    {
      "epoch": 0.2827888186789457,
      "grad_norm": 13.5625,
      "learning_rate": 1.6832024859545565e-05,
      "loss": 0.673,
      "num_input_tokens_seen": 1580471552,
      "step": 21280,
      "train_runtime": 25264.4543,
      "train_tokens_per_second": 62557.122
    },
    {
      "epoch": 0.28292170816140755,
      "grad_norm": 10.75,
      "learning_rate": 1.6828881347404057e-05,
      "loss": 0.6396,
      "num_input_tokens_seen": 1580705792,
      "step": 21290,
      "train_runtime": 25268.4914,
      "train_tokens_per_second": 62556.398
    },
    {
      "epoch": 0.2830545976438695,
      "grad_norm": 12.8125,
      "learning_rate": 1.682573657027317e-05,
      "loss": 0.5136,
      "num_input_tokens_seen": 1580821760,
      "step": 21300,
      "train_runtime": 25270.8366,
      "train_tokens_per_second": 62555.181
    },
    {
      "epoch": 0.2831874871263314,
      "grad_norm": 20.5,
      "learning_rate": 1.682259052873545e-05,
      "loss": 0.8416,
      "num_input_tokens_seen": 1582969088,
      "step": 21310,
      "train_runtime": 25304.7157,
      "train_tokens_per_second": 62556.288
    },
    {
      "epoch": 0.2833203766087933,
      "grad_norm": 21.75,
      "learning_rate": 1.6819443223373672e-05,
      "loss": 0.9007,
      "num_input_tokens_seen": 1583813376,
      "step": 21320,
      "train_runtime": 25317.3942,
      "train_tokens_per_second": 62558.309
    },
    {
      "epoch": 0.2834532660912552,
      "grad_norm": 19.125,
      "learning_rate": 1.681629465477084e-05,
      "loss": 0.7616,
      "num_input_tokens_seen": 1584291328,
      "step": 21330,
      "train_runtime": 25324.2675,
      "train_tokens_per_second": 62560.204
    },
    {
      "epoch": 0.28358615557371714,
      "grad_norm": 42.5,
      "learning_rate": 1.6813144823510206e-05,
      "loss": 0.6094,
      "num_input_tokens_seen": 1584532224,
      "step": 21340,
      "train_runtime": 25328.3305,
      "train_tokens_per_second": 62559.679
    },
    {
      "epoch": 0.283719045056179,
      "grad_norm": 15.6875,
      "learning_rate": 1.6809993730175246e-05,
      "loss": 0.5789,
      "num_input_tokens_seen": 1584650240,
      "step": 21350,
      "train_runtime": 25330.7583,
      "train_tokens_per_second": 62558.342
    },
    {
      "epoch": 0.28385193453864094,
      "grad_norm": 26.0,
      "learning_rate": 1.680684137534967e-05,
      "loss": 0.8423,
      "num_input_tokens_seen": 1586615808,
      "step": 21360,
      "train_runtime": 25362.8879,
      "train_tokens_per_second": 62556.591
    },
    {
      "epoch": 0.28398482402110287,
      "grad_norm": 44.25,
      "learning_rate": 1.6803687759617427e-05,
      "loss": 0.8974,
      "num_input_tokens_seen": 1587486464,
      "step": 21370,
      "train_runtime": 25375.7993,
      "train_tokens_per_second": 62559.072
    },
    {
      "epoch": 0.28411771350356474,
      "grad_norm": 23.0,
      "learning_rate": 1.680053288356269e-05,
      "loss": 0.7209,
      "num_input_tokens_seen": 1587956224,
      "step": 21380,
      "train_runtime": 25383.0364,
      "train_tokens_per_second": 62559.743
    },
    {
      "epoch": 0.28425060298602667,
      "grad_norm": 13.5625,
      "learning_rate": 1.6797376747769875e-05,
      "loss": 0.6587,
      "num_input_tokens_seen": 1588186880,
      "step": 21390,
      "train_runtime": 25386.5122,
      "train_tokens_per_second": 62560.263
    },
    {
      "epoch": 0.2843834924684886,
      "grad_norm": 11.0625,
      "learning_rate": 1.6794219352823634e-05,
      "loss": 0.5635,
      "num_input_tokens_seen": 1588300800,
      "step": 21400,
      "train_runtime": 25388.8777,
      "train_tokens_per_second": 62558.921
    },
    {
      "epoch": 0.28451638195095047,
      "grad_norm": 21.25,
      "learning_rate": 1.6791060699308836e-05,
      "loss": 0.7891,
      "num_input_tokens_seen": 1590377728,
      "step": 21410,
      "train_runtime": 25421.622,
      "train_tokens_per_second": 62560.041
    },
    {
      "epoch": 0.2846492714334124,
      "grad_norm": 26.25,
      "learning_rate": 1.6787900787810598e-05,
      "loss": 0.9706,
      "num_input_tokens_seen": 1591191040,
      "step": 21420,
      "train_runtime": 25433.9832,
      "train_tokens_per_second": 62561.614
    },
    {
      "epoch": 0.2847821609158743,
      "grad_norm": 48.5,
      "learning_rate": 1.6784739618914268e-05,
      "loss": 0.6998,
      "num_input_tokens_seen": 1591621632,
      "step": 21430,
      "train_runtime": 25440.3927,
      "train_tokens_per_second": 62562.778
    },
    {
      "epoch": 0.2849150503983362,
      "grad_norm": 15.625,
      "learning_rate": 1.678157719320542e-05,
      "loss": 0.6357,
      "num_input_tokens_seen": 1591844096,
      "step": 21440,
      "train_runtime": 25443.75,
      "train_tokens_per_second": 62563.266
    },
    {
      "epoch": 0.28504793988079813,
      "grad_norm": 14.1875,
      "learning_rate": 1.677841351126987e-05,
      "loss": 0.5177,
      "num_input_tokens_seen": 1591959808,
      "step": 21450,
      "train_runtime": 25446.0897,
      "train_tokens_per_second": 62562.061
    },
    {
      "epoch": 0.28518082936326006,
      "grad_norm": 48.0,
      "learning_rate": 1.677524857369366e-05,
      "loss": 0.7722,
      "num_input_tokens_seen": 1593928960,
      "step": 21460,
      "train_runtime": 25478.8878,
      "train_tokens_per_second": 62558.812
    },
    {
      "epoch": 0.28531371884572193,
      "grad_norm": 23.875,
      "learning_rate": 1.677208238106306e-05,
      "loss": 0.9432,
      "num_input_tokens_seen": 1594782208,
      "step": 21470,
      "train_runtime": 25491.6119,
      "train_tokens_per_second": 62561.058
    },
    {
      "epoch": 0.28544660832818386,
      "grad_norm": 24.375,
      "learning_rate": 1.676891493396459e-05,
      "loss": 0.7757,
      "num_input_tokens_seen": 1595238912,
      "step": 21480,
      "train_runtime": 25498.269,
      "train_tokens_per_second": 62562.636
    },
    {
      "epoch": 0.2855794978106458,
      "grad_norm": 13.0,
      "learning_rate": 1.676574623298498e-05,
      "loss": 0.6189,
      "num_input_tokens_seen": 1595464704,
      "step": 21490,
      "train_runtime": 25501.7111,
      "train_tokens_per_second": 62563.045
    },
    {
      "epoch": 0.28571238729310766,
      "grad_norm": 16.875,
      "learning_rate": 1.6762576278711213e-05,
      "loss": 0.6143,
      "num_input_tokens_seen": 1595581184,
      "step": 21500,
      "train_runtime": 25504.1197,
      "train_tokens_per_second": 62561.704
    },
    {
      "epoch": 0.2858452767755696,
      "grad_norm": 35.75,
      "learning_rate": 1.675940507173049e-05,
      "loss": 0.8894,
      "num_input_tokens_seen": 1597649664,
      "step": 21510,
      "train_runtime": 25537.8754,
      "train_tokens_per_second": 62560.007
    },
    {
      "epoch": 0.2859781662580315,
      "grad_norm": 17.5,
      "learning_rate": 1.6756232612630252e-05,
      "loss": 0.9277,
      "num_input_tokens_seen": 1598471424,
      "step": 21520,
      "train_runtime": 25550.3391,
      "train_tokens_per_second": 62561.652
    },
    {
      "epoch": 0.2861110557404934,
      "grad_norm": 15.625,
      "learning_rate": 1.6753058901998164e-05,
      "loss": 0.8026,
      "num_input_tokens_seen": 1598917376,
      "step": 21530,
      "train_runtime": 25556.9228,
      "train_tokens_per_second": 62562.985
    },
    {
      "epoch": 0.2862439452229553,
      "grad_norm": 29.5,
      "learning_rate": 1.674988394042213e-05,
      "loss": 0.6877,
      "num_input_tokens_seen": 1599151360,
      "step": 21540,
      "train_runtime": 25560.3912,
      "train_tokens_per_second": 62563.65
    },
    {
      "epoch": 0.28637683470541725,
      "grad_norm": 8.9375,
      "learning_rate": 1.674670772849029e-05,
      "loss": 0.5166,
      "num_input_tokens_seen": 1599268864,
      "step": 21550,
      "train_runtime": 25563.3252,
      "train_tokens_per_second": 62561.066
    },
    {
      "epoch": 0.2865097241878792,
      "grad_norm": 12.625,
      "learning_rate": 1.6743530266790996e-05,
      "loss": 0.8408,
      "num_input_tokens_seen": 1601315840,
      "step": 21560,
      "train_runtime": 25596.1019,
      "train_tokens_per_second": 62560.926
    },
    {
      "epoch": 0.28664261367034105,
      "grad_norm": 33.75,
      "learning_rate": 1.6740351555912852e-05,
      "loss": 0.8962,
      "num_input_tokens_seen": 1602215168,
      "step": 21570,
      "train_runtime": 25609.1691,
      "train_tokens_per_second": 62564.122
    },
    {
      "epoch": 0.286775503152803,
      "grad_norm": 13.125,
      "learning_rate": 1.6737171596444688e-05,
      "loss": 0.7984,
      "num_input_tokens_seen": 1602712576,
      "step": 21580,
      "train_runtime": 25616.2766,
      "train_tokens_per_second": 62566.18
    },
    {
      "epoch": 0.2869083926352649,
      "grad_norm": 12.125,
      "learning_rate": 1.6733990388975556e-05,
      "loss": 0.6377,
      "num_input_tokens_seen": 1602958336,
      "step": 21590,
      "train_runtime": 25619.8763,
      "train_tokens_per_second": 62566.982
    },
    {
      "epoch": 0.2870412821177268,
      "grad_norm": 7.71875,
      "learning_rate": 1.6730807934094753e-05,
      "loss": 0.5176,
      "num_input_tokens_seen": 1603073536,
      "step": 21600,
      "train_runtime": 25622.3032,
      "train_tokens_per_second": 62565.552
    },
    {
      "epoch": 0.2871741716001887,
      "grad_norm": 31.125,
      "learning_rate": 1.6727624232391798e-05,
      "loss": 0.8632,
      "num_input_tokens_seen": 1605033728,
      "step": 21610,
      "train_runtime": 25654.3162,
      "train_tokens_per_second": 62563.887
    },
    {
      "epoch": 0.28730706108265064,
      "grad_norm": 26.125,
      "learning_rate": 1.672443928445644e-05,
      "loss": 0.8973,
      "num_input_tokens_seen": 1605883392,
      "step": 21620,
      "train_runtime": 25667.1934,
      "train_tokens_per_second": 62565.602
    },
    {
      "epoch": 0.2874399505651125,
      "grad_norm": 12.3125,
      "learning_rate": 1.672125309087867e-05,
      "loss": 0.7608,
      "num_input_tokens_seen": 1606332928,
      "step": 21630,
      "train_runtime": 25673.8567,
      "train_tokens_per_second": 62566.873
    },
    {
      "epoch": 0.28757284004757444,
      "grad_norm": 21.125,
      "learning_rate": 1.6718065652248694e-05,
      "loss": 0.6188,
      "num_input_tokens_seen": 1606560768,
      "step": 21640,
      "train_runtime": 25677.3363,
      "train_tokens_per_second": 62567.268
    },
    {
      "epoch": 0.28770572953003637,
      "grad_norm": 8.4375,
      "learning_rate": 1.6714876969156957e-05,
      "loss": 0.4877,
      "num_input_tokens_seen": 1606678016,
      "step": 21650,
      "train_runtime": 25679.7476,
      "train_tokens_per_second": 62565.958
    },
    {
      "epoch": 0.28783861901249824,
      "grad_norm": 34.0,
      "learning_rate": 1.6711687042194142e-05,
      "loss": 0.807,
      "num_input_tokens_seen": 1608846336,
      "step": 21660,
      "train_runtime": 25713.6233,
      "train_tokens_per_second": 62567.858
    },
    {
      "epoch": 0.28797150849496017,
      "grad_norm": 22.25,
      "learning_rate": 1.6708495871951146e-05,
      "loss": 0.864,
      "num_input_tokens_seen": 1609701120,
      "step": 21670,
      "train_runtime": 25726.209,
      "train_tokens_per_second": 62570.475
    },
    {
      "epoch": 0.2881043979774221,
      "grad_norm": 26.375,
      "learning_rate": 1.670530345901911e-05,
      "loss": 0.7586,
      "num_input_tokens_seen": 1610165248,
      "step": 21680,
      "train_runtime": 25732.8745,
      "train_tokens_per_second": 62572.304
    },
    {
      "epoch": 0.288237287459884,
      "grad_norm": 25.75,
      "learning_rate": 1.6702109803989398e-05,
      "loss": 0.5968,
      "num_input_tokens_seen": 1610397440,
      "step": 21690,
      "train_runtime": 25736.3691,
      "train_tokens_per_second": 62572.83
    },
    {
      "epoch": 0.2883701769423459,
      "grad_norm": 19.125,
      "learning_rate": 1.669891490745361e-05,
      "loss": 0.5637,
      "num_input_tokens_seen": 1610514944,
      "step": 21700,
      "train_runtime": 25738.8241,
      "train_tokens_per_second": 62571.427
    },
    {
      "epoch": 0.28850306642480783,
      "grad_norm": 48.5,
      "learning_rate": 1.6695718770003564e-05,
      "loss": 0.8252,
      "num_input_tokens_seen": 1612541952,
      "step": 21710,
      "train_runtime": 25771.5542,
      "train_tokens_per_second": 62570.613
    },
    {
      "epoch": 0.2886359559072697,
      "grad_norm": 23.625,
      "learning_rate": 1.669252139223132e-05,
      "loss": 0.8909,
      "num_input_tokens_seen": 1613367808,
      "step": 21720,
      "train_runtime": 25784.222,
      "train_tokens_per_second": 62571.902
    },
    {
      "epoch": 0.28876884538973163,
      "grad_norm": 26.75,
      "learning_rate": 1.6689322774729168e-05,
      "loss": 0.7572,
      "num_input_tokens_seen": 1613819648,
      "step": 21730,
      "train_runtime": 25790.9135,
      "train_tokens_per_second": 62573.187
    },
    {
      "epoch": 0.28890173487219356,
      "grad_norm": 31.375,
      "learning_rate": 1.6686122918089616e-05,
      "loss": 0.6558,
      "num_input_tokens_seen": 1614041088,
      "step": 21740,
      "train_runtime": 25794.8389,
      "train_tokens_per_second": 62572.249
    },
    {
      "epoch": 0.28903462435465543,
      "grad_norm": 29.625,
      "learning_rate": 1.6682921822905414e-05,
      "loss": 0.5548,
      "num_input_tokens_seen": 1614155008,
      "step": 21750,
      "train_runtime": 25797.3115,
      "train_tokens_per_second": 62570.668
    },
    {
      "epoch": 0.28916751383711736,
      "grad_norm": 55.75,
      "learning_rate": 1.6679719489769535e-05,
      "loss": 0.81,
      "num_input_tokens_seen": 1616209664,
      "step": 21760,
      "train_runtime": 25829.9294,
      "train_tokens_per_second": 62571.199
    },
    {
      "epoch": 0.2893004033195793,
      "grad_norm": 30.625,
      "learning_rate": 1.667651591927518e-05,
      "loss": 0.9191,
      "num_input_tokens_seen": 1617039872,
      "step": 21770,
      "train_runtime": 25842.5045,
      "train_tokens_per_second": 62572.878
    },
    {
      "epoch": 0.28943329280204116,
      "grad_norm": 14.25,
      "learning_rate": 1.667331111201579e-05,
      "loss": 0.7971,
      "num_input_tokens_seen": 1617496064,
      "step": 21780,
      "train_runtime": 25849.0763,
      "train_tokens_per_second": 62574.618
    },
    {
      "epoch": 0.2895661822845031,
      "grad_norm": 12.125,
      "learning_rate": 1.6670105068585015e-05,
      "loss": 0.6759,
      "num_input_tokens_seen": 1617722112,
      "step": 21790,
      "train_runtime": 25853.0235,
      "train_tokens_per_second": 62573.807
    },
    {
      "epoch": 0.289699071766965,
      "grad_norm": 19.0,
      "learning_rate": 1.6666897789576757e-05,
      "loss": 0.6018,
      "num_input_tokens_seen": 1617838336,
      "step": 21800,
      "train_runtime": 25855.378,
      "train_tokens_per_second": 62572.604
    },
    {
      "epoch": 0.2898319612494269,
      "grad_norm": 87.5,
      "learning_rate": 1.6663689275585128e-05,
      "loss": 0.8859,
      "num_input_tokens_seen": 1619842560,
      "step": 21810,
      "train_runtime": 25887.4916,
      "train_tokens_per_second": 62572.403
    },
    {
      "epoch": 0.2899648507318888,
      "grad_norm": 14.0,
      "learning_rate": 1.6660479527204484e-05,
      "loss": 0.9213,
      "num_input_tokens_seen": 1620659712,
      "step": 21820,
      "train_runtime": 25899.8007,
      "train_tokens_per_second": 62574.215
    },
    {
      "epoch": 0.29009774021435075,
      "grad_norm": 19.0,
      "learning_rate": 1.665726854502939e-05,
      "loss": 0.7261,
      "num_input_tokens_seen": 1621131008,
      "step": 21830,
      "train_runtime": 25906.6864,
      "train_tokens_per_second": 62575.776
    },
    {
      "epoch": 0.2902306296968126,
      "grad_norm": 22.125,
      "learning_rate": 1.6654056329654666e-05,
      "loss": 0.6362,
      "num_input_tokens_seen": 1621357824,
      "step": 21840,
      "train_runtime": 25910.1139,
      "train_tokens_per_second": 62576.252
    },
    {
      "epoch": 0.29036351917927455,
      "grad_norm": 9.3125,
      "learning_rate": 1.6650842881675343e-05,
      "loss": 0.5383,
      "num_input_tokens_seen": 1621470720,
      "step": 21850,
      "train_runtime": 25912.5536,
      "train_tokens_per_second": 62574.718
    },
    {
      "epoch": 0.2904964086617365,
      "grad_norm": 104.5,
      "learning_rate": 1.6647628201686676e-05,
      "loss": 0.8527,
      "num_input_tokens_seen": 1623604480,
      "step": 21860,
      "train_runtime": 25947.2017,
      "train_tokens_per_second": 62573.394
    },
    {
      "epoch": 0.29062929814419836,
      "grad_norm": 11.9375,
      "learning_rate": 1.6644412290284156e-05,
      "loss": 0.9181,
      "num_input_tokens_seen": 1624467712,
      "step": 21870,
      "train_runtime": 25960.0574,
      "train_tokens_per_second": 62575.659
    },
    {
      "epoch": 0.2907621876266603,
      "grad_norm": 21.75,
      "learning_rate": 1.6641195148063513e-05,
      "loss": 0.8256,
      "num_input_tokens_seen": 1624959488,
      "step": 21880,
      "train_runtime": 25967.0379,
      "train_tokens_per_second": 62577.776
    },
    {
      "epoch": 0.2908950771091222,
      "grad_norm": 18.25,
      "learning_rate": 1.6637976775620687e-05,
      "loss": 0.7013,
      "num_input_tokens_seen": 1625195008,
      "step": 21890,
      "train_runtime": 25970.5744,
      "train_tokens_per_second": 62578.324
    },
    {
      "epoch": 0.2910279665915841,
      "grad_norm": 12.5625,
      "learning_rate": 1.663475717355185e-05,
      "loss": 0.5675,
      "num_input_tokens_seen": 1625309184,
      "step": 21900,
      "train_runtime": 25973.4526,
      "train_tokens_per_second": 62575.785
    },
    {
      "epoch": 0.291160856074046,
      "grad_norm": 24.375,
      "learning_rate": 1.6631536342453408e-05,
      "loss": 0.8476,
      "num_input_tokens_seen": 1627554816,
      "step": 21910,
      "train_runtime": 26008.0176,
      "train_tokens_per_second": 62578.965
    },
    {
      "epoch": 0.29129374555650794,
      "grad_norm": 18.75,
      "learning_rate": 1.6628314282921994e-05,
      "loss": 0.9116,
      "num_input_tokens_seen": 1628387328,
      "step": 21920,
      "train_runtime": 26020.7599,
      "train_tokens_per_second": 62580.314
    },
    {
      "epoch": 0.2914266350389698,
      "grad_norm": 16.125,
      "learning_rate": 1.6625090995554463e-05,
      "loss": 0.7827,
      "num_input_tokens_seen": 1628845056,
      "step": 21930,
      "train_runtime": 26027.2896,
      "train_tokens_per_second": 62582.2
    },
    {
      "epoch": 0.29155952452143175,
      "grad_norm": 28.0,
      "learning_rate": 1.6621866480947897e-05,
      "loss": 0.5933,
      "num_input_tokens_seen": 1629066752,
      "step": 21940,
      "train_runtime": 26030.6937,
      "train_tokens_per_second": 62582.533
    },
    {
      "epoch": 0.2916924140038937,
      "grad_norm": 10.6875,
      "learning_rate": 1.661864073969961e-05,
      "loss": 0.5893,
      "num_input_tokens_seen": 1629179648,
      "step": 21950,
      "train_runtime": 26033.6054,
      "train_tokens_per_second": 62579.87
    },
    {
      "epoch": 0.29182530348635555,
      "grad_norm": 53.0,
      "learning_rate": 1.6615413772407145e-05,
      "loss": 0.7935,
      "num_input_tokens_seen": 1631222528,
      "step": 21960,
      "train_runtime": 26066.5498,
      "train_tokens_per_second": 62579.15
    },
    {
      "epoch": 0.2919581929688175,
      "grad_norm": 18.875,
      "learning_rate": 1.661218557966827e-05,
      "loss": 0.9418,
      "num_input_tokens_seen": 1632032512,
      "step": 21970,
      "train_runtime": 26078.7208,
      "train_tokens_per_second": 62581.003
    },
    {
      "epoch": 0.2920910824512794,
      "grad_norm": 19.625,
      "learning_rate": 1.660895616208097e-05,
      "loss": 0.7661,
      "num_input_tokens_seen": 1632481280,
      "step": 21980,
      "train_runtime": 26085.301,
      "train_tokens_per_second": 62582.421
    },
    {
      "epoch": 0.2922239719337413,
      "grad_norm": 12.0,
      "learning_rate": 1.660572552024348e-05,
      "loss": 0.6287,
      "num_input_tokens_seen": 1632712704,
      "step": 21990,
      "train_runtime": 26088.7801,
      "train_tokens_per_second": 62582.946
    },
    {
      "epoch": 0.2923568614162032,
      "grad_norm": 31.125,
      "learning_rate": 1.6602493654754232e-05,
      "loss": 0.5034,
      "num_input_tokens_seen": 1632829440,
      "step": 22000,
      "train_runtime": 26091.2269,
      "train_tokens_per_second": 62581.551
    },
    {
      "epoch": 0.29248975089866514,
      "grad_norm": 22.75,
      "learning_rate": 1.6599260566211916e-05,
      "loss": 0.8894,
      "num_input_tokens_seen": 1634921216,
      "step": 22010,
      "train_runtime": 26124.9335,
      "train_tokens_per_second": 62580.876
    },
    {
      "epoch": 0.292622640381127,
      "grad_norm": 28.25,
      "learning_rate": 1.6596026255215417e-05,
      "loss": 0.8585,
      "num_input_tokens_seen": 1635780864,
      "step": 22020,
      "train_runtime": 26137.7397,
      "train_tokens_per_second": 62583.103
    },
    {
      "epoch": 0.29275552986358894,
      "grad_norm": 376.0,
      "learning_rate": 1.6592790722363874e-05,
      "loss": 0.77,
      "num_input_tokens_seen": 1636267776,
      "step": 22030,
      "train_runtime": 26144.793,
      "train_tokens_per_second": 62584.843
    },
    {
      "epoch": 0.29288841934605087,
      "grad_norm": 13.8125,
      "learning_rate": 1.6589553968256633e-05,
      "loss": 0.7373,
      "num_input_tokens_seen": 1636510720,
      "step": 22040,
      "train_runtime": 26148.3756,
      "train_tokens_per_second": 62585.56
    },
    {
      "epoch": 0.2930213088285128,
      "grad_norm": 11.375,
      "learning_rate": 1.6586315993493277e-05,
      "loss": 0.509,
      "num_input_tokens_seen": 1636626432,
      "step": 22050,
      "train_runtime": 26150.8244,
      "train_tokens_per_second": 62584.124
    },
    {
      "epoch": 0.29315419831097467,
      "grad_norm": 13.5625,
      "learning_rate": 1.658307679867362e-05,
      "loss": 0.8517,
      "num_input_tokens_seen": 1638657792,
      "step": 22060,
      "train_runtime": 26183.7655,
      "train_tokens_per_second": 62582.969
    },
    {
      "epoch": 0.2932870877934366,
      "grad_norm": 17.875,
      "learning_rate": 1.657983638439768e-05,
      "loss": 0.9266,
      "num_input_tokens_seen": 1639488768,
      "step": 22070,
      "train_runtime": 26196.4908,
      "train_tokens_per_second": 62584.29
    },
    {
      "epoch": 0.2934199772758985,
      "grad_norm": 19.0,
      "learning_rate": 1.6576594751265722e-05,
      "loss": 0.7011,
      "num_input_tokens_seen": 1639954176,
      "step": 22080,
      "train_runtime": 26203.2392,
      "train_tokens_per_second": 62585.933
    },
    {
      "epoch": 0.2935528667583604,
      "grad_norm": 9.3125,
      "learning_rate": 1.6573351899878227e-05,
      "loss": 0.6407,
      "num_input_tokens_seen": 1640187648,
      "step": 22090,
      "train_runtime": 26206.7379,
      "train_tokens_per_second": 62586.486
    },
    {
      "epoch": 0.2936857562408223,
      "grad_norm": 11.8125,
      "learning_rate": 1.657010783083591e-05,
      "loss": 0.568,
      "num_input_tokens_seen": 1640305408,
      "step": 22100,
      "train_runtime": 26209.1998,
      "train_tokens_per_second": 62585.101
    },
    {
      "epoch": 0.29381864572328426,
      "grad_norm": 14.625,
      "learning_rate": 1.65668625447397e-05,
      "loss": 0.8168,
      "num_input_tokens_seen": 1642396672,
      "step": 22110,
      "train_runtime": 26244.2946,
      "train_tokens_per_second": 62581.094
    },
    {
      "epoch": 0.29395153520574613,
      "grad_norm": 12.6875,
      "learning_rate": 1.6563616042190757e-05,
      "loss": 0.8944,
      "num_input_tokens_seen": 1643235072,
      "step": 22120,
      "train_runtime": 26256.9879,
      "train_tokens_per_second": 62582.771
    },
    {
      "epoch": 0.29408442468820806,
      "grad_norm": 12.3125,
      "learning_rate": 1.656036832379047e-05,
      "loss": 0.7008,
      "num_input_tokens_seen": 1643726848,
      "step": 22130,
      "train_runtime": 26263.6263,
      "train_tokens_per_second": 62585.678
    },
    {
      "epoch": 0.29421731417067,
      "grad_norm": 15.5,
      "learning_rate": 1.655711939014045e-05,
      "loss": 0.7148,
      "num_input_tokens_seen": 1643949824,
      "step": 22140,
      "train_runtime": 26267.1087,
      "train_tokens_per_second": 62585.869
    },
    {
      "epoch": 0.29435020365313186,
      "grad_norm": 12.9375,
      "learning_rate": 1.6553869241842532e-05,
      "loss": 0.5119,
      "num_input_tokens_seen": 1644066816,
      "step": 22150,
      "train_runtime": 26269.5475,
      "train_tokens_per_second": 62584.512
    },
    {
      "epoch": 0.2944830931355938,
      "grad_norm": 15.25,
      "learning_rate": 1.655061787949878e-05,
      "loss": 0.7943,
      "num_input_tokens_seen": 1646015744,
      "step": 22160,
      "train_runtime": 26301.6158,
      "train_tokens_per_second": 62582.305
    },
    {
      "epoch": 0.2946159826180557,
      "grad_norm": 16.5,
      "learning_rate": 1.654736530371147e-05,
      "loss": 0.9532,
      "num_input_tokens_seen": 1646882304,
      "step": 22170,
      "train_runtime": 26314.4464,
      "train_tokens_per_second": 62584.722
    },
    {
      "epoch": 0.2947488721005176,
      "grad_norm": 64.5,
      "learning_rate": 1.6544111515083124e-05,
      "loss": 0.7511,
      "num_input_tokens_seen": 1647364352,
      "step": 22180,
      "train_runtime": 26321.406,
      "train_tokens_per_second": 62586.488
    },
    {
      "epoch": 0.2948817615829795,
      "grad_norm": 13.75,
      "learning_rate": 1.654085651421647e-05,
      "loss": 0.6657,
      "num_input_tokens_seen": 1647596544,
      "step": 22190,
      "train_runtime": 26325.4633,
      "train_tokens_per_second": 62585.662
    },
    {
      "epoch": 0.29501465106544145,
      "grad_norm": 11.25,
      "learning_rate": 1.653760030171447e-05,
      "loss": 0.522,
      "num_input_tokens_seen": 1647716352,
      "step": 22200,
      "train_runtime": 26327.8856,
      "train_tokens_per_second": 62584.454
    },
    {
      "epoch": 0.2951475405479033,
      "grad_norm": 19.875,
      "learning_rate": 1.653434287818031e-05,
      "loss": 0.8394,
      "num_input_tokens_seen": 1649805568,
      "step": 22210,
      "train_runtime": 26361.5512,
      "train_tokens_per_second": 62583.782
    },
    {
      "epoch": 0.29528043003036525,
      "grad_norm": 30.125,
      "learning_rate": 1.6531084244217398e-05,
      "loss": 0.9607,
      "num_input_tokens_seen": 1650659584,
      "step": 22220,
      "train_runtime": 26374.3926,
      "train_tokens_per_second": 62585.691
    },
    {
      "epoch": 0.2954133195128272,
      "grad_norm": 21.0,
      "learning_rate": 1.6527824400429364e-05,
      "loss": 0.7616,
      "num_input_tokens_seen": 1651114752,
      "step": 22230,
      "train_runtime": 26381.088,
      "train_tokens_per_second": 62587.061
    },
    {
      "epoch": 0.29554620899528905,
      "grad_norm": 25.875,
      "learning_rate": 1.652456334742007e-05,
      "loss": 0.5923,
      "num_input_tokens_seen": 1651351296,
      "step": 22240,
      "train_runtime": 26385.1486,
      "train_tokens_per_second": 62586.394
    },
    {
      "epoch": 0.295679098477751,
      "grad_norm": 9.375,
      "learning_rate": 1.6521301085793595e-05,
      "loss": 0.4605,
      "num_input_tokens_seen": 1651467008,
      "step": 22250,
      "train_runtime": 26387.6211,
      "train_tokens_per_second": 62584.914
    },
    {
      "epoch": 0.2958119879602129,
      "grad_norm": 42.5,
      "learning_rate": 1.651803761615424e-05,
      "loss": 0.8869,
      "num_input_tokens_seen": 1653420032,
      "step": 22260,
      "train_runtime": 26418.7222,
      "train_tokens_per_second": 62585.163
    },
    {
      "epoch": 0.2959448774426748,
      "grad_norm": 25.125,
      "learning_rate": 1.651477293910654e-05,
      "loss": 0.9013,
      "num_input_tokens_seen": 1654268416,
      "step": 22270,
      "train_runtime": 26431.4745,
      "train_tokens_per_second": 62587.065
    },
    {
      "epoch": 0.2960777669251367,
      "grad_norm": 13.25,
      "learning_rate": 1.6511507055255244e-05,
      "loss": 0.8181,
      "num_input_tokens_seen": 1654739200,
      "step": 22280,
      "train_runtime": 26438.1748,
      "train_tokens_per_second": 62589.01
    },
    {
      "epoch": 0.29621065640759864,
      "grad_norm": 976.0,
      "learning_rate": 1.6508239965205325e-05,
      "loss": 0.7234,
      "num_input_tokens_seen": 1654966272,
      "step": 22290,
      "train_runtime": 26442.1104,
      "train_tokens_per_second": 62588.282
    },
    {
      "epoch": 0.2963435458900605,
      "grad_norm": 12.875,
      "learning_rate": 1.6504971669561987e-05,
      "loss": 0.524,
      "num_input_tokens_seen": 1655084288,
      "step": 22300,
      "train_runtime": 26444.4704,
      "train_tokens_per_second": 62587.16
    },
    {
      "epoch": 0.29647643537252244,
      "grad_norm": 17.5,
      "learning_rate": 1.6501702168930652e-05,
      "loss": 0.8459,
      "num_input_tokens_seen": 1657137920,
      "step": 22310,
      "train_runtime": 26477.6362,
      "train_tokens_per_second": 62586.324
    },
    {
      "epoch": 0.29660932485498437,
      "grad_norm": 28.875,
      "learning_rate": 1.6498431463916963e-05,
      "loss": 0.922,
      "num_input_tokens_seen": 1657991168,
      "step": 22320,
      "train_runtime": 26490.3882,
      "train_tokens_per_second": 62588.406
    },
    {
      "epoch": 0.29674221433744624,
      "grad_norm": 29.25,
      "learning_rate": 1.649515955512679e-05,
      "loss": 0.8117,
      "num_input_tokens_seen": 1658434560,
      "step": 22330,
      "train_runtime": 26497.0424,
      "train_tokens_per_second": 62589.422
    },
    {
      "epoch": 0.29687510381990817,
      "grad_norm": 34.75,
      "learning_rate": 1.649188644316623e-05,
      "loss": 0.6305,
      "num_input_tokens_seen": 1658669056,
      "step": 22340,
      "train_runtime": 26500.5469,
      "train_tokens_per_second": 62589.993
    },
    {
      "epoch": 0.2970079933023701,
      "grad_norm": 16.375,
      "learning_rate": 1.648861212864159e-05,
      "loss": 0.6178,
      "num_input_tokens_seen": 1658784000,
      "step": 22350,
      "train_runtime": 26503.0412,
      "train_tokens_per_second": 62588.44
    },
    {
      "epoch": 0.297140882784832,
      "grad_norm": 18.25,
      "learning_rate": 1.6485336612159413e-05,
      "loss": 0.8905,
      "num_input_tokens_seen": 1660700416,
      "step": 22360,
      "train_runtime": 26534.2458,
      "train_tokens_per_second": 62587.059
    },
    {
      "epoch": 0.2972737722672939,
      "grad_norm": 22.875,
      "learning_rate": 1.6482059894326456e-05,
      "loss": 0.9986,
      "num_input_tokens_seen": 1661519104,
      "step": 22370,
      "train_runtime": 26546.6435,
      "train_tokens_per_second": 62588.67
    },
    {
      "epoch": 0.29740666174975583,
      "grad_norm": 23.0,
      "learning_rate": 1.6478781975749704e-05,
      "loss": 0.7926,
      "num_input_tokens_seen": 1661980672,
      "step": 22380,
      "train_runtime": 26553.3846,
      "train_tokens_per_second": 62590.163
    },
    {
      "epoch": 0.2975395512322177,
      "grad_norm": 11.9375,
      "learning_rate": 1.647550285703636e-05,
      "loss": 0.6531,
      "num_input_tokens_seen": 1662214656,
      "step": 22390,
      "train_runtime": 26556.9169,
      "train_tokens_per_second": 62590.649
    },
    {
      "epoch": 0.29767244071467963,
      "grad_norm": 26.5,
      "learning_rate": 1.647222253879385e-05,
      "loss": 0.5465,
      "num_input_tokens_seen": 1662331136,
      "step": 22400,
      "train_runtime": 26559.2852,
      "train_tokens_per_second": 62589.453
    },
    {
      "epoch": 0.29780533019714156,
      "grad_norm": 23.5,
      "learning_rate": 1.6468941021629833e-05,
      "loss": 0.7489,
      "num_input_tokens_seen": 1664308992,
      "step": 22410,
      "train_runtime": 26592.9629,
      "train_tokens_per_second": 62584.564
    },
    {
      "epoch": 0.29793821967960343,
      "grad_norm": 10.4375,
      "learning_rate": 1.646565830615217e-05,
      "loss": 0.9058,
      "num_input_tokens_seen": 1665162496,
      "step": 22420,
      "train_runtime": 26605.4881,
      "train_tokens_per_second": 62587.181
    },
    {
      "epoch": 0.29807110916206536,
      "grad_norm": 18.75,
      "learning_rate": 1.646237439296896e-05,
      "loss": 0.7417,
      "num_input_tokens_seen": 1665608448,
      "step": 22430,
      "train_runtime": 26612.2721,
      "train_tokens_per_second": 62587.984
    },
    {
      "epoch": 0.2982039986445273,
      "grad_norm": 13.625,
      "learning_rate": 1.6459089282688518e-05,
      "loss": 0.7056,
      "num_input_tokens_seen": 1665838080,
      "step": 22440,
      "train_runtime": 26615.8065,
      "train_tokens_per_second": 62588.3
    },
    {
      "epoch": 0.29833688812698916,
      "grad_norm": 7.21875,
      "learning_rate": 1.645580297591938e-05,
      "loss": 0.4822,
      "num_input_tokens_seen": 1665957376,
      "step": 22450,
      "train_runtime": 26618.244,
      "train_tokens_per_second": 62587.05
    },
    {
      "epoch": 0.2984697776094511,
      "grad_norm": 14.125,
      "learning_rate": 1.6452515473270304e-05,
      "loss": 0.7705,
      "num_input_tokens_seen": 1668021504,
      "step": 22460,
      "train_runtime": 26651.4421,
      "train_tokens_per_second": 62586.538
    },
    {
      "epoch": 0.298602667091913,
      "grad_norm": 19.0,
      "learning_rate": 1.6449226775350276e-05,
      "loss": 0.9277,
      "num_input_tokens_seen": 1668842752,
      "step": 22470,
      "train_runtime": 26663.8551,
      "train_tokens_per_second": 62588.202
    },
    {
      "epoch": 0.2987355565743749,
      "grad_norm": 32.0,
      "learning_rate": 1.6445936882768494e-05,
      "loss": 0.6825,
      "num_input_tokens_seen": 1669307904,
      "step": 22480,
      "train_runtime": 26670.3959,
      "train_tokens_per_second": 62590.293
    },
    {
      "epoch": 0.2988684460568368,
      "grad_norm": 6.6875,
      "learning_rate": 1.6442645796134386e-05,
      "loss": 0.5945,
      "num_input_tokens_seen": 1669537024,
      "step": 22490,
      "train_runtime": 26673.8285,
      "train_tokens_per_second": 62590.828
    },
    {
      "epoch": 0.29900133553929875,
      "grad_norm": 12.0625,
      "learning_rate": 1.6439353516057583e-05,
      "loss": 0.5808,
      "num_input_tokens_seen": 1669652736,
      "step": 22500,
      "train_runtime": 26676.2178,
      "train_tokens_per_second": 62589.56
    },
    {
      "epoch": 0.2991342250217607,
      "grad_norm": 31.125,
      "learning_rate": 1.6436060043147967e-05,
      "loss": 0.8136,
      "num_input_tokens_seen": 1671681792,
      "step": 22510,
      "train_runtime": 26708.9704,
      "train_tokens_per_second": 62588.777
    },
    {
      "epoch": 0.29926711450422255,
      "grad_norm": 12.6875,
      "learning_rate": 1.6432765378015613e-05,
      "loss": 0.8965,
      "num_input_tokens_seen": 1672528384,
      "step": 22520,
      "train_runtime": 26721.9316,
      "train_tokens_per_second": 62590.1
    },
    {
      "epoch": 0.2994000039866845,
      "grad_norm": 18.75,
      "learning_rate": 1.6429469521270836e-05,
      "loss": 0.748,
      "num_input_tokens_seen": 1672993280,
      "step": 22530,
      "train_runtime": 26728.6935,
      "train_tokens_per_second": 62591.659
    },
    {
      "epoch": 0.2995328934691464,
      "grad_norm": 15.4375,
      "learning_rate": 1.642617247352416e-05,
      "loss": 0.5843,
      "num_input_tokens_seen": 1673228288,
      "step": 22540,
      "train_runtime": 26732.2496,
      "train_tokens_per_second": 62592.124
    },
    {
      "epoch": 0.2996657829516083,
      "grad_norm": 22.25,
      "learning_rate": 1.6422874235386333e-05,
      "loss": 0.5593,
      "num_input_tokens_seen": 1673344768,
      "step": 22550,
      "train_runtime": 26734.6486,
      "train_tokens_per_second": 62590.865
    },
    {
      "epoch": 0.2997986724340702,
      "grad_norm": 22.0,
      "learning_rate": 1.6419574807468323e-05,
      "loss": 0.8014,
      "num_input_tokens_seen": 1675430400,
      "step": 22560,
      "train_runtime": 26768.2093,
      "train_tokens_per_second": 62590.306
    },
    {
      "epoch": 0.29993156191653214,
      "grad_norm": 14.875,
      "learning_rate": 1.641627419038133e-05,
      "loss": 0.9363,
      "num_input_tokens_seen": 1676301312,
      "step": 22570,
      "train_runtime": 26781.4487,
      "train_tokens_per_second": 62591.883
    },
    {
      "epoch": 0.300064451398994,
      "grad_norm": 18.125,
      "learning_rate": 1.641297238473675e-05,
      "loss": 0.7196,
      "num_input_tokens_seen": 1676780544,
      "step": 22580,
      "train_runtime": 26788.2467,
      "train_tokens_per_second": 62593.889
    },
    {
      "epoch": 0.30019734088145594,
      "grad_norm": 25.375,
      "learning_rate": 1.640966939114622e-05,
      "loss": 0.6834,
      "num_input_tokens_seen": 1677014272,
      "step": 22590,
      "train_runtime": 26791.7206,
      "train_tokens_per_second": 62594.497
    },
    {
      "epoch": 0.30033023036391787,
      "grad_norm": 63.0,
      "learning_rate": 1.640636521022159e-05,
      "loss": 0.6104,
      "num_input_tokens_seen": 1677131008,
      "step": 22600,
      "train_runtime": 26794.0846,
      "train_tokens_per_second": 62593.331
    },
    {
      "epoch": 0.30046311984637974,
      "grad_norm": 18.75,
      "learning_rate": 1.640305984257493e-05,
      "loss": 0.8784,
      "num_input_tokens_seen": 1679112960,
      "step": 22610,
      "train_runtime": 26826.8666,
      "train_tokens_per_second": 62590.722
    },
    {
      "epoch": 0.3005960093288417,
      "grad_norm": 35.5,
      "learning_rate": 1.6399753288818526e-05,
      "loss": 0.8867,
      "num_input_tokens_seen": 1679954688,
      "step": 22620,
      "train_runtime": 26839.5803,
      "train_tokens_per_second": 62592.435
    },
    {
      "epoch": 0.3007288988113036,
      "grad_norm": 16.5,
      "learning_rate": 1.6396445549564894e-05,
      "loss": 0.7746,
      "num_input_tokens_seen": 1680428800,
      "step": 22630,
      "train_runtime": 26846.216,
      "train_tokens_per_second": 62594.624
    },
    {
      "epoch": 0.3008617882937655,
      "grad_norm": 11.5625,
      "learning_rate": 1.639313662542676e-05,
      "loss": 0.6735,
      "num_input_tokens_seen": 1680647168,
      "step": 22640,
      "train_runtime": 26849.5811,
      "train_tokens_per_second": 62594.912
    },
    {
      "epoch": 0.3009946777762274,
      "grad_norm": 8.5625,
      "learning_rate": 1.6389826517017074e-05,
      "loss": 0.5278,
      "num_input_tokens_seen": 1680757760,
      "step": 22650,
      "train_runtime": 26851.9443,
      "train_tokens_per_second": 62593.522
    },
    {
      "epoch": 0.30112756725868933,
      "grad_norm": 17.375,
      "learning_rate": 1.6386515224949e-05,
      "loss": 0.8839,
      "num_input_tokens_seen": 1682760704,
      "step": 22660,
      "train_runtime": 26884.1232,
      "train_tokens_per_second": 62593.103
    },
    {
      "epoch": 0.3012604567411512,
      "grad_norm": 15.8125,
      "learning_rate": 1.6383202749835928e-05,
      "loss": 0.8776,
      "num_input_tokens_seen": 1683623424,
      "step": 22670,
      "train_runtime": 26896.4346,
      "train_tokens_per_second": 62596.528
    },
    {
      "epoch": 0.30139334622361313,
      "grad_norm": 18.25,
      "learning_rate": 1.6379889092291465e-05,
      "loss": 0.7805,
      "num_input_tokens_seen": 1684052736,
      "step": 22680,
      "train_runtime": 26902.7029,
      "train_tokens_per_second": 62597.901
    },
    {
      "epoch": 0.30152623570607506,
      "grad_norm": 15.5,
      "learning_rate": 1.6376574252929436e-05,
      "loss": 0.738,
      "num_input_tokens_seen": 1684272128,
      "step": 22690,
      "train_runtime": 26906.5952,
      "train_tokens_per_second": 62597.0
    },
    {
      "epoch": 0.30165912518853694,
      "grad_norm": 10.0625,
      "learning_rate": 1.6373258232363885e-05,
      "loss": 0.4521,
      "num_input_tokens_seen": 1684386560,
      "step": 22700,
      "train_runtime": 26908.9764,
      "train_tokens_per_second": 62595.713
    },
    {
      "epoch": 0.30179201467099886,
      "grad_norm": 16.0,
      "learning_rate": 1.636994103120908e-05,
      "loss": 0.82,
      "num_input_tokens_seen": 1686513408,
      "step": 22710,
      "train_runtime": 26943.4836,
      "train_tokens_per_second": 62594.482
    },
    {
      "epoch": 0.3019249041534608,
      "grad_norm": 17.375,
      "learning_rate": 1.6366622650079495e-05,
      "loss": 0.8797,
      "num_input_tokens_seen": 1687352832,
      "step": 22720,
      "train_runtime": 26956.1694,
      "train_tokens_per_second": 62596.165
    },
    {
      "epoch": 0.30205779363592267,
      "grad_norm": 28.375,
      "learning_rate": 1.6363303089589836e-05,
      "loss": 0.7885,
      "num_input_tokens_seen": 1687826944,
      "step": 22730,
      "train_runtime": 26962.9689,
      "train_tokens_per_second": 62597.964
    },
    {
      "epoch": 0.3021906831183846,
      "grad_norm": 27.125,
      "learning_rate": 1.6359982350355016e-05,
      "loss": 0.6555,
      "num_input_tokens_seen": 1688056064,
      "step": 22740,
      "train_runtime": 26966.9982,
      "train_tokens_per_second": 62597.107
    },
    {
      "epoch": 0.3023235726008465,
      "grad_norm": 11.9375,
      "learning_rate": 1.6356660432990177e-05,
      "loss": 0.5871,
      "num_input_tokens_seen": 1688171008,
      "step": 22750,
      "train_runtime": 26969.4251,
      "train_tokens_per_second": 62595.736
    },
    {
      "epoch": 0.3024564620833084,
      "grad_norm": 20.875,
      "learning_rate": 1.6353337338110676e-05,
      "loss": 0.8604,
      "num_input_tokens_seen": 1690212864,
      "step": 22760,
      "train_runtime": 27002.0745,
      "train_tokens_per_second": 62595.667
    },
    {
      "epoch": 0.3025893515657703,
      "grad_norm": 21.75,
      "learning_rate": 1.635001306633209e-05,
      "loss": 0.9504,
      "num_input_tokens_seen": 1691071488,
      "step": 22770,
      "train_runtime": 27014.897,
      "train_tokens_per_second": 62597.74
    },
    {
      "epoch": 0.30272224104823225,
      "grad_norm": 18.5,
      "learning_rate": 1.6346687618270197e-05,
      "loss": 0.6926,
      "num_input_tokens_seen": 1691559936,
      "step": 22780,
      "train_runtime": 27021.6677,
      "train_tokens_per_second": 62600.131
    },
    {
      "epoch": 0.3028551305306941,
      "grad_norm": 17.25,
      "learning_rate": 1.6343360994541022e-05,
      "loss": 0.6456,
      "num_input_tokens_seen": 1691802624,
      "step": 22790,
      "train_runtime": 27025.2199,
      "train_tokens_per_second": 62600.883
    },
    {
      "epoch": 0.30298802001315606,
      "grad_norm": 11.625,
      "learning_rate": 1.634003319576078e-05,
      "loss": 0.5737,
      "num_input_tokens_seen": 1691916288,
      "step": 22800,
      "train_runtime": 27027.6649,
      "train_tokens_per_second": 62599.425
    },
    {
      "epoch": 0.303120909495618,
      "grad_norm": 17.625,
      "learning_rate": 1.6336704222545926e-05,
      "loss": 0.8702,
      "num_input_tokens_seen": 1693966336,
      "step": 22810,
      "train_runtime": 27060.2838,
      "train_tokens_per_second": 62599.725
    },
    {
      "epoch": 0.30325379897807986,
      "grad_norm": 19.375,
      "learning_rate": 1.6333374075513117e-05,
      "loss": 0.892,
      "num_input_tokens_seen": 1694842112,
      "step": 22820,
      "train_runtime": 27073.3159,
      "train_tokens_per_second": 62601.94
    },
    {
      "epoch": 0.3033866884605418,
      "grad_norm": 14.4375,
      "learning_rate": 1.633004275527923e-05,
      "loss": 0.7488,
      "num_input_tokens_seen": 1695326976,
      "step": 22830,
      "train_runtime": 27080.1933,
      "train_tokens_per_second": 62603.947
    },
    {
      "epoch": 0.3035195779430037,
      "grad_norm": 16.75,
      "learning_rate": 1.632671026246137e-05,
      "loss": 0.6684,
      "num_input_tokens_seen": 1695559680,
      "step": 22840,
      "train_runtime": 27083.7326,
      "train_tokens_per_second": 62604.358
    },
    {
      "epoch": 0.3036524674254656,
      "grad_norm": 36.75,
      "learning_rate": 1.6323376597676847e-05,
      "loss": 0.529,
      "num_input_tokens_seen": 1695675904,
      "step": 22850,
      "train_runtime": 27086.1417,
      "train_tokens_per_second": 62603.08
    },
    {
      "epoch": 0.3037853569079275,
      "grad_norm": 35.5,
      "learning_rate": 1.6320041761543194e-05,
      "loss": 0.7802,
      "num_input_tokens_seen": 1697775104,
      "step": 22860,
      "train_runtime": 27118.4551,
      "train_tokens_per_second": 62605.893
    },
    {
      "epoch": 0.30391824639038945,
      "grad_norm": 14.875,
      "learning_rate": 1.631670575467816e-05,
      "loss": 0.8953,
      "num_input_tokens_seen": 1698621184,
      "step": 22870,
      "train_runtime": 27131.336,
      "train_tokens_per_second": 62607.355
    },
    {
      "epoch": 0.3040511358728513,
      "grad_norm": 19.25,
      "learning_rate": 1.6313368577699707e-05,
      "loss": 0.7939,
      "num_input_tokens_seen": 1699100416,
      "step": 22880,
      "train_runtime": 27137.9816,
      "train_tokens_per_second": 62609.683
    },
    {
      "epoch": 0.30418402535531325,
      "grad_norm": 15.75,
      "learning_rate": 1.631003023122603e-05,
      "loss": 0.6599,
      "num_input_tokens_seen": 1699331328,
      "step": 22890,
      "train_runtime": 27141.5077,
      "train_tokens_per_second": 62610.056
    },
    {
      "epoch": 0.3043169148377752,
      "grad_norm": 10.9375,
      "learning_rate": 1.630669071587551e-05,
      "loss": 0.4968,
      "num_input_tokens_seen": 1699447040,
      "step": 22900,
      "train_runtime": 27143.8864,
      "train_tokens_per_second": 62608.833
    },
    {
      "epoch": 0.30444980432023705,
      "grad_norm": 154.0,
      "learning_rate": 1.630335003226677e-05,
      "loss": 0.804,
      "num_input_tokens_seen": 1701472256,
      "step": 22910,
      "train_runtime": 27176.0902,
      "train_tokens_per_second": 62609.163
    },
    {
      "epoch": 0.304582693802699,
      "grad_norm": 15.25,
      "learning_rate": 1.6300008181018644e-05,
      "loss": 0.9093,
      "num_input_tokens_seen": 1702338304,
      "step": 22920,
      "train_runtime": 27188.6613,
      "train_tokens_per_second": 62612.068
    },
    {
      "epoch": 0.3047155832851609,
      "grad_norm": 21.875,
      "learning_rate": 1.6296665162750178e-05,
      "loss": 0.7588,
      "num_input_tokens_seen": 1702808320,
      "step": 22930,
      "train_runtime": 27195.339,
      "train_tokens_per_second": 62613.977
    },
    {
      "epoch": 0.3048484727676228,
      "grad_norm": 9.6875,
      "learning_rate": 1.6293320978080637e-05,
      "loss": 0.6335,
      "num_input_tokens_seen": 1703036160,
      "step": 22940,
      "train_runtime": 27199.3179,
      "train_tokens_per_second": 62613.194
    },
    {
      "epoch": 0.3049813622500847,
      "grad_norm": 10.0625,
      "learning_rate": 1.6289975627629503e-05,
      "loss": 0.5911,
      "num_input_tokens_seen": 1703151360,
      "step": 22950,
      "train_runtime": 27201.7492,
      "train_tokens_per_second": 62611.832
    },
    {
      "epoch": 0.30511425173254664,
      "grad_norm": 23.0,
      "learning_rate": 1.628662911201647e-05,
      "loss": 0.7623,
      "num_input_tokens_seen": 1705181440,
      "step": 22960,
      "train_runtime": 27235.3123,
      "train_tokens_per_second": 62609.212
    },
    {
      "epoch": 0.3052471412150085,
      "grad_norm": 24.25,
      "learning_rate": 1.628328143186145e-05,
      "loss": 0.9467,
      "num_input_tokens_seen": 1706042368,
      "step": 22970,
      "train_runtime": 27248.1036,
      "train_tokens_per_second": 62611.417
    },
    {
      "epoch": 0.30538003069747044,
      "grad_norm": 15.9375,
      "learning_rate": 1.6279932587784576e-05,
      "loss": 0.7647,
      "num_input_tokens_seen": 1706545408,
      "step": 22980,
      "train_runtime": 27255.098,
      "train_tokens_per_second": 62613.806
    },
    {
      "epoch": 0.30551292017993237,
      "grad_norm": 38.5,
      "learning_rate": 1.6276582580406183e-05,
      "loss": 0.6667,
      "num_input_tokens_seen": 1706784000,
      "step": 22990,
      "train_runtime": 27258.6458,
      "train_tokens_per_second": 62614.409
    },
    {
      "epoch": 0.3056458096623943,
      "grad_norm": 25.0,
      "learning_rate": 1.6273231410346833e-05,
      "loss": 0.5292,
      "num_input_tokens_seen": 1706904576,
      "step": 23000,
      "train_runtime": 27261.6021,
      "train_tokens_per_second": 62612.042
    },
    {
      "epoch": 0.30577869914485617,
      "grad_norm": 17.75,
      "learning_rate": 1.6269879078227305e-05,
      "loss": 0.9001,
      "num_input_tokens_seen": 1708815104,
      "step": 23010,
      "train_runtime": 27292.5286,
      "train_tokens_per_second": 62611.095
    },
    {
      "epoch": 0.3059115886273181,
      "grad_norm": 14.9375,
      "learning_rate": 1.626652558466858e-05,
      "loss": 0.9175,
      "num_input_tokens_seen": 1709616640,
      "step": 23020,
      "train_runtime": 27304.595,
      "train_tokens_per_second": 62612.781
    },
    {
      "epoch": 0.30604447810978,
      "grad_norm": 26.0,
      "learning_rate": 1.6263170930291873e-05,
      "loss": 0.7328,
      "num_input_tokens_seen": 1710046976,
      "step": 23030,
      "train_runtime": 27310.7877,
      "train_tokens_per_second": 62614.341
    },
    {
      "epoch": 0.3061773675922419,
      "grad_norm": 57.0,
      "learning_rate": 1.62598151157186e-05,
      "loss": 0.6636,
      "num_input_tokens_seen": 1710270208,
      "step": 23040,
      "train_runtime": 27314.1873,
      "train_tokens_per_second": 62614.721
    },
    {
      "epoch": 0.30631025707470383,
      "grad_norm": 14.9375,
      "learning_rate": 1.625645814157039e-05,
      "loss": 0.484,
      "num_input_tokens_seen": 1710383104,
      "step": 23050,
      "train_runtime": 27317.0266,
      "train_tokens_per_second": 62612.345
    },
    {
      "epoch": 0.30644314655716576,
      "grad_norm": 31.625,
      "learning_rate": 1.62531000084691e-05,
      "loss": 0.8714,
      "num_input_tokens_seen": 1712455424,
      "step": 23060,
      "train_runtime": 27349.8542,
      "train_tokens_per_second": 62612.964
    },
    {
      "epoch": 0.30657603603962763,
      "grad_norm": 23.25,
      "learning_rate": 1.624974071703679e-05,
      "loss": 0.8592,
      "num_input_tokens_seen": 1713278464,
      "step": 23070,
      "train_runtime": 27362.4031,
      "train_tokens_per_second": 62614.327
    },
    {
      "epoch": 0.30670892552208956,
      "grad_norm": 12.125,
      "learning_rate": 1.624638026789574e-05,
      "loss": 0.7566,
      "num_input_tokens_seen": 1713757440,
      "step": 23080,
      "train_runtime": 27369.1062,
      "train_tokens_per_second": 62616.493
    },
    {
      "epoch": 0.3068418150045515,
      "grad_norm": 24.5,
      "learning_rate": 1.6243018661668446e-05,
      "loss": 0.6741,
      "num_input_tokens_seen": 1713985280,
      "step": 23090,
      "train_runtime": 27372.5993,
      "train_tokens_per_second": 62616.826
    },
    {
      "epoch": 0.30697470448701336,
      "grad_norm": 10.0625,
      "learning_rate": 1.6239655898977612e-05,
      "loss": 0.5235,
      "num_input_tokens_seen": 1714102016,
      "step": 23100,
      "train_runtime": 27375.0729,
      "train_tokens_per_second": 62615.432
    },
    {
      "epoch": 0.3071075939694753,
      "grad_norm": 13.6875,
      "learning_rate": 1.6236291980446157e-05,
      "loss": 0.9108,
      "num_input_tokens_seen": 1716055808,
      "step": 23110,
      "train_runtime": 27406.2507,
      "train_tokens_per_second": 62615.49
    },
    {
      "epoch": 0.3072404834519372,
      "grad_norm": 18.0,
      "learning_rate": 1.6232926906697222e-05,
      "loss": 0.9421,
      "num_input_tokens_seen": 1716919296,
      "step": 23120,
      "train_runtime": 27419.0232,
      "train_tokens_per_second": 62617.814
    },
    {
      "epoch": 0.3073733729343991,
      "grad_norm": 21.75,
      "learning_rate": 1.622956067835416e-05,
      "loss": 0.7677,
      "num_input_tokens_seen": 1717400832,
      "step": 23130,
      "train_runtime": 27425.7402,
      "train_tokens_per_second": 62620.036
    },
    {
      "epoch": 0.307506262416861,
      "grad_norm": 24.875,
      "learning_rate": 1.6226193296040527e-05,
      "loss": 0.6494,
      "num_input_tokens_seen": 1717633792,
      "step": 23140,
      "train_runtime": 27429.1961,
      "train_tokens_per_second": 62620.639
    },
    {
      "epoch": 0.30763915189932295,
      "grad_norm": 10.0625,
      "learning_rate": 1.6222824760380103e-05,
      "loss": 0.4983,
      "num_input_tokens_seen": 1717746688,
      "step": 23150,
      "train_runtime": 27432.116,
      "train_tokens_per_second": 62618.089
    },
    {
      "epoch": 0.3077720413817848,
      "grad_norm": 18.875,
      "learning_rate": 1.621945507199688e-05,
      "loss": 0.7618,
      "num_input_tokens_seen": 1719756544,
      "step": 23160,
      "train_runtime": 27465.0273,
      "train_tokens_per_second": 62616.233
    },
    {
      "epoch": 0.30790493086424675,
      "grad_norm": 18.75,
      "learning_rate": 1.6216084231515062e-05,
      "loss": 0.8846,
      "num_input_tokens_seen": 1720575232,
      "step": 23170,
      "train_runtime": 27477.574,
      "train_tokens_per_second": 62617.436
    },
    {
      "epoch": 0.3080378203467087,
      "grad_norm": 14.0625,
      "learning_rate": 1.621271223955907e-05,
      "loss": 0.7102,
      "num_input_tokens_seen": 1721040896,
      "step": 23180,
      "train_runtime": 27484.0764,
      "train_tokens_per_second": 62619.565
    },
    {
      "epoch": 0.30817070982917055,
      "grad_norm": 8.375,
      "learning_rate": 1.6209339096753535e-05,
      "loss": 0.6251,
      "num_input_tokens_seen": 1721270016,
      "step": 23190,
      "train_runtime": 27487.5017,
      "train_tokens_per_second": 62620.097
    },
    {
      "epoch": 0.3083035993116325,
      "grad_norm": 7.3125,
      "learning_rate": 1.6205964803723294e-05,
      "loss": 0.575,
      "num_input_tokens_seen": 1721386240,
      "step": 23200,
      "train_runtime": 27489.9618,
      "train_tokens_per_second": 62618.721
    },
    {
      "epoch": 0.3084364887940944,
      "grad_norm": 20.0,
      "learning_rate": 1.6202589361093418e-05,
      "loss": 0.7999,
      "num_input_tokens_seen": 1723370496,
      "step": 23210,
      "train_runtime": 27522.1742,
      "train_tokens_per_second": 62617.527
    },
    {
      "epoch": 0.3085693782765563,
      "grad_norm": 27.0,
      "learning_rate": 1.6199212769489167e-05,
      "loss": 0.8888,
      "num_input_tokens_seen": 1724223744,
      "step": 23220,
      "train_runtime": 27535.0266,
      "train_tokens_per_second": 62619.287
    },
    {
      "epoch": 0.3087022677590182,
      "grad_norm": 93.0,
      "learning_rate": 1.6195835029536026e-05,
      "loss": 0.7237,
      "num_input_tokens_seen": 1724679680,
      "step": 23230,
      "train_runtime": 27541.5553,
      "train_tokens_per_second": 62620.998
    },
    {
      "epoch": 0.30883515724148014,
      "grad_norm": 28.75,
      "learning_rate": 1.6192456141859697e-05,
      "loss": 0.655,
      "num_input_tokens_seen": 1724913664,
      "step": 23240,
      "train_runtime": 27545.0794,
      "train_tokens_per_second": 62621.481
    },
    {
      "epoch": 0.308968046723942,
      "grad_norm": 14.75,
      "learning_rate": 1.618907610708608e-05,
      "loss": 0.484,
      "num_input_tokens_seen": 1725032192,
      "step": 23250,
      "train_runtime": 27547.4856,
      "train_tokens_per_second": 62620.314
    },
    {
      "epoch": 0.30910093620640394,
      "grad_norm": 13.125,
      "learning_rate": 1.6185694925841303e-05,
      "loss": 0.8276,
      "num_input_tokens_seen": 1727182592,
      "step": 23260,
      "train_runtime": 27582.0695,
      "train_tokens_per_second": 62619.761
    },
    {
      "epoch": 0.30923382568886587,
      "grad_norm": 22.375,
      "learning_rate": 1.61823125987517e-05,
      "loss": 0.9183,
      "num_input_tokens_seen": 1728010496,
      "step": 23270,
      "train_runtime": 27594.8932,
      "train_tokens_per_second": 62620.663
    },
    {
      "epoch": 0.30936671517132774,
      "grad_norm": 36.75,
      "learning_rate": 1.6178929126443812e-05,
      "loss": 0.7749,
      "num_input_tokens_seen": 1728497152,
      "step": 23280,
      "train_runtime": 27601.7135,
      "train_tokens_per_second": 62622.82
    },
    {
      "epoch": 0.3094996046537897,
      "grad_norm": 14.625,
      "learning_rate": 1.6175544509544404e-05,
      "loss": 0.6266,
      "num_input_tokens_seen": 1728728576,
      "step": 23290,
      "train_runtime": 27605.189,
      "train_tokens_per_second": 62623.32
    },
    {
      "epoch": 0.3096324941362516,
      "grad_norm": 17.625,
      "learning_rate": 1.6172158748680437e-05,
      "loss": 0.473,
      "num_input_tokens_seen": 1728846848,
      "step": 23300,
      "train_runtime": 27607.6085,
      "train_tokens_per_second": 62622.115
    },
    {
      "epoch": 0.3097653836187135,
      "grad_norm": 13.4375,
      "learning_rate": 1.61687718444791e-05,
      "loss": 0.8539,
      "num_input_tokens_seen": 1730885120,
      "step": 23310,
      "train_runtime": 27641.1201,
      "train_tokens_per_second": 62619.934
    },
    {
      "epoch": 0.3098982731011754,
      "grad_norm": 26.25,
      "learning_rate": 1.6165383797567785e-05,
      "loss": 0.9102,
      "num_input_tokens_seen": 1731700480,
      "step": 23320,
      "train_runtime": 27654.1016,
      "train_tokens_per_second": 62620.023
    },
    {
      "epoch": 0.31003116258363733,
      "grad_norm": 15.375,
      "learning_rate": 1.6161994608574096e-05,
      "loss": 0.7359,
      "num_input_tokens_seen": 1732163328,
      "step": 23330,
      "train_runtime": 27660.7857,
      "train_tokens_per_second": 62621.624
    },
    {
      "epoch": 0.3101640520660992,
      "grad_norm": 15.1875,
      "learning_rate": 1.6158604278125852e-05,
      "loss": 0.6866,
      "num_input_tokens_seen": 1732400384,
      "step": 23340,
      "train_runtime": 27664.3128,
      "train_tokens_per_second": 62622.209
    },
    {
      "epoch": 0.31029694154856113,
      "grad_norm": 24.0,
      "learning_rate": 1.615521280685108e-05,
      "loss": 0.5955,
      "num_input_tokens_seen": 1732518144,
      "step": 23350,
      "train_runtime": 27666.7901,
      "train_tokens_per_second": 62620.858
    },
    {
      "epoch": 0.31042983103102306,
      "grad_norm": 33.75,
      "learning_rate": 1.615182019537802e-05,
      "loss": 0.9236,
      "num_input_tokens_seen": 1734615552,
      "step": 23360,
      "train_runtime": 27700.7318,
      "train_tokens_per_second": 62619.846
    },
    {
      "epoch": 0.31056272051348494,
      "grad_norm": 191.0,
      "learning_rate": 1.6148426444335128e-05,
      "loss": 0.8882,
      "num_input_tokens_seen": 1735467264,
      "step": 23370,
      "train_runtime": 27714.1755,
      "train_tokens_per_second": 62620.202
    },
    {
      "epoch": 0.31069560999594686,
      "grad_norm": 11.3125,
      "learning_rate": 1.614503155435106e-05,
      "loss": 0.7273,
      "num_input_tokens_seen": 1735946240,
      "step": 23380,
      "train_runtime": 27721.01,
      "train_tokens_per_second": 62622.041
    },
    {
      "epoch": 0.3108284994784088,
      "grad_norm": 14.5625,
      "learning_rate": 1.6141635526054688e-05,
      "loss": 0.6509,
      "num_input_tokens_seen": 1736180992,
      "step": 23390,
      "train_runtime": 27724.5372,
      "train_tokens_per_second": 62622.542
    },
    {
      "epoch": 0.31096138896087067,
      "grad_norm": 10.6875,
      "learning_rate": 1.61382383600751e-05,
      "loss": 0.5442,
      "num_input_tokens_seen": 1736297216,
      "step": 23400,
      "train_runtime": 27726.9755,
      "train_tokens_per_second": 62621.227
    },
    {
      "epoch": 0.3110942784433326,
      "grad_norm": 31.0,
      "learning_rate": 1.613484005704159e-05,
      "loss": 0.8047,
      "num_input_tokens_seen": 1738436608,
      "step": 23410,
      "train_runtime": 27762.0082,
      "train_tokens_per_second": 62619.267
    },
    {
      "epoch": 0.3112271679257945,
      "grad_norm": 13.4375,
      "learning_rate": 1.6131440617583664e-05,
      "loss": 0.8791,
      "num_input_tokens_seen": 1739298560,
      "step": 23420,
      "train_runtime": 27775.3696,
      "train_tokens_per_second": 62620.177
    },
    {
      "epoch": 0.3113600574082564,
      "grad_norm": 32.0,
      "learning_rate": 1.6128040042331032e-05,
      "loss": 0.778,
      "num_input_tokens_seen": 1739773696,
      "step": 23430,
      "train_runtime": 27782.1466,
      "train_tokens_per_second": 62622.004
    },
    {
      "epoch": 0.3114929468907183,
      "grad_norm": 19.0,
      "learning_rate": 1.6124638331913627e-05,
      "loss": 0.6824,
      "num_input_tokens_seen": 1740006912,
      "step": 23440,
      "train_runtime": 27785.6497,
      "train_tokens_per_second": 62622.502
    },
    {
      "epoch": 0.31162583637318025,
      "grad_norm": 12.25,
      "learning_rate": 1.612123548696159e-05,
      "loss": 0.5293,
      "num_input_tokens_seen": 1740125184,
      "step": 23450,
      "train_runtime": 27788.0372,
      "train_tokens_per_second": 62621.378
    },
    {
      "epoch": 0.3117587258556422,
      "grad_norm": 16.875,
      "learning_rate": 1.6117831508105254e-05,
      "loss": 0.8793,
      "num_input_tokens_seen": 1742298880,
      "step": 23460,
      "train_runtime": 27822.6076,
      "train_tokens_per_second": 62621.696
    },
    {
      "epoch": 0.31189161533810406,
      "grad_norm": 34.0,
      "learning_rate": 1.611442639597518e-05,
      "loss": 0.9225,
      "num_input_tokens_seen": 1743116800,
      "step": 23470,
      "train_runtime": 27834.9767,
      "train_tokens_per_second": 62623.253
    },
    {
      "epoch": 0.312024504820566,
      "grad_norm": 18.5,
      "learning_rate": 1.6111020151202144e-05,
      "loss": 0.8052,
      "num_input_tokens_seen": 1743563264,
      "step": 23480,
      "train_runtime": 27841.3218,
      "train_tokens_per_second": 62625.018
    },
    {
      "epoch": 0.3121573943030279,
      "grad_norm": 13.0625,
      "learning_rate": 1.6107612774417112e-05,
      "loss": 0.5975,
      "num_input_tokens_seen": 1743785984,
      "step": 23490,
      "train_runtime": 27845.1373,
      "train_tokens_per_second": 62624.435
    },
    {
      "epoch": 0.3122902837854898,
      "grad_norm": 16.875,
      "learning_rate": 1.6104204266251276e-05,
      "loss": 0.5728,
      "num_input_tokens_seen": 1743900672,
      "step": 23500,
      "train_runtime": 27847.5117,
      "train_tokens_per_second": 62623.214
    },
    {
      "epoch": 0.3124231732679517,
      "grad_norm": 21.125,
      "learning_rate": 1.610079462733603e-05,
      "loss": 0.8213,
      "num_input_tokens_seen": 1745975040,
      "step": 23510,
      "train_runtime": 27881.2961,
      "train_tokens_per_second": 62621.731
    },
    {
      "epoch": 0.31255606275041364,
      "grad_norm": 15.875,
      "learning_rate": 1.6097383858302978e-05,
      "loss": 0.9129,
      "num_input_tokens_seen": 1746817024,
      "step": 23520,
      "train_runtime": 27894.0735,
      "train_tokens_per_second": 62623.232
    },
    {
      "epoch": 0.3126889522328755,
      "grad_norm": 284.0,
      "learning_rate": 1.609397195978394e-05,
      "loss": 0.7577,
      "num_input_tokens_seen": 1747286784,
      "step": 23530,
      "train_runtime": 27900.8295,
      "train_tokens_per_second": 62624.904
    },
    {
      "epoch": 0.31282184171533745,
      "grad_norm": 17.375,
      "learning_rate": 1.609055893241093e-05,
      "loss": 0.6693,
      "num_input_tokens_seen": 1747517952,
      "step": 23540,
      "train_runtime": 27904.8912,
      "train_tokens_per_second": 62624.073
    },
    {
      "epoch": 0.3129547311977994,
      "grad_norm": 23.25,
      "learning_rate": 1.6087144776816187e-05,
      "loss": 0.5328,
      "num_input_tokens_seen": 1747632128,
      "step": 23550,
      "train_runtime": 27907.3089,
      "train_tokens_per_second": 62622.739
    },
    {
      "epoch": 0.31308762068026125,
      "grad_norm": 15.9375,
      "learning_rate": 1.6083729493632153e-05,
      "loss": 0.8339,
      "num_input_tokens_seen": 1749661696,
      "step": 23560,
      "train_runtime": 27940.5423,
      "train_tokens_per_second": 62620.893
    },
    {
      "epoch": 0.3132205101627232,
      "grad_norm": 19.75,
      "learning_rate": 1.608031308349148e-05,
      "loss": 0.9266,
      "num_input_tokens_seen": 1750546176,
      "step": 23570,
      "train_runtime": 27953.2879,
      "train_tokens_per_second": 62623.981
    },
    {
      "epoch": 0.3133533996451851,
      "grad_norm": 14.375,
      "learning_rate": 1.607689554702702e-05,
      "loss": 0.7872,
      "num_input_tokens_seen": 1751040512,
      "step": 23580,
      "train_runtime": 27960.174,
      "train_tokens_per_second": 62626.238
    },
    {
      "epoch": 0.313486289127647,
      "grad_norm": 15.875,
      "learning_rate": 1.6073476884871853e-05,
      "loss": 0.6943,
      "num_input_tokens_seen": 1751275008,
      "step": 23590,
      "train_runtime": 27963.6334,
      "train_tokens_per_second": 62626.876
    },
    {
      "epoch": 0.3136191786101089,
      "grad_norm": 36.75,
      "learning_rate": 1.6070057097659243e-05,
      "loss": 0.4969,
      "num_input_tokens_seen": 1751388160,
      "step": 23600,
      "train_runtime": 27965.9763,
      "train_tokens_per_second": 62625.676
    },
    {
      "epoch": 0.31375206809257083,
      "grad_norm": 23.375,
      "learning_rate": 1.6066636186022687e-05,
      "loss": 0.8311,
      "num_input_tokens_seen": 1753409536,
      "step": 23610,
      "train_runtime": 27998.9787,
      "train_tokens_per_second": 62624.053
    },
    {
      "epoch": 0.3138849575750327,
      "grad_norm": 28.625,
      "learning_rate": 1.6063214150595872e-05,
      "loss": 0.8592,
      "num_input_tokens_seen": 1754233344,
      "step": 23620,
      "train_runtime": 28011.8999,
      "train_tokens_per_second": 62624.576
    },
    {
      "epoch": 0.31401784705749464,
      "grad_norm": 29.75,
      "learning_rate": 1.60597909920127e-05,
      "loss": 0.732,
      "num_input_tokens_seen": 1754690304,
      "step": 23630,
      "train_runtime": 28018.4766,
      "train_tokens_per_second": 62626.185
    },
    {
      "epoch": 0.31415073653995657,
      "grad_norm": 17.75,
      "learning_rate": 1.605636671090728e-05,
      "loss": 0.6175,
      "num_input_tokens_seen": 1754917888,
      "step": 23640,
      "train_runtime": 28021.9072,
      "train_tokens_per_second": 62626.64
    },
    {
      "epoch": 0.31428362602241844,
      "grad_norm": 13.875,
      "learning_rate": 1.6052941307913927e-05,
      "loss": 0.5622,
      "num_input_tokens_seen": 1755030528,
      "step": 23650,
      "train_runtime": 28024.281,
      "train_tokens_per_second": 62625.354
    },
    {
      "epoch": 0.31441651550488037,
      "grad_norm": 29.5,
      "learning_rate": 1.6049514783667176e-05,
      "loss": 0.8619,
      "num_input_tokens_seen": 1757059584,
      "step": 23660,
      "train_runtime": 28058.1001,
      "train_tokens_per_second": 62622.187
    },
    {
      "epoch": 0.3145494049873423,
      "grad_norm": 15.0,
      "learning_rate": 1.604608713880175e-05,
      "loss": 0.9006,
      "num_input_tokens_seen": 1757910016,
      "step": 23670,
      "train_runtime": 28070.9488,
      "train_tokens_per_second": 62623.819
    },
    {
      "epoch": 0.31468229446980417,
      "grad_norm": 17.5,
      "learning_rate": 1.6042658373952595e-05,
      "loss": 0.7656,
      "num_input_tokens_seen": 1758365952,
      "step": 23680,
      "train_runtime": 28077.3172,
      "train_tokens_per_second": 62625.854
    },
    {
      "epoch": 0.3148151839522661,
      "grad_norm": 11.0,
      "learning_rate": 1.6039228489754857e-05,
      "loss": 0.672,
      "num_input_tokens_seen": 1758587648,
      "step": 23690,
      "train_runtime": 28080.668,
      "train_tokens_per_second": 62626.275
    },
    {
      "epoch": 0.314948073434728,
      "grad_norm": 10.75,
      "learning_rate": 1.603579748684389e-05,
      "loss": 0.5189,
      "num_input_tokens_seen": 1758699008,
      "step": 23700,
      "train_runtime": 28083.0579,
      "train_tokens_per_second": 62624.911
    },
    {
      "epoch": 0.3150809629171899,
      "grad_norm": 16.125,
      "learning_rate": 1.603236536585526e-05,
      "loss": 0.8431,
      "num_input_tokens_seen": 1760631808,
      "step": 23710,
      "train_runtime": 28113.9582,
      "train_tokens_per_second": 62624.828
    },
    {
      "epoch": 0.31521385239965183,
      "grad_norm": 15.6875,
      "learning_rate": 1.602893212742474e-05,
      "loss": 0.8743,
      "num_input_tokens_seen": 1761428736,
      "step": 23720,
      "train_runtime": 28126.5297,
      "train_tokens_per_second": 62625.171
    },
    {
      "epoch": 0.31534674188211376,
      "grad_norm": 18.125,
      "learning_rate": 1.6025497772188296e-05,
      "loss": 0.7304,
      "num_input_tokens_seen": 1761880576,
      "step": 23730,
      "train_runtime": 28132.8468,
      "train_tokens_per_second": 62627.17
    },
    {
      "epoch": 0.31547963136457563,
      "grad_norm": 10.1875,
      "learning_rate": 1.602206230078212e-05,
      "loss": 0.7033,
      "num_input_tokens_seen": 1762108160,
      "step": 23740,
      "train_runtime": 28136.232,
      "train_tokens_per_second": 62627.724
    },
    {
      "epoch": 0.31561252084703756,
      "grad_norm": 16.25,
      "learning_rate": 1.60186257138426e-05,
      "loss": 0.4913,
      "num_input_tokens_seen": 1762221056,
      "step": 23750,
      "train_runtime": 28138.6147,
      "train_tokens_per_second": 62626.433
    },
    {
      "epoch": 0.3157454103294995,
      "grad_norm": 17.125,
      "learning_rate": 1.601518801200634e-05,
      "loss": 0.8883,
      "num_input_tokens_seen": 1764259072,
      "step": 23760,
      "train_runtime": 28172.67,
      "train_tokens_per_second": 62623.07
    },
    {
      "epoch": 0.31587829981196136,
      "grad_norm": 19.0,
      "learning_rate": 1.601174919591013e-05,
      "loss": 0.894,
      "num_input_tokens_seen": 1765122816,
      "step": 23770,
      "train_runtime": 28185.0831,
      "train_tokens_per_second": 62626.135
    },
    {
      "epoch": 0.3160111892944233,
      "grad_norm": 12.5625,
      "learning_rate": 1.6008309266190988e-05,
      "loss": 0.7338,
      "num_input_tokens_seen": 1765558272,
      "step": 23780,
      "train_runtime": 28191.864,
      "train_tokens_per_second": 62626.518
    },
    {
      "epoch": 0.3161440787768852,
      "grad_norm": 15.75,
      "learning_rate": 1.6004868223486132e-05,
      "loss": 0.6504,
      "num_input_tokens_seen": 1765780992,
      "step": 23790,
      "train_runtime": 28195.234,
      "train_tokens_per_second": 62626.932
    },
    {
      "epoch": 0.3162769682593471,
      "grad_norm": 11.8125,
      "learning_rate": 1.6001426068432978e-05,
      "loss": 0.5583,
      "num_input_tokens_seen": 1765898496,
      "step": 23800,
      "train_runtime": 28197.6619,
      "train_tokens_per_second": 62625.706
    },
    {
      "epoch": 0.316409857741809,
      "grad_norm": 22.875,
      "learning_rate": 1.599798280166916e-05,
      "loss": 0.8092,
      "num_input_tokens_seen": 1767873280,
      "step": 23810,
      "train_runtime": 28230.3034,
      "train_tokens_per_second": 62623.248
    },
    {
      "epoch": 0.31654274722427095,
      "grad_norm": 15.8125,
      "learning_rate": 1.5994538423832517e-05,
      "loss": 0.8424,
      "num_input_tokens_seen": 1768711168,
      "step": 23820,
      "train_runtime": 28242.9326,
      "train_tokens_per_second": 62624.912
    },
    {
      "epoch": 0.3166756367067328,
      "grad_norm": 31.625,
      "learning_rate": 1.599109293556108e-05,
      "loss": 0.7694,
      "num_input_tokens_seen": 1769176576,
      "step": 23830,
      "train_runtime": 28250.0463,
      "train_tokens_per_second": 62625.617
    },
    {
      "epoch": 0.31680852618919475,
      "grad_norm": 17.875,
      "learning_rate": 1.5987646337493093e-05,
      "loss": 0.578,
      "num_input_tokens_seen": 1769402112,
      "step": 23840,
      "train_runtime": 28253.4503,
      "train_tokens_per_second": 62626.054
    },
    {
      "epoch": 0.3169414156716567,
      "grad_norm": 13.4375,
      "learning_rate": 1.598419863026702e-05,
      "loss": 0.5735,
      "num_input_tokens_seen": 1769516800,
      "step": 23850,
      "train_runtime": 28255.8139,
      "train_tokens_per_second": 62624.875
    },
    {
      "epoch": 0.31707430515411855,
      "grad_norm": 13.75,
      "learning_rate": 1.5980749814521507e-05,
      "loss": 0.8564,
      "num_input_tokens_seen": 1771468032,
      "step": 23860,
      "train_runtime": 28287.6161,
      "train_tokens_per_second": 62623.447
    },
    {
      "epoch": 0.3172071946365805,
      "grad_norm": 40.0,
      "learning_rate": 1.597729989089542e-05,
      "loss": 0.8944,
      "num_input_tokens_seen": 1772308480,
      "step": 23870,
      "train_runtime": 28300.1679,
      "train_tokens_per_second": 62625.37
    },
    {
      "epoch": 0.3173400841190424,
      "grad_norm": 18.875,
      "learning_rate": 1.5973848860027826e-05,
      "loss": 0.7716,
      "num_input_tokens_seen": 1772780288,
      "step": 23880,
      "train_runtime": 28307.2603,
      "train_tokens_per_second": 62626.346
    },
    {
      "epoch": 0.3174729736015043,
      "grad_norm": 13.875,
      "learning_rate": 1.5970396722558e-05,
      "loss": 0.6617,
      "num_input_tokens_seen": 1772999168,
      "step": 23890,
      "train_runtime": 28310.648,
      "train_tokens_per_second": 62626.584
    },
    {
      "epoch": 0.3176058630839662,
      "grad_norm": 14.4375,
      "learning_rate": 1.5966943479125422e-05,
      "loss": 0.5387,
      "num_input_tokens_seen": 1773112832,
      "step": 23900,
      "train_runtime": 28312.9857,
      "train_tokens_per_second": 62625.428
    },
    {
      "epoch": 0.31773875256642814,
      "grad_norm": 11.0,
      "learning_rate": 1.5963489130369765e-05,
      "loss": 0.8564,
      "num_input_tokens_seen": 1775400448,
      "step": 23910,
      "train_runtime": 28348.7031,
      "train_tokens_per_second": 62627.219
    },
    {
      "epoch": 0.31787164204889,
      "grad_norm": 39.0,
      "learning_rate": 1.5960033676930927e-05,
      "loss": 0.8921,
      "num_input_tokens_seen": 1776245504,
      "step": 23920,
      "train_runtime": 28361.674,
      "train_tokens_per_second": 62628.373
    },
    {
      "epoch": 0.31800453153135194,
      "grad_norm": 21.125,
      "learning_rate": 1.5956577119448992e-05,
      "loss": 0.8308,
      "num_input_tokens_seen": 1776734720,
      "step": 23930,
      "train_runtime": 28368.499,
      "train_tokens_per_second": 62630.551
    },
    {
      "epoch": 0.31813742101381387,
      "grad_norm": 21.375,
      "learning_rate": 1.5953119458564258e-05,
      "loss": 0.6623,
      "num_input_tokens_seen": 1776963840,
      "step": 23940,
      "train_runtime": 28371.9957,
      "train_tokens_per_second": 62630.908
    },
    {
      "epoch": 0.3182703104962758,
      "grad_norm": 13.5625,
      "learning_rate": 1.594966069491723e-05,
      "loss": 0.5212,
      "num_input_tokens_seen": 1777080320,
      "step": 23950,
      "train_runtime": 28374.4479,
      "train_tokens_per_second": 62629.6
    },
    {
      "epoch": 0.31840319997873767,
      "grad_norm": 23.25,
      "learning_rate": 1.594620082914861e-05,
      "loss": 0.9425,
      "num_input_tokens_seen": 1779081216,
      "step": 23960,
      "train_runtime": 28406.1887,
      "train_tokens_per_second": 62630.057
    },
    {
      "epoch": 0.3185360894611996,
      "grad_norm": 29.125,
      "learning_rate": 1.5942739861899306e-05,
      "loss": 0.9909,
      "num_input_tokens_seen": 1779903232,
      "step": 23970,
      "train_runtime": 28418.3431,
      "train_tokens_per_second": 62632.196
    },
    {
      "epoch": 0.31866897894366153,
      "grad_norm": 16.75,
      "learning_rate": 1.5939277793810434e-05,
      "loss": 0.7606,
      "num_input_tokens_seen": 1780345344,
      "step": 23980,
      "train_runtime": 28424.6708,
      "train_tokens_per_second": 62633.807
    },
    {
      "epoch": 0.3188018684261234,
      "grad_norm": 12.0625,
      "learning_rate": 1.593581462552331e-05,
      "loss": 0.6,
      "num_input_tokens_seen": 1780571392,
      "step": 23990,
      "train_runtime": 28428.0466,
      "train_tokens_per_second": 62634.321
    },
    {
      "epoch": 0.31893475790858533,
      "grad_norm": 12.5625,
      "learning_rate": 1.5932350357679455e-05,
      "loss": 0.5233,
      "num_input_tokens_seen": 1780684288,
      "step": 24000,
      "train_runtime": 28430.4097,
      "train_tokens_per_second": 62633.086
    },
    {
      "epoch": 0.31906764739104726,
      "grad_norm": 18.5,
      "learning_rate": 1.5928884990920598e-05,
      "loss": 0.847,
      "num_input_tokens_seen": 1782806784,
      "step": 24010,
      "train_runtime": 28464.8926,
      "train_tokens_per_second": 62631.776
    },
    {
      "epoch": 0.31920053687350913,
      "grad_norm": 15.625,
      "learning_rate": 1.592541852588866e-05,
      "loss": 0.9507,
      "num_input_tokens_seen": 1783678208,
      "step": 24020,
      "train_runtime": 28477.9949,
      "train_tokens_per_second": 62633.56
    },
    {
      "epoch": 0.31933342635597106,
      "grad_norm": 55.75,
      "learning_rate": 1.5921950963225777e-05,
      "loss": 0.8036,
      "num_input_tokens_seen": 1784183040,
      "step": 24030,
      "train_runtime": 28485.1018,
      "train_tokens_per_second": 62635.656
    },
    {
      "epoch": 0.319466315838433,
      "grad_norm": 17.25,
      "learning_rate": 1.591848230357428e-05,
      "loss": 0.6832,
      "num_input_tokens_seen": 1784411904,
      "step": 24040,
      "train_runtime": 28488.58,
      "train_tokens_per_second": 62636.042
    },
    {
      "epoch": 0.31959920532089486,
      "grad_norm": 10.9375,
      "learning_rate": 1.5915012547576713e-05,
      "loss": 0.5101,
      "num_input_tokens_seen": 1784527104,
      "step": 24050,
      "train_runtime": 28490.9732,
      "train_tokens_per_second": 62634.824
    },
    {
      "epoch": 0.3197320948033568,
      "grad_norm": 19.125,
      "learning_rate": 1.5911541695875815e-05,
      "loss": 0.8154,
      "num_input_tokens_seen": 1786578176,
      "step": 24060,
      "train_runtime": 28524.4994,
      "train_tokens_per_second": 62633.112
    },
    {
      "epoch": 0.3198649842858187,
      "grad_norm": 19.5,
      "learning_rate": 1.590806974911453e-05,
      "loss": 0.9543,
      "num_input_tokens_seen": 1787419648,
      "step": 24070,
      "train_runtime": 28536.8797,
      "train_tokens_per_second": 62635.427
    },
    {
      "epoch": 0.3199978737682806,
      "grad_norm": 20.5,
      "learning_rate": 1.5904596707936005e-05,
      "loss": 0.739,
      "num_input_tokens_seen": 1787864832,
      "step": 24080,
      "train_runtime": 28543.3198,
      "train_tokens_per_second": 62636.892
    },
    {
      "epoch": 0.3201307632507425,
      "grad_norm": 11.125,
      "learning_rate": 1.590112257298359e-05,
      "loss": 0.6225,
      "num_input_tokens_seen": 1788083712,
      "step": 24090,
      "train_runtime": 28546.6673,
      "train_tokens_per_second": 62637.214
    },
    {
      "epoch": 0.32026365273320445,
      "grad_norm": 65.0,
      "learning_rate": 1.589764734490084e-05,
      "loss": 0.577,
      "num_input_tokens_seen": 1788198656,
      "step": 24100,
      "train_runtime": 28549.0143,
      "train_tokens_per_second": 62636.091
    },
    {
      "epoch": 0.3203965422156663,
      "grad_norm": 15.8125,
      "learning_rate": 1.58941710243315e-05,
      "loss": 0.8616,
      "num_input_tokens_seen": 1790349568,
      "step": 24110,
      "train_runtime": 28584.1639,
      "train_tokens_per_second": 62634.316
    },
    {
      "epoch": 0.32052943169812825,
      "grad_norm": 26.5,
      "learning_rate": 1.589069361191954e-05,
      "loss": 0.8861,
      "num_input_tokens_seen": 1791176960,
      "step": 24120,
      "train_runtime": 28597.3031,
      "train_tokens_per_second": 62634.471
    },
    {
      "epoch": 0.3206623211805902,
      "grad_norm": 13.3125,
      "learning_rate": 1.5887215108309114e-05,
      "loss": 0.676,
      "num_input_tokens_seen": 1791637504,
      "step": 24130,
      "train_runtime": 28603.9418,
      "train_tokens_per_second": 62636.035
    },
    {
      "epoch": 0.32079521066305206,
      "grad_norm": 12.8125,
      "learning_rate": 1.588373551414459e-05,
      "loss": 0.6197,
      "num_input_tokens_seen": 1791861504,
      "step": 24140,
      "train_runtime": 28607.3315,
      "train_tokens_per_second": 62636.444
    },
    {
      "epoch": 0.320928100145514,
      "grad_norm": 12.625,
      "learning_rate": 1.5880254830070518e-05,
      "loss": 0.5302,
      "num_input_tokens_seen": 1791975680,
      "step": 24150,
      "train_runtime": 28609.6823,
      "train_tokens_per_second": 62635.288
    },
    {
      "epoch": 0.3210609896279759,
      "grad_norm": 13.5625,
      "learning_rate": 1.5876773056731678e-05,
      "loss": 0.881,
      "num_input_tokens_seen": 1794097664,
      "step": 24160,
      "train_runtime": 28641.6096,
      "train_tokens_per_second": 62639.554
    },
    {
      "epoch": 0.3211938791104378,
      "grad_norm": 20.875,
      "learning_rate": 1.587329019477303e-05,
      "loss": 0.879,
      "num_input_tokens_seen": 1794946560,
      "step": 24170,
      "train_runtime": 28654.6386,
      "train_tokens_per_second": 62640.698
    },
    {
      "epoch": 0.3213267685928997,
      "grad_norm": 16.375,
      "learning_rate": 1.5869806244839744e-05,
      "loss": 0.758,
      "num_input_tokens_seen": 1795425280,
      "step": 24180,
      "train_runtime": 28661.372,
      "train_tokens_per_second": 62642.684
    },
    {
      "epoch": 0.32145965807536164,
      "grad_norm": 23.625,
      "learning_rate": 1.5866321207577192e-05,
      "loss": 0.6783,
      "num_input_tokens_seen": 1795655936,
      "step": 24190,
      "train_runtime": 28664.8117,
      "train_tokens_per_second": 62643.214
    },
    {
      "epoch": 0.3215925475578235,
      "grad_norm": 18.25,
      "learning_rate": 1.5862835083630943e-05,
      "loss": 0.5549,
      "num_input_tokens_seen": 1795769344,
      "step": 24200,
      "train_runtime": 28667.1654,
      "train_tokens_per_second": 62642.027
    },
    {
      "epoch": 0.32172543704028544,
      "grad_norm": 14.0,
      "learning_rate": 1.5859347873646774e-05,
      "loss": 0.8242,
      "num_input_tokens_seen": 1797720832,
      "step": 24210,
      "train_runtime": 28698.5484,
      "train_tokens_per_second": 62641.525
    },
    {
      "epoch": 0.3218583265227474,
      "grad_norm": 11.0625,
      "learning_rate": 1.5855859578270656e-05,
      "loss": 0.9045,
      "num_input_tokens_seen": 1798512896,
      "step": 24220,
      "train_runtime": 28710.8341,
      "train_tokens_per_second": 62642.307
    },
    {
      "epoch": 0.32199121600520925,
      "grad_norm": 44.75,
      "learning_rate": 1.585237019814877e-05,
      "loss": 0.6962,
      "num_input_tokens_seen": 1798957312,
      "step": 24230,
      "train_runtime": 28717.357,
      "train_tokens_per_second": 62643.554
    },
    {
      "epoch": 0.3221241054876712,
      "grad_norm": 12.8125,
      "learning_rate": 1.5848879733927492e-05,
      "loss": 0.6223,
      "num_input_tokens_seen": 1799183360,
      "step": 24240,
      "train_runtime": 28720.761,
      "train_tokens_per_second": 62644.0
    },
    {
      "epoch": 0.3222569949701331,
      "grad_norm": 16.625,
      "learning_rate": 1.5845388186253393e-05,
      "loss": 0.5206,
      "num_input_tokens_seen": 1799299328,
      "step": 24250,
      "train_runtime": 28723.6802,
      "train_tokens_per_second": 62641.671
    },
    {
      "epoch": 0.322389884452595,
      "grad_norm": 21.5,
      "learning_rate": 1.5841895555773253e-05,
      "loss": 0.9236,
      "num_input_tokens_seen": 1801337344,
      "step": 24260,
      "train_runtime": 28757.223,
      "train_tokens_per_second": 62639.475
    },
    {
      "epoch": 0.3225227739350569,
      "grad_norm": 13.125,
      "learning_rate": 1.5838401843134058e-05,
      "loss": 0.9099,
      "num_input_tokens_seen": 1802155008,
      "step": 24270,
      "train_runtime": 28769.5121,
      "train_tokens_per_second": 62641.139
    },
    {
      "epoch": 0.32265566341751883,
      "grad_norm": 24.25,
      "learning_rate": 1.583490704898298e-05,
      "loss": 0.6837,
      "num_input_tokens_seen": 1802619904,
      "step": 24280,
      "train_runtime": 28775.974,
      "train_tokens_per_second": 62643.228
    },
    {
      "epoch": 0.3227885528999807,
      "grad_norm": 16.875,
      "learning_rate": 1.5831411173967397e-05,
      "loss": 0.588,
      "num_input_tokens_seen": 1802845696,
      "step": 24290,
      "train_runtime": 28779.4182,
      "train_tokens_per_second": 62643.577
    },
    {
      "epoch": 0.32292144238244264,
      "grad_norm": 19.125,
      "learning_rate": 1.5827914218734894e-05,
      "loss": 0.5312,
      "num_input_tokens_seen": 1802958592,
      "step": 24300,
      "train_runtime": 28781.8026,
      "train_tokens_per_second": 62642.31
    },
    {
      "epoch": 0.32305433186490456,
      "grad_norm": 29.125,
      "learning_rate": 1.5824416183933246e-05,
      "loss": 0.8529,
      "num_input_tokens_seen": 1804822016,
      "step": 24310,
      "train_runtime": 28811.6545,
      "train_tokens_per_second": 62642.082
    },
    {
      "epoch": 0.32318722134736644,
      "grad_norm": 29.875,
      "learning_rate": 1.582091707021044e-05,
      "loss": 0.9097,
      "num_input_tokens_seen": 1805658880,
      "step": 24320,
      "train_runtime": 28823.9094,
      "train_tokens_per_second": 62644.482
    },
    {
      "epoch": 0.32332011082982837,
      "grad_norm": 15.6875,
      "learning_rate": 1.5817416878214648e-05,
      "loss": 0.7395,
      "num_input_tokens_seen": 1806098176,
      "step": 24330,
      "train_runtime": 28830.3318,
      "train_tokens_per_second": 62645.764
    },
    {
      "epoch": 0.3234530003122903,
      "grad_norm": 24.375,
      "learning_rate": 1.5813915608594252e-05,
      "loss": 0.608,
      "num_input_tokens_seen": 1806319616,
      "step": 24340,
      "train_runtime": 28833.732,
      "train_tokens_per_second": 62646.057
    },
    {
      "epoch": 0.32358588979475217,
      "grad_norm": 8.6875,
      "learning_rate": 1.5810413261997834e-05,
      "loss": 0.4848,
      "num_input_tokens_seen": 1806433536,
      "step": 24350,
      "train_runtime": 28836.0971,
      "train_tokens_per_second": 62644.869
    },
    {
      "epoch": 0.3237187792772141,
      "grad_norm": 28.125,
      "learning_rate": 1.5806909839074162e-05,
      "loss": 0.8371,
      "num_input_tokens_seen": 1808360192,
      "step": 24360,
      "train_runtime": 28867.5857,
      "train_tokens_per_second": 62643.278
    },
    {
      "epoch": 0.323851668759676,
      "grad_norm": 22.125,
      "learning_rate": 1.580340534047223e-05,
      "loss": 0.9057,
      "num_input_tokens_seen": 1809161728,
      "step": 24370,
      "train_runtime": 28879.768,
      "train_tokens_per_second": 62644.607
    },
    {
      "epoch": 0.3239845582421379,
      "grad_norm": 13.375,
      "learning_rate": 1.5799899766841202e-05,
      "loss": 0.8407,
      "num_input_tokens_seen": 1809619712,
      "step": 24380,
      "train_runtime": 28886.282,
      "train_tokens_per_second": 62646.335
    },
    {
      "epoch": 0.3241174477245998,
      "grad_norm": 11.875,
      "learning_rate": 1.579639311883046e-05,
      "loss": 0.5896,
      "num_input_tokens_seen": 1809848320,
      "step": 24390,
      "train_runtime": 28889.7652,
      "train_tokens_per_second": 62646.695
    },
    {
      "epoch": 0.32425033720706176,
      "grad_norm": 20.75,
      "learning_rate": 1.5792885397089583e-05,
      "loss": 0.5515,
      "num_input_tokens_seen": 1809960192,
      "step": 24400,
      "train_runtime": 28892.761,
      "train_tokens_per_second": 62644.072
    },
    {
      "epoch": 0.3243832266895237,
      "grad_norm": 19.5,
      "learning_rate": 1.5789376602268333e-05,
      "loss": 0.7405,
      "num_input_tokens_seen": 1811840768,
      "step": 24410,
      "train_runtime": 28922.9377,
      "train_tokens_per_second": 62643.732
    },
    {
      "epoch": 0.32451611617198556,
      "grad_norm": 122.5,
      "learning_rate": 1.5785866735016693e-05,
      "loss": 0.9246,
      "num_input_tokens_seen": 1812648192,
      "step": 24420,
      "train_runtime": 28935.0062,
      "train_tokens_per_second": 62645.509
    },
    {
      "epoch": 0.3246490056544475,
      "grad_norm": 19.5,
      "learning_rate": 1.5782355795984833e-05,
      "loss": 0.6964,
      "num_input_tokens_seen": 1813081856,
      "step": 24430,
      "train_runtime": 28941.3508,
      "train_tokens_per_second": 62646.76
    },
    {
      "epoch": 0.3247818951369094,
      "grad_norm": 18.5,
      "learning_rate": 1.577884378582312e-05,
      "loss": 0.6481,
      "num_input_tokens_seen": 1813300224,
      "step": 24440,
      "train_runtime": 28944.7088,
      "train_tokens_per_second": 62647.036
    },
    {
      "epoch": 0.3249147846193713,
      "grad_norm": 18.0,
      "learning_rate": 1.577533070518213e-05,
      "loss": 0.5233,
      "num_input_tokens_seen": 1813412352,
      "step": 24450,
      "train_runtime": 28947.1066,
      "train_tokens_per_second": 62645.721
    },
    {
      "epoch": 0.3250476741018332,
      "grad_norm": 17.0,
      "learning_rate": 1.5771816554712618e-05,
      "loss": 0.8067,
      "num_input_tokens_seen": 1815508736,
      "step": 24460,
      "train_runtime": 28982.4398,
      "train_tokens_per_second": 62641.68
    },
    {
      "epoch": 0.32518056358429515,
      "grad_norm": 10.6875,
      "learning_rate": 1.576830133506556e-05,
      "loss": 0.9644,
      "num_input_tokens_seen": 1816366848,
      "step": 24470,
      "train_runtime": 28995.352,
      "train_tokens_per_second": 62643.38
    },
    {
      "epoch": 0.325313453066757,
      "grad_norm": 26.25,
      "learning_rate": 1.5764785046892114e-05,
      "loss": 0.8133,
      "num_input_tokens_seen": 1816851456,
      "step": 24480,
      "train_runtime": 29002.7525,
      "train_tokens_per_second": 62644.104
    },
    {
      "epoch": 0.32544634254921895,
      "grad_norm": 55.25,
      "learning_rate": 1.576126769084364e-05,
      "loss": 0.6048,
      "num_input_tokens_seen": 1817080832,
      "step": 24490,
      "train_runtime": 29006.2464,
      "train_tokens_per_second": 62644.467
    },
    {
      "epoch": 0.3255792320316809,
      "grad_norm": 16.625,
      "learning_rate": 1.5757749267571698e-05,
      "loss": 0.4638,
      "num_input_tokens_seen": 1817199360,
      "step": 24500,
      "train_runtime": 29008.6295,
      "train_tokens_per_second": 62643.406
    },
    {
      "epoch": 0.32571212151414275,
      "grad_norm": 25.0,
      "learning_rate": 1.5754229777728046e-05,
      "loss": 0.8286,
      "num_input_tokens_seen": 1819205120,
      "step": 24510,
      "train_runtime": 29041.9827,
      "train_tokens_per_second": 62640.528
    },
    {
      "epoch": 0.3258450109966047,
      "grad_norm": 11.8125,
      "learning_rate": 1.575070922196464e-05,
      "loss": 0.9224,
      "num_input_tokens_seen": 1820023040,
      "step": 24520,
      "train_runtime": 29054.5808,
      "train_tokens_per_second": 62641.518
    },
    {
      "epoch": 0.3259779004790666,
      "grad_norm": 18.0,
      "learning_rate": 1.5747187600933625e-05,
      "loss": 0.741,
      "num_input_tokens_seen": 1820501760,
      "step": 24530,
      "train_runtime": 29061.3181,
      "train_tokens_per_second": 62643.468
    },
    {
      "epoch": 0.3261107899615285,
      "grad_norm": 18.0,
      "learning_rate": 1.5743664915287355e-05,
      "loss": 0.5485,
      "num_input_tokens_seen": 1820732160,
      "step": 24540,
      "train_runtime": 29065.5397,
      "train_tokens_per_second": 62642.297
    },
    {
      "epoch": 0.3262436794439904,
      "grad_norm": 7.1875,
      "learning_rate": 1.5740141165678375e-05,
      "loss": 0.5345,
      "num_input_tokens_seen": 1820846080,
      "step": 24550,
      "train_runtime": 29068.1338,
      "train_tokens_per_second": 62640.625
    },
    {
      "epoch": 0.32637656892645234,
      "grad_norm": 15.25,
      "learning_rate": 1.5736616352759425e-05,
      "loss": 0.7838,
      "num_input_tokens_seen": 1822854912,
      "step": 24560,
      "train_runtime": 29100.3986,
      "train_tokens_per_second": 62640.204
    },
    {
      "epoch": 0.3265094584089142,
      "grad_norm": 32.5,
      "learning_rate": 1.5733090477183454e-05,
      "loss": 0.9039,
      "num_input_tokens_seen": 1823674112,
      "step": 24570,
      "train_runtime": 29112.5146,
      "train_tokens_per_second": 62642.274
    },
    {
      "epoch": 0.32664234789137614,
      "grad_norm": 42.5,
      "learning_rate": 1.572956353960359e-05,
      "loss": 0.6823,
      "num_input_tokens_seen": 1824129024,
      "step": 24580,
      "train_runtime": 29118.8446,
      "train_tokens_per_second": 62644.279
    },
    {
      "epoch": 0.32677523737383807,
      "grad_norm": 55.25,
      "learning_rate": 1.572603554067317e-05,
      "loss": 0.672,
      "num_input_tokens_seen": 1824358400,
      "step": 24590,
      "train_runtime": 29122.8121,
      "train_tokens_per_second": 62643.621
    },
    {
      "epoch": 0.32690812685629994,
      "grad_norm": 58.0,
      "learning_rate": 1.5722506481045722e-05,
      "loss": 0.5622,
      "num_input_tokens_seen": 1824473088,
      "step": 24600,
      "train_runtime": 29125.2587,
      "train_tokens_per_second": 62642.296
    },
    {
      "epoch": 0.32704101633876187,
      "grad_norm": 27.375,
      "learning_rate": 1.5718976361374976e-05,
      "loss": 0.7894,
      "num_input_tokens_seen": 1826467584,
      "step": 24610,
      "train_runtime": 29158.0969,
      "train_tokens_per_second": 62640.151
    },
    {
      "epoch": 0.3271739058212238,
      "grad_norm": 12.9375,
      "learning_rate": 1.571544518231486e-05,
      "loss": 0.9153,
      "num_input_tokens_seen": 1827297280,
      "step": 24620,
      "train_runtime": 29170.3592,
      "train_tokens_per_second": 62642.262
    },
    {
      "epoch": 0.32730679530368567,
      "grad_norm": 15.9375,
      "learning_rate": 1.571191294451948e-05,
      "loss": 0.7709,
      "num_input_tokens_seen": 1827759104,
      "step": 24630,
      "train_runtime": 29176.7082,
      "train_tokens_per_second": 62644.459
    },
    {
      "epoch": 0.3274396847861476,
      "grad_norm": 22.0,
      "learning_rate": 1.5708379648643163e-05,
      "loss": 0.5968,
      "num_input_tokens_seen": 1827987712,
      "step": 24640,
      "train_runtime": 29180.1143,
      "train_tokens_per_second": 62644.981
    },
    {
      "epoch": 0.32757257426860953,
      "grad_norm": 37.0,
      "learning_rate": 1.5704845295340415e-05,
      "loss": 0.6646,
      "num_input_tokens_seen": 1828099328,
      "step": 24650,
      "train_runtime": 29182.4765,
      "train_tokens_per_second": 62643.735
    },
    {
      "epoch": 0.3277054637510714,
      "grad_norm": 20.375,
      "learning_rate": 1.5701309885265947e-05,
      "loss": 0.8182,
      "num_input_tokens_seen": 1830140160,
      "step": 24660,
      "train_runtime": 29216.2802,
      "train_tokens_per_second": 62641.108
    },
    {
      "epoch": 0.32783835323353333,
      "grad_norm": 14.25,
      "learning_rate": 1.5697773419074657e-05,
      "loss": 0.8582,
      "num_input_tokens_seen": 1830984448,
      "step": 24670,
      "train_runtime": 29228.5327,
      "train_tokens_per_second": 62643.735
    },
    {
      "epoch": 0.32797124271599526,
      "grad_norm": 15.5625,
      "learning_rate": 1.5694235897421652e-05,
      "loss": 0.7872,
      "num_input_tokens_seen": 1831429888,
      "step": 24680,
      "train_runtime": 29235.0071,
      "train_tokens_per_second": 62645.098
    },
    {
      "epoch": 0.32810413219845713,
      "grad_norm": 18.5,
      "learning_rate": 1.5690697320962218e-05,
      "loss": 0.6381,
      "num_input_tokens_seen": 1831650816,
      "step": 24690,
      "train_runtime": 29238.4187,
      "train_tokens_per_second": 62645.345
    },
    {
      "epoch": 0.32823702168091906,
      "grad_norm": 8.9375,
      "learning_rate": 1.5687157690351852e-05,
      "loss": 0.4628,
      "num_input_tokens_seen": 1831768576,
      "step": 24700,
      "train_runtime": 29240.9104,
      "train_tokens_per_second": 62644.034
    },
    {
      "epoch": 0.328369911163381,
      "grad_norm": 39.25,
      "learning_rate": 1.568361700624623e-05,
      "loss": 0.8715,
      "num_input_tokens_seen": 1833864704,
      "step": 24710,
      "train_runtime": 29274.3811,
      "train_tokens_per_second": 62644.013
    },
    {
      "epoch": 0.32850280064584286,
      "grad_norm": 10.8125,
      "learning_rate": 1.568007526930124e-05,
      "loss": 0.9208,
      "num_input_tokens_seen": 1834690816,
      "step": 24720,
      "train_runtime": 29286.8305,
      "train_tokens_per_second": 62645.591
    },
    {
      "epoch": 0.3286356901283048,
      "grad_norm": 20.625,
      "learning_rate": 1.5676532480172958e-05,
      "loss": 0.7703,
      "num_input_tokens_seen": 1835165440,
      "step": 24730,
      "train_runtime": 29293.5492,
      "train_tokens_per_second": 62647.425
    },
    {
      "epoch": 0.3287685796107667,
      "grad_norm": 28.125,
      "learning_rate": 1.5672988639517645e-05,
      "loss": 0.7,
      "num_input_tokens_seen": 1835397120,
      "step": 24740,
      "train_runtime": 29297.0057,
      "train_tokens_per_second": 62647.942
    },
    {
      "epoch": 0.3289014690932286,
      "grad_norm": 21.375,
      "learning_rate": 1.5669443747991777e-05,
      "loss": 0.5092,
      "num_input_tokens_seen": 1835509760,
      "step": 24750,
      "train_runtime": 29299.8672,
      "train_tokens_per_second": 62645.668
    },
    {
      "epoch": 0.3290343585756905,
      "grad_norm": 15.0,
      "learning_rate": 1.5665897806252008e-05,
      "loss": 0.8648,
      "num_input_tokens_seen": 1837600512,
      "step": 24760,
      "train_runtime": 29333.3434,
      "train_tokens_per_second": 62645.45
    },
    {
      "epoch": 0.32916724805815245,
      "grad_norm": 15.6875,
      "learning_rate": 1.5662350814955187e-05,
      "loss": 0.929,
      "num_input_tokens_seen": 1838426624,
      "step": 24770,
      "train_runtime": 29346.1397,
      "train_tokens_per_second": 62646.285
    },
    {
      "epoch": 0.3293001375406143,
      "grad_norm": 22.875,
      "learning_rate": 1.565880277475837e-05,
      "loss": 0.7973,
      "num_input_tokens_seen": 1838923776,
      "step": 24780,
      "train_runtime": 29353.1583,
      "train_tokens_per_second": 62648.242
    },
    {
      "epoch": 0.32943302702307625,
      "grad_norm": 18.625,
      "learning_rate": 1.56552536863188e-05,
      "loss": 0.6664,
      "num_input_tokens_seen": 1839172352,
      "step": 24790,
      "train_runtime": 29356.806,
      "train_tokens_per_second": 62648.925
    },
    {
      "epoch": 0.3295659165055382,
      "grad_norm": 9.625,
      "learning_rate": 1.5651703550293907e-05,
      "loss": 0.5025,
      "num_input_tokens_seen": 1839288832,
      "step": 24800,
      "train_runtime": 29359.2672,
      "train_tokens_per_second": 62647.641
    },
    {
      "epoch": 0.32969880598800005,
      "grad_norm": 38.75,
      "learning_rate": 1.5648152367341332e-05,
      "loss": 0.856,
      "num_input_tokens_seen": 1841350912,
      "step": 24810,
      "train_runtime": 29392.6259,
      "train_tokens_per_second": 62646.696
    },
    {
      "epoch": 0.329831695470462,
      "grad_norm": 27.125,
      "learning_rate": 1.5644600138118892e-05,
      "loss": 0.8616,
      "num_input_tokens_seen": 1842219264,
      "step": 24820,
      "train_runtime": 29405.4285,
      "train_tokens_per_second": 62648.951
    },
    {
      "epoch": 0.3299645849529239,
      "grad_norm": 23.75,
      "learning_rate": 1.5641046863284607e-05,
      "loss": 0.7747,
      "num_input_tokens_seen": 1842718208,
      "step": 24830,
      "train_runtime": 29412.8722,
      "train_tokens_per_second": 62650.06
    },
    {
      "epoch": 0.3300974744353858,
      "grad_norm": 16.125,
      "learning_rate": 1.5637492543496694e-05,
      "loss": 0.6386,
      "num_input_tokens_seen": 1842950656,
      "step": 24840,
      "train_runtime": 29416.4478,
      "train_tokens_per_second": 62650.347
    },
    {
      "epoch": 0.3302303639178477,
      "grad_norm": 16.375,
      "learning_rate": 1.5633937179413553e-05,
      "loss": 0.504,
      "num_input_tokens_seen": 1843069696,
      "step": 24850,
      "train_runtime": 29418.9121,
      "train_tokens_per_second": 62649.145
    },
    {
      "epoch": 0.33036325340030964,
      "grad_norm": 22.125,
      "learning_rate": 1.563038077169379e-05,
      "loss": 0.794,
      "num_input_tokens_seen": 1845137664,
      "step": 24860,
      "train_runtime": 29452.0225,
      "train_tokens_per_second": 62648.929
    },
    {
      "epoch": 0.3304961428827715,
      "grad_norm": 26.75,
      "learning_rate": 1.562682332099619e-05,
      "loss": 0.9391,
      "num_input_tokens_seen": 1845970432,
      "step": 24870,
      "train_runtime": 29464.6233,
      "train_tokens_per_second": 62650.4
    },
    {
      "epoch": 0.33062903236523344,
      "grad_norm": 36.0,
      "learning_rate": 1.5623264827979742e-05,
      "loss": 0.7175,
      "num_input_tokens_seen": 1846447360,
      "step": 24880,
      "train_runtime": 29471.3268,
      "train_tokens_per_second": 62652.332
    },
    {
      "epoch": 0.3307619218476954,
      "grad_norm": 41.25,
      "learning_rate": 1.5619705293303633e-05,
      "loss": 0.6621,
      "num_input_tokens_seen": 1846680576,
      "step": 24890,
      "train_runtime": 29475.3197,
      "train_tokens_per_second": 62651.757
    },
    {
      "epoch": 0.3308948113301573,
      "grad_norm": 8.4375,
      "learning_rate": 1.5616144717627223e-05,
      "loss": 0.4656,
      "num_input_tokens_seen": 1846796032,
      "step": 24900,
      "train_runtime": 29477.6917,
      "train_tokens_per_second": 62650.633
    },
    {
      "epoch": 0.3310277008126192,
      "grad_norm": 33.5,
      "learning_rate": 1.5612583101610086e-05,
      "loss": 0.852,
      "num_input_tokens_seen": 1848807936,
      "step": 24910,
      "train_runtime": 29510.9384,
      "train_tokens_per_second": 62648.226
    },
    {
      "epoch": 0.3311605902950811,
      "grad_norm": 25.125,
      "learning_rate": 1.5609020445911974e-05,
      "loss": 0.9025,
      "num_input_tokens_seen": 1849695232,
      "step": 24920,
      "train_runtime": 29524.3657,
      "train_tokens_per_second": 62649.787
    },
    {
      "epoch": 0.33129347977754303,
      "grad_norm": 18.875,
      "learning_rate": 1.560545675119284e-05,
      "loss": 0.7943,
      "num_input_tokens_seen": 1850183424,
      "step": 24930,
      "train_runtime": 29531.5419,
      "train_tokens_per_second": 62651.095
    },
    {
      "epoch": 0.3314263692600049,
      "grad_norm": 14.8125,
      "learning_rate": 1.5601892018112822e-05,
      "loss": 0.6741,
      "num_input_tokens_seen": 1850430464,
      "step": 24940,
      "train_runtime": 29535.7576,
      "train_tokens_per_second": 62650.516
    },
    {
      "epoch": 0.33155925874246683,
      "grad_norm": 9.0625,
      "learning_rate": 1.5598326247332262e-05,
      "loss": 0.5191,
      "num_input_tokens_seen": 1850549504,
      "step": 24950,
      "train_runtime": 29538.1752,
      "train_tokens_per_second": 62649.419
    },
    {
      "epoch": 0.33169214822492876,
      "grad_norm": 18.125,
      "learning_rate": 1.5594759439511684e-05,
      "loss": 0.7984,
      "num_input_tokens_seen": 1852629248,
      "step": 24960,
      "train_runtime": 29571.4626,
      "train_tokens_per_second": 62649.226
    },
    {
      "epoch": 0.33182503770739064,
      "grad_norm": 37.75,
      "learning_rate": 1.5591191595311807e-05,
      "loss": 0.8655,
      "num_input_tokens_seen": 1853540096,
      "step": 24970,
      "train_runtime": 29584.5463,
      "train_tokens_per_second": 62652.308
    },
    {
      "epoch": 0.33195792718985256,
      "grad_norm": 96.0,
      "learning_rate": 1.5587622715393544e-05,
      "loss": 0.7101,
      "num_input_tokens_seen": 1854006528,
      "step": 24980,
      "train_runtime": 29591.2746,
      "train_tokens_per_second": 62653.825
    },
    {
      "epoch": 0.3320908166723145,
      "grad_norm": 16.5,
      "learning_rate": 1.5584052800417998e-05,
      "loss": 0.6591,
      "num_input_tokens_seen": 1854241024,
      "step": 24990,
      "train_runtime": 29595.2685,
      "train_tokens_per_second": 62653.293
    },
    {
      "epoch": 0.33222370615477637,
      "grad_norm": 14.875,
      "learning_rate": 1.558048185104646e-05,
      "loss": 0.5543,
      "num_input_tokens_seen": 1854358528,
      "step": 25000,
      "train_runtime": 29597.6606,
      "train_tokens_per_second": 62652.199
    },
    {
      "epoch": 0.33222370615477637,
      "eval_accuracy": 0.8183,
      "eval_balanced_accuracy": 0.8190838132368258,
      "eval_invalid_f1": 0.8327657616198804,
      "eval_invalid_precision": 0.8543909348441926,
      "eval_invalid_recall": 0.8122082585278276,
      "eval_loss": 0.3649894595146179,
      "eval_macro_f1": 0.8169302261848719,
      "eval_mcc": 0.6351136309156951,
      "eval_runtime": 78.3039,
      "eval_samples_per_second": 127.708,
      "eval_steps_per_second": 31.927,
      "eval_valid_f1": 0.8010946907498633,
      "eval_valid_precision": 0.777683315621679,
      "eval_valid_recall": 0.8259593679458239,
      "num_input_tokens_seen": 1854358528,
      "step": 25000
    },
    {
      "epoch": 0.3323565956372383,
      "grad_norm": 28.75,
      "learning_rate": 1.557690986794042e-05,
      "loss": 0.8797,
      "num_input_tokens_seen": 1856325632,
      "step": 25010,
      "train_runtime": 29711.7271,
      "train_tokens_per_second": 62477.877
    },
    {
      "epoch": 0.3324894851197002,
      "grad_norm": 14.9375,
      "learning_rate": 1.5573336851761553e-05,
      "loss": 0.9163,
      "num_input_tokens_seen": 1857132544,
      "step": 25020,
      "train_runtime": 29724.3219,
      "train_tokens_per_second": 62478.55
    },
    {
      "epoch": 0.3326223746021621,
      "grad_norm": 13.9375,
      "learning_rate": 1.5569762803171732e-05,
      "loss": 0.6924,
      "num_input_tokens_seen": 1857614336,
      "step": 25030,
      "train_runtime": 29730.9504,
      "train_tokens_per_second": 62480.826
    },
    {
      "epoch": 0.332755264084624,
      "grad_norm": 17.75,
      "learning_rate": 1.5566187722833016e-05,
      "loss": 0.6126,
      "num_input_tokens_seen": 1857834240,
      "step": 25040,
      "train_runtime": 29734.2854,
      "train_tokens_per_second": 62481.214
    },
    {
      "epoch": 0.33288815356708595,
      "grad_norm": 8.125,
      "learning_rate": 1.5562611611407662e-05,
      "loss": 0.5234,
      "num_input_tokens_seen": 1857946368,
      "step": 25050,
      "train_runtime": 29737.186,
      "train_tokens_per_second": 62478.89
    },
    {
      "epoch": 0.3330210430495478,
      "grad_norm": 40.75,
      "learning_rate": 1.5559034469558106e-05,
      "loss": 0.7485,
      "num_input_tokens_seen": 1859918848,
      "step": 25060,
      "train_runtime": 29768.8212,
      "train_tokens_per_second": 62478.754
    },
    {
      "epoch": 0.33315393253200976,
      "grad_norm": 20.625,
      "learning_rate": 1.555545629794698e-05,
      "loss": 0.9073,
      "num_input_tokens_seen": 1860725760,
      "step": 25070,
      "train_runtime": 29780.7602,
      "train_tokens_per_second": 62480.801
    },
    {
      "epoch": 0.3332868220144717,
      "grad_norm": 21.5,
      "learning_rate": 1.5551877097237112e-05,
      "loss": 0.7858,
      "num_input_tokens_seen": 1861172480,
      "step": 25080,
      "train_runtime": 29787.0906,
      "train_tokens_per_second": 62482.52
    },
    {
      "epoch": 0.33341971149693356,
      "grad_norm": 10.4375,
      "learning_rate": 1.5548296868091522e-05,
      "loss": 0.7122,
      "num_input_tokens_seen": 1861402368,
      "step": 25090,
      "train_runtime": 29790.4583,
      "train_tokens_per_second": 62483.173
    },
    {
      "epoch": 0.3335526009793955,
      "grad_norm": 12.5,
      "learning_rate": 1.5544715611173406e-05,
      "loss": 0.5015,
      "num_input_tokens_seen": 1861517312,
      "step": 25100,
      "train_runtime": 29793.3967,
      "train_tokens_per_second": 62480.869
    },
    {
      "epoch": 0.3336854904618574,
      "grad_norm": 17.0,
      "learning_rate": 1.5541133327146164e-05,
      "loss": 0.8675,
      "num_input_tokens_seen": 1863543040,
      "step": 25110,
      "train_runtime": 29826.1769,
      "train_tokens_per_second": 62480.118
    },
    {
      "epoch": 0.3338183799443193,
      "grad_norm": 16.125,
      "learning_rate": 1.5537550016673384e-05,
      "loss": 0.9286,
      "num_input_tokens_seen": 1864382464,
      "step": 25120,
      "train_runtime": 29838.8104,
      "train_tokens_per_second": 62481.796
    },
    {
      "epoch": 0.3339512694267812,
      "grad_norm": 15.5,
      "learning_rate": 1.553396568041884e-05,
      "loss": 0.7725,
      "num_input_tokens_seen": 1864840192,
      "step": 25130,
      "train_runtime": 29845.3574,
      "train_tokens_per_second": 62483.426
    },
    {
      "epoch": 0.33408415890924315,
      "grad_norm": 29.375,
      "learning_rate": 1.55303803190465e-05,
      "loss": 0.6219,
      "num_input_tokens_seen": 1865064704,
      "step": 25140,
      "train_runtime": 29848.7668,
      "train_tokens_per_second": 62483.811
    },
    {
      "epoch": 0.334217048391705,
      "grad_norm": 11.4375,
      "learning_rate": 1.552679393322052e-05,
      "loss": 0.5623,
      "num_input_tokens_seen": 1865180672,
      "step": 25150,
      "train_runtime": 29851.2375,
      "train_tokens_per_second": 62482.524
    },
    {
      "epoch": 0.33434993787416695,
      "grad_norm": 15.3125,
      "learning_rate": 1.5523206523605248e-05,
      "loss": 0.9062,
      "num_input_tokens_seen": 1867123456,
      "step": 25160,
      "train_runtime": 29883.012,
      "train_tokens_per_second": 62481.1
    },
    {
      "epoch": 0.3344828273566289,
      "grad_norm": 63.5,
      "learning_rate": 1.551961809086521e-05,
      "loss": 0.911,
      "num_input_tokens_seen": 1867958784,
      "step": 25170,
      "train_runtime": 29895.6388,
      "train_tokens_per_second": 62482.652
    },
    {
      "epoch": 0.33461571683909075,
      "grad_norm": 15.5625,
      "learning_rate": 1.5516028635665144e-05,
      "loss": 0.7742,
      "num_input_tokens_seen": 1868422400,
      "step": 25180,
      "train_runtime": 29902.2944,
      "train_tokens_per_second": 62484.249
    },
    {
      "epoch": 0.3347486063215527,
      "grad_norm": 16.5,
      "learning_rate": 1.5512438158669957e-05,
      "loss": 0.6883,
      "num_input_tokens_seen": 1868656384,
      "step": 25190,
      "train_runtime": 29905.8005,
      "train_tokens_per_second": 62484.747
    },
    {
      "epoch": 0.3348814958040146,
      "grad_norm": 10.375,
      "learning_rate": 1.550884666054475e-05,
      "loss": 0.5236,
      "num_input_tokens_seen": 1868770304,
      "step": 25200,
      "train_runtime": 29908.2262,
      "train_tokens_per_second": 62483.488
    },
    {
      "epoch": 0.3350143852864765,
      "grad_norm": 32.5,
      "learning_rate": 1.5505254141954827e-05,
      "loss": 0.8659,
      "num_input_tokens_seen": 1870901248,
      "step": 25210,
      "train_runtime": 29942.9192,
      "train_tokens_per_second": 62482.259
    },
    {
      "epoch": 0.3351472747689384,
      "grad_norm": 13.375,
      "learning_rate": 1.5501660603565664e-05,
      "loss": 0.8922,
      "num_input_tokens_seen": 1871738624,
      "step": 25220,
      "train_runtime": 29955.524,
      "train_tokens_per_second": 62483.922
    },
    {
      "epoch": 0.33528016425140034,
      "grad_norm": 30.25,
      "learning_rate": 1.549806604604293e-05,
      "loss": 0.7348,
      "num_input_tokens_seen": 1872170240,
      "step": 25230,
      "train_runtime": 29961.9949,
      "train_tokens_per_second": 62484.833
    },
    {
      "epoch": 0.3354130537338622,
      "grad_norm": 33.5,
      "learning_rate": 1.5494470470052484e-05,
      "loss": 0.6113,
      "num_input_tokens_seen": 1872391424,
      "step": 25240,
      "train_runtime": 29965.4268,
      "train_tokens_per_second": 62485.058
    },
    {
      "epoch": 0.33554594321632414,
      "grad_norm": 14.6875,
      "learning_rate": 1.549087387626038e-05,
      "loss": 0.6603,
      "num_input_tokens_seen": 1872507904,
      "step": 25250,
      "train_runtime": 29968.4466,
      "train_tokens_per_second": 62482.648
    },
    {
      "epoch": 0.33567883269878607,
      "grad_norm": 13.9375,
      "learning_rate": 1.5487276265332848e-05,
      "loss": 0.8254,
      "num_input_tokens_seen": 1874578944,
      "step": 25260,
      "train_runtime": 30002.8404,
      "train_tokens_per_second": 62480.049
    },
    {
      "epoch": 0.33581172218124794,
      "grad_norm": 14.5625,
      "learning_rate": 1.5483677637936318e-05,
      "loss": 0.9726,
      "num_input_tokens_seen": 1875421952,
      "step": 25270,
      "train_runtime": 30015.6288,
      "train_tokens_per_second": 62481.515
    },
    {
      "epoch": 0.33594461166370987,
      "grad_norm": 12.625,
      "learning_rate": 1.5480077994737402e-05,
      "loss": 0.7418,
      "num_input_tokens_seen": 1875883008,
      "step": 25280,
      "train_runtime": 30022.2144,
      "train_tokens_per_second": 62483.166
    },
    {
      "epoch": 0.3360775011461718,
      "grad_norm": 29.125,
      "learning_rate": 1.5476477336402903e-05,
      "loss": 0.6221,
      "num_input_tokens_seen": 1876115712,
      "step": 25290,
      "train_runtime": 30025.7043,
      "train_tokens_per_second": 62483.654
    },
    {
      "epoch": 0.33621039062863367,
      "grad_norm": 14.625,
      "learning_rate": 1.5472875663599815e-05,
      "loss": 0.557,
      "num_input_tokens_seen": 1876234240,
      "step": 25300,
      "train_runtime": 30028.1477,
      "train_tokens_per_second": 62482.517
    },
    {
      "epoch": 0.3363432801110956,
      "grad_norm": 26.0,
      "learning_rate": 1.5469272976995307e-05,
      "loss": 0.919,
      "num_input_tokens_seen": 1878243072,
      "step": 25310,
      "train_runtime": 30060.4598,
      "train_tokens_per_second": 62482.18
    },
    {
      "epoch": 0.33647616959355753,
      "grad_norm": 20.875,
      "learning_rate": 1.546566927725675e-05,
      "loss": 0.9457,
      "num_input_tokens_seen": 1879147264,
      "step": 25320,
      "train_runtime": 30073.8758,
      "train_tokens_per_second": 62484.373
    },
    {
      "epoch": 0.3366090590760194,
      "grad_norm": 13.375,
      "learning_rate": 1.5462064565051692e-05,
      "loss": 0.7523,
      "num_input_tokens_seen": 1879625472,
      "step": 25330,
      "train_runtime": 30080.8445,
      "train_tokens_per_second": 62485.795
    },
    {
      "epoch": 0.33674194855848133,
      "grad_norm": 162.0,
      "learning_rate": 1.545845884104788e-05,
      "loss": 0.6327,
      "num_input_tokens_seen": 1879869696,
      "step": 25340,
      "train_runtime": 30084.4606,
      "train_tokens_per_second": 62486.402
    },
    {
      "epoch": 0.33687483804094326,
      "grad_norm": 10.75,
      "learning_rate": 1.545485210591324e-05,
      "loss": 0.5747,
      "num_input_tokens_seen": 1879988992,
      "step": 25350,
      "train_runtime": 30086.8824,
      "train_tokens_per_second": 62485.337
    },
    {
      "epoch": 0.3370077275234052,
      "grad_norm": 26.125,
      "learning_rate": 1.5451244360315885e-05,
      "loss": 0.8587,
      "num_input_tokens_seen": 1881995264,
      "step": 25360,
      "train_runtime": 30119.5891,
      "train_tokens_per_second": 62484.095
    },
    {
      "epoch": 0.33714061700586706,
      "grad_norm": 18.25,
      "learning_rate": 1.5447635604924127e-05,
      "loss": 0.939,
      "num_input_tokens_seen": 1882821120,
      "step": 25370,
      "train_runtime": 30132.3035,
      "train_tokens_per_second": 62485.137
    },
    {
      "epoch": 0.337273506488329,
      "grad_norm": 14.75,
      "learning_rate": 1.5444025840406444e-05,
      "loss": 0.8094,
      "num_input_tokens_seen": 1883251968,
      "step": 25380,
      "train_runtime": 30138.6533,
      "train_tokens_per_second": 62486.268
    },
    {
      "epoch": 0.3374063959707909,
      "grad_norm": 32.25,
      "learning_rate": 1.544041506743152e-05,
      "loss": 0.707,
      "num_input_tokens_seen": 1883467264,
      "step": 25390,
      "train_runtime": 30141.9417,
      "train_tokens_per_second": 62486.594
    },
    {
      "epoch": 0.3375392854532528,
      "grad_norm": 24.625,
      "learning_rate": 1.543680328666821e-05,
      "loss": 0.5687,
      "num_input_tokens_seen": 1883581952,
      "step": 25400,
      "train_runtime": 30144.3078,
      "train_tokens_per_second": 62485.494
    },
    {
      "epoch": 0.3376721749357147,
      "grad_norm": 30.5,
      "learning_rate": 1.543319049878558e-05,
      "loss": 0.8249,
      "num_input_tokens_seen": 1885575168,
      "step": 25410,
      "train_runtime": 30177.3255,
      "train_tokens_per_second": 62483.177
    },
    {
      "epoch": 0.33780506441817665,
      "grad_norm": 24.625,
      "learning_rate": 1.542957670445285e-05,
      "loss": 0.855,
      "num_input_tokens_seen": 1886393088,
      "step": 25420,
      "train_runtime": 30189.59,
      "train_tokens_per_second": 62484.886
    },
    {
      "epoch": 0.3379379539006385,
      "grad_norm": 21.5,
      "learning_rate": 1.5425961904339447e-05,
      "loss": 0.7304,
      "num_input_tokens_seen": 1886851584,
      "step": 25430,
      "train_runtime": 30196.1088,
      "train_tokens_per_second": 62486.58
    },
    {
      "epoch": 0.33807084338310045,
      "grad_norm": 11.8125,
      "learning_rate": 1.542234609911499e-05,
      "loss": 0.5971,
      "num_input_tokens_seen": 1887079936,
      "step": 25440,
      "train_runtime": 30199.5779,
      "train_tokens_per_second": 62486.964
    },
    {
      "epoch": 0.3382037328655624,
      "grad_norm": 15.5,
      "learning_rate": 1.5418729289449266e-05,
      "loss": 0.4857,
      "num_input_tokens_seen": 1887195392,
      "step": 25450,
      "train_runtime": 30202.5657,
      "train_tokens_per_second": 62484.605
    },
    {
      "epoch": 0.33833662234802425,
      "grad_norm": 19.625,
      "learning_rate": 1.541511147601226e-05,
      "loss": 0.9373,
      "num_input_tokens_seen": 1889207808,
      "step": 25460,
      "train_runtime": 30235.3397,
      "train_tokens_per_second": 62483.433
    },
    {
      "epoch": 0.3384695118304862,
      "grad_norm": 50.5,
      "learning_rate": 1.5411492659474138e-05,
      "loss": 0.9252,
      "num_input_tokens_seen": 1890080768,
      "step": 25470,
      "train_runtime": 30248.2884,
      "train_tokens_per_second": 62485.544
    },
    {
      "epoch": 0.3386024013129481,
      "grad_norm": 13.25,
      "learning_rate": 1.5407872840505256e-05,
      "loss": 0.7584,
      "num_input_tokens_seen": 1890549248,
      "step": 25480,
      "train_runtime": 30255.2387,
      "train_tokens_per_second": 62486.674
    },
    {
      "epoch": 0.33873529079541,
      "grad_norm": 18.0,
      "learning_rate": 1.540425201977615e-05,
      "loss": 0.6401,
      "num_input_tokens_seen": 1890778624,
      "step": 25490,
      "train_runtime": 30258.7,
      "train_tokens_per_second": 62487.107
    },
    {
      "epoch": 0.3388681802778719,
      "grad_norm": 11.625,
      "learning_rate": 1.5400630197957544e-05,
      "loss": 0.5832,
      "num_input_tokens_seen": 1890893824,
      "step": 25500,
      "train_runtime": 30261.6494,
      "train_tokens_per_second": 62484.824
    },
    {
      "epoch": 0.33900106976033384,
      "grad_norm": 17.125,
      "learning_rate": 1.5397007375720354e-05,
      "loss": 0.8146,
      "num_input_tokens_seen": 1892827904,
      "step": 25510,
      "train_runtime": 30292.9598,
      "train_tokens_per_second": 62484.086
    },
    {
      "epoch": 0.3391339592427957,
      "grad_norm": 32.5,
      "learning_rate": 1.539338355373567e-05,
      "loss": 0.9042,
      "num_input_tokens_seen": 1893633536,
      "step": 25520,
      "train_runtime": 30305.2437,
      "train_tokens_per_second": 62485.343
    },
    {
      "epoch": 0.33926684872525764,
      "grad_norm": 19.5,
      "learning_rate": 1.5389758732674773e-05,
      "loss": 0.7042,
      "num_input_tokens_seen": 1894103040,
      "step": 25530,
      "train_runtime": 30311.7447,
      "train_tokens_per_second": 62487.43
    },
    {
      "epoch": 0.33939973820771957,
      "grad_norm": 11.5,
      "learning_rate": 1.5386132913209137e-05,
      "loss": 0.5943,
      "num_input_tokens_seen": 1894326016,
      "step": 25540,
      "train_runtime": 30315.0853,
      "train_tokens_per_second": 62487.9
    },
    {
      "epoch": 0.33953262769018144,
      "grad_norm": 18.0,
      "learning_rate": 1.53825060960104e-05,
      "loss": 0.4756,
      "num_input_tokens_seen": 1894436352,
      "step": 25550,
      "train_runtime": 30317.4646,
      "train_tokens_per_second": 62486.635
    },
    {
      "epoch": 0.33966551717264337,
      "grad_norm": 29.625,
      "learning_rate": 1.53788782817504e-05,
      "loss": 0.8703,
      "num_input_tokens_seen": 1896485888,
      "step": 25560,
      "train_runtime": 30349.4796,
      "train_tokens_per_second": 62488.251
    },
    {
      "epoch": 0.3397984066551053,
      "grad_norm": 12.3125,
      "learning_rate": 1.537524947110117e-05,
      "loss": 0.9065,
      "num_input_tokens_seen": 1897346048,
      "step": 25570,
      "train_runtime": 30362.5864,
      "train_tokens_per_second": 62489.606
    },
    {
      "epoch": 0.3399312961375672,
      "grad_norm": 17.625,
      "learning_rate": 1.53716196647349e-05,
      "loss": 0.8049,
      "num_input_tokens_seen": 1897826560,
      "step": 25580,
      "train_runtime": 30369.2468,
      "train_tokens_per_second": 62491.723
    },
    {
      "epoch": 0.3400641856200291,
      "grad_norm": 17.25,
      "learning_rate": 1.536798886332399e-05,
      "loss": 0.6718,
      "num_input_tokens_seen": 1898059520,
      "step": 25590,
      "train_runtime": 30372.7237,
      "train_tokens_per_second": 62492.239
    },
    {
      "epoch": 0.34019707510249103,
      "grad_norm": 8.75,
      "learning_rate": 1.5364357067541e-05,
      "loss": 0.5588,
      "num_input_tokens_seen": 1898176256,
      "step": 25600,
      "train_runtime": 30375.1282,
      "train_tokens_per_second": 62491.136
    },
    {
      "epoch": 0.3403299645849529,
      "grad_norm": 23.625,
      "learning_rate": 1.5360724278058703e-05,
      "loss": 0.7965,
      "num_input_tokens_seen": 1900224768,
      "step": 25610,
      "train_runtime": 30408.128,
      "train_tokens_per_second": 62490.686
    },
    {
      "epoch": 0.34046285406741483,
      "grad_norm": 16.875,
      "learning_rate": 1.5357090495550034e-05,
      "loss": 0.9288,
      "num_input_tokens_seen": 1901053440,
      "step": 25620,
      "train_runtime": 30420.6958,
      "train_tokens_per_second": 62492.109
    },
    {
      "epoch": 0.34059574354987676,
      "grad_norm": 9.9375,
      "learning_rate": 1.535345572068812e-05,
      "loss": 0.6836,
      "num_input_tokens_seen": 1901518592,
      "step": 25630,
      "train_runtime": 30427.3116,
      "train_tokens_per_second": 62493.809
    },
    {
      "epoch": 0.34072863303233863,
      "grad_norm": 17.5,
      "learning_rate": 1.5349819954146266e-05,
      "loss": 0.6405,
      "num_input_tokens_seen": 1901749504,
      "step": 25640,
      "train_runtime": 30431.332,
      "train_tokens_per_second": 62493.14
    },
    {
      "epoch": 0.34086152251480056,
      "grad_norm": 7.21875,
      "learning_rate": 1.534618319659797e-05,
      "loss": 0.4983,
      "num_input_tokens_seen": 1901864704,
      "step": 25650,
      "train_runtime": 30433.696,
      "train_tokens_per_second": 62492.071
    },
    {
      "epoch": 0.3409944119972625,
      "grad_norm": 17.5,
      "learning_rate": 1.5342545448716904e-05,
      "loss": 0.858,
      "num_input_tokens_seen": 1904019456,
      "step": 25660,
      "train_runtime": 30468.3098,
      "train_tokens_per_second": 62491.798
    },
    {
      "epoch": 0.34112730147972437,
      "grad_norm": 24.75,
      "learning_rate": 1.533890671117694e-05,
      "loss": 0.8938,
      "num_input_tokens_seen": 1904849920,
      "step": 25670,
      "train_runtime": 30480.9168,
      "train_tokens_per_second": 62493.196
    },
    {
      "epoch": 0.3412601909621863,
      "grad_norm": 19.625,
      "learning_rate": 1.5335266984652107e-05,
      "loss": 0.7238,
      "num_input_tokens_seen": 1905305856,
      "step": 25680,
      "train_runtime": 30487.5745,
      "train_tokens_per_second": 62494.504
    },
    {
      "epoch": 0.3413930804446482,
      "grad_norm": 36.25,
      "learning_rate": 1.533162626981664e-05,
      "loss": 0.6723,
      "num_input_tokens_seen": 1905545216,
      "step": 25690,
      "train_runtime": 30491.5287,
      "train_tokens_per_second": 62494.25
    },
    {
      "epoch": 0.3415259699271101,
      "grad_norm": 8.0625,
      "learning_rate": 1.5327984567344954e-05,
      "loss": 0.5405,
      "num_input_tokens_seen": 1905661184,
      "step": 25700,
      "train_runtime": 30493.9145,
      "train_tokens_per_second": 62493.163
    },
    {
      "epoch": 0.341658859409572,
      "grad_norm": 41.0,
      "learning_rate": 1.5324341877911632e-05,
      "loss": 0.8281,
      "num_input_tokens_seen": 1907778560,
      "step": 25710,
      "train_runtime": 30527.9518,
      "train_tokens_per_second": 62492.845
    },
    {
      "epoch": 0.34179174889203395,
      "grad_norm": 41.75,
      "learning_rate": 1.5320698202191452e-05,
      "loss": 0.9137,
      "num_input_tokens_seen": 1908652800,
      "step": 25720,
      "train_runtime": 30540.8489,
      "train_tokens_per_second": 62495.08
    },
    {
      "epoch": 0.3419246383744958,
      "grad_norm": 13.1875,
      "learning_rate": 1.531705354085937e-05,
      "loss": 0.7672,
      "num_input_tokens_seen": 1909140736,
      "step": 25730,
      "train_runtime": 30548.1728,
      "train_tokens_per_second": 62496.07
    },
    {
      "epoch": 0.34205752785695775,
      "grad_norm": 20.0,
      "learning_rate": 1.531340789459054e-05,
      "loss": 0.6847,
      "num_input_tokens_seen": 1909381376,
      "step": 25740,
      "train_runtime": 30551.7746,
      "train_tokens_per_second": 62496.578
    },
    {
      "epoch": 0.3421904173394197,
      "grad_norm": 22.125,
      "learning_rate": 1.530976126406027e-05,
      "loss": 0.5553,
      "num_input_tokens_seen": 1909503488,
      "step": 25750,
      "train_runtime": 30554.2019,
      "train_tokens_per_second": 62495.61
    },
    {
      "epoch": 0.34232330682188156,
      "grad_norm": 18.75,
      "learning_rate": 1.5306113649944072e-05,
      "loss": 0.8397,
      "num_input_tokens_seen": 1911473664,
      "step": 25760,
      "train_runtime": 30586.1939,
      "train_tokens_per_second": 62494.656
    },
    {
      "epoch": 0.3424561963043435,
      "grad_norm": 21.875,
      "learning_rate": 1.5302465052917637e-05,
      "loss": 0.9245,
      "num_input_tokens_seen": 1912291584,
      "step": 25770,
      "train_runtime": 30598.8513,
      "train_tokens_per_second": 62495.535
    },
    {
      "epoch": 0.3425890857868054,
      "grad_norm": 31.25,
      "learning_rate": 1.5298815473656834e-05,
      "loss": 0.7284,
      "num_input_tokens_seen": 1912756736,
      "step": 25780,
      "train_runtime": 30605.5091,
      "train_tokens_per_second": 62497.138
    },
    {
      "epoch": 0.3427219752692673,
      "grad_norm": 14.4375,
      "learning_rate": 1.5295164912837707e-05,
      "loss": 0.6643,
      "num_input_tokens_seen": 1912989440,
      "step": 25790,
      "train_runtime": 30609.5536,
      "train_tokens_per_second": 62496.483
    },
    {
      "epoch": 0.3428548647517292,
      "grad_norm": 28.625,
      "learning_rate": 1.52915133711365e-05,
      "loss": 0.5769,
      "num_input_tokens_seen": 1913105920,
      "step": 25800,
      "train_runtime": 30612.044,
      "train_tokens_per_second": 62495.204
    },
    {
      "epoch": 0.34298775423419114,
      "grad_norm": 29.875,
      "learning_rate": 1.5287860849229623e-05,
      "loss": 0.8317,
      "num_input_tokens_seen": 1915124480,
      "step": 25810,
      "train_runtime": 30644.3223,
      "train_tokens_per_second": 62495.247
    },
    {
      "epoch": 0.3431206437166531,
      "grad_norm": 22.25,
      "learning_rate": 1.5284207347793677e-05,
      "loss": 0.94,
      "num_input_tokens_seen": 1915948800,
      "step": 25820,
      "train_runtime": 30656.9477,
      "train_tokens_per_second": 62496.398
    },
    {
      "epoch": 0.34325353319911495,
      "grad_norm": 27.375,
      "learning_rate": 1.5280552867505435e-05,
      "loss": 0.742,
      "num_input_tokens_seen": 1916439296,
      "step": 25830,
      "train_runtime": 30663.7952,
      "train_tokens_per_second": 62498.438
    },
    {
      "epoch": 0.3433864226815769,
      "grad_norm": 17.125,
      "learning_rate": 1.527689740904186e-05,
      "loss": 0.6401,
      "num_input_tokens_seen": 1916667648,
      "step": 25840,
      "train_runtime": 30667.7092,
      "train_tokens_per_second": 62497.907
    },
    {
      "epoch": 0.3435193121640388,
      "grad_norm": 20.25,
      "learning_rate": 1.5273240973080097e-05,
      "loss": 0.507,
      "num_input_tokens_seen": 1916777472,
      "step": 25850,
      "train_runtime": 30670.0385,
      "train_tokens_per_second": 62496.742
    },
    {
      "epoch": 0.3436522016465007,
      "grad_norm": 27.125,
      "learning_rate": 1.5269583560297462e-05,
      "loss": 0.774,
      "num_input_tokens_seen": 1918848000,
      "step": 25860,
      "train_runtime": 30703.096,
      "train_tokens_per_second": 62496.89
    },
    {
      "epoch": 0.3437850911289626,
      "grad_norm": 16.25,
      "learning_rate": 1.526592517137146e-05,
      "loss": 0.8742,
      "num_input_tokens_seen": 1919714048,
      "step": 25870,
      "train_runtime": 30715.7222,
      "train_tokens_per_second": 62499.395
    },
    {
      "epoch": 0.34391798061142453,
      "grad_norm": 28.75,
      "learning_rate": 1.5262265806979777e-05,
      "loss": 0.8196,
      "num_input_tokens_seen": 1920186624,
      "step": 25880,
      "train_runtime": 30722.4263,
      "train_tokens_per_second": 62501.139
    },
    {
      "epoch": 0.3440508700938864,
      "grad_norm": 12.5625,
      "learning_rate": 1.5258605467800278e-05,
      "loss": 0.6016,
      "num_input_tokens_seen": 1920410112,
      "step": 25890,
      "train_runtime": 30725.8541,
      "train_tokens_per_second": 62501.44
    },
    {
      "epoch": 0.34418375957634834,
      "grad_norm": 14.5,
      "learning_rate": 1.5254944154511002e-05,
      "loss": 0.5084,
      "num_input_tokens_seen": 1920526848,
      "step": 25900,
      "train_runtime": 30728.2349,
      "train_tokens_per_second": 62500.396
    },
    {
      "epoch": 0.34431664905881026,
      "grad_norm": 15.8125,
      "learning_rate": 1.525128186779018e-05,
      "loss": 0.8526,
      "num_input_tokens_seen": 1922510336,
      "step": 25910,
      "train_runtime": 30760.7017,
      "train_tokens_per_second": 62498.91
    },
    {
      "epoch": 0.34444953854127214,
      "grad_norm": 18.875,
      "learning_rate": 1.5247618608316215e-05,
      "loss": 0.9443,
      "num_input_tokens_seen": 1923375104,
      "step": 25920,
      "train_runtime": 30774.1037,
      "train_tokens_per_second": 62499.793
    },
    {
      "epoch": 0.34458242802373407,
      "grad_norm": 22.625,
      "learning_rate": 1.52439543767677e-05,
      "loss": 0.7656,
      "num_input_tokens_seen": 1923830272,
      "step": 25930,
      "train_runtime": 30780.7466,
      "train_tokens_per_second": 62501.092
    },
    {
      "epoch": 0.344715317506196,
      "grad_norm": 14.4375,
      "learning_rate": 1.5240289173823396e-05,
      "loss": 0.6895,
      "num_input_tokens_seen": 1924061952,
      "step": 25940,
      "train_runtime": 30784.1866,
      "train_tokens_per_second": 62501.634
    },
    {
      "epoch": 0.34484820698865787,
      "grad_norm": 16.375,
      "learning_rate": 1.5236623000162243e-05,
      "loss": 0.5118,
      "num_input_tokens_seen": 1924177920,
      "step": 25950,
      "train_runtime": 30786.579,
      "train_tokens_per_second": 62500.544
    },
    {
      "epoch": 0.3449810964711198,
      "grad_norm": 1768.0,
      "learning_rate": 1.5232955856463378e-05,
      "loss": 0.8374,
      "num_input_tokens_seen": 1926253056,
      "step": 25960,
      "train_runtime": 30818.8744,
      "train_tokens_per_second": 62502.382
    },
    {
      "epoch": 0.3451139859535817,
      "grad_norm": 17.125,
      "learning_rate": 1.5229287743406097e-05,
      "loss": 0.8814,
      "num_input_tokens_seen": 1927086336,
      "step": 25970,
      "train_runtime": 30831.8395,
      "train_tokens_per_second": 62503.125
    },
    {
      "epoch": 0.3452468754360436,
      "grad_norm": 18.125,
      "learning_rate": 1.5225618661669897e-05,
      "loss": 0.7462,
      "num_input_tokens_seen": 1927581696,
      "step": 25980,
      "train_runtime": 30838.8783,
      "train_tokens_per_second": 62504.922
    },
    {
      "epoch": 0.3453797649185055,
      "grad_norm": 55.75,
      "learning_rate": 1.5221948611934433e-05,
      "loss": 0.63,
      "num_input_tokens_seen": 1927826944,
      "step": 25990,
      "train_runtime": 30842.4798,
      "train_tokens_per_second": 62505.575
    },
    {
      "epoch": 0.34551265440096746,
      "grad_norm": 11.6875,
      "learning_rate": 1.5218277594879552e-05,
      "loss": 0.5768,
      "num_input_tokens_seen": 1927942656,
      "step": 26000,
      "train_runtime": 30844.9006,
      "train_tokens_per_second": 62504.421
    },
    {
      "epoch": 0.34564554388342933,
      "grad_norm": 21.875,
      "learning_rate": 1.5214605611185275e-05,
      "loss": 0.8206,
      "num_input_tokens_seen": 1930063616,
      "step": 26010,
      "train_runtime": 30879.1218,
      "train_tokens_per_second": 62503.838
    },
    {
      "epoch": 0.34577843336589126,
      "grad_norm": 19.25,
      "learning_rate": 1.521093266153181e-05,
      "loss": 0.8508,
      "num_input_tokens_seen": 1930910720,
      "step": 26020,
      "train_runtime": 30891.9538,
      "train_tokens_per_second": 62505.296
    },
    {
      "epoch": 0.3459113228483532,
      "grad_norm": 25.5,
      "learning_rate": 1.5207258746599533e-05,
      "loss": 0.7503,
      "num_input_tokens_seen": 1931365888,
      "step": 26030,
      "train_runtime": 30898.5359,
      "train_tokens_per_second": 62506.712
    },
    {
      "epoch": 0.34604421233081506,
      "grad_norm": 29.5,
      "learning_rate": 1.5203583867069007e-05,
      "loss": 0.6383,
      "num_input_tokens_seen": 1931590656,
      "step": 26040,
      "train_runtime": 30902.0115,
      "train_tokens_per_second": 62506.955
    },
    {
      "epoch": 0.346177101813277,
      "grad_norm": 14.625,
      "learning_rate": 1.5199908023620966e-05,
      "loss": 0.566,
      "num_input_tokens_seen": 1931706880,
      "step": 26050,
      "train_runtime": 30904.397,
      "train_tokens_per_second": 62505.891
    },
    {
      "epoch": 0.3463099912957389,
      "grad_norm": 35.25,
      "learning_rate": 1.5196231216936332e-05,
      "loss": 0.7837,
      "num_input_tokens_seen": 1933748992,
      "step": 26060,
      "train_runtime": 30938.1306,
      "train_tokens_per_second": 62503.744
    },
    {
      "epoch": 0.3464428807782008,
      "grad_norm": 11.125,
      "learning_rate": 1.5192553447696199e-05,
      "loss": 0.9014,
      "num_input_tokens_seen": 1934586624,
      "step": 26070,
      "train_runtime": 30951.1794,
      "train_tokens_per_second": 62504.456
    },
    {
      "epoch": 0.3465757702606627,
      "grad_norm": 70.5,
      "learning_rate": 1.5188874716581839e-05,
      "loss": 0.7425,
      "num_input_tokens_seen": 1935060480,
      "step": 26080,
      "train_runtime": 30957.7472,
      "train_tokens_per_second": 62506.502
    },
    {
      "epoch": 0.34670865974312465,
      "grad_norm": 20.25,
      "learning_rate": 1.5185195024274709e-05,
      "loss": 0.5911,
      "num_input_tokens_seen": 1935290880,
      "step": 26090,
      "train_runtime": 30961.1745,
      "train_tokens_per_second": 62507.024
    },
    {
      "epoch": 0.3468415492255865,
      "grad_norm": 37.25,
      "learning_rate": 1.5181514371456436e-05,
      "loss": 0.5254,
      "num_input_tokens_seen": 1935406848,
      "step": 26100,
      "train_runtime": 30963.5997,
      "train_tokens_per_second": 62505.874
    },
    {
      "epoch": 0.34697443870804845,
      "grad_norm": 17.125,
      "learning_rate": 1.5177832758808828e-05,
      "loss": 0.9172,
      "num_input_tokens_seen": 1937387008,
      "step": 26110,
      "train_runtime": 30996.5193,
      "train_tokens_per_second": 62503.373
    },
    {
      "epoch": 0.3471073281905104,
      "grad_norm": 9.3125,
      "learning_rate": 1.517415018701387e-05,
      "loss": 0.8683,
      "num_input_tokens_seen": 1938251008,
      "step": 26120,
      "train_runtime": 31009.4635,
      "train_tokens_per_second": 62505.145
    },
    {
      "epoch": 0.34724021767297225,
      "grad_norm": 31.875,
      "learning_rate": 1.5170466656753726e-05,
      "loss": 0.722,
      "num_input_tokens_seen": 1938692608,
      "step": 26130,
      "train_runtime": 31016.0069,
      "train_tokens_per_second": 62506.196
    },
    {
      "epoch": 0.3473731071554342,
      "grad_norm": 34.5,
      "learning_rate": 1.516678216871074e-05,
      "loss": 0.7245,
      "num_input_tokens_seen": 1938924800,
      "step": 26140,
      "train_runtime": 31019.4927,
      "train_tokens_per_second": 62506.657
    },
    {
      "epoch": 0.3475059966378961,
      "grad_norm": 5.65625,
      "learning_rate": 1.516309672356743e-05,
      "loss": 0.5521,
      "num_input_tokens_seen": 1939040000,
      "step": 26150,
      "train_runtime": 31021.8724,
      "train_tokens_per_second": 62505.576
    },
    {
      "epoch": 0.347638886120358,
      "grad_norm": 14.4375,
      "learning_rate": 1.5159410322006487e-05,
      "loss": 0.8545,
      "num_input_tokens_seen": 1941111552,
      "step": 26160,
      "train_runtime": 31054.7175,
      "train_tokens_per_second": 62506.173
    },
    {
      "epoch": 0.3477717756028199,
      "grad_norm": 26.375,
      "learning_rate": 1.5155722964710789e-05,
      "loss": 0.9963,
      "num_input_tokens_seen": 1942012672,
      "step": 26170,
      "train_runtime": 31067.726,
      "train_tokens_per_second": 62509.006
    },
    {
      "epoch": 0.34790466508528184,
      "grad_norm": 27.75,
      "learning_rate": 1.5152034652363385e-05,
      "loss": 0.7727,
      "num_input_tokens_seen": 1942510080,
      "step": 26180,
      "train_runtime": 31074.7732,
      "train_tokens_per_second": 62510.837
    },
    {
      "epoch": 0.3480375545677437,
      "grad_norm": 11.0,
      "learning_rate": 1.5148345385647503e-05,
      "loss": 0.67,
      "num_input_tokens_seen": 1942742784,
      "step": 26190,
      "train_runtime": 31078.2928,
      "train_tokens_per_second": 62511.245
    },
    {
      "epoch": 0.34817044405020564,
      "grad_norm": 24.375,
      "learning_rate": 1.5144655165246542e-05,
      "loss": 0.5013,
      "num_input_tokens_seen": 1942857472,
      "step": 26200,
      "train_runtime": 31081.2613,
      "train_tokens_per_second": 62508.965
    },
    {
      "epoch": 0.34830333353266757,
      "grad_norm": 51.0,
      "learning_rate": 1.5140963991844087e-05,
      "loss": 0.8152,
      "num_input_tokens_seen": 1944790272,
      "step": 26210,
      "train_runtime": 31112.8347,
      "train_tokens_per_second": 62507.653
    },
    {
      "epoch": 0.34843622301512944,
      "grad_norm": 10.0625,
      "learning_rate": 1.5137271866123895e-05,
      "loss": 0.9211,
      "num_input_tokens_seen": 1945633024,
      "step": 26220,
      "train_runtime": 31125.4105,
      "train_tokens_per_second": 62509.474
    },
    {
      "epoch": 0.34856911249759137,
      "grad_norm": 16.375,
      "learning_rate": 1.5133578788769896e-05,
      "loss": 0.7501,
      "num_input_tokens_seen": 1946095872,
      "step": 26230,
      "train_runtime": 31131.9378,
      "train_tokens_per_second": 62511.235
    },
    {
      "epoch": 0.3487020019800533,
      "grad_norm": 12.375,
      "learning_rate": 1.51298847604662e-05,
      "loss": 0.7218,
      "num_input_tokens_seen": 1946324992,
      "step": 26240,
      "train_runtime": 31135.4638,
      "train_tokens_per_second": 62511.514
    },
    {
      "epoch": 0.3488348914625152,
      "grad_norm": 27.875,
      "learning_rate": 1.51261897818971e-05,
      "loss": 0.5689,
      "num_input_tokens_seen": 1946441216,
      "step": 26250,
      "train_runtime": 31137.8552,
      "train_tokens_per_second": 62510.446
    },
    {
      "epoch": 0.3489677809449771,
      "grad_norm": 12.5625,
      "learning_rate": 1.512249385374705e-05,
      "loss": 0.7813,
      "num_input_tokens_seen": 1948608768,
      "step": 26260,
      "train_runtime": 31172.3646,
      "train_tokens_per_second": 62510.778
    },
    {
      "epoch": 0.34910067042743903,
      "grad_norm": 19.0,
      "learning_rate": 1.5118796976700689e-05,
      "loss": 0.8891,
      "num_input_tokens_seen": 1949468416,
      "step": 26270,
      "train_runtime": 31185.4256,
      "train_tokens_per_second": 62512.163
    },
    {
      "epoch": 0.3492335599099009,
      "grad_norm": 15.0625,
      "learning_rate": 1.5115099151442831e-05,
      "loss": 0.7426,
      "num_input_tokens_seen": 1949948928,
      "step": 26280,
      "train_runtime": 31192.3129,
      "train_tokens_per_second": 62513.765
    },
    {
      "epoch": 0.34936644939236283,
      "grad_norm": 10.8125,
      "learning_rate": 1.5111400378658467e-05,
      "loss": 0.6383,
      "num_input_tokens_seen": 1950181376,
      "step": 26290,
      "train_runtime": 31195.8154,
      "train_tokens_per_second": 62514.198
    },
    {
      "epoch": 0.34949933887482476,
      "grad_norm": 14.375,
      "learning_rate": 1.5107700659032757e-05,
      "loss": 0.4997,
      "num_input_tokens_seen": 1950299648,
      "step": 26300,
      "train_runtime": 31198.2195,
      "train_tokens_per_second": 62513.172
    },
    {
      "epoch": 0.3496322283572867,
      "grad_norm": 24.75,
      "learning_rate": 1.5103999993251051e-05,
      "loss": 0.823,
      "num_input_tokens_seen": 1952449536,
      "step": 26310,
      "train_runtime": 31232.7664,
      "train_tokens_per_second": 62512.859
    },
    {
      "epoch": 0.34976511783974856,
      "grad_norm": 32.25,
      "learning_rate": 1.5100298381998852e-05,
      "loss": 0.9191,
      "num_input_tokens_seen": 1953275648,
      "step": 26320,
      "train_runtime": 31245.2257,
      "train_tokens_per_second": 62514.372
    },
    {
      "epoch": 0.3498980073222105,
      "grad_norm": 24.5,
      "learning_rate": 1.5096595825961858e-05,
      "loss": 0.8262,
      "num_input_tokens_seen": 1953711616,
      "step": 26330,
      "train_runtime": 31251.5768,
      "train_tokens_per_second": 62515.617
    },
    {
      "epoch": 0.3500308968046724,
      "grad_norm": 41.25,
      "learning_rate": 1.5092892325825933e-05,
      "loss": 0.6775,
      "num_input_tokens_seen": 1953926656,
      "step": 26340,
      "train_runtime": 31254.8947,
      "train_tokens_per_second": 62515.861
    },
    {
      "epoch": 0.3501637862871343,
      "grad_norm": 12.25,
      "learning_rate": 1.5089187882277116e-05,
      "loss": 0.4804,
      "num_input_tokens_seen": 1954040832,
      "step": 26350,
      "train_runtime": 31258.1265,
      "train_tokens_per_second": 62513.05
    },
    {
      "epoch": 0.3502966757695962,
      "grad_norm": 21.5,
      "learning_rate": 1.5085482496001627e-05,
      "loss": 0.857,
      "num_input_tokens_seen": 1956171264,
      "step": 26360,
      "train_runtime": 31292.0066,
      "train_tokens_per_second": 62513.449
    },
    {
      "epoch": 0.35042956525205815,
      "grad_norm": 13.9375,
      "learning_rate": 1.5081776167685852e-05,
      "loss": 0.9191,
      "num_input_tokens_seen": 1957026304,
      "step": 26370,
      "train_runtime": 31304.9048,
      "train_tokens_per_second": 62515.006
    },
    {
      "epoch": 0.35056245473452,
      "grad_norm": 14.8125,
      "learning_rate": 1.5078068898016351e-05,
      "loss": 0.7191,
      "num_input_tokens_seen": 1957492736,
      "step": 26380,
      "train_runtime": 31311.5217,
      "train_tokens_per_second": 62516.691
    },
    {
      "epoch": 0.35069534421698195,
      "grad_norm": 20.5,
      "learning_rate": 1.5074360687679868e-05,
      "loss": 0.6066,
      "num_input_tokens_seen": 1957718528,
      "step": 26390,
      "train_runtime": 31314.9729,
      "train_tokens_per_second": 62517.012
    },
    {
      "epoch": 0.3508282336994439,
      "grad_norm": 13.125,
      "learning_rate": 1.5070651537363315e-05,
      "loss": 0.4724,
      "num_input_tokens_seen": 1957832960,
      "step": 26400,
      "train_runtime": 31317.3681,
      "train_tokens_per_second": 62515.884
    },
    {
      "epoch": 0.35096112318190575,
      "grad_norm": 44.5,
      "learning_rate": 1.506694144775378e-05,
      "loss": 0.8959,
      "num_input_tokens_seen": 1959853056,
      "step": 26410,
      "train_runtime": 31349.4651,
      "train_tokens_per_second": 62516.316
    },
    {
      "epoch": 0.3510940126643677,
      "grad_norm": 23.0,
      "learning_rate": 1.5063230419538524e-05,
      "loss": 0.9176,
      "num_input_tokens_seen": 1960706816,
      "step": 26420,
      "train_runtime": 31362.85,
      "train_tokens_per_second": 62516.857
    },
    {
      "epoch": 0.3512269021468296,
      "grad_norm": 20.375,
      "learning_rate": 1.5059518453404978e-05,
      "loss": 0.7938,
      "num_input_tokens_seen": 1961159168,
      "step": 26430,
      "train_runtime": 31369.4816,
      "train_tokens_per_second": 62518.061
    },
    {
      "epoch": 0.3513597916292915,
      "grad_norm": 14.1875,
      "learning_rate": 1.5055805550040754e-05,
      "loss": 0.5962,
      "num_input_tokens_seen": 1961399040,
      "step": 26440,
      "train_runtime": 31373.0189,
      "train_tokens_per_second": 62518.658
    },
    {
      "epoch": 0.3514926811117534,
      "grad_norm": 18.25,
      "learning_rate": 1.5052091710133634e-05,
      "loss": 0.5434,
      "num_input_tokens_seen": 1961519104,
      "step": 26450,
      "train_runtime": 31375.4275,
      "train_tokens_per_second": 62517.685
    },
    {
      "epoch": 0.35162557059421534,
      "grad_norm": 26.0,
      "learning_rate": 1.5048376934371574e-05,
      "loss": 0.8682,
      "num_input_tokens_seen": 1963441664,
      "step": 26460,
      "train_runtime": 31406.8803,
      "train_tokens_per_second": 62516.291
    },
    {
      "epoch": 0.3517584600766772,
      "grad_norm": 24.5,
      "learning_rate": 1.5044661223442702e-05,
      "loss": 0.902,
      "num_input_tokens_seen": 1964305408,
      "step": 26470,
      "train_runtime": 31419.8163,
      "train_tokens_per_second": 62518.042
    },
    {
      "epoch": 0.35189134955913914,
      "grad_norm": 13.3125,
      "learning_rate": 1.504094457803532e-05,
      "loss": 0.7757,
      "num_input_tokens_seen": 1964783360,
      "step": 26480,
      "train_runtime": 31426.5907,
      "train_tokens_per_second": 62519.774
    },
    {
      "epoch": 0.3520242390416011,
      "grad_norm": 31.0,
      "learning_rate": 1.5037226998837904e-05,
      "loss": 0.6868,
      "num_input_tokens_seen": 1965017088,
      "step": 26490,
      "train_runtime": 31430.101,
      "train_tokens_per_second": 62520.228
    },
    {
      "epoch": 0.35215712852406295,
      "grad_norm": 7.40625,
      "learning_rate": 1.5033508486539106e-05,
      "loss": 0.5076,
      "num_input_tokens_seen": 1965132544,
      "step": 26500,
      "train_runtime": 31432.4939,
      "train_tokens_per_second": 62519.142
    },
    {
      "epoch": 0.3522900180065249,
      "grad_norm": 58.75,
      "learning_rate": 1.5029789041827738e-05,
      "loss": 0.7693,
      "num_input_tokens_seen": 1967213568,
      "step": 26510,
      "train_runtime": 31468.044,
      "train_tokens_per_second": 62514.644
    },
    {
      "epoch": 0.3524229074889868,
      "grad_norm": 42.75,
      "learning_rate": 1.5026068665392806e-05,
      "loss": 0.9291,
      "num_input_tokens_seen": 1968097280,
      "step": 26520,
      "train_runtime": 31481.095,
      "train_tokens_per_second": 62516.799
    },
    {
      "epoch": 0.3525557969714487,
      "grad_norm": 18.25,
      "learning_rate": 1.5022347357923467e-05,
      "loss": 0.8094,
      "num_input_tokens_seen": 1968590080,
      "step": 26530,
      "train_runtime": 31488.1548,
      "train_tokens_per_second": 62518.432
    },
    {
      "epoch": 0.3526886864539106,
      "grad_norm": 31.0,
      "learning_rate": 1.5018625120109062e-05,
      "loss": 0.7009,
      "num_input_tokens_seen": 1968829952,
      "step": 26540,
      "train_runtime": 31491.7307,
      "train_tokens_per_second": 62518.951
    },
    {
      "epoch": 0.35282157593637253,
      "grad_norm": 24.875,
      "learning_rate": 1.5014901952639104e-05,
      "loss": 0.5361,
      "num_input_tokens_seen": 1968948992,
      "step": 26550,
      "train_runtime": 31494.169,
      "train_tokens_per_second": 62517.89
    },
    {
      "epoch": 0.3529544654188344,
      "grad_norm": 27.375,
      "learning_rate": 1.5011177856203278e-05,
      "loss": 0.8783,
      "num_input_tokens_seen": 1970966528,
      "step": 26560,
      "train_runtime": 31527.6998,
      "train_tokens_per_second": 62515.393
    },
    {
      "epoch": 0.35308735490129634,
      "grad_norm": 15.4375,
      "learning_rate": 1.500745283149144e-05,
      "loss": 0.9085,
      "num_input_tokens_seen": 1971857664,
      "step": 26570,
      "train_runtime": 31540.9442,
      "train_tokens_per_second": 62517.395
    },
    {
      "epoch": 0.35322024438375826,
      "grad_norm": 30.25,
      "learning_rate": 1.5003726879193617e-05,
      "loss": 0.7913,
      "num_input_tokens_seen": 1972367360,
      "step": 26580,
      "train_runtime": 31548.0917,
      "train_tokens_per_second": 62519.387
    },
    {
      "epoch": 0.35335313386622014,
      "grad_norm": 10.0,
      "learning_rate": 1.5000000000000002e-05,
      "loss": 0.621,
      "num_input_tokens_seen": 1972608000,
      "step": 26590,
      "train_runtime": 31552.2688,
      "train_tokens_per_second": 62518.737
    },
    {
      "epoch": 0.35348602334868207,
      "grad_norm": 10.8125,
      "learning_rate": 1.4996272194600972e-05,
      "loss": 0.4931,
      "num_input_tokens_seen": 1972727296,
      "step": 26600,
      "train_runtime": 31554.718,
      "train_tokens_per_second": 62517.665
    },
    {
      "epoch": 0.353618912831144,
      "grad_norm": 53.75,
      "learning_rate": 1.4992543463687069e-05,
      "loss": 0.8131,
      "num_input_tokens_seen": 1974834688,
      "step": 26610,
      "train_runtime": 31588.8846,
      "train_tokens_per_second": 62516.759
    },
    {
      "epoch": 0.35375180231360587,
      "grad_norm": 24.25,
      "learning_rate": 1.4988813807949009e-05,
      "loss": 0.9113,
      "num_input_tokens_seen": 1975681792,
      "step": 26620,
      "train_runtime": 31601.4822,
      "train_tokens_per_second": 62518.643
    },
    {
      "epoch": 0.3538846917960678,
      "grad_norm": 22.875,
      "learning_rate": 1.4985083228077675e-05,
      "loss": 0.7151,
      "num_input_tokens_seen": 1976179968,
      "step": 26630,
      "train_runtime": 31608.3507,
      "train_tokens_per_second": 62520.819
    },
    {
      "epoch": 0.3540175812785297,
      "grad_norm": 15.625,
      "learning_rate": 1.4981351724764122e-05,
      "loss": 0.6738,
      "num_input_tokens_seen": 1976419072,
      "step": 26640,
      "train_runtime": 31611.8421,
      "train_tokens_per_second": 62521.477
    },
    {
      "epoch": 0.3541504707609916,
      "grad_norm": 15.8125,
      "learning_rate": 1.4977619298699581e-05,
      "loss": 0.532,
      "num_input_tokens_seen": 1976537600,
      "step": 26650,
      "train_runtime": 31614.8434,
      "train_tokens_per_second": 62519.291
    },
    {
      "epoch": 0.3542833602434535,
      "grad_norm": 15.0,
      "learning_rate": 1.497388595057545e-05,
      "loss": 0.7548,
      "num_input_tokens_seen": 1978633728,
      "step": 26660,
      "train_runtime": 31648.3501,
      "train_tokens_per_second": 62519.333
    },
    {
      "epoch": 0.35441624972591546,
      "grad_norm": 26.625,
      "learning_rate": 1.4970151681083298e-05,
      "loss": 0.9192,
      "num_input_tokens_seen": 1979480576,
      "step": 26670,
      "train_runtime": 31661.4724,
      "train_tokens_per_second": 62520.168
    },
    {
      "epoch": 0.35454913920837733,
      "grad_norm": 54.0,
      "learning_rate": 1.4966416490914865e-05,
      "loss": 0.7747,
      "num_input_tokens_seen": 1979984896,
      "step": 26680,
      "train_runtime": 31668.549,
      "train_tokens_per_second": 62522.122
    },
    {
      "epoch": 0.35468202869083926,
      "grad_norm": 18.375,
      "learning_rate": 1.4962680380762062e-05,
      "loss": 0.5982,
      "num_input_tokens_seen": 1980229632,
      "step": 26690,
      "train_runtime": 31672.6135,
      "train_tokens_per_second": 62521.826
    },
    {
      "epoch": 0.3548149181733012,
      "grad_norm": 16.875,
      "learning_rate": 1.495894335131697e-05,
      "loss": 0.529,
      "num_input_tokens_seen": 1980345344,
      "step": 26700,
      "train_runtime": 31675.0303,
      "train_tokens_per_second": 62520.709
    },
    {
      "epoch": 0.35494780765576306,
      "grad_norm": 20.5,
      "learning_rate": 1.495520540327184e-05,
      "loss": 0.764,
      "num_input_tokens_seen": 1982341888,
      "step": 26710,
      "train_runtime": 31706.867,
      "train_tokens_per_second": 62520.901
    },
    {
      "epoch": 0.355080697138225,
      "grad_norm": 15.75,
      "learning_rate": 1.4951466537319093e-05,
      "loss": 0.8531,
      "num_input_tokens_seen": 1983183104,
      "step": 26720,
      "train_runtime": 31719.7998,
      "train_tokens_per_second": 62521.93
    },
    {
      "epoch": 0.3552135866206869,
      "grad_norm": 11.9375,
      "learning_rate": 1.4947726754151321e-05,
      "loss": 0.7431,
      "num_input_tokens_seen": 1983638784,
      "step": 26730,
      "train_runtime": 31726.5134,
      "train_tokens_per_second": 62523.063
    },
    {
      "epoch": 0.3553464761031488,
      "grad_norm": 15.8125,
      "learning_rate": 1.4943986054461287e-05,
      "loss": 0.658,
      "num_input_tokens_seen": 1983875840,
      "step": 26740,
      "train_runtime": 31730.1462,
      "train_tokens_per_second": 62523.375
    },
    {
      "epoch": 0.3554793655856107,
      "grad_norm": 10.375,
      "learning_rate": 1.4940244438941923e-05,
      "loss": 0.4981,
      "num_input_tokens_seen": 1983992832,
      "step": 26750,
      "train_runtime": 31732.8361,
      "train_tokens_per_second": 62521.762
    },
    {
      "epoch": 0.35561225506807265,
      "grad_norm": 19.25,
      "learning_rate": 1.4936501908286326e-05,
      "loss": 0.7925,
      "num_input_tokens_seen": 1986059008,
      "step": 26760,
      "train_runtime": 31765.4919,
      "train_tokens_per_second": 62522.533
    },
    {
      "epoch": 0.3557451445505346,
      "grad_norm": 75.5,
      "learning_rate": 1.4932758463187767e-05,
      "loss": 0.9177,
      "num_input_tokens_seen": 1986894080,
      "step": 26770,
      "train_runtime": 31778.4676,
      "train_tokens_per_second": 62523.282
    },
    {
      "epoch": 0.35587803403299645,
      "grad_norm": 19.75,
      "learning_rate": 1.4929014104339689e-05,
      "loss": 0.7488,
      "num_input_tokens_seen": 1987341056,
      "step": 26780,
      "train_runtime": 31785.0668,
      "train_tokens_per_second": 62524.363
    },
    {
      "epoch": 0.3560109235154584,
      "grad_norm": 36.25,
      "learning_rate": 1.49252688324357e-05,
      "loss": 0.672,
      "num_input_tokens_seen": 1987576832,
      "step": 26790,
      "train_runtime": 31788.5712,
      "train_tokens_per_second": 62524.887
    },
    {
      "epoch": 0.3561438129979203,
      "grad_norm": 8.125,
      "learning_rate": 1.4921522648169575e-05,
      "loss": 0.5661,
      "num_input_tokens_seen": 1987692544,
      "step": 26800,
      "train_runtime": 31790.9732,
      "train_tokens_per_second": 62523.803
    },
    {
      "epoch": 0.3562767024803822,
      "grad_norm": 38.75,
      "learning_rate": 1.4917775552235266e-05,
      "loss": 0.8873,
      "num_input_tokens_seen": 1989673984,
      "step": 26810,
      "train_runtime": 31821.8277,
      "train_tokens_per_second": 62525.447
    },
    {
      "epoch": 0.3564095919628441,
      "grad_norm": 29.0,
      "learning_rate": 1.4914027545326888e-05,
      "loss": 0.9111,
      "num_input_tokens_seen": 1990530304,
      "step": 26820,
      "train_runtime": 31834.8017,
      "train_tokens_per_second": 62526.864
    },
    {
      "epoch": 0.35654248144530604,
      "grad_norm": 29.5,
      "learning_rate": 1.491027862813872e-05,
      "loss": 0.7426,
      "num_input_tokens_seen": 1991023360,
      "step": 26830,
      "train_runtime": 31842.1664,
      "train_tokens_per_second": 62527.886
    },
    {
      "epoch": 0.3566753709277679,
      "grad_norm": 22.25,
      "learning_rate": 1.4906528801365227e-05,
      "loss": 0.6048,
      "num_input_tokens_seen": 1991247872,
      "step": 26840,
      "train_runtime": 31845.5681,
      "train_tokens_per_second": 62528.257
    },
    {
      "epoch": 0.35680826041022984,
      "grad_norm": 24.375,
      "learning_rate": 1.490277806570102e-05,
      "loss": 0.5416,
      "num_input_tokens_seen": 1991362304,
      "step": 26850,
      "train_runtime": 31847.9944,
      "train_tokens_per_second": 62527.087
    },
    {
      "epoch": 0.35694114989269177,
      "grad_norm": 24.875,
      "learning_rate": 1.4899026421840894e-05,
      "loss": 0.7915,
      "num_input_tokens_seen": 1993519104,
      "step": 26860,
      "train_runtime": 31882.6872,
      "train_tokens_per_second": 62526.696
    },
    {
      "epoch": 0.35707403937515364,
      "grad_norm": 27.625,
      "learning_rate": 1.4895273870479808e-05,
      "loss": 0.9612,
      "num_input_tokens_seen": 1994390528,
      "step": 26870,
      "train_runtime": 31895.3931,
      "train_tokens_per_second": 62529.109
    },
    {
      "epoch": 0.35720692885761557,
      "grad_norm": 21.875,
      "learning_rate": 1.4891520412312887e-05,
      "loss": 0.7284,
      "num_input_tokens_seen": 1994844928,
      "step": 26880,
      "train_runtime": 31902.7205,
      "train_tokens_per_second": 62528.991
    },
    {
      "epoch": 0.3573398183400775,
      "grad_norm": 22.875,
      "learning_rate": 1.4887766048035425e-05,
      "loss": 0.6645,
      "num_input_tokens_seen": 1995074048,
      "step": 26890,
      "train_runtime": 31906.1684,
      "train_tokens_per_second": 62529.415
    },
    {
      "epoch": 0.35747270782253937,
      "grad_norm": 17.25,
      "learning_rate": 1.488401077834289e-05,
      "loss": 0.5644,
      "num_input_tokens_seen": 1995188224,
      "step": 26900,
      "train_runtime": 31908.5712,
      "train_tokens_per_second": 62528.285
    },
    {
      "epoch": 0.3576055973050013,
      "grad_norm": 21.75,
      "learning_rate": 1.4880254603930908e-05,
      "loss": 0.8193,
      "num_input_tokens_seen": 1997262080,
      "step": 26910,
      "train_runtime": 31940.0578,
      "train_tokens_per_second": 62531.574
    },
    {
      "epoch": 0.35773848678746323,
      "grad_norm": 14.0,
      "learning_rate": 1.4876497525495275e-05,
      "loss": 0.9465,
      "num_input_tokens_seen": 1998134272,
      "step": 26920,
      "train_runtime": 31953.0099,
      "train_tokens_per_second": 62533.523
    },
    {
      "epoch": 0.3578713762699251,
      "grad_norm": 13.375,
      "learning_rate": 1.4872739543731957e-05,
      "loss": 0.7226,
      "num_input_tokens_seen": 1998591232,
      "step": 26930,
      "train_runtime": 31959.5185,
      "train_tokens_per_second": 62535.086
    },
    {
      "epoch": 0.35800426575238703,
      "grad_norm": 11.3125,
      "learning_rate": 1.4868980659337088e-05,
      "loss": 0.6325,
      "num_input_tokens_seen": 1998816000,
      "step": 26940,
      "train_runtime": 31962.9237,
      "train_tokens_per_second": 62535.456
    },
    {
      "epoch": 0.35813715523484896,
      "grad_norm": 8.125,
      "learning_rate": 1.4865220873006971e-05,
      "loss": 0.5627,
      "num_input_tokens_seen": 1998930944,
      "step": 26950,
      "train_runtime": 31965.3247,
      "train_tokens_per_second": 62534.354
    },
    {
      "epoch": 0.35827004471731083,
      "grad_norm": 19.625,
      "learning_rate": 1.4861460185438067e-05,
      "loss": 0.805,
      "num_input_tokens_seen": 2001067264,
      "step": 26960,
      "train_runtime": 31999.5819,
      "train_tokens_per_second": 62534.169
    },
    {
      "epoch": 0.35840293419977276,
      "grad_norm": 25.25,
      "learning_rate": 1.4857698597327012e-05,
      "loss": 0.8994,
      "num_input_tokens_seen": 2001979648,
      "step": 26970,
      "train_runtime": 32013.0005,
      "train_tokens_per_second": 62536.458
    },
    {
      "epoch": 0.3585358236822347,
      "grad_norm": 19.125,
      "learning_rate": 1.485393610937061e-05,
      "loss": 0.7227,
      "num_input_tokens_seen": 2002477056,
      "step": 26980,
      "train_runtime": 32020.083,
      "train_tokens_per_second": 62538.159
    },
    {
      "epoch": 0.35866871316469656,
      "grad_norm": 24.5,
      "learning_rate": 1.4850172722265828e-05,
      "loss": 0.6665,
      "num_input_tokens_seen": 2002721792,
      "step": 26990,
      "train_runtime": 32023.6668,
      "train_tokens_per_second": 62538.803
    },
    {
      "epoch": 0.3588016026471585,
      "grad_norm": 27.125,
      "learning_rate": 1.4846408436709792e-05,
      "loss": 0.4817,
      "num_input_tokens_seen": 2002840064,
      "step": 27000,
      "train_runtime": 32026.577,
      "train_tokens_per_second": 62536.813
    },
    {
      "epoch": 0.3589344921296204,
      "grad_norm": 16.125,
      "learning_rate": 1.4842643253399808e-05,
      "loss": 0.8401,
      "num_input_tokens_seen": 2004977408,
      "step": 27010,
      "train_runtime": 32061.5541,
      "train_tokens_per_second": 62535.253
    },
    {
      "epoch": 0.3590673816120823,
      "grad_norm": 32.25,
      "learning_rate": 1.4838877173033344e-05,
      "loss": 0.9312,
      "num_input_tokens_seen": 2005826304,
      "step": 27020,
      "train_runtime": 32074.3307,
      "train_tokens_per_second": 62536.809
    },
    {
      "epoch": 0.3592002710945442,
      "grad_norm": 31.875,
      "learning_rate": 1.483511019630803e-05,
      "loss": 0.746,
      "num_input_tokens_seen": 2006306816,
      "step": 27030,
      "train_runtime": 32081.1336,
      "train_tokens_per_second": 62538.526
    },
    {
      "epoch": 0.35933316057700615,
      "grad_norm": 13.375,
      "learning_rate": 1.4831342323921667e-05,
      "loss": 0.6428,
      "num_input_tokens_seen": 2006540032,
      "step": 27040,
      "train_runtime": 32084.6474,
      "train_tokens_per_second": 62538.946
    },
    {
      "epoch": 0.359466050059468,
      "grad_norm": 16.375,
      "learning_rate": 1.4827573556572219e-05,
      "loss": 0.5527,
      "num_input_tokens_seen": 2006654464,
      "step": 27050,
      "train_runtime": 32087.5301,
      "train_tokens_per_second": 62536.894
    },
    {
      "epoch": 0.35959893954192995,
      "grad_norm": 10.625,
      "learning_rate": 1.4823803894957816e-05,
      "loss": 0.8714,
      "num_input_tokens_seen": 2008640768,
      "step": 27060,
      "train_runtime": 32119.4651,
      "train_tokens_per_second": 62536.557
    },
    {
      "epoch": 0.3597318290243919,
      "grad_norm": 34.75,
      "learning_rate": 1.4820033339776754e-05,
      "loss": 0.9039,
      "num_input_tokens_seen": 2009499392,
      "step": 27070,
      "train_runtime": 32131.9572,
      "train_tokens_per_second": 62538.966
    },
    {
      "epoch": 0.35986471850685375,
      "grad_norm": 20.75,
      "learning_rate": 1.4816261891727495e-05,
      "loss": 0.7612,
      "num_input_tokens_seen": 2009952000,
      "step": 27080,
      "train_runtime": 32138.4482,
      "train_tokens_per_second": 62540.418
    },
    {
      "epoch": 0.3599976079893157,
      "grad_norm": 23.375,
      "learning_rate": 1.4812489551508667e-05,
      "loss": 0.6073,
      "num_input_tokens_seen": 2010167808,
      "step": 27090,
      "train_runtime": 32141.751,
      "train_tokens_per_second": 62540.706
    },
    {
      "epoch": 0.3601304974717776,
      "grad_norm": 15.125,
      "learning_rate": 1.4808716319819061e-05,
      "loss": 0.4971,
      "num_input_tokens_seen": 2010281728,
      "step": 27100,
      "train_runtime": 32144.146,
      "train_tokens_per_second": 62539.591
    },
    {
      "epoch": 0.3602633869542395,
      "grad_norm": 47.0,
      "learning_rate": 1.4804942197357638e-05,
      "loss": 0.8593,
      "num_input_tokens_seen": 2012338688,
      "step": 27110,
      "train_runtime": 32177.2011,
      "train_tokens_per_second": 62539.271
    },
    {
      "epoch": 0.3603962764367014,
      "grad_norm": 28.25,
      "learning_rate": 1.4801167184823514e-05,
      "loss": 0.9138,
      "num_input_tokens_seen": 2013181696,
      "step": 27120,
      "train_runtime": 32190.2519,
      "train_tokens_per_second": 62540.104
    },
    {
      "epoch": 0.36052916591916334,
      "grad_norm": 25.875,
      "learning_rate": 1.4797391282915981e-05,
      "loss": 0.7144,
      "num_input_tokens_seen": 2013625344,
      "step": 27130,
      "train_runtime": 32196.6847,
      "train_tokens_per_second": 62541.388
    },
    {
      "epoch": 0.3606620554016252,
      "grad_norm": 9.625,
      "learning_rate": 1.4793614492334492e-05,
      "loss": 0.6277,
      "num_input_tokens_seen": 2013843968,
      "step": 27140,
      "train_runtime": 32200.0124,
      "train_tokens_per_second": 62541.714
    },
    {
      "epoch": 0.36079494488408714,
      "grad_norm": 11.375,
      "learning_rate": 1.4789836813778665e-05,
      "loss": 0.4663,
      "num_input_tokens_seen": 2013960192,
      "step": 27150,
      "train_runtime": 32202.4196,
      "train_tokens_per_second": 62540.648
    },
    {
      "epoch": 0.36092783436654907,
      "grad_norm": 26.375,
      "learning_rate": 1.4786058247948272e-05,
      "loss": 0.7646,
      "num_input_tokens_seen": 2016022016,
      "step": 27160,
      "train_runtime": 32234.3999,
      "train_tokens_per_second": 62542.564
    },
    {
      "epoch": 0.36106072384901094,
      "grad_norm": 14.0,
      "learning_rate": 1.4782278795543268e-05,
      "loss": 0.903,
      "num_input_tokens_seen": 2016879616,
      "step": 27170,
      "train_runtime": 32247.4792,
      "train_tokens_per_second": 62543.792
    },
    {
      "epoch": 0.3611936133314729,
      "grad_norm": 15.75,
      "learning_rate": 1.477849845726376e-05,
      "loss": 0.7222,
      "num_input_tokens_seen": 2017362688,
      "step": 27180,
      "train_runtime": 32254.3275,
      "train_tokens_per_second": 62545.489
    },
    {
      "epoch": 0.3613265028139348,
      "grad_norm": 25.875,
      "learning_rate": 1.477471723381002e-05,
      "loss": 0.6782,
      "num_input_tokens_seen": 2017600768,
      "step": 27190,
      "train_runtime": 32257.9098,
      "train_tokens_per_second": 62545.924
    },
    {
      "epoch": 0.3614593922963967,
      "grad_norm": 8.5625,
      "learning_rate": 1.4770935125882493e-05,
      "loss": 0.5277,
      "num_input_tokens_seen": 2017720064,
      "step": 27200,
      "train_runtime": 32260.3535,
      "train_tokens_per_second": 62544.884
    },
    {
      "epoch": 0.3615922817788586,
      "grad_norm": 23.25,
      "learning_rate": 1.476715213418177e-05,
      "loss": 0.8364,
      "num_input_tokens_seen": 2019747072,
      "step": 27210,
      "train_runtime": 32292.9097,
      "train_tokens_per_second": 62544.598
    },
    {
      "epoch": 0.36172517126132053,
      "grad_norm": 13.3125,
      "learning_rate": 1.4763368259408627e-05,
      "loss": 0.9067,
      "num_input_tokens_seen": 2020584448,
      "step": 27220,
      "train_runtime": 32305.9935,
      "train_tokens_per_second": 62545.188
    },
    {
      "epoch": 0.3618580607437824,
      "grad_norm": 19.875,
      "learning_rate": 1.4759583502263985e-05,
      "loss": 0.7148,
      "num_input_tokens_seen": 2021054464,
      "step": 27230,
      "train_runtime": 32312.6743,
      "train_tokens_per_second": 62546.803
    },
    {
      "epoch": 0.36199095022624433,
      "grad_norm": 18.875,
      "learning_rate": 1.4755797863448939e-05,
      "loss": 0.6295,
      "num_input_tokens_seen": 2021285888,
      "step": 27240,
      "train_runtime": 32316.1631,
      "train_tokens_per_second": 62547.211
    },
    {
      "epoch": 0.36212383970870626,
      "grad_norm": 13.5,
      "learning_rate": 1.4752011343664747e-05,
      "loss": 0.5032,
      "num_input_tokens_seen": 2021404672,
      "step": 27250,
      "train_runtime": 32318.5991,
      "train_tokens_per_second": 62546.172
    },
    {
      "epoch": 0.3622567291911682,
      "grad_norm": 32.75,
      "learning_rate": 1.4748223943612825e-05,
      "loss": 0.7799,
      "num_input_tokens_seen": 2023447040,
      "step": 27260,
      "train_runtime": 32350.9273,
      "train_tokens_per_second": 62546.802
    },
    {
      "epoch": 0.36238961867363007,
      "grad_norm": 20.25,
      "learning_rate": 1.4744435663994758e-05,
      "loss": 0.9312,
      "num_input_tokens_seen": 2024273152,
      "step": 27270,
      "train_runtime": 32363.3922,
      "train_tokens_per_second": 62548.238
    },
    {
      "epoch": 0.362522508156092,
      "grad_norm": 18.625,
      "learning_rate": 1.4740646505512288e-05,
      "loss": 0.8321,
      "num_input_tokens_seen": 2024709376,
      "step": 27280,
      "train_runtime": 32370.3312,
      "train_tokens_per_second": 62548.306
    },
    {
      "epoch": 0.3626553976385539,
      "grad_norm": 16.375,
      "learning_rate": 1.4736856468867327e-05,
      "loss": 0.6159,
      "num_input_tokens_seen": 2024937472,
      "step": 27290,
      "train_runtime": 32373.7778,
      "train_tokens_per_second": 62548.693
    },
    {
      "epoch": 0.3627882871210158,
      "grad_norm": 19.375,
      "learning_rate": 1.473306555476194e-05,
      "loss": 0.599,
      "num_input_tokens_seen": 2025051648,
      "step": 27300,
      "train_runtime": 32376.2023,
      "train_tokens_per_second": 62547.535
    },
    {
      "epoch": 0.3629211766034777,
      "grad_norm": 23.5,
      "learning_rate": 1.4729273763898367e-05,
      "loss": 0.8291,
      "num_input_tokens_seen": 2027105536,
      "step": 27310,
      "train_runtime": 32408.8209,
      "train_tokens_per_second": 62547.957
    },
    {
      "epoch": 0.36305406608593965,
      "grad_norm": 24.75,
      "learning_rate": 1.4725481096978997e-05,
      "loss": 0.9004,
      "num_input_tokens_seen": 2027943424,
      "step": 27320,
      "train_runtime": 32421.4941,
      "train_tokens_per_second": 62549.351
    },
    {
      "epoch": 0.3631869555684015,
      "grad_norm": 10.4375,
      "learning_rate": 1.472168755470639e-05,
      "loss": 0.7539,
      "num_input_tokens_seen": 2028405248,
      "step": 27330,
      "train_runtime": 32428.8201,
      "train_tokens_per_second": 62549.462
    },
    {
      "epoch": 0.36331984505086345,
      "grad_norm": 22.25,
      "learning_rate": 1.4717893137783264e-05,
      "loss": 0.6428,
      "num_input_tokens_seen": 2028644864,
      "step": 27340,
      "train_runtime": 32432.4657,
      "train_tokens_per_second": 62549.819
    },
    {
      "epoch": 0.3634527345333254,
      "grad_norm": 8.4375,
      "learning_rate": 1.4714097846912505e-05,
      "loss": 0.511,
      "num_input_tokens_seen": 2028759808,
      "step": 27350,
      "train_runtime": 32435.1972,
      "train_tokens_per_second": 62548.095
    },
    {
      "epoch": 0.36358562401578726,
      "grad_norm": 15.125,
      "learning_rate": 1.4710301682797155e-05,
      "loss": 0.9113,
      "num_input_tokens_seen": 2030752768,
      "step": 27360,
      "train_runtime": 32467.5901,
      "train_tokens_per_second": 62547.074
    },
    {
      "epoch": 0.3637185134982492,
      "grad_norm": 14.6875,
      "learning_rate": 1.470650464614042e-05,
      "loss": 0.924,
      "num_input_tokens_seen": 2031604992,
      "step": 27370,
      "train_runtime": 32480.3803,
      "train_tokens_per_second": 62548.682
    },
    {
      "epoch": 0.3638514029807111,
      "grad_norm": 27.875,
      "learning_rate": 1.4702706737645666e-05,
      "loss": 0.7827,
      "num_input_tokens_seen": 2032072448,
      "step": 27380,
      "train_runtime": 32487.441,
      "train_tokens_per_second": 62549.477
    },
    {
      "epoch": 0.363984292463173,
      "grad_norm": 13.6875,
      "learning_rate": 1.469890795801642e-05,
      "loss": 0.6701,
      "num_input_tokens_seen": 2032299264,
      "step": 27390,
      "train_runtime": 32490.869,
      "train_tokens_per_second": 62549.859
    },
    {
      "epoch": 0.3641171819456349,
      "grad_norm": 5.6875,
      "learning_rate": 1.4695108307956376e-05,
      "loss": 0.4963,
      "num_input_tokens_seen": 2032410624,
      "step": 27400,
      "train_runtime": 32493.2375,
      "train_tokens_per_second": 62548.727
    },
    {
      "epoch": 0.36425007142809684,
      "grad_norm": 33.25,
      "learning_rate": 1.4691307788169384e-05,
      "loss": 0.8053,
      "num_input_tokens_seen": 2034390784,
      "step": 27410,
      "train_runtime": 32526.2236,
      "train_tokens_per_second": 62546.172
    },
    {
      "epoch": 0.3643829609105587,
      "grad_norm": 14.375,
      "learning_rate": 1.4687506399359454e-05,
      "loss": 0.8813,
      "num_input_tokens_seen": 2035213056,
      "step": 27420,
      "train_runtime": 32538.636,
      "train_tokens_per_second": 62547.584
    },
    {
      "epoch": 0.36451585039302065,
      "grad_norm": 18.25,
      "learning_rate": 1.4683704142230761e-05,
      "loss": 0.7074,
      "num_input_tokens_seen": 2035663104,
      "step": 27430,
      "train_runtime": 32545.2235,
      "train_tokens_per_second": 62548.752
    },
    {
      "epoch": 0.3646487398754826,
      "grad_norm": 26.875,
      "learning_rate": 1.4679901017487642e-05,
      "loss": 0.727,
      "num_input_tokens_seen": 2035885312,
      "step": 27440,
      "train_runtime": 32548.6254,
      "train_tokens_per_second": 62549.041
    },
    {
      "epoch": 0.36478162935794445,
      "grad_norm": 14.75,
      "learning_rate": 1.4676097025834586e-05,
      "loss": 0.5433,
      "num_input_tokens_seen": 2035998720,
      "step": 27450,
      "train_runtime": 32551.57,
      "train_tokens_per_second": 62546.867
    },
    {
      "epoch": 0.3649145188404064,
      "grad_norm": 24.875,
      "learning_rate": 1.4672292167976254e-05,
      "loss": 0.9,
      "num_input_tokens_seen": 2038141696,
      "step": 27460,
      "train_runtime": 32585.9943,
      "train_tokens_per_second": 62546.555
    },
    {
      "epoch": 0.3650474083228683,
      "grad_norm": 80.0,
      "learning_rate": 1.4668486444617462e-05,
      "loss": 0.9381,
      "num_input_tokens_seen": 2039001344,
      "step": 27470,
      "train_runtime": 32599.0566,
      "train_tokens_per_second": 62547.864
    },
    {
      "epoch": 0.3651802978053302,
      "grad_norm": 13.625,
      "learning_rate": 1.4664679856463181e-05,
      "loss": 0.7856,
      "num_input_tokens_seen": 2039489280,
      "step": 27480,
      "train_runtime": 32605.9678,
      "train_tokens_per_second": 62549.571
    },
    {
      "epoch": 0.3653131872877921,
      "grad_norm": 14.125,
      "learning_rate": 1.4660872404218553e-05,
      "loss": 0.68,
      "num_input_tokens_seen": 2039732992,
      "step": 27490,
      "train_runtime": 32609.5109,
      "train_tokens_per_second": 62550.248
    },
    {
      "epoch": 0.36544607677025404,
      "grad_norm": 28.5,
      "learning_rate": 1.4657064088588871e-05,
      "loss": 0.5132,
      "num_input_tokens_seen": 2039851520,
      "step": 27500,
      "train_runtime": 32612.458,
      "train_tokens_per_second": 62548.23
    },
    {
      "epoch": 0.3655789662527159,
      "grad_norm": 25.5,
      "learning_rate": 1.4653254910279593e-05,
      "loss": 0.8442,
      "num_input_tokens_seen": 2041817600,
      "step": 27510,
      "train_runtime": 32645.5009,
      "train_tokens_per_second": 62545.145
    },
    {
      "epoch": 0.36571185573517784,
      "grad_norm": 29.25,
      "learning_rate": 1.4649444869996337e-05,
      "loss": 0.903,
      "num_input_tokens_seen": 2042645504,
      "step": 27520,
      "train_runtime": 32658.0126,
      "train_tokens_per_second": 62546.534
    },
    {
      "epoch": 0.36584474521763977,
      "grad_norm": 17.0,
      "learning_rate": 1.4645633968444872e-05,
      "loss": 0.7947,
      "num_input_tokens_seen": 2043113216,
      "step": 27530,
      "train_runtime": 32664.6796,
      "train_tokens_per_second": 62548.087
    },
    {
      "epoch": 0.36597763470010164,
      "grad_norm": 11.625,
      "learning_rate": 1.4641822206331142e-05,
      "loss": 0.674,
      "num_input_tokens_seen": 2043343360,
      "step": 27540,
      "train_runtime": 32668.1676,
      "train_tokens_per_second": 62548.453
    },
    {
      "epoch": 0.36611052418256357,
      "grad_norm": 11.1875,
      "learning_rate": 1.4638009584361241e-05,
      "loss": 0.6478,
      "num_input_tokens_seen": 2043457792,
      "step": 27550,
      "train_runtime": 32670.5478,
      "train_tokens_per_second": 62547.399
    },
    {
      "epoch": 0.3662434136650255,
      "grad_norm": 36.25,
      "learning_rate": 1.4634196103241417e-05,
      "loss": 0.8649,
      "num_input_tokens_seen": 2045483008,
      "step": 27560,
      "train_runtime": 32702.6325,
      "train_tokens_per_second": 62547.962
    },
    {
      "epoch": 0.36637630314748737,
      "grad_norm": 18.625,
      "learning_rate": 1.4630381763678093e-05,
      "loss": 0.9353,
      "num_input_tokens_seen": 2046308352,
      "step": 27570,
      "train_runtime": 32714.9888,
      "train_tokens_per_second": 62549.566
    },
    {
      "epoch": 0.3665091926299493,
      "grad_norm": 23.5,
      "learning_rate": 1.4626566566377828e-05,
      "loss": 0.8064,
      "num_input_tokens_seen": 2046764032,
      "step": 27580,
      "train_runtime": 32721.4334,
      "train_tokens_per_second": 62551.173
    },
    {
      "epoch": 0.3666420821124112,
      "grad_norm": 23.625,
      "learning_rate": 1.4622750512047366e-05,
      "loss": 0.6792,
      "num_input_tokens_seen": 2046977792,
      "step": 27590,
      "train_runtime": 32724.7573,
      "train_tokens_per_second": 62551.351
    },
    {
      "epoch": 0.3667749715948731,
      "grad_norm": 12.3125,
      "learning_rate": 1.4618933601393588e-05,
      "loss": 0.5172,
      "num_input_tokens_seen": 2047088128,
      "step": 27600,
      "train_runtime": 32727.1406,
      "train_tokens_per_second": 62550.168
    },
    {
      "epoch": 0.36690786107733503,
      "grad_norm": 24.25,
      "learning_rate": 1.461511583512355e-05,
      "loss": 0.8394,
      "num_input_tokens_seen": 2049204736,
      "step": 27610,
      "train_runtime": 32760.8388,
      "train_tokens_per_second": 62550.436
    },
    {
      "epoch": 0.36704075055979696,
      "grad_norm": 23.25,
      "learning_rate": 1.4611297213944455e-05,
      "loss": 0.9422,
      "num_input_tokens_seen": 2050032896,
      "step": 27620,
      "train_runtime": 32773.8084,
      "train_tokens_per_second": 62550.951
    },
    {
      "epoch": 0.36717364004225883,
      "grad_norm": 13.5625,
      "learning_rate": 1.460747773856367e-05,
      "loss": 0.7891,
      "num_input_tokens_seen": 2050502656,
      "step": 27630,
      "train_runtime": 32780.5211,
      "train_tokens_per_second": 62552.473
    },
    {
      "epoch": 0.36730652952472076,
      "grad_norm": 16.875,
      "learning_rate": 1.4603657409688716e-05,
      "loss": 0.6861,
      "num_input_tokens_seen": 2050730240,
      "step": 27640,
      "train_runtime": 32783.9861,
      "train_tokens_per_second": 62552.803
    },
    {
      "epoch": 0.3674394190071827,
      "grad_norm": 19.75,
      "learning_rate": 1.459983622802728e-05,
      "loss": 0.4115,
      "num_input_tokens_seen": 2050843904,
      "step": 27650,
      "train_runtime": 32786.36,
      "train_tokens_per_second": 62551.741
    },
    {
      "epoch": 0.36757230848964456,
      "grad_norm": 18.625,
      "learning_rate": 1.4596014194287195e-05,
      "loss": 0.7613,
      "num_input_tokens_seen": 2052875520,
      "step": 27660,
      "train_runtime": 32818.8185,
      "train_tokens_per_second": 62551.78
    },
    {
      "epoch": 0.3677051979721065,
      "grad_norm": 37.25,
      "learning_rate": 1.4592191309176466e-05,
      "loss": 0.8678,
      "num_input_tokens_seen": 2053719040,
      "step": 27670,
      "train_runtime": 32831.8612,
      "train_tokens_per_second": 62552.623
    },
    {
      "epoch": 0.3678380874545684,
      "grad_norm": 12.875,
      "learning_rate": 1.4588367573403242e-05,
      "loss": 0.8193,
      "num_input_tokens_seen": 2054193664,
      "step": 27680,
      "train_runtime": 32838.4932,
      "train_tokens_per_second": 62554.443
    },
    {
      "epoch": 0.3679709769370303,
      "grad_norm": 12.0625,
      "learning_rate": 1.4584542987675838e-05,
      "loss": 0.5864,
      "num_input_tokens_seen": 2054422272,
      "step": 27690,
      "train_runtime": 32841.9233,
      "train_tokens_per_second": 62554.871
    },
    {
      "epoch": 0.3681038664194922,
      "grad_norm": 22.125,
      "learning_rate": 1.4580717552702727e-05,
      "loss": 0.5957,
      "num_input_tokens_seen": 2054534912,
      "step": 27700,
      "train_runtime": 32844.3195,
      "train_tokens_per_second": 62553.737
    },
    {
      "epoch": 0.36823675590195415,
      "grad_norm": 17.0,
      "learning_rate": 1.4576891269192537e-05,
      "loss": 0.7816,
      "num_input_tokens_seen": 2056548096,
      "step": 27710,
      "train_runtime": 32877.6342,
      "train_tokens_per_second": 62551.584
    },
    {
      "epoch": 0.3683696453844161,
      "grad_norm": 26.5,
      "learning_rate": 1.4573064137854047e-05,
      "loss": 0.8912,
      "num_input_tokens_seen": 2057379584,
      "step": 27720,
      "train_runtime": 32890.2423,
      "train_tokens_per_second": 62552.886
    },
    {
      "epoch": 0.36850253486687795,
      "grad_norm": 21.0,
      "learning_rate": 1.4569236159396205e-05,
      "loss": 0.795,
      "num_input_tokens_seen": 2057844736,
      "step": 27730,
      "train_runtime": 32896.9196,
      "train_tokens_per_second": 62554.329
    },
    {
      "epoch": 0.3686354243493399,
      "grad_norm": 27.25,
      "learning_rate": 1.4565407334528107e-05,
      "loss": 0.6641,
      "num_input_tokens_seen": 2058080000,
      "step": 27740,
      "train_runtime": 32900.4248,
      "train_tokens_per_second": 62554.815
    },
    {
      "epoch": 0.3687683138318018,
      "grad_norm": 22.75,
      "learning_rate": 1.4561577663959009e-05,
      "loss": 0.58,
      "num_input_tokens_seen": 2058197504,
      "step": 27750,
      "train_runtime": 32902.8124,
      "train_tokens_per_second": 62553.847
    },
    {
      "epoch": 0.3689012033142637,
      "grad_norm": 11.0,
      "learning_rate": 1.4557747148398322e-05,
      "loss": 0.822,
      "num_input_tokens_seen": 2060325120,
      "step": 27760,
      "train_runtime": 32937.7794,
      "train_tokens_per_second": 62552.035
    },
    {
      "epoch": 0.3690340927967256,
      "grad_norm": 12.4375,
      "learning_rate": 1.4553915788555621e-05,
      "loss": 0.8954,
      "num_input_tokens_seen": 2061154560,
      "step": 27770,
      "train_runtime": 32950.5692,
      "train_tokens_per_second": 62552.927
    },
    {
      "epoch": 0.36916698227918754,
      "grad_norm": 15.125,
      "learning_rate": 1.4550083585140624e-05,
      "loss": 0.8048,
      "num_input_tokens_seen": 2061634048,
      "step": 27780,
      "train_runtime": 32957.88,
      "train_tokens_per_second": 62553.6
    },
    {
      "epoch": 0.3692998717616494,
      "grad_norm": 9.75,
      "learning_rate": 1.4546250538863216e-05,
      "loss": 0.6118,
      "num_input_tokens_seen": 2061863936,
      "step": 27790,
      "train_runtime": 32961.37,
      "train_tokens_per_second": 62553.951
    },
    {
      "epoch": 0.36943276124411134,
      "grad_norm": 33.5,
      "learning_rate": 1.4542416650433433e-05,
      "loss": 0.6151,
      "num_input_tokens_seen": 2061981440,
      "step": 27800,
      "train_runtime": 32963.7555,
      "train_tokens_per_second": 62552.989
    },
    {
      "epoch": 0.36956565072657327,
      "grad_norm": 29.125,
      "learning_rate": 1.4538581920561468e-05,
      "loss": 0.7825,
      "num_input_tokens_seen": 2064067328,
      "step": 27810,
      "train_runtime": 32997.1557,
      "train_tokens_per_second": 62552.886
    },
    {
      "epoch": 0.36969854020903514,
      "grad_norm": 24.75,
      "learning_rate": 1.4534746349957672e-05,
      "loss": 0.9725,
      "num_input_tokens_seen": 2064922880,
      "step": 27820,
      "train_runtime": 33009.7771,
      "train_tokens_per_second": 62554.887
    },
    {
      "epoch": 0.36983142969149707,
      "grad_norm": 35.75,
      "learning_rate": 1.4530909939332553e-05,
      "loss": 0.795,
      "num_input_tokens_seen": 2065386496,
      "step": 27830,
      "train_runtime": 33016.9682,
      "train_tokens_per_second": 62555.304
    },
    {
      "epoch": 0.369964319173959,
      "grad_norm": 17.875,
      "learning_rate": 1.4527072689396767e-05,
      "loss": 0.6231,
      "num_input_tokens_seen": 2065606656,
      "step": 27840,
      "train_runtime": 33020.55,
      "train_tokens_per_second": 62555.186
    },
    {
      "epoch": 0.3700972086564209,
      "grad_norm": 9.5625,
      "learning_rate": 1.4523234600861134e-05,
      "loss": 0.5262,
      "num_input_tokens_seen": 2065721088,
      "step": 27850,
      "train_runtime": 33023.0166,
      "train_tokens_per_second": 62553.979
    },
    {
      "epoch": 0.3702300981388828,
      "grad_norm": 35.25,
      "learning_rate": 1.4519395674436623e-05,
      "loss": 0.9068,
      "num_input_tokens_seen": 2067684096,
      "step": 27860,
      "train_runtime": 33053.8071,
      "train_tokens_per_second": 62555.097
    },
    {
      "epoch": 0.37036298762134473,
      "grad_norm": 14.375,
      "learning_rate": 1.4515555910834363e-05,
      "loss": 0.811,
      "num_input_tokens_seen": 2068504832,
      "step": 27870,
      "train_runtime": 33065.8157,
      "train_tokens_per_second": 62557.2
    },
    {
      "epoch": 0.3704958771038066,
      "grad_norm": 24.5,
      "learning_rate": 1.4511715310765632e-05,
      "loss": 0.6888,
      "num_input_tokens_seen": 2068952576,
      "step": 27880,
      "train_runtime": 33072.1715,
      "train_tokens_per_second": 62558.716
    },
    {
      "epoch": 0.37062876658626853,
      "grad_norm": 22.875,
      "learning_rate": 1.4507873874941876e-05,
      "loss": 0.6134,
      "num_input_tokens_seen": 2069174272,
      "step": 27890,
      "train_runtime": 33075.4897,
      "train_tokens_per_second": 62559.142
    },
    {
      "epoch": 0.37076165606873046,
      "grad_norm": 13.1875,
      "learning_rate": 1.4504031604074676e-05,
      "loss": 0.6455,
      "num_input_tokens_seen": 2069286400,
      "step": 27900,
      "train_runtime": 33077.8059,
      "train_tokens_per_second": 62558.152
    },
    {
      "epoch": 0.37089454555119233,
      "grad_norm": 12.4375,
      "learning_rate": 1.4500188498875783e-05,
      "loss": 0.7776,
      "num_input_tokens_seen": 2071316224,
      "step": 27910,
      "train_runtime": 33110.3445,
      "train_tokens_per_second": 62557.979
    },
    {
      "epoch": 0.37102743503365426,
      "grad_norm": 20.375,
      "learning_rate": 1.4496344560057098e-05,
      "loss": 0.8942,
      "num_input_tokens_seen": 2072103424,
      "step": 27920,
      "train_runtime": 33122.3125,
      "train_tokens_per_second": 62559.141
    },
    {
      "epoch": 0.3711603245161162,
      "grad_norm": 30.375,
      "learning_rate": 1.4492499788330679e-05,
      "loss": 0.737,
      "num_input_tokens_seen": 2072550400,
      "step": 27930,
      "train_runtime": 33128.5993,
      "train_tokens_per_second": 62560.761
    },
    {
      "epoch": 0.37129321399857806,
      "grad_norm": 17.0,
      "learning_rate": 1.448865418440873e-05,
      "loss": 0.6647,
      "num_input_tokens_seen": 2072773376,
      "step": 27940,
      "train_runtime": 33131.9639,
      "train_tokens_per_second": 62561.138
    },
    {
      "epoch": 0.37142610348104,
      "grad_norm": 15.5625,
      "learning_rate": 1.4484807749003622e-05,
      "loss": 0.497,
      "num_input_tokens_seen": 2072886528,
      "step": 27950,
      "train_runtime": 33134.3479,
      "train_tokens_per_second": 62560.052
    },
    {
      "epoch": 0.3715589929635019,
      "grad_norm": 16.125,
      "learning_rate": 1.4480960482827863e-05,
      "loss": 0.8291,
      "num_input_tokens_seen": 2074957056,
      "step": 27960,
      "train_runtime": 33168.2111,
      "train_tokens_per_second": 62558.606
    },
    {
      "epoch": 0.3716918824459638,
      "grad_norm": 15.125,
      "learning_rate": 1.4477112386594135e-05,
      "loss": 0.8824,
      "num_input_tokens_seen": 2075788800,
      "step": 27970,
      "train_runtime": 33181.4595,
      "train_tokens_per_second": 62558.695
    },
    {
      "epoch": 0.3718247719284257,
      "grad_norm": 18.0,
      "learning_rate": 1.4473263461015257e-05,
      "loss": 0.7793,
      "num_input_tokens_seen": 2076276480,
      "step": 27980,
      "train_runtime": 33188.2793,
      "train_tokens_per_second": 62560.534
    },
    {
      "epoch": 0.37195766141088765,
      "grad_norm": 65.0,
      "learning_rate": 1.4469413706804211e-05,
      "loss": 0.6566,
      "num_input_tokens_seen": 2076504576,
      "step": 27990,
      "train_runtime": 33191.7274,
      "train_tokens_per_second": 62560.907
    },
    {
      "epoch": 0.3720905508933495,
      "grad_norm": 19.75,
      "learning_rate": 1.4465563124674129e-05,
      "loss": 0.5956,
      "num_input_tokens_seen": 2076617984,
      "step": 28000,
      "train_runtime": 33194.1401,
      "train_tokens_per_second": 62559.776
    },
    {
      "epoch": 0.37222344037581145,
      "grad_norm": 15.75,
      "learning_rate": 1.4461711715338295e-05,
      "loss": 0.7998,
      "num_input_tokens_seen": 2078640384,
      "step": 28010,
      "train_runtime": 33226.4791,
      "train_tokens_per_second": 62559.755
    },
    {
      "epoch": 0.3723563298582734,
      "grad_norm": 20.125,
      "learning_rate": 1.4457859479510153e-05,
      "loss": 0.9061,
      "num_input_tokens_seen": 2079432448,
      "step": 28020,
      "train_runtime": 33239.087,
      "train_tokens_per_second": 62559.854
    },
    {
      "epoch": 0.37248921934073526,
      "grad_norm": 31.75,
      "learning_rate": 1.4454006417903292e-05,
      "loss": 0.6984,
      "num_input_tokens_seen": 2079875328,
      "step": 28030,
      "train_runtime": 33245.6359,
      "train_tokens_per_second": 62560.853
    },
    {
      "epoch": 0.3726221088231972,
      "grad_norm": 10.5,
      "learning_rate": 1.4450152531231458e-05,
      "loss": 0.6264,
      "num_input_tokens_seen": 2080110592,
      "step": 28040,
      "train_runtime": 33249.1128,
      "train_tokens_per_second": 62561.386
    },
    {
      "epoch": 0.3727549983056591,
      "grad_norm": 11.3125,
      "learning_rate": 1.4446297820208548e-05,
      "loss": 0.479,
      "num_input_tokens_seen": 2080225280,
      "step": 28050,
      "train_runtime": 33251.5461,
      "train_tokens_per_second": 62560.257
    },
    {
      "epoch": 0.372887887788121,
      "grad_norm": 14.375,
      "learning_rate": 1.4442442285548612e-05,
      "loss": 0.7427,
      "num_input_tokens_seen": 2082385152,
      "step": 28060,
      "train_runtime": 33285.5447,
      "train_tokens_per_second": 62561.246
    },
    {
      "epoch": 0.3730207772705829,
      "grad_norm": 14.375,
      "learning_rate": 1.4438585927965856e-05,
      "loss": 0.9527,
      "num_input_tokens_seen": 2083276032,
      "step": 28070,
      "train_runtime": 33299.0778,
      "train_tokens_per_second": 62562.574
    },
    {
      "epoch": 0.37315366675304484,
      "grad_norm": 19.75,
      "learning_rate": 1.4434728748174635e-05,
      "loss": 0.7385,
      "num_input_tokens_seen": 2083768832,
      "step": 28080,
      "train_runtime": 33305.9488,
      "train_tokens_per_second": 62564.464
    },
    {
      "epoch": 0.3732865562355067,
      "grad_norm": 24.125,
      "learning_rate": 1.443087074688946e-05,
      "loss": 0.6674,
      "num_input_tokens_seen": 2084001024,
      "step": 28090,
      "train_runtime": 33309.4197,
      "train_tokens_per_second": 62564.915
    },
    {
      "epoch": 0.37341944571796865,
      "grad_norm": 21.0,
      "learning_rate": 1.4427011924824984e-05,
      "loss": 0.5223,
      "num_input_tokens_seen": 2084117248,
      "step": 28100,
      "train_runtime": 33311.8383,
      "train_tokens_per_second": 62563.862
    },
    {
      "epoch": 0.3735523352004306,
      "grad_norm": 30.75,
      "learning_rate": 1.442315228269603e-05,
      "loss": 0.8168,
      "num_input_tokens_seen": 2086104320,
      "step": 28110,
      "train_runtime": 33344.3436,
      "train_tokens_per_second": 62562.465
    },
    {
      "epoch": 0.37368522468289245,
      "grad_norm": 16.375,
      "learning_rate": 1.4419291821217551e-05,
      "loss": 0.9357,
      "num_input_tokens_seen": 2086961408,
      "step": 28120,
      "train_runtime": 33357.7391,
      "train_tokens_per_second": 62563.035
    },
    {
      "epoch": 0.3738181141653544,
      "grad_norm": 11.75,
      "learning_rate": 1.4415430541104669e-05,
      "loss": 0.7769,
      "num_input_tokens_seen": 2087437312,
      "step": 28130,
      "train_runtime": 33364.5449,
      "train_tokens_per_second": 62564.537
    },
    {
      "epoch": 0.3739510036478163,
      "grad_norm": 20.375,
      "learning_rate": 1.441156844307265e-05,
      "loss": 0.6544,
      "num_input_tokens_seen": 2087674112,
      "step": 28140,
      "train_runtime": 33368.0713,
      "train_tokens_per_second": 62565.022
    },
    {
      "epoch": 0.3740838931302782,
      "grad_norm": 24.375,
      "learning_rate": 1.4407705527836917e-05,
      "loss": 0.5353,
      "num_input_tokens_seen": 2087791616,
      "step": 28150,
      "train_runtime": 33370.457,
      "train_tokens_per_second": 62564.07
    },
    {
      "epoch": 0.3742167826127401,
      "grad_norm": 20.25,
      "learning_rate": 1.4403841796113036e-05,
      "loss": 0.7988,
      "num_input_tokens_seen": 2089752576,
      "step": 28160,
      "train_runtime": 33401.6765,
      "train_tokens_per_second": 62564.302
    },
    {
      "epoch": 0.37434967209520204,
      "grad_norm": 28.625,
      "learning_rate": 1.439997724861673e-05,
      "loss": 0.9631,
      "num_input_tokens_seen": 2090573568,
      "step": 28170,
      "train_runtime": 33414.163,
      "train_tokens_per_second": 62565.493
    },
    {
      "epoch": 0.3744825615776639,
      "grad_norm": 26.125,
      "learning_rate": 1.4396111886063872e-05,
      "loss": 0.7419,
      "num_input_tokens_seen": 2091075584,
      "step": 28180,
      "train_runtime": 33421.0556,
      "train_tokens_per_second": 62567.61
    },
    {
      "epoch": 0.37461545106012584,
      "grad_norm": 13.625,
      "learning_rate": 1.4392245709170492e-05,
      "loss": 0.6445,
      "num_input_tokens_seen": 2091304192,
      "step": 28190,
      "train_runtime": 33425.4252,
      "train_tokens_per_second": 62566.27
    },
    {
      "epoch": 0.37474834054258777,
      "grad_norm": 9.125,
      "learning_rate": 1.4388378718652755e-05,
      "loss": 0.4577,
      "num_input_tokens_seen": 2091419648,
      "step": 28200,
      "train_runtime": 33427.8408,
      "train_tokens_per_second": 62565.203
    },
    {
      "epoch": 0.3748812300250497,
      "grad_norm": 13.5,
      "learning_rate": 1.438451091522699e-05,
      "loss": 0.7571,
      "num_input_tokens_seen": 2093430528,
      "step": 28210,
      "train_runtime": 33460.5918,
      "train_tokens_per_second": 62564.062
    },
    {
      "epoch": 0.37501411950751157,
      "grad_norm": 14.0625,
      "learning_rate": 1.4380642299609674e-05,
      "loss": 0.9379,
      "num_input_tokens_seen": 2094246144,
      "step": 28220,
      "train_runtime": 33473.1143,
      "train_tokens_per_second": 62565.022
    },
    {
      "epoch": 0.3751470089899735,
      "grad_norm": 24.125,
      "learning_rate": 1.4376772872517435e-05,
      "loss": 0.7022,
      "num_input_tokens_seen": 2094704640,
      "step": 28230,
      "train_runtime": 33479.8612,
      "train_tokens_per_second": 62566.109
    },
    {
      "epoch": 0.3752798984724354,
      "grad_norm": 26.5,
      "learning_rate": 1.4372902634667047e-05,
      "loss": 0.6544,
      "num_input_tokens_seen": 2094937856,
      "step": 28240,
      "train_runtime": 33483.8719,
      "train_tokens_per_second": 62565.58
    },
    {
      "epoch": 0.3754127879548973,
      "grad_norm": 10.0625,
      "learning_rate": 1.4369031586775441e-05,
      "loss": 0.4261,
      "num_input_tokens_seen": 2095052800,
      "step": 28250,
      "train_runtime": 33486.2703,
      "train_tokens_per_second": 62564.531
    },
    {
      "epoch": 0.3755456774373592,
      "grad_norm": 19.875,
      "learning_rate": 1.4365159729559688e-05,
      "loss": 0.7361,
      "num_input_tokens_seen": 2097228544,
      "step": 28260,
      "train_runtime": 33521.8377,
      "train_tokens_per_second": 62563.054
    },
    {
      "epoch": 0.37567856691982116,
      "grad_norm": 14.25,
      "learning_rate": 1.4361287063737022e-05,
      "loss": 0.9108,
      "num_input_tokens_seen": 2098087680,
      "step": 28270,
      "train_runtime": 33534.6532,
      "train_tokens_per_second": 62564.765
    },
    {
      "epoch": 0.37581145640228303,
      "grad_norm": 24.0,
      "learning_rate": 1.4357413590024814e-05,
      "loss": 0.7345,
      "num_input_tokens_seen": 2098546432,
      "step": 28280,
      "train_runtime": 33541.9075,
      "train_tokens_per_second": 62564.91
    },
    {
      "epoch": 0.37594434588474496,
      "grad_norm": 34.0,
      "learning_rate": 1.4353539309140594e-05,
      "loss": 0.6385,
      "num_input_tokens_seen": 2098781184,
      "step": 28290,
      "train_runtime": 33545.4418,
      "train_tokens_per_second": 62565.316
    },
    {
      "epoch": 0.3760772353672069,
      "grad_norm": 25.375,
      "learning_rate": 1.4349664221802036e-05,
      "loss": 0.5954,
      "num_input_tokens_seen": 2098895360,
      "step": 28300,
      "train_runtime": 33547.8253,
      "train_tokens_per_second": 62564.275
    },
    {
      "epoch": 0.37621012484966876,
      "grad_norm": 25.0,
      "learning_rate": 1.4345788328726966e-05,
      "loss": 0.7722,
      "num_input_tokens_seen": 2101020416,
      "step": 28310,
      "train_runtime": 33580.8476,
      "train_tokens_per_second": 62566.033
    },
    {
      "epoch": 0.3763430143321307,
      "grad_norm": 65.5,
      "learning_rate": 1.4341911630633359e-05,
      "loss": 0.8827,
      "num_input_tokens_seen": 2101854208,
      "step": 28320,
      "train_runtime": 33593.5294,
      "train_tokens_per_second": 62567.234
    },
    {
      "epoch": 0.3764759038145926,
      "grad_norm": 13.75,
      "learning_rate": 1.4338034128239338e-05,
      "loss": 0.7787,
      "num_input_tokens_seen": 2102376960,
      "step": 28330,
      "train_runtime": 33600.4868,
      "train_tokens_per_second": 62569.836
    },
    {
      "epoch": 0.3766087932970545,
      "grad_norm": 19.375,
      "learning_rate": 1.4334155822263176e-05,
      "loss": 0.6499,
      "num_input_tokens_seen": 2102619648,
      "step": 28340,
      "train_runtime": 33604.0829,
      "train_tokens_per_second": 62570.362
    },
    {
      "epoch": 0.3767416827795164,
      "grad_norm": 13.8125,
      "learning_rate": 1.4330276713423297e-05,
      "loss": 0.5466,
      "num_input_tokens_seen": 2102739200,
      "step": 28350,
      "train_runtime": 33607.0138,
      "train_tokens_per_second": 62568.463
    },
    {
      "epoch": 0.37687457226197835,
      "grad_norm": 34.75,
      "learning_rate": 1.432639680243827e-05,
      "loss": 0.7919,
      "num_input_tokens_seen": 2104818176,
      "step": 28360,
      "train_runtime": 33640.0131,
      "train_tokens_per_second": 62568.887
    },
    {
      "epoch": 0.3770074617444402,
      "grad_norm": 80.5,
      "learning_rate": 1.4322516090026808e-05,
      "loss": 0.8938,
      "num_input_tokens_seen": 2105653248,
      "step": 28370,
      "train_runtime": 33652.3519,
      "train_tokens_per_second": 62570.76
    },
    {
      "epoch": 0.37714035122690215,
      "grad_norm": 14.375,
      "learning_rate": 1.4318634576907786e-05,
      "loss": 0.7145,
      "num_input_tokens_seen": 2106117888,
      "step": 28380,
      "train_runtime": 33658.9304,
      "train_tokens_per_second": 62572.336
    },
    {
      "epoch": 0.3772732407093641,
      "grad_norm": 15.5625,
      "learning_rate": 1.4314752263800218e-05,
      "loss": 0.644,
      "num_input_tokens_seen": 2106341376,
      "step": 28390,
      "train_runtime": 33662.3526,
      "train_tokens_per_second": 62572.613
    },
    {
      "epoch": 0.37740613019182595,
      "grad_norm": 17.125,
      "learning_rate": 1.4310869151423266e-05,
      "loss": 0.4203,
      "num_input_tokens_seen": 2106451712,
      "step": 28400,
      "train_runtime": 33665.5502,
      "train_tokens_per_second": 62569.948
    },
    {
      "epoch": 0.3775390196742879,
      "grad_norm": 16.0,
      "learning_rate": 1.4306985240496245e-05,
      "loss": 0.7929,
      "num_input_tokens_seen": 2108481024,
      "step": 28410,
      "train_runtime": 33698.4313,
      "train_tokens_per_second": 62569.115
    },
    {
      "epoch": 0.3776719091567498,
      "grad_norm": 31.625,
      "learning_rate": 1.4303100531738613e-05,
      "loss": 0.8862,
      "num_input_tokens_seen": 2109306368,
      "step": 28420,
      "train_runtime": 33710.9979,
      "train_tokens_per_second": 62570.274
    },
    {
      "epoch": 0.3778047986392117,
      "grad_norm": 21.75,
      "learning_rate": 1.4299215025869977e-05,
      "loss": 0.7891,
      "num_input_tokens_seen": 2109742592,
      "step": 28430,
      "train_runtime": 33717.478,
      "train_tokens_per_second": 62571.186
    },
    {
      "epoch": 0.3779376881216736,
      "grad_norm": 31.0,
      "learning_rate": 1.4295328723610091e-05,
      "loss": 0.6408,
      "num_input_tokens_seen": 2109968640,
      "step": 28440,
      "train_runtime": 33720.9176,
      "train_tokens_per_second": 62571.507
    },
    {
      "epoch": 0.37807057760413554,
      "grad_norm": 15.4375,
      "learning_rate": 1.429144162567886e-05,
      "loss": 0.5299,
      "num_input_tokens_seen": 2110084608,
      "step": 28450,
      "train_runtime": 33723.56,
      "train_tokens_per_second": 62570.043
    },
    {
      "epoch": 0.3782034670865974,
      "grad_norm": 78.5,
      "learning_rate": 1.4287553732796336e-05,
      "loss": 0.7836,
      "num_input_tokens_seen": 2112152832,
      "step": 28460,
      "train_runtime": 33757.0511,
      "train_tokens_per_second": 62569.234
    },
    {
      "epoch": 0.37833635656905934,
      "grad_norm": 15.3125,
      "learning_rate": 1.4283665045682712e-05,
      "loss": 1.0297,
      "num_input_tokens_seen": 2113005568,
      "step": 28470,
      "train_runtime": 33771.0038,
      "train_tokens_per_second": 62568.634
    },
    {
      "epoch": 0.37846924605152127,
      "grad_norm": 20.25,
      "learning_rate": 1.4279775565058335e-05,
      "loss": 0.8036,
      "num_input_tokens_seen": 2113482496,
      "step": 28480,
      "train_runtime": 33777.9646,
      "train_tokens_per_second": 62569.859
    },
    {
      "epoch": 0.37860213553398314,
      "grad_norm": 21.875,
      "learning_rate": 1.4275885291643699e-05,
      "loss": 0.6576,
      "num_input_tokens_seen": 2113713920,
      "step": 28490,
      "train_runtime": 33781.5288,
      "train_tokens_per_second": 62570.108
    },
    {
      "epoch": 0.37873502501644507,
      "grad_norm": 10.875,
      "learning_rate": 1.4271994226159435e-05,
      "loss": 0.5284,
      "num_input_tokens_seen": 2113830912,
      "step": 28500,
      "train_runtime": 33784.0246,
      "train_tokens_per_second": 62568.949
    },
    {
      "epoch": 0.378867914498907,
      "grad_norm": 26.125,
      "learning_rate": 1.426810236932634e-05,
      "loss": 0.8375,
      "num_input_tokens_seen": 2115883520,
      "step": 28510,
      "train_runtime": 33816.6875,
      "train_tokens_per_second": 62569.213
    },
    {
      "epoch": 0.3790008039813689,
      "grad_norm": 49.75,
      "learning_rate": 1.4264209721865338e-05,
      "loss": 1.0009,
      "num_input_tokens_seen": 2116715520,
      "step": 28520,
      "train_runtime": 33829.7373,
      "train_tokens_per_second": 62569.671
    },
    {
      "epoch": 0.3791336934638308,
      "grad_norm": 22.375,
      "learning_rate": 1.4260316284497506e-05,
      "loss": 0.7076,
      "num_input_tokens_seen": 2117175040,
      "step": 28530,
      "train_runtime": 33836.1575,
      "train_tokens_per_second": 62571.379
    },
    {
      "epoch": 0.37926658294629273,
      "grad_norm": 12.0625,
      "learning_rate": 1.425642205794407e-05,
      "loss": 0.6719,
      "num_input_tokens_seen": 2117393408,
      "step": 28540,
      "train_runtime": 33839.5424,
      "train_tokens_per_second": 62571.573
    },
    {
      "epoch": 0.3793994724287546,
      "grad_norm": 8.875,
      "learning_rate": 1.4252527042926403e-05,
      "loss": 0.4938,
      "num_input_tokens_seen": 2117507328,
      "step": 28550,
      "train_runtime": 33842.1168,
      "train_tokens_per_second": 62570.18
    },
    {
      "epoch": 0.37953236191121653,
      "grad_norm": 12.8125,
      "learning_rate": 1.4248631240166022e-05,
      "loss": 0.8372,
      "num_input_tokens_seen": 2119580928,
      "step": 28560,
      "train_runtime": 33874.977,
      "train_tokens_per_second": 62570.697
    },
    {
      "epoch": 0.37966525139367846,
      "grad_norm": 29.25,
      "learning_rate": 1.4244734650384585e-05,
      "loss": 0.8863,
      "num_input_tokens_seen": 2120401664,
      "step": 28570,
      "train_runtime": 33887.4163,
      "train_tokens_per_second": 62571.948
    },
    {
      "epoch": 0.37979814087614033,
      "grad_norm": 21.625,
      "learning_rate": 1.4240837274303904e-05,
      "loss": 0.7281,
      "num_input_tokens_seen": 2120869376,
      "step": 28580,
      "train_runtime": 33893.9077,
      "train_tokens_per_second": 62573.764
    },
    {
      "epoch": 0.37993103035860226,
      "grad_norm": 25.875,
      "learning_rate": 1.4236939112645935e-05,
      "loss": 0.6989,
      "num_input_tokens_seen": 2121096448,
      "step": 28590,
      "train_runtime": 33898.1271,
      "train_tokens_per_second": 62572.674
    },
    {
      "epoch": 0.3800639198410642,
      "grad_norm": 9.0,
      "learning_rate": 1.4233040166132775e-05,
      "loss": 0.4927,
      "num_input_tokens_seen": 2121211136,
      "step": 28600,
      "train_runtime": 33900.8136,
      "train_tokens_per_second": 62571.098
    },
    {
      "epoch": 0.38019680932352606,
      "grad_norm": 17.875,
      "learning_rate": 1.4229140435486665e-05,
      "loss": 0.8417,
      "num_input_tokens_seen": 2123191296,
      "step": 28610,
      "train_runtime": 33932.9767,
      "train_tokens_per_second": 62570.146
    },
    {
      "epoch": 0.380329698805988,
      "grad_norm": 31.875,
      "learning_rate": 1.4225239921430008e-05,
      "loss": 0.9091,
      "num_input_tokens_seen": 2124031488,
      "step": 28620,
      "train_runtime": 33945.5093,
      "train_tokens_per_second": 62571.796
    },
    {
      "epoch": 0.3804625882884499,
      "grad_norm": 17.375,
      "learning_rate": 1.4221338624685326e-05,
      "loss": 0.7829,
      "num_input_tokens_seen": 2124468992,
      "step": 28630,
      "train_runtime": 33952.0974,
      "train_tokens_per_second": 62572.541
    },
    {
      "epoch": 0.3805954777709118,
      "grad_norm": 22.625,
      "learning_rate": 1.4217436545975305e-05,
      "loss": 0.6717,
      "num_input_tokens_seen": 2124700672,
      "step": 28640,
      "train_runtime": 33956.3956,
      "train_tokens_per_second": 62571.443
    },
    {
      "epoch": 0.3807283672533737,
      "grad_norm": 33.75,
      "learning_rate": 1.4213533686022769e-05,
      "loss": 0.543,
      "num_input_tokens_seen": 2124815872,
      "step": 28650,
      "train_runtime": 33958.9537,
      "train_tokens_per_second": 62570.122
    },
    {
      "epoch": 0.38086125673583565,
      "grad_norm": 18.875,
      "learning_rate": 1.4209630045550692e-05,
      "loss": 0.7631,
      "num_input_tokens_seen": 2126886144,
      "step": 28660,
      "train_runtime": 33994.0307,
      "train_tokens_per_second": 62566.46
    },
    {
      "epoch": 0.3809941462182976,
      "grad_norm": 59.25,
      "learning_rate": 1.4205725625282185e-05,
      "loss": 0.9551,
      "num_input_tokens_seen": 2127768064,
      "step": 28670,
      "train_runtime": 34007.0434,
      "train_tokens_per_second": 62568.452
    },
    {
      "epoch": 0.38112703570075945,
      "grad_norm": 14.4375,
      "learning_rate": 1.4201820425940507e-05,
      "loss": 0.7198,
      "num_input_tokens_seen": 2128273664,
      "step": 28680,
      "train_runtime": 34014.7125,
      "train_tokens_per_second": 62569.209
    },
    {
      "epoch": 0.3812599251832214,
      "grad_norm": 16.125,
      "learning_rate": 1.4197914448249063e-05,
      "loss": 0.6367,
      "num_input_tokens_seen": 2128515328,
      "step": 28690,
      "train_runtime": 34018.2727,
      "train_tokens_per_second": 62569.765
    },
    {
      "epoch": 0.3813928146656833,
      "grad_norm": 27.0,
      "learning_rate": 1.4194007692931397e-05,
      "loss": 0.55,
      "num_input_tokens_seen": 2128632320,
      "step": 28700,
      "train_runtime": 34020.7395,
      "train_tokens_per_second": 62568.667
    },
    {
      "epoch": 0.3815257041481452,
      "grad_norm": 46.75,
      "learning_rate": 1.4190100160711205e-05,
      "loss": 0.8449,
      "num_input_tokens_seen": 2130610944,
      "step": 28710,
      "train_runtime": 34052.877,
      "train_tokens_per_second": 62567.722
    },
    {
      "epoch": 0.3816585936306071,
      "grad_norm": 14.3125,
      "learning_rate": 1.4186191852312323e-05,
      "loss": 0.8495,
      "num_input_tokens_seen": 2131431424,
      "step": 28720,
      "train_runtime": 34065.385,
      "train_tokens_per_second": 62568.834
    },
    {
      "epoch": 0.38179148311306904,
      "grad_norm": 86.0,
      "learning_rate": 1.4182282768458724e-05,
      "loss": 0.717,
      "num_input_tokens_seen": 2131883520,
      "step": 28730,
      "train_runtime": 34071.8584,
      "train_tokens_per_second": 62570.215
    },
    {
      "epoch": 0.3819243725955309,
      "grad_norm": 22.625,
      "learning_rate": 1.4178372909874536e-05,
      "loss": 0.5333,
      "num_input_tokens_seen": 2132104704,
      "step": 28740,
      "train_runtime": 34075.3311,
      "train_tokens_per_second": 62570.33
    },
    {
      "epoch": 0.38205726207799284,
      "grad_norm": 13.9375,
      "learning_rate": 1.4174462277284025e-05,
      "loss": 0.5187,
      "num_input_tokens_seen": 2132221184,
      "step": 28750,
      "train_runtime": 34078.6233,
      "train_tokens_per_second": 62567.703
    },
    {
      "epoch": 0.38219015156045477,
      "grad_norm": 36.5,
      "learning_rate": 1.4170550871411598e-05,
      "loss": 0.8339,
      "num_input_tokens_seen": 2134323712,
      "step": 28760,
      "train_runtime": 34113.179,
      "train_tokens_per_second": 62565.958
    },
    {
      "epoch": 0.38232304104291664,
      "grad_norm": 37.25,
      "learning_rate": 1.4166638692981813e-05,
      "loss": 0.8659,
      "num_input_tokens_seen": 2135209728,
      "step": 28770,
      "train_runtime": 34126.3423,
      "train_tokens_per_second": 62567.787
    },
    {
      "epoch": 0.3824559305253786,
      "grad_norm": 26.75,
      "learning_rate": 1.4162725742719363e-05,
      "loss": 0.716,
      "num_input_tokens_seen": 2135696128,
      "step": 28780,
      "train_runtime": 34133.3525,
      "train_tokens_per_second": 62569.187
    },
    {
      "epoch": 0.3825888200078405,
      "grad_norm": 17.625,
      "learning_rate": 1.4158812021349086e-05,
      "loss": 0.6572,
      "num_input_tokens_seen": 2135923456,
      "step": 28790,
      "train_runtime": 34136.9537,
      "train_tokens_per_second": 62569.246
    },
    {
      "epoch": 0.3827217094903024,
      "grad_norm": 6.53125,
      "learning_rate": 1.4154897529595965e-05,
      "loss": 0.4955,
      "num_input_tokens_seen": 2136036352,
      "step": 28800,
      "train_runtime": 34140.3098,
      "train_tokens_per_second": 62566.402
    },
    {
      "epoch": 0.3828545989727643,
      "grad_norm": 14.0625,
      "learning_rate": 1.4150982268185127e-05,
      "loss": 0.8114,
      "num_input_tokens_seen": 2138067456,
      "step": 28810,
      "train_runtime": 34172.0631,
      "train_tokens_per_second": 62567.702
    },
    {
      "epoch": 0.38298748845522623,
      "grad_norm": 16.75,
      "learning_rate": 1.4147066237841838e-05,
      "loss": 0.8402,
      "num_input_tokens_seen": 2138925056,
      "step": 28820,
      "train_runtime": 34184.7652,
      "train_tokens_per_second": 62569.541
    },
    {
      "epoch": 0.3831203779376881,
      "grad_norm": 17.25,
      "learning_rate": 1.4143149439291508e-05,
      "loss": 0.6491,
      "num_input_tokens_seen": 2139360768,
      "step": 28830,
      "train_runtime": 34191.2573,
      "train_tokens_per_second": 62570.404
    },
    {
      "epoch": 0.38325326742015003,
      "grad_norm": 23.125,
      "learning_rate": 1.4139231873259686e-05,
      "loss": 0.5549,
      "num_input_tokens_seen": 2139584256,
      "step": 28840,
      "train_runtime": 34194.7615,
      "train_tokens_per_second": 62570.527
    },
    {
      "epoch": 0.38338615690261196,
      "grad_norm": 27.125,
      "learning_rate": 1.4135313540472068e-05,
      "loss": 0.5256,
      "num_input_tokens_seen": 2139695104,
      "step": 28850,
      "train_runtime": 34198.0812,
      "train_tokens_per_second": 62567.695
    },
    {
      "epoch": 0.38351904638507384,
      "grad_norm": 21.75,
      "learning_rate": 1.4131394441654492e-05,
      "loss": 0.7566,
      "num_input_tokens_seen": 2141804544,
      "step": 28860,
      "train_runtime": 34230.9584,
      "train_tokens_per_second": 62569.225
    },
    {
      "epoch": 0.38365193586753576,
      "grad_norm": 25.75,
      "learning_rate": 1.4127474577532932e-05,
      "loss": 0.9157,
      "num_input_tokens_seen": 2142651904,
      "step": 28870,
      "train_runtime": 34243.2712,
      "train_tokens_per_second": 62571.473
    },
    {
      "epoch": 0.3837848253499977,
      "grad_norm": 16.25,
      "learning_rate": 1.4123553948833513e-05,
      "loss": 0.6877,
      "num_input_tokens_seen": 2143094016,
      "step": 28880,
      "train_runtime": 34249.716,
      "train_tokens_per_second": 62572.607
    },
    {
      "epoch": 0.38391771483245957,
      "grad_norm": 14.4375,
      "learning_rate": 1.4119632556282495e-05,
      "loss": 0.6173,
      "num_input_tokens_seen": 2143319040,
      "step": 28890,
      "train_runtime": 34253.3014,
      "train_tokens_per_second": 62572.627
    },
    {
      "epoch": 0.3840506043149215,
      "grad_norm": 16.75,
      "learning_rate": 1.4115710400606276e-05,
      "loss": 0.4246,
      "num_input_tokens_seen": 2143431936,
      "step": 28900,
      "train_runtime": 34255.9185,
      "train_tokens_per_second": 62571.142
    },
    {
      "epoch": 0.3841834937973834,
      "grad_norm": 11.875,
      "learning_rate": 1.4111787482531408e-05,
      "loss": 0.7697,
      "num_input_tokens_seen": 2145504256,
      "step": 28910,
      "train_runtime": 34290.7888,
      "train_tokens_per_second": 62567.947
    },
    {
      "epoch": 0.3843163832798453,
      "grad_norm": 19.875,
      "learning_rate": 1.4107863802784571e-05,
      "loss": 0.8902,
      "num_input_tokens_seen": 2146348800,
      "step": 28920,
      "train_runtime": 34303.8085,
      "train_tokens_per_second": 62568.819
    },
    {
      "epoch": 0.3844492727623072,
      "grad_norm": 19.75,
      "learning_rate": 1.4103939362092594e-05,
      "loss": 0.8133,
      "num_input_tokens_seen": 2146841856,
      "step": 28930,
      "train_runtime": 34310.8646,
      "train_tokens_per_second": 62570.322
    },
    {
      "epoch": 0.38458216224476915,
      "grad_norm": 9.3125,
      "learning_rate": 1.4100014161182447e-05,
      "loss": 0.6836,
      "num_input_tokens_seen": 2147074560,
      "step": 28940,
      "train_runtime": 34314.4518,
      "train_tokens_per_second": 62570.563
    },
    {
      "epoch": 0.384715051727231,
      "grad_norm": 20.0,
      "learning_rate": 1.4096088200781233e-05,
      "loss": 0.5767,
      "num_input_tokens_seen": 2147190784,
      "step": 28950,
      "train_runtime": 34317.171,
      "train_tokens_per_second": 62568.992
    },
    {
      "epoch": 0.38484794120969296,
      "grad_norm": 58.25,
      "learning_rate": 1.4092161481616202e-05,
      "loss": 0.8245,
      "num_input_tokens_seen": 2149209600,
      "step": 28960,
      "train_runtime": 34349.2721,
      "train_tokens_per_second": 62569.291
    },
    {
      "epoch": 0.3849808306921549,
      "grad_norm": 17.5,
      "learning_rate": 1.4088234004414749e-05,
      "loss": 0.9177,
      "num_input_tokens_seen": 2150042368,
      "step": 28970,
      "train_runtime": 34361.7514,
      "train_tokens_per_second": 62570.803
    },
    {
      "epoch": 0.38511372017461676,
      "grad_norm": 14.875,
      "learning_rate": 1.40843057699044e-05,
      "loss": 0.6527,
      "num_input_tokens_seen": 2150502912,
      "step": 28980,
      "train_runtime": 34368.1762,
      "train_tokens_per_second": 62572.506
    },
    {
      "epoch": 0.3852466096570787,
      "grad_norm": 13.125,
      "learning_rate": 1.4080376778812828e-05,
      "loss": 0.6612,
      "num_input_tokens_seen": 2150716416,
      "step": 28990,
      "train_runtime": 34371.7049,
      "train_tokens_per_second": 62572.294
    },
    {
      "epoch": 0.3853794991395406,
      "grad_norm": 23.75,
      "learning_rate": 1.4076447031867839e-05,
      "loss": 0.5784,
      "num_input_tokens_seen": 2150828800,
      "step": 29000,
      "train_runtime": 34374.4785,
      "train_tokens_per_second": 62570.514
    },
    {
      "epoch": 0.3855123886220025,
      "grad_norm": 25.75,
      "learning_rate": 1.4072516529797387e-05,
      "loss": 0.8413,
      "num_input_tokens_seen": 2152916992,
      "step": 29010,
      "train_runtime": 34407.8414,
      "train_tokens_per_second": 62570.533
    },
    {
      "epoch": 0.3856452781044644,
      "grad_norm": 40.0,
      "learning_rate": 1.4068585273329563e-05,
      "loss": 0.9018,
      "num_input_tokens_seen": 2153784320,
      "step": 29020,
      "train_runtime": 34420.6294,
      "train_tokens_per_second": 62572.485
    },
    {
      "epoch": 0.38577816758692635,
      "grad_norm": 22.625,
      "learning_rate": 1.4064653263192594e-05,
      "loss": 0.7436,
      "num_input_tokens_seen": 2154264320,
      "step": 29030,
      "train_runtime": 34427.4018,
      "train_tokens_per_second": 62574.118
    },
    {
      "epoch": 0.3859110570693882,
      "grad_norm": 10.9375,
      "learning_rate": 1.4060720500114853e-05,
      "loss": 0.6178,
      "num_input_tokens_seen": 2154498816,
      "step": 29040,
      "train_runtime": 34431.7754,
      "train_tokens_per_second": 62572.981
    },
    {
      "epoch": 0.38604394655185015,
      "grad_norm": 73.0,
      "learning_rate": 1.405678698482485e-05,
      "loss": 0.5685,
      "num_input_tokens_seen": 2154614272,
      "step": 29050,
      "train_runtime": 34434.3475,
      "train_tokens_per_second": 62571.66
    },
    {
      "epoch": 0.3861768360343121,
      "grad_norm": 22.25,
      "learning_rate": 1.4052852718051226e-05,
      "loss": 0.851,
      "num_input_tokens_seen": 2156647168,
      "step": 29060,
      "train_runtime": 34467.5358,
      "train_tokens_per_second": 62570.39
    },
    {
      "epoch": 0.38630972551677395,
      "grad_norm": 15.3125,
      "learning_rate": 1.4048917700522782e-05,
      "loss": 0.8425,
      "num_input_tokens_seen": 2157523456,
      "step": 29070,
      "train_runtime": 34480.2523,
      "train_tokens_per_second": 62572.728
    },
    {
      "epoch": 0.3864426149992359,
      "grad_norm": 25.25,
      "learning_rate": 1.4044981932968433e-05,
      "loss": 0.7621,
      "num_input_tokens_seen": 2157989376,
      "step": 29080,
      "train_runtime": 34486.9126,
      "train_tokens_per_second": 62574.154
    },
    {
      "epoch": 0.3865755044816978,
      "grad_norm": 32.75,
      "learning_rate": 1.404104541611725e-05,
      "loss": 0.6371,
      "num_input_tokens_seen": 2158214656,
      "step": 29090,
      "train_runtime": 34490.4126,
      "train_tokens_per_second": 62574.336
    },
    {
      "epoch": 0.3867083939641597,
      "grad_norm": 25.0,
      "learning_rate": 1.4037108150698437e-05,
      "loss": 0.4905,
      "num_input_tokens_seen": 2158329600,
      "step": 29100,
      "train_runtime": 34493.812,
      "train_tokens_per_second": 62571.501
    },
    {
      "epoch": 0.3868412834466216,
      "grad_norm": 18.125,
      "learning_rate": 1.4033170137441339e-05,
      "loss": 0.8791,
      "num_input_tokens_seen": 2160230144,
      "step": 29110,
      "train_runtime": 34525.5504,
      "train_tokens_per_second": 62569.028
    },
    {
      "epoch": 0.38697417292908354,
      "grad_norm": 15.0,
      "learning_rate": 1.4029231377075432e-05,
      "loss": 0.9145,
      "num_input_tokens_seen": 2161050368,
      "step": 29120,
      "train_runtime": 34538.0617,
      "train_tokens_per_second": 62570.111
    },
    {
      "epoch": 0.3871070624115454,
      "grad_norm": 35.5,
      "learning_rate": 1.4025291870330344e-05,
      "loss": 0.7705,
      "num_input_tokens_seen": 2161503744,
      "step": 29130,
      "train_runtime": 34544.6862,
      "train_tokens_per_second": 62571.237
    },
    {
      "epoch": 0.38723995189400734,
      "grad_norm": 16.75,
      "learning_rate": 1.402135161793583e-05,
      "loss": 0.6025,
      "num_input_tokens_seen": 2161737216,
      "step": 29140,
      "train_runtime": 34548.2745,
      "train_tokens_per_second": 62571.496
    },
    {
      "epoch": 0.38737284137646927,
      "grad_norm": 16.0,
      "learning_rate": 1.4017410620621787e-05,
      "loss": 0.5838,
      "num_input_tokens_seen": 2161850880,
      "step": 29150,
      "train_runtime": 34552.0568,
      "train_tokens_per_second": 62567.936
    },
    {
      "epoch": 0.3875057308589312,
      "grad_norm": 24.5,
      "learning_rate": 1.4013468879118248e-05,
      "loss": 0.8495,
      "num_input_tokens_seen": 2163946752,
      "step": 29160,
      "train_runtime": 34585.3912,
      "train_tokens_per_second": 62568.231
    },
    {
      "epoch": 0.38763862034139307,
      "grad_norm": 25.25,
      "learning_rate": 1.4009526394155386e-05,
      "loss": 0.8682,
      "num_input_tokens_seen": 2164828672,
      "step": 29170,
      "train_runtime": 34598.424,
      "train_tokens_per_second": 62570.153
    },
    {
      "epoch": 0.387771509823855,
      "grad_norm": 19.625,
      "learning_rate": 1.400558316646351e-05,
      "loss": 0.6477,
      "num_input_tokens_seen": 2165320960,
      "step": 29180,
      "train_runtime": 34605.3023,
      "train_tokens_per_second": 62571.942
    },
    {
      "epoch": 0.3879043993063169,
      "grad_norm": 14.75,
      "learning_rate": 1.4001639196773072e-05,
      "loss": 0.6161,
      "num_input_tokens_seen": 2165554688,
      "step": 29190,
      "train_runtime": 34609.6374,
      "train_tokens_per_second": 62570.857
    },
    {
      "epoch": 0.3880372887887788,
      "grad_norm": 23.75,
      "learning_rate": 1.3997694485814654e-05,
      "loss": 0.4941,
      "num_input_tokens_seen": 2165670656,
      "step": 29200,
      "train_runtime": 34612.2496,
      "train_tokens_per_second": 62569.486
    },
    {
      "epoch": 0.38817017827124073,
      "grad_norm": 15.1875,
      "learning_rate": 1.399374903431898e-05,
      "loss": 0.8362,
      "num_input_tokens_seen": 2167752448,
      "step": 29210,
      "train_runtime": 34645.0638,
      "train_tokens_per_second": 62570.312
    },
    {
      "epoch": 0.38830306775370266,
      "grad_norm": 12.5625,
      "learning_rate": 1.3989802843016908e-05,
      "loss": 0.8882,
      "num_input_tokens_seen": 2168585472,
      "step": 29220,
      "train_runtime": 34657.6818,
      "train_tokens_per_second": 62571.567
    },
    {
      "epoch": 0.38843595723616453,
      "grad_norm": 25.625,
      "learning_rate": 1.3985855912639438e-05,
      "loss": 0.7027,
      "num_input_tokens_seen": 2169030400,
      "step": 29230,
      "train_runtime": 34664.0706,
      "train_tokens_per_second": 62572.87
    },
    {
      "epoch": 0.38856884671862646,
      "grad_norm": 10.0,
      "learning_rate": 1.3981908243917699e-05,
      "loss": 0.6413,
      "num_input_tokens_seen": 2169253888,
      "step": 29240,
      "train_runtime": 34667.4865,
      "train_tokens_per_second": 62573.152
    },
    {
      "epoch": 0.3887017362010884,
      "grad_norm": 8.75,
      "learning_rate": 1.3977959837582967e-05,
      "loss": 0.5252,
      "num_input_tokens_seen": 2169368064,
      "step": 29250,
      "train_runtime": 34670.1046,
      "train_tokens_per_second": 62571.72
    },
    {
      "epoch": 0.38883462568355026,
      "grad_norm": 23.5,
      "learning_rate": 1.3974010694366642e-05,
      "loss": 0.9098,
      "num_input_tokens_seen": 2171408640,
      "step": 29260,
      "train_runtime": 34703.9042,
      "train_tokens_per_second": 62569.578
    },
    {
      "epoch": 0.3889675151660122,
      "grad_norm": 11.6875,
      "learning_rate": 1.3970060815000273e-05,
      "loss": 0.8646,
      "num_input_tokens_seen": 2172234496,
      "step": 29270,
      "train_runtime": 34716.5502,
      "train_tokens_per_second": 62570.575
    },
    {
      "epoch": 0.3891004046484741,
      "grad_norm": 22.375,
      "learning_rate": 1.3966110200215537e-05,
      "loss": 0.6543,
      "num_input_tokens_seen": 2172719616,
      "step": 29280,
      "train_runtime": 34723.2966,
      "train_tokens_per_second": 62572.389
    },
    {
      "epoch": 0.389233294130936,
      "grad_norm": 16.0,
      "learning_rate": 1.3962158850744253e-05,
      "loss": 0.7135,
      "num_input_tokens_seen": 2172955136,
      "step": 29290,
      "train_runtime": 34726.8672,
      "train_tokens_per_second": 62572.737
    },
    {
      "epoch": 0.3893661836133979,
      "grad_norm": 10.8125,
      "learning_rate": 1.3958206767318373e-05,
      "loss": 0.5408,
      "num_input_tokens_seen": 2173069312,
      "step": 29300,
      "train_runtime": 34729.3461,
      "train_tokens_per_second": 62571.558
    },
    {
      "epoch": 0.38949907309585985,
      "grad_norm": 15.1875,
      "learning_rate": 1.3954253950669982e-05,
      "loss": 0.8176,
      "num_input_tokens_seen": 2174999808,
      "step": 29310,
      "train_runtime": 34760.7608,
      "train_tokens_per_second": 62570.547
    },
    {
      "epoch": 0.3896319625783217,
      "grad_norm": 20.375,
      "learning_rate": 1.395030040153131e-05,
      "loss": 0.8831,
      "num_input_tokens_seen": 2175838976,
      "step": 29320,
      "train_runtime": 34773.2809,
      "train_tokens_per_second": 62572.151
    },
    {
      "epoch": 0.38976485206078365,
      "grad_norm": 22.625,
      "learning_rate": 1.3946346120634707e-05,
      "loss": 0.7588,
      "num_input_tokens_seen": 2176320768,
      "step": 29330,
      "train_runtime": 34779.9665,
      "train_tokens_per_second": 62573.975
    },
    {
      "epoch": 0.3898977415432456,
      "grad_norm": 13.4375,
      "learning_rate": 1.3942391108712677e-05,
      "loss": 0.6574,
      "num_input_tokens_seen": 2176551168,
      "step": 29340,
      "train_runtime": 34783.5407,
      "train_tokens_per_second": 62574.169
    },
    {
      "epoch": 0.39003063102570745,
      "grad_norm": 14.3125,
      "learning_rate": 1.3938435366497847e-05,
      "loss": 0.5548,
      "num_input_tokens_seen": 2176666880,
      "step": 29350,
      "train_runtime": 34786.087,
      "train_tokens_per_second": 62572.915
    },
    {
      "epoch": 0.3901635205081694,
      "grad_norm": 19.75,
      "learning_rate": 1.3934478894722984e-05,
      "loss": 0.8481,
      "num_input_tokens_seen": 2178668544,
      "step": 29360,
      "train_runtime": 34818.9521,
      "train_tokens_per_second": 62571.342
    },
    {
      "epoch": 0.3902964099906313,
      "grad_norm": 28.125,
      "learning_rate": 1.3930521694120988e-05,
      "loss": 0.8977,
      "num_input_tokens_seen": 2179489024,
      "step": 29370,
      "train_runtime": 34831.3975,
      "train_tokens_per_second": 62572.54
    },
    {
      "epoch": 0.3904292994730932,
      "grad_norm": 24.0,
      "learning_rate": 1.39265637654249e-05,
      "loss": 0.7663,
      "num_input_tokens_seen": 2179959040,
      "step": 29380,
      "train_runtime": 34837.9774,
      "train_tokens_per_second": 62574.214
    },
    {
      "epoch": 0.3905621889555551,
      "grad_norm": 26.25,
      "learning_rate": 1.3922605109367884e-05,
      "loss": 0.7155,
      "num_input_tokens_seen": 2180184320,
      "step": 29390,
      "train_runtime": 34841.4335,
      "train_tokens_per_second": 62574.472
    },
    {
      "epoch": 0.39069507843801704,
      "grad_norm": 35.0,
      "learning_rate": 1.3918645726683246e-05,
      "loss": 0.5149,
      "num_input_tokens_seen": 2180297216,
      "step": 29400,
      "train_runtime": 34843.9356,
      "train_tokens_per_second": 62573.219
    },
    {
      "epoch": 0.3908279679204789,
      "grad_norm": 62.5,
      "learning_rate": 1.3914685618104432e-05,
      "loss": 0.8307,
      "num_input_tokens_seen": 2182412288,
      "step": 29410,
      "train_runtime": 34877.8405,
      "train_tokens_per_second": 62573.034
    },
    {
      "epoch": 0.39096085740294084,
      "grad_norm": 13.875,
      "learning_rate": 1.3910724784365014e-05,
      "loss": 0.8646,
      "num_input_tokens_seen": 2183276544,
      "step": 29420,
      "train_runtime": 34890.6026,
      "train_tokens_per_second": 62574.916
    },
    {
      "epoch": 0.39109374688540277,
      "grad_norm": 16.875,
      "learning_rate": 1.3906763226198701e-05,
      "loss": 0.7302,
      "num_input_tokens_seen": 2183727104,
      "step": 29430,
      "train_runtime": 34898.0492,
      "train_tokens_per_second": 62574.475
    },
    {
      "epoch": 0.39122663636786464,
      "grad_norm": 30.625,
      "learning_rate": 1.3902800944339337e-05,
      "loss": 0.6184,
      "num_input_tokens_seen": 2183951872,
      "step": 29440,
      "train_runtime": 34901.5561,
      "train_tokens_per_second": 62574.627
    },
    {
      "epoch": 0.3913595258503266,
      "grad_norm": 10.4375,
      "learning_rate": 1.3898837939520894e-05,
      "loss": 0.4754,
      "num_input_tokens_seen": 2184062720,
      "step": 29450,
      "train_runtime": 34904.1139,
      "train_tokens_per_second": 62573.218
    },
    {
      "epoch": 0.3914924153327885,
      "grad_norm": 18.5,
      "learning_rate": 1.3894874212477497e-05,
      "loss": 0.7928,
      "num_input_tokens_seen": 2186066432,
      "step": 29460,
      "train_runtime": 34936.281,
      "train_tokens_per_second": 62572.958
    },
    {
      "epoch": 0.3916253048152504,
      "grad_norm": 16.5,
      "learning_rate": 1.3890909763943376e-05,
      "loss": 0.9119,
      "num_input_tokens_seen": 2186916352,
      "step": 29470,
      "train_runtime": 34949.0482,
      "train_tokens_per_second": 62574.418
    },
    {
      "epoch": 0.3917581942977123,
      "grad_norm": 11.5,
      "learning_rate": 1.3886944594652916e-05,
      "loss": 0.7127,
      "num_input_tokens_seen": 2187379200,
      "step": 29480,
      "train_runtime": 34956.4206,
      "train_tokens_per_second": 62574.462
    },
    {
      "epoch": 0.39189108378017423,
      "grad_norm": 10.75,
      "learning_rate": 1.388297870534063e-05,
      "loss": 0.6196,
      "num_input_tokens_seen": 2187613184,
      "step": 29490,
      "train_runtime": 34960.0021,
      "train_tokens_per_second": 62574.744
    },
    {
      "epoch": 0.3920239732626361,
      "grad_norm": 9.75,
      "learning_rate": 1.3879012096741161e-05,
      "loss": 0.5256,
      "num_input_tokens_seen": 2187727872,
      "step": 29500,
      "train_runtime": 34962.6313,
      "train_tokens_per_second": 62573.319
    },
    {
      "epoch": 0.39215686274509803,
      "grad_norm": 15.0625,
      "learning_rate": 1.3875044769589291e-05,
      "loss": 0.8227,
      "num_input_tokens_seen": 2189811968,
      "step": 29510,
      "train_runtime": 34996.0415,
      "train_tokens_per_second": 62573.133
    },
    {
      "epoch": 0.39228975222755996,
      "grad_norm": 13.875,
      "learning_rate": 1.3871076724619931e-05,
      "loss": 0.8794,
      "num_input_tokens_seen": 2190633984,
      "step": 29520,
      "train_runtime": 35008.6208,
      "train_tokens_per_second": 62574.13
    },
    {
      "epoch": 0.39242264171002184,
      "grad_norm": 40.0,
      "learning_rate": 1.3867107962568125e-05,
      "loss": 0.7424,
      "num_input_tokens_seen": 2191103232,
      "step": 29530,
      "train_runtime": 35016.1019,
      "train_tokens_per_second": 62574.162
    },
    {
      "epoch": 0.39255553119248376,
      "grad_norm": 13.25,
      "learning_rate": 1.386313848416905e-05,
      "loss": 0.6752,
      "num_input_tokens_seen": 2191340544,
      "step": 29540,
      "train_runtime": 35019.709,
      "train_tokens_per_second": 62574.493
    },
    {
      "epoch": 0.3926884206749457,
      "grad_norm": 8.75,
      "learning_rate": 1.3859168290158021e-05,
      "loss": 0.5527,
      "num_input_tokens_seen": 2191455744,
      "step": 29550,
      "train_runtime": 35022.2626,
      "train_tokens_per_second": 62573.22
    },
    {
      "epoch": 0.39282131015740757,
      "grad_norm": 14.875,
      "learning_rate": 1.3855197381270476e-05,
      "loss": 0.8166,
      "num_input_tokens_seen": 2193496064,
      "step": 29560,
      "train_runtime": 35054.851,
      "train_tokens_per_second": 62573.253
    },
    {
      "epoch": 0.3929541996398695,
      "grad_norm": 26.625,
      "learning_rate": 1.3851225758241992e-05,
      "loss": 0.9336,
      "num_input_tokens_seen": 2194386688,
      "step": 29570,
      "train_runtime": 35068.128,
      "train_tokens_per_second": 62574.959
    },
    {
      "epoch": 0.3930870891223314,
      "grad_norm": 142.0,
      "learning_rate": 1.3847253421808278e-05,
      "loss": 0.7659,
      "num_input_tokens_seen": 2194893824,
      "step": 29580,
      "train_runtime": 35075.3851,
      "train_tokens_per_second": 62576.471
    },
    {
      "epoch": 0.3932199786047933,
      "grad_norm": 17.375,
      "learning_rate": 1.384328037270517e-05,
      "loss": 0.6599,
      "num_input_tokens_seen": 2195140608,
      "step": 29590,
      "train_runtime": 35079.1329,
      "train_tokens_per_second": 62576.821
    },
    {
      "epoch": 0.3933528680872552,
      "grad_norm": 29.0,
      "learning_rate": 1.3839306611668645e-05,
      "loss": 0.5325,
      "num_input_tokens_seen": 2195262464,
      "step": 29600,
      "train_runtime": 35082.5593,
      "train_tokens_per_second": 62574.182
    },
    {
      "epoch": 0.39348575756971715,
      "grad_norm": 27.0,
      "learning_rate": 1.3835332139434805e-05,
      "loss": 0.8518,
      "num_input_tokens_seen": 2197208576,
      "step": 29610,
      "train_runtime": 35114.2368,
      "train_tokens_per_second": 62573.155
    },
    {
      "epoch": 0.3936186470521791,
      "grad_norm": 13.375,
      "learning_rate": 1.3831356956739885e-05,
      "loss": 0.8336,
      "num_input_tokens_seen": 2198078464,
      "step": 29620,
      "train_runtime": 35127.0446,
      "train_tokens_per_second": 62575.104
    },
    {
      "epoch": 0.39375153653464096,
      "grad_norm": 17.125,
      "learning_rate": 1.3827381064320254e-05,
      "loss": 0.7379,
      "num_input_tokens_seen": 2198549248,
      "step": 29630,
      "train_runtime": 35133.7898,
      "train_tokens_per_second": 62576.49
    },
    {
      "epoch": 0.3938844260171029,
      "grad_norm": 24.125,
      "learning_rate": 1.3823404462912406e-05,
      "loss": 0.6321,
      "num_input_tokens_seen": 2198784768,
      "step": 29640,
      "train_runtime": 35138.1404,
      "train_tokens_per_second": 62575.445
    },
    {
      "epoch": 0.3940173154995648,
      "grad_norm": 7.1875,
      "learning_rate": 1.3819427153252976e-05,
      "loss": 0.4625,
      "num_input_tokens_seen": 2198897920,
      "step": 29650,
      "train_runtime": 35140.7892,
      "train_tokens_per_second": 62573.948
    },
    {
      "epoch": 0.3941502049820267,
      "grad_norm": 28.625,
      "learning_rate": 1.381544913607872e-05,
      "loss": 0.8053,
      "num_input_tokens_seen": 2200932352,
      "step": 29660,
      "train_runtime": 35174.4881,
      "train_tokens_per_second": 62571.837
    },
    {
      "epoch": 0.3942830944644886,
      "grad_norm": 13.9375,
      "learning_rate": 1.3811470412126538e-05,
      "loss": 0.9198,
      "num_input_tokens_seen": 2201768192,
      "step": 29670,
      "train_runtime": 35186.8076,
      "train_tokens_per_second": 62573.684
    },
    {
      "epoch": 0.39441598394695054,
      "grad_norm": 15.5625,
      "learning_rate": 1.3807490982133444e-05,
      "loss": 0.7717,
      "num_input_tokens_seen": 2202240512,
      "step": 29680,
      "train_runtime": 35193.3531,
      "train_tokens_per_second": 62575.467
    },
    {
      "epoch": 0.3945488734294124,
      "grad_norm": 26.125,
      "learning_rate": 1.3803510846836598e-05,
      "loss": 0.6353,
      "num_input_tokens_seen": 2202462464,
      "step": 29690,
      "train_runtime": 35196.8677,
      "train_tokens_per_second": 62575.525
    },
    {
      "epoch": 0.39468176291187435,
      "grad_norm": 11.625,
      "learning_rate": 1.3799530006973283e-05,
      "loss": 0.5449,
      "num_input_tokens_seen": 2202575872,
      "step": 29700,
      "train_runtime": 35199.5897,
      "train_tokens_per_second": 62573.908
    },
    {
      "epoch": 0.3948146523943363,
      "grad_norm": 20.875,
      "learning_rate": 1.3795548463280916e-05,
      "loss": 0.7902,
      "num_input_tokens_seen": 2204581888,
      "step": 29710,
      "train_runtime": 35231.3671,
      "train_tokens_per_second": 62574.406
    },
    {
      "epoch": 0.39494754187679815,
      "grad_norm": 95.5,
      "learning_rate": 1.379156621649704e-05,
      "loss": 0.8968,
      "num_input_tokens_seen": 2205411840,
      "step": 29720,
      "train_runtime": 35244.5822,
      "train_tokens_per_second": 62574.492
    },
    {
      "epoch": 0.3950804313592601,
      "grad_norm": 20.375,
      "learning_rate": 1.378758326735933e-05,
      "loss": 0.7073,
      "num_input_tokens_seen": 2205903872,
      "step": 29730,
      "train_runtime": 35251.5464,
      "train_tokens_per_second": 62576.088
    },
    {
      "epoch": 0.395213320841722,
      "grad_norm": 13.5,
      "learning_rate": 1.3783599616605591e-05,
      "loss": 0.65,
      "num_input_tokens_seen": 2206136064,
      "step": 29740,
      "train_runtime": 35255.1281,
      "train_tokens_per_second": 62576.317
    },
    {
      "epoch": 0.3953462103241839,
      "grad_norm": 26.0,
      "learning_rate": 1.3779615264973763e-05,
      "loss": 0.4714,
      "num_input_tokens_seen": 2206251776,
      "step": 29750,
      "train_runtime": 35257.657,
      "train_tokens_per_second": 62575.11
    },
    {
      "epoch": 0.3954790998066458,
      "grad_norm": 14.9375,
      "learning_rate": 1.3775630213201907e-05,
      "loss": 0.8159,
      "num_input_tokens_seen": 2208271104,
      "step": 29760,
      "train_runtime": 35290.6613,
      "train_tokens_per_second": 62573.809
    },
    {
      "epoch": 0.39561198928910773,
      "grad_norm": 11.25,
      "learning_rate": 1.3771644462028225e-05,
      "loss": 0.8351,
      "num_input_tokens_seen": 2209091840,
      "step": 29770,
      "train_runtime": 35303.4565,
      "train_tokens_per_second": 62574.378
    },
    {
      "epoch": 0.3957448787715696,
      "grad_norm": 25.875,
      "learning_rate": 1.3767658012191034e-05,
      "loss": 0.793,
      "num_input_tokens_seen": 2209579776,
      "step": 29780,
      "train_runtime": 35311.0217,
      "train_tokens_per_second": 62574.79
    },
    {
      "epoch": 0.39587776825403154,
      "grad_norm": 13.3125,
      "learning_rate": 1.3763670864428793e-05,
      "loss": 0.573,
      "num_input_tokens_seen": 2209808128,
      "step": 29790,
      "train_runtime": 35314.5752,
      "train_tokens_per_second": 62574.96
    },
    {
      "epoch": 0.39601065773649347,
      "grad_norm": 16.625,
      "learning_rate": 1.3759683019480084e-05,
      "loss": 0.4677,
      "num_input_tokens_seen": 2209921792,
      "step": 29800,
      "train_runtime": 35317.1602,
      "train_tokens_per_second": 62573.598
    },
    {
      "epoch": 0.39614354721895534,
      "grad_norm": 17.75,
      "learning_rate": 1.3755694478083622e-05,
      "loss": 0.7646,
      "num_input_tokens_seen": 2211869184,
      "step": 29810,
      "train_runtime": 35349.035,
      "train_tokens_per_second": 62572.265
    },
    {
      "epoch": 0.39627643670141727,
      "grad_norm": 15.9375,
      "learning_rate": 1.3751705240978244e-05,
      "loss": 0.8764,
      "num_input_tokens_seen": 2212690176,
      "step": 29820,
      "train_runtime": 35361.63,
      "train_tokens_per_second": 62573.195
    },
    {
      "epoch": 0.3964093261838792,
      "grad_norm": 16.625,
      "learning_rate": 1.3747715308902924e-05,
      "loss": 0.7671,
      "num_input_tokens_seen": 2213180160,
      "step": 29830,
      "train_runtime": 35369.245,
      "train_tokens_per_second": 62573.577
    },
    {
      "epoch": 0.39654221566634107,
      "grad_norm": 16.0,
      "learning_rate": 1.3743724682596762e-05,
      "loss": 0.5554,
      "num_input_tokens_seen": 2213408768,
      "step": 29840,
      "train_runtime": 35372.7631,
      "train_tokens_per_second": 62573.816
    },
    {
      "epoch": 0.396675105148803,
      "grad_norm": 15.4375,
      "learning_rate": 1.3739733362798984e-05,
      "loss": 0.6153,
      "num_input_tokens_seen": 2213522176,
      "step": 29850,
      "train_runtime": 35375.3236,
      "train_tokens_per_second": 62572.493
    },
    {
      "epoch": 0.3968079946312649,
      "grad_norm": 23.625,
      "learning_rate": 1.3735741350248945e-05,
      "loss": 0.8111,
      "num_input_tokens_seen": 2215604224,
      "step": 29860,
      "train_runtime": 35409.5197,
      "train_tokens_per_second": 62570.864
    },
    {
      "epoch": 0.3969408841137268,
      "grad_norm": 40.75,
      "learning_rate": 1.3731748645686139e-05,
      "loss": 1.0069,
      "num_input_tokens_seen": 2216448256,
      "step": 29870,
      "train_runtime": 35422.2844,
      "train_tokens_per_second": 62572.143
    },
    {
      "epoch": 0.39707377359618873,
      "grad_norm": 14.0625,
      "learning_rate": 1.3727755249850163e-05,
      "loss": 0.742,
      "num_input_tokens_seen": 2216909824,
      "step": 29880,
      "train_runtime": 35428.9402,
      "train_tokens_per_second": 62573.416
    },
    {
      "epoch": 0.39720666307865066,
      "grad_norm": 17.0,
      "learning_rate": 1.3723761163480773e-05,
      "loss": 0.5809,
      "num_input_tokens_seen": 2217136128,
      "step": 29890,
      "train_runtime": 35433.1141,
      "train_tokens_per_second": 62572.432
    },
    {
      "epoch": 0.39733955256111253,
      "grad_norm": 35.25,
      "learning_rate": 1.3719766387317828e-05,
      "loss": 0.515,
      "num_input_tokens_seen": 2217249536,
      "step": 29900,
      "train_runtime": 35435.6418,
      "train_tokens_per_second": 62571.169
    },
    {
      "epoch": 0.39747244204357446,
      "grad_norm": 32.25,
      "learning_rate": 1.371577092210133e-05,
      "loss": 0.8775,
      "num_input_tokens_seen": 2219368192,
      "step": 29910,
      "train_runtime": 35469.707,
      "train_tokens_per_second": 62570.807
    },
    {
      "epoch": 0.3976053315260364,
      "grad_norm": 17.875,
      "learning_rate": 1.37117747685714e-05,
      "loss": 0.8796,
      "num_input_tokens_seen": 2220272896,
      "step": 29920,
      "train_runtime": 35482.6525,
      "train_tokens_per_second": 62573.476
    },
    {
      "epoch": 0.39773822100849826,
      "grad_norm": 10.8125,
      "learning_rate": 1.3707777927468295e-05,
      "loss": 0.7261,
      "num_input_tokens_seen": 2220755968,
      "step": 29930,
      "train_runtime": 35489.5915,
      "train_tokens_per_second": 62574.853
    },
    {
      "epoch": 0.3978711104909602,
      "grad_norm": 14.4375,
      "learning_rate": 1.3703780399532388e-05,
      "loss": 0.6152,
      "num_input_tokens_seen": 2220996352,
      "step": 29940,
      "train_runtime": 35493.9423,
      "train_tokens_per_second": 62573.955
    },
    {
      "epoch": 0.3980039999734221,
      "grad_norm": 18.25,
      "learning_rate": 1.369978218550419e-05,
      "loss": 0.5169,
      "num_input_tokens_seen": 2221109760,
      "step": 29950,
      "train_runtime": 35496.5353,
      "train_tokens_per_second": 62572.579
    },
    {
      "epoch": 0.398136889455884,
      "grad_norm": 25.375,
      "learning_rate": 1.3695783286124328e-05,
      "loss": 0.8235,
      "num_input_tokens_seen": 2223281664,
      "step": 29960,
      "train_runtime": 35530.7714,
      "train_tokens_per_second": 62573.414
    },
    {
      "epoch": 0.3982697789383459,
      "grad_norm": 15.3125,
      "learning_rate": 1.3691783702133569e-05,
      "loss": 0.9416,
      "num_input_tokens_seen": 2224151808,
      "step": 29970,
      "train_runtime": 35543.4536,
      "train_tokens_per_second": 62575.568
    },
    {
      "epoch": 0.39840266842080785,
      "grad_norm": 21.75,
      "learning_rate": 1.3687783434272799e-05,
      "loss": 0.6441,
      "num_input_tokens_seen": 2224611072,
      "step": 29980,
      "train_runtime": 35550.8982,
      "train_tokens_per_second": 62575.383
    },
    {
      "epoch": 0.3985355579032697,
      "grad_norm": 14.6875,
      "learning_rate": 1.3683782483283032e-05,
      "loss": 0.6868,
      "num_input_tokens_seen": 2224841472,
      "step": 29990,
      "train_runtime": 35554.4607,
      "train_tokens_per_second": 62575.593
    },
    {
      "epoch": 0.39866844738573165,
      "grad_norm": 48.25,
      "learning_rate": 1.3679780849905406e-05,
      "loss": 0.5089,
      "num_input_tokens_seen": 2224954624,
      "step": 30000,
      "train_runtime": 35556.9962,
      "train_tokens_per_second": 62574.313
    },
    {
      "epoch": 0.39866844738573165,
      "eval_accuracy": 0.82,
      "eval_balanced_accuracy": 0.8183229247297883,
      "eval_invalid_f1": 0.8375451263537906,
      "eval_invalid_precision": 0.8421052631578947,
      "eval_invalid_recall": 0.8330341113105925,
      "eval_loss": 0.3634049892425537,
      "eval_macro_f1": 0.8178757021903482,
      "eval_mcc": 0.6358115352434813,
      "eval_runtime": 84.5952,
      "eval_samples_per_second": 118.21,
      "eval_steps_per_second": 29.553,
      "eval_valid_f1": 0.7982062780269058,
      "eval_valid_precision": 0.7928730512249443,
      "eval_valid_recall": 0.8036117381489842,
      "num_input_tokens_seen": 2224954624,
      "step": 30000
    },
    {
      "epoch": 0.3988013368681936,
      "grad_norm": 23.625,
      "learning_rate": 1.367577853488119e-05,
      "loss": 0.8121,
      "num_input_tokens_seen": 2227021056,
      "step": 30010,
      "train_runtime": 35679.2519,
      "train_tokens_per_second": 62417.818
    },
    {
      "epoch": 0.39893422635065545,
      "grad_norm": 29.0,
      "learning_rate": 1.3671775538951777e-05,
      "loss": 0.8602,
      "num_input_tokens_seen": 2227850496,
      "step": 30020,
      "train_runtime": 35691.8671,
      "train_tokens_per_second": 62418.996
    },
    {
      "epoch": 0.3990671158331174,
      "grad_norm": 23.0,
      "learning_rate": 1.3667771862858688e-05,
      "loss": 0.7294,
      "num_input_tokens_seen": 2228346624,
      "step": 30030,
      "train_runtime": 35698.7744,
      "train_tokens_per_second": 62420.816
    },
    {
      "epoch": 0.3992000053155793,
      "grad_norm": 21.5,
      "learning_rate": 1.3663767507343565e-05,
      "loss": 0.5959,
      "num_input_tokens_seen": 2228582912,
      "step": 30040,
      "train_runtime": 35703.2537,
      "train_tokens_per_second": 62419.603
    },
    {
      "epoch": 0.3993328947980412,
      "grad_norm": 8.375,
      "learning_rate": 1.365976247314818e-05,
      "loss": 0.5458,
      "num_input_tokens_seen": 2228699648,
      "step": 30050,
      "train_runtime": 35705.8607,
      "train_tokens_per_second": 62418.315
    },
    {
      "epoch": 0.3994657842805031,
      "grad_norm": 29.5,
      "learning_rate": 1.3655756761014427e-05,
      "loss": 0.8034,
      "num_input_tokens_seen": 2230824960,
      "step": 30060,
      "train_runtime": 35739.1178,
      "train_tokens_per_second": 62419.699
    },
    {
      "epoch": 0.39959867376296504,
      "grad_norm": 28.625,
      "learning_rate": 1.3651750371684336e-05,
      "loss": 0.8334,
      "num_input_tokens_seen": 2231688704,
      "step": 30070,
      "train_runtime": 35752.471,
      "train_tokens_per_second": 62420.544
    },
    {
      "epoch": 0.3997315632454269,
      "grad_norm": 21.75,
      "learning_rate": 1.364774330590005e-05,
      "loss": 0.7605,
      "num_input_tokens_seen": 2232189696,
      "step": 30080,
      "train_runtime": 35760.44,
      "train_tokens_per_second": 62420.644
    },
    {
      "epoch": 0.39986445272788884,
      "grad_norm": 18.125,
      "learning_rate": 1.3643735564403842e-05,
      "loss": 0.6233,
      "num_input_tokens_seen": 2232438272,
      "step": 30090,
      "train_runtime": 35764.1417,
      "train_tokens_per_second": 62421.134
    },
    {
      "epoch": 0.39999734221035077,
      "grad_norm": 16.625,
      "learning_rate": 1.363972714793811e-05,
      "loss": 0.5479,
      "num_input_tokens_seen": 2232555264,
      "step": 30100,
      "train_runtime": 35766.7013,
      "train_tokens_per_second": 62419.938
    },
    {
      "epoch": 0.4001302316928127,
      "grad_norm": 24.375,
      "learning_rate": 1.3635718057245382e-05,
      "loss": 0.8302,
      "num_input_tokens_seen": 2234628608,
      "step": 30110,
      "train_runtime": 35799.714,
      "train_tokens_per_second": 62420.292
    },
    {
      "epoch": 0.40026312117527457,
      "grad_norm": 17.25,
      "learning_rate": 1.3631708293068303e-05,
      "loss": 0.9071,
      "num_input_tokens_seen": 2235442688,
      "step": 30120,
      "train_runtime": 35812.0576,
      "train_tokens_per_second": 62421.509
    },
    {
      "epoch": 0.4003960106577365,
      "grad_norm": 26.125,
      "learning_rate": 1.3627697856149644e-05,
      "loss": 0.6584,
      "num_input_tokens_seen": 2235915520,
      "step": 30130,
      "train_runtime": 35818.5635,
      "train_tokens_per_second": 62423.372
    },
    {
      "epoch": 0.40052890014019843,
      "grad_norm": 13.1875,
      "learning_rate": 1.3623686747232303e-05,
      "loss": 0.6561,
      "num_input_tokens_seen": 2236146176,
      "step": 30140,
      "train_runtime": 35822.1127,
      "train_tokens_per_second": 62423.626
    },
    {
      "epoch": 0.4006617896226603,
      "grad_norm": 9.625,
      "learning_rate": 1.3619674967059308e-05,
      "loss": 0.4898,
      "num_input_tokens_seen": 2236264448,
      "step": 30150,
      "train_runtime": 35824.7779,
      "train_tokens_per_second": 62422.284
    },
    {
      "epoch": 0.40079467910512223,
      "grad_norm": 17.5,
      "learning_rate": 1.36156625163738e-05,
      "loss": 0.8075,
      "num_input_tokens_seen": 2238325760,
      "step": 30160,
      "train_runtime": 35858.8824,
      "train_tokens_per_second": 62420.399
    },
    {
      "epoch": 0.40092756858758416,
      "grad_norm": 30.25,
      "learning_rate": 1.3611649395919049e-05,
      "loss": 0.9306,
      "num_input_tokens_seen": 2239159040,
      "step": 30170,
      "train_runtime": 35871.6298,
      "train_tokens_per_second": 62421.447
    },
    {
      "epoch": 0.40106045807004603,
      "grad_norm": 13.75,
      "learning_rate": 1.3607635606438454e-05,
      "loss": 0.7499,
      "num_input_tokens_seen": 2239643904,
      "step": 30180,
      "train_runtime": 35878.4544,
      "train_tokens_per_second": 62423.088
    },
    {
      "epoch": 0.40119334755250796,
      "grad_norm": 32.25,
      "learning_rate": 1.3603621148675531e-05,
      "loss": 0.6574,
      "num_input_tokens_seen": 2239874560,
      "step": 30190,
      "train_runtime": 35882.0077,
      "train_tokens_per_second": 62423.334
    },
    {
      "epoch": 0.4013262370349699,
      "grad_norm": 9.0625,
      "learning_rate": 1.3599606023373926e-05,
      "loss": 0.4267,
      "num_input_tokens_seen": 2239989760,
      "step": 30200,
      "train_runtime": 35884.6016,
      "train_tokens_per_second": 62422.032
    },
    {
      "epoch": 0.40145912651743176,
      "grad_norm": 16.5,
      "learning_rate": 1.3595590231277398e-05,
      "loss": 0.7819,
      "num_input_tokens_seen": 2241988096,
      "step": 30210,
      "train_runtime": 35916.7081,
      "train_tokens_per_second": 62421.87
    },
    {
      "epoch": 0.4015920159998937,
      "grad_norm": 31.625,
      "learning_rate": 1.3591573773129843e-05,
      "loss": 0.9452,
      "num_input_tokens_seen": 2242825984,
      "step": 30220,
      "train_runtime": 35929.9257,
      "train_tokens_per_second": 62422.227
    },
    {
      "epoch": 0.4017249054823556,
      "grad_norm": 18.375,
      "learning_rate": 1.3587556649675272e-05,
      "loss": 0.681,
      "num_input_tokens_seen": 2243299840,
      "step": 30230,
      "train_runtime": 35936.6664,
      "train_tokens_per_second": 62423.704
    },
    {
      "epoch": 0.4018577949648175,
      "grad_norm": 11.3125,
      "learning_rate": 1.3583538861657818e-05,
      "loss": 0.665,
      "num_input_tokens_seen": 2243527936,
      "step": 30240,
      "train_runtime": 35940.1778,
      "train_tokens_per_second": 62423.952
    },
    {
      "epoch": 0.4019906844472794,
      "grad_norm": 11.875,
      "learning_rate": 1.3579520409821749e-05,
      "loss": 0.4926,
      "num_input_tokens_seen": 2243645184,
      "step": 30250,
      "train_runtime": 35942.8214,
      "train_tokens_per_second": 62422.623
    },
    {
      "epoch": 0.40212357392974135,
      "grad_norm": 27.0,
      "learning_rate": 1.357550129491144e-05,
      "loss": 0.8431,
      "num_input_tokens_seen": 2245793792,
      "step": 30260,
      "train_runtime": 35978.7978,
      "train_tokens_per_second": 62419.923
    },
    {
      "epoch": 0.4022564634122032,
      "grad_norm": 50.75,
      "learning_rate": 1.3571481517671396e-05,
      "loss": 0.8682,
      "num_input_tokens_seen": 2246675712,
      "step": 30270,
      "train_runtime": 35992.7451,
      "train_tokens_per_second": 62420.238
    },
    {
      "epoch": 0.40238935289466515,
      "grad_norm": 19.375,
      "learning_rate": 1.3567461078846252e-05,
      "loss": 0.7124,
      "num_input_tokens_seen": 2247158784,
      "step": 30280,
      "train_runtime": 35999.7656,
      "train_tokens_per_second": 62421.484
    },
    {
      "epoch": 0.4025222423771271,
      "grad_norm": 30.75,
      "learning_rate": 1.3563439979180747e-05,
      "loss": 0.6542,
      "num_input_tokens_seen": 2247397376,
      "step": 30290,
      "train_runtime": 36003.3859,
      "train_tokens_per_second": 62421.834
    },
    {
      "epoch": 0.40265513185958896,
      "grad_norm": 11.75,
      "learning_rate": 1.3559418219419767e-05,
      "loss": 0.5231,
      "num_input_tokens_seen": 2247513344,
      "step": 30300,
      "train_runtime": 36005.9777,
      "train_tokens_per_second": 62420.561
    },
    {
      "epoch": 0.4027880213420509,
      "grad_norm": 27.5,
      "learning_rate": 1.3555395800308298e-05,
      "loss": 0.8143,
      "num_input_tokens_seen": 2249513984,
      "step": 30310,
      "train_runtime": 36039.3991,
      "train_tokens_per_second": 62418.188
    },
    {
      "epoch": 0.4029209108245128,
      "grad_norm": 13.0,
      "learning_rate": 1.3551372722591458e-05,
      "loss": 0.864,
      "num_input_tokens_seen": 2250356224,
      "step": 30320,
      "train_runtime": 36052.2091,
      "train_tokens_per_second": 62419.371
    },
    {
      "epoch": 0.4030538003069747,
      "grad_norm": 22.625,
      "learning_rate": 1.354734898701449e-05,
      "loss": 0.7213,
      "num_input_tokens_seen": 2250852352,
      "step": 30330,
      "train_runtime": 36059.2085,
      "train_tokens_per_second": 62421.014
    },
    {
      "epoch": 0.4031866897894366,
      "grad_norm": 15.9375,
      "learning_rate": 1.3543324594322751e-05,
      "loss": 0.6308,
      "num_input_tokens_seen": 2251086080,
      "step": 30340,
      "train_runtime": 36062.8209,
      "train_tokens_per_second": 62421.242
    },
    {
      "epoch": 0.40331957927189854,
      "grad_norm": 9.0,
      "learning_rate": 1.353929954526173e-05,
      "loss": 0.6171,
      "num_input_tokens_seen": 2251201024,
      "step": 30350,
      "train_runtime": 36065.4553,
      "train_tokens_per_second": 62419.87
    },
    {
      "epoch": 0.4034524687543604,
      "grad_norm": 20.125,
      "learning_rate": 1.3535273840577028e-05,
      "loss": 0.883,
      "num_input_tokens_seen": 2253328384,
      "step": 30360,
      "train_runtime": 36099.0088,
      "train_tokens_per_second": 62420.783
    },
    {
      "epoch": 0.40358535823682234,
      "grad_norm": 21.25,
      "learning_rate": 1.3531247481014366e-05,
      "loss": 0.9787,
      "num_input_tokens_seen": 2254161152,
      "step": 30370,
      "train_runtime": 36111.6651,
      "train_tokens_per_second": 62421.967
    },
    {
      "epoch": 0.4037182477192843,
      "grad_norm": 19.0,
      "learning_rate": 1.3527220467319598e-05,
      "loss": 0.7198,
      "num_input_tokens_seen": 2254637312,
      "step": 30380,
      "train_runtime": 36118.3971,
      "train_tokens_per_second": 62423.515
    },
    {
      "epoch": 0.40385113720174615,
      "grad_norm": 23.25,
      "learning_rate": 1.352319280023869e-05,
      "loss": 0.6065,
      "num_input_tokens_seen": 2254872320,
      "step": 30390,
      "train_runtime": 36122.6179,
      "train_tokens_per_second": 62422.727
    },
    {
      "epoch": 0.4039840266842081,
      "grad_norm": 13.8125,
      "learning_rate": 1.351916448051773e-05,
      "loss": 0.5725,
      "num_input_tokens_seen": 2254987264,
      "step": 30400,
      "train_runtime": 36125.2141,
      "train_tokens_per_second": 62421.423
    },
    {
      "epoch": 0.40411691616667,
      "grad_norm": 11.8125,
      "learning_rate": 1.3515135508902932e-05,
      "loss": 0.838,
      "num_input_tokens_seen": 2256894720,
      "step": 30410,
      "train_runtime": 36156.8513,
      "train_tokens_per_second": 62419.559
    },
    {
      "epoch": 0.4042498056491319,
      "grad_norm": 16.75,
      "learning_rate": 1.3511105886140623e-05,
      "loss": 0.8491,
      "num_input_tokens_seen": 2257705984,
      "step": 30420,
      "train_runtime": 36168.9853,
      "train_tokens_per_second": 62421.048
    },
    {
      "epoch": 0.4043826951315938,
      "grad_norm": 82.0,
      "learning_rate": 1.3507075612977259e-05,
      "loss": 0.7623,
      "num_input_tokens_seen": 2258160896,
      "step": 30430,
      "train_runtime": 36175.4153,
      "train_tokens_per_second": 62422.529
    },
    {
      "epoch": 0.40451558461405573,
      "grad_norm": 11.5,
      "learning_rate": 1.3503044690159408e-05,
      "loss": 0.6237,
      "num_input_tokens_seen": 2258385664,
      "step": 30440,
      "train_runtime": 36179.6014,
      "train_tokens_per_second": 62421.519
    },
    {
      "epoch": 0.4046484740965176,
      "grad_norm": 8.5625,
      "learning_rate": 1.3499013118433766e-05,
      "loss": 0.6046,
      "num_input_tokens_seen": 2258499840,
      "step": 30450,
      "train_runtime": 36182.2457,
      "train_tokens_per_second": 62420.112
    },
    {
      "epoch": 0.40478136357897954,
      "grad_norm": 60.75,
      "learning_rate": 1.3494980898547144e-05,
      "loss": 0.7785,
      "num_input_tokens_seen": 2260566016,
      "step": 30460,
      "train_runtime": 36215.9027,
      "train_tokens_per_second": 62419.154
    },
    {
      "epoch": 0.40491425306144146,
      "grad_norm": 18.625,
      "learning_rate": 1.349094803124647e-05,
      "loss": 0.8411,
      "num_input_tokens_seen": 2261390336,
      "step": 30470,
      "train_runtime": 36228.6063,
      "train_tokens_per_second": 62420.02
    },
    {
      "epoch": 0.40504714254390334,
      "grad_norm": 29.875,
      "learning_rate": 1.3486914517278806e-05,
      "loss": 0.8127,
      "num_input_tokens_seen": 2261864704,
      "step": 30480,
      "train_runtime": 36236.0343,
      "train_tokens_per_second": 62420.316
    },
    {
      "epoch": 0.40518003202636527,
      "grad_norm": 20.375,
      "learning_rate": 1.3482880357391316e-05,
      "loss": 0.6741,
      "num_input_tokens_seen": 2262095616,
      "step": 30490,
      "train_runtime": 36239.6763,
      "train_tokens_per_second": 62420.415
    },
    {
      "epoch": 0.4053129215088272,
      "grad_norm": 15.125,
      "learning_rate": 1.34788455523313e-05,
      "loss": 0.4291,
      "num_input_tokens_seen": 2262213888,
      "step": 30500,
      "train_runtime": 36242.3235,
      "train_tokens_per_second": 62419.119
    },
    {
      "epoch": 0.40544581099128907,
      "grad_norm": 12.8125,
      "learning_rate": 1.3474810102846165e-05,
      "loss": 0.9227,
      "num_input_tokens_seen": 2264334592,
      "step": 30510,
      "train_runtime": 36276.6005,
      "train_tokens_per_second": 62418.599
    },
    {
      "epoch": 0.405578700473751,
      "grad_norm": 11.5625,
      "learning_rate": 1.3470774009683445e-05,
      "loss": 0.8661,
      "num_input_tokens_seen": 2265198080,
      "step": 30520,
      "train_runtime": 36289.4795,
      "train_tokens_per_second": 62420.242
    },
    {
      "epoch": 0.4057115899562129,
      "grad_norm": 21.875,
      "learning_rate": 1.3466737273590784e-05,
      "loss": 0.8138,
      "num_input_tokens_seen": 2265683968,
      "step": 30530,
      "train_runtime": 36296.2822,
      "train_tokens_per_second": 62421.929
    },
    {
      "epoch": 0.4058444794386748,
      "grad_norm": 12.6875,
      "learning_rate": 1.3462699895315956e-05,
      "loss": 0.5538,
      "num_input_tokens_seen": 2265905664,
      "step": 30540,
      "train_runtime": 36299.7873,
      "train_tokens_per_second": 62422.009
    },
    {
      "epoch": 0.4059773689211367,
      "grad_norm": 9.5625,
      "learning_rate": 1.3458661875606849e-05,
      "loss": 0.5168,
      "num_input_tokens_seen": 2266021376,
      "step": 30550,
      "train_runtime": 36302.2976,
      "train_tokens_per_second": 62420.88
    },
    {
      "epoch": 0.40611025840359866,
      "grad_norm": 34.25,
      "learning_rate": 1.3454623215211471e-05,
      "loss": 0.8839,
      "num_input_tokens_seen": 2268111616,
      "step": 30560,
      "train_runtime": 36335.3684,
      "train_tokens_per_second": 62421.594
    },
    {
      "epoch": 0.4062431478860606,
      "grad_norm": 21.125,
      "learning_rate": 1.3450583914877947e-05,
      "loss": 0.872,
      "num_input_tokens_seen": 2268985856,
      "step": 30570,
      "train_runtime": 36348.4224,
      "train_tokens_per_second": 62423.228
    },
    {
      "epoch": 0.40637603736852246,
      "grad_norm": 18.25,
      "learning_rate": 1.3446543975354519e-05,
      "loss": 0.6738,
      "num_input_tokens_seen": 2269453056,
      "step": 30580,
      "train_runtime": 36355.271,
      "train_tokens_per_second": 62424.32
    },
    {
      "epoch": 0.4065089268509844,
      "grad_norm": 12.5625,
      "learning_rate": 1.344250339738955e-05,
      "loss": 0.6052,
      "num_input_tokens_seen": 2269682432,
      "step": 30590,
      "train_runtime": 36358.7769,
      "train_tokens_per_second": 62424.609
    },
    {
      "epoch": 0.4066418163334463,
      "grad_norm": 19.375,
      "learning_rate": 1.3438462181731527e-05,
      "loss": 0.5912,
      "num_input_tokens_seen": 2269795072,
      "step": 30600,
      "train_runtime": 36361.386,
      "train_tokens_per_second": 62423.228
    },
    {
      "epoch": 0.4067747058159082,
      "grad_norm": 15.25,
      "learning_rate": 1.3434420329129044e-05,
      "loss": 0.7838,
      "num_input_tokens_seen": 2271801088,
      "step": 30610,
      "train_runtime": 36394.1521,
      "train_tokens_per_second": 62422.146
    },
    {
      "epoch": 0.4069075952983701,
      "grad_norm": 13.3125,
      "learning_rate": 1.343037784033082e-05,
      "loss": 0.8958,
      "num_input_tokens_seen": 2272654336,
      "step": 30620,
      "train_runtime": 36406.7742,
      "train_tokens_per_second": 62423.941
    },
    {
      "epoch": 0.40704048478083205,
      "grad_norm": 27.75,
      "learning_rate": 1.3426334716085687e-05,
      "loss": 0.7554,
      "num_input_tokens_seen": 2273121792,
      "step": 30630,
      "train_runtime": 36413.3675,
      "train_tokens_per_second": 62425.476
    },
    {
      "epoch": 0.4071733742632939,
      "grad_norm": 27.125,
      "learning_rate": 1.34222909571426e-05,
      "loss": 0.6056,
      "num_input_tokens_seen": 2273350656,
      "step": 30640,
      "train_runtime": 36416.872,
      "train_tokens_per_second": 62425.753
    },
    {
      "epoch": 0.40730626374575585,
      "grad_norm": 6.40625,
      "learning_rate": 1.3418246564250629e-05,
      "loss": 0.549,
      "num_input_tokens_seen": 2273463808,
      "step": 30650,
      "train_runtime": 36419.4099,
      "train_tokens_per_second": 62424.51
    },
    {
      "epoch": 0.4074391532282178,
      "grad_norm": 24.875,
      "learning_rate": 1.3414201538158965e-05,
      "loss": 0.8705,
      "num_input_tokens_seen": 2275463424,
      "step": 30660,
      "train_runtime": 36453.2154,
      "train_tokens_per_second": 62421.474
    },
    {
      "epoch": 0.40757204271067965,
      "grad_norm": 20.75,
      "learning_rate": 1.3410155879616906e-05,
      "loss": 0.9116,
      "num_input_tokens_seen": 2276355072,
      "step": 30670,
      "train_runtime": 36466.2983,
      "train_tokens_per_second": 62423.53
    },
    {
      "epoch": 0.4077049321931416,
      "grad_norm": 19.125,
      "learning_rate": 1.3406109589373883e-05,
      "loss": 0.754,
      "num_input_tokens_seen": 2276831488,
      "step": 30680,
      "train_runtime": 36473.1312,
      "train_tokens_per_second": 62424.898
    },
    {
      "epoch": 0.4078378216756035,
      "grad_norm": 16.75,
      "learning_rate": 1.3402062668179427e-05,
      "loss": 0.5527,
      "num_input_tokens_seen": 2277057280,
      "step": 30690,
      "train_runtime": 36476.5498,
      "train_tokens_per_second": 62425.237
    },
    {
      "epoch": 0.4079707111580654,
      "grad_norm": 11.9375,
      "learning_rate": 1.3398015116783196e-05,
      "loss": 0.5866,
      "num_input_tokens_seen": 2277169920,
      "step": 30700,
      "train_runtime": 36479.1837,
      "train_tokens_per_second": 62423.818
    },
    {
      "epoch": 0.4081036006405273,
      "grad_norm": 18.75,
      "learning_rate": 1.3393966935934965e-05,
      "loss": 0.894,
      "num_input_tokens_seen": 2279200512,
      "step": 30710,
      "train_runtime": 36512.6053,
      "train_tokens_per_second": 62422.292
    },
    {
      "epoch": 0.40823649012298924,
      "grad_norm": 15.5,
      "learning_rate": 1.3389918126384625e-05,
      "loss": 0.8713,
      "num_input_tokens_seen": 2280059648,
      "step": 30720,
      "train_runtime": 36525.2694,
      "train_tokens_per_second": 62424.171
    },
    {
      "epoch": 0.4083693796054511,
      "grad_norm": 12.3125,
      "learning_rate": 1.3385868688882178e-05,
      "loss": 0.698,
      "num_input_tokens_seen": 2280568576,
      "step": 30730,
      "train_runtime": 36532.4895,
      "train_tokens_per_second": 62425.764
    },
    {
      "epoch": 0.40850226908791304,
      "grad_norm": 12.0,
      "learning_rate": 1.3381818624177747e-05,
      "loss": 0.6999,
      "num_input_tokens_seen": 2280827904,
      "step": 30740,
      "train_runtime": 36536.2867,
      "train_tokens_per_second": 62426.374
    },
    {
      "epoch": 0.40863515857037497,
      "grad_norm": 10.8125,
      "learning_rate": 1.3377767933021571e-05,
      "loss": 0.5104,
      "num_input_tokens_seen": 2280946432,
      "step": 30750,
      "train_runtime": 36538.7008,
      "train_tokens_per_second": 62425.494
    },
    {
      "epoch": 0.40876804805283684,
      "grad_norm": 16.75,
      "learning_rate": 1.3373716616164005e-05,
      "loss": 0.8183,
      "num_input_tokens_seen": 2283050496,
      "step": 30760,
      "train_runtime": 36573.1132,
      "train_tokens_per_second": 62424.286
    },
    {
      "epoch": 0.40890093753529877,
      "grad_norm": 29.0,
      "learning_rate": 1.3369664674355518e-05,
      "loss": 0.8293,
      "num_input_tokens_seen": 2283944960,
      "step": 30770,
      "train_runtime": 36586.1085,
      "train_tokens_per_second": 62426.562
    },
    {
      "epoch": 0.4090338270177607,
      "grad_norm": 23.5,
      "learning_rate": 1.3365612108346699e-05,
      "loss": 0.78,
      "num_input_tokens_seen": 2284428288,
      "step": 30780,
      "train_runtime": 36592.928,
      "train_tokens_per_second": 62428.136
    },
    {
      "epoch": 0.40916671650022257,
      "grad_norm": 15.375,
      "learning_rate": 1.3361558918888247e-05,
      "loss": 0.6581,
      "num_input_tokens_seen": 2284653056,
      "step": 30790,
      "train_runtime": 36597.2173,
      "train_tokens_per_second": 62426.961
    },
    {
      "epoch": 0.4092996059826845,
      "grad_norm": 17.125,
      "learning_rate": 1.3357505106730977e-05,
      "loss": 0.5387,
      "num_input_tokens_seen": 2284770560,
      "step": 30800,
      "train_runtime": 36599.6537,
      "train_tokens_per_second": 62426.016
    },
    {
      "epoch": 0.40943249546514643,
      "grad_norm": 15.5,
      "learning_rate": 1.3353450672625824e-05,
      "loss": 0.7665,
      "num_input_tokens_seen": 2286788352,
      "step": 30810,
      "train_runtime": 36631.5523,
      "train_tokens_per_second": 62426.739
    },
    {
      "epoch": 0.4095653849476083,
      "grad_norm": 28.625,
      "learning_rate": 1.3349395617323842e-05,
      "loss": 0.864,
      "num_input_tokens_seen": 2287620608,
      "step": 30820,
      "train_runtime": 36644.0465,
      "train_tokens_per_second": 62428.166
    },
    {
      "epoch": 0.40969827443007023,
      "grad_norm": 15.875,
      "learning_rate": 1.3345339941576184e-05,
      "loss": 0.789,
      "num_input_tokens_seen": 2288098560,
      "step": 30830,
      "train_runtime": 36650.6051,
      "train_tokens_per_second": 62430.035
    },
    {
      "epoch": 0.40983116391253216,
      "grad_norm": 32.5,
      "learning_rate": 1.3341283646134131e-05,
      "loss": 0.6127,
      "num_input_tokens_seen": 2288322560,
      "step": 30840,
      "train_runtime": 36654.66,
      "train_tokens_per_second": 62429.24
    },
    {
      "epoch": 0.40996405339499403,
      "grad_norm": 37.25,
      "learning_rate": 1.3337226731749077e-05,
      "loss": 0.4802,
      "num_input_tokens_seen": 2288438272,
      "step": 30850,
      "train_runtime": 36657.1065,
      "train_tokens_per_second": 62428.23
    },
    {
      "epoch": 0.41009694287745596,
      "grad_norm": 17.0,
      "learning_rate": 1.3333169199172526e-05,
      "loss": 0.8561,
      "num_input_tokens_seen": 2290430720,
      "step": 30860,
      "train_runtime": 36689.8668,
      "train_tokens_per_second": 62426.793
    },
    {
      "epoch": 0.4102298323599179,
      "grad_norm": 12.75,
      "learning_rate": 1.3329111049156105e-05,
      "loss": 0.8854,
      "num_input_tokens_seen": 2291279616,
      "step": 30870,
      "train_runtime": 36702.6845,
      "train_tokens_per_second": 62428.121
    },
    {
      "epoch": 0.41036272184237976,
      "grad_norm": 12.1875,
      "learning_rate": 1.3325052282451547e-05,
      "loss": 0.7488,
      "num_input_tokens_seen": 2291745024,
      "step": 30880,
      "train_runtime": 36710.1384,
      "train_tokens_per_second": 62428.123
    },
    {
      "epoch": 0.4104956113248417,
      "grad_norm": 15.0625,
      "learning_rate": 1.3320992899810703e-05,
      "loss": 0.6558,
      "num_input_tokens_seen": 2291979008,
      "step": 30890,
      "train_runtime": 36713.6838,
      "train_tokens_per_second": 62428.467
    },
    {
      "epoch": 0.4106285008073036,
      "grad_norm": 10.5625,
      "learning_rate": 1.3316932901985535e-05,
      "loss": 0.5863,
      "num_input_tokens_seen": 2292097280,
      "step": 30900,
      "train_runtime": 36716.419,
      "train_tokens_per_second": 62427.038
    },
    {
      "epoch": 0.4107613902897655,
      "grad_norm": 23.125,
      "learning_rate": 1.3312872289728125e-05,
      "loss": 0.8509,
      "num_input_tokens_seen": 2294157056,
      "step": 30910,
      "train_runtime": 36750.1079,
      "train_tokens_per_second": 62425.859
    },
    {
      "epoch": 0.4108942797722274,
      "grad_norm": 24.0,
      "learning_rate": 1.3308811063790665e-05,
      "loss": 0.8886,
      "num_input_tokens_seen": 2294989824,
      "step": 30920,
      "train_runtime": 36762.6467,
      "train_tokens_per_second": 62427.22
    },
    {
      "epoch": 0.41102716925468935,
      "grad_norm": 16.0,
      "learning_rate": 1.3304749224925458e-05,
      "loss": 0.7521,
      "num_input_tokens_seen": 2295459584,
      "step": 30930,
      "train_runtime": 36769.3752,
      "train_tokens_per_second": 62428.572
    },
    {
      "epoch": 0.4111600587371512,
      "grad_norm": 12.9375,
      "learning_rate": 1.3300686773884925e-05,
      "loss": 0.6328,
      "num_input_tokens_seen": 2295690752,
      "step": 30940,
      "train_runtime": 36772.8498,
      "train_tokens_per_second": 62428.96
    },
    {
      "epoch": 0.41129294821961315,
      "grad_norm": 35.25,
      "learning_rate": 1.3296623711421599e-05,
      "loss": 0.5677,
      "num_input_tokens_seen": 2295809024,
      "step": 30950,
      "train_runtime": 36776.1263,
      "train_tokens_per_second": 62426.614
    },
    {
      "epoch": 0.4114258377020751,
      "grad_norm": 18.875,
      "learning_rate": 1.3292560038288125e-05,
      "loss": 0.8434,
      "num_input_tokens_seen": 2297889792,
      "step": 30960,
      "train_runtime": 36808.6028,
      "train_tokens_per_second": 62428.063
    },
    {
      "epoch": 0.41155872718453695,
      "grad_norm": 14.5625,
      "learning_rate": 1.3288495755237266e-05,
      "loss": 0.9062,
      "num_input_tokens_seen": 2298709248,
      "step": 30970,
      "train_runtime": 36821.0351,
      "train_tokens_per_second": 62429.24
    },
    {
      "epoch": 0.4116916166669989,
      "grad_norm": 22.0,
      "learning_rate": 1.3284430863021888e-05,
      "loss": 0.7234,
      "num_input_tokens_seen": 2299190272,
      "step": 30980,
      "train_runtime": 36827.748,
      "train_tokens_per_second": 62430.922
    },
    {
      "epoch": 0.4118245061494608,
      "grad_norm": 15.375,
      "learning_rate": 1.328036536239498e-05,
      "loss": 0.655,
      "num_input_tokens_seen": 2299412224,
      "step": 30990,
      "train_runtime": 36831.3343,
      "train_tokens_per_second": 62430.869
    },
    {
      "epoch": 0.4119573956319227,
      "grad_norm": 19.0,
      "learning_rate": 1.3276299254109641e-05,
      "loss": 0.4699,
      "num_input_tokens_seen": 2299522048,
      "step": 31000,
      "train_runtime": 36834.8512,
      "train_tokens_per_second": 62427.89
    },
    {
      "epoch": 0.4120902851143846,
      "grad_norm": 24.5,
      "learning_rate": 1.3272232538919079e-05,
      "loss": 0.8127,
      "num_input_tokens_seen": 2301657856,
      "step": 31010,
      "train_runtime": 36868.5911,
      "train_tokens_per_second": 62428.69
    },
    {
      "epoch": 0.41222317459684654,
      "grad_norm": 16.5,
      "learning_rate": 1.3268165217576613e-05,
      "loss": 0.8674,
      "num_input_tokens_seen": 2302509056,
      "step": 31020,
      "train_runtime": 36881.0597,
      "train_tokens_per_second": 62430.664
    },
    {
      "epoch": 0.41235606407930847,
      "grad_norm": 13.25,
      "learning_rate": 1.3264097290835687e-05,
      "loss": 0.7474,
      "num_input_tokens_seen": 2302969344,
      "step": 31030,
      "train_runtime": 36887.4374,
      "train_tokens_per_second": 62432.348
    },
    {
      "epoch": 0.41248895356177034,
      "grad_norm": 13.5,
      "learning_rate": 1.3260028759449844e-05,
      "loss": 0.6788,
      "num_input_tokens_seen": 2303190784,
      "step": 31040,
      "train_runtime": 36890.8007,
      "train_tokens_per_second": 62432.659
    },
    {
      "epoch": 0.4126218430442323,
      "grad_norm": 15.625,
      "learning_rate": 1.3255959624172739e-05,
      "loss": 0.4746,
      "num_input_tokens_seen": 2303308288,
      "step": 31050,
      "train_runtime": 36893.7715,
      "train_tokens_per_second": 62430.817
    },
    {
      "epoch": 0.4127547325266942,
      "grad_norm": 45.25,
      "learning_rate": 1.3251889885758148e-05,
      "loss": 0.91,
      "num_input_tokens_seen": 2305436160,
      "step": 31060,
      "train_runtime": 36927.4713,
      "train_tokens_per_second": 62431.466
    },
    {
      "epoch": 0.4128876220091561,
      "grad_norm": 41.0,
      "learning_rate": 1.324781954495995e-05,
      "loss": 0.9623,
      "num_input_tokens_seen": 2306272768,
      "step": 31070,
      "train_runtime": 36940.1824,
      "train_tokens_per_second": 62432.631
    },
    {
      "epoch": 0.413020511491618,
      "grad_norm": 23.5,
      "learning_rate": 1.3243748602532145e-05,
      "loss": 0.7457,
      "num_input_tokens_seen": 2306736128,
      "step": 31080,
      "train_runtime": 36946.6764,
      "train_tokens_per_second": 62434.199
    },
    {
      "epoch": 0.41315340097407993,
      "grad_norm": 26.0,
      "learning_rate": 1.3239677059228835e-05,
      "loss": 0.6799,
      "num_input_tokens_seen": 2306956800,
      "step": 31090,
      "train_runtime": 36950.062,
      "train_tokens_per_second": 62434.45
    },
    {
      "epoch": 0.4132862904565418,
      "grad_norm": 12.1875,
      "learning_rate": 1.3235604915804235e-05,
      "loss": 0.4916,
      "num_input_tokens_seen": 2307071744,
      "step": 31100,
      "train_runtime": 36953.4564,
      "train_tokens_per_second": 62431.826
    },
    {
      "epoch": 0.41341917993900373,
      "grad_norm": 24.125,
      "learning_rate": 1.3231532173012676e-05,
      "loss": 0.9022,
      "num_input_tokens_seen": 2309030144,
      "step": 31110,
      "train_runtime": 36986.4571,
      "train_tokens_per_second": 62429.071
    },
    {
      "epoch": 0.41355206942146566,
      "grad_norm": 33.25,
      "learning_rate": 1.3227458831608598e-05,
      "loss": 0.8921,
      "num_input_tokens_seen": 2309879040,
      "step": 31120,
      "train_runtime": 36998.9038,
      "train_tokens_per_second": 62431.013
    },
    {
      "epoch": 0.41368495890392754,
      "grad_norm": 27.25,
      "learning_rate": 1.322338489234655e-05,
      "loss": 0.7682,
      "num_input_tokens_seen": 2310339328,
      "step": 31130,
      "train_runtime": 37005.4043,
      "train_tokens_per_second": 62432.484
    },
    {
      "epoch": 0.41381784838638946,
      "grad_norm": 26.125,
      "learning_rate": 1.3219310355981195e-05,
      "loss": 0.6411,
      "num_input_tokens_seen": 2310559488,
      "step": 31140,
      "train_runtime": 37008.7514,
      "train_tokens_per_second": 62432.787
    },
    {
      "epoch": 0.4139507378688514,
      "grad_norm": 10.375,
      "learning_rate": 1.32152352232673e-05,
      "loss": 0.5303,
      "num_input_tokens_seen": 2310671104,
      "step": 31150,
      "train_runtime": 37011.6844,
      "train_tokens_per_second": 62430.855
    },
    {
      "epoch": 0.41408362735131327,
      "grad_norm": 14.5625,
      "learning_rate": 1.3211159494959751e-05,
      "loss": 0.7968,
      "num_input_tokens_seen": 2312641792,
      "step": 31160,
      "train_runtime": 37044.3876,
      "train_tokens_per_second": 62428.938
    },
    {
      "epoch": 0.4142165168337752,
      "grad_norm": 160.0,
      "learning_rate": 1.3207083171813538e-05,
      "loss": 0.8718,
      "num_input_tokens_seen": 2313484032,
      "step": 31170,
      "train_runtime": 37057.036,
      "train_tokens_per_second": 62430.358
    },
    {
      "epoch": 0.4143494063162371,
      "grad_norm": 56.75,
      "learning_rate": 1.3203006254583764e-05,
      "loss": 0.6765,
      "num_input_tokens_seen": 2313967616,
      "step": 31180,
      "train_runtime": 37063.8487,
      "train_tokens_per_second": 62431.93
    },
    {
      "epoch": 0.414482295798699,
      "grad_norm": 55.0,
      "learning_rate": 1.3198928744025641e-05,
      "loss": 0.6635,
      "num_input_tokens_seen": 2314204160,
      "step": 31190,
      "train_runtime": 37067.3326,
      "train_tokens_per_second": 62432.444
    },
    {
      "epoch": 0.4146151852811609,
      "grad_norm": 18.375,
      "learning_rate": 1.3194850640894496e-05,
      "loss": 0.5064,
      "num_input_tokens_seen": 2314317824,
      "step": 31200,
      "train_runtime": 37069.6914,
      "train_tokens_per_second": 62431.537
    },
    {
      "epoch": 0.41474807476362285,
      "grad_norm": 14.9375,
      "learning_rate": 1.3190771945945755e-05,
      "loss": 0.9245,
      "num_input_tokens_seen": 2316358656,
      "step": 31210,
      "train_runtime": 37102.2012,
      "train_tokens_per_second": 62431.839
    },
    {
      "epoch": 0.4148809642460847,
      "grad_norm": 20.5,
      "learning_rate": 1.3186692659934963e-05,
      "loss": 0.8715,
      "num_input_tokens_seen": 2317211904,
      "step": 31220,
      "train_runtime": 37114.7783,
      "train_tokens_per_second": 62433.672
    },
    {
      "epoch": 0.41501385372854666,
      "grad_norm": 22.125,
      "learning_rate": 1.3182612783617772e-05,
      "loss": 0.6884,
      "num_input_tokens_seen": 2317682688,
      "step": 31230,
      "train_runtime": 37121.3661,
      "train_tokens_per_second": 62435.275
    },
    {
      "epoch": 0.4151467432110086,
      "grad_norm": 13.6875,
      "learning_rate": 1.3178532317749941e-05,
      "loss": 0.6551,
      "num_input_tokens_seen": 2317915136,
      "step": 31240,
      "train_runtime": 37124.8005,
      "train_tokens_per_second": 62435.76
    },
    {
      "epoch": 0.41527963269347046,
      "grad_norm": 15.75,
      "learning_rate": 1.3174451263087344e-05,
      "loss": 0.6536,
      "num_input_tokens_seen": 2318032384,
      "step": 31250,
      "train_runtime": 37127.2321,
      "train_tokens_per_second": 62434.829
    },
    {
      "epoch": 0.4154125221759324,
      "grad_norm": 27.875,
      "learning_rate": 1.3170369620385952e-05,
      "loss": 0.8032,
      "num_input_tokens_seen": 2319945728,
      "step": 31260,
      "train_runtime": 37158.7394,
      "train_tokens_per_second": 62433.381
    },
    {
      "epoch": 0.4155454116583943,
      "grad_norm": 16.25,
      "learning_rate": 1.3166287390401857e-05,
      "loss": 0.8306,
      "num_input_tokens_seen": 2320751104,
      "step": 31270,
      "train_runtime": 37171.3776,
      "train_tokens_per_second": 62433.82
    },
    {
      "epoch": 0.4156783011408562,
      "grad_norm": 18.5,
      "learning_rate": 1.3162204573891258e-05,
      "loss": 0.7578,
      "num_input_tokens_seen": 2321219840,
      "step": 31280,
      "train_runtime": 37177.7572,
      "train_tokens_per_second": 62435.715
    },
    {
      "epoch": 0.4158111906233181,
      "grad_norm": 25.625,
      "learning_rate": 1.3158121171610458e-05,
      "loss": 0.6878,
      "num_input_tokens_seen": 2321443072,
      "step": 31290,
      "train_runtime": 37181.163,
      "train_tokens_per_second": 62435.999
    },
    {
      "epoch": 0.41594408010578005,
      "grad_norm": 19.625,
      "learning_rate": 1.3154037184315875e-05,
      "loss": 0.4871,
      "num_input_tokens_seen": 2321556224,
      "step": 31300,
      "train_runtime": 37183.5871,
      "train_tokens_per_second": 62434.972
    },
    {
      "epoch": 0.4160769695882419,
      "grad_norm": 26.5,
      "learning_rate": 1.3149952612764024e-05,
      "loss": 0.8079,
      "num_input_tokens_seen": 2323468544,
      "step": 31310,
      "train_runtime": 37215.1036,
      "train_tokens_per_second": 62433.483
    },
    {
      "epoch": 0.41620985907070385,
      "grad_norm": 27.5,
      "learning_rate": 1.3145867457711541e-05,
      "loss": 0.8701,
      "num_input_tokens_seen": 2324271104,
      "step": 31320,
      "train_runtime": 37227.4334,
      "train_tokens_per_second": 62434.363
    },
    {
      "epoch": 0.4163427485531658,
      "grad_norm": 25.125,
      "learning_rate": 1.3141781719915159e-05,
      "loss": 0.752,
      "num_input_tokens_seen": 2324690944,
      "step": 31330,
      "train_runtime": 37233.4128,
      "train_tokens_per_second": 62435.613
    },
    {
      "epoch": 0.41647563803562765,
      "grad_norm": 11.9375,
      "learning_rate": 1.3137695400131731e-05,
      "loss": 0.5998,
      "num_input_tokens_seen": 2324904192,
      "step": 31340,
      "train_runtime": 37236.7024,
      "train_tokens_per_second": 62435.824
    },
    {
      "epoch": 0.4166085275180896,
      "grad_norm": 6.28125,
      "learning_rate": 1.3133608499118209e-05,
      "loss": 0.4892,
      "num_input_tokens_seen": 2325015808,
      "step": 31350,
      "train_runtime": 37239.0783,
      "train_tokens_per_second": 62434.838
    },
    {
      "epoch": 0.4167414170005515,
      "grad_norm": 35.25,
      "learning_rate": 1.3129521017631653e-05,
      "loss": 0.7786,
      "num_input_tokens_seen": 2326988800,
      "step": 31360,
      "train_runtime": 37270.9411,
      "train_tokens_per_second": 62434.399
    },
    {
      "epoch": 0.4168743064830134,
      "grad_norm": 35.0,
      "learning_rate": 1.3125432956429233e-05,
      "loss": 0.9555,
      "num_input_tokens_seen": 2327813120,
      "step": 31370,
      "train_runtime": 37284.1878,
      "train_tokens_per_second": 62434.326
    },
    {
      "epoch": 0.4170071959654753,
      "grad_norm": 35.25,
      "learning_rate": 1.3121344316268228e-05,
      "loss": 0.7445,
      "num_input_tokens_seen": 2328306944,
      "step": 31380,
      "train_runtime": 37291.1687,
      "train_tokens_per_second": 62435.88
    },
    {
      "epoch": 0.41714008544793724,
      "grad_norm": 16.5,
      "learning_rate": 1.311725509790602e-05,
      "loss": 0.7026,
      "num_input_tokens_seen": 2328550400,
      "step": 31390,
      "train_runtime": 37294.7993,
      "train_tokens_per_second": 62436.33
    },
    {
      "epoch": 0.4172729749303991,
      "grad_norm": 35.25,
      "learning_rate": 1.3113165302100104e-05,
      "loss": 0.5889,
      "num_input_tokens_seen": 2328672512,
      "step": 31400,
      "train_runtime": 37297.2495,
      "train_tokens_per_second": 62435.502
    },
    {
      "epoch": 0.41740586441286104,
      "grad_norm": 13.6875,
      "learning_rate": 1.3109074929608075e-05,
      "loss": 0.8882,
      "num_input_tokens_seen": 2330742016,
      "step": 31410,
      "train_runtime": 37329.7688,
      "train_tokens_per_second": 62436.551
    },
    {
      "epoch": 0.41753875389532297,
      "grad_norm": 13.6875,
      "learning_rate": 1.3104983981187637e-05,
      "loss": 0.9713,
      "num_input_tokens_seen": 2331571200,
      "step": 31420,
      "train_runtime": 37342.077,
      "train_tokens_per_second": 62438.177
    },
    {
      "epoch": 0.41767164337778484,
      "grad_norm": 13.375,
      "learning_rate": 1.3100892457596602e-05,
      "loss": 0.7736,
      "num_input_tokens_seen": 2332016384,
      "step": 31430,
      "train_runtime": 37348.4671,
      "train_tokens_per_second": 62439.414
    },
    {
      "epoch": 0.41780453286024677,
      "grad_norm": 19.25,
      "learning_rate": 1.3096800359592889e-05,
      "loss": 0.633,
      "num_input_tokens_seen": 2332247808,
      "step": 31440,
      "train_runtime": 37352.0046,
      "train_tokens_per_second": 62439.696
    },
    {
      "epoch": 0.4179374223427087,
      "grad_norm": 10.0,
      "learning_rate": 1.3092707687934525e-05,
      "loss": 0.6003,
      "num_input_tokens_seen": 2332363008,
      "step": 31450,
      "train_runtime": 37355.1805,
      "train_tokens_per_second": 62437.471
    },
    {
      "epoch": 0.41807031182517057,
      "grad_norm": 21.375,
      "learning_rate": 1.3088614443379639e-05,
      "loss": 0.8007,
      "num_input_tokens_seen": 2334537216,
      "step": 31460,
      "train_runtime": 37388.8248,
      "train_tokens_per_second": 62439.438
    },
    {
      "epoch": 0.4182032013076325,
      "grad_norm": 23.25,
      "learning_rate": 1.3084520626686468e-05,
      "loss": 0.8682,
      "num_input_tokens_seen": 2335367424,
      "step": 31470,
      "train_runtime": 37401.0439,
      "train_tokens_per_second": 62441.236
    },
    {
      "epoch": 0.41833609079009443,
      "grad_norm": 16.25,
      "learning_rate": 1.3080426238613356e-05,
      "loss": 0.7819,
      "num_input_tokens_seen": 2335816704,
      "step": 31480,
      "train_runtime": 37407.4438,
      "train_tokens_per_second": 62442.564
    },
    {
      "epoch": 0.4184689802725563,
      "grad_norm": 17.75,
      "learning_rate": 1.3076331279918748e-05,
      "loss": 0.7003,
      "num_input_tokens_seen": 2336033536,
      "step": 31490,
      "train_runtime": 37410.8781,
      "train_tokens_per_second": 62442.628
    },
    {
      "epoch": 0.41860186975501823,
      "grad_norm": 8.5625,
      "learning_rate": 1.3072235751361202e-05,
      "loss": 0.5336,
      "num_input_tokens_seen": 2336147968,
      "step": 31500,
      "train_runtime": 37414.0115,
      "train_tokens_per_second": 62440.457
    },
    {
      "epoch": 0.41873475923748016,
      "grad_norm": 28.875,
      "learning_rate": 1.306813965369938e-05,
      "loss": 0.8606,
      "num_input_tokens_seen": 2338321920,
      "step": 31510,
      "train_runtime": 37448.8558,
      "train_tokens_per_second": 62440.41
    },
    {
      "epoch": 0.4188676487199421,
      "grad_norm": 15.25,
      "learning_rate": 1.3064042987692042e-05,
      "loss": 0.8817,
      "num_input_tokens_seen": 2339151616,
      "step": 31520,
      "train_runtime": 37461.4818,
      "train_tokens_per_second": 62441.513
    },
    {
      "epoch": 0.41900053820240396,
      "grad_norm": 22.625,
      "learning_rate": 1.3059945754098066e-05,
      "loss": 0.7612,
      "num_input_tokens_seen": 2339636480,
      "step": 31530,
      "train_runtime": 37468.2499,
      "train_tokens_per_second": 62443.175
    },
    {
      "epoch": 0.4191334276848659,
      "grad_norm": 17.875,
      "learning_rate": 1.3055847953676421e-05,
      "loss": 0.6358,
      "num_input_tokens_seen": 2339873280,
      "step": 31540,
      "train_runtime": 37471.7985,
      "train_tokens_per_second": 62443.581
    },
    {
      "epoch": 0.4192663171673278,
      "grad_norm": 9.5,
      "learning_rate": 1.3051749587186192e-05,
      "loss": 0.5652,
      "num_input_tokens_seen": 2339995136,
      "step": 31550,
      "train_runtime": 37474.892,
      "train_tokens_per_second": 62441.678
    },
    {
      "epoch": 0.4193992066497897,
      "grad_norm": 36.5,
      "learning_rate": 1.3047650655386566e-05,
      "loss": 0.7783,
      "num_input_tokens_seen": 2342032384,
      "step": 31560,
      "train_runtime": 37507.7909,
      "train_tokens_per_second": 62441.224
    },
    {
      "epoch": 0.4195320961322516,
      "grad_norm": 16.125,
      "learning_rate": 1.3043551159036828e-05,
      "loss": 0.9202,
      "num_input_tokens_seen": 2342874112,
      "step": 31570,
      "train_runtime": 37520.4472,
      "train_tokens_per_second": 62442.596
    },
    {
      "epoch": 0.41966498561471355,
      "grad_norm": 28.75,
      "learning_rate": 1.303945109889638e-05,
      "loss": 0.7672,
      "num_input_tokens_seen": 2343345152,
      "step": 31580,
      "train_runtime": 37527.1686,
      "train_tokens_per_second": 62443.964
    },
    {
      "epoch": 0.4197978750971754,
      "grad_norm": 52.0,
      "learning_rate": 1.3035350475724715e-05,
      "loss": 0.6346,
      "num_input_tokens_seen": 2343569152,
      "step": 31590,
      "train_runtime": 37530.6457,
      "train_tokens_per_second": 62444.147
    },
    {
      "epoch": 0.41993076457963735,
      "grad_norm": 8.8125,
      "learning_rate": 1.3031249290281443e-05,
      "loss": 0.5188,
      "num_input_tokens_seen": 2343683584,
      "step": 31600,
      "train_runtime": 37533.2084,
      "train_tokens_per_second": 62442.932
    },
    {
      "epoch": 0.4200636540620993,
      "grad_norm": 20.375,
      "learning_rate": 1.3027147543326268e-05,
      "loss": 0.7964,
      "num_input_tokens_seen": 2345811456,
      "step": 31610,
      "train_runtime": 37567.8111,
      "train_tokens_per_second": 62442.058
    },
    {
      "epoch": 0.42019654354456115,
      "grad_norm": 27.375,
      "learning_rate": 1.3023045235619003e-05,
      "loss": 0.9202,
      "num_input_tokens_seen": 2346671104,
      "step": 31620,
      "train_runtime": 37581.0475,
      "train_tokens_per_second": 62442.94
    },
    {
      "epoch": 0.4203294330270231,
      "grad_norm": 26.375,
      "learning_rate": 1.3018942367919562e-05,
      "loss": 0.7187,
      "num_input_tokens_seen": 2347116288,
      "step": 31630,
      "train_runtime": 37587.5713,
      "train_tokens_per_second": 62443.946
    },
    {
      "epoch": 0.420462322509485,
      "grad_norm": 10.3125,
      "learning_rate": 1.3014838940987971e-05,
      "loss": 0.6207,
      "num_input_tokens_seen": 2347336192,
      "step": 31640,
      "train_runtime": 37591.0302,
      "train_tokens_per_second": 62444.051
    },
    {
      "epoch": 0.4205952119919469,
      "grad_norm": 9.1875,
      "learning_rate": 1.3010734955584345e-05,
      "loss": 0.5492,
      "num_input_tokens_seen": 2347452672,
      "step": 31650,
      "train_runtime": 37593.4991,
      "train_tokens_per_second": 62443.048
    },
    {
      "epoch": 0.4207281014744088,
      "grad_norm": 20.0,
      "learning_rate": 1.3006630412468916e-05,
      "loss": 0.8434,
      "num_input_tokens_seen": 2349531136,
      "step": 31660,
      "train_runtime": 37626.9164,
      "train_tokens_per_second": 62442.83
    },
    {
      "epoch": 0.42086099095687074,
      "grad_norm": 61.0,
      "learning_rate": 1.3002525312402015e-05,
      "loss": 0.9196,
      "num_input_tokens_seen": 2350407168,
      "step": 31670,
      "train_runtime": 37639.836,
      "train_tokens_per_second": 62444.671
    },
    {
      "epoch": 0.4209938804393326,
      "grad_norm": 21.375,
      "learning_rate": 1.299841965614407e-05,
      "loss": 0.7555,
      "num_input_tokens_seen": 2350887936,
      "step": 31680,
      "train_runtime": 37646.7109,
      "train_tokens_per_second": 62446.038
    },
    {
      "epoch": 0.42112676992179454,
      "grad_norm": 31.75,
      "learning_rate": 1.2994313444455622e-05,
      "loss": 0.6897,
      "num_input_tokens_seen": 2351111424,
      "step": 31690,
      "train_runtime": 37650.2401,
      "train_tokens_per_second": 62446.12
    },
    {
      "epoch": 0.42125965940425647,
      "grad_norm": 12.375,
      "learning_rate": 1.2990206678097306e-05,
      "loss": 0.5176,
      "num_input_tokens_seen": 2351225600,
      "step": 31700,
      "train_runtime": 37652.8805,
      "train_tokens_per_second": 62444.774
    },
    {
      "epoch": 0.42139254888671834,
      "grad_norm": 84.0,
      "learning_rate": 1.2986099357829866e-05,
      "loss": 0.8377,
      "num_input_tokens_seen": 2353265152,
      "step": 31710,
      "train_runtime": 37686.9685,
      "train_tokens_per_second": 62442.41
    },
    {
      "epoch": 0.42152543836918027,
      "grad_norm": 22.625,
      "learning_rate": 1.2981991484414152e-05,
      "loss": 0.9167,
      "num_input_tokens_seen": 2354093568,
      "step": 31720,
      "train_runtime": 37699.5394,
      "train_tokens_per_second": 62443.563
    },
    {
      "epoch": 0.4216583278516422,
      "grad_norm": 17.125,
      "learning_rate": 1.29778830586111e-05,
      "loss": 0.815,
      "num_input_tokens_seen": 2354565376,
      "step": 31730,
      "train_runtime": 37706.1921,
      "train_tokens_per_second": 62445.059
    },
    {
      "epoch": 0.4217912173341041,
      "grad_norm": 26.0,
      "learning_rate": 1.2973774081181767e-05,
      "loss": 0.6919,
      "num_input_tokens_seen": 2354797568,
      "step": 31740,
      "train_runtime": 37710.3687,
      "train_tokens_per_second": 62444.3
    },
    {
      "epoch": 0.421924106816566,
      "grad_norm": 17.0,
      "learning_rate": 1.29696645528873e-05,
      "loss": 0.4911,
      "num_input_tokens_seen": 2354911744,
      "step": 31750,
      "train_runtime": 37712.909,
      "train_tokens_per_second": 62443.121
    },
    {
      "epoch": 0.42205699629902793,
      "grad_norm": 31.125,
      "learning_rate": 1.2965554474488956e-05,
      "loss": 0.83,
      "num_input_tokens_seen": 2356989440,
      "step": 31760,
      "train_runtime": 37746.2453,
      "train_tokens_per_second": 62443.017
    },
    {
      "epoch": 0.4221898857814898,
      "grad_norm": 33.5,
      "learning_rate": 1.296144384674809e-05,
      "loss": 0.8418,
      "num_input_tokens_seen": 2357857024,
      "step": 31770,
      "train_runtime": 37759.2861,
      "train_tokens_per_second": 62444.428
    },
    {
      "epoch": 0.42232277526395173,
      "grad_norm": 17.25,
      "learning_rate": 1.2957332670426156e-05,
      "loss": 0.6825,
      "num_input_tokens_seen": 2358357248,
      "step": 31780,
      "train_runtime": 37766.2327,
      "train_tokens_per_second": 62446.187
    },
    {
      "epoch": 0.42245566474641366,
      "grad_norm": 40.5,
      "learning_rate": 1.2953220946284715e-05,
      "loss": 0.654,
      "num_input_tokens_seen": 2358594304,
      "step": 31790,
      "train_runtime": 37769.8217,
      "train_tokens_per_second": 62446.53
    },
    {
      "epoch": 0.42258855422887553,
      "grad_norm": 12.8125,
      "learning_rate": 1.294910867508543e-05,
      "loss": 0.5252,
      "num_input_tokens_seen": 2358712832,
      "step": 31800,
      "train_runtime": 37772.9692,
      "train_tokens_per_second": 62444.464
    },
    {
      "epoch": 0.42272144371133746,
      "grad_norm": 11.25,
      "learning_rate": 1.2944995857590054e-05,
      "loss": 0.8622,
      "num_input_tokens_seen": 2360881152,
      "step": 31810,
      "train_runtime": 37807.5694,
      "train_tokens_per_second": 62444.669
    },
    {
      "epoch": 0.4228543331937994,
      "grad_norm": 43.75,
      "learning_rate": 1.294088249456046e-05,
      "loss": 1.0757,
      "num_input_tokens_seen": 2361724160,
      "step": 31820,
      "train_runtime": 37820.3701,
      "train_tokens_per_second": 62445.824
    },
    {
      "epoch": 0.42298722267626127,
      "grad_norm": 15.125,
      "learning_rate": 1.2936768586758608e-05,
      "loss": 0.7077,
      "num_input_tokens_seen": 2362198016,
      "step": 31830,
      "train_runtime": 37827.2152,
      "train_tokens_per_second": 62447.05
    },
    {
      "epoch": 0.4231201121587232,
      "grad_norm": 19.625,
      "learning_rate": 1.2932654134946558e-05,
      "loss": 0.6574,
      "num_input_tokens_seen": 2362426880,
      "step": 31840,
      "train_runtime": 37831.2917,
      "train_tokens_per_second": 62446.371
    },
    {
      "epoch": 0.4232530016411851,
      "grad_norm": 7.65625,
      "learning_rate": 1.2928539139886482e-05,
      "loss": 0.5378,
      "num_input_tokens_seen": 2362540800,
      "step": 31850,
      "train_runtime": 37833.8274,
      "train_tokens_per_second": 62445.197
    },
    {
      "epoch": 0.423385891123647,
      "grad_norm": 26.125,
      "learning_rate": 1.2924423602340644e-05,
      "loss": 0.7694,
      "num_input_tokens_seen": 2364660992,
      "step": 31860,
      "train_runtime": 37868.6262,
      "train_tokens_per_second": 62443.802
    },
    {
      "epoch": 0.4235187806061089,
      "grad_norm": 26.0,
      "learning_rate": 1.2920307523071407e-05,
      "loss": 0.9095,
      "num_input_tokens_seen": 2365497856,
      "step": 31870,
      "train_runtime": 37881.2805,
      "train_tokens_per_second": 62445.034
    },
    {
      "epoch": 0.42365167008857085,
      "grad_norm": 14.3125,
      "learning_rate": 1.2916190902841247e-05,
      "loss": 0.7521,
      "num_input_tokens_seen": 2365963776,
      "step": 31880,
      "train_runtime": 37887.8981,
      "train_tokens_per_second": 62446.425
    },
    {
      "epoch": 0.4237845595710327,
      "grad_norm": 24.25,
      "learning_rate": 1.2912073742412722e-05,
      "loss": 0.6988,
      "num_input_tokens_seen": 2366188288,
      "step": 31890,
      "train_runtime": 37891.2834,
      "train_tokens_per_second": 62446.771
    },
    {
      "epoch": 0.42391744905349465,
      "grad_norm": 29.75,
      "learning_rate": 1.2907956042548503e-05,
      "loss": 0.5699,
      "num_input_tokens_seen": 2366299904,
      "step": 31900,
      "train_runtime": 37893.698,
      "train_tokens_per_second": 62445.737
    },
    {
      "epoch": 0.4240503385359566,
      "grad_norm": 21.375,
      "learning_rate": 1.2903837804011357e-05,
      "loss": 0.8289,
      "num_input_tokens_seen": 2368311552,
      "step": 31910,
      "train_runtime": 37927.4799,
      "train_tokens_per_second": 62443.156
    },
    {
      "epoch": 0.42418322801841846,
      "grad_norm": 13.6875,
      "learning_rate": 1.2899719027564151e-05,
      "loss": 0.9421,
      "num_input_tokens_seen": 2369171456,
      "step": 31920,
      "train_runtime": 37940.0802,
      "train_tokens_per_second": 62445.083
    },
    {
      "epoch": 0.4243161175008804,
      "grad_norm": 38.5,
      "learning_rate": 1.2895599713969851e-05,
      "loss": 0.7755,
      "num_input_tokens_seen": 2369625600,
      "step": 31930,
      "train_runtime": 37946.6858,
      "train_tokens_per_second": 62446.181
    },
    {
      "epoch": 0.4244490069833423,
      "grad_norm": 20.25,
      "learning_rate": 1.2891479863991522e-05,
      "loss": 0.7241,
      "num_input_tokens_seen": 2369855232,
      "step": 31940,
      "train_runtime": 37950.1443,
      "train_tokens_per_second": 62446.541
    },
    {
      "epoch": 0.4245818964658042,
      "grad_norm": 16.125,
      "learning_rate": 1.2887359478392334e-05,
      "loss": 0.5224,
      "num_input_tokens_seen": 2369970944,
      "step": 31950,
      "train_runtime": 37952.636,
      "train_tokens_per_second": 62445.49
    },
    {
      "epoch": 0.4247147859482661,
      "grad_norm": 13.25,
      "learning_rate": 1.2883238557935548e-05,
      "loss": 0.9025,
      "num_input_tokens_seen": 2372083200,
      "step": 31960,
      "train_runtime": 37986.6366,
      "train_tokens_per_second": 62445.202
    },
    {
      "epoch": 0.42484767543072804,
      "grad_norm": 22.5,
      "learning_rate": 1.2879117103384527e-05,
      "loss": 0.8607,
      "num_input_tokens_seen": 2372923648,
      "step": 31970,
      "train_runtime": 37999.3576,
      "train_tokens_per_second": 62446.415
    },
    {
      "epoch": 0.42498056491319,
      "grad_norm": 21.625,
      "learning_rate": 1.2874995115502735e-05,
      "loss": 0.816,
      "num_input_tokens_seen": 2373386496,
      "step": 31980,
      "train_runtime": 38005.9038,
      "train_tokens_per_second": 62447.837
    },
    {
      "epoch": 0.42511345439565185,
      "grad_norm": 23.125,
      "learning_rate": 1.2870872595053738e-05,
      "loss": 0.6695,
      "num_input_tokens_seen": 2373614080,
      "step": 31990,
      "train_runtime": 38009.3116,
      "train_tokens_per_second": 62448.226
    },
    {
      "epoch": 0.4252463438781138,
      "grad_norm": 31.0,
      "learning_rate": 1.2866749542801187e-05,
      "loss": 0.5108,
      "num_input_tokens_seen": 2373729024,
      "step": 32000,
      "train_runtime": 38012.2968,
      "train_tokens_per_second": 62446.346
    },
    {
      "epoch": 0.4253792333605757,
      "grad_norm": 41.75,
      "learning_rate": 1.2862625959508848e-05,
      "loss": 0.8439,
      "num_input_tokens_seen": 2375672064,
      "step": 32010,
      "train_runtime": 38044.2395,
      "train_tokens_per_second": 62444.988
    },
    {
      "epoch": 0.4255121228430376,
      "grad_norm": 12.125,
      "learning_rate": 1.2858501845940575e-05,
      "loss": 0.8365,
      "num_input_tokens_seen": 2376519168,
      "step": 32020,
      "train_runtime": 38056.8557,
      "train_tokens_per_second": 62446.545
    },
    {
      "epoch": 0.4256450123254995,
      "grad_norm": 13.4375,
      "learning_rate": 1.2854377202860322e-05,
      "loss": 0.7329,
      "num_input_tokens_seen": 2376978944,
      "step": 32030,
      "train_runtime": 38063.3227,
      "train_tokens_per_second": 62448.015
    },
    {
      "epoch": 0.42577790180796143,
      "grad_norm": 10.8125,
      "learning_rate": 1.2850252031032148e-05,
      "loss": 0.6784,
      "num_input_tokens_seen": 2377196544,
      "step": 32040,
      "train_runtime": 38066.7978,
      "train_tokens_per_second": 62448.03
    },
    {
      "epoch": 0.4259107912904233,
      "grad_norm": 19.25,
      "learning_rate": 1.28461263312202e-05,
      "loss": 0.5027,
      "num_input_tokens_seen": 2377307648,
      "step": 32050,
      "train_runtime": 38069.5466,
      "train_tokens_per_second": 62446.44
    },
    {
      "epoch": 0.42604368077288524,
      "grad_norm": 15.375,
      "learning_rate": 1.2842000104188724e-05,
      "loss": 0.7969,
      "num_input_tokens_seen": 2379276032,
      "step": 32060,
      "train_runtime": 38102.326,
      "train_tokens_per_second": 62444.378
    },
    {
      "epoch": 0.42617657025534716,
      "grad_norm": 11.9375,
      "learning_rate": 1.283787335070207e-05,
      "loss": 0.8815,
      "num_input_tokens_seen": 2380100352,
      "step": 32070,
      "train_runtime": 38114.8366,
      "train_tokens_per_second": 62445.509
    },
    {
      "epoch": 0.42630945973780904,
      "grad_norm": 22.625,
      "learning_rate": 1.2833746071524686e-05,
      "loss": 0.7043,
      "num_input_tokens_seen": 2380551168,
      "step": 32080,
      "train_runtime": 38123.0739,
      "train_tokens_per_second": 62443.841
    },
    {
      "epoch": 0.42644234922027097,
      "grad_norm": 17.75,
      "learning_rate": 1.2829618267421111e-05,
      "loss": 0.7393,
      "num_input_tokens_seen": 2380790016,
      "step": 32090,
      "train_runtime": 38127.3008,
      "train_tokens_per_second": 62443.183
    },
    {
      "epoch": 0.4265752387027329,
      "grad_norm": 12.0,
      "learning_rate": 1.2825489939155979e-05,
      "loss": 0.5278,
      "num_input_tokens_seen": 2380905472,
      "step": 32100,
      "train_runtime": 38132.385,
      "train_tokens_per_second": 62437.885
    },
    {
      "epoch": 0.42670812818519477,
      "grad_norm": 26.25,
      "learning_rate": 1.2821361087494033e-05,
      "loss": 0.8303,
      "num_input_tokens_seen": 2382743552,
      "step": 32110,
      "train_runtime": 38162.8534,
      "train_tokens_per_second": 62436.2
    },
    {
      "epoch": 0.4268410176676567,
      "grad_norm": 16.25,
      "learning_rate": 1.2817231713200105e-05,
      "loss": 0.9585,
      "num_input_tokens_seen": 2383563520,
      "step": 32120,
      "train_runtime": 38175.3233,
      "train_tokens_per_second": 62437.284
    },
    {
      "epoch": 0.4269739071501186,
      "grad_norm": 104.5,
      "learning_rate": 1.281310181703912e-05,
      "loss": 0.7535,
      "num_input_tokens_seen": 2384012544,
      "step": 32130,
      "train_runtime": 38182.5945,
      "train_tokens_per_second": 62437.154
    },
    {
      "epoch": 0.4271067966325805,
      "grad_norm": 39.0,
      "learning_rate": 1.2808971399776106e-05,
      "loss": 0.6942,
      "num_input_tokens_seen": 2384238848,
      "step": 32140,
      "train_runtime": 38187.7972,
      "train_tokens_per_second": 62434.574
    },
    {
      "epoch": 0.4272396861150424,
      "grad_norm": 8.625,
      "learning_rate": 1.2804840462176192e-05,
      "loss": 0.5692,
      "num_input_tokens_seen": 2384351744,
      "step": 32150,
      "train_runtime": 38192.6537,
      "train_tokens_per_second": 62429.591
    },
    {
      "epoch": 0.42737257559750436,
      "grad_norm": 18.875,
      "learning_rate": 1.280070900500459e-05,
      "loss": 0.8061,
      "num_input_tokens_seen": 2386361856,
      "step": 32160,
      "train_runtime": 38225.6546,
      "train_tokens_per_second": 62428.28
    },
    {
      "epoch": 0.42750546507996623,
      "grad_norm": 42.0,
      "learning_rate": 1.2796577029026617e-05,
      "loss": 0.9152,
      "num_input_tokens_seen": 2387203072,
      "step": 32170,
      "train_runtime": 38238.4042,
      "train_tokens_per_second": 62429.464
    },
    {
      "epoch": 0.42763835456242816,
      "grad_norm": 20.25,
      "learning_rate": 1.2792444535007687e-05,
      "loss": 0.7191,
      "num_input_tokens_seen": 2387674112,
      "step": 32180,
      "train_runtime": 38245.3818,
      "train_tokens_per_second": 62430.39
    },
    {
      "epoch": 0.4277712440448901,
      "grad_norm": 14.3125,
      "learning_rate": 1.2788311523713308e-05,
      "loss": 0.6966,
      "num_input_tokens_seen": 2387906304,
      "step": 32190,
      "train_runtime": 38250.598,
      "train_tokens_per_second": 62427.947
    },
    {
      "epoch": 0.42790413352735196,
      "grad_norm": 13.125,
      "learning_rate": 1.2784177995909082e-05,
      "loss": 0.4337,
      "num_input_tokens_seen": 2388020736,
      "step": 32200,
      "train_runtime": 38254.0336,
      "train_tokens_per_second": 62425.332
    },
    {
      "epoch": 0.4280370230098139,
      "grad_norm": 21.75,
      "learning_rate": 1.2780043952360707e-05,
      "loss": 0.7399,
      "num_input_tokens_seen": 2390118656,
      "step": 32210,
      "train_runtime": 38288.6888,
      "train_tokens_per_second": 62423.623
    },
    {
      "epoch": 0.4281699124922758,
      "grad_norm": 16.75,
      "learning_rate": 1.2775909393833978e-05,
      "loss": 0.9745,
      "num_input_tokens_seen": 2390966528,
      "step": 32220,
      "train_runtime": 38301.9606,
      "train_tokens_per_second": 62424.129
    },
    {
      "epoch": 0.4283028019747377,
      "grad_norm": 28.375,
      "learning_rate": 1.2771774321094787e-05,
      "loss": 0.7587,
      "num_input_tokens_seen": 2391482112,
      "step": 32230,
      "train_runtime": 38309.6659,
      "train_tokens_per_second": 62425.032
    },
    {
      "epoch": 0.4284356914571996,
      "grad_norm": 20.875,
      "learning_rate": 1.2767638734909121e-05,
      "loss": 0.5839,
      "num_input_tokens_seen": 2391721472,
      "step": 32240,
      "train_runtime": 38315.2259,
      "train_tokens_per_second": 62422.22
    },
    {
      "epoch": 0.42856858093966155,
      "grad_norm": 8.75,
      "learning_rate": 1.2763502636043055e-05,
      "loss": 0.5841,
      "num_input_tokens_seen": 2391839232,
      "step": 32250,
      "train_runtime": 38319.4862,
      "train_tokens_per_second": 62418.353
    },
    {
      "epoch": 0.4287014704221234,
      "grad_norm": 20.625,
      "learning_rate": 1.275936602526277e-05,
      "loss": 0.8501,
      "num_input_tokens_seen": 2393901312,
      "step": 32260,
      "train_runtime": 38351.8351,
      "train_tokens_per_second": 62419.472
    },
    {
      "epoch": 0.42883435990458535,
      "grad_norm": 19.75,
      "learning_rate": 1.2755228903334529e-05,
      "loss": 0.897,
      "num_input_tokens_seen": 2394741504,
      "step": 32270,
      "train_runtime": 38364.0825,
      "train_tokens_per_second": 62421.446
    },
    {
      "epoch": 0.4289672493870473,
      "grad_norm": 20.125,
      "learning_rate": 1.2751091271024706e-05,
      "loss": 0.7193,
      "num_input_tokens_seen": 2395191552,
      "step": 32280,
      "train_runtime": 38370.4223,
      "train_tokens_per_second": 62422.861
    },
    {
      "epoch": 0.42910013886950915,
      "grad_norm": 11.25,
      "learning_rate": 1.2746953129099756e-05,
      "loss": 0.6667,
      "num_input_tokens_seen": 2395417856,
      "step": 32290,
      "train_runtime": 38374.0174,
      "train_tokens_per_second": 62422.911
    },
    {
      "epoch": 0.4292330283519711,
      "grad_norm": 18.5,
      "learning_rate": 1.2742814478326233e-05,
      "loss": 0.4753,
      "num_input_tokens_seen": 2395534080,
      "step": 32300,
      "train_runtime": 38376.6539,
      "train_tokens_per_second": 62421.651
    },
    {
      "epoch": 0.429365917834433,
      "grad_norm": 23.375,
      "learning_rate": 1.273867531947078e-05,
      "loss": 0.8172,
      "num_input_tokens_seen": 2397554176,
      "step": 32310,
      "train_runtime": 38409.6776,
      "train_tokens_per_second": 62420.575
    },
    {
      "epoch": 0.4294988073168949,
      "grad_norm": 13.1875,
      "learning_rate": 1.2734535653300148e-05,
      "loss": 0.9526,
      "num_input_tokens_seen": 2398397696,
      "step": 32320,
      "train_runtime": 38422.2353,
      "train_tokens_per_second": 62422.128
    },
    {
      "epoch": 0.4296316967993568,
      "grad_norm": 13.625,
      "learning_rate": 1.273039548058117e-05,
      "loss": 0.7406,
      "num_input_tokens_seen": 2398864128,
      "step": 32330,
      "train_runtime": 38429.3348,
      "train_tokens_per_second": 62422.734
    },
    {
      "epoch": 0.42976458628181874,
      "grad_norm": 11.625,
      "learning_rate": 1.2726254802080775e-05,
      "loss": 0.604,
      "num_input_tokens_seen": 2399099392,
      "step": 32340,
      "train_runtime": 38433.1301,
      "train_tokens_per_second": 62422.691
    },
    {
      "epoch": 0.4298974757642806,
      "grad_norm": 10.75,
      "learning_rate": 1.272211361856599e-05,
      "loss": 0.5978,
      "num_input_tokens_seen": 2399218176,
      "step": 32350,
      "train_runtime": 38436.6553,
      "train_tokens_per_second": 62420.056
    },
    {
      "epoch": 0.43003036524674254,
      "grad_norm": 35.75,
      "learning_rate": 1.271797193080393e-05,
      "loss": 0.8812,
      "num_input_tokens_seen": 2401251072,
      "step": 32360,
      "train_runtime": 38470.6639,
      "train_tokens_per_second": 62417.719
    },
    {
      "epoch": 0.43016325472920447,
      "grad_norm": 35.0,
      "learning_rate": 1.2713829739561803e-05,
      "loss": 0.9763,
      "num_input_tokens_seen": 2402098688,
      "step": 32370,
      "train_runtime": 38483.3128,
      "train_tokens_per_second": 62419.228
    },
    {
      "epoch": 0.43029614421166634,
      "grad_norm": 20.625,
      "learning_rate": 1.2709687045606916e-05,
      "loss": 0.7514,
      "num_input_tokens_seen": 2402550272,
      "step": 32380,
      "train_runtime": 38490.0865,
      "train_tokens_per_second": 62419.976
    },
    {
      "epoch": 0.43042903369412827,
      "grad_norm": 11.75,
      "learning_rate": 1.2705543849706663e-05,
      "loss": 0.5634,
      "num_input_tokens_seen": 2402777600,
      "step": 32390,
      "train_runtime": 38494.6059,
      "train_tokens_per_second": 62418.553
    },
    {
      "epoch": 0.4305619231765902,
      "grad_norm": 39.5,
      "learning_rate": 1.2701400152628537e-05,
      "loss": 0.4757,
      "num_input_tokens_seen": 2402891776,
      "step": 32400,
      "train_runtime": 38500.1168,
      "train_tokens_per_second": 62412.584
    },
    {
      "epoch": 0.4306948126590521,
      "grad_norm": 23.75,
      "learning_rate": 1.2697255955140126e-05,
      "loss": 0.841,
      "num_input_tokens_seen": 2404850944,
      "step": 32410,
      "train_runtime": 38531.5778,
      "train_tokens_per_second": 62412.47
    },
    {
      "epoch": 0.430827702141514,
      "grad_norm": 24.875,
      "learning_rate": 1.2693111258009096e-05,
      "loss": 0.8925,
      "num_input_tokens_seen": 2405681664,
      "step": 32420,
      "train_runtime": 38544.5658,
      "train_tokens_per_second": 62412.992
    },
    {
      "epoch": 0.43096059162397593,
      "grad_norm": 16.25,
      "learning_rate": 1.268896606200322e-05,
      "loss": 0.7654,
      "num_input_tokens_seen": 2406130944,
      "step": 32430,
      "train_runtime": 38551.7907,
      "train_tokens_per_second": 62412.949
    },
    {
      "epoch": 0.4310934811064378,
      "grad_norm": 19.5,
      "learning_rate": 1.268482036789036e-05,
      "loss": 0.6331,
      "num_input_tokens_seen": 2406359808,
      "step": 32440,
      "train_runtime": 38556.3454,
      "train_tokens_per_second": 62411.512
    },
    {
      "epoch": 0.43122637058889973,
      "grad_norm": 26.875,
      "learning_rate": 1.2680674176438465e-05,
      "loss": 0.5105,
      "num_input_tokens_seen": 2406474240,
      "step": 32450,
      "train_runtime": 38559.5585,
      "train_tokens_per_second": 62409.279
    },
    {
      "epoch": 0.43135926007136166,
      "grad_norm": 26.5,
      "learning_rate": 1.2676527488415584e-05,
      "loss": 0.8672,
      "num_input_tokens_seen": 2408367104,
      "step": 32460,
      "train_runtime": 38590.9643,
      "train_tokens_per_second": 62407.539
    },
    {
      "epoch": 0.4314921495538236,
      "grad_norm": 12.5,
      "learning_rate": 1.267238030458985e-05,
      "loss": 0.9279,
      "num_input_tokens_seen": 2409203456,
      "step": 32470,
      "train_runtime": 38603.3596,
      "train_tokens_per_second": 62409.165
    },
    {
      "epoch": 0.43162503903628546,
      "grad_norm": 10.5,
      "learning_rate": 1.2668232625729494e-05,
      "loss": 0.7853,
      "num_input_tokens_seen": 2409635072,
      "step": 32480,
      "train_runtime": 38610.9071,
      "train_tokens_per_second": 62408.145
    },
    {
      "epoch": 0.4317579285187474,
      "grad_norm": 18.0,
      "learning_rate": 1.2664084452602835e-05,
      "loss": 0.6801,
      "num_input_tokens_seen": 2409861632,
      "step": 32490,
      "train_runtime": 38614.7515,
      "train_tokens_per_second": 62407.798
    },
    {
      "epoch": 0.4318908180012093,
      "grad_norm": 9.875,
      "learning_rate": 1.2659935785978289e-05,
      "loss": 0.5223,
      "num_input_tokens_seen": 2409977344,
      "step": 32500,
      "train_runtime": 38619.2041,
      "train_tokens_per_second": 62403.599
    },
    {
      "epoch": 0.4320237074836712,
      "grad_norm": 19.125,
      "learning_rate": 1.2655786626624356e-05,
      "loss": 0.8764,
      "num_input_tokens_seen": 2412016896,
      "step": 32510,
      "train_runtime": 38650.7193,
      "train_tokens_per_second": 62405.485
    },
    {
      "epoch": 0.4321565969661331,
      "grad_norm": 28.625,
      "learning_rate": 1.2651636975309634e-05,
      "loss": 0.9021,
      "num_input_tokens_seen": 2412879872,
      "step": 32520,
      "train_runtime": 38663.3464,
      "train_tokens_per_second": 62407.425
    },
    {
      "epoch": 0.43228948644859505,
      "grad_norm": 14.25,
      "learning_rate": 1.2647486832802803e-05,
      "loss": 0.7409,
      "num_input_tokens_seen": 2413332736,
      "step": 32530,
      "train_runtime": 38671.29,
      "train_tokens_per_second": 62406.316
    },
    {
      "epoch": 0.4324223759310569,
      "grad_norm": 25.25,
      "learning_rate": 1.2643336199872642e-05,
      "loss": 0.6546,
      "num_input_tokens_seen": 2413561600,
      "step": 32540,
      "train_runtime": 38675.1343,
      "train_tokens_per_second": 62406.03
    },
    {
      "epoch": 0.43255526541351885,
      "grad_norm": 16.375,
      "learning_rate": 1.2639185077288023e-05,
      "loss": 0.5823,
      "num_input_tokens_seen": 2413676032,
      "step": 32550,
      "train_runtime": 38678.273,
      "train_tokens_per_second": 62403.925
    },
    {
      "epoch": 0.4326881548959808,
      "grad_norm": 51.75,
      "learning_rate": 1.2635033465817903e-05,
      "loss": 0.7454,
      "num_input_tokens_seen": 2415714048,
      "step": 32560,
      "train_runtime": 38711.7798,
      "train_tokens_per_second": 62402.557
    },
    {
      "epoch": 0.43282104437844265,
      "grad_norm": 18.5,
      "learning_rate": 1.2630881366231326e-05,
      "loss": 0.8282,
      "num_input_tokens_seen": 2416508928,
      "step": 32570,
      "train_runtime": 38724.7865,
      "train_tokens_per_second": 62402.124
    },
    {
      "epoch": 0.4329539338609046,
      "grad_norm": 71.0,
      "learning_rate": 1.2626728779297437e-05,
      "loss": 0.7457,
      "num_input_tokens_seen": 2416932864,
      "step": 32580,
      "train_runtime": 38730.9323,
      "train_tokens_per_second": 62403.168
    },
    {
      "epoch": 0.4330868233433665,
      "grad_norm": 14.6875,
      "learning_rate": 1.2622575705785463e-05,
      "loss": 0.6842,
      "num_input_tokens_seen": 2417146624,
      "step": 32590,
      "train_runtime": 38734.4606,
      "train_tokens_per_second": 62403.002
    },
    {
      "epoch": 0.4332197128258284,
      "grad_norm": 22.875,
      "learning_rate": 1.2618422146464725e-05,
      "loss": 0.539,
      "num_input_tokens_seen": 2417259520,
      "step": 32600,
      "train_runtime": 38737.3166,
      "train_tokens_per_second": 62401.316
    },
    {
      "epoch": 0.4333526023082903,
      "grad_norm": 13.4375,
      "learning_rate": 1.2614268102104634e-05,
      "loss": 0.8199,
      "num_input_tokens_seen": 2419317760,
      "step": 32610,
      "train_runtime": 38770.9979,
      "train_tokens_per_second": 62400.193
    },
    {
      "epoch": 0.43348549179075224,
      "grad_norm": 14.625,
      "learning_rate": 1.2610113573474687e-05,
      "loss": 0.8922,
      "num_input_tokens_seen": 2420127232,
      "step": 32620,
      "train_runtime": 38783.5882,
      "train_tokens_per_second": 62400.808
    },
    {
      "epoch": 0.4336183812732141,
      "grad_norm": 66.5,
      "learning_rate": 1.2605958561344473e-05,
      "loss": 0.7328,
      "num_input_tokens_seen": 2420586752,
      "step": 32630,
      "train_runtime": 38790.4136,
      "train_tokens_per_second": 62401.674
    },
    {
      "epoch": 0.43375127075567604,
      "grad_norm": 22.625,
      "learning_rate": 1.2601803066483674e-05,
      "loss": 0.6991,
      "num_input_tokens_seen": 2420826112,
      "step": 32640,
      "train_runtime": 38794.7059,
      "train_tokens_per_second": 62400.94
    },
    {
      "epoch": 0.433884160238138,
      "grad_norm": 9.125,
      "learning_rate": 1.2597647089662056e-05,
      "loss": 0.5187,
      "num_input_tokens_seen": 2420947712,
      "step": 32650,
      "train_runtime": 38798.94,
      "train_tokens_per_second": 62397.264
    },
    {
      "epoch": 0.43401704972059985,
      "grad_norm": 19.125,
      "learning_rate": 1.2593490631649477e-05,
      "loss": 0.7795,
      "num_input_tokens_seen": 2423037696,
      "step": 32660,
      "train_runtime": 38830.9673,
      "train_tokens_per_second": 62399.622
    },
    {
      "epoch": 0.4341499392030618,
      "grad_norm": 19.0,
      "learning_rate": 1.2589333693215886e-05,
      "loss": 0.9064,
      "num_input_tokens_seen": 2423863296,
      "step": 32670,
      "train_runtime": 38843.7522,
      "train_tokens_per_second": 62400.339
    },
    {
      "epoch": 0.4342828286855237,
      "grad_norm": 27.25,
      "learning_rate": 1.2585176275131314e-05,
      "loss": 0.7819,
      "num_input_tokens_seen": 2424367616,
      "step": 32680,
      "train_runtime": 38850.7344,
      "train_tokens_per_second": 62402.105
    },
    {
      "epoch": 0.4344157181679856,
      "grad_norm": 9.375,
      "learning_rate": 1.2581018378165892e-05,
      "loss": 0.6299,
      "num_input_tokens_seen": 2424602880,
      "step": 32690,
      "train_runtime": 38855.1263,
      "train_tokens_per_second": 62401.107
    },
    {
      "epoch": 0.4345486076504475,
      "grad_norm": 24.625,
      "learning_rate": 1.2576860003089828e-05,
      "loss": 0.5914,
      "num_input_tokens_seen": 2424720128,
      "step": 32700,
      "train_runtime": 38859.4125,
      "train_tokens_per_second": 62397.241
    },
    {
      "epoch": 0.43468149713290943,
      "grad_norm": 26.75,
      "learning_rate": 1.2572701150673426e-05,
      "loss": 0.8642,
      "num_input_tokens_seen": 2426921472,
      "step": 32710,
      "train_runtime": 38893.4644,
      "train_tokens_per_second": 62399.21
    },
    {
      "epoch": 0.4348143866153713,
      "grad_norm": 49.75,
      "learning_rate": 1.256854182168708e-05,
      "loss": 0.9314,
      "num_input_tokens_seen": 2427751680,
      "step": 32720,
      "train_runtime": 38906.3473,
      "train_tokens_per_second": 62399.887
    },
    {
      "epoch": 0.43494727609783324,
      "grad_norm": 20.375,
      "learning_rate": 1.256438201690126e-05,
      "loss": 0.7345,
      "num_input_tokens_seen": 2428209152,
      "step": 32730,
      "train_runtime": 38913.2888,
      "train_tokens_per_second": 62400.512
    },
    {
      "epoch": 0.43508016558029516,
      "grad_norm": 15.5,
      "learning_rate": 1.2560221737086537e-05,
      "loss": 0.6687,
      "num_input_tokens_seen": 2428433408,
      "step": 32740,
      "train_runtime": 38917.5217,
      "train_tokens_per_second": 62399.487
    },
    {
      "epoch": 0.43521305506275704,
      "grad_norm": 15.375,
      "learning_rate": 1.2556060983013567e-05,
      "loss": 0.5353,
      "num_input_tokens_seen": 2428547584,
      "step": 32750,
      "train_runtime": 38920.8628,
      "train_tokens_per_second": 62397.064
    },
    {
      "epoch": 0.43534594454521897,
      "grad_norm": 31.5,
      "learning_rate": 1.2551899755453092e-05,
      "loss": 0.8309,
      "num_input_tokens_seen": 2430543616,
      "step": 32760,
      "train_runtime": 38953.2686,
      "train_tokens_per_second": 62396.397
    },
    {
      "epoch": 0.4354788340276809,
      "grad_norm": 13.0625,
      "learning_rate": 1.2547738055175947e-05,
      "loss": 0.9129,
      "num_input_tokens_seen": 2431394304,
      "step": 32770,
      "train_runtime": 38967.0158,
      "train_tokens_per_second": 62396.215
    },
    {
      "epoch": 0.43561172351014277,
      "grad_norm": 21.25,
      "learning_rate": 1.254357588295304e-05,
      "loss": 0.7366,
      "num_input_tokens_seen": 2431848192,
      "step": 32780,
      "train_runtime": 38974.4047,
      "train_tokens_per_second": 62396.032
    },
    {
      "epoch": 0.4357446129926047,
      "grad_norm": 25.75,
      "learning_rate": 1.2539413239555383e-05,
      "loss": 0.6352,
      "num_input_tokens_seen": 2432081408,
      "step": 32790,
      "train_runtime": 38978.5162,
      "train_tokens_per_second": 62395.433
    },
    {
      "epoch": 0.4358775024750666,
      "grad_norm": 18.375,
      "learning_rate": 1.2535250125754065e-05,
      "loss": 0.5978,
      "num_input_tokens_seen": 2432196608,
      "step": 32800,
      "train_runtime": 38981.4872,
      "train_tokens_per_second": 62393.633
    },
    {
      "epoch": 0.4360103919575285,
      "grad_norm": 34.25,
      "learning_rate": 1.253108654232027e-05,
      "loss": 0.8058,
      "num_input_tokens_seen": 2434204160,
      "step": 32810,
      "train_runtime": 39015.7043,
      "train_tokens_per_second": 62390.368
    },
    {
      "epoch": 0.4361432814399904,
      "grad_norm": 14.8125,
      "learning_rate": 1.2526922490025264e-05,
      "loss": 0.9306,
      "num_input_tokens_seen": 2435036928,
      "step": 32820,
      "train_runtime": 39028.7905,
      "train_tokens_per_second": 62390.786
    },
    {
      "epoch": 0.43627617092245236,
      "grad_norm": 17.875,
      "learning_rate": 1.2522757969640393e-05,
      "loss": 0.8076,
      "num_input_tokens_seen": 2435533568,
      "step": 32830,
      "train_runtime": 39035.7055,
      "train_tokens_per_second": 62392.457
    },
    {
      "epoch": 0.43640906040491423,
      "grad_norm": 12.25,
      "learning_rate": 1.2518592981937107e-05,
      "loss": 0.6101,
      "num_input_tokens_seen": 2435755776,
      "step": 32840,
      "train_runtime": 39039.4017,
      "train_tokens_per_second": 62392.241
    },
    {
      "epoch": 0.43654194988737616,
      "grad_norm": 28.5,
      "learning_rate": 1.251442752768693e-05,
      "loss": 0.5664,
      "num_input_tokens_seen": 2435869952,
      "step": 32850,
      "train_runtime": 39042.5532,
      "train_tokens_per_second": 62390.13
    },
    {
      "epoch": 0.4366748393698381,
      "grad_norm": 21.125,
      "learning_rate": 1.2510261607661471e-05,
      "loss": 0.8179,
      "num_input_tokens_seen": 2437887488,
      "step": 32860,
      "train_runtime": 39074.9094,
      "train_tokens_per_second": 62390.1
    },
    {
      "epoch": 0.43680772885229996,
      "grad_norm": 31.5,
      "learning_rate": 1.2506095222632436e-05,
      "loss": 0.8951,
      "num_input_tokens_seen": 2438724608,
      "step": 32870,
      "train_runtime": 39087.4131,
      "train_tokens_per_second": 62391.558
    },
    {
      "epoch": 0.4369406183347619,
      "grad_norm": 15.25,
      "learning_rate": 1.2501928373371606e-05,
      "loss": 0.727,
      "num_input_tokens_seen": 2439183616,
      "step": 32880,
      "train_runtime": 39094.0195,
      "train_tokens_per_second": 62392.756
    },
    {
      "epoch": 0.4370735078172238,
      "grad_norm": 19.875,
      "learning_rate": 1.2497761060650853e-05,
      "loss": 0.6389,
      "num_input_tokens_seen": 2439417088,
      "step": 32890,
      "train_runtime": 39099.1389,
      "train_tokens_per_second": 62390.558
    },
    {
      "epoch": 0.4372063972996857,
      "grad_norm": 9.75,
      "learning_rate": 1.2493593285242136e-05,
      "loss": 0.544,
      "num_input_tokens_seen": 2439534592,
      "step": 32900,
      "train_runtime": 39102.3104,
      "train_tokens_per_second": 62388.503
    },
    {
      "epoch": 0.4373392867821476,
      "grad_norm": 23.625,
      "learning_rate": 1.2489425047917499e-05,
      "loss": 0.8217,
      "num_input_tokens_seen": 2441540608,
      "step": 32910,
      "train_runtime": 39134.9171,
      "train_tokens_per_second": 62387.78
    },
    {
      "epoch": 0.43747217626460955,
      "grad_norm": 29.375,
      "learning_rate": 1.2485256349449069e-05,
      "loss": 0.9021,
      "num_input_tokens_seen": 2442393344,
      "step": 32920,
      "train_runtime": 39147.9267,
      "train_tokens_per_second": 62388.83
    },
    {
      "epoch": 0.4376050657470715,
      "grad_norm": 31.125,
      "learning_rate": 1.248108719060906e-05,
      "loss": 0.7831,
      "num_input_tokens_seen": 2442895360,
      "step": 32930,
      "train_runtime": 39154.8008,
      "train_tokens_per_second": 62390.698
    },
    {
      "epoch": 0.43773795522953335,
      "grad_norm": 13.0625,
      "learning_rate": 1.2476917572169776e-05,
      "loss": 0.6344,
      "num_input_tokens_seen": 2443127808,
      "step": 32940,
      "train_runtime": 39158.5798,
      "train_tokens_per_second": 62390.613
    },
    {
      "epoch": 0.4378708447119953,
      "grad_norm": 48.75,
      "learning_rate": 1.2472747494903594e-05,
      "loss": 0.4733,
      "num_input_tokens_seen": 2443240704,
      "step": 32950,
      "train_runtime": 39162.6168,
      "train_tokens_per_second": 62387.065
    },
    {
      "epoch": 0.4380037341944572,
      "grad_norm": 112.5,
      "learning_rate": 1.246857695958299e-05,
      "loss": 0.836,
      "num_input_tokens_seen": 2445328384,
      "step": 32960,
      "train_runtime": 39197.1816,
      "train_tokens_per_second": 62385.311
    },
    {
      "epoch": 0.4381366236769191,
      "grad_norm": 43.5,
      "learning_rate": 1.2464405966980513e-05,
      "loss": 0.9178,
      "num_input_tokens_seen": 2446156800,
      "step": 32970,
      "train_runtime": 39209.8527,
      "train_tokens_per_second": 62386.279
    },
    {
      "epoch": 0.438269513159381,
      "grad_norm": 16.5,
      "learning_rate": 1.246023451786881e-05,
      "loss": 0.8192,
      "num_input_tokens_seen": 2446631424,
      "step": 32980,
      "train_runtime": 39216.577,
      "train_tokens_per_second": 62387.684
    },
    {
      "epoch": 0.43840240264184294,
      "grad_norm": 26.375,
      "learning_rate": 1.2456062613020597e-05,
      "loss": 0.6443,
      "num_input_tokens_seen": 2446862080,
      "step": 32990,
      "train_runtime": 39220.3072,
      "train_tokens_per_second": 62387.632
    },
    {
      "epoch": 0.4385352921243048,
      "grad_norm": 7.96875,
      "learning_rate": 1.2451890253208685e-05,
      "loss": 0.5331,
      "num_input_tokens_seen": 2446977536,
      "step": 33000,
      "train_runtime": 39224.3285,
      "train_tokens_per_second": 62384.179
    },
    {
      "epoch": 0.43866818160676674,
      "grad_norm": 33.5,
      "learning_rate": 1.244771743920597e-05,
      "loss": 0.8625,
      "num_input_tokens_seen": 2448940288,
      "step": 33010,
      "train_runtime": 39256.0089,
      "train_tokens_per_second": 62383.833
    },
    {
      "epoch": 0.43880107108922867,
      "grad_norm": 19.625,
      "learning_rate": 1.2443544171785423e-05,
      "loss": 0.9226,
      "num_input_tokens_seen": 2449773568,
      "step": 33020,
      "train_runtime": 39268.7182,
      "train_tokens_per_second": 62384.862
    },
    {
      "epoch": 0.43893396057169054,
      "grad_norm": 19.625,
      "learning_rate": 1.2439370451720105e-05,
      "loss": 0.7533,
      "num_input_tokens_seen": 2450272512,
      "step": 33030,
      "train_runtime": 39275.5947,
      "train_tokens_per_second": 62386.643
    },
    {
      "epoch": 0.43906685005415247,
      "grad_norm": 23.875,
      "learning_rate": 1.2435196279783166e-05,
      "loss": 0.6588,
      "num_input_tokens_seen": 2450498560,
      "step": 33040,
      "train_runtime": 39279.3179,
      "train_tokens_per_second": 62386.485
    },
    {
      "epoch": 0.4391997395366144,
      "grad_norm": 13.5,
      "learning_rate": 1.2431021656747827e-05,
      "loss": 0.5888,
      "num_input_tokens_seen": 2450611200,
      "step": 33050,
      "train_runtime": 39283.2933,
      "train_tokens_per_second": 62383.039
    },
    {
      "epoch": 0.43933262901907627,
      "grad_norm": 15.0625,
      "learning_rate": 1.2426846583387404e-05,
      "loss": 0.841,
      "num_input_tokens_seen": 2452642304,
      "step": 33060,
      "train_runtime": 39315.2637,
      "train_tokens_per_second": 62383.972
    },
    {
      "epoch": 0.4394655185015382,
      "grad_norm": 27.625,
      "learning_rate": 1.242267106047529e-05,
      "loss": 0.9415,
      "num_input_tokens_seen": 2453430272,
      "step": 33070,
      "train_runtime": 39327.3186,
      "train_tokens_per_second": 62384.886
    },
    {
      "epoch": 0.43959840798400013,
      "grad_norm": 30.5,
      "learning_rate": 1.2418495088784968e-05,
      "loss": 0.6854,
      "num_input_tokens_seen": 2453862912,
      "step": 33080,
      "train_runtime": 39333.6913,
      "train_tokens_per_second": 62385.778
    },
    {
      "epoch": 0.439731297466462,
      "grad_norm": 14.5625,
      "learning_rate": 1.2414318669089995e-05,
      "loss": 0.6609,
      "num_input_tokens_seen": 2454085120,
      "step": 33090,
      "train_runtime": 39337.2466,
      "train_tokens_per_second": 62385.788
    },
    {
      "epoch": 0.43986418694892393,
      "grad_norm": 8.75,
      "learning_rate": 1.2410141802164017e-05,
      "loss": 0.5715,
      "num_input_tokens_seen": 2454195968,
      "step": 33100,
      "train_runtime": 39340.1636,
      "train_tokens_per_second": 62383.98
    },
    {
      "epoch": 0.43999707643138586,
      "grad_norm": 35.5,
      "learning_rate": 1.240596448878076e-05,
      "loss": 0.8722,
      "num_input_tokens_seen": 2456085760,
      "step": 33110,
      "train_runtime": 39370.3928,
      "train_tokens_per_second": 62384.081
    },
    {
      "epoch": 0.44012996591384773,
      "grad_norm": 25.75,
      "learning_rate": 1.2401786729714034e-05,
      "loss": 0.839,
      "num_input_tokens_seen": 2456891648,
      "step": 33120,
      "train_runtime": 39382.8296,
      "train_tokens_per_second": 62384.843
    },
    {
      "epoch": 0.44026285539630966,
      "grad_norm": 33.25,
      "learning_rate": 1.2397608525737733e-05,
      "loss": 0.6737,
      "num_input_tokens_seen": 2457344256,
      "step": 33130,
      "train_runtime": 39390.2437,
      "train_tokens_per_second": 62384.591
    },
    {
      "epoch": 0.4403957448787716,
      "grad_norm": 32.0,
      "learning_rate": 1.2393429877625832e-05,
      "loss": 0.6811,
      "num_input_tokens_seen": 2457567488,
      "step": 33140,
      "train_runtime": 39393.9394,
      "train_tokens_per_second": 62384.406
    },
    {
      "epoch": 0.44052863436123346,
      "grad_norm": 11.25,
      "learning_rate": 1.238925078615239e-05,
      "loss": 0.4963,
      "num_input_tokens_seen": 2457683968,
      "step": 33150,
      "train_runtime": 39396.825,
      "train_tokens_per_second": 62382.793
    },
    {
      "epoch": 0.4406615238436954,
      "grad_norm": 15.5,
      "learning_rate": 1.238507125209154e-05,
      "loss": 0.8641,
      "num_input_tokens_seen": 2459789056,
      "step": 33160,
      "train_runtime": 39430.4751,
      "train_tokens_per_second": 62382.942
    },
    {
      "epoch": 0.4407944133261573,
      "grad_norm": 13.9375,
      "learning_rate": 1.2380891276217514e-05,
      "loss": 0.8938,
      "num_input_tokens_seen": 2460605952,
      "step": 33170,
      "train_runtime": 39442.7269,
      "train_tokens_per_second": 62384.276
    },
    {
      "epoch": 0.4409273028086192,
      "grad_norm": 13.1875,
      "learning_rate": 1.2376710859304607e-05,
      "loss": 0.768,
      "num_input_tokens_seen": 2461070336,
      "step": 33180,
      "train_runtime": 39449.2625,
      "train_tokens_per_second": 62385.712
    },
    {
      "epoch": 0.4410601922910811,
      "grad_norm": 9.75,
      "learning_rate": 1.23725300021272e-05,
      "loss": 0.6412,
      "num_input_tokens_seen": 2461291264,
      "step": 33190,
      "train_runtime": 39453.7291,
      "train_tokens_per_second": 62384.249
    },
    {
      "epoch": 0.44119308177354305,
      "grad_norm": 7.4375,
      "learning_rate": 1.2368348705459773e-05,
      "loss": 0.5532,
      "num_input_tokens_seen": 2461406720,
      "step": 33200,
      "train_runtime": 39456.5352,
      "train_tokens_per_second": 62382.739
    },
    {
      "epoch": 0.4413259712560049,
      "grad_norm": 14.3125,
      "learning_rate": 1.2364166970076864e-05,
      "loss": 0.8067,
      "num_input_tokens_seen": 2463528448,
      "step": 33210,
      "train_runtime": 39489.6307,
      "train_tokens_per_second": 62384.186
    },
    {
      "epoch": 0.44145886073846685,
      "grad_norm": 38.5,
      "learning_rate": 1.2359984796753103e-05,
      "loss": 0.9194,
      "num_input_tokens_seen": 2464423680,
      "step": 33220,
      "train_runtime": 39502.4123,
      "train_tokens_per_second": 62386.663
    },
    {
      "epoch": 0.4415917502209288,
      "grad_norm": 26.625,
      "learning_rate": 1.2355802186263204e-05,
      "loss": 0.7518,
      "num_input_tokens_seen": 2464908800,
      "step": 33230,
      "train_runtime": 39509.2823,
      "train_tokens_per_second": 62388.094
    },
    {
      "epoch": 0.44172463970339065,
      "grad_norm": 12.625,
      "learning_rate": 1.2351619139381959e-05,
      "loss": 0.6165,
      "num_input_tokens_seen": 2465148672,
      "step": 33240,
      "train_runtime": 39513.9568,
      "train_tokens_per_second": 62386.784
    },
    {
      "epoch": 0.4418575291858526,
      "grad_norm": 12.3125,
      "learning_rate": 1.2347435656884234e-05,
      "loss": 0.5294,
      "num_input_tokens_seen": 2465263616,
      "step": 33250,
      "train_runtime": 39516.7813,
      "train_tokens_per_second": 62385.233
    },
    {
      "epoch": 0.4419904186683145,
      "grad_norm": 12.875,
      "learning_rate": 1.2343251739544987e-05,
      "loss": 0.7811,
      "num_input_tokens_seen": 2467367680,
      "step": 33260,
      "train_runtime": 39551.0956,
      "train_tokens_per_second": 62384.307
    },
    {
      "epoch": 0.4421233081507764,
      "grad_norm": 19.75,
      "learning_rate": 1.2339067388139245e-05,
      "loss": 0.9116,
      "num_input_tokens_seen": 2468232704,
      "step": 33270,
      "train_runtime": 39563.8628,
      "train_tokens_per_second": 62386.039
    },
    {
      "epoch": 0.4422561976332383,
      "grad_norm": 12.3125,
      "learning_rate": 1.2334882603442128e-05,
      "loss": 0.6784,
      "num_input_tokens_seen": 2468689664,
      "step": 33280,
      "train_runtime": 39571.2934,
      "train_tokens_per_second": 62385.872
    },
    {
      "epoch": 0.44238908711570024,
      "grad_norm": 24.75,
      "learning_rate": 1.2330697386228832e-05,
      "loss": 0.6748,
      "num_input_tokens_seen": 2468923136,
      "step": 33290,
      "train_runtime": 39575.103,
      "train_tokens_per_second": 62385.766
    },
    {
      "epoch": 0.4425219765981621,
      "grad_norm": 20.25,
      "learning_rate": 1.2326511737274626e-05,
      "loss": 0.5013,
      "num_input_tokens_seen": 2469038336,
      "step": 33300,
      "train_runtime": 39577.9841,
      "train_tokens_per_second": 62384.136
    },
    {
      "epoch": 0.44265486608062404,
      "grad_norm": 22.5,
      "learning_rate": 1.2322325657354866e-05,
      "loss": 0.8725,
      "num_input_tokens_seen": 2471039488,
      "step": 33310,
      "train_runtime": 39610.0914,
      "train_tokens_per_second": 62384.09
    },
    {
      "epoch": 0.44278775556308597,
      "grad_norm": 51.25,
      "learning_rate": 1.2318139147244985e-05,
      "loss": 0.8955,
      "num_input_tokens_seen": 2471848960,
      "step": 33320,
      "train_runtime": 39622.5308,
      "train_tokens_per_second": 62384.934
    },
    {
      "epoch": 0.44292064504554784,
      "grad_norm": 13.8125,
      "learning_rate": 1.2313952207720498e-05,
      "loss": 0.7022,
      "num_input_tokens_seen": 2472311040,
      "step": 33330,
      "train_runtime": 39629.1791,
      "train_tokens_per_second": 62386.128
    },
    {
      "epoch": 0.4430535345280098,
      "grad_norm": 27.5,
      "learning_rate": 1.2309764839556997e-05,
      "loss": 0.6134,
      "num_input_tokens_seen": 2472539904,
      "step": 33340,
      "train_runtime": 39633.9331,
      "train_tokens_per_second": 62384.419
    },
    {
      "epoch": 0.4431864240104717,
      "grad_norm": 8.125,
      "learning_rate": 1.2305577043530156e-05,
      "loss": 0.5126,
      "num_input_tokens_seen": 2472659456,
      "step": 33350,
      "train_runtime": 39636.8838,
      "train_tokens_per_second": 62382.791
    },
    {
      "epoch": 0.4433193134929336,
      "grad_norm": 21.0,
      "learning_rate": 1.2301388820415727e-05,
      "loss": 0.7771,
      "num_input_tokens_seen": 2474748160,
      "step": 33360,
      "train_runtime": 39670.9377,
      "train_tokens_per_second": 62381.892
    },
    {
      "epoch": 0.4434522029753955,
      "grad_norm": 37.25,
      "learning_rate": 1.2297200170989538e-05,
      "loss": 0.899,
      "num_input_tokens_seen": 2475621376,
      "step": 33370,
      "train_runtime": 39683.5175,
      "train_tokens_per_second": 62384.121
    },
    {
      "epoch": 0.44358509245785743,
      "grad_norm": 15.125,
      "learning_rate": 1.2293011096027501e-05,
      "loss": 0.7555,
      "num_input_tokens_seen": 2476068864,
      "step": 33380,
      "train_runtime": 39689.9923,
      "train_tokens_per_second": 62385.219
    },
    {
      "epoch": 0.4437179819403193,
      "grad_norm": 10.0625,
      "learning_rate": 1.2288821596305605e-05,
      "loss": 0.5954,
      "num_input_tokens_seen": 2476290560,
      "step": 33390,
      "train_runtime": 39694.6024,
      "train_tokens_per_second": 62383.559
    },
    {
      "epoch": 0.44385087142278123,
      "grad_norm": 19.75,
      "learning_rate": 1.2284631672599921e-05,
      "loss": 0.5932,
      "num_input_tokens_seen": 2476402688,
      "step": 33400,
      "train_runtime": 39697.3902,
      "train_tokens_per_second": 62382.002
    },
    {
      "epoch": 0.44398376090524316,
      "grad_norm": 21.5,
      "learning_rate": 1.228044132568659e-05,
      "loss": 0.8915,
      "num_input_tokens_seen": 2478468352,
      "step": 33410,
      "train_runtime": 39730.7616,
      "train_tokens_per_second": 62381.597
    },
    {
      "epoch": 0.4441166503877051,
      "grad_norm": 19.0,
      "learning_rate": 1.2276250556341836e-05,
      "loss": 0.942,
      "num_input_tokens_seen": 2479355904,
      "step": 33420,
      "train_runtime": 39743.9321,
      "train_tokens_per_second": 62383.256
    },
    {
      "epoch": 0.44424953987016697,
      "grad_norm": 11.5,
      "learning_rate": 1.2272059365341961e-05,
      "loss": 0.728,
      "num_input_tokens_seen": 2479860992,
      "step": 33430,
      "train_runtime": 39750.9472,
      "train_tokens_per_second": 62384.954
    },
    {
      "epoch": 0.4443824293526289,
      "grad_norm": 17.5,
      "learning_rate": 1.2267867753463352e-05,
      "loss": 0.6433,
      "num_input_tokens_seen": 2480096000,
      "step": 33440,
      "train_runtime": 39754.8147,
      "train_tokens_per_second": 62384.796
    },
    {
      "epoch": 0.4445153188350908,
      "grad_norm": 484.0,
      "learning_rate": 1.2263675721482462e-05,
      "loss": 0.522,
      "num_input_tokens_seen": 2480213248,
      "step": 33450,
      "train_runtime": 39758.7383,
      "train_tokens_per_second": 62381.588
    },
    {
      "epoch": 0.4446482083175527,
      "grad_norm": 15.6875,
      "learning_rate": 1.2259483270175831e-05,
      "loss": 0.7758,
      "num_input_tokens_seen": 2482413056,
      "step": 33460,
      "train_runtime": 39792.2353,
      "train_tokens_per_second": 62384.358
    },
    {
      "epoch": 0.4447810978000146,
      "grad_norm": 27.25,
      "learning_rate": 1.225529040032007e-05,
      "loss": 0.8353,
      "num_input_tokens_seen": 2483254528,
      "step": 33470,
      "train_runtime": 39804.9403,
      "train_tokens_per_second": 62385.586
    },
    {
      "epoch": 0.44491398728247655,
      "grad_norm": 15.6875,
      "learning_rate": 1.2251097112691873e-05,
      "loss": 0.7284,
      "num_input_tokens_seen": 2483757056,
      "step": 33480,
      "train_runtime": 39811.8726,
      "train_tokens_per_second": 62387.346
    },
    {
      "epoch": 0.4450468767649384,
      "grad_norm": 17.625,
      "learning_rate": 1.224690340806801e-05,
      "loss": 0.6433,
      "num_input_tokens_seen": 2483983360,
      "step": 33490,
      "train_runtime": 39815.5495,
      "train_tokens_per_second": 62387.268
    },
    {
      "epoch": 0.44517976624740035,
      "grad_norm": 8.5625,
      "learning_rate": 1.2242709287225323e-05,
      "loss": 0.6073,
      "num_input_tokens_seen": 2484100352,
      "step": 33500,
      "train_runtime": 39819.091,
      "train_tokens_per_second": 62384.657
    },
    {
      "epoch": 0.4453126557298623,
      "grad_norm": 26.625,
      "learning_rate": 1.2238514750940742e-05,
      "loss": 0.7964,
      "num_input_tokens_seen": 2486183680,
      "step": 33510,
      "train_runtime": 39852.732,
      "train_tokens_per_second": 62384.272
    },
    {
      "epoch": 0.44544554521232416,
      "grad_norm": 133.0,
      "learning_rate": 1.2234319799991261e-05,
      "loss": 0.8338,
      "num_input_tokens_seen": 2487052032,
      "step": 33520,
      "train_runtime": 39865.6538,
      "train_tokens_per_second": 62385.833
    },
    {
      "epoch": 0.4455784346947861,
      "grad_norm": 322.0,
      "learning_rate": 1.223012443515396e-05,
      "loss": 0.7771,
      "num_input_tokens_seen": 2487516672,
      "step": 33530,
      "train_runtime": 39872.4813,
      "train_tokens_per_second": 62386.804
    },
    {
      "epoch": 0.445711324177248,
      "grad_norm": 23.0,
      "learning_rate": 1.2225928657205994e-05,
      "loss": 0.5891,
      "num_input_tokens_seen": 2487745280,
      "step": 33540,
      "train_runtime": 39877.0376,
      "train_tokens_per_second": 62385.408
    },
    {
      "epoch": 0.4458442136597099,
      "grad_norm": 17.25,
      "learning_rate": 1.2221732466924592e-05,
      "loss": 0.5838,
      "num_input_tokens_seen": 2487862272,
      "step": 33550,
      "train_runtime": 39879.9268,
      "train_tokens_per_second": 62383.822
    },
    {
      "epoch": 0.4459771031421718,
      "grad_norm": 13.0625,
      "learning_rate": 1.2217535865087062e-05,
      "loss": 0.8772,
      "num_input_tokens_seen": 2489894656,
      "step": 33560,
      "train_runtime": 39913.6263,
      "train_tokens_per_second": 62382.071
    },
    {
      "epoch": 0.44610999262463374,
      "grad_norm": 102.5,
      "learning_rate": 1.2213338852470788e-05,
      "loss": 0.8239,
      "num_input_tokens_seen": 2490772224,
      "step": 33570,
      "train_runtime": 39926.7047,
      "train_tokens_per_second": 62383.616
    },
    {
      "epoch": 0.4462428821070956,
      "grad_norm": 36.25,
      "learning_rate": 1.2209141429853226e-05,
      "loss": 0.675,
      "num_input_tokens_seen": 2491279872,
      "step": 33580,
      "train_runtime": 39933.8774,
      "train_tokens_per_second": 62385.123
    },
    {
      "epoch": 0.44637577158955755,
      "grad_norm": 20.25,
      "learning_rate": 1.2204943598011912e-05,
      "loss": 0.6428,
      "num_input_tokens_seen": 2491513088,
      "step": 33590,
      "train_runtime": 39937.8551,
      "train_tokens_per_second": 62384.749
    },
    {
      "epoch": 0.4465086610720195,
      "grad_norm": 9.1875,
      "learning_rate": 1.2200745357724458e-05,
      "loss": 0.4755,
      "num_input_tokens_seen": 2491626240,
      "step": 33600,
      "train_runtime": 39940.7235,
      "train_tokens_per_second": 62383.102
    },
    {
      "epoch": 0.44664155055448135,
      "grad_norm": 36.0,
      "learning_rate": 1.2196546709768551e-05,
      "loss": 0.821,
      "num_input_tokens_seen": 2493638656,
      "step": 33610,
      "train_runtime": 39974.5295,
      "train_tokens_per_second": 62380.688
    },
    {
      "epoch": 0.4467744400369433,
      "grad_norm": 37.5,
      "learning_rate": 1.2192347654921952e-05,
      "loss": 0.9785,
      "num_input_tokens_seen": 2494477568,
      "step": 33620,
      "train_runtime": 39987.1035,
      "train_tokens_per_second": 62382.052
    },
    {
      "epoch": 0.4469073295194052,
      "grad_norm": 31.375,
      "learning_rate": 1.2188148193962497e-05,
      "loss": 0.8161,
      "num_input_tokens_seen": 2494942720,
      "step": 33630,
      "train_runtime": 39993.9184,
      "train_tokens_per_second": 62383.053
    },
    {
      "epoch": 0.4470402190018671,
      "grad_norm": 17.25,
      "learning_rate": 1.21839483276681e-05,
      "loss": 0.6624,
      "num_input_tokens_seen": 2495168768,
      "step": 33640,
      "train_runtime": 39997.5298,
      "train_tokens_per_second": 62383.072
    },
    {
      "epoch": 0.447173108484329,
      "grad_norm": 15.8125,
      "learning_rate": 1.2179748056816754e-05,
      "loss": 0.5214,
      "num_input_tokens_seen": 2495280640,
      "step": 33650,
      "train_runtime": 40000.1319,
      "train_tokens_per_second": 62381.81
    },
    {
      "epoch": 0.44730599796679094,
      "grad_norm": 23.75,
      "learning_rate": 1.2175547382186511e-05,
      "loss": 0.7837,
      "num_input_tokens_seen": 2497255936,
      "step": 33660,
      "train_runtime": 40033.2452,
      "train_tokens_per_second": 62379.553
    },
    {
      "epoch": 0.4474388874492528,
      "grad_norm": 29.25,
      "learning_rate": 1.217134630455552e-05,
      "loss": 0.908,
      "num_input_tokens_seen": 2498107136,
      "step": 33670,
      "train_runtime": 40046.4804,
      "train_tokens_per_second": 62380.192
    },
    {
      "epoch": 0.44757177693171474,
      "grad_norm": 16.625,
      "learning_rate": 1.2167144824701984e-05,
      "loss": 0.7673,
      "num_input_tokens_seen": 2498588928,
      "step": 33680,
      "train_runtime": 40053.1388,
      "train_tokens_per_second": 62381.851
    },
    {
      "epoch": 0.44770466641417667,
      "grad_norm": 17.0,
      "learning_rate": 1.2162942943404192e-05,
      "loss": 0.6733,
      "num_input_tokens_seen": 2498808320,
      "step": 33690,
      "train_runtime": 40056.7186,
      "train_tokens_per_second": 62381.753
    },
    {
      "epoch": 0.44783755589663854,
      "grad_norm": 16.25,
      "learning_rate": 1.2158740661440509e-05,
      "loss": 0.4586,
      "num_input_tokens_seen": 2498921472,
      "step": 33700,
      "train_runtime": 40059.4381,
      "train_tokens_per_second": 62380.343
    },
    {
      "epoch": 0.44797044537910047,
      "grad_norm": 17.375,
      "learning_rate": 1.2154537979589366e-05,
      "loss": 0.8367,
      "num_input_tokens_seen": 2500993280,
      "step": 33710,
      "train_runtime": 40093.7434,
      "train_tokens_per_second": 62378.642
    },
    {
      "epoch": 0.4481033348615624,
      "grad_norm": 20.375,
      "learning_rate": 1.2150334898629277e-05,
      "loss": 0.9124,
      "num_input_tokens_seen": 2501840896,
      "step": 33720,
      "train_runtime": 40106.0154,
      "train_tokens_per_second": 62380.69
    },
    {
      "epoch": 0.44823622434402427,
      "grad_norm": 23.25,
      "learning_rate": 1.2146131419338823e-05,
      "loss": 0.777,
      "num_input_tokens_seen": 2502281216,
      "step": 33730,
      "train_runtime": 40112.554,
      "train_tokens_per_second": 62381.498
    },
    {
      "epoch": 0.4483691138264862,
      "grad_norm": 11.875,
      "learning_rate": 1.2141927542496656e-05,
      "loss": 0.616,
      "num_input_tokens_seen": 2502512384,
      "step": 33740,
      "train_runtime": 40116.2018,
      "train_tokens_per_second": 62381.588
    },
    {
      "epoch": 0.4485020033089481,
      "grad_norm": 13.875,
      "learning_rate": 1.2137723268881512e-05,
      "loss": 0.4521,
      "num_input_tokens_seen": 2502628352,
      "step": 33750,
      "train_runtime": 40119.0108,
      "train_tokens_per_second": 62380.111
    },
    {
      "epoch": 0.44863489279141,
      "grad_norm": 22.875,
      "learning_rate": 1.2133518599272194e-05,
      "loss": 0.88,
      "num_input_tokens_seen": 2504832256,
      "step": 33760,
      "train_runtime": 40155.0452,
      "train_tokens_per_second": 62379.017
    },
    {
      "epoch": 0.44876778227387193,
      "grad_norm": 41.75,
      "learning_rate": 1.2129313534447583e-05,
      "loss": 0.9033,
      "num_input_tokens_seen": 2505707520,
      "step": 33770,
      "train_runtime": 40169.1448,
      "train_tokens_per_second": 62378.911
    },
    {
      "epoch": 0.44890067175633386,
      "grad_norm": 14.5625,
      "learning_rate": 1.2125108075186622e-05,
      "loss": 0.7763,
      "num_input_tokens_seen": 2506192640,
      "step": 33780,
      "train_runtime": 40176.149,
      "train_tokens_per_second": 62380.111
    },
    {
      "epoch": 0.44903356123879573,
      "grad_norm": 26.75,
      "learning_rate": 1.212090222226834e-05,
      "loss": 0.6373,
      "num_input_tokens_seen": 2506425600,
      "step": 33790,
      "train_runtime": 40179.884,
      "train_tokens_per_second": 62380.11
    },
    {
      "epoch": 0.44916645072125766,
      "grad_norm": 11.5625,
      "learning_rate": 1.2116695976471835e-05,
      "loss": 0.5494,
      "num_input_tokens_seen": 2506543872,
      "step": 33800,
      "train_runtime": 40182.7332,
      "train_tokens_per_second": 62378.631
    },
    {
      "epoch": 0.4492993402037196,
      "grad_norm": 42.0,
      "learning_rate": 1.2112489338576274e-05,
      "loss": 0.8198,
      "num_input_tokens_seen": 2508641280,
      "step": 33810,
      "train_runtime": 40216.9075,
      "train_tokens_per_second": 62377.777
    },
    {
      "epoch": 0.44943222968618146,
      "grad_norm": 23.0,
      "learning_rate": 1.21082823093609e-05,
      "loss": 0.9301,
      "num_input_tokens_seen": 2509519104,
      "step": 33820,
      "train_runtime": 40230.1203,
      "train_tokens_per_second": 62379.11
    },
    {
      "epoch": 0.4495651191686434,
      "grad_norm": 50.25,
      "learning_rate": 1.2104074889605029e-05,
      "loss": 0.821,
      "num_input_tokens_seen": 2510025472,
      "step": 33830,
      "train_runtime": 40237.3349,
      "train_tokens_per_second": 62380.51
    },
    {
      "epoch": 0.4496980086511053,
      "grad_norm": 11.125,
      "learning_rate": 1.2099867080088043e-05,
      "loss": 0.6448,
      "num_input_tokens_seen": 2510260992,
      "step": 33840,
      "train_runtime": 40241.0367,
      "train_tokens_per_second": 62380.624
    },
    {
      "epoch": 0.4498308981335672,
      "grad_norm": 10.9375,
      "learning_rate": 1.2095658881589405e-05,
      "loss": 0.5212,
      "num_input_tokens_seen": 2510373120,
      "step": 33850,
      "train_runtime": 40243.7287,
      "train_tokens_per_second": 62379.238
    },
    {
      "epoch": 0.4499637876160291,
      "grad_norm": 30.75,
      "learning_rate": 1.2091450294888646e-05,
      "loss": 0.7599,
      "num_input_tokens_seen": 2512523264,
      "step": 33860,
      "train_runtime": 40277.5245,
      "train_tokens_per_second": 62380.28
    },
    {
      "epoch": 0.45009667709849105,
      "grad_norm": 24.75,
      "learning_rate": 1.208724132076537e-05,
      "loss": 0.9196,
      "num_input_tokens_seen": 2513360896,
      "step": 33870,
      "train_runtime": 40290.3285,
      "train_tokens_per_second": 62381.246
    },
    {
      "epoch": 0.450229566580953,
      "grad_norm": 60.25,
      "learning_rate": 1.2083031959999252e-05,
      "loss": 0.6898,
      "num_input_tokens_seen": 2513821184,
      "step": 33880,
      "train_runtime": 40297.7422,
      "train_tokens_per_second": 62381.192
    },
    {
      "epoch": 0.45036245606341485,
      "grad_norm": 18.5,
      "learning_rate": 1.2078822213370036e-05,
      "loss": 0.6571,
      "num_input_tokens_seen": 2514052096,
      "step": 33890,
      "train_runtime": 40301.4562,
      "train_tokens_per_second": 62381.173
    },
    {
      "epoch": 0.4504953455458768,
      "grad_norm": 15.0625,
      "learning_rate": 1.2074612081657541e-05,
      "loss": 0.5777,
      "num_input_tokens_seen": 2514167296,
      "step": 33900,
      "train_runtime": 40304.2552,
      "train_tokens_per_second": 62379.699
    },
    {
      "epoch": 0.4506282350283387,
      "grad_norm": 24.5,
      "learning_rate": 1.2070401565641658e-05,
      "loss": 0.8654,
      "num_input_tokens_seen": 2516237312,
      "step": 33910,
      "train_runtime": 40337.1472,
      "train_tokens_per_second": 62380.151
    },
    {
      "epoch": 0.4507611245108006,
      "grad_norm": 29.0,
      "learning_rate": 1.2066190666102345e-05,
      "loss": 0.9179,
      "num_input_tokens_seen": 2517067264,
      "step": 33920,
      "train_runtime": 40349.7248,
      "train_tokens_per_second": 62381.275
    },
    {
      "epoch": 0.4508940139932625,
      "grad_norm": 20.0,
      "learning_rate": 1.2061979383819638e-05,
      "loss": 0.7942,
      "num_input_tokens_seen": 2517554176,
      "step": 33930,
      "train_runtime": 40357.2497,
      "train_tokens_per_second": 62381.708
    },
    {
      "epoch": 0.45102690347572444,
      "grad_norm": 18.5,
      "learning_rate": 1.2057767719573631e-05,
      "loss": 0.6994,
      "num_input_tokens_seen": 2517793792,
      "step": 33940,
      "train_runtime": 40360.9921,
      "train_tokens_per_second": 62381.861
    },
    {
      "epoch": 0.4511597929581863,
      "grad_norm": 15.875,
      "learning_rate": 1.2053555674144506e-05,
      "loss": 0.468,
      "num_input_tokens_seen": 2517908480,
      "step": 33950,
      "train_runtime": 40363.9084,
      "train_tokens_per_second": 62380.195
    },
    {
      "epoch": 0.45129268244064824,
      "grad_norm": 89.0,
      "learning_rate": 1.2049343248312504e-05,
      "loss": 0.9222,
      "num_input_tokens_seen": 2519889664,
      "step": 33960,
      "train_runtime": 40395.8106,
      "train_tokens_per_second": 62379.975
    },
    {
      "epoch": 0.45142557192311017,
      "grad_norm": 32.25,
      "learning_rate": 1.204513044285794e-05,
      "loss": 0.9634,
      "num_input_tokens_seen": 2520717312,
      "step": 33970,
      "train_runtime": 40408.3618,
      "train_tokens_per_second": 62381.081
    },
    {
      "epoch": 0.45155846140557204,
      "grad_norm": 13.0,
      "learning_rate": 1.2040917258561195e-05,
      "loss": 0.7262,
      "num_input_tokens_seen": 2521180160,
      "step": 33980,
      "train_runtime": 40415.0984,
      "train_tokens_per_second": 62382.136
    },
    {
      "epoch": 0.45169135088803397,
      "grad_norm": 24.75,
      "learning_rate": 1.203670369620273e-05,
      "loss": 0.6387,
      "num_input_tokens_seen": 2521414912,
      "step": 33990,
      "train_runtime": 40418.7412,
      "train_tokens_per_second": 62382.322
    },
    {
      "epoch": 0.4518242403704959,
      "grad_norm": 36.0,
      "learning_rate": 1.2032489756563064e-05,
      "loss": 0.5365,
      "num_input_tokens_seen": 2521527296,
      "step": 34000,
      "train_runtime": 40421.537,
      "train_tokens_per_second": 62380.787
    },
    {
      "epoch": 0.4519571298529578,
      "grad_norm": 162.0,
      "learning_rate": 1.2028275440422792e-05,
      "loss": 0.8003,
      "num_input_tokens_seen": 2523508992,
      "step": 34010,
      "train_runtime": 40454.8741,
      "train_tokens_per_second": 62378.367
    },
    {
      "epoch": 0.4520900193354197,
      "grad_norm": 11.625,
      "learning_rate": 1.202406074856258e-05,
      "loss": 0.9454,
      "num_input_tokens_seen": 2524404480,
      "step": 34020,
      "train_runtime": 40467.6874,
      "train_tokens_per_second": 62380.745
    },
    {
      "epoch": 0.45222290881788163,
      "grad_norm": 14.125,
      "learning_rate": 1.2019845681763166e-05,
      "loss": 0.7197,
      "num_input_tokens_seen": 2524878592,
      "step": 34030,
      "train_runtime": 40474.5432,
      "train_tokens_per_second": 62381.892
    },
    {
      "epoch": 0.4523557983003435,
      "grad_norm": 23.25,
      "learning_rate": 1.2015630240805349e-05,
      "loss": 0.7046,
      "num_input_tokens_seen": 2525108224,
      "step": 34040,
      "train_runtime": 40478.2433,
      "train_tokens_per_second": 62381.863
    },
    {
      "epoch": 0.45248868778280543,
      "grad_norm": 31.625,
      "learning_rate": 1.2011414426470005e-05,
      "loss": 0.6074,
      "num_input_tokens_seen": 2525224960,
      "step": 34050,
      "train_runtime": 40480.7382,
      "train_tokens_per_second": 62380.902
    },
    {
      "epoch": 0.45262157726526736,
      "grad_norm": 28.0,
      "learning_rate": 1.2007198239538072e-05,
      "loss": 0.8682,
      "num_input_tokens_seen": 2527270400,
      "step": 34060,
      "train_runtime": 40513.6935,
      "train_tokens_per_second": 62380.647
    },
    {
      "epoch": 0.45275446674772923,
      "grad_norm": 22.5,
      "learning_rate": 1.200298168079056e-05,
      "loss": 0.898,
      "num_input_tokens_seen": 2528157440,
      "step": 34070,
      "train_runtime": 40526.3919,
      "train_tokens_per_second": 62382.988
    },
    {
      "epoch": 0.45288735623019116,
      "grad_norm": 43.75,
      "learning_rate": 1.1998764751008553e-05,
      "loss": 0.726,
      "num_input_tokens_seen": 2528659456,
      "step": 34080,
      "train_runtime": 40533.1774,
      "train_tokens_per_second": 62384.931
    },
    {
      "epoch": 0.4530202457126531,
      "grad_norm": 15.3125,
      "learning_rate": 1.1994547450973201e-05,
      "loss": 0.6602,
      "num_input_tokens_seen": 2528891136,
      "step": 34090,
      "train_runtime": 40536.672,
      "train_tokens_per_second": 62385.268
    },
    {
      "epoch": 0.45315313519511496,
      "grad_norm": 9.25,
      "learning_rate": 1.199032978146572e-05,
      "loss": 0.5407,
      "num_input_tokens_seen": 2529005824,
      "step": 34100,
      "train_runtime": 40539.0953,
      "train_tokens_per_second": 62384.368
    },
    {
      "epoch": 0.4532860246775769,
      "grad_norm": 15.6875,
      "learning_rate": 1.198611174326739e-05,
      "loss": 0.795,
      "num_input_tokens_seen": 2531140352,
      "step": 34110,
      "train_runtime": 40574.7854,
      "train_tokens_per_second": 62382.101
    },
    {
      "epoch": 0.4534189141600388,
      "grad_norm": 26.75,
      "learning_rate": 1.1981893337159571e-05,
      "loss": 0.8164,
      "num_input_tokens_seen": 2531978496,
      "step": 34120,
      "train_runtime": 40587.6106,
      "train_tokens_per_second": 62383.039
    },
    {
      "epoch": 0.4535518036425007,
      "grad_norm": 13.5,
      "learning_rate": 1.1977674563923687e-05,
      "loss": 0.8259,
      "num_input_tokens_seen": 2532466944,
      "step": 34130,
      "train_runtime": 40594.3553,
      "train_tokens_per_second": 62384.707
    },
    {
      "epoch": 0.4536846931249626,
      "grad_norm": 23.0,
      "learning_rate": 1.1973455424341223e-05,
      "loss": 0.5709,
      "num_input_tokens_seen": 2532697088,
      "step": 34140,
      "train_runtime": 40597.7919,
      "train_tokens_per_second": 62385.095
    },
    {
      "epoch": 0.45381758260742455,
      "grad_norm": 12.625,
      "learning_rate": 1.1969235919193735e-05,
      "loss": 0.5647,
      "num_input_tokens_seen": 2532819200,
      "step": 34150,
      "train_runtime": 40600.1761,
      "train_tokens_per_second": 62384.439
    },
    {
      "epoch": 0.4539504720898864,
      "grad_norm": 41.5,
      "learning_rate": 1.1965016049262857e-05,
      "loss": 0.8366,
      "num_input_tokens_seen": 2534901504,
      "step": 34160,
      "train_runtime": 40633.8998,
      "train_tokens_per_second": 62383.909
    },
    {
      "epoch": 0.45408336157234835,
      "grad_norm": 18.125,
      "learning_rate": 1.1960795815330273e-05,
      "loss": 0.8623,
      "num_input_tokens_seen": 2535740416,
      "step": 34170,
      "train_runtime": 40646.9302,
      "train_tokens_per_second": 62384.549
    },
    {
      "epoch": 0.4542162510548103,
      "grad_norm": 22.125,
      "learning_rate": 1.195657521817775e-05,
      "loss": 0.8276,
      "num_input_tokens_seen": 2536213248,
      "step": 34180,
      "train_runtime": 40653.7985,
      "train_tokens_per_second": 62385.64
    },
    {
      "epoch": 0.45434914053727216,
      "grad_norm": 38.5,
      "learning_rate": 1.1952354258587116e-05,
      "loss": 0.6958,
      "num_input_tokens_seen": 2536453376,
      "step": 34190,
      "train_runtime": 40657.373,
      "train_tokens_per_second": 62386.062
    },
    {
      "epoch": 0.4544820300197341,
      "grad_norm": 7.75,
      "learning_rate": 1.1948132937340264e-05,
      "loss": 0.5555,
      "num_input_tokens_seen": 2536571904,
      "step": 34200,
      "train_runtime": 40659.7351,
      "train_tokens_per_second": 62385.352
    },
    {
      "epoch": 0.454614919502196,
      "grad_norm": 27.75,
      "learning_rate": 1.1943911255219156e-05,
      "loss": 0.8379,
      "num_input_tokens_seen": 2538652160,
      "step": 34210,
      "train_runtime": 40693.5168,
      "train_tokens_per_second": 62384.683
    },
    {
      "epoch": 0.4547478089846579,
      "grad_norm": 90.5,
      "learning_rate": 1.193968921300582e-05,
      "loss": 0.8747,
      "num_input_tokens_seen": 2539516160,
      "step": 34220,
      "train_runtime": 40707.0193,
      "train_tokens_per_second": 62385.215
    },
    {
      "epoch": 0.4548806984671198,
      "grad_norm": 13.625,
      "learning_rate": 1.1935466811482354e-05,
      "loss": 0.7304,
      "num_input_tokens_seen": 2540008960,
      "step": 34230,
      "train_runtime": 40713.8141,
      "train_tokens_per_second": 62386.908
    },
    {
      "epoch": 0.45501358794958174,
      "grad_norm": 15.1875,
      "learning_rate": 1.1931244051430921e-05,
      "loss": 0.6258,
      "num_input_tokens_seen": 2540237568,
      "step": 34240,
      "train_runtime": 40717.2552,
      "train_tokens_per_second": 62387.25
    },
    {
      "epoch": 0.4551464774320436,
      "grad_norm": 27.625,
      "learning_rate": 1.1927020933633749e-05,
      "loss": 0.5662,
      "num_input_tokens_seen": 2540353280,
      "step": 34250,
      "train_runtime": 40719.6372,
      "train_tokens_per_second": 62386.442
    },
    {
      "epoch": 0.45527936691450555,
      "grad_norm": 35.25,
      "learning_rate": 1.1922797458873127e-05,
      "loss": 0.7726,
      "num_input_tokens_seen": 2542337280,
      "step": 34260,
      "train_runtime": 40751.69,
      "train_tokens_per_second": 62386.058
    },
    {
      "epoch": 0.4554122563969675,
      "grad_norm": 16.5,
      "learning_rate": 1.1918573627931425e-05,
      "loss": 0.9599,
      "num_input_tokens_seen": 2543178496,
      "step": 34270,
      "train_runtime": 40764.7073,
      "train_tokens_per_second": 62386.772
    },
    {
      "epoch": 0.45554514587942935,
      "grad_norm": 13.8125,
      "learning_rate": 1.1914349441591064e-05,
      "loss": 0.8171,
      "num_input_tokens_seen": 2543652096,
      "step": 34280,
      "train_runtime": 40771.4972,
      "train_tokens_per_second": 62387.998
    },
    {
      "epoch": 0.4556780353618913,
      "grad_norm": 10.625,
      "learning_rate": 1.1910124900634541e-05,
      "loss": 0.6421,
      "num_input_tokens_seen": 2543884544,
      "step": 34290,
      "train_runtime": 40775.0017,
      "train_tokens_per_second": 62388.337
    },
    {
      "epoch": 0.4558109248443532,
      "grad_norm": 10.125,
      "learning_rate": 1.1905900005844413e-05,
      "loss": 0.5491,
      "num_input_tokens_seen": 2544001024,
      "step": 34300,
      "train_runtime": 40777.3907,
      "train_tokens_per_second": 62387.538
    },
    {
      "epoch": 0.4559438143268151,
      "grad_norm": 21.5,
      "learning_rate": 1.1901674758003302e-05,
      "loss": 0.8794,
      "num_input_tokens_seen": 2545986304,
      "step": 34310,
      "train_runtime": 40809.4861,
      "train_tokens_per_second": 62387.12
    },
    {
      "epoch": 0.456076703809277,
      "grad_norm": 20.625,
      "learning_rate": 1.1897449157893899e-05,
      "loss": 0.8724,
      "num_input_tokens_seen": 2546818560,
      "step": 34320,
      "train_runtime": 40821.8389,
      "train_tokens_per_second": 62388.629
    },
    {
      "epoch": 0.45620959329173894,
      "grad_norm": 24.625,
      "learning_rate": 1.1893223206298956e-05,
      "loss": 0.7972,
      "num_input_tokens_seen": 2547277312,
      "step": 34330,
      "train_runtime": 40828.2757,
      "train_tokens_per_second": 62390.029
    },
    {
      "epoch": 0.4563424827742008,
      "grad_norm": 42.25,
      "learning_rate": 1.1888996904001298e-05,
      "loss": 0.7191,
      "num_input_tokens_seen": 2547491840,
      "step": 34340,
      "train_runtime": 40831.5616,
      "train_tokens_per_second": 62390.262
    },
    {
      "epoch": 0.45647537225666274,
      "grad_norm": 17.25,
      "learning_rate": 1.1884770251783811e-05,
      "loss": 0.5113,
      "num_input_tokens_seen": 2547602944,
      "step": 34350,
      "train_runtime": 40833.9138,
      "train_tokens_per_second": 62389.389
    },
    {
      "epoch": 0.45660826173912467,
      "grad_norm": 54.5,
      "learning_rate": 1.1880543250429435e-05,
      "loss": 0.846,
      "num_input_tokens_seen": 2549717504,
      "step": 34360,
      "train_runtime": 40867.6142,
      "train_tokens_per_second": 62389.683
    },
    {
      "epoch": 0.4567411512215866,
      "grad_norm": 12.875,
      "learning_rate": 1.1876315900721193e-05,
      "loss": 0.884,
      "num_input_tokens_seen": 2550542848,
      "step": 34370,
      "train_runtime": 40880.2073,
      "train_tokens_per_second": 62390.653
    },
    {
      "epoch": 0.45687404070404847,
      "grad_norm": 36.25,
      "learning_rate": 1.1872088203442162e-05,
      "loss": 0.8041,
      "num_input_tokens_seen": 2551012608,
      "step": 34380,
      "train_runtime": 40886.8609,
      "train_tokens_per_second": 62391.99
    },
    {
      "epoch": 0.4570069301865104,
      "grad_norm": 21.25,
      "learning_rate": 1.1867860159375484e-05,
      "loss": 0.6068,
      "num_input_tokens_seen": 2551235072,
      "step": 34390,
      "train_runtime": 40890.8446,
      "train_tokens_per_second": 62391.352
    },
    {
      "epoch": 0.4571398196689723,
      "grad_norm": 16.125,
      "learning_rate": 1.1863631769304367e-05,
      "loss": 0.5294,
      "num_input_tokens_seen": 2551343872,
      "step": 34400,
      "train_runtime": 40893.1486,
      "train_tokens_per_second": 62390.497
    },
    {
      "epoch": 0.4572727091514342,
      "grad_norm": 16.0,
      "learning_rate": 1.1859403034012083e-05,
      "loss": 0.8444,
      "num_input_tokens_seen": 2553517824,
      "step": 34410,
      "train_runtime": 40926.8137,
      "train_tokens_per_second": 62392.295
    },
    {
      "epoch": 0.4574055986338961,
      "grad_norm": 16.75,
      "learning_rate": 1.1855173954281964e-05,
      "loss": 0.9098,
      "num_input_tokens_seen": 2554364416,
      "step": 34420,
      "train_runtime": 40939.4242,
      "train_tokens_per_second": 62393.755
    },
    {
      "epoch": 0.45753848811635806,
      "grad_norm": 15.3125,
      "learning_rate": 1.1850944530897413e-05,
      "loss": 0.7617,
      "num_input_tokens_seen": 2554821376,
      "step": 34430,
      "train_runtime": 40945.9977,
      "train_tokens_per_second": 62394.899
    },
    {
      "epoch": 0.45767137759881993,
      "grad_norm": 19.75,
      "learning_rate": 1.1846714764641893e-05,
      "loss": 0.6969,
      "num_input_tokens_seen": 2555045120,
      "step": 34440,
      "train_runtime": 40949.3676,
      "train_tokens_per_second": 62395.228
    },
    {
      "epoch": 0.45780426708128186,
      "grad_norm": 18.625,
      "learning_rate": 1.1842484656298932e-05,
      "loss": 0.5548,
      "num_input_tokens_seen": 2555159296,
      "step": 34450,
      "train_runtime": 40952.3055,
      "train_tokens_per_second": 62393.54
    },
    {
      "epoch": 0.4579371565637438,
      "grad_norm": 22.5,
      "learning_rate": 1.1838254206652118e-05,
      "loss": 0.8525,
      "num_input_tokens_seen": 2557050880,
      "step": 34460,
      "train_runtime": 40983.2998,
      "train_tokens_per_second": 62392.508
    },
    {
      "epoch": 0.45807004604620566,
      "grad_norm": 31.125,
      "learning_rate": 1.1834023416485099e-05,
      "loss": 0.8448,
      "num_input_tokens_seen": 2557908992,
      "step": 34470,
      "train_runtime": 40995.6407,
      "train_tokens_per_second": 62394.658
    },
    {
      "epoch": 0.4582029355286676,
      "grad_norm": 35.75,
      "learning_rate": 1.1829792286581598e-05,
      "loss": 0.7566,
      "num_input_tokens_seen": 2558361856,
      "step": 34480,
      "train_runtime": 41002.0874,
      "train_tokens_per_second": 62395.893
    },
    {
      "epoch": 0.4583358250111295,
      "grad_norm": 11.9375,
      "learning_rate": 1.182556081772539e-05,
      "loss": 0.6203,
      "num_input_tokens_seen": 2558582784,
      "step": 34490,
      "train_runtime": 41005.4707,
      "train_tokens_per_second": 62396.133
    },
    {
      "epoch": 0.4584687144935914,
      "grad_norm": 7.5,
      "learning_rate": 1.1821329010700319e-05,
      "loss": 0.5386,
      "num_input_tokens_seen": 2558697216,
      "step": 34500,
      "train_runtime": 41007.7887,
      "train_tokens_per_second": 62395.396
    },
    {
      "epoch": 0.4586016039760533,
      "grad_norm": 20.0,
      "learning_rate": 1.1817096866290294e-05,
      "loss": 0.8484,
      "num_input_tokens_seen": 2560804864,
      "step": 34510,
      "train_runtime": 41041.1295,
      "train_tokens_per_second": 62396.062
    },
    {
      "epoch": 0.45873449345851525,
      "grad_norm": 28.625,
      "learning_rate": 1.1812864385279275e-05,
      "loss": 0.8263,
      "num_input_tokens_seen": 2561662464,
      "step": 34520,
      "train_runtime": 41053.8221,
      "train_tokens_per_second": 62397.661
    },
    {
      "epoch": 0.4588673829409771,
      "grad_norm": 12.0625,
      "learning_rate": 1.1808631568451292e-05,
      "loss": 0.7544,
      "num_input_tokens_seen": 2562149120,
      "step": 34530,
      "train_runtime": 41060.6857,
      "train_tokens_per_second": 62399.083
    },
    {
      "epoch": 0.45900027242343905,
      "grad_norm": 10.1875,
      "learning_rate": 1.1804398416590439e-05,
      "loss": 0.6587,
      "num_input_tokens_seen": 2562389504,
      "step": 34540,
      "train_runtime": 41064.2486,
      "train_tokens_per_second": 62399.522
    },
    {
      "epoch": 0.459133161905901,
      "grad_norm": 11.0,
      "learning_rate": 1.1800164930480869e-05,
      "loss": 0.5416,
      "num_input_tokens_seen": 2562506240,
      "step": 34550,
      "train_runtime": 41067.15,
      "train_tokens_per_second": 62397.956
    },
    {
      "epoch": 0.45926605138836285,
      "grad_norm": 26.625,
      "learning_rate": 1.1795931110906797e-05,
      "loss": 0.8615,
      "num_input_tokens_seen": 2564470272,
      "step": 34560,
      "train_runtime": 41098.4739,
      "train_tokens_per_second": 62398.187
    },
    {
      "epoch": 0.4593989408708248,
      "grad_norm": 22.125,
      "learning_rate": 1.1791696958652501e-05,
      "loss": 0.8741,
      "num_input_tokens_seen": 2565314560,
      "step": 34570,
      "train_runtime": 41110.9525,
      "train_tokens_per_second": 62399.784
    },
    {
      "epoch": 0.4595318303532867,
      "grad_norm": 16.375,
      "learning_rate": 1.178746247450232e-05,
      "loss": 0.7391,
      "num_input_tokens_seen": 2565776896,
      "step": 34580,
      "train_runtime": 41117.6338,
      "train_tokens_per_second": 62400.889
    },
    {
      "epoch": 0.4596647198357486,
      "grad_norm": 16.125,
      "learning_rate": 1.1783227659240652e-05,
      "loss": 0.6515,
      "num_input_tokens_seen": 2565998848,
      "step": 34590,
      "train_runtime": 41120.9974,
      "train_tokens_per_second": 62401.182
    },
    {
      "epoch": 0.4597976093182105,
      "grad_norm": 8.6875,
      "learning_rate": 1.1778992513651963e-05,
      "loss": 0.6224,
      "num_input_tokens_seen": 2566112256,
      "step": 34600,
      "train_runtime": 41123.3757,
      "train_tokens_per_second": 62400.331
    },
    {
      "epoch": 0.45993049880067244,
      "grad_norm": 29.125,
      "learning_rate": 1.1774757038520775e-05,
      "loss": 0.8977,
      "num_input_tokens_seen": 2568263936,
      "step": 34610,
      "train_runtime": 41157.1983,
      "train_tokens_per_second": 62401.331
    },
    {
      "epoch": 0.4600633882831343,
      "grad_norm": 20.875,
      "learning_rate": 1.1770521234631671e-05,
      "loss": 0.8734,
      "num_input_tokens_seen": 2569060608,
      "step": 34620,
      "train_runtime": 41169.3513,
      "train_tokens_per_second": 62402.261
    },
    {
      "epoch": 0.46019627776559624,
      "grad_norm": 79.0,
      "learning_rate": 1.1766285102769293e-05,
      "loss": 0.6956,
      "num_input_tokens_seen": 2569518336,
      "step": 34630,
      "train_runtime": 41175.7448,
      "train_tokens_per_second": 62403.688
    },
    {
      "epoch": 0.46032916724805817,
      "grad_norm": 103.0,
      "learning_rate": 1.176204864371835e-05,
      "loss": 0.5965,
      "num_input_tokens_seen": 2569746944,
      "step": 34640,
      "train_runtime": 41179.1896,
      "train_tokens_per_second": 62404.019
    },
    {
      "epoch": 0.46046205673052004,
      "grad_norm": 19.375,
      "learning_rate": 1.1757811858263606e-05,
      "loss": 0.5113,
      "num_input_tokens_seen": 2569861888,
      "step": 34650,
      "train_runtime": 41181.5964,
      "train_tokens_per_second": 62403.163
    },
    {
      "epoch": 0.46059494621298197,
      "grad_norm": 55.5,
      "learning_rate": 1.1753574747189892e-05,
      "loss": 0.8575,
      "num_input_tokens_seen": 2571946240,
      "step": 34660,
      "train_runtime": 41215.7215,
      "train_tokens_per_second": 62402.068
    },
    {
      "epoch": 0.4607278356954439,
      "grad_norm": 92.5,
      "learning_rate": 1.1749337311282095e-05,
      "loss": 0.8648,
      "num_input_tokens_seen": 2572775680,
      "step": 34670,
      "train_runtime": 41228.3724,
      "train_tokens_per_second": 62403.038
    },
    {
      "epoch": 0.4608607251779058,
      "grad_norm": 26.5,
      "learning_rate": 1.1745099551325154e-05,
      "loss": 0.7157,
      "num_input_tokens_seen": 2573240064,
      "step": 34680,
      "train_runtime": 41235.093,
      "train_tokens_per_second": 62404.129
    },
    {
      "epoch": 0.4609936146603677,
      "grad_norm": 11.6875,
      "learning_rate": 1.1740861468104086e-05,
      "loss": 0.6218,
      "num_input_tokens_seen": 2573468416,
      "step": 34690,
      "train_runtime": 41239.0731,
      "train_tokens_per_second": 62403.644
    },
    {
      "epoch": 0.46112650414282963,
      "grad_norm": 21.875,
      "learning_rate": 1.1736623062403951e-05,
      "loss": 0.537,
      "num_input_tokens_seen": 2573582080,
      "step": 34700,
      "train_runtime": 41241.4578,
      "train_tokens_per_second": 62402.791
    },
    {
      "epoch": 0.4612593936252915,
      "grad_norm": 27.0,
      "learning_rate": 1.1732384335009882e-05,
      "loss": 0.777,
      "num_input_tokens_seen": 2575617536,
      "step": 34710,
      "train_runtime": 41275.3024,
      "train_tokens_per_second": 62400.937
    },
    {
      "epoch": 0.46139228310775343,
      "grad_norm": 26.375,
      "learning_rate": 1.172814528670706e-05,
      "loss": 0.9163,
      "num_input_tokens_seen": 2576428288,
      "step": 34720,
      "train_runtime": 41287.6249,
      "train_tokens_per_second": 62401.95
    },
    {
      "epoch": 0.46152517259021536,
      "grad_norm": 21.875,
      "learning_rate": 1.1723905918280734e-05,
      "loss": 0.7625,
      "num_input_tokens_seen": 2576894208,
      "step": 34730,
      "train_runtime": 41294.0516,
      "train_tokens_per_second": 62403.521
    },
    {
      "epoch": 0.46165806207267723,
      "grad_norm": 13.9375,
      "learning_rate": 1.1719666230516208e-05,
      "loss": 0.6371,
      "num_input_tokens_seen": 2577122304,
      "step": 34740,
      "train_runtime": 41298.0113,
      "train_tokens_per_second": 62403.061
    },
    {
      "epoch": 0.46179095155513916,
      "grad_norm": 25.0,
      "learning_rate": 1.1715426224198844e-05,
      "loss": 0.5181,
      "num_input_tokens_seen": 2577240320,
      "step": 34750,
      "train_runtime": 41300.4214,
      "train_tokens_per_second": 62402.277
    },
    {
      "epoch": 0.4619238410376011,
      "grad_norm": 28.125,
      "learning_rate": 1.1711185900114072e-05,
      "loss": 0.778,
      "num_input_tokens_seen": 2579288576,
      "step": 34760,
      "train_runtime": 41334.8448,
      "train_tokens_per_second": 62399.861
    },
    {
      "epoch": 0.46205673052006296,
      "grad_norm": 16.25,
      "learning_rate": 1.170694525904737e-05,
      "loss": 0.879,
      "num_input_tokens_seen": 2580201984,
      "step": 34770,
      "train_runtime": 41347.8473,
      "train_tokens_per_second": 62402.329
    },
    {
      "epoch": 0.4621896200025249,
      "grad_norm": 18.125,
      "learning_rate": 1.170270430178428e-05,
      "loss": 0.7539,
      "num_input_tokens_seen": 2580681472,
      "step": 34780,
      "train_runtime": 41354.7091,
      "train_tokens_per_second": 62403.57
    },
    {
      "epoch": 0.4623225094849868,
      "grad_norm": 63.75,
      "learning_rate": 1.16984630291104e-05,
      "loss": 0.5791,
      "num_input_tokens_seen": 2580915968,
      "step": 34790,
      "train_runtime": 41358.662,
      "train_tokens_per_second": 62403.275
    },
    {
      "epoch": 0.4624553989674487,
      "grad_norm": 6.5,
      "learning_rate": 1.1694221441811389e-05,
      "loss": 0.5133,
      "num_input_tokens_seen": 2581029120,
      "step": 34800,
      "train_runtime": 41361.0319,
      "train_tokens_per_second": 62402.435
    },
    {
      "epoch": 0.4625882884499106,
      "grad_norm": 18.25,
      "learning_rate": 1.168997954067296e-05,
      "loss": 0.8497,
      "num_input_tokens_seen": 2583090688,
      "step": 34810,
      "train_runtime": 41393.6582,
      "train_tokens_per_second": 62403.054
    },
    {
      "epoch": 0.46272117793237255,
      "grad_norm": 21.75,
      "learning_rate": 1.1685737326480894e-05,
      "loss": 0.8648,
      "num_input_tokens_seen": 2583917056,
      "step": 34820,
      "train_runtime": 41406.2657,
      "train_tokens_per_second": 62404.011
    },
    {
      "epoch": 0.4628540674148345,
      "grad_norm": 40.0,
      "learning_rate": 1.1681494800021018e-05,
      "loss": 0.6541,
      "num_input_tokens_seen": 2584403456,
      "step": 34830,
      "train_runtime": 41412.948,
      "train_tokens_per_second": 62405.687
    },
    {
      "epoch": 0.46298695689729635,
      "grad_norm": 12.375,
      "learning_rate": 1.1677251962079224e-05,
      "loss": 0.596,
      "num_input_tokens_seen": 2584642560,
      "step": 34840,
      "train_runtime": 41416.4999,
      "train_tokens_per_second": 62406.108
    },
    {
      "epoch": 0.4631198463797583,
      "grad_norm": 25.375,
      "learning_rate": 1.1673008813441462e-05,
      "loss": 0.5129,
      "num_input_tokens_seen": 2584759040,
      "step": 34850,
      "train_runtime": 41418.9024,
      "train_tokens_per_second": 62405.3
    },
    {
      "epoch": 0.4632527358622202,
      "grad_norm": 21.75,
      "learning_rate": 1.1668765354893732e-05,
      "loss": 0.8163,
      "num_input_tokens_seen": 2586898432,
      "step": 34860,
      "train_runtime": 41454.0689,
      "train_tokens_per_second": 62403.969
    },
    {
      "epoch": 0.4633856253446821,
      "grad_norm": 14.875,
      "learning_rate": 1.1664521587222102e-05,
      "loss": 0.9352,
      "num_input_tokens_seen": 2587763968,
      "step": 34870,
      "train_runtime": 41466.9504,
      "train_tokens_per_second": 62405.456
    },
    {
      "epoch": 0.463518514827144,
      "grad_norm": 12.75,
      "learning_rate": 1.1660277511212693e-05,
      "loss": 0.7106,
      "num_input_tokens_seen": 2588227584,
      "step": 34880,
      "train_runtime": 41473.635,
      "train_tokens_per_second": 62406.577
    },
    {
      "epoch": 0.46365140430960594,
      "grad_norm": 42.5,
      "learning_rate": 1.1656033127651674e-05,
      "loss": 0.6412,
      "num_input_tokens_seen": 2588451840,
      "step": 34890,
      "train_runtime": 41477.0235,
      "train_tokens_per_second": 62406.885
    },
    {
      "epoch": 0.4637842937920678,
      "grad_norm": 24.375,
      "learning_rate": 1.1651788437325287e-05,
      "loss": 0.5386,
      "num_input_tokens_seen": 2588566272,
      "step": 34900,
      "train_runtime": 41479.3711,
      "train_tokens_per_second": 62406.112
    },
    {
      "epoch": 0.46391718327452974,
      "grad_norm": 21.875,
      "learning_rate": 1.1647543441019819e-05,
      "loss": 0.8732,
      "num_input_tokens_seen": 2590542848,
      "step": 34910,
      "train_runtime": 41512.0302,
      "train_tokens_per_second": 62404.629
    },
    {
      "epoch": 0.46405007275699167,
      "grad_norm": 22.625,
      "learning_rate": 1.164329813952162e-05,
      "loss": 0.8743,
      "num_input_tokens_seen": 2591379456,
      "step": 34920,
      "train_runtime": 41524.6332,
      "train_tokens_per_second": 62405.836
    },
    {
      "epoch": 0.46418296223945354,
      "grad_norm": 22.25,
      "learning_rate": 1.1639052533617096e-05,
      "loss": 0.7204,
      "num_input_tokens_seen": 2591841536,
      "step": 34930,
      "train_runtime": 41531.2743,
      "train_tokens_per_second": 62406.983
    },
    {
      "epoch": 0.4643158517219155,
      "grad_norm": 14.125,
      "learning_rate": 1.1634806624092707e-05,
      "loss": 0.6528,
      "num_input_tokens_seen": 2592068608,
      "step": 34940,
      "train_runtime": 41534.7268,
      "train_tokens_per_second": 62407.263
    },
    {
      "epoch": 0.4644487412043774,
      "grad_norm": 24.875,
      "learning_rate": 1.1630560411734964e-05,
      "loss": 0.5834,
      "num_input_tokens_seen": 2592184320,
      "step": 34950,
      "train_runtime": 41537.1024,
      "train_tokens_per_second": 62406.479
    },
    {
      "epoch": 0.4645816306868393,
      "grad_norm": 56.5,
      "learning_rate": 1.1626313897330443e-05,
      "loss": 0.8012,
      "num_input_tokens_seen": 2594220544,
      "step": 34960,
      "train_runtime": 41569.8581,
      "train_tokens_per_second": 62406.288
    },
    {
      "epoch": 0.4647145201693012,
      "grad_norm": 24.625,
      "learning_rate": 1.1622067081665778e-05,
      "loss": 0.854,
      "num_input_tokens_seen": 2595072768,
      "step": 34970,
      "train_runtime": 41582.3177,
      "train_tokens_per_second": 62408.084
    },
    {
      "epoch": 0.46484740965176313,
      "grad_norm": 14.25,
      "learning_rate": 1.161781996552765e-05,
      "loss": 0.7643,
      "num_input_tokens_seen": 2595539712,
      "step": 34980,
      "train_runtime": 41589.0475,
      "train_tokens_per_second": 62409.213
    },
    {
      "epoch": 0.464980299134225,
      "grad_norm": 15.9375,
      "learning_rate": 1.1613572549702795e-05,
      "loss": 0.6368,
      "num_input_tokens_seen": 2595768320,
      "step": 34990,
      "train_runtime": 41593.0932,
      "train_tokens_per_second": 62408.639
    },
    {
      "epoch": 0.46511318861668693,
      "grad_norm": 10.5,
      "learning_rate": 1.1609324834978018e-05,
      "loss": 0.6159,
      "num_input_tokens_seen": 2595882496,
      "step": 35000,
      "train_runtime": 41595.485,
      "train_tokens_per_second": 62407.795
    },
    {
      "epoch": 0.46511318861668693,
      "eval_accuracy": 0.8195,
      "eval_balanced_accuracy": 0.8177123902233425,
      "eval_invalid_f1": 0.837226079898999,
      "eval_invalid_precision": 0.8410944011596304,
      "eval_invalid_recall": 0.8333931777378815,
      "eval_loss": 0.36159074306488037,
      "eval_macro_f1": 0.8173337222522713,
      "eval_mcc": 0.6347108746549454,
      "eval_runtime": 77.5755,
      "eval_samples_per_second": 128.907,
      "eval_steps_per_second": 32.227,
      "eval_valid_f1": 0.7974413646055436,
      "eval_valid_precision": 0.7929033697835305,
      "eval_valid_recall": 0.8020316027088036,
      "num_input_tokens_seen": 2595882496,
      "step": 35000
    },
    {
      "epoch": 0.46524607809914886,
      "grad_norm": 14.3125,
      "learning_rate": 1.1605076822140163e-05,
      "loss": 0.8453,
      "num_input_tokens_seen": 2597981184,
      "step": 35010,
      "train_runtime": 41710.0135,
      "train_tokens_per_second": 62286.75
    },
    {
      "epoch": 0.46537896758161074,
      "grad_norm": 16.375,
      "learning_rate": 1.1600828511976138e-05,
      "loss": 0.887,
      "num_input_tokens_seen": 2598843648,
      "step": 35020,
      "train_runtime": 41722.7902,
      "train_tokens_per_second": 62288.347
    },
    {
      "epoch": 0.46551185706407266,
      "grad_norm": 12.125,
      "learning_rate": 1.1596579905272906e-05,
      "loss": 0.6973,
      "num_input_tokens_seen": 2599296768,
      "step": 35030,
      "train_runtime": 41729.3804,
      "train_tokens_per_second": 62289.369
    },
    {
      "epoch": 0.4656447465465346,
      "grad_norm": 12.625,
      "learning_rate": 1.1592331002817485e-05,
      "loss": 0.5896,
      "num_input_tokens_seen": 2599522816,
      "step": 35040,
      "train_runtime": 41732.8297,
      "train_tokens_per_second": 62289.637
    },
    {
      "epoch": 0.46577763602899647,
      "grad_norm": 19.0,
      "learning_rate": 1.1588081805396943e-05,
      "loss": 0.5125,
      "num_input_tokens_seen": 2599634432,
      "step": 35050,
      "train_runtime": 41735.1936,
      "train_tokens_per_second": 62288.783
    },
    {
      "epoch": 0.4659105255114584,
      "grad_norm": 14.6875,
      "learning_rate": 1.1583832313798404e-05,
      "loss": 0.7888,
      "num_input_tokens_seen": 2601697792,
      "step": 35060,
      "train_runtime": 41768.9008,
      "train_tokens_per_second": 62287.916
    },
    {
      "epoch": 0.4660434149939203,
      "grad_norm": 18.75,
      "learning_rate": 1.1579582528809055e-05,
      "loss": 0.9031,
      "num_input_tokens_seen": 2602499328,
      "step": 35070,
      "train_runtime": 41781.3552,
      "train_tokens_per_second": 62288.533
    },
    {
      "epoch": 0.4661763044763822,
      "grad_norm": 30.5,
      "learning_rate": 1.1575332451216128e-05,
      "loss": 0.7321,
      "num_input_tokens_seen": 2602940928,
      "step": 35080,
      "train_runtime": 41787.8764,
      "train_tokens_per_second": 62289.38
    },
    {
      "epoch": 0.4663091939588441,
      "grad_norm": 21.0,
      "learning_rate": 1.1571082081806912e-05,
      "loss": 0.647,
      "num_input_tokens_seen": 2603166720,
      "step": 35090,
      "train_runtime": 41791.2416,
      "train_tokens_per_second": 62289.767
    },
    {
      "epoch": 0.46644208344130605,
      "grad_norm": 11.1875,
      "learning_rate": 1.156683142136875e-05,
      "loss": 0.5184,
      "num_input_tokens_seen": 2603282176,
      "step": 35100,
      "train_runtime": 41793.6207,
      "train_tokens_per_second": 62288.984
    },
    {
      "epoch": 0.4665749729237679,
      "grad_norm": 18.0,
      "learning_rate": 1.1562580470689033e-05,
      "loss": 0.8031,
      "num_input_tokens_seen": 2605312000,
      "step": 35110,
      "train_runtime": 41825.3008,
      "train_tokens_per_second": 62290.335
    },
    {
      "epoch": 0.46670786240622986,
      "grad_norm": 18.375,
      "learning_rate": 1.1558329230555221e-05,
      "loss": 0.9691,
      "num_input_tokens_seen": 2606153728,
      "step": 35120,
      "train_runtime": 41838.5069,
      "train_tokens_per_second": 62290.792
    },
    {
      "epoch": 0.4668407518886918,
      "grad_norm": 34.75,
      "learning_rate": 1.1554077701754813e-05,
      "loss": 0.7691,
      "num_input_tokens_seen": 2606657024,
      "step": 35130,
      "train_runtime": 41845.4457,
      "train_tokens_per_second": 62292.49
    },
    {
      "epoch": 0.46697364137115366,
      "grad_norm": 19.875,
      "learning_rate": 1.154982588507537e-05,
      "loss": 0.6098,
      "num_input_tokens_seen": 2606889472,
      "step": 35140,
      "train_runtime": 41848.9561,
      "train_tokens_per_second": 62292.82
    },
    {
      "epoch": 0.4671065308536156,
      "grad_norm": 10.3125,
      "learning_rate": 1.15455737813045e-05,
      "loss": 0.4848,
      "num_input_tokens_seen": 2607001600,
      "step": 35150,
      "train_runtime": 41851.2967,
      "train_tokens_per_second": 62292.015
    },
    {
      "epoch": 0.4672394203360775,
      "grad_norm": 17.0,
      "learning_rate": 1.1541321391229868e-05,
      "loss": 0.8765,
      "num_input_tokens_seen": 2608939008,
      "step": 35160,
      "train_runtime": 41881.7989,
      "train_tokens_per_second": 62292.907
    },
    {
      "epoch": 0.4673723098185394,
      "grad_norm": 20.375,
      "learning_rate": 1.1537068715639195e-05,
      "loss": 0.9048,
      "num_input_tokens_seen": 2609754624,
      "step": 35170,
      "train_runtime": 41894.4656,
      "train_tokens_per_second": 62293.541
    },
    {
      "epoch": 0.4675051993010013,
      "grad_norm": 14.5,
      "learning_rate": 1.1532815755320243e-05,
      "loss": 0.7832,
      "num_input_tokens_seen": 2610194688,
      "step": 35180,
      "train_runtime": 41900.887,
      "train_tokens_per_second": 62294.497
    },
    {
      "epoch": 0.46763808878346325,
      "grad_norm": 20.875,
      "learning_rate": 1.1528562511060843e-05,
      "loss": 0.674,
      "num_input_tokens_seen": 2610412032,
      "step": 35190,
      "train_runtime": 41904.2513,
      "train_tokens_per_second": 62294.682
    },
    {
      "epoch": 0.4677709782659251,
      "grad_norm": 14.875,
      "learning_rate": 1.1524308983648871e-05,
      "loss": 0.5651,
      "num_input_tokens_seen": 2610525696,
      "step": 35200,
      "train_runtime": 41906.6267,
      "train_tokens_per_second": 62293.864
    },
    {
      "epoch": 0.46790386774838705,
      "grad_norm": 17.625,
      "learning_rate": 1.1520055173872247e-05,
      "loss": 0.8401,
      "num_input_tokens_seen": 2612657664,
      "step": 35210,
      "train_runtime": 41940.7046,
      "train_tokens_per_second": 62294.081
    },
    {
      "epoch": 0.468036757230849,
      "grad_norm": 14.875,
      "learning_rate": 1.1515801082518958e-05,
      "loss": 0.9634,
      "num_input_tokens_seen": 2613520384,
      "step": 35220,
      "train_runtime": 41953.983,
      "train_tokens_per_second": 62294.929
    },
    {
      "epoch": 0.46816964671331085,
      "grad_norm": 31.75,
      "learning_rate": 1.1511546710377033e-05,
      "loss": 0.7252,
      "num_input_tokens_seen": 2614012672,
      "step": 35230,
      "train_runtime": 41961.0084,
      "train_tokens_per_second": 62296.231
    },
    {
      "epoch": 0.4683025361957728,
      "grad_norm": 24.0,
      "learning_rate": 1.1507292058234563e-05,
      "loss": 0.6706,
      "num_input_tokens_seen": 2614259456,
      "step": 35240,
      "train_runtime": 41964.584,
      "train_tokens_per_second": 62296.804
    },
    {
      "epoch": 0.4684354256782347,
      "grad_norm": 15.375,
      "learning_rate": 1.1503037126879675e-05,
      "loss": 0.5621,
      "num_input_tokens_seen": 2614381824,
      "step": 35250,
      "train_runtime": 41966.9636,
      "train_tokens_per_second": 62296.187
    },
    {
      "epoch": 0.4685683151606966,
      "grad_norm": 36.5,
      "learning_rate": 1.1498781917100563e-05,
      "loss": 0.8574,
      "num_input_tokens_seen": 2616492544,
      "step": 35260,
      "train_runtime": 41999.1473,
      "train_tokens_per_second": 62298.706
    },
    {
      "epoch": 0.4687012046431585,
      "grad_norm": 11.3125,
      "learning_rate": 1.1494526429685466e-05,
      "loss": 0.8437,
      "num_input_tokens_seen": 2617292800,
      "step": 35270,
      "train_runtime": 42011.7867,
      "train_tokens_per_second": 62299.012
    },
    {
      "epoch": 0.46883409412562044,
      "grad_norm": 23.875,
      "learning_rate": 1.1490270665422672e-05,
      "loss": 0.7389,
      "num_input_tokens_seen": 2617772544,
      "step": 35280,
      "train_runtime": 42019.1329,
      "train_tokens_per_second": 62299.537
    },
    {
      "epoch": 0.4689669836080823,
      "grad_norm": 25.5,
      "learning_rate": 1.1486014625100525e-05,
      "loss": 0.6338,
      "num_input_tokens_seen": 2618009344,
      "step": 35290,
      "train_runtime": 42022.7063,
      "train_tokens_per_second": 62299.875
    },
    {
      "epoch": 0.46909987309054424,
      "grad_norm": 18.125,
      "learning_rate": 1.1481758309507425e-05,
      "loss": 0.551,
      "num_input_tokens_seen": 2618126080,
      "step": 35300,
      "train_runtime": 42025.0719,
      "train_tokens_per_second": 62299.146
    },
    {
      "epoch": 0.46923276257300617,
      "grad_norm": 18.625,
      "learning_rate": 1.1477501719431805e-05,
      "loss": 0.8168,
      "num_input_tokens_seen": 2620228608,
      "step": 35310,
      "train_runtime": 42059.9422,
      "train_tokens_per_second": 62297.485
    },
    {
      "epoch": 0.4693656520554681,
      "grad_norm": 36.25,
      "learning_rate": 1.147324485566217e-05,
      "loss": 0.8829,
      "num_input_tokens_seen": 2621066496,
      "step": 35320,
      "train_runtime": 42072.7672,
      "train_tokens_per_second": 62298.41
    },
    {
      "epoch": 0.46949854153792997,
      "grad_norm": 17.0,
      "learning_rate": 1.1468987718987057e-05,
      "loss": 0.8346,
      "num_input_tokens_seen": 2621569792,
      "step": 35330,
      "train_runtime": 42080.4052,
      "train_tokens_per_second": 62299.062
    },
    {
      "epoch": 0.4696314310203919,
      "grad_norm": 29.125,
      "learning_rate": 1.1464730310195074e-05,
      "loss": 0.6681,
      "num_input_tokens_seen": 2621805824,
      "step": 35340,
      "train_runtime": 42083.983,
      "train_tokens_per_second": 62299.375
    },
    {
      "epoch": 0.4697643205028538,
      "grad_norm": 9.1875,
      "learning_rate": 1.146047263007486e-05,
      "loss": 0.4916,
      "num_input_tokens_seen": 2621922048,
      "step": 35350,
      "train_runtime": 42086.3092,
      "train_tokens_per_second": 62298.693
    },
    {
      "epoch": 0.4698972099853157,
      "grad_norm": 23.125,
      "learning_rate": 1.1456214679415112e-05,
      "loss": 0.7442,
      "num_input_tokens_seen": 2623957504,
      "step": 35360,
      "train_runtime": 42118.9902,
      "train_tokens_per_second": 62298.68
    },
    {
      "epoch": 0.47003009946777763,
      "grad_norm": 23.0,
      "learning_rate": 1.145195645900458e-05,
      "loss": 0.8878,
      "num_input_tokens_seen": 2624792320,
      "step": 35370,
      "train_runtime": 42131.6464,
      "train_tokens_per_second": 62299.78
    },
    {
      "epoch": 0.47016298895023956,
      "grad_norm": 24.5,
      "learning_rate": 1.1447697969632061e-05,
      "loss": 0.7768,
      "num_input_tokens_seen": 2625260544,
      "step": 35380,
      "train_runtime": 42138.2797,
      "train_tokens_per_second": 62301.085
    },
    {
      "epoch": 0.47029587843270143,
      "grad_norm": 14.375,
      "learning_rate": 1.1443439212086402e-05,
      "loss": 0.577,
      "num_input_tokens_seen": 2625488640,
      "step": 35390,
      "train_runtime": 42141.6821,
      "train_tokens_per_second": 62301.468
    },
    {
      "epoch": 0.47042876791516336,
      "grad_norm": 14.3125,
      "learning_rate": 1.1439180187156503e-05,
      "loss": 0.4862,
      "num_input_tokens_seen": 2625600000,
      "step": 35400,
      "train_runtime": 42144.5956,
      "train_tokens_per_second": 62299.803
    },
    {
      "epoch": 0.4705616573976253,
      "grad_norm": 13.5,
      "learning_rate": 1.1434920895631301e-05,
      "loss": 0.8288,
      "num_input_tokens_seen": 2627653888,
      "step": 35410,
      "train_runtime": 42177.8612,
      "train_tokens_per_second": 62299.363
    },
    {
      "epoch": 0.47069454688008716,
      "grad_norm": 30.375,
      "learning_rate": 1.1430661338299805e-05,
      "loss": 0.9412,
      "num_input_tokens_seen": 2628491264,
      "step": 35420,
      "train_runtime": 42190.4867,
      "train_tokens_per_second": 62300.568
    },
    {
      "epoch": 0.4708274363625491,
      "grad_norm": 43.0,
      "learning_rate": 1.1426401515951047e-05,
      "loss": 0.779,
      "num_input_tokens_seen": 2628943872,
      "step": 35430,
      "train_runtime": 42197.1274,
      "train_tokens_per_second": 62301.489
    },
    {
      "epoch": 0.470960325845011,
      "grad_norm": 27.25,
      "learning_rate": 1.1422141429374129e-05,
      "loss": 0.643,
      "num_input_tokens_seen": 2629165824,
      "step": 35440,
      "train_runtime": 42200.5544,
      "train_tokens_per_second": 62301.689
    },
    {
      "epoch": 0.4710932153274729,
      "grad_norm": 424.0,
      "learning_rate": 1.1417881079358193e-05,
      "loss": 0.498,
      "num_input_tokens_seen": 2629279232,
      "step": 35450,
      "train_runtime": 42203.5188,
      "train_tokens_per_second": 62300.0
    },
    {
      "epoch": 0.4712261048099348,
      "grad_norm": 22.0,
      "learning_rate": 1.1413620466692431e-05,
      "loss": 0.801,
      "num_input_tokens_seen": 2631323904,
      "step": 35460,
      "train_runtime": 42236.6571,
      "train_tokens_per_second": 62299.53
    },
    {
      "epoch": 0.47135899429239675,
      "grad_norm": 18.75,
      "learning_rate": 1.140935959216608e-05,
      "loss": 0.8997,
      "num_input_tokens_seen": 2632185344,
      "step": 35470,
      "train_runtime": 42249.6478,
      "train_tokens_per_second": 62300.764
    },
    {
      "epoch": 0.4714918837748586,
      "grad_norm": 31.125,
      "learning_rate": 1.1405098456568433e-05,
      "loss": 0.7838,
      "num_input_tokens_seen": 2632690944,
      "step": 35480,
      "train_runtime": 42256.8791,
      "train_tokens_per_second": 62302.068
    },
    {
      "epoch": 0.47162477325732055,
      "grad_norm": 20.5,
      "learning_rate": 1.1400837060688827e-05,
      "loss": 0.6283,
      "num_input_tokens_seen": 2632940032,
      "step": 35490,
      "train_runtime": 42260.5474,
      "train_tokens_per_second": 62302.554
    },
    {
      "epoch": 0.4717576627397825,
      "grad_norm": 40.25,
      "learning_rate": 1.1396575405316646e-05,
      "loss": 0.6234,
      "num_input_tokens_seen": 2633057280,
      "step": 35500,
      "train_runtime": 42262.951,
      "train_tokens_per_second": 62301.785
    },
    {
      "epoch": 0.47189055222224435,
      "grad_norm": 30.875,
      "learning_rate": 1.1392313491241325e-05,
      "loss": 0.8729,
      "num_input_tokens_seen": 2635044864,
      "step": 35510,
      "train_runtime": 42295.0399,
      "train_tokens_per_second": 62301.51
    },
    {
      "epoch": 0.4720234417047063,
      "grad_norm": 21.625,
      "learning_rate": 1.1388051319252345e-05,
      "loss": 0.8063,
      "num_input_tokens_seen": 2635898368,
      "step": 35520,
      "train_runtime": 42308.0245,
      "train_tokens_per_second": 62302.563
    },
    {
      "epoch": 0.4721563311871682,
      "grad_norm": 19.25,
      "learning_rate": 1.1383788890139231e-05,
      "loss": 0.7561,
      "num_input_tokens_seen": 2636360704,
      "step": 35530,
      "train_runtime": 42314.4254,
      "train_tokens_per_second": 62304.065
    },
    {
      "epoch": 0.4722892206696301,
      "grad_norm": 18.0,
      "learning_rate": 1.1379526204691567e-05,
      "loss": 0.6356,
      "num_input_tokens_seen": 2636585472,
      "step": 35540,
      "train_runtime": 42317.8017,
      "train_tokens_per_second": 62304.405
    },
    {
      "epoch": 0.472422110152092,
      "grad_norm": 21.5,
      "learning_rate": 1.1375263263698973e-05,
      "loss": 0.5698,
      "num_input_tokens_seen": 2636697344,
      "step": 35550,
      "train_runtime": 42320.1373,
      "train_tokens_per_second": 62303.61
    },
    {
      "epoch": 0.47255499963455394,
      "grad_norm": 21.625,
      "learning_rate": 1.1371000067951126e-05,
      "loss": 0.7926,
      "num_input_tokens_seen": 2638754048,
      "step": 35560,
      "train_runtime": 42352.9569,
      "train_tokens_per_second": 62303.892
    },
    {
      "epoch": 0.4726878891170158,
      "grad_norm": 13.5,
      "learning_rate": 1.136673661823774e-05,
      "loss": 0.9138,
      "num_input_tokens_seen": 2639614208,
      "step": 35570,
      "train_runtime": 42366.4593,
      "train_tokens_per_second": 62304.338
    },
    {
      "epoch": 0.47282077859947774,
      "grad_norm": 32.5,
      "learning_rate": 1.136247291534858e-05,
      "loss": 0.7849,
      "num_input_tokens_seen": 2640105984,
      "step": 35580,
      "train_runtime": 42373.3217,
      "train_tokens_per_second": 62305.854
    },
    {
      "epoch": 0.47295366808193967,
      "grad_norm": 63.5,
      "learning_rate": 1.1358208960073464e-05,
      "loss": 0.5846,
      "num_input_tokens_seen": 2640342016,
      "step": 35590,
      "train_runtime": 42376.8021,
      "train_tokens_per_second": 62306.306
    },
    {
      "epoch": 0.47308655756440154,
      "grad_norm": 15.1875,
      "learning_rate": 1.1353944753202248e-05,
      "loss": 0.5156,
      "num_input_tokens_seen": 2640455936,
      "step": 35600,
      "train_runtime": 42379.1462,
      "train_tokens_per_second": 62305.548
    },
    {
      "epoch": 0.4732194470468635,
      "grad_norm": 43.5,
      "learning_rate": 1.1349680295524841e-05,
      "loss": 0.8696,
      "num_input_tokens_seen": 2642451712,
      "step": 35610,
      "train_runtime": 42411.7868,
      "train_tokens_per_second": 62304.654
    },
    {
      "epoch": 0.4733523365293254,
      "grad_norm": 14.625,
      "learning_rate": 1.1345415587831191e-05,
      "loss": 0.8445,
      "num_input_tokens_seen": 2643296512,
      "step": 35620,
      "train_runtime": 42424.7635,
      "train_tokens_per_second": 62305.51
    },
    {
      "epoch": 0.4734852260117873,
      "grad_norm": 19.125,
      "learning_rate": 1.1341150630911302e-05,
      "loss": 0.7846,
      "num_input_tokens_seen": 2643764992,
      "step": 35630,
      "train_runtime": 42431.53,
      "train_tokens_per_second": 62306.615
    },
    {
      "epoch": 0.4736181154942492,
      "grad_norm": 1976.0,
      "learning_rate": 1.1336885425555215e-05,
      "loss": 0.5924,
      "num_input_tokens_seen": 2643990784,
      "step": 35640,
      "train_runtime": 42434.9143,
      "train_tokens_per_second": 62306.967
    },
    {
      "epoch": 0.47375100497671113,
      "grad_norm": 45.25,
      "learning_rate": 1.1332619972553028e-05,
      "loss": 0.5476,
      "num_input_tokens_seen": 2644102144,
      "step": 35650,
      "train_runtime": 42437.2456,
      "train_tokens_per_second": 62306.168
    },
    {
      "epoch": 0.473883894459173,
      "grad_norm": 16.25,
      "learning_rate": 1.1328354272694872e-05,
      "loss": 0.8391,
      "num_input_tokens_seen": 2646232320,
      "step": 35660,
      "train_runtime": 42470.7956,
      "train_tokens_per_second": 62307.105
    },
    {
      "epoch": 0.47401678394163493,
      "grad_norm": 10.5,
      "learning_rate": 1.1324088326770933e-05,
      "loss": 0.8463,
      "num_input_tokens_seen": 2647054080,
      "step": 35670,
      "train_runtime": 42483.2777,
      "train_tokens_per_second": 62308.142
    },
    {
      "epoch": 0.47414967342409686,
      "grad_norm": 38.5,
      "learning_rate": 1.1319822135571436e-05,
      "loss": 0.8171,
      "num_input_tokens_seen": 2647536384,
      "step": 35680,
      "train_runtime": 42490.3141,
      "train_tokens_per_second": 62309.174
    },
    {
      "epoch": 0.47428256290655874,
      "grad_norm": 15.1875,
      "learning_rate": 1.1315555699886657e-05,
      "loss": 0.6947,
      "num_input_tokens_seen": 2647760128,
      "step": 35690,
      "train_runtime": 42493.6882,
      "train_tokens_per_second": 62309.492
    },
    {
      "epoch": 0.47441545238902066,
      "grad_norm": 24.625,
      "learning_rate": 1.1311289020506914e-05,
      "loss": 0.5838,
      "num_input_tokens_seen": 2647875584,
      "step": 35700,
      "train_runtime": 42496.0523,
      "train_tokens_per_second": 62308.743
    },
    {
      "epoch": 0.4745483418714826,
      "grad_norm": 31.625,
      "learning_rate": 1.1307022098222575e-05,
      "loss": 0.8344,
      "num_input_tokens_seen": 2649952256,
      "step": 35710,
      "train_runtime": 42529.3566,
      "train_tokens_per_second": 62308.778
    },
    {
      "epoch": 0.47468123135394447,
      "grad_norm": 34.25,
      "learning_rate": 1.1302754933824046e-05,
      "loss": 0.9324,
      "num_input_tokens_seen": 2650838528,
      "step": 35720,
      "train_runtime": 42542.9167,
      "train_tokens_per_second": 62309.751
    },
    {
      "epoch": 0.4748141208364064,
      "grad_norm": 12.125,
      "learning_rate": 1.1298487528101783e-05,
      "loss": 0.6819,
      "num_input_tokens_seen": 2651320320,
      "step": 35730,
      "train_runtime": 42549.4994,
      "train_tokens_per_second": 62311.434
    },
    {
      "epoch": 0.4749470103188683,
      "grad_norm": 20.875,
      "learning_rate": 1.1294219881846284e-05,
      "loss": 0.6453,
      "num_input_tokens_seen": 2651543808,
      "step": 35740,
      "train_runtime": 42552.8282,
      "train_tokens_per_second": 62311.811
    },
    {
      "epoch": 0.4750798998013302,
      "grad_norm": 12.125,
      "learning_rate": 1.1289951995848089e-05,
      "loss": 0.5233,
      "num_input_tokens_seen": 2651656192,
      "step": 35750,
      "train_runtime": 42555.3058,
      "train_tokens_per_second": 62310.824
    },
    {
      "epoch": 0.4752127892837921,
      "grad_norm": 12.9375,
      "learning_rate": 1.1285683870897792e-05,
      "loss": 0.8181,
      "num_input_tokens_seen": 2653700352,
      "step": 35760,
      "train_runtime": 42588.9985,
      "train_tokens_per_second": 62309.527
    },
    {
      "epoch": 0.47534567876625405,
      "grad_norm": 19.375,
      "learning_rate": 1.1281415507786023e-05,
      "loss": 0.9403,
      "num_input_tokens_seen": 2654574848,
      "step": 35770,
      "train_runtime": 42601.9335,
      "train_tokens_per_second": 62311.135
    },
    {
      "epoch": 0.475478568248716,
      "grad_norm": 15.0625,
      "learning_rate": 1.1277146907303456e-05,
      "loss": 0.7186,
      "num_input_tokens_seen": 2655069184,
      "step": 35780,
      "train_runtime": 42608.7252,
      "train_tokens_per_second": 62312.805
    },
    {
      "epoch": 0.47561145773117786,
      "grad_norm": 19.25,
      "learning_rate": 1.1272878070240812e-05,
      "loss": 0.5934,
      "num_input_tokens_seen": 2655296768,
      "step": 35790,
      "train_runtime": 42612.1956,
      "train_tokens_per_second": 62313.071
    },
    {
      "epoch": 0.4757443472136398,
      "grad_norm": 12.4375,
      "learning_rate": 1.1268608997388855e-05,
      "loss": 0.4842,
      "num_input_tokens_seen": 2655414016,
      "step": 35800,
      "train_runtime": 42615.184,
      "train_tokens_per_second": 62311.452
    },
    {
      "epoch": 0.4758772366961017,
      "grad_norm": 67.0,
      "learning_rate": 1.1264339689538392e-05,
      "loss": 0.8581,
      "num_input_tokens_seen": 2657441792,
      "step": 35810,
      "train_runtime": 42648.6322,
      "train_tokens_per_second": 62310.129
    },
    {
      "epoch": 0.4760101261785636,
      "grad_norm": 20.875,
      "learning_rate": 1.126007014748028e-05,
      "loss": 0.9116,
      "num_input_tokens_seen": 2658306304,
      "step": 35820,
      "train_runtime": 42661.5604,
      "train_tokens_per_second": 62311.511
    },
    {
      "epoch": 0.4761430156610255,
      "grad_norm": 27.125,
      "learning_rate": 1.1255800372005408e-05,
      "loss": 0.7664,
      "num_input_tokens_seen": 2658776320,
      "step": 35830,
      "train_runtime": 42668.3674,
      "train_tokens_per_second": 62312.586
    },
    {
      "epoch": 0.47627590514348744,
      "grad_norm": 21.25,
      "learning_rate": 1.1251530363904713e-05,
      "loss": 0.6211,
      "num_input_tokens_seen": 2659010816,
      "step": 35840,
      "train_runtime": 42671.8848,
      "train_tokens_per_second": 62312.945
    },
    {
      "epoch": 0.4764087946259493,
      "grad_norm": 6.5,
      "learning_rate": 1.1247260123969178e-05,
      "loss": 0.4482,
      "num_input_tokens_seen": 2659125504,
      "step": 35850,
      "train_runtime": 42674.2547,
      "train_tokens_per_second": 62312.172
    },
    {
      "epoch": 0.47654168410841125,
      "grad_norm": 18.625,
      "learning_rate": 1.1242989652989824e-05,
      "loss": 0.8255,
      "num_input_tokens_seen": 2661164800,
      "step": 35860,
      "train_runtime": 42707.0379,
      "train_tokens_per_second": 62312.09
    },
    {
      "epoch": 0.4766745735908732,
      "grad_norm": 19.0,
      "learning_rate": 1.1238718951757726e-05,
      "loss": 0.8861,
      "num_input_tokens_seen": 2662011136,
      "step": 35870,
      "train_runtime": 42719.5392,
      "train_tokens_per_second": 62313.667
    },
    {
      "epoch": 0.47680746307333505,
      "grad_norm": 27.0,
      "learning_rate": 1.1234448021063983e-05,
      "loss": 0.7462,
      "num_input_tokens_seen": 2662462208,
      "step": 35880,
      "train_runtime": 42726.0543,
      "train_tokens_per_second": 62314.722
    },
    {
      "epoch": 0.476940352555797,
      "grad_norm": 12.625,
      "learning_rate": 1.1230176861699754e-05,
      "loss": 0.7325,
      "num_input_tokens_seen": 2662692352,
      "step": 35890,
      "train_runtime": 42729.4966,
      "train_tokens_per_second": 62315.088
    },
    {
      "epoch": 0.4770732420382589,
      "grad_norm": 11.9375,
      "learning_rate": 1.122590547445623e-05,
      "loss": 0.5467,
      "num_input_tokens_seen": 2662812160,
      "step": 35900,
      "train_runtime": 42731.8667,
      "train_tokens_per_second": 62314.436
    },
    {
      "epoch": 0.4772061315207208,
      "grad_norm": 34.25,
      "learning_rate": 1.1221633860124648e-05,
      "loss": 0.8467,
      "num_input_tokens_seen": 2664844544,
      "step": 35910,
      "train_runtime": 42765.0955,
      "train_tokens_per_second": 62313.541
    },
    {
      "epoch": 0.4773390210031827,
      "grad_norm": 33.25,
      "learning_rate": 1.1217362019496286e-05,
      "loss": 0.8596,
      "num_input_tokens_seen": 2665688832,
      "step": 35920,
      "train_runtime": 42778.017,
      "train_tokens_per_second": 62314.455
    },
    {
      "epoch": 0.47747191048564464,
      "grad_norm": 18.125,
      "learning_rate": 1.1213089953362466e-05,
      "loss": 0.7629,
      "num_input_tokens_seen": 2666171648,
      "step": 35930,
      "train_runtime": 42784.8619,
      "train_tokens_per_second": 62315.771
    },
    {
      "epoch": 0.4776047999681065,
      "grad_norm": 17.375,
      "learning_rate": 1.1208817662514546e-05,
      "loss": 0.6702,
      "num_input_tokens_seen": 2666402816,
      "step": 35940,
      "train_runtime": 42788.7854,
      "train_tokens_per_second": 62315.459
    },
    {
      "epoch": 0.47773768945056844,
      "grad_norm": 6.40625,
      "learning_rate": 1.1204545147743934e-05,
      "loss": 0.5319,
      "num_input_tokens_seen": 2666515456,
      "step": 35950,
      "train_runtime": 42791.1548,
      "train_tokens_per_second": 62314.641
    },
    {
      "epoch": 0.47787057893303037,
      "grad_norm": 21.0,
      "learning_rate": 1.1200272409842077e-05,
      "loss": 0.7612,
      "num_input_tokens_seen": 2668754944,
      "step": 35960,
      "train_runtime": 42825.0629,
      "train_tokens_per_second": 62317.595
    },
    {
      "epoch": 0.47800346841549224,
      "grad_norm": 28.25,
      "learning_rate": 1.1195999449600457e-05,
      "loss": 0.9284,
      "num_input_tokens_seen": 2669576448,
      "step": 35970,
      "train_runtime": 42837.8992,
      "train_tokens_per_second": 62318.099
    },
    {
      "epoch": 0.47813635789795417,
      "grad_norm": 23.75,
      "learning_rate": 1.1191726267810603e-05,
      "loss": 0.7428,
      "num_input_tokens_seen": 2670127360,
      "step": 35980,
      "train_runtime": 42845.0629,
      "train_tokens_per_second": 62320.538
    },
    {
      "epoch": 0.4782692473804161,
      "grad_norm": 40.25,
      "learning_rate": 1.1187452865264086e-05,
      "loss": 0.6611,
      "num_input_tokens_seen": 2670378496,
      "step": 35990,
      "train_runtime": 42848.6983,
      "train_tokens_per_second": 62321.111
    },
    {
      "epoch": 0.47840213686287797,
      "grad_norm": 12.9375,
      "learning_rate": 1.1183179242752513e-05,
      "loss": 0.538,
      "num_input_tokens_seen": 2670495744,
      "step": 36000,
      "train_runtime": 42851.663,
      "train_tokens_per_second": 62319.536
    },
    {
      "epoch": 0.4785350263453399,
      "grad_norm": 33.5,
      "learning_rate": 1.1178905401067535e-05,
      "loss": 0.8066,
      "num_input_tokens_seen": 2672663296,
      "step": 36010,
      "train_runtime": 42886.6142,
      "train_tokens_per_second": 62319.289
    },
    {
      "epoch": 0.4786679158278018,
      "grad_norm": 15.75,
      "learning_rate": 1.1174631341000843e-05,
      "loss": 0.9075,
      "num_input_tokens_seen": 2673505792,
      "step": 36020,
      "train_runtime": 42899.2653,
      "train_tokens_per_second": 62320.55
    },
    {
      "epoch": 0.4788008053102637,
      "grad_norm": 19.875,
      "learning_rate": 1.1170357063344171e-05,
      "loss": 0.7329,
      "num_input_tokens_seen": 2673953280,
      "step": 36030,
      "train_runtime": 42905.7613,
      "train_tokens_per_second": 62321.544
    },
    {
      "epoch": 0.47893369479272563,
      "grad_norm": 17.5,
      "learning_rate": 1.1166082568889286e-05,
      "loss": 0.6522,
      "num_input_tokens_seen": 2674171648,
      "step": 36040,
      "train_runtime": 42909.1215,
      "train_tokens_per_second": 62321.753
    },
    {
      "epoch": 0.47906658427518756,
      "grad_norm": 10.0625,
      "learning_rate": 1.1161807858428007e-05,
      "loss": 0.4818,
      "num_input_tokens_seen": 2674284544,
      "step": 36050,
      "train_runtime": 42912.1208,
      "train_tokens_per_second": 62320.027
    },
    {
      "epoch": 0.47919947375764943,
      "grad_norm": 33.5,
      "learning_rate": 1.115753293275218e-05,
      "loss": 0.8302,
      "num_input_tokens_seen": 2676404480,
      "step": 36060,
      "train_runtime": 42945.1715,
      "train_tokens_per_second": 62321.429
    },
    {
      "epoch": 0.47933236324011136,
      "grad_norm": 29.5,
      "learning_rate": 1.1153257792653697e-05,
      "loss": 0.9713,
      "num_input_tokens_seen": 2677266688,
      "step": 36070,
      "train_runtime": 42958.015,
      "train_tokens_per_second": 62322.868
    },
    {
      "epoch": 0.4794652527225733,
      "grad_norm": 27.375,
      "learning_rate": 1.1148982438924491e-05,
      "loss": 0.7557,
      "num_input_tokens_seen": 2677718784,
      "step": 36080,
      "train_runtime": 42964.6209,
      "train_tokens_per_second": 62323.808
    },
    {
      "epoch": 0.47959814220503516,
      "grad_norm": 18.875,
      "learning_rate": 1.1144706872356539e-05,
      "loss": 0.6963,
      "num_input_tokens_seen": 2677949184,
      "step": 36090,
      "train_runtime": 42968.6419,
      "train_tokens_per_second": 62323.338
    },
    {
      "epoch": 0.4797310316874971,
      "grad_norm": 26.375,
      "learning_rate": 1.1140431093741842e-05,
      "loss": 0.5717,
      "num_input_tokens_seen": 2678064384,
      "step": 36100,
      "train_runtime": 42971.0256,
      "train_tokens_per_second": 62322.561
    },
    {
      "epoch": 0.479863921169959,
      "grad_norm": 17.875,
      "learning_rate": 1.1136155103872452e-05,
      "loss": 0.785,
      "num_input_tokens_seen": 2680061696,
      "step": 36110,
      "train_runtime": 43002.8838,
      "train_tokens_per_second": 62322.837
    },
    {
      "epoch": 0.4799968106524209,
      "grad_norm": 39.0,
      "learning_rate": 1.113187890354046e-05,
      "loss": 0.9024,
      "num_input_tokens_seen": 2680858880,
      "step": 36120,
      "train_runtime": 43014.9225,
      "train_tokens_per_second": 62323.927
    },
    {
      "epoch": 0.4801297001348828,
      "grad_norm": 22.125,
      "learning_rate": 1.1127602493538e-05,
      "loss": 0.8015,
      "num_input_tokens_seen": 2681313024,
      "step": 36130,
      "train_runtime": 43021.4355,
      "train_tokens_per_second": 62325.048
    },
    {
      "epoch": 0.48026258961734475,
      "grad_norm": 8.75,
      "learning_rate": 1.1123325874657228e-05,
      "loss": 0.626,
      "num_input_tokens_seen": 2681551104,
      "step": 36140,
      "train_runtime": 43024.9429,
      "train_tokens_per_second": 62325.501
    },
    {
      "epoch": 0.4803954790998066,
      "grad_norm": 10.0625,
      "learning_rate": 1.1119049047690356e-05,
      "loss": 0.5172,
      "num_input_tokens_seen": 2681665280,
      "step": 36150,
      "train_runtime": 43027.3037,
      "train_tokens_per_second": 62324.735
    },
    {
      "epoch": 0.48052836858226855,
      "grad_norm": 20.25,
      "learning_rate": 1.1114772013429625e-05,
      "loss": 0.7556,
      "num_input_tokens_seen": 2683639040,
      "step": 36160,
      "train_runtime": 43060.5214,
      "train_tokens_per_second": 62322.493
    },
    {
      "epoch": 0.4806612580647305,
      "grad_norm": 25.625,
      "learning_rate": 1.1110494772667317e-05,
      "loss": 0.9075,
      "num_input_tokens_seen": 2684495360,
      "step": 36170,
      "train_runtime": 43073.5071,
      "train_tokens_per_second": 62323.585
    },
    {
      "epoch": 0.48079414754719235,
      "grad_norm": 13.8125,
      "learning_rate": 1.1106217326195756e-05,
      "loss": 0.7388,
      "num_input_tokens_seen": 2685011968,
      "step": 36180,
      "train_runtime": 43080.4759,
      "train_tokens_per_second": 62325.495
    },
    {
      "epoch": 0.4809270370296543,
      "grad_norm": 38.75,
      "learning_rate": 1.1101939674807301e-05,
      "loss": 0.6386,
      "num_input_tokens_seen": 2685240832,
      "step": 36190,
      "train_runtime": 43083.9485,
      "train_tokens_per_second": 62325.783
    },
    {
      "epoch": 0.4810599265121162,
      "grad_norm": 16.25,
      "learning_rate": 1.1097661819294345e-05,
      "loss": 0.5742,
      "num_input_tokens_seen": 2685351424,
      "step": 36200,
      "train_runtime": 43086.2665,
      "train_tokens_per_second": 62324.997
    },
    {
      "epoch": 0.4811928159945781,
      "grad_norm": 30.5,
      "learning_rate": 1.1093383760449325e-05,
      "loss": 0.7883,
      "num_input_tokens_seen": 2687477504,
      "step": 36210,
      "train_runtime": 43119.4732,
      "train_tokens_per_second": 62326.307
    },
    {
      "epoch": 0.48132570547704,
      "grad_norm": 16.5,
      "learning_rate": 1.1089105499064713e-05,
      "loss": 0.9385,
      "num_input_tokens_seen": 2688319744,
      "step": 36220,
      "train_runtime": 43132.811,
      "train_tokens_per_second": 62326.56
    },
    {
      "epoch": 0.48145859495950194,
      "grad_norm": 11.4375,
      "learning_rate": 1.1084827035933021e-05,
      "loss": 0.7548,
      "num_input_tokens_seen": 2688803584,
      "step": 36230,
      "train_runtime": 43139.7266,
      "train_tokens_per_second": 62327.785
    },
    {
      "epoch": 0.48159148444196387,
      "grad_norm": 18.75,
      "learning_rate": 1.1080548371846791e-05,
      "loss": 0.635,
      "num_input_tokens_seen": 2689037056,
      "step": 36240,
      "train_runtime": 43143.2291,
      "train_tokens_per_second": 62328.136
    },
    {
      "epoch": 0.48172437392442574,
      "grad_norm": 9.8125,
      "learning_rate": 1.1076269507598611e-05,
      "loss": 0.5644,
      "num_input_tokens_seen": 2689154816,
      "step": 36250,
      "train_runtime": 43145.5954,
      "train_tokens_per_second": 62327.447
    },
    {
      "epoch": 0.48185726340688767,
      "grad_norm": 26.125,
      "learning_rate": 1.1071990443981102e-05,
      "loss": 0.8373,
      "num_input_tokens_seen": 2691389952,
      "step": 36260,
      "train_runtime": 43179.8686,
      "train_tokens_per_second": 62329.739
    },
    {
      "epoch": 0.4819901528893496,
      "grad_norm": 12.5,
      "learning_rate": 1.106771118178692e-05,
      "loss": 0.9064,
      "num_input_tokens_seen": 2692248320,
      "step": 36270,
      "train_runtime": 43192.6028,
      "train_tokens_per_second": 62331.236
    },
    {
      "epoch": 0.48212304237181147,
      "grad_norm": 16.375,
      "learning_rate": 1.1063431721808763e-05,
      "loss": 0.7017,
      "num_input_tokens_seen": 2692710400,
      "step": 36280,
      "train_runtime": 43199.1716,
      "train_tokens_per_second": 62332.455
    },
    {
      "epoch": 0.4822559318542734,
      "grad_norm": 10.875,
      "learning_rate": 1.1059152064839365e-05,
      "loss": 0.6643,
      "num_input_tokens_seen": 2692938496,
      "step": 36290,
      "train_runtime": 43202.5929,
      "train_tokens_per_second": 62332.798
    },
    {
      "epoch": 0.48238882133673533,
      "grad_norm": 13.625,
      "learning_rate": 1.105487221167149e-05,
      "loss": 0.5075,
      "num_input_tokens_seen": 2693053696,
      "step": 36300,
      "train_runtime": 43204.9367,
      "train_tokens_per_second": 62332.083
    },
    {
      "epoch": 0.4825217108191972,
      "grad_norm": 15.0625,
      "learning_rate": 1.1050592163097944e-05,
      "loss": 0.7989,
      "num_input_tokens_seen": 2695062272,
      "step": 36310,
      "train_runtime": 43237.8839,
      "train_tokens_per_second": 62331.04
    },
    {
      "epoch": 0.48265460030165913,
      "grad_norm": 40.0,
      "learning_rate": 1.1046311919911565e-05,
      "loss": 0.9284,
      "num_input_tokens_seen": 2695881728,
      "step": 36320,
      "train_runtime": 43250.107,
      "train_tokens_per_second": 62332.371
    },
    {
      "epoch": 0.48278748978412106,
      "grad_norm": 152.0,
      "learning_rate": 1.1042031482905233e-05,
      "loss": 0.7505,
      "num_input_tokens_seen": 2696327168,
      "step": 36330,
      "train_runtime": 43256.4979,
      "train_tokens_per_second": 62333.46
    },
    {
      "epoch": 0.48292037926658293,
      "grad_norm": 13.75,
      "learning_rate": 1.1037750852871862e-05,
      "loss": 0.6595,
      "num_input_tokens_seen": 2696550656,
      "step": 36340,
      "train_runtime": 43260.4592,
      "train_tokens_per_second": 62332.918
    },
    {
      "epoch": 0.48305326874904486,
      "grad_norm": 27.125,
      "learning_rate": 1.1033470030604401e-05,
      "loss": 0.4616,
      "num_input_tokens_seen": 2696665088,
      "step": 36350,
      "train_runtime": 43262.8035,
      "train_tokens_per_second": 62332.185
    },
    {
      "epoch": 0.4831861582315068,
      "grad_norm": 19.625,
      "learning_rate": 1.102918901689583e-05,
      "loss": 0.7836,
      "num_input_tokens_seen": 2698697216,
      "step": 36360,
      "train_runtime": 43295.5758,
      "train_tokens_per_second": 62331.94
    },
    {
      "epoch": 0.48331904771396866,
      "grad_norm": 20.875,
      "learning_rate": 1.1024907812539168e-05,
      "loss": 0.8917,
      "num_input_tokens_seen": 2699528704,
      "step": 36370,
      "train_runtime": 43308.1179,
      "train_tokens_per_second": 62333.088
    },
    {
      "epoch": 0.4834519371964306,
      "grad_norm": 44.75,
      "learning_rate": 1.1020626418327476e-05,
      "loss": 0.7721,
      "num_input_tokens_seen": 2700019712,
      "step": 36380,
      "train_runtime": 43315.3499,
      "train_tokens_per_second": 62334.016
    },
    {
      "epoch": 0.4835848266788925,
      "grad_norm": 11.6875,
      "learning_rate": 1.101634483505384e-05,
      "loss": 0.6371,
      "num_input_tokens_seen": 2700244736,
      "step": 36390,
      "train_runtime": 43318.7488,
      "train_tokens_per_second": 62334.32
    },
    {
      "epoch": 0.4837177161613544,
      "grad_norm": 8.625,
      "learning_rate": 1.1012063063511387e-05,
      "loss": 0.4895,
      "num_input_tokens_seen": 2700357120,
      "step": 36400,
      "train_runtime": 43321.0955,
      "train_tokens_per_second": 62333.537
    },
    {
      "epoch": 0.4838506056438163,
      "grad_norm": 14.0625,
      "learning_rate": 1.1007781104493276e-05,
      "loss": 0.8363,
      "num_input_tokens_seen": 2702323712,
      "step": 36410,
      "train_runtime": 43353.2761,
      "train_tokens_per_second": 62332.63
    },
    {
      "epoch": 0.48398349512627825,
      "grad_norm": 18.0,
      "learning_rate": 1.1003498958792697e-05,
      "loss": 0.9229,
      "num_input_tokens_seen": 2703167744,
      "step": 36420,
      "train_runtime": 43365.7835,
      "train_tokens_per_second": 62334.115
    },
    {
      "epoch": 0.4841163846087401,
      "grad_norm": 17.25,
      "learning_rate": 1.0999216627202888e-05,
      "loss": 0.7209,
      "num_input_tokens_seen": 2703639552,
      "step": 36430,
      "train_runtime": 43372.4709,
      "train_tokens_per_second": 62335.382
    },
    {
      "epoch": 0.48424927409120205,
      "grad_norm": 11.6875,
      "learning_rate": 1.0994934110517108e-05,
      "loss": 0.6773,
      "num_input_tokens_seen": 2703873536,
      "step": 36440,
      "train_runtime": 43376.0113,
      "train_tokens_per_second": 62335.689
    },
    {
      "epoch": 0.484382163573664,
      "grad_norm": 14.0,
      "learning_rate": 1.0990651409528657e-05,
      "loss": 0.5518,
      "num_input_tokens_seen": 2703991552,
      "step": 36450,
      "train_runtime": 43378.3768,
      "train_tokens_per_second": 62335.01
    },
    {
      "epoch": 0.48451505305612586,
      "grad_norm": 13.9375,
      "learning_rate": 1.0986368525030866e-05,
      "loss": 0.8549,
      "num_input_tokens_seen": 2705967360,
      "step": 36460,
      "train_runtime": 43411.418,
      "train_tokens_per_second": 62333.079
    },
    {
      "epoch": 0.4846479425385878,
      "grad_norm": 23.75,
      "learning_rate": 1.0982085457817098e-05,
      "loss": 0.9264,
      "num_input_tokens_seen": 2706796288,
      "step": 36470,
      "train_runtime": 43424.0161,
      "train_tokens_per_second": 62334.085
    },
    {
      "epoch": 0.4847808320210497,
      "grad_norm": 25.625,
      "learning_rate": 1.0977802208680757e-05,
      "loss": 0.7306,
      "num_input_tokens_seen": 2707273984,
      "step": 36480,
      "train_runtime": 43430.7728,
      "train_tokens_per_second": 62335.386
    },
    {
      "epoch": 0.4849137215035116,
      "grad_norm": 20.75,
      "learning_rate": 1.0973518778415276e-05,
      "loss": 0.6104,
      "num_input_tokens_seen": 2707502848,
      "step": 36490,
      "train_runtime": 43434.221,
      "train_tokens_per_second": 62335.706
    },
    {
      "epoch": 0.4850466109859735,
      "grad_norm": 7.59375,
      "learning_rate": 1.096923516781412e-05,
      "loss": 0.5485,
      "num_input_tokens_seen": 2707619072,
      "step": 36500,
      "train_runtime": 43436.5656,
      "train_tokens_per_second": 62335.017
    },
    {
      "epoch": 0.48517950046843544,
      "grad_norm": 12.1875,
      "learning_rate": 1.0964951377670797e-05,
      "loss": 0.7934,
      "num_input_tokens_seen": 2709720832,
      "step": 36510,
      "train_runtime": 43470.9476,
      "train_tokens_per_second": 62334.064
    },
    {
      "epoch": 0.4853123899508973,
      "grad_norm": 36.25,
      "learning_rate": 1.0960667408778832e-05,
      "loss": 0.9129,
      "num_input_tokens_seen": 2710619904,
      "step": 36520,
      "train_runtime": 43483.8755,
      "train_tokens_per_second": 62336.208
    },
    {
      "epoch": 0.48544527943335924,
      "grad_norm": 18.125,
      "learning_rate": 1.0956383261931796e-05,
      "loss": 0.7121,
      "num_input_tokens_seen": 2711091712,
      "step": 36530,
      "train_runtime": 43490.6793,
      "train_tokens_per_second": 62337.304
    },
    {
      "epoch": 0.4855781689158212,
      "grad_norm": 23.0,
      "learning_rate": 1.095209893792329e-05,
      "loss": 0.5948,
      "num_input_tokens_seen": 2711331328,
      "step": 36540,
      "train_runtime": 43494.2758,
      "train_tokens_per_second": 62337.659
    },
    {
      "epoch": 0.48571105839828305,
      "grad_norm": 36.75,
      "learning_rate": 1.0947814437546945e-05,
      "loss": 0.484,
      "num_input_tokens_seen": 2711451904,
      "step": 36550,
      "train_runtime": 43496.6603,
      "train_tokens_per_second": 62337.014
    },
    {
      "epoch": 0.485843947880745,
      "grad_norm": 21.125,
      "learning_rate": 1.0943529761596426e-05,
      "loss": 0.8582,
      "num_input_tokens_seen": 2713447936,
      "step": 36560,
      "train_runtime": 43528.9605,
      "train_tokens_per_second": 62336.612
    },
    {
      "epoch": 0.4859768373632069,
      "grad_norm": 36.0,
      "learning_rate": 1.0939244910865432e-05,
      "loss": 0.9093,
      "num_input_tokens_seen": 2714267648,
      "step": 36570,
      "train_runtime": 43541.413,
      "train_tokens_per_second": 62337.611
    },
    {
      "epoch": 0.4861097268456688,
      "grad_norm": 22.75,
      "learning_rate": 1.0934959886147693e-05,
      "loss": 0.7563,
      "num_input_tokens_seen": 2714719744,
      "step": 36580,
      "train_runtime": 43547.9342,
      "train_tokens_per_second": 62338.657
    },
    {
      "epoch": 0.4862426163281307,
      "grad_norm": 79.5,
      "learning_rate": 1.0930674688236971e-05,
      "loss": 0.6786,
      "num_input_tokens_seen": 2714949888,
      "step": 36590,
      "train_runtime": 43551.8751,
      "train_tokens_per_second": 62338.301
    },
    {
      "epoch": 0.48637550581059263,
      "grad_norm": 18.5,
      "learning_rate": 1.0926389317927062e-05,
      "loss": 0.5432,
      "num_input_tokens_seen": 2715063552,
      "step": 36600,
      "train_runtime": 43554.2099,
      "train_tokens_per_second": 62337.569
    },
    {
      "epoch": 0.4865083952930545,
      "grad_norm": 17.625,
      "learning_rate": 1.0922103776011796e-05,
      "loss": 0.8128,
      "num_input_tokens_seen": 2717169664,
      "step": 36610,
      "train_runtime": 43589.1438,
      "train_tokens_per_second": 62335.926
    },
    {
      "epoch": 0.48664128477551644,
      "grad_norm": 42.75,
      "learning_rate": 1.0917818063285026e-05,
      "loss": 0.906,
      "num_input_tokens_seen": 2718038528,
      "step": 36620,
      "train_runtime": 43602.1721,
      "train_tokens_per_second": 62337.228
    },
    {
      "epoch": 0.48677417425797836,
      "grad_norm": 15.0625,
      "learning_rate": 1.0913532180540642e-05,
      "loss": 0.7771,
      "num_input_tokens_seen": 2718531072,
      "step": 36630,
      "train_runtime": 43609.1748,
      "train_tokens_per_second": 62338.512
    },
    {
      "epoch": 0.48690706374044024,
      "grad_norm": 12.0,
      "learning_rate": 1.0909246128572569e-05,
      "loss": 0.6299,
      "num_input_tokens_seen": 2718763520,
      "step": 36640,
      "train_runtime": 43612.6715,
      "train_tokens_per_second": 62338.844
    },
    {
      "epoch": 0.48703995322290217,
      "grad_norm": 30.625,
      "learning_rate": 1.0904959908174758e-05,
      "loss": 0.5016,
      "num_input_tokens_seen": 2718881024,
      "step": 36650,
      "train_runtime": 43615.5631,
      "train_tokens_per_second": 62337.405
    },
    {
      "epoch": 0.4871728427053641,
      "grad_norm": 18.5,
      "learning_rate": 1.0900673520141195e-05,
      "loss": 0.7787,
      "num_input_tokens_seen": 2720941056,
      "step": 36660,
      "train_runtime": 43649.3091,
      "train_tokens_per_second": 62336.406
    },
    {
      "epoch": 0.48730573218782597,
      "grad_norm": 16.625,
      "learning_rate": 1.089638696526589e-05,
      "loss": 0.87,
      "num_input_tokens_seen": 2721744896,
      "step": 36670,
      "train_runtime": 43661.7791,
      "train_tokens_per_second": 62337.013
    },
    {
      "epoch": 0.4874386216702879,
      "grad_norm": 15.625,
      "learning_rate": 1.0892100244342893e-05,
      "loss": 0.74,
      "num_input_tokens_seen": 2722192384,
      "step": 36680,
      "train_runtime": 43668.3423,
      "train_tokens_per_second": 62337.891
    },
    {
      "epoch": 0.4875715111527498,
      "grad_norm": 16.125,
      "learning_rate": 1.088781335816628e-05,
      "loss": 0.7119,
      "num_input_tokens_seen": 2722427904,
      "step": 36690,
      "train_runtime": 43671.785,
      "train_tokens_per_second": 62338.37
    },
    {
      "epoch": 0.4877044006352117,
      "grad_norm": 10.875,
      "learning_rate": 1.0883526307530163e-05,
      "loss": 0.5504,
      "num_input_tokens_seen": 2722541568,
      "step": 36700,
      "train_runtime": 43674.6753,
      "train_tokens_per_second": 62336.847
    },
    {
      "epoch": 0.4878372901176736,
      "grad_norm": 21.625,
      "learning_rate": 1.0879239093228674e-05,
      "loss": 0.9052,
      "num_input_tokens_seen": 2724602624,
      "step": 36710,
      "train_runtime": 43708.5639,
      "train_tokens_per_second": 62335.67
    },
    {
      "epoch": 0.48797017960013556,
      "grad_norm": 33.5,
      "learning_rate": 1.0874951716055984e-05,
      "loss": 0.8822,
      "num_input_tokens_seen": 2725456640,
      "step": 36720,
      "train_runtime": 43721.5621,
      "train_tokens_per_second": 62336.671
    },
    {
      "epoch": 0.4881030690825975,
      "grad_norm": 38.75,
      "learning_rate": 1.0870664176806287e-05,
      "loss": 0.753,
      "num_input_tokens_seen": 2725943808,
      "step": 36730,
      "train_runtime": 43728.3316,
      "train_tokens_per_second": 62338.162
    },
    {
      "epoch": 0.48823595856505936,
      "grad_norm": 17.625,
      "learning_rate": 1.0866376476273817e-05,
      "loss": 0.5718,
      "num_input_tokens_seen": 2726173696,
      "step": 36740,
      "train_runtime": 43731.7678,
      "train_tokens_per_second": 62338.52
    },
    {
      "epoch": 0.4883688480475213,
      "grad_norm": 91.0,
      "learning_rate": 1.0862088615252828e-05,
      "loss": 0.5379,
      "num_input_tokens_seen": 2726292224,
      "step": 36750,
      "train_runtime": 43734.1381,
      "train_tokens_per_second": 62337.852
    },
    {
      "epoch": 0.4885017375299832,
      "grad_norm": 82.5,
      "learning_rate": 1.085780059453761e-05,
      "loss": 0.9292,
      "num_input_tokens_seen": 2728387328,
      "step": 36760,
      "train_runtime": 43767.2043,
      "train_tokens_per_second": 62338.625
    },
    {
      "epoch": 0.4886346270124451,
      "grad_norm": 14.125,
      "learning_rate": 1.0853512414922487e-05,
      "loss": 0.867,
      "num_input_tokens_seen": 2729264640,
      "step": 36770,
      "train_runtime": 43780.4528,
      "train_tokens_per_second": 62339.799
    },
    {
      "epoch": 0.488767516494907,
      "grad_norm": 30.875,
      "learning_rate": 1.0849224077201797e-05,
      "loss": 0.7136,
      "num_input_tokens_seen": 2729752832,
      "step": 36780,
      "train_runtime": 43787.2105,
      "train_tokens_per_second": 62341.328
    },
    {
      "epoch": 0.48890040597736895,
      "grad_norm": 18.25,
      "learning_rate": 1.0844935582169915e-05,
      "loss": 0.6432,
      "num_input_tokens_seen": 2729986304,
      "step": 36790,
      "train_runtime": 43790.7202,
      "train_tokens_per_second": 62341.663
    },
    {
      "epoch": 0.4890332954598308,
      "grad_norm": 8.375,
      "learning_rate": 1.0840646930621252e-05,
      "loss": 0.5427,
      "num_input_tokens_seen": 2730102016,
      "step": 36800,
      "train_runtime": 43793.0537,
      "train_tokens_per_second": 62340.983
    },
    {
      "epoch": 0.48916618494229275,
      "grad_norm": 25.0,
      "learning_rate": 1.083635812335024e-05,
      "loss": 0.8517,
      "num_input_tokens_seen": 2732203776,
      "step": 36810,
      "train_runtime": 43826.2355,
      "train_tokens_per_second": 62341.74
    },
    {
      "epoch": 0.4892990744247547,
      "grad_norm": 12.0,
      "learning_rate": 1.0832069161151348e-05,
      "loss": 0.9074,
      "num_input_tokens_seen": 2733057536,
      "step": 36820,
      "train_runtime": 43838.7969,
      "train_tokens_per_second": 62343.352
    },
    {
      "epoch": 0.48943196390721655,
      "grad_norm": 21.375,
      "learning_rate": 1.0827780044819061e-05,
      "loss": 0.7798,
      "num_input_tokens_seen": 2733534976,
      "step": 36830,
      "train_runtime": 43845.7043,
      "train_tokens_per_second": 62344.419
    },
    {
      "epoch": 0.4895648533896785,
      "grad_norm": 30.0,
      "learning_rate": 1.08234907751479e-05,
      "loss": 0.7032,
      "num_input_tokens_seen": 2733773312,
      "step": 36840,
      "train_runtime": 43849.2397,
      "train_tokens_per_second": 62344.828
    },
    {
      "epoch": 0.4896977428721404,
      "grad_norm": 10.6875,
      "learning_rate": 1.0819201352932418e-05,
      "loss": 0.5144,
      "num_input_tokens_seen": 2733893632,
      "step": 36850,
      "train_runtime": 43851.6352,
      "train_tokens_per_second": 62344.166
    },
    {
      "epoch": 0.4898306323546023,
      "grad_norm": 34.5,
      "learning_rate": 1.081491177896719e-05,
      "loss": 0.7989,
      "num_input_tokens_seen": 2735949824,
      "step": 36860,
      "train_runtime": 43883.3299,
      "train_tokens_per_second": 62345.994
    },
    {
      "epoch": 0.4899635218370642,
      "grad_norm": 31.5,
      "learning_rate": 1.081062205404682e-05,
      "loss": 0.8858,
      "num_input_tokens_seen": 2736775424,
      "step": 36870,
      "train_runtime": 43895.8028,
      "train_tokens_per_second": 62347.087
    },
    {
      "epoch": 0.49009641131952614,
      "grad_norm": 12.375,
      "learning_rate": 1.0806332178965945e-05,
      "loss": 0.6875,
      "num_input_tokens_seen": 2737218304,
      "step": 36880,
      "train_runtime": 43902.9641,
      "train_tokens_per_second": 62347.005
    },
    {
      "epoch": 0.490229300801988,
      "grad_norm": 346.0,
      "learning_rate": 1.0802042154519224e-05,
      "loss": 0.5882,
      "num_input_tokens_seen": 2737442048,
      "step": 36890,
      "train_runtime": 43906.3693,
      "train_tokens_per_second": 62347.265
    },
    {
      "epoch": 0.49036219028444994,
      "grad_norm": 15.9375,
      "learning_rate": 1.0797751981501341e-05,
      "loss": 0.4593,
      "num_input_tokens_seen": 2737555712,
      "step": 36900,
      "train_runtime": 43908.6613,
      "train_tokens_per_second": 62346.599
    },
    {
      "epoch": 0.49049507976691187,
      "grad_norm": 31.125,
      "learning_rate": 1.0793461660707018e-05,
      "loss": 0.7855,
      "num_input_tokens_seen": 2739470080,
      "step": 36910,
      "train_runtime": 43939.7306,
      "train_tokens_per_second": 62346.083
    },
    {
      "epoch": 0.49062796924937374,
      "grad_norm": 23.875,
      "learning_rate": 1.0789171192931002e-05,
      "loss": 0.8236,
      "num_input_tokens_seen": 2740270848,
      "step": 36920,
      "train_runtime": 43951.8178,
      "train_tokens_per_second": 62347.156
    },
    {
      "epoch": 0.49076085873183567,
      "grad_norm": 38.5,
      "learning_rate": 1.0784880578968056e-05,
      "loss": 0.7522,
      "num_input_tokens_seen": 2740734720,
      "step": 36930,
      "train_runtime": 43958.7632,
      "train_tokens_per_second": 62347.858
    },
    {
      "epoch": 0.4908937482142976,
      "grad_norm": 16.5,
      "learning_rate": 1.0780589819612983e-05,
      "loss": 0.6892,
      "num_input_tokens_seen": 2740962560,
      "step": 36940,
      "train_runtime": 43962.2143,
      "train_tokens_per_second": 62348.146
    },
    {
      "epoch": 0.49102663769675947,
      "grad_norm": 85.5,
      "learning_rate": 1.0776298915660605e-05,
      "loss": 0.5914,
      "num_input_tokens_seen": 2741079296,
      "step": 36950,
      "train_runtime": 43964.53,
      "train_tokens_per_second": 62347.517
    },
    {
      "epoch": 0.4911595271792214,
      "grad_norm": 11.5625,
      "learning_rate": 1.0772007867905776e-05,
      "loss": 0.8554,
      "num_input_tokens_seen": 2742996992,
      "step": 36960,
      "train_runtime": 43995.7735,
      "train_tokens_per_second": 62346.83
    },
    {
      "epoch": 0.49129241666168333,
      "grad_norm": 13.75,
      "learning_rate": 1.0767716677143372e-05,
      "loss": 0.8457,
      "num_input_tokens_seen": 2743823104,
      "step": 36970,
      "train_runtime": 44008.5797,
      "train_tokens_per_second": 62347.459
    },
    {
      "epoch": 0.4914253061441452,
      "grad_norm": 31.375,
      "learning_rate": 1.0763425344168302e-05,
      "loss": 0.8095,
      "num_input_tokens_seen": 2744288000,
      "step": 36980,
      "train_runtime": 44015.0495,
      "train_tokens_per_second": 62348.856
    },
    {
      "epoch": 0.49155819562660713,
      "grad_norm": 14.5,
      "learning_rate": 1.075913386977549e-05,
      "loss": 0.625,
      "num_input_tokens_seen": 2744513280,
      "step": 36990,
      "train_runtime": 44018.4638,
      "train_tokens_per_second": 62349.138
    },
    {
      "epoch": 0.49169108510906906,
      "grad_norm": 12.5625,
      "learning_rate": 1.0754842254759898e-05,
      "loss": 0.5209,
      "num_input_tokens_seen": 2744627712,
      "step": 37000,
      "train_runtime": 44020.7918,
      "train_tokens_per_second": 62348.44
    },
    {
      "epoch": 0.49182397459153093,
      "grad_norm": 33.0,
      "learning_rate": 1.0750550499916513e-05,
      "loss": 0.8001,
      "num_input_tokens_seen": 2746691840,
      "step": 37010,
      "train_runtime": 44054.1134,
      "train_tokens_per_second": 62348.136
    },
    {
      "epoch": 0.49195686407399286,
      "grad_norm": 19.875,
      "learning_rate": 1.0746258606040338e-05,
      "loss": 0.8774,
      "num_input_tokens_seen": 2747587840,
      "step": 37020,
      "train_runtime": 44067.2018,
      "train_tokens_per_second": 62349.95
    },
    {
      "epoch": 0.4920897535564548,
      "grad_norm": 25.5,
      "learning_rate": 1.074196657392641e-05,
      "loss": 0.8229,
      "num_input_tokens_seen": 2748076288,
      "step": 37030,
      "train_runtime": 44074.3189,
      "train_tokens_per_second": 62350.964
    },
    {
      "epoch": 0.49222264303891666,
      "grad_norm": 30.125,
      "learning_rate": 1.0737674404369793e-05,
      "loss": 0.611,
      "num_input_tokens_seen": 2748314880,
      "step": 37040,
      "train_runtime": 44077.89,
      "train_tokens_per_second": 62351.326
    },
    {
      "epoch": 0.4923555325213786,
      "grad_norm": 15.75,
      "learning_rate": 1.0733382098165567e-05,
      "loss": 0.5593,
      "num_input_tokens_seen": 2748434176,
      "step": 37050,
      "train_runtime": 44080.8578,
      "train_tokens_per_second": 62349.834
    },
    {
      "epoch": 0.4924884220038405,
      "grad_norm": 21.25,
      "learning_rate": 1.0729089656108848e-05,
      "loss": 0.8944,
      "num_input_tokens_seen": 2750635264,
      "step": 37060,
      "train_runtime": 44114.5524,
      "train_tokens_per_second": 62352.106
    },
    {
      "epoch": 0.4926213114863024,
      "grad_norm": 60.75,
      "learning_rate": 1.0724797078994772e-05,
      "loss": 0.998,
      "num_input_tokens_seen": 2751504128,
      "step": 37070,
      "train_runtime": 44127.2901,
      "train_tokens_per_second": 62353.798
    },
    {
      "epoch": 0.4927542009687643,
      "grad_norm": 51.25,
      "learning_rate": 1.07205043676185e-05,
      "loss": 0.736,
      "num_input_tokens_seen": 2751955456,
      "step": 37080,
      "train_runtime": 44133.791,
      "train_tokens_per_second": 62354.84
    },
    {
      "epoch": 0.49288709045122625,
      "grad_norm": 18.875,
      "learning_rate": 1.0716211522775217e-05,
      "loss": 0.7096,
      "num_input_tokens_seen": 2752176896,
      "step": 37090,
      "train_runtime": 44137.2054,
      "train_tokens_per_second": 62355.033
    },
    {
      "epoch": 0.4930199799336881,
      "grad_norm": 8.25,
      "learning_rate": 1.0711918545260139e-05,
      "loss": 0.5111,
      "num_input_tokens_seen": 2752288512,
      "step": 37100,
      "train_runtime": 44140.1256,
      "train_tokens_per_second": 62353.436
    },
    {
      "epoch": 0.49315286941615005,
      "grad_norm": 14.375,
      "learning_rate": 1.0707625435868497e-05,
      "loss": 0.814,
      "num_input_tokens_seen": 2754327552,
      "step": 37110,
      "train_runtime": 44173.3638,
      "train_tokens_per_second": 62352.678
    },
    {
      "epoch": 0.493285758898612,
      "grad_norm": 35.25,
      "learning_rate": 1.070333219539555e-05,
      "loss": 0.9046,
      "num_input_tokens_seen": 2755142656,
      "step": 37120,
      "train_runtime": 44186.0036,
      "train_tokens_per_second": 62353.289
    },
    {
      "epoch": 0.49341864838107385,
      "grad_norm": 15.25,
      "learning_rate": 1.0699038824636587e-05,
      "loss": 0.7071,
      "num_input_tokens_seen": 2755615232,
      "step": 37130,
      "train_runtime": 44192.639,
      "train_tokens_per_second": 62354.62
    },
    {
      "epoch": 0.4935515378635358,
      "grad_norm": 10.4375,
      "learning_rate": 1.0694745324386918e-05,
      "loss": 0.639,
      "num_input_tokens_seen": 2755842816,
      "step": 37140,
      "train_runtime": 44196.6273,
      "train_tokens_per_second": 62354.143
    },
    {
      "epoch": 0.4936844273459977,
      "grad_norm": 12.3125,
      "learning_rate": 1.0690451695441868e-05,
      "loss": 0.5153,
      "num_input_tokens_seen": 2755959552,
      "step": 37150,
      "train_runtime": 44198.9919,
      "train_tokens_per_second": 62353.448
    },
    {
      "epoch": 0.4938173168284596,
      "grad_norm": 34.25,
      "learning_rate": 1.0686157938596799e-05,
      "loss": 0.8556,
      "num_input_tokens_seen": 2758000640,
      "step": 37160,
      "train_runtime": 44231.9456,
      "train_tokens_per_second": 62353.139
    },
    {
      "epoch": 0.4939502063109215,
      "grad_norm": 20.625,
      "learning_rate": 1.0681864054647091e-05,
      "loss": 0.9038,
      "num_input_tokens_seen": 2758851328,
      "step": 37170,
      "train_runtime": 44244.8847,
      "train_tokens_per_second": 62354.131
    },
    {
      "epoch": 0.49408309579338344,
      "grad_norm": 16.125,
      "learning_rate": 1.0677570044388149e-05,
      "loss": 0.7258,
      "num_input_tokens_seen": 2759319296,
      "step": 37180,
      "train_runtime": 44251.6107,
      "train_tokens_per_second": 62355.229
    },
    {
      "epoch": 0.49421598527584537,
      "grad_norm": 15.9375,
      "learning_rate": 1.0673275908615397e-05,
      "loss": 0.6409,
      "num_input_tokens_seen": 2759547648,
      "step": 37190,
      "train_runtime": 44255.0327,
      "train_tokens_per_second": 62355.567
    },
    {
      "epoch": 0.49434887475830724,
      "grad_norm": 12.0,
      "learning_rate": 1.0668981648124285e-05,
      "loss": 0.4865,
      "num_input_tokens_seen": 2759664896,
      "step": 37200,
      "train_runtime": 44257.4055,
      "train_tokens_per_second": 62354.873
    },
    {
      "epoch": 0.4944817642407692,
      "grad_norm": 25.0,
      "learning_rate": 1.0664687263710287e-05,
      "loss": 0.8381,
      "num_input_tokens_seen": 2761679872,
      "step": 37210,
      "train_runtime": 44290.705,
      "train_tokens_per_second": 62353.486
    },
    {
      "epoch": 0.4946146537232311,
      "grad_norm": 18.625,
      "learning_rate": 1.0660392756168903e-05,
      "loss": 0.9225,
      "num_input_tokens_seen": 2762500608,
      "step": 37220,
      "train_runtime": 44303.266,
      "train_tokens_per_second": 62354.333
    },
    {
      "epoch": 0.494747543205693,
      "grad_norm": 16.875,
      "learning_rate": 1.0656098126295647e-05,
      "loss": 0.7435,
      "num_input_tokens_seen": 2762976256,
      "step": 37230,
      "train_runtime": 44309.9895,
      "train_tokens_per_second": 62355.606
    },
    {
      "epoch": 0.4948804326881549,
      "grad_norm": 11.0,
      "learning_rate": 1.0651803374886068e-05,
      "loss": 0.6616,
      "num_input_tokens_seen": 2763211264,
      "step": 37240,
      "train_runtime": 44313.4297,
      "train_tokens_per_second": 62356.069
    },
    {
      "epoch": 0.49501332217061683,
      "grad_norm": 9.8125,
      "learning_rate": 1.0647508502735719e-05,
      "loss": 0.5474,
      "num_input_tokens_seen": 2763322112,
      "step": 37250,
      "train_runtime": 44316.4166,
      "train_tokens_per_second": 62354.367
    },
    {
      "epoch": 0.4951462116530787,
      "grad_norm": 96.5,
      "learning_rate": 1.06432135106402e-05,
      "loss": 0.7677,
      "num_input_tokens_seen": 2765384704,
      "step": 37260,
      "train_runtime": 44349.856,
      "train_tokens_per_second": 62353.86
    },
    {
      "epoch": 0.49527910113554063,
      "grad_norm": 47.5,
      "learning_rate": 1.063891839939511e-05,
      "loss": 0.8814,
      "num_input_tokens_seen": 2766209792,
      "step": 37270,
      "train_runtime": 44362.4091,
      "train_tokens_per_second": 62354.815
    },
    {
      "epoch": 0.49541199061800256,
      "grad_norm": 14.625,
      "learning_rate": 1.0634623169796083e-05,
      "loss": 0.7327,
      "num_input_tokens_seen": 2766668800,
      "step": 37280,
      "train_runtime": 44369.036,
      "train_tokens_per_second": 62355.847
    },
    {
      "epoch": 0.49554488010046444,
      "grad_norm": 19.375,
      "learning_rate": 1.0630327822638776e-05,
      "loss": 0.6608,
      "num_input_tokens_seen": 2766892288,
      "step": 37290,
      "train_runtime": 44372.4695,
      "train_tokens_per_second": 62356.058
    },
    {
      "epoch": 0.49567776958292636,
      "grad_norm": 30.5,
      "learning_rate": 1.0626032358718858e-05,
      "loss": 0.6259,
      "num_input_tokens_seen": 2767007744,
      "step": 37300,
      "train_runtime": 44374.7842,
      "train_tokens_per_second": 62355.407
    },
    {
      "epoch": 0.4958106590653883,
      "grad_norm": 31.625,
      "learning_rate": 1.0621736778832028e-05,
      "loss": 0.8068,
      "num_input_tokens_seen": 2768900096,
      "step": 37310,
      "train_runtime": 44405.9191,
      "train_tokens_per_second": 62354.302
    },
    {
      "epoch": 0.49594354854785017,
      "grad_norm": 34.75,
      "learning_rate": 1.0617441083774003e-05,
      "loss": 0.8445,
      "num_input_tokens_seen": 2769721600,
      "step": 37320,
      "train_runtime": 44418.7294,
      "train_tokens_per_second": 62354.814
    },
    {
      "epoch": 0.4960764380303121,
      "grad_norm": 27.5,
      "learning_rate": 1.0613145274340524e-05,
      "loss": 0.7373,
      "num_input_tokens_seen": 2770191360,
      "step": 37330,
      "train_runtime": 44425.2066,
      "train_tokens_per_second": 62356.297
    },
    {
      "epoch": 0.496209327512774,
      "grad_norm": 26.375,
      "learning_rate": 1.060884935132735e-05,
      "loss": 0.6587,
      "num_input_tokens_seen": 2770422784,
      "step": 37340,
      "train_runtime": 44428.615,
      "train_tokens_per_second": 62356.722
    },
    {
      "epoch": 0.4963422169952359,
      "grad_norm": 16.875,
      "learning_rate": 1.0604553315530266e-05,
      "loss": 0.553,
      "num_input_tokens_seen": 2770537728,
      "step": 37350,
      "train_runtime": 44430.9525,
      "train_tokens_per_second": 62356.028
    },
    {
      "epoch": 0.4964751064776978,
      "grad_norm": 15.4375,
      "learning_rate": 1.0600257167745065e-05,
      "loss": 0.7889,
      "num_input_tokens_seen": 2772580096,
      "step": 37360,
      "train_runtime": 44463.9395,
      "train_tokens_per_second": 62355.7
    },
    {
      "epoch": 0.49660799596015975,
      "grad_norm": 27.0,
      "learning_rate": 1.059596090876758e-05,
      "loss": 0.8706,
      "num_input_tokens_seen": 2773400064,
      "step": 37370,
      "train_runtime": 44476.8618,
      "train_tokens_per_second": 62356.02
    },
    {
      "epoch": 0.4967408854426216,
      "grad_norm": 16.25,
      "learning_rate": 1.0591664539393648e-05,
      "loss": 0.7152,
      "num_input_tokens_seen": 2773893376,
      "step": 37380,
      "train_runtime": 44483.6896,
      "train_tokens_per_second": 62357.538
    },
    {
      "epoch": 0.49687377492508356,
      "grad_norm": 16.5,
      "learning_rate": 1.0587368060419137e-05,
      "loss": 0.6421,
      "num_input_tokens_seen": 2774119168,
      "step": 37390,
      "train_runtime": 44487.1434,
      "train_tokens_per_second": 62357.773
    },
    {
      "epoch": 0.4970066644075455,
      "grad_norm": 29.25,
      "learning_rate": 1.0583071472639932e-05,
      "loss": 0.5637,
      "num_input_tokens_seen": 2774236672,
      "step": 37400,
      "train_runtime": 44489.5204,
      "train_tokens_per_second": 62357.082
    },
    {
      "epoch": 0.49713955389000736,
      "grad_norm": 20.875,
      "learning_rate": 1.0578774776851934e-05,
      "loss": 0.8525,
      "num_input_tokens_seen": 2776196096,
      "step": 37410,
      "train_runtime": 44521.9534,
      "train_tokens_per_second": 62355.667
    },
    {
      "epoch": 0.4972724433724693,
      "grad_norm": 35.25,
      "learning_rate": 1.0574477973851071e-05,
      "loss": 0.8873,
      "num_input_tokens_seen": 2777038848,
      "step": 37420,
      "train_runtime": 44534.3989,
      "train_tokens_per_second": 62357.165
    },
    {
      "epoch": 0.4974053328549312,
      "grad_norm": 18.75,
      "learning_rate": 1.0570181064433285e-05,
      "loss": 0.8005,
      "num_input_tokens_seen": 2777513984,
      "step": 37430,
      "train_runtime": 44541.2119,
      "train_tokens_per_second": 62358.294
    },
    {
      "epoch": 0.4975382223373931,
      "grad_norm": 11.1875,
      "learning_rate": 1.056588404939454e-05,
      "loss": 0.7282,
      "num_input_tokens_seen": 2777749248,
      "step": 37440,
      "train_runtime": 44544.7813,
      "train_tokens_per_second": 62358.579
    },
    {
      "epoch": 0.497671111819855,
      "grad_norm": 10.375,
      "learning_rate": 1.0561586929530823e-05,
      "loss": 0.5773,
      "num_input_tokens_seen": 2777867520,
      "step": 37450,
      "train_runtime": 44547.1643,
      "train_tokens_per_second": 62357.898
    },
    {
      "epoch": 0.49780400130231695,
      "grad_norm": 19.125,
      "learning_rate": 1.0557289705638132e-05,
      "loss": 0.853,
      "num_input_tokens_seen": 2779918336,
      "step": 37460,
      "train_runtime": 44579.4428,
      "train_tokens_per_second": 62358.75
    },
    {
      "epoch": 0.4979368907847788,
      "grad_norm": 32.5,
      "learning_rate": 1.0552992378512492e-05,
      "loss": 0.8717,
      "num_input_tokens_seen": 2780730368,
      "step": 37470,
      "train_runtime": 44591.8837,
      "train_tokens_per_second": 62359.563
    },
    {
      "epoch": 0.49806978026724075,
      "grad_norm": 11.75,
      "learning_rate": 1.0548694948949947e-05,
      "loss": 0.7234,
      "num_input_tokens_seen": 2781201664,
      "step": 37480,
      "train_runtime": 44598.7418,
      "train_tokens_per_second": 62360.541
    },
    {
      "epoch": 0.4982026697497027,
      "grad_norm": 15.6875,
      "learning_rate": 1.0544397417746555e-05,
      "loss": 0.6658,
      "num_input_tokens_seen": 2781433088,
      "step": 37490,
      "train_runtime": 44603.0018,
      "train_tokens_per_second": 62359.773
    },
    {
      "epoch": 0.49833555923216455,
      "grad_norm": 10.5,
      "learning_rate": 1.0540099785698397e-05,
      "loss": 0.5895,
      "num_input_tokens_seen": 2781553408,
      "step": 37500,
      "train_runtime": 44605.6821,
      "train_tokens_per_second": 62358.724
    },
    {
      "epoch": 0.4984684487146265,
      "grad_norm": 23.0,
      "learning_rate": 1.0535802053601568e-05,
      "loss": 0.8646,
      "num_input_tokens_seen": 2783489792,
      "step": 37510,
      "train_runtime": 44636.3023,
      "train_tokens_per_second": 62359.327
    },
    {
      "epoch": 0.4986013381970884,
      "grad_norm": 33.5,
      "learning_rate": 1.0531504222252187e-05,
      "loss": 0.9133,
      "num_input_tokens_seen": 2784285440,
      "step": 37520,
      "train_runtime": 44648.3915,
      "train_tokens_per_second": 62360.263
    },
    {
      "epoch": 0.4987342276795503,
      "grad_norm": 19.375,
      "learning_rate": 1.0527206292446387e-05,
      "loss": 0.7424,
      "num_input_tokens_seen": 2784729856,
      "step": 37530,
      "train_runtime": 44654.6904,
      "train_tokens_per_second": 62361.419
    },
    {
      "epoch": 0.4988671171620122,
      "grad_norm": 23.625,
      "learning_rate": 1.0522908264980321e-05,
      "loss": 0.6413,
      "num_input_tokens_seen": 2784961536,
      "step": 37540,
      "train_runtime": 44658.1236,
      "train_tokens_per_second": 62361.813
    },
    {
      "epoch": 0.49900000664447414,
      "grad_norm": 10.875,
      "learning_rate": 1.0518610140650163e-05,
      "loss": 0.5662,
      "num_input_tokens_seen": 2785078272,
      "step": 37550,
      "train_runtime": 44661.009,
      "train_tokens_per_second": 62360.397
    },
    {
      "epoch": 0.499132896126936,
      "grad_norm": 21.25,
      "learning_rate": 1.0514311920252102e-05,
      "loss": 0.8124,
      "num_input_tokens_seen": 2787140096,
      "step": 37560,
      "train_runtime": 44694.5036,
      "train_tokens_per_second": 62359.795
    },
    {
      "epoch": 0.49926578560939794,
      "grad_norm": 17.625,
      "learning_rate": 1.0510013604582343e-05,
      "loss": 0.976,
      "num_input_tokens_seen": 2788008704,
      "step": 37570,
      "train_runtime": 44707.6367,
      "train_tokens_per_second": 62360.905
    },
    {
      "epoch": 0.49939867509185987,
      "grad_norm": 22.5,
      "learning_rate": 1.0505715194437114e-05,
      "loss": 0.7457,
      "num_input_tokens_seen": 2788496128,
      "step": 37580,
      "train_runtime": 44714.576,
      "train_tokens_per_second": 62362.128
    },
    {
      "epoch": 0.49953156457432174,
      "grad_norm": 52.0,
      "learning_rate": 1.0501416690612651e-05,
      "loss": 0.6313,
      "num_input_tokens_seen": 2788733952,
      "step": 37590,
      "train_runtime": 44718.0787,
      "train_tokens_per_second": 62362.562
    },
    {
      "epoch": 0.49966445405678367,
      "grad_norm": 17.625,
      "learning_rate": 1.049711809390522e-05,
      "loss": 0.5116,
      "num_input_tokens_seen": 2788847616,
      "step": 37600,
      "train_runtime": 44720.9936,
      "train_tokens_per_second": 62361.039
    },
    {
      "epoch": 0.4997973435392456,
      "grad_norm": 29.0,
      "learning_rate": 1.0492819405111092e-05,
      "loss": 0.7893,
      "num_input_tokens_seen": 2790832384,
      "step": 37610,
      "train_runtime": 44753.387,
      "train_tokens_per_second": 62360.25
    },
    {
      "epoch": 0.49993023302170747,
      "grad_norm": 22.0,
      "learning_rate": 1.0488520625026563e-05,
      "loss": 0.9476,
      "num_input_tokens_seen": 2791697920,
      "step": 37620,
      "train_runtime": 44766.4404,
      "train_tokens_per_second": 62361.4
    },
    {
      "epoch": 0.5000631225041694,
      "grad_norm": 19.75,
      "learning_rate": 1.0484221754447948e-05,
      "loss": 0.7123,
      "num_input_tokens_seen": 2792181248,
      "step": 37630,
      "train_runtime": 44773.374,
      "train_tokens_per_second": 62362.538
    },
    {
      "epoch": 0.5001960119866313,
      "grad_norm": 31.25,
      "learning_rate": 1.0479922794171567e-05,
      "loss": 0.6624,
      "num_input_tokens_seen": 2792413440,
      "step": 37640,
      "train_runtime": 44776.9044,
      "train_tokens_per_second": 62362.807
    },
    {
      "epoch": 0.5003289014690933,
      "grad_norm": 12.875,
      "learning_rate": 1.0475623744993768e-05,
      "loss": 0.5002,
      "num_input_tokens_seen": 2792529152,
      "step": 37650,
      "train_runtime": 44779.8636,
      "train_tokens_per_second": 62361.27
    },
    {
      "epoch": 0.5004617909515552,
      "grad_norm": 12.125,
      "learning_rate": 1.0471324607710915e-05,
      "loss": 0.8102,
      "num_input_tokens_seen": 2794688512,
      "step": 37660,
      "train_runtime": 44813.5925,
      "train_tokens_per_second": 62362.519
    },
    {
      "epoch": 0.500594680434017,
      "grad_norm": 25.875,
      "learning_rate": 1.046702538311938e-05,
      "loss": 0.8544,
      "num_input_tokens_seen": 2795548928,
      "step": 37670,
      "train_runtime": 44826.3676,
      "train_tokens_per_second": 62363.941
    },
    {
      "epoch": 0.5007275699164789,
      "grad_norm": 27.375,
      "learning_rate": 1.0462726072015556e-05,
      "loss": 0.7689,
      "num_input_tokens_seen": 2796044288,
      "step": 37680,
      "train_runtime": 44833.3034,
      "train_tokens_per_second": 62365.342
    },
    {
      "epoch": 0.5008604593989409,
      "grad_norm": 23.5,
      "learning_rate": 1.0458426675195852e-05,
      "loss": 0.6507,
      "num_input_tokens_seen": 2796273152,
      "step": 37690,
      "train_runtime": 44836.7008,
      "train_tokens_per_second": 62365.721
    },
    {
      "epoch": 0.5009933488814028,
      "grad_norm": 8.0625,
      "learning_rate": 1.0454127193456695e-05,
      "loss": 0.5135,
      "num_input_tokens_seen": 2796386304,
      "step": 37700,
      "train_runtime": 44839.7016,
      "train_tokens_per_second": 62364.07
    },
    {
      "epoch": 0.5011262383638647,
      "grad_norm": 19.5,
      "learning_rate": 1.0449827627594524e-05,
      "loss": 0.7569,
      "num_input_tokens_seen": 2798381824,
      "step": 37710,
      "train_runtime": 44872.9944,
      "train_tokens_per_second": 62362.271
    },
    {
      "epoch": 0.5012591278463266,
      "grad_norm": 28.25,
      "learning_rate": 1.04455279784058e-05,
      "loss": 0.821,
      "num_input_tokens_seen": 2799200768,
      "step": 37720,
      "train_runtime": 44885.398,
      "train_tokens_per_second": 62363.283
    },
    {
      "epoch": 0.5013920173287885,
      "grad_norm": 31.25,
      "learning_rate": 1.0441228246686988e-05,
      "loss": 0.7075,
      "num_input_tokens_seen": 2799701248,
      "step": 37730,
      "train_runtime": 44892.1708,
      "train_tokens_per_second": 62365.023
    },
    {
      "epoch": 0.5015249068112504,
      "grad_norm": 19.375,
      "learning_rate": 1.0436928433234579e-05,
      "loss": 0.5933,
      "num_input_tokens_seen": 2799931136,
      "step": 37740,
      "train_runtime": 44895.5899,
      "train_tokens_per_second": 62365.394
    },
    {
      "epoch": 0.5016577962937123,
      "grad_norm": 15.3125,
      "learning_rate": 1.043262853884507e-05,
      "loss": 0.5353,
      "num_input_tokens_seen": 2800048640,
      "step": 37750,
      "train_runtime": 44898.4931,
      "train_tokens_per_second": 62363.978
    },
    {
      "epoch": 0.5017906857761743,
      "grad_norm": 18.0,
      "learning_rate": 1.0428328564314984e-05,
      "loss": 0.7918,
      "num_input_tokens_seen": 2802195456,
      "step": 37760,
      "train_runtime": 44931.8696,
      "train_tokens_per_second": 62365.432
    },
    {
      "epoch": 0.5019235752586362,
      "grad_norm": 31.125,
      "learning_rate": 1.0424028510440853e-05,
      "loss": 0.9454,
      "num_input_tokens_seen": 2803054848,
      "step": 37770,
      "train_runtime": 44944.8474,
      "train_tokens_per_second": 62366.545
    },
    {
      "epoch": 0.5020564647410981,
      "grad_norm": 12.1875,
      "learning_rate": 1.0419728378019218e-05,
      "loss": 0.6538,
      "num_input_tokens_seen": 2803578624,
      "step": 37780,
      "train_runtime": 44952.0349,
      "train_tokens_per_second": 62368.225
    },
    {
      "epoch": 0.5021893542235599,
      "grad_norm": 20.0,
      "learning_rate": 1.0415428167846645e-05,
      "loss": 0.7198,
      "num_input_tokens_seen": 2803822080,
      "step": 37790,
      "train_runtime": 44955.6602,
      "train_tokens_per_second": 62368.611
    },
    {
      "epoch": 0.5023222437060219,
      "grad_norm": 16.5,
      "learning_rate": 1.0411127880719707e-05,
      "loss": 0.494,
      "num_input_tokens_seen": 2803941888,
      "step": 37800,
      "train_runtime": 44958.6107,
      "train_tokens_per_second": 62367.183
    },
    {
      "epoch": 0.5024551331884838,
      "grad_norm": 20.25,
      "learning_rate": 1.0406827517434996e-05,
      "loss": 0.9239,
      "num_input_tokens_seen": 2805922816,
      "step": 37810,
      "train_runtime": 44991.4225,
      "train_tokens_per_second": 62365.728
    },
    {
      "epoch": 0.5025880226709457,
      "grad_norm": 18.875,
      "learning_rate": 1.0402527078789117e-05,
      "loss": 0.8822,
      "num_input_tokens_seen": 2806792960,
      "step": 37820,
      "train_runtime": 45004.0521,
      "train_tokens_per_second": 62367.561
    },
    {
      "epoch": 0.5027209121534076,
      "grad_norm": 22.375,
      "learning_rate": 1.0398226565578686e-05,
      "loss": 0.7016,
      "num_input_tokens_seen": 2807257088,
      "step": 37830,
      "train_runtime": 45010.7043,
      "train_tokens_per_second": 62368.655
    },
    {
      "epoch": 0.5028538016358696,
      "grad_norm": 21.0,
      "learning_rate": 1.0393925978600333e-05,
      "loss": 0.6164,
      "num_input_tokens_seen": 2807488256,
      "step": 37840,
      "train_runtime": 45014.1521,
      "train_tokens_per_second": 62369.013
    },
    {
      "epoch": 0.5029866911183314,
      "grad_norm": 8.0625,
      "learning_rate": 1.0389625318650704e-05,
      "loss": 0.4322,
      "num_input_tokens_seen": 2807602688,
      "step": 37850,
      "train_runtime": 45017.0287,
      "train_tokens_per_second": 62367.57
    },
    {
      "epoch": 0.5031195806007933,
      "grad_norm": 28.125,
      "learning_rate": 1.038532458652646e-05,
      "loss": 0.8412,
      "num_input_tokens_seen": 2809708544,
      "step": 37860,
      "train_runtime": 45051.3276,
      "train_tokens_per_second": 62366.831
    },
    {
      "epoch": 0.5032524700832552,
      "grad_norm": 12.75,
      "learning_rate": 1.0381023783024274e-05,
      "loss": 0.8666,
      "num_input_tokens_seen": 2810543872,
      "step": 37870,
      "train_runtime": 45063.9927,
      "train_tokens_per_second": 62367.84
    },
    {
      "epoch": 0.5033853595657172,
      "grad_norm": 29.0,
      "learning_rate": 1.0376722908940826e-05,
      "loss": 0.7624,
      "num_input_tokens_seen": 2810995200,
      "step": 37880,
      "train_runtime": 45070.6294,
      "train_tokens_per_second": 62368.67
    },
    {
      "epoch": 0.5035182490481791,
      "grad_norm": 21.375,
      "learning_rate": 1.037242196507282e-05,
      "loss": 0.5755,
      "num_input_tokens_seen": 2811216640,
      "step": 37890,
      "train_runtime": 45074.0523,
      "train_tokens_per_second": 62368.846
    },
    {
      "epoch": 0.503651138530641,
      "grad_norm": 7.90625,
      "learning_rate": 1.0368120952216965e-05,
      "loss": 0.4984,
      "num_input_tokens_seen": 2811329536,
      "step": 37900,
      "train_runtime": 45076.4243,
      "train_tokens_per_second": 62368.069
    },
    {
      "epoch": 0.503784028013103,
      "grad_norm": 17.625,
      "learning_rate": 1.0363819871169989e-05,
      "loss": 0.8272,
      "num_input_tokens_seen": 2813371648,
      "step": 37910,
      "train_runtime": 45110.5408,
      "train_tokens_per_second": 62366.17
    },
    {
      "epoch": 0.5039169174955648,
      "grad_norm": 25.5,
      "learning_rate": 1.0359518722728623e-05,
      "loss": 0.8838,
      "num_input_tokens_seen": 2814219008,
      "step": 37920,
      "train_runtime": 45123.329,
      "train_tokens_per_second": 62367.274
    },
    {
      "epoch": 0.5040498069780267,
      "grad_norm": 32.5,
      "learning_rate": 1.035521750768962e-05,
      "loss": 0.741,
      "num_input_tokens_seen": 2814699264,
      "step": 37930,
      "train_runtime": 45130.3026,
      "train_tokens_per_second": 62368.278
    },
    {
      "epoch": 0.5041826964604886,
      "grad_norm": 101.5,
      "learning_rate": 1.0350916226849743e-05,
      "loss": 0.6559,
      "num_input_tokens_seen": 2814930432,
      "step": 37940,
      "train_runtime": 45133.8116,
      "train_tokens_per_second": 62368.551
    },
    {
      "epoch": 0.5043155859429506,
      "grad_norm": 17.125,
      "learning_rate": 1.0346614881005762e-05,
      "loss": 0.5759,
      "num_input_tokens_seen": 2815049216,
      "step": 37950,
      "train_runtime": 45136.212,
      "train_tokens_per_second": 62367.866
    },
    {
      "epoch": 0.5044484754254125,
      "grad_norm": 23.5,
      "learning_rate": 1.0342313470954464e-05,
      "loss": 0.8137,
      "num_input_tokens_seen": 2817108480,
      "step": 37960,
      "train_runtime": 45169.7891,
      "train_tokens_per_second": 62367.094
    },
    {
      "epoch": 0.5045813649078744,
      "grad_norm": 41.75,
      "learning_rate": 1.033801199749265e-05,
      "loss": 0.9301,
      "num_input_tokens_seen": 2817932800,
      "step": 37970,
      "train_runtime": 45182.6131,
      "train_tokens_per_second": 62367.637
    },
    {
      "epoch": 0.5047142543903362,
      "grad_norm": 53.25,
      "learning_rate": 1.0333710461417129e-05,
      "loss": 0.7446,
      "num_input_tokens_seen": 2818354688,
      "step": 37980,
      "train_runtime": 45188.9235,
      "train_tokens_per_second": 62368.264
    },
    {
      "epoch": 0.5048471438727982,
      "grad_norm": 31.25,
      "learning_rate": 1.032940886352472e-05,
      "loss": 0.6611,
      "num_input_tokens_seen": 2818580736,
      "step": 37990,
      "train_runtime": 45192.3288,
      "train_tokens_per_second": 62368.566
    },
    {
      "epoch": 0.5049800333552601,
      "grad_norm": 13.375,
      "learning_rate": 1.0325107204612253e-05,
      "loss": 0.5754,
      "num_input_tokens_seen": 2818693888,
      "step": 38000,
      "train_runtime": 45194.6698,
      "train_tokens_per_second": 62367.839
    },
    {
      "epoch": 0.505112922837722,
      "grad_norm": 18.125,
      "learning_rate": 1.0320805485476574e-05,
      "loss": 0.8646,
      "num_input_tokens_seen": 2820726528,
      "step": 38010,
      "train_runtime": 45226.2843,
      "train_tokens_per_second": 62369.186
    },
    {
      "epoch": 0.505245812320184,
      "grad_norm": 28.625,
      "learning_rate": 1.0316503706914545e-05,
      "loss": 0.9489,
      "num_input_tokens_seen": 2821562368,
      "step": 38020,
      "train_runtime": 45239.0038,
      "train_tokens_per_second": 62370.126
    },
    {
      "epoch": 0.5053787018026459,
      "grad_norm": 21.375,
      "learning_rate": 1.0312201869723024e-05,
      "loss": 0.7254,
      "num_input_tokens_seen": 2822027008,
      "step": 38030,
      "train_runtime": 45246.3911,
      "train_tokens_per_second": 62370.212
    },
    {
      "epoch": 0.5055115912851077,
      "grad_norm": 14.375,
      "learning_rate": 1.0307899974698889e-05,
      "loss": 0.6353,
      "num_input_tokens_seen": 2822266624,
      "step": 38040,
      "train_runtime": 45249.9861,
      "train_tokens_per_second": 62370.552
    },
    {
      "epoch": 0.5056444807675696,
      "grad_norm": 17.25,
      "learning_rate": 1.0303598022639033e-05,
      "loss": 0.5614,
      "num_input_tokens_seen": 2822383360,
      "step": 38050,
      "train_runtime": 45252.6846,
      "train_tokens_per_second": 62369.413
    },
    {
      "epoch": 0.5057773702500316,
      "grad_norm": 18.0,
      "learning_rate": 1.0299296014340345e-05,
      "loss": 0.7675,
      "num_input_tokens_seen": 2824487680,
      "step": 38060,
      "train_runtime": 45286.4322,
      "train_tokens_per_second": 62369.402
    },
    {
      "epoch": 0.5059102597324935,
      "grad_norm": 42.75,
      "learning_rate": 1.0294993950599743e-05,
      "loss": 0.9349,
      "num_input_tokens_seen": 2825380864,
      "step": 38070,
      "train_runtime": 45299.4408,
      "train_tokens_per_second": 62371.208
    },
    {
      "epoch": 0.5060431492149554,
      "grad_norm": 26.5,
      "learning_rate": 1.0290691832214141e-05,
      "loss": 0.8016,
      "num_input_tokens_seen": 2825847040,
      "step": 38080,
      "train_runtime": 45306.7791,
      "train_tokens_per_second": 62371.396
    },
    {
      "epoch": 0.5061760386974173,
      "grad_norm": 28.125,
      "learning_rate": 1.0286389659980471e-05,
      "loss": 0.6237,
      "num_input_tokens_seen": 2826078208,
      "step": 38090,
      "train_runtime": 45310.2364,
      "train_tokens_per_second": 62371.738
    },
    {
      "epoch": 0.5063089281798792,
      "grad_norm": 34.0,
      "learning_rate": 1.0282087434695669e-05,
      "loss": 0.5535,
      "num_input_tokens_seen": 2826195712,
      "step": 38100,
      "train_runtime": 45312.5757,
      "train_tokens_per_second": 62371.112
    },
    {
      "epoch": 0.5064418176623411,
      "grad_norm": 29.375,
      "learning_rate": 1.0277785157156683e-05,
      "loss": 0.7629,
      "num_input_tokens_seen": 2828240384,
      "step": 38110,
      "train_runtime": 45346.9146,
      "train_tokens_per_second": 62368.971
    },
    {
      "epoch": 0.506574707144803,
      "grad_norm": 14.0,
      "learning_rate": 1.0273482828160475e-05,
      "loss": 0.9251,
      "num_input_tokens_seen": 2829068544,
      "step": 38120,
      "train_runtime": 45360.0778,
      "train_tokens_per_second": 62369.129
    },
    {
      "epoch": 0.506707596627265,
      "grad_norm": 14.6875,
      "learning_rate": 1.0269180448504015e-05,
      "loss": 0.718,
      "num_input_tokens_seen": 2829552640,
      "step": 38130,
      "train_runtime": 45366.9702,
      "train_tokens_per_second": 62370.324
    },
    {
      "epoch": 0.5068404861097269,
      "grad_norm": 19.375,
      "learning_rate": 1.0264878018984272e-05,
      "loss": 0.6267,
      "num_input_tokens_seen": 2829794560,
      "step": 38140,
      "train_runtime": 45370.5538,
      "train_tokens_per_second": 62370.73
    },
    {
      "epoch": 0.5069733755921888,
      "grad_norm": 18.25,
      "learning_rate": 1.0260575540398241e-05,
      "loss": 0.5964,
      "num_input_tokens_seen": 2829914112,
      "step": 38150,
      "train_runtime": 45372.9362,
      "train_tokens_per_second": 62370.09
    },
    {
      "epoch": 0.5071062650746506,
      "grad_norm": 19.375,
      "learning_rate": 1.0256273013542915e-05,
      "loss": 0.9367,
      "num_input_tokens_seen": 2831970304,
      "step": 38160,
      "train_runtime": 45406.8238,
      "train_tokens_per_second": 62368.826
    },
    {
      "epoch": 0.5072391545571125,
      "grad_norm": 34.0,
      "learning_rate": 1.0251970439215295e-05,
      "loss": 0.9141,
      "num_input_tokens_seen": 2832781568,
      "step": 38170,
      "train_runtime": 45419.3642,
      "train_tokens_per_second": 62369.468
    },
    {
      "epoch": 0.5073720440395745,
      "grad_norm": 18.375,
      "learning_rate": 1.0247667818212399e-05,
      "loss": 0.6963,
      "num_input_tokens_seen": 2833260544,
      "step": 38180,
      "train_runtime": 45426.01,
      "train_tokens_per_second": 62370.887
    },
    {
      "epoch": 0.5075049335220364,
      "grad_norm": 34.75,
      "learning_rate": 1.024336515133125e-05,
      "loss": 0.6354,
      "num_input_tokens_seen": 2833485312,
      "step": 38190,
      "train_runtime": 45430.0257,
      "train_tokens_per_second": 62370.322
    },
    {
      "epoch": 0.5076378230044983,
      "grad_norm": 14.625,
      "learning_rate": 1.0239062439368876e-05,
      "loss": 0.4929,
      "num_input_tokens_seen": 2833600256,
      "step": 38200,
      "train_runtime": 45432.3422,
      "train_tokens_per_second": 62369.671
    },
    {
      "epoch": 0.5077707124869603,
      "grad_norm": 23.625,
      "learning_rate": 1.0234759683122314e-05,
      "loss": 0.8879,
      "num_input_tokens_seen": 2835664384,
      "step": 38210,
      "train_runtime": 45466.3424,
      "train_tokens_per_second": 62368.43
    },
    {
      "epoch": 0.5079036019694221,
      "grad_norm": 52.0,
      "learning_rate": 1.0230456883388617e-05,
      "loss": 0.8669,
      "num_input_tokens_seen": 2836521984,
      "step": 38220,
      "train_runtime": 45479.1921,
      "train_tokens_per_second": 62369.665
    },
    {
      "epoch": 0.508036491451884,
      "grad_norm": 29.375,
      "learning_rate": 1.0226154040964839e-05,
      "loss": 0.7533,
      "num_input_tokens_seen": 2836988928,
      "step": 38230,
      "train_runtime": 45485.9312,
      "train_tokens_per_second": 62370.69
    },
    {
      "epoch": 0.5081693809343459,
      "grad_norm": 13.8125,
      "learning_rate": 1.022185115664804e-05,
      "loss": 0.6142,
      "num_input_tokens_seen": 2837219584,
      "step": 38240,
      "train_runtime": 45489.3807,
      "train_tokens_per_second": 62371.031
    },
    {
      "epoch": 0.5083022704168079,
      "grad_norm": 19.875,
      "learning_rate": 1.0217548231235293e-05,
      "loss": 0.4936,
      "num_input_tokens_seen": 2837339136,
      "step": 38250,
      "train_runtime": 45492.2883,
      "train_tokens_per_second": 62369.673
    },
    {
      "epoch": 0.5084351598992698,
      "grad_norm": 19.25,
      "learning_rate": 1.0213245265523672e-05,
      "loss": 0.883,
      "num_input_tokens_seen": 2839382272,
      "step": 38260,
      "train_runtime": 45524.4639,
      "train_tokens_per_second": 62370.471
    },
    {
      "epoch": 0.5085680493817317,
      "grad_norm": 18.25,
      "learning_rate": 1.020894226031027e-05,
      "loss": 0.9289,
      "num_input_tokens_seen": 2840240384,
      "step": 38270,
      "train_runtime": 45537.0371,
      "train_tokens_per_second": 62372.095
    },
    {
      "epoch": 0.5087009388641935,
      "grad_norm": 18.75,
      "learning_rate": 1.0204639216392177e-05,
      "loss": 0.6544,
      "num_input_tokens_seen": 2840698368,
      "step": 38280,
      "train_runtime": 45543.5461,
      "train_tokens_per_second": 62373.236
    },
    {
      "epoch": 0.5088338283466555,
      "grad_norm": 28.125,
      "learning_rate": 1.0200336134566496e-05,
      "loss": 0.6123,
      "num_input_tokens_seen": 2840911872,
      "step": 38290,
      "train_runtime": 45547.3943,
      "train_tokens_per_second": 62372.654
    },
    {
      "epoch": 0.5089667178291174,
      "grad_norm": 8.1875,
      "learning_rate": 1.0196033015630328e-05,
      "loss": 0.5211,
      "num_input_tokens_seen": 2841027840,
      "step": 38300,
      "train_runtime": 45549.6873,
      "train_tokens_per_second": 62372.06
    },
    {
      "epoch": 0.5090996073115793,
      "grad_norm": 24.375,
      "learning_rate": 1.0191729860380796e-05,
      "loss": 0.8048,
      "num_input_tokens_seen": 2843073792,
      "step": 38310,
      "train_runtime": 45584.0155,
      "train_tokens_per_second": 62369.973
    },
    {
      "epoch": 0.5092324967940413,
      "grad_norm": 10.375,
      "learning_rate": 1.0187426669615015e-05,
      "loss": 0.9551,
      "num_input_tokens_seen": 2843919872,
      "step": 38320,
      "train_runtime": 45596.5692,
      "train_tokens_per_second": 62371.357
    },
    {
      "epoch": 0.5093653862765032,
      "grad_norm": 39.75,
      "learning_rate": 1.0183123444130117e-05,
      "loss": 0.7805,
      "num_input_tokens_seen": 2844399360,
      "step": 38330,
      "train_runtime": 45603.2641,
      "train_tokens_per_second": 62372.714
    },
    {
      "epoch": 0.509498275758965,
      "grad_norm": 17.5,
      "learning_rate": 1.017882018472323e-05,
      "loss": 0.6537,
      "num_input_tokens_seen": 2844630272,
      "step": 38340,
      "train_runtime": 45606.7284,
      "train_tokens_per_second": 62373.04
    },
    {
      "epoch": 0.5096311652414269,
      "grad_norm": 10.0,
      "learning_rate": 1.0174516892191502e-05,
      "loss": 0.4736,
      "num_input_tokens_seen": 2844747008,
      "step": 38350,
      "train_runtime": 45609.6095,
      "train_tokens_per_second": 62371.659
    },
    {
      "epoch": 0.5097640547238889,
      "grad_norm": 26.25,
      "learning_rate": 1.0170213567332072e-05,
      "loss": 0.8119,
      "num_input_tokens_seen": 2846785536,
      "step": 38360,
      "train_runtime": 45642.7854,
      "train_tokens_per_second": 62370.986
    },
    {
      "epoch": 0.5098969442063508,
      "grad_norm": 23.375,
      "learning_rate": 1.0165910210942098e-05,
      "loss": 0.8898,
      "num_input_tokens_seen": 2847626752,
      "step": 38370,
      "train_runtime": 45655.4556,
      "train_tokens_per_second": 62372.102
    },
    {
      "epoch": 0.5100298336888127,
      "grad_norm": 17.25,
      "learning_rate": 1.0161606823818738e-05,
      "loss": 0.6966,
      "num_input_tokens_seen": 2848083712,
      "step": 38380,
      "train_runtime": 45662.1149,
      "train_tokens_per_second": 62373.014
    },
    {
      "epoch": 0.5101627231712746,
      "grad_norm": 15.0625,
      "learning_rate": 1.0157303406759157e-05,
      "loss": 0.6169,
      "num_input_tokens_seen": 2848317184,
      "step": 38390,
      "train_runtime": 45666.1148,
      "train_tokens_per_second": 62372.663
    },
    {
      "epoch": 0.5102956126537366,
      "grad_norm": 12.8125,
      "learning_rate": 1.0152999960560519e-05,
      "loss": 0.5375,
      "num_input_tokens_seen": 2848432128,
      "step": 38400,
      "train_runtime": 45668.4907,
      "train_tokens_per_second": 62371.935
    },
    {
      "epoch": 0.5104285021361984,
      "grad_norm": 24.875,
      "learning_rate": 1.0148696486020003e-05,
      "loss": 0.7974,
      "num_input_tokens_seen": 2850487808,
      "step": 38410,
      "train_runtime": 45701.0045,
      "train_tokens_per_second": 62372.542
    },
    {
      "epoch": 0.5105613916186603,
      "grad_norm": 35.0,
      "learning_rate": 1.0144392983934788e-05,
      "loss": 0.9454,
      "num_input_tokens_seen": 2851324160,
      "step": 38420,
      "train_runtime": 45713.8987,
      "train_tokens_per_second": 62373.244
    },
    {
      "epoch": 0.5106942811011222,
      "grad_norm": 23.625,
      "learning_rate": 1.0140089455102062e-05,
      "loss": 0.7661,
      "num_input_tokens_seen": 2851784960,
      "step": 38430,
      "train_runtime": 45720.7173,
      "train_tokens_per_second": 62374.02
    },
    {
      "epoch": 0.5108271705835842,
      "grad_norm": 18.125,
      "learning_rate": 1.0135785900319013e-05,
      "loss": 0.6801,
      "num_input_tokens_seen": 2852018688,
      "step": 38440,
      "train_runtime": 45724.1497,
      "train_tokens_per_second": 62374.45
    },
    {
      "epoch": 0.5109600600660461,
      "grad_norm": 11.375,
      "learning_rate": 1.0131482320382837e-05,
      "loss": 0.5472,
      "num_input_tokens_seen": 2852133888,
      "step": 38450,
      "train_runtime": 45726.5207,
      "train_tokens_per_second": 62373.735
    },
    {
      "epoch": 0.511092949548508,
      "grad_norm": 21.5,
      "learning_rate": 1.0127178716090734e-05,
      "loss": 0.8245,
      "num_input_tokens_seen": 2854142208,
      "step": 38460,
      "train_runtime": 45758.6805,
      "train_tokens_per_second": 62373.787
    },
    {
      "epoch": 0.5112258390309699,
      "grad_norm": 28.5,
      "learning_rate": 1.0122875088239907e-05,
      "loss": 0.8516,
      "num_input_tokens_seen": 2854972928,
      "step": 38470,
      "train_runtime": 45771.645,
      "train_tokens_per_second": 62374.27
    },
    {
      "epoch": 0.5113587285134318,
      "grad_norm": 36.75,
      "learning_rate": 1.0118571437627567e-05,
      "loss": 0.7672,
      "num_input_tokens_seen": 2855450368,
      "step": 38480,
      "train_runtime": 45778.3427,
      "train_tokens_per_second": 62375.573
    },
    {
      "epoch": 0.5114916179958937,
      "grad_norm": 38.5,
      "learning_rate": 1.0114267765050923e-05,
      "loss": 0.7073,
      "num_input_tokens_seen": 2855684096,
      "step": 38490,
      "train_runtime": 45781.8609,
      "train_tokens_per_second": 62375.885
    },
    {
      "epoch": 0.5116245074783556,
      "grad_norm": 46.5,
      "learning_rate": 1.0109964071307194e-05,
      "loss": 0.5909,
      "num_input_tokens_seen": 2855804672,
      "step": 38500,
      "train_runtime": 45784.2854,
      "train_tokens_per_second": 62375.216
    },
    {
      "epoch": 0.5117573969608176,
      "grad_norm": 15.4375,
      "learning_rate": 1.0105660357193605e-05,
      "loss": 0.8633,
      "num_input_tokens_seen": 2857751296,
      "step": 38510,
      "train_runtime": 45814.8326,
      "train_tokens_per_second": 62376.116
    },
    {
      "epoch": 0.5118902864432795,
      "grad_norm": 13.0625,
      "learning_rate": 1.0101356623507374e-05,
      "loss": 0.8522,
      "num_input_tokens_seen": 2858611200,
      "step": 38520,
      "train_runtime": 45827.4282,
      "train_tokens_per_second": 62377.736
    },
    {
      "epoch": 0.5120231759257413,
      "grad_norm": 26.125,
      "learning_rate": 1.0097052871045734e-05,
      "loss": 0.8293,
      "num_input_tokens_seen": 2859079168,
      "step": 38530,
      "train_runtime": 45834.5826,
      "train_tokens_per_second": 62378.209
    },
    {
      "epoch": 0.5121560654082032,
      "grad_norm": 15.875,
      "learning_rate": 1.0092749100605911e-05,
      "loss": 0.5675,
      "num_input_tokens_seen": 2859304448,
      "step": 38540,
      "train_runtime": 45837.9645,
      "train_tokens_per_second": 62378.521
    },
    {
      "epoch": 0.5122889548906652,
      "grad_norm": 15.6875,
      "learning_rate": 1.0088445312985152e-05,
      "loss": 0.4667,
      "num_input_tokens_seen": 2859417600,
      "step": 38550,
      "train_runtime": 45840.35,
      "train_tokens_per_second": 62377.744
    },
    {
      "epoch": 0.5124218443731271,
      "grad_norm": 25.75,
      "learning_rate": 1.0084141508980683e-05,
      "loss": 0.8217,
      "num_input_tokens_seen": 2861511168,
      "step": 38560,
      "train_runtime": 45874.0381,
      "train_tokens_per_second": 62377.573
    },
    {
      "epoch": 0.512554733855589,
      "grad_norm": 19.875,
      "learning_rate": 1.0079837689389749e-05,
      "loss": 0.9087,
      "num_input_tokens_seen": 2862373888,
      "step": 38570,
      "train_runtime": 45886.9668,
      "train_tokens_per_second": 62378.799
    },
    {
      "epoch": 0.512687623338051,
      "grad_norm": 27.0,
      "learning_rate": 1.0075533855009596e-05,
      "loss": 0.8251,
      "num_input_tokens_seen": 2862849024,
      "step": 38580,
      "train_runtime": 45894.5856,
      "train_tokens_per_second": 62378.797
    },
    {
      "epoch": 0.5128205128205128,
      "grad_norm": 20.0,
      "learning_rate": 1.0071230006637467e-05,
      "loss": 0.6819,
      "num_input_tokens_seen": 2863094784,
      "step": 38590,
      "train_runtime": 45898.1669,
      "train_tokens_per_second": 62379.284
    },
    {
      "epoch": 0.5129534023029747,
      "grad_norm": 13.6875,
      "learning_rate": 1.0066926145070618e-05,
      "loss": 0.4991,
      "num_input_tokens_seen": 2863215360,
      "step": 38600,
      "train_runtime": 45900.5802,
      "train_tokens_per_second": 62378.631
    },
    {
      "epoch": 0.5130862917854366,
      "grad_norm": 18.625,
      "learning_rate": 1.0062622271106298e-05,
      "loss": 0.7718,
      "num_input_tokens_seen": 2865155328,
      "step": 38610,
      "train_runtime": 45932.8457,
      "train_tokens_per_second": 62377.048
    },
    {
      "epoch": 0.5132191812678986,
      "grad_norm": 21.25,
      "learning_rate": 1.0058318385541756e-05,
      "loss": 0.8345,
      "num_input_tokens_seen": 2866002176,
      "step": 38620,
      "train_runtime": 45946.0638,
      "train_tokens_per_second": 62377.534
    },
    {
      "epoch": 0.5133520707503605,
      "grad_norm": 24.0,
      "learning_rate": 1.005401448917426e-05,
      "loss": 0.7104,
      "num_input_tokens_seen": 2866452480,
      "step": 38630,
      "train_runtime": 45952.5063,
      "train_tokens_per_second": 62378.589
    },
    {
      "epoch": 0.5134849602328224,
      "grad_norm": 25.75,
      "learning_rate": 1.0049710582801055e-05,
      "loss": 0.6382,
      "num_input_tokens_seen": 2866684416,
      "step": 38640,
      "train_runtime": 45955.8954,
      "train_tokens_per_second": 62379.035
    },
    {
      "epoch": 0.5136178497152842,
      "grad_norm": 8.4375,
      "learning_rate": 1.0045406667219411e-05,
      "loss": 0.5339,
      "num_input_tokens_seen": 2866797056,
      "step": 38650,
      "train_runtime": 45958.2317,
      "train_tokens_per_second": 62378.315
    },
    {
      "epoch": 0.5137507391977462,
      "grad_norm": 14.3125,
      "learning_rate": 1.0041102743226588e-05,
      "loss": 0.8013,
      "num_input_tokens_seen": 2868864256,
      "step": 38660,
      "train_runtime": 45992.0907,
      "train_tokens_per_second": 62377.339
    },
    {
      "epoch": 0.5138836286802081,
      "grad_norm": 24.375,
      "learning_rate": 1.0036798811619846e-05,
      "loss": 0.9336,
      "num_input_tokens_seen": 2869702656,
      "step": 38670,
      "train_runtime": 46004.6363,
      "train_tokens_per_second": 62378.553
    },
    {
      "epoch": 0.51401651816267,
      "grad_norm": 16.875,
      "learning_rate": 1.003249487319645e-05,
      "loss": 0.7016,
      "num_input_tokens_seen": 2870173696,
      "step": 38680,
      "train_runtime": 46011.3298,
      "train_tokens_per_second": 62379.716
    },
    {
      "epoch": 0.514149407645132,
      "grad_norm": 640.0,
      "learning_rate": 1.0028190928753672e-05,
      "loss": 0.5705,
      "num_input_tokens_seen": 2870400256,
      "step": 38690,
      "train_runtime": 46014.8093,
      "train_tokens_per_second": 62379.923
    },
    {
      "epoch": 0.5142822971275939,
      "grad_norm": 16.375,
      "learning_rate": 1.0023886979088773e-05,
      "loss": 0.6202,
      "num_input_tokens_seen": 2870513664,
      "step": 38700,
      "train_runtime": 46017.7477,
      "train_tokens_per_second": 62378.404
    },
    {
      "epoch": 0.5144151866100557,
      "grad_norm": 57.25,
      "learning_rate": 1.0019583024999029e-05,
      "loss": 0.7544,
      "num_input_tokens_seen": 2872504832,
      "step": 38710,
      "train_runtime": 46050.2993,
      "train_tokens_per_second": 62377.55
    },
    {
      "epoch": 0.5145480760925176,
      "grad_norm": 24.875,
      "learning_rate": 1.0015279067281702e-05,
      "loss": 0.8795,
      "num_input_tokens_seen": 2873321984,
      "step": 38720,
      "train_runtime": 46062.6807,
      "train_tokens_per_second": 62378.523
    },
    {
      "epoch": 0.5146809655749796,
      "grad_norm": 28.75,
      "learning_rate": 1.0010975106734062e-05,
      "loss": 0.6917,
      "num_input_tokens_seen": 2873767936,
      "step": 38730,
      "train_runtime": 46069.2751,
      "train_tokens_per_second": 62379.274
    },
    {
      "epoch": 0.5148138550574415,
      "grad_norm": 24.625,
      "learning_rate": 1.0006671144153384e-05,
      "loss": 0.662,
      "num_input_tokens_seen": 2873987584,
      "step": 38740,
      "train_runtime": 46072.5966,
      "train_tokens_per_second": 62379.544
    },
    {
      "epoch": 0.5149467445399034,
      "grad_norm": 23.125,
      "learning_rate": 1.0002367180336934e-05,
      "loss": 0.5028,
      "num_input_tokens_seen": 2874101760,
      "step": 38750,
      "train_runtime": 46074.9156,
      "train_tokens_per_second": 62378.883
    },
    {
      "epoch": 0.5150796340223653,
      "grad_norm": 15.375,
      "learning_rate": 9.998063216081986e-06,
      "loss": 0.8179,
      "num_input_tokens_seen": 2876124672,
      "step": 38760,
      "train_runtime": 46108.4693,
      "train_tokens_per_second": 62377.362
    },
    {
      "epoch": 0.5152125235048272,
      "grad_norm": 17.25,
      "learning_rate": 9.99375925218581e-06,
      "loss": 0.8704,
      "num_input_tokens_seen": 2876987136,
      "step": 38770,
      "train_runtime": 46121.4996,
      "train_tokens_per_second": 62378.439
    },
    {
      "epoch": 0.5153454129872891,
      "grad_norm": 14.625,
      "learning_rate": 9.989455289445678e-06,
      "loss": 0.7449,
      "num_input_tokens_seen": 2877474816,
      "step": 38780,
      "train_runtime": 46128.4893,
      "train_tokens_per_second": 62379.559
    },
    {
      "epoch": 0.515478302469751,
      "grad_norm": 22.0,
      "learning_rate": 9.985151328658855e-06,
      "loss": 0.6076,
      "num_input_tokens_seen": 2877712896,
      "step": 38790,
      "train_runtime": 46132.0627,
      "train_tokens_per_second": 62379.888
    },
    {
      "epoch": 0.5156111919522129,
      "grad_norm": 13.875,
      "learning_rate": 9.98084737062262e-06,
      "loss": 0.5211,
      "num_input_tokens_seen": 2877831680,
      "step": 38800,
      "train_runtime": 46134.5132,
      "train_tokens_per_second": 62379.149
    },
    {
      "epoch": 0.5157440814346749,
      "grad_norm": 15.0625,
      "learning_rate": 9.976543416134237e-06,
      "loss": 0.8059,
      "num_input_tokens_seen": 2879910400,
      "step": 38810,
      "train_runtime": 46167.767,
      "train_tokens_per_second": 62379.244
    },
    {
      "epoch": 0.5158769709171368,
      "grad_norm": 30.375,
      "learning_rate": 9.97223946599098e-06,
      "loss": 0.893,
      "num_input_tokens_seen": 2880752640,
      "step": 38820,
      "train_runtime": 46180.3529,
      "train_tokens_per_second": 62380.481
    },
    {
      "epoch": 0.5160098603995986,
      "grad_norm": 49.0,
      "learning_rate": 9.967935520990115e-06,
      "loss": 0.761,
      "num_input_tokens_seen": 2881200128,
      "step": 38830,
      "train_runtime": 46186.8049,
      "train_tokens_per_second": 62381.456
    },
    {
      "epoch": 0.5161427498820605,
      "grad_norm": 12.9375,
      "learning_rate": 9.963631581928903e-06,
      "loss": 0.5784,
      "num_input_tokens_seen": 2881419520,
      "step": 38840,
      "train_runtime": 46190.1228,
      "train_tokens_per_second": 62381.725
    },
    {
      "epoch": 0.5162756393645225,
      "grad_norm": 13.3125,
      "learning_rate": 9.959327649604621e-06,
      "loss": 0.5166,
      "num_input_tokens_seen": 2881533696,
      "step": 38850,
      "train_runtime": 46192.4817,
      "train_tokens_per_second": 62381.011
    },
    {
      "epoch": 0.5164085288469844,
      "grad_norm": 32.0,
      "learning_rate": 9.955023724814524e-06,
      "loss": 0.8024,
      "num_input_tokens_seen": 2883647232,
      "step": 38860,
      "train_runtime": 46225.6636,
      "train_tokens_per_second": 62381.954
    },
    {
      "epoch": 0.5165414183294463,
      "grad_norm": 28.5,
      "learning_rate": 9.950719808355886e-06,
      "loss": 0.8567,
      "num_input_tokens_seen": 2884440832,
      "step": 38870,
      "train_runtime": 46237.7387,
      "train_tokens_per_second": 62382.827
    },
    {
      "epoch": 0.5166743078119083,
      "grad_norm": 25.125,
      "learning_rate": 9.946415901025963e-06,
      "loss": 0.7419,
      "num_input_tokens_seen": 2884888576,
      "step": 38880,
      "train_runtime": 46244.3097,
      "train_tokens_per_second": 62383.645
    },
    {
      "epoch": 0.5168071972943702,
      "grad_norm": 14.6875,
      "learning_rate": 9.942112003622018e-06,
      "loss": 0.5661,
      "num_input_tokens_seen": 2885113856,
      "step": 38890,
      "train_runtime": 46247.7623,
      "train_tokens_per_second": 62383.858
    },
    {
      "epoch": 0.516940086776832,
      "grad_norm": 12.625,
      "learning_rate": 9.937808116941303e-06,
      "loss": 0.5553,
      "num_input_tokens_seen": 2885225984,
      "step": 38900,
      "train_runtime": 46250.1733,
      "train_tokens_per_second": 62383.031
    },
    {
      "epoch": 0.5170729762592939,
      "grad_norm": 23.125,
      "learning_rate": 9.933504241781084e-06,
      "loss": 0.8722,
      "num_input_tokens_seen": 2887177472,
      "step": 38910,
      "train_runtime": 46282.1474,
      "train_tokens_per_second": 62382.098
    },
    {
      "epoch": 0.5172058657417559,
      "grad_norm": 28.875,
      "learning_rate": 9.929200378938613e-06,
      "loss": 0.9337,
      "num_input_tokens_seen": 2888007168,
      "step": 38920,
      "train_runtime": 46294.4072,
      "train_tokens_per_second": 62383.5
    },
    {
      "epoch": 0.5173387552242178,
      "grad_norm": 17.625,
      "learning_rate": 9.924896529211137e-06,
      "loss": 0.7509,
      "num_input_tokens_seen": 2888437504,
      "step": 38930,
      "train_runtime": 46300.6104,
      "train_tokens_per_second": 62384.437
    },
    {
      "epoch": 0.5174716447066797,
      "grad_norm": 11.625,
      "learning_rate": 9.92059269339591e-06,
      "loss": 0.5705,
      "num_input_tokens_seen": 2888655616,
      "step": 38940,
      "train_runtime": 46303.929,
      "train_tokens_per_second": 62384.676
    },
    {
      "epoch": 0.5176045341891417,
      "grad_norm": 38.25,
      "learning_rate": 9.916288872290176e-06,
      "loss": 0.6104,
      "num_input_tokens_seen": 2888772352,
      "step": 38950,
      "train_runtime": 46306.2839,
      "train_tokens_per_second": 62384.025
    },
    {
      "epoch": 0.5177374236716035,
      "grad_norm": 17.875,
      "learning_rate": 9.911985066691186e-06,
      "loss": 0.7261,
      "num_input_tokens_seen": 2890893312,
      "step": 38960,
      "train_runtime": 46340.3066,
      "train_tokens_per_second": 62383.992
    },
    {
      "epoch": 0.5178703131540654,
      "grad_norm": 18.5,
      "learning_rate": 9.907681277396177e-06,
      "loss": 0.8888,
      "num_input_tokens_seen": 2891744256,
      "step": 38970,
      "train_runtime": 46353.0568,
      "train_tokens_per_second": 62385.19
    },
    {
      "epoch": 0.5180032026365273,
      "grad_norm": 26.25,
      "learning_rate": 9.903377505202389e-06,
      "loss": 0.7714,
      "num_input_tokens_seen": 2892232960,
      "step": 38980,
      "train_runtime": 46359.9039,
      "train_tokens_per_second": 62386.518
    },
    {
      "epoch": 0.5181360921189893,
      "grad_norm": 12.5625,
      "learning_rate": 9.899073750907055e-06,
      "loss": 0.5938,
      "num_input_tokens_seen": 2892462848,
      "step": 38990,
      "train_runtime": 46363.4012,
      "train_tokens_per_second": 62386.77
    },
    {
      "epoch": 0.5182689816014512,
      "grad_norm": 12.625,
      "learning_rate": 9.894770015307407e-06,
      "loss": 0.524,
      "num_input_tokens_seen": 2892579840,
      "step": 39000,
      "train_runtime": 46365.7828,
      "train_tokens_per_second": 62386.089
    },
    {
      "epoch": 0.5184018710839131,
      "grad_norm": 24.125,
      "learning_rate": 9.890466299200678e-06,
      "loss": 0.7382,
      "num_input_tokens_seen": 2894714112,
      "step": 39010,
      "train_runtime": 46399.8662,
      "train_tokens_per_second": 62386.26
    },
    {
      "epoch": 0.5185347605663749,
      "grad_norm": 26.75,
      "learning_rate": 9.886162603384086e-06,
      "loss": 0.9164,
      "num_input_tokens_seen": 2895522048,
      "step": 39020,
      "train_runtime": 46412.3266,
      "train_tokens_per_second": 62386.919
    },
    {
      "epoch": 0.5186676500488369,
      "grad_norm": 21.5,
      "learning_rate": 9.881858928654861e-06,
      "loss": 0.7043,
      "num_input_tokens_seen": 2895972096,
      "step": 39030,
      "train_runtime": 46418.8539,
      "train_tokens_per_second": 62387.841
    },
    {
      "epoch": 0.5188005395312988,
      "grad_norm": 11.125,
      "learning_rate": 9.877555275810216e-06,
      "loss": 0.648,
      "num_input_tokens_seen": 2896198400,
      "step": 39040,
      "train_runtime": 46422.271,
      "train_tokens_per_second": 62388.124
    },
    {
      "epoch": 0.5189334290137607,
      "grad_norm": 8.4375,
      "learning_rate": 9.873251645647364e-06,
      "loss": 0.5352,
      "num_input_tokens_seen": 2896313600,
      "step": 39050,
      "train_runtime": 46424.6447,
      "train_tokens_per_second": 62387.416
    },
    {
      "epoch": 0.5190663184962226,
      "grad_norm": 21.375,
      "learning_rate": 9.86894803896351e-06,
      "loss": 0.8294,
      "num_input_tokens_seen": 2898366976,
      "step": 39060,
      "train_runtime": 46457.8195,
      "train_tokens_per_second": 62387.064
    },
    {
      "epoch": 0.5191992079786846,
      "grad_norm": 18.5,
      "learning_rate": 9.86464445655587e-06,
      "loss": 0.987,
      "num_input_tokens_seen": 2899225600,
      "step": 39070,
      "train_runtime": 46470.7089,
      "train_tokens_per_second": 62388.237
    },
    {
      "epoch": 0.5193320974611464,
      "grad_norm": 12.5,
      "learning_rate": 9.860340899221636e-06,
      "loss": 0.7379,
      "num_input_tokens_seen": 2899709952,
      "step": 39080,
      "train_runtime": 46478.3867,
      "train_tokens_per_second": 62388.352
    },
    {
      "epoch": 0.5194649869436083,
      "grad_norm": 24.375,
      "learning_rate": 9.856037367758002e-06,
      "loss": 0.7065,
      "num_input_tokens_seen": 2899951872,
      "step": 39090,
      "train_runtime": 46481.9843,
      "train_tokens_per_second": 62388.728
    },
    {
      "epoch": 0.5195978764260702,
      "grad_norm": 10.5,
      "learning_rate": 9.851733862962165e-06,
      "loss": 0.6157,
      "num_input_tokens_seen": 2900072448,
      "step": 39100,
      "train_runtime": 46484.4023,
      "train_tokens_per_second": 62388.076
    },
    {
      "epoch": 0.5197307659085322,
      "grad_norm": 17.5,
      "learning_rate": 9.847430385631304e-06,
      "loss": 0.7989,
      "num_input_tokens_seen": 2902157056,
      "step": 39110,
      "train_runtime": 46517.2656,
      "train_tokens_per_second": 62388.814
    },
    {
      "epoch": 0.5198636553909941,
      "grad_norm": 22.75,
      "learning_rate": 9.843126936562609e-06,
      "loss": 0.8959,
      "num_input_tokens_seen": 2903002880,
      "step": 39120,
      "train_runtime": 46529.5792,
      "train_tokens_per_second": 62390.482
    },
    {
      "epoch": 0.519996544873456,
      "grad_norm": 22.625,
      "learning_rate": 9.838823516553245e-06,
      "loss": 0.8256,
      "num_input_tokens_seen": 2903481344,
      "step": 39130,
      "train_runtime": 46537.1218,
      "train_tokens_per_second": 62390.651
    },
    {
      "epoch": 0.5201294343559179,
      "grad_norm": 21.25,
      "learning_rate": 9.83452012640039e-06,
      "loss": 0.6747,
      "num_input_tokens_seen": 2903716864,
      "step": 39140,
      "train_runtime": 46540.7118,
      "train_tokens_per_second": 62390.899
    },
    {
      "epoch": 0.5202623238383798,
      "grad_norm": 12.9375,
      "learning_rate": 9.830216766901208e-06,
      "loss": 0.615,
      "num_input_tokens_seen": 2903831552,
      "step": 39150,
      "train_runtime": 46543.1244,
      "train_tokens_per_second": 62390.129
    },
    {
      "epoch": 0.5203952133208417,
      "grad_norm": 30.5,
      "learning_rate": 9.825913438852851e-06,
      "loss": 0.7438,
      "num_input_tokens_seen": 2905867776,
      "step": 39160,
      "train_runtime": 46575.7801,
      "train_tokens_per_second": 62390.104
    },
    {
      "epoch": 0.5205281028033036,
      "grad_norm": 15.9375,
      "learning_rate": 9.82161014305248e-06,
      "loss": 0.8973,
      "num_input_tokens_seen": 2906716416,
      "step": 39170,
      "train_runtime": 46588.9578,
      "train_tokens_per_second": 62390.673
    },
    {
      "epoch": 0.5206609922857656,
      "grad_norm": 30.375,
      "learning_rate": 9.817306880297239e-06,
      "loss": 0.8214,
      "num_input_tokens_seen": 2907190272,
      "step": 39180,
      "train_runtime": 46595.694,
      "train_tokens_per_second": 62391.823
    },
    {
      "epoch": 0.5207938817682275,
      "grad_norm": 25.875,
      "learning_rate": 9.813003651384271e-06,
      "loss": 0.664,
      "num_input_tokens_seen": 2907421952,
      "step": 39190,
      "train_runtime": 46599.1625,
      "train_tokens_per_second": 62392.15
    },
    {
      "epoch": 0.5209267712506893,
      "grad_norm": 12.1875,
      "learning_rate": 9.808700457110705e-06,
      "loss": 0.5112,
      "num_input_tokens_seen": 2907539200,
      "step": 39200,
      "train_runtime": 46601.5336,
      "train_tokens_per_second": 62391.492
    },
    {
      "epoch": 0.5210596607331512,
      "grad_norm": 16.375,
      "learning_rate": 9.804397298273678e-06,
      "loss": 0.8767,
      "num_input_tokens_seen": 2909457664,
      "step": 39210,
      "train_runtime": 46633.1966,
      "train_tokens_per_second": 62390.269
    },
    {
      "epoch": 0.5211925502156132,
      "grad_norm": 22.5,
      "learning_rate": 9.800094175670303e-06,
      "loss": 0.8943,
      "num_input_tokens_seen": 2910311680,
      "step": 39220,
      "train_runtime": 46646.1951,
      "train_tokens_per_second": 62391.191
    },
    {
      "epoch": 0.5213254396980751,
      "grad_norm": 18.375,
      "learning_rate": 9.795791090097703e-06,
      "loss": 0.7516,
      "num_input_tokens_seen": 2910809856,
      "step": 39230,
      "train_runtime": 46653.7293,
      "train_tokens_per_second": 62391.794
    },
    {
      "epoch": 0.521458329180537,
      "grad_norm": 13.375,
      "learning_rate": 9.791488042352984e-06,
      "loss": 0.6795,
      "num_input_tokens_seen": 2911048448,
      "step": 39240,
      "train_runtime": 46657.3019,
      "train_tokens_per_second": 62392.13
    },
    {
      "epoch": 0.521591218662999,
      "grad_norm": 14.75,
      "learning_rate": 9.787185033233242e-06,
      "loss": 0.5873,
      "num_input_tokens_seen": 2911163136,
      "step": 39250,
      "train_runtime": 46659.6767,
      "train_tokens_per_second": 62391.413
    },
    {
      "epoch": 0.5217241081454608,
      "grad_norm": 19.375,
      "learning_rate": 9.782882063535579e-06,
      "loss": 0.7559,
      "num_input_tokens_seen": 2913341440,
      "step": 39260,
      "train_runtime": 46694.4273,
      "train_tokens_per_second": 62391.63
    },
    {
      "epoch": 0.5218569976279227,
      "grad_norm": 29.375,
      "learning_rate": 9.778579134057071e-06,
      "loss": 0.9285,
      "num_input_tokens_seen": 2914205952,
      "step": 39270,
      "train_runtime": 46707.6555,
      "train_tokens_per_second": 62392.469
    },
    {
      "epoch": 0.5219898871103846,
      "grad_norm": 24.625,
      "learning_rate": 9.77427624559481e-06,
      "loss": 0.7679,
      "num_input_tokens_seen": 2914705920,
      "step": 39280,
      "train_runtime": 46715.36,
      "train_tokens_per_second": 62392.882
    },
    {
      "epoch": 0.5221227765928466,
      "grad_norm": 16.75,
      "learning_rate": 9.76997339894586e-06,
      "loss": 0.6535,
      "num_input_tokens_seen": 2914937088,
      "step": 39290,
      "train_runtime": 46718.8619,
      "train_tokens_per_second": 62393.153
    },
    {
      "epoch": 0.5222556660753085,
      "grad_norm": 20.5,
      "learning_rate": 9.765670594907287e-06,
      "loss": 0.6001,
      "num_input_tokens_seen": 2915052032,
      "step": 39300,
      "train_runtime": 46721.2505,
      "train_tokens_per_second": 62392.423
    },
    {
      "epoch": 0.5223885555577704,
      "grad_norm": 34.25,
      "learning_rate": 9.761367834276149e-06,
      "loss": 0.8191,
      "num_input_tokens_seen": 2917127168,
      "step": 39310,
      "train_runtime": 46754.0547,
      "train_tokens_per_second": 62393.031
    },
    {
      "epoch": 0.5225214450402323,
      "grad_norm": 39.25,
      "learning_rate": 9.757065117849486e-06,
      "loss": 0.8649,
      "num_input_tokens_seen": 2917989632,
      "step": 39320,
      "train_runtime": 46766.8973,
      "train_tokens_per_second": 62394.339
    },
    {
      "epoch": 0.5226543345226942,
      "grad_norm": 22.5,
      "learning_rate": 9.75276244642435e-06,
      "loss": 0.7735,
      "num_input_tokens_seen": 2918461440,
      "step": 39330,
      "train_runtime": 46773.6766,
      "train_tokens_per_second": 62395.382
    },
    {
      "epoch": 0.5227872240051561,
      "grad_norm": 25.25,
      "learning_rate": 9.748459820797761e-06,
      "loss": 0.6487,
      "num_input_tokens_seen": 2918696448,
      "step": 39340,
      "train_runtime": 46777.215,
      "train_tokens_per_second": 62395.687
    },
    {
      "epoch": 0.522920113487618,
      "grad_norm": 16.375,
      "learning_rate": 9.74415724176675e-06,
      "loss": 0.5684,
      "num_input_tokens_seen": 2918816512,
      "step": 39350,
      "train_runtime": 46780.2392,
      "train_tokens_per_second": 62394.22
    },
    {
      "epoch": 0.52305300297008,
      "grad_norm": 16.625,
      "learning_rate": 9.739854710128325e-06,
      "loss": 0.8106,
      "num_input_tokens_seen": 2920987392,
      "step": 39360,
      "train_runtime": 46813.2561,
      "train_tokens_per_second": 62396.587
    },
    {
      "epoch": 0.5231858924525419,
      "grad_norm": 20.625,
      "learning_rate": 9.7355522266795e-06,
      "loss": 0.8189,
      "num_input_tokens_seen": 2921852672,
      "step": 39370,
      "train_runtime": 46826.2207,
      "train_tokens_per_second": 62397.79
    },
    {
      "epoch": 0.5233187819350038,
      "grad_norm": 40.25,
      "learning_rate": 9.731249792217263e-06,
      "loss": 0.7293,
      "num_input_tokens_seen": 2922316544,
      "step": 39380,
      "train_runtime": 46832.9303,
      "train_tokens_per_second": 62398.755
    },
    {
      "epoch": 0.5234516714174656,
      "grad_norm": 13.8125,
      "learning_rate": 9.72694740753861e-06,
      "loss": 0.6871,
      "num_input_tokens_seen": 2922543104,
      "step": 39390,
      "train_runtime": 46836.3208,
      "train_tokens_per_second": 62399.075
    },
    {
      "epoch": 0.5235845608999276,
      "grad_norm": 15.8125,
      "learning_rate": 9.722645073440515e-06,
      "loss": 0.5288,
      "num_input_tokens_seen": 2922654464,
      "step": 39400,
      "train_runtime": 46839.2903,
      "train_tokens_per_second": 62397.497
    },
    {
      "epoch": 0.5237174503823895,
      "grad_norm": 22.375,
      "learning_rate": 9.718342790719945e-06,
      "loss": 0.7524,
      "num_input_tokens_seen": 2924683776,
      "step": 39410,
      "train_runtime": 46873.2222,
      "train_tokens_per_second": 62395.62
    },
    {
      "epoch": 0.5238503398648514,
      "grad_norm": 94.0,
      "learning_rate": 9.714040560173862e-06,
      "loss": 0.9507,
      "num_input_tokens_seen": 2925578752,
      "step": 39420,
      "train_runtime": 46886.2504,
      "train_tokens_per_second": 62397.371
    },
    {
      "epoch": 0.5239832293473133,
      "grad_norm": 18.0,
      "learning_rate": 9.709738382599212e-06,
      "loss": 0.7043,
      "num_input_tokens_seen": 2926043392,
      "step": 39430,
      "train_runtime": 46892.9001,
      "train_tokens_per_second": 62398.431
    },
    {
      "epoch": 0.5241161188297753,
      "grad_norm": 19.875,
      "learning_rate": 9.705436258792943e-06,
      "loss": 0.6878,
      "num_input_tokens_seen": 2926271744,
      "step": 39440,
      "train_runtime": 46897.0201,
      "train_tokens_per_second": 62397.818
    },
    {
      "epoch": 0.5242490083122371,
      "grad_norm": 24.0,
      "learning_rate": 9.701134189551978e-06,
      "loss": 0.5552,
      "num_input_tokens_seen": 2926387968,
      "step": 39450,
      "train_runtime": 46899.4665,
      "train_tokens_per_second": 62397.042
    },
    {
      "epoch": 0.524381897794699,
      "grad_norm": 17.75,
      "learning_rate": 9.696832175673242e-06,
      "loss": 0.8342,
      "num_input_tokens_seen": 2928459264,
      "step": 39460,
      "train_runtime": 46932.4347,
      "train_tokens_per_second": 62397.344
    },
    {
      "epoch": 0.5245147872771609,
      "grad_norm": 16.375,
      "learning_rate": 9.692530217953642e-06,
      "loss": 0.8955,
      "num_input_tokens_seen": 2929280256,
      "step": 39470,
      "train_runtime": 46944.9262,
      "train_tokens_per_second": 62398.229
    },
    {
      "epoch": 0.5246476767596229,
      "grad_norm": 154.0,
      "learning_rate": 9.688228317190074e-06,
      "loss": 0.7487,
      "num_input_tokens_seen": 2929735680,
      "step": 39480,
      "train_runtime": 46951.4623,
      "train_tokens_per_second": 62399.242
    },
    {
      "epoch": 0.5247805662420848,
      "grad_norm": 19.625,
      "learning_rate": 9.683926474179432e-06,
      "loss": 0.6087,
      "num_input_tokens_seen": 2929961728,
      "step": 39490,
      "train_runtime": 46954.8929,
      "train_tokens_per_second": 62399.498
    },
    {
      "epoch": 0.5249134557245467,
      "grad_norm": 24.75,
      "learning_rate": 9.679624689718594e-06,
      "loss": 0.4691,
      "num_input_tokens_seen": 2930079232,
      "step": 39500,
      "train_runtime": 46957.2748,
      "train_tokens_per_second": 62398.835
    },
    {
      "epoch": 0.5250463452070085,
      "grad_norm": 36.75,
      "learning_rate": 9.675322964604424e-06,
      "loss": 0.8925,
      "num_input_tokens_seen": 2932202240,
      "step": 39510,
      "train_runtime": 46990.6889,
      "train_tokens_per_second": 62399.644
    },
    {
      "epoch": 0.5251792346894705,
      "grad_norm": 20.875,
      "learning_rate": 9.671021299633777e-06,
      "loss": 0.8427,
      "num_input_tokens_seen": 2933047296,
      "step": 39520,
      "train_runtime": 47003.3742,
      "train_tokens_per_second": 62400.782
    },
    {
      "epoch": 0.5253121241719324,
      "grad_norm": 44.5,
      "learning_rate": 9.666719695603504e-06,
      "loss": 0.7847,
      "num_input_tokens_seen": 2933509376,
      "step": 39530,
      "train_runtime": 47010.1274,
      "train_tokens_per_second": 62401.647
    },
    {
      "epoch": 0.5254450136543943,
      "grad_norm": 15.9375,
      "learning_rate": 9.662418153310433e-06,
      "loss": 0.6112,
      "num_input_tokens_seen": 2933739776,
      "step": 39540,
      "train_runtime": 47013.6261,
      "train_tokens_per_second": 62401.904
    },
    {
      "epoch": 0.5255779031368563,
      "grad_norm": 15.0625,
      "learning_rate": 9.658116673551393e-06,
      "loss": 0.5821,
      "num_input_tokens_seen": 2933856000,
      "step": 39550,
      "train_runtime": 47016.0557,
      "train_tokens_per_second": 62401.151
    },
    {
      "epoch": 0.5257107926193182,
      "grad_norm": 42.75,
      "learning_rate": 9.653815257123189e-06,
      "loss": 0.7527,
      "num_input_tokens_seen": 2935868672,
      "step": 39560,
      "train_runtime": 47050.0922,
      "train_tokens_per_second": 62398.787
    },
    {
      "epoch": 0.52584368210178,
      "grad_norm": 15.1875,
      "learning_rate": 9.649513904822622e-06,
      "loss": 0.9042,
      "num_input_tokens_seen": 2936720128,
      "step": 39570,
      "train_runtime": 47063.1399,
      "train_tokens_per_second": 62399.579
    },
    {
      "epoch": 0.5259765715842419,
      "grad_norm": 22.75,
      "learning_rate": 9.64521261744648e-06,
      "loss": 0.7547,
      "num_input_tokens_seen": 2937179392,
      "step": 39580,
      "train_runtime": 47069.963,
      "train_tokens_per_second": 62400.291
    },
    {
      "epoch": 0.5261094610667039,
      "grad_norm": 23.0,
      "learning_rate": 9.640911395791531e-06,
      "loss": 0.6242,
      "num_input_tokens_seen": 2937411072,
      "step": 39590,
      "train_runtime": 47073.463,
      "train_tokens_per_second": 62400.573
    },
    {
      "epoch": 0.5262423505491658,
      "grad_norm": 35.75,
      "learning_rate": 9.63661024065455e-06,
      "loss": 0.5201,
      "num_input_tokens_seen": 2937527808,
      "step": 39600,
      "train_runtime": 47075.8893,
      "train_tokens_per_second": 62399.837
    },
    {
      "epoch": 0.5263752400316277,
      "grad_norm": 15.1875,
      "learning_rate": 9.63230915283228e-06,
      "loss": 0.7434,
      "num_input_tokens_seen": 2939692288,
      "step": 39610,
      "train_runtime": 47111.2315,
      "train_tokens_per_second": 62398.969
    },
    {
      "epoch": 0.5265081295140897,
      "grad_norm": 18.875,
      "learning_rate": 9.628008133121462e-06,
      "loss": 0.8515,
      "num_input_tokens_seen": 2940561152,
      "step": 39620,
      "train_runtime": 47124.3894,
      "train_tokens_per_second": 62399.984
    },
    {
      "epoch": 0.5266410189965515,
      "grad_norm": 20.125,
      "learning_rate": 9.62370718231882e-06,
      "loss": 0.7354,
      "num_input_tokens_seen": 2941054208,
      "step": 39630,
      "train_runtime": 47131.3015,
      "train_tokens_per_second": 62401.294
    },
    {
      "epoch": 0.5267739084790134,
      "grad_norm": 7.53125,
      "learning_rate": 9.619406301221063e-06,
      "loss": 0.6942,
      "num_input_tokens_seen": 2941282304,
      "step": 39640,
      "train_runtime": 47134.791,
      "train_tokens_per_second": 62401.514
    },
    {
      "epoch": 0.5269067979614753,
      "grad_norm": 6.46875,
      "learning_rate": 9.615105490624901e-06,
      "loss": 0.4875,
      "num_input_tokens_seen": 2941401088,
      "step": 39650,
      "train_runtime": 47137.2047,
      "train_tokens_per_second": 62400.838
    },
    {
      "epoch": 0.5270396874439373,
      "grad_norm": 19.75,
      "learning_rate": 9.61080475132701e-06,
      "loss": 0.7971,
      "num_input_tokens_seen": 2943512832,
      "step": 39660,
      "train_runtime": 47172.1466,
      "train_tokens_per_second": 62399.383
    },
    {
      "epoch": 0.5271725769263992,
      "grad_norm": 23.0,
      "learning_rate": 9.606504084124073e-06,
      "loss": 0.8961,
      "num_input_tokens_seen": 2944345600,
      "step": 39670,
      "train_runtime": 47184.6542,
      "train_tokens_per_second": 62400.491
    },
    {
      "epoch": 0.5273054664088611,
      "grad_norm": 15.5,
      "learning_rate": 9.60220348981274e-06,
      "loss": 0.7047,
      "num_input_tokens_seen": 2944788480,
      "step": 39680,
      "train_runtime": 47191.1521,
      "train_tokens_per_second": 62401.284
    },
    {
      "epoch": 0.5274383558913229,
      "grad_norm": 21.375,
      "learning_rate": 9.597902969189668e-06,
      "loss": 0.6432,
      "num_input_tokens_seen": 2945017856,
      "step": 39690,
      "train_runtime": 47194.6433,
      "train_tokens_per_second": 62401.528
    },
    {
      "epoch": 0.5275712453737849,
      "grad_norm": 18.125,
      "learning_rate": 9.593602523051484e-06,
      "loss": 0.5072,
      "num_input_tokens_seen": 2945133312,
      "step": 39700,
      "train_runtime": 47197.0119,
      "train_tokens_per_second": 62400.843
    },
    {
      "epoch": 0.5277041348562468,
      "grad_norm": 17.875,
      "learning_rate": 9.58930215219481e-06,
      "loss": 0.8449,
      "num_input_tokens_seen": 2947250176,
      "step": 39710,
      "train_runtime": 47231.6956,
      "train_tokens_per_second": 62399.838
    },
    {
      "epoch": 0.5278370243387087,
      "grad_norm": 151.0,
      "learning_rate": 9.585001857416251e-06,
      "loss": 0.8784,
      "num_input_tokens_seen": 2948084736,
      "step": 39720,
      "train_runtime": 47244.5284,
      "train_tokens_per_second": 62400.554
    },
    {
      "epoch": 0.5279699138211706,
      "grad_norm": 19.625,
      "learning_rate": 9.580701639512393e-06,
      "loss": 0.7522,
      "num_input_tokens_seen": 2948526336,
      "step": 39730,
      "train_runtime": 47251.0631,
      "train_tokens_per_second": 62401.27
    },
    {
      "epoch": 0.5281028033036326,
      "grad_norm": 13.5625,
      "learning_rate": 9.576401499279822e-06,
      "loss": 0.6452,
      "num_input_tokens_seen": 2948750848,
      "step": 39740,
      "train_runtime": 47254.4471,
      "train_tokens_per_second": 62401.552
    },
    {
      "epoch": 0.5282356927860944,
      "grad_norm": 14.0625,
      "learning_rate": 9.57210143751509e-06,
      "loss": 0.4988,
      "num_input_tokens_seen": 2948861184,
      "step": 39750,
      "train_runtime": 47256.8323,
      "train_tokens_per_second": 62400.737
    },
    {
      "epoch": 0.5283685822685563,
      "grad_norm": 20.375,
      "learning_rate": 9.567801455014759e-06,
      "loss": 0.7872,
      "num_input_tokens_seen": 2950908928,
      "step": 39760,
      "train_runtime": 47290.7459,
      "train_tokens_per_second": 62399.289
    },
    {
      "epoch": 0.5285014717510182,
      "grad_norm": 14.3125,
      "learning_rate": 9.563501552575349e-06,
      "loss": 0.886,
      "num_input_tokens_seen": 2951752704,
      "step": 39770,
      "train_runtime": 47304.1746,
      "train_tokens_per_second": 62399.413
    },
    {
      "epoch": 0.5286343612334802,
      "grad_norm": 27.875,
      "learning_rate": 9.559201730993387e-06,
      "loss": 0.7382,
      "num_input_tokens_seen": 2952239872,
      "step": 39780,
      "train_runtime": 47310.9817,
      "train_tokens_per_second": 62400.732
    },
    {
      "epoch": 0.5287672507159421,
      "grad_norm": 13.75,
      "learning_rate": 9.554901991065376e-06,
      "loss": 0.6757,
      "num_input_tokens_seen": 2952477696,
      "step": 39790,
      "train_runtime": 47314.5255,
      "train_tokens_per_second": 62401.084
    },
    {
      "epoch": 0.528900140198404,
      "grad_norm": 55.25,
      "learning_rate": 9.550602333587796e-06,
      "loss": 0.6202,
      "num_input_tokens_seen": 2952594432,
      "step": 39800,
      "train_runtime": 47316.9643,
      "train_tokens_per_second": 62400.335
    },
    {
      "epoch": 0.529033029680866,
      "grad_norm": 15.3125,
      "learning_rate": 9.54630275935713e-06,
      "loss": 0.8258,
      "num_input_tokens_seen": 2954644224,
      "step": 39810,
      "train_runtime": 47350.0551,
      "train_tokens_per_second": 62400.017
    },
    {
      "epoch": 0.5291659191633278,
      "grad_norm": 20.0,
      "learning_rate": 9.542003269169831e-06,
      "loss": 0.9244,
      "num_input_tokens_seen": 2955478528,
      "step": 39820,
      "train_runtime": 47362.6252,
      "train_tokens_per_second": 62401.071
    },
    {
      "epoch": 0.5292988086457897,
      "grad_norm": 17.625,
      "learning_rate": 9.537703863822346e-06,
      "loss": 0.7184,
      "num_input_tokens_seen": 2955958784,
      "step": 39830,
      "train_runtime": 47370.0545,
      "train_tokens_per_second": 62401.423
    },
    {
      "epoch": 0.5294316981282516,
      "grad_norm": 18.25,
      "learning_rate": 9.533404544111092e-06,
      "loss": 0.5943,
      "num_input_tokens_seen": 2956184320,
      "step": 39840,
      "train_runtime": 47373.4865,
      "train_tokens_per_second": 62401.663
    },
    {
      "epoch": 0.5295645876107136,
      "grad_norm": 17.75,
      "learning_rate": 9.529105310832491e-06,
      "loss": 0.5826,
      "num_input_tokens_seen": 2956295680,
      "step": 39850,
      "train_runtime": 47375.9127,
      "train_tokens_per_second": 62400.817
    },
    {
      "epoch": 0.5296974770931755,
      "grad_norm": 24.25,
      "learning_rate": 9.52480616478293e-06,
      "loss": 0.8043,
      "num_input_tokens_seen": 2958257664,
      "step": 39860,
      "train_runtime": 47407.6982,
      "train_tokens_per_second": 62400.365
    },
    {
      "epoch": 0.5298303665756374,
      "grad_norm": 21.0,
      "learning_rate": 9.520507106758791e-06,
      "loss": 0.9587,
      "num_input_tokens_seen": 2959071488,
      "step": 39870,
      "train_runtime": 47420.7927,
      "train_tokens_per_second": 62400.296
    },
    {
      "epoch": 0.5299632560580992,
      "grad_norm": 16.25,
      "learning_rate": 9.516208137556437e-06,
      "loss": 0.7204,
      "num_input_tokens_seen": 2959518720,
      "step": 39880,
      "train_runtime": 47427.297,
      "train_tokens_per_second": 62401.168
    },
    {
      "epoch": 0.5300961455405612,
      "grad_norm": 17.75,
      "learning_rate": 9.511909257972204e-06,
      "loss": 0.6085,
      "num_input_tokens_seen": 2959747584,
      "step": 39890,
      "train_runtime": 47430.6925,
      "train_tokens_per_second": 62401.526
    },
    {
      "epoch": 0.5302290350230231,
      "grad_norm": 16.125,
      "learning_rate": 9.507610468802435e-06,
      "loss": 0.5442,
      "num_input_tokens_seen": 2959866112,
      "step": 39900,
      "train_runtime": 47433.1649,
      "train_tokens_per_second": 62400.772
    },
    {
      "epoch": 0.530361924505485,
      "grad_norm": 12.5,
      "learning_rate": 9.503311770843432e-06,
      "loss": 0.852,
      "num_input_tokens_seen": 2961827840,
      "step": 39910,
      "train_runtime": 47464.5318,
      "train_tokens_per_second": 62400.865
    },
    {
      "epoch": 0.530494813987947,
      "grad_norm": 32.25,
      "learning_rate": 9.4990131648915e-06,
      "loss": 0.8848,
      "num_input_tokens_seen": 2962650880,
      "step": 39920,
      "train_runtime": 47477.2322,
      "train_tokens_per_second": 62401.508
    },
    {
      "epoch": 0.5306277034704089,
      "grad_norm": 17.75,
      "learning_rate": 9.494714651742909e-06,
      "loss": 0.7368,
      "num_input_tokens_seen": 2963173632,
      "step": 39930,
      "train_runtime": 47484.2455,
      "train_tokens_per_second": 62403.3
    },
    {
      "epoch": 0.5307605929528707,
      "grad_norm": 462.0,
      "learning_rate": 9.490416232193923e-06,
      "loss": 0.5943,
      "num_input_tokens_seen": 2963410944,
      "step": 39940,
      "train_runtime": 47488.4659,
      "train_tokens_per_second": 62402.752
    },
    {
      "epoch": 0.5308934824353326,
      "grad_norm": 7.375,
      "learning_rate": 9.48611790704079e-06,
      "loss": 0.5375,
      "num_input_tokens_seen": 2963529216,
      "step": 39950,
      "train_runtime": 47490.9651,
      "train_tokens_per_second": 62401.958
    },
    {
      "epoch": 0.5310263719177946,
      "grad_norm": 32.5,
      "learning_rate": 9.481819677079725e-06,
      "loss": 0.8465,
      "num_input_tokens_seen": 2965580288,
      "step": 39960,
      "train_runtime": 47525.0619,
      "train_tokens_per_second": 62400.346
    },
    {
      "epoch": 0.5311592614002565,
      "grad_norm": 59.0,
      "learning_rate": 9.47752154310695e-06,
      "loss": 0.897,
      "num_input_tokens_seen": 2966435840,
      "step": 39970,
      "train_runtime": 47537.778,
      "train_tokens_per_second": 62401.651
    },
    {
      "epoch": 0.5312921508827184,
      "grad_norm": 20.875,
      "learning_rate": 9.473223505918648e-06,
      "loss": 0.7053,
      "num_input_tokens_seen": 2966880768,
      "step": 39980,
      "train_runtime": 47544.3735,
      "train_tokens_per_second": 62402.353
    },
    {
      "epoch": 0.5314250403651803,
      "grad_norm": 29.375,
      "learning_rate": 9.468925566310997e-06,
      "loss": 0.6277,
      "num_input_tokens_seen": 2967109632,
      "step": 39990,
      "train_runtime": 47548.4808,
      "train_tokens_per_second": 62401.776
    },
    {
      "epoch": 0.5315579298476422,
      "grad_norm": 12.8125,
      "learning_rate": 9.464627725080143e-06,
      "loss": 0.5272,
      "num_input_tokens_seen": 2967226112,
      "step": 40000,
      "train_runtime": 47550.8642,
      "train_tokens_per_second": 62401.097
    },
    {
      "epoch": 0.5315579298476422,
      "eval_accuracy": 0.8185,
      "eval_balanced_accuracy": 0.8182007367751296,
      "eval_invalid_f1": 0.8343826991513824,
      "eval_invalid_precision": 0.8483948784561143,
      "eval_invalid_recall": 0.8208258527827648,
      "eval_loss": 0.3613389730453491,
      "eval_macro_f1": 0.8168152849810668,
      "eval_mcc": 0.6341748054553771,
      "eval_runtime": 78.1557,
      "eval_samples_per_second": 127.95,
      "eval_steps_per_second": 31.987,
      "eval_valid_f1": 0.7992478708107511,
      "eval_valid_precision": 0.7835610496638473,
      "eval_valid_recall": 0.8155756207674943,
      "num_input_tokens_seen": 2967226112,
      "step": 40000
    },
    {
      "epoch": 0.5316908193301041,
      "grad_norm": 22.0,
      "learning_rate": 9.460329983022236e-06,
      "loss": 0.7926,
      "num_input_tokens_seen": 2969357824,
      "step": 40010,
      "train_runtime": 47667.1308,
      "train_tokens_per_second": 62293.614
    },
    {
      "epoch": 0.531823708812566,
      "grad_norm": 12.3125,
      "learning_rate": 9.456032340933383e-06,
      "loss": 0.9528,
      "num_input_tokens_seen": 2970211584,
      "step": 40020,
      "train_runtime": 47679.8015,
      "train_tokens_per_second": 62294.965
    },
    {
      "epoch": 0.531956598295028,
      "grad_norm": 20.5,
      "learning_rate": 9.451734799609693e-06,
      "loss": 0.7404,
      "num_input_tokens_seen": 2970670592,
      "step": 40030,
      "train_runtime": 47686.5211,
      "train_tokens_per_second": 62295.813
    },
    {
      "epoch": 0.5320894877774899,
      "grad_norm": 27.25,
      "learning_rate": 9.447437359847241e-06,
      "loss": 0.6595,
      "num_input_tokens_seen": 2970899712,
      "step": 40040,
      "train_runtime": 47689.968,
      "train_tokens_per_second": 62296.115
    },
    {
      "epoch": 0.5322223772599518,
      "grad_norm": 13.875,
      "learning_rate": 9.443140022442086e-06,
      "loss": 0.5242,
      "num_input_tokens_seen": 2971017984,
      "step": 40050,
      "train_runtime": 47693.0487,
      "train_tokens_per_second": 62294.57
    },
    {
      "epoch": 0.5323552667424136,
      "grad_norm": 18.375,
      "learning_rate": 9.438842788190282e-06,
      "loss": 0.8574,
      "num_input_tokens_seen": 2973191936,
      "step": 40060,
      "train_runtime": 47726.7831,
      "train_tokens_per_second": 62296.089
    },
    {
      "epoch": 0.5324881562248756,
      "grad_norm": 19.375,
      "learning_rate": 9.434545657887843e-06,
      "loss": 0.8856,
      "num_input_tokens_seen": 2974035456,
      "step": 40070,
      "train_runtime": 47739.4271,
      "train_tokens_per_second": 62297.259
    },
    {
      "epoch": 0.5326210457073375,
      "grad_norm": 15.5625,
      "learning_rate": 9.430248632330784e-06,
      "loss": 0.7556,
      "num_input_tokens_seen": 2974510848,
      "step": 40080,
      "train_runtime": 47746.2605,
      "train_tokens_per_second": 62298.3
    },
    {
      "epoch": 0.5327539351897994,
      "grad_norm": 10.0625,
      "learning_rate": 9.425951712315083e-06,
      "loss": 0.6563,
      "num_input_tokens_seen": 2974749184,
      "step": 40090,
      "train_runtime": 47750.4295,
      "train_tokens_per_second": 62297.852
    },
    {
      "epoch": 0.5328868246722613,
      "grad_norm": 13.25,
      "learning_rate": 9.42165489863671e-06,
      "loss": 0.5953,
      "num_input_tokens_seen": 2974866432,
      "step": 40100,
      "train_runtime": 47752.8405,
      "train_tokens_per_second": 62297.162
    },
    {
      "epoch": 0.5330197141547233,
      "grad_norm": 38.5,
      "learning_rate": 9.41735819209161e-06,
      "loss": 0.7833,
      "num_input_tokens_seen": 2976855808,
      "step": 40110,
      "train_runtime": 47786.1246,
      "train_tokens_per_second": 62295.401
    },
    {
      "epoch": 0.5331526036371851,
      "grad_norm": 27.625,
      "learning_rate": 9.413061593475706e-06,
      "loss": 0.9703,
      "num_input_tokens_seen": 2977705984,
      "step": 40120,
      "train_runtime": 47798.7996,
      "train_tokens_per_second": 62296.669
    },
    {
      "epoch": 0.533285493119647,
      "grad_norm": 24.0,
      "learning_rate": 9.40876510358491e-06,
      "loss": 0.7132,
      "num_input_tokens_seen": 2978177024,
      "step": 40130,
      "train_runtime": 47805.6033,
      "train_tokens_per_second": 62297.656
    },
    {
      "epoch": 0.5334183826021089,
      "grad_norm": 29.625,
      "learning_rate": 9.40446872321511e-06,
      "loss": 0.648,
      "num_input_tokens_seen": 2978415104,
      "step": 40140,
      "train_runtime": 47809.1411,
      "train_tokens_per_second": 62298.026
    },
    {
      "epoch": 0.5335512720845709,
      "grad_norm": 31.625,
      "learning_rate": 9.400172453162166e-06,
      "loss": 0.6015,
      "num_input_tokens_seen": 2978531840,
      "step": 40150,
      "train_runtime": 47811.5695,
      "train_tokens_per_second": 62297.303
    },
    {
      "epoch": 0.5336841615670328,
      "grad_norm": 24.5,
      "learning_rate": 9.395876294221924e-06,
      "loss": 0.7703,
      "num_input_tokens_seen": 2980626176,
      "step": 40160,
      "train_runtime": 47845.1158,
      "train_tokens_per_second": 62297.397
    },
    {
      "epoch": 0.5338170510494947,
      "grad_norm": 16.0,
      "learning_rate": 9.391580247190215e-06,
      "loss": 0.9062,
      "num_input_tokens_seen": 2981449728,
      "step": 40170,
      "train_runtime": 47857.9758,
      "train_tokens_per_second": 62297.865
    },
    {
      "epoch": 0.5339499405319565,
      "grad_norm": 27.5,
      "learning_rate": 9.38728431286284e-06,
      "loss": 0.7246,
      "num_input_tokens_seen": 2981895424,
      "step": 40180,
      "train_runtime": 47864.398,
      "train_tokens_per_second": 62298.818
    },
    {
      "epoch": 0.5340828300144185,
      "grad_norm": 33.5,
      "learning_rate": 9.382988492035581e-06,
      "loss": 0.6608,
      "num_input_tokens_seen": 2982124544,
      "step": 40190,
      "train_runtime": 47867.8625,
      "train_tokens_per_second": 62299.096
    },
    {
      "epoch": 0.5342157194968804,
      "grad_norm": 9.5,
      "learning_rate": 9.378692785504205e-06,
      "loss": 0.5105,
      "num_input_tokens_seen": 2982239488,
      "step": 40200,
      "train_runtime": 47870.278,
      "train_tokens_per_second": 62298.353
    },
    {
      "epoch": 0.5343486089793423,
      "grad_norm": 37.5,
      "learning_rate": 9.374397194064445e-06,
      "loss": 0.8316,
      "num_input_tokens_seen": 2984221440,
      "step": 40210,
      "train_runtime": 47902.7585,
      "train_tokens_per_second": 62297.486
    },
    {
      "epoch": 0.5344814984618043,
      "grad_norm": 18.75,
      "learning_rate": 9.370101718512031e-06,
      "loss": 0.9091,
      "num_input_tokens_seen": 2985063680,
      "step": 40220,
      "train_runtime": 47915.4897,
      "train_tokens_per_second": 62298.511
    },
    {
      "epoch": 0.5346143879442662,
      "grad_norm": 82.5,
      "learning_rate": 9.365806359642653e-06,
      "loss": 0.7621,
      "num_input_tokens_seen": 2985517824,
      "step": 40230,
      "train_runtime": 47922.5823,
      "train_tokens_per_second": 62298.768
    },
    {
      "epoch": 0.534747277426728,
      "grad_norm": 28.125,
      "learning_rate": 9.361511118251998e-06,
      "loss": 0.5849,
      "num_input_tokens_seen": 2985751552,
      "step": 40240,
      "train_runtime": 47926.0288,
      "train_tokens_per_second": 62299.164
    },
    {
      "epoch": 0.5348801669091899,
      "grad_norm": 11.1875,
      "learning_rate": 9.35721599513571e-06,
      "loss": 0.5349,
      "num_input_tokens_seen": 2985871104,
      "step": 40250,
      "train_runtime": 47928.4246,
      "train_tokens_per_second": 62298.545
    },
    {
      "epoch": 0.5350130563916519,
      "grad_norm": 17.25,
      "learning_rate": 9.35292099108943e-06,
      "loss": 0.796,
      "num_input_tokens_seen": 2987894272,
      "step": 40260,
      "train_runtime": 47961.9359,
      "train_tokens_per_second": 62297.199
    },
    {
      "epoch": 0.5351459458741138,
      "grad_norm": 24.375,
      "learning_rate": 9.348626106908767e-06,
      "loss": 0.9324,
      "num_input_tokens_seen": 2988750848,
      "step": 40270,
      "train_runtime": 47974.9837,
      "train_tokens_per_second": 62298.111
    },
    {
      "epoch": 0.5352788353565757,
      "grad_norm": 11.9375,
      "learning_rate": 9.344331343389304e-06,
      "loss": 0.7082,
      "num_input_tokens_seen": 2989248768,
      "step": 40280,
      "train_runtime": 47982.4815,
      "train_tokens_per_second": 62298.753
    },
    {
      "epoch": 0.5354117248390377,
      "grad_norm": 11.625,
      "learning_rate": 9.340036701326618e-06,
      "loss": 0.6629,
      "num_input_tokens_seen": 2989487872,
      "step": 40290,
      "train_runtime": 47986.0576,
      "train_tokens_per_second": 62299.093
    },
    {
      "epoch": 0.5355446143214996,
      "grad_norm": 632.0,
      "learning_rate": 9.335742181516246e-06,
      "loss": 0.488,
      "num_input_tokens_seen": 2989605632,
      "step": 40300,
      "train_runtime": 47988.4714,
      "train_tokens_per_second": 62298.413
    },
    {
      "epoch": 0.5356775038039614,
      "grad_norm": 28.75,
      "learning_rate": 9.331447784753715e-06,
      "loss": 0.8722,
      "num_input_tokens_seen": 2991656448,
      "step": 40310,
      "train_runtime": 48020.9947,
      "train_tokens_per_second": 62298.927
    },
    {
      "epoch": 0.5358103932864233,
      "grad_norm": 15.125,
      "learning_rate": 9.327153511834516e-06,
      "loss": 0.8869,
      "num_input_tokens_seen": 2992474112,
      "step": 40320,
      "train_runtime": 48034.0714,
      "train_tokens_per_second": 62298.99
    },
    {
      "epoch": 0.5359432827688853,
      "grad_norm": 17.125,
      "learning_rate": 9.322859363554133e-06,
      "loss": 0.7463,
      "num_input_tokens_seen": 2992924416,
      "step": 40330,
      "train_runtime": 48040.3207,
      "train_tokens_per_second": 62300.259
    },
    {
      "epoch": 0.5360761722513472,
      "grad_norm": 13.25,
      "learning_rate": 9.318565340708014e-06,
      "loss": 0.6448,
      "num_input_tokens_seen": 2993145088,
      "step": 40340,
      "train_runtime": 48043.6634,
      "train_tokens_per_second": 62300.517
    },
    {
      "epoch": 0.5362090617338091,
      "grad_norm": 6.71875,
      "learning_rate": 9.314271444091589e-06,
      "loss": 0.5122,
      "num_input_tokens_seen": 2993256192,
      "step": 40350,
      "train_runtime": 48045.9883,
      "train_tokens_per_second": 62299.815
    },
    {
      "epoch": 0.536341951216271,
      "grad_norm": 13.8125,
      "learning_rate": 9.309977674500267e-06,
      "loss": 0.7883,
      "num_input_tokens_seen": 2995350016,
      "step": 40360,
      "train_runtime": 48080.8349,
      "train_tokens_per_second": 62298.211
    },
    {
      "epoch": 0.5364748406987329,
      "grad_norm": 19.625,
      "learning_rate": 9.305684032729423e-06,
      "loss": 0.8506,
      "num_input_tokens_seen": 2996188672,
      "step": 40370,
      "train_runtime": 48093.2766,
      "train_tokens_per_second": 62299.533
    },
    {
      "epoch": 0.5366077301811948,
      "grad_norm": 16.0,
      "learning_rate": 9.301390519574424e-06,
      "loss": 0.7341,
      "num_input_tokens_seen": 2996669440,
      "step": 40380,
      "train_runtime": 48099.9946,
      "train_tokens_per_second": 62300.827
    },
    {
      "epoch": 0.5367406196636567,
      "grad_norm": 19.875,
      "learning_rate": 9.297097135830601e-06,
      "loss": 0.5891,
      "num_input_tokens_seen": 2996891648,
      "step": 40390,
      "train_runtime": 48104.0801,
      "train_tokens_per_second": 62300.155
    },
    {
      "epoch": 0.5368735091461186,
      "grad_norm": 14.5625,
      "learning_rate": 9.292803882293267e-06,
      "loss": 0.4539,
      "num_input_tokens_seen": 2997006848,
      "step": 40400,
      "train_runtime": 48106.4803,
      "train_tokens_per_second": 62299.441
    },
    {
      "epoch": 0.5370063986285806,
      "grad_norm": 18.875,
      "learning_rate": 9.288510759757705e-06,
      "loss": 0.8372,
      "num_input_tokens_seen": 2999055360,
      "step": 40410,
      "train_runtime": 48139.4816,
      "train_tokens_per_second": 62299.287
    },
    {
      "epoch": 0.5371392881110425,
      "grad_norm": 18.125,
      "learning_rate": 9.284217769019183e-06,
      "loss": 0.8405,
      "num_input_tokens_seen": 2999891456,
      "step": 40420,
      "train_runtime": 48151.8356,
      "train_tokens_per_second": 62300.667
    },
    {
      "epoch": 0.5372721775935043,
      "grad_norm": 17.875,
      "learning_rate": 9.279924910872938e-06,
      "loss": 0.6953,
      "num_input_tokens_seen": 3000352000,
      "step": 40430,
      "train_runtime": 48158.3355,
      "train_tokens_per_second": 62301.821
    },
    {
      "epoch": 0.5374050670759662,
      "grad_norm": 18.875,
      "learning_rate": 9.275632186114178e-06,
      "loss": 0.6578,
      "num_input_tokens_seen": 3000564480,
      "step": 40440,
      "train_runtime": 48162.3246,
      "train_tokens_per_second": 62301.073
    },
    {
      "epoch": 0.5375379565584282,
      "grad_norm": 7.59375,
      "learning_rate": 9.271339595538102e-06,
      "loss": 0.5299,
      "num_input_tokens_seen": 3000676096,
      "step": 40450,
      "train_runtime": 48164.6722,
      "train_tokens_per_second": 62300.353
    },
    {
      "epoch": 0.5376708460408901,
      "grad_norm": 17.375,
      "learning_rate": 9.267047139939865e-06,
      "loss": 0.8158,
      "num_input_tokens_seen": 3002691072,
      "step": 40460,
      "train_runtime": 48197.635,
      "train_tokens_per_second": 62299.552
    },
    {
      "epoch": 0.537803735523352,
      "grad_norm": 19.125,
      "learning_rate": 9.26275482011461e-06,
      "loss": 0.9587,
      "num_input_tokens_seen": 3003548928,
      "step": 40470,
      "train_runtime": 48210.2797,
      "train_tokens_per_second": 62301.006
    },
    {
      "epoch": 0.537936625005814,
      "grad_norm": 17.625,
      "learning_rate": 9.25846263685745e-06,
      "loss": 0.7993,
      "num_input_tokens_seen": 3004008960,
      "step": 40480,
      "train_runtime": 48216.7826,
      "train_tokens_per_second": 62302.145
    },
    {
      "epoch": 0.5380695144882758,
      "grad_norm": 15.0,
      "learning_rate": 9.254170590963478e-06,
      "loss": 0.6836,
      "num_input_tokens_seen": 3004241152,
      "step": 40490,
      "train_runtime": 48220.8675,
      "train_tokens_per_second": 62301.682
    },
    {
      "epoch": 0.5382024039707377,
      "grad_norm": 27.0,
      "learning_rate": 9.249878683227749e-06,
      "loss": 0.5565,
      "num_input_tokens_seen": 3004356864,
      "step": 40500,
      "train_runtime": 48223.2167,
      "train_tokens_per_second": 62301.046
    },
    {
      "epoch": 0.5383352934531996,
      "grad_norm": 31.125,
      "learning_rate": 9.24558691444531e-06,
      "loss": 0.8188,
      "num_input_tokens_seen": 3006487296,
      "step": 40510,
      "train_runtime": 48257.3874,
      "train_tokens_per_second": 62301.079
    },
    {
      "epoch": 0.5384681829356616,
      "grad_norm": 31.375,
      "learning_rate": 9.241295285411166e-06,
      "loss": 0.8736,
      "num_input_tokens_seen": 3007311616,
      "step": 40520,
      "train_runtime": 48269.6804,
      "train_tokens_per_second": 62302.29
    },
    {
      "epoch": 0.5386010724181235,
      "grad_norm": 30.625,
      "learning_rate": 9.237003796920303e-06,
      "loss": 0.6864,
      "num_input_tokens_seen": 3007744000,
      "step": 40530,
      "train_runtime": 48276.0095,
      "train_tokens_per_second": 62303.078
    },
    {
      "epoch": 0.5387339619005854,
      "grad_norm": 15.625,
      "learning_rate": 9.232712449767685e-06,
      "loss": 0.662,
      "num_input_tokens_seen": 3007981312,
      "step": 40540,
      "train_runtime": 48279.5128,
      "train_tokens_per_second": 62303.473
    },
    {
      "epoch": 0.5388668513830472,
      "grad_norm": 27.125,
      "learning_rate": 9.228421244748244e-06,
      "loss": 0.4929,
      "num_input_tokens_seen": 3008095744,
      "step": 40550,
      "train_runtime": 48282.4727,
      "train_tokens_per_second": 62302.023
    },
    {
      "epoch": 0.5389997408655092,
      "grad_norm": 44.0,
      "learning_rate": 9.224130182656888e-06,
      "loss": 0.8569,
      "num_input_tokens_seen": 3010130944,
      "step": 40560,
      "train_runtime": 48316.1768,
      "train_tokens_per_second": 62300.686
    },
    {
      "epoch": 0.5391326303479711,
      "grad_norm": 11.875,
      "learning_rate": 9.219839264288492e-06,
      "loss": 0.8833,
      "num_input_tokens_seen": 3010983936,
      "step": 40570,
      "train_runtime": 48328.9564,
      "train_tokens_per_second": 62301.861
    },
    {
      "epoch": 0.539265519830433,
      "grad_norm": 20.125,
      "learning_rate": 9.215548490437922e-06,
      "loss": 0.7604,
      "num_input_tokens_seen": 3011469312,
      "step": 40580,
      "train_runtime": 48335.7518,
      "train_tokens_per_second": 62303.144
    },
    {
      "epoch": 0.539398409312895,
      "grad_norm": 13.75,
      "learning_rate": 9.211257861899995e-06,
      "loss": 0.6334,
      "num_input_tokens_seen": 3011708416,
      "step": 40590,
      "train_runtime": 48339.2703,
      "train_tokens_per_second": 62303.556
    },
    {
      "epoch": 0.5395312987953569,
      "grad_norm": 8.75,
      "learning_rate": 9.20696737946952e-06,
      "loss": 0.5285,
      "num_input_tokens_seen": 3011822592,
      "step": 40600,
      "train_runtime": 48341.6718,
      "train_tokens_per_second": 62302.822
    },
    {
      "epoch": 0.5396641882778187,
      "grad_norm": 22.25,
      "learning_rate": 9.202677043941265e-06,
      "loss": 0.8329,
      "num_input_tokens_seen": 3013832960,
      "step": 40610,
      "train_runtime": 48373.7765,
      "train_tokens_per_second": 62303.032
    },
    {
      "epoch": 0.5397970777602806,
      "grad_norm": 41.5,
      "learning_rate": 9.19838685610998e-06,
      "loss": 0.8857,
      "num_input_tokens_seen": 3014653184,
      "step": 40620,
      "train_runtime": 48386.151,
      "train_tokens_per_second": 62304.05
    },
    {
      "epoch": 0.5399299672427426,
      "grad_norm": 20.25,
      "learning_rate": 9.19409681677038e-06,
      "loss": 0.742,
      "num_input_tokens_seen": 3015112192,
      "step": 40630,
      "train_runtime": 48393.3994,
      "train_tokens_per_second": 62304.203
    },
    {
      "epoch": 0.5400628567252045,
      "grad_norm": 31.5,
      "learning_rate": 9.189806926717156e-06,
      "loss": 0.6982,
      "num_input_tokens_seen": 3015343360,
      "step": 40640,
      "train_runtime": 48396.847,
      "train_tokens_per_second": 62304.541
    },
    {
      "epoch": 0.5401957462076664,
      "grad_norm": 8.5625,
      "learning_rate": 9.185517186744978e-06,
      "loss": 0.5307,
      "num_input_tokens_seen": 3015457792,
      "step": 40650,
      "train_runtime": 48399.2128,
      "train_tokens_per_second": 62303.86
    },
    {
      "epoch": 0.5403286356901283,
      "grad_norm": 39.75,
      "learning_rate": 9.181227597648477e-06,
      "loss": 0.9042,
      "num_input_tokens_seen": 3017407744,
      "step": 40660,
      "train_runtime": 48431.2724,
      "train_tokens_per_second": 62302.88
    },
    {
      "epoch": 0.5404615251725902,
      "grad_norm": 21.25,
      "learning_rate": 9.176938160222266e-06,
      "loss": 0.9074,
      "num_input_tokens_seen": 3018243072,
      "step": 40670,
      "train_runtime": 48443.5741,
      "train_tokens_per_second": 62304.302
    },
    {
      "epoch": 0.5405944146550521,
      "grad_norm": 84.5,
      "learning_rate": 9.17264887526092e-06,
      "loss": 0.7336,
      "num_input_tokens_seen": 3018679808,
      "step": 40680,
      "train_runtime": 48449.8084,
      "train_tokens_per_second": 62305.299
    },
    {
      "epoch": 0.540727304137514,
      "grad_norm": 19.375,
      "learning_rate": 9.168359743558989e-06,
      "loss": 0.6585,
      "num_input_tokens_seen": 3018894848,
      "step": 40690,
      "train_runtime": 48453.8293,
      "train_tokens_per_second": 62304.567
    },
    {
      "epoch": 0.540860193619976,
      "grad_norm": 11.1875,
      "learning_rate": 9.164070765911003e-06,
      "loss": 0.5553,
      "num_input_tokens_seen": 3019006720,
      "step": 40700,
      "train_runtime": 48456.1813,
      "train_tokens_per_second": 62303.851
    },
    {
      "epoch": 0.5409930831024379,
      "grad_norm": 24.625,
      "learning_rate": 9.159781943111454e-06,
      "loss": 0.8069,
      "num_input_tokens_seen": 3021071360,
      "step": 40710,
      "train_runtime": 48489.0265,
      "train_tokens_per_second": 62304.228
    },
    {
      "epoch": 0.5411259725848998,
      "grad_norm": 28.0,
      "learning_rate": 9.15549327595481e-06,
      "loss": 0.8442,
      "num_input_tokens_seen": 3021901312,
      "step": 40720,
      "train_runtime": 48501.7558,
      "train_tokens_per_second": 62304.988
    },
    {
      "epoch": 0.5412588620673616,
      "grad_norm": 23.625,
      "learning_rate": 9.1512047652355e-06,
      "loss": 0.7337,
      "num_input_tokens_seen": 3022384640,
      "step": 40730,
      "train_runtime": 48508.4299,
      "train_tokens_per_second": 62306.379
    },
    {
      "epoch": 0.5413917515498236,
      "grad_norm": 19.5,
      "learning_rate": 9.146916411747943e-06,
      "loss": 0.6341,
      "num_input_tokens_seen": 3022616832,
      "step": 40740,
      "train_runtime": 48512.6387,
      "train_tokens_per_second": 62305.76
    },
    {
      "epoch": 0.5415246410322855,
      "grad_norm": 13.25,
      "learning_rate": 9.14262821628651e-06,
      "loss": 0.5262,
      "num_input_tokens_seen": 3022736384,
      "step": 40750,
      "train_runtime": 48515.0598,
      "train_tokens_per_second": 62305.115
    },
    {
      "epoch": 0.5416575305147474,
      "grad_norm": 14.25,
      "learning_rate": 9.138340179645559e-06,
      "loss": 0.8448,
      "num_input_tokens_seen": 3024755200,
      "step": 40760,
      "train_runtime": 48548.0911,
      "train_tokens_per_second": 62304.308
    },
    {
      "epoch": 0.5417904199972093,
      "grad_norm": 16.5,
      "learning_rate": 9.134052302619407e-06,
      "loss": 0.9449,
      "num_input_tokens_seen": 3025611264,
      "step": 40770,
      "train_runtime": 48560.8254,
      "train_tokens_per_second": 62305.598
    },
    {
      "epoch": 0.5419233094796713,
      "grad_norm": 27.625,
      "learning_rate": 9.129764586002343e-06,
      "loss": 0.6903,
      "num_input_tokens_seen": 3026085376,
      "step": 40780,
      "train_runtime": 48567.4089,
      "train_tokens_per_second": 62306.914
    },
    {
      "epoch": 0.5420561989621332,
      "grad_norm": 17.125,
      "learning_rate": 9.125477030588631e-06,
      "loss": 0.6266,
      "num_input_tokens_seen": 3026322176,
      "step": 40790,
      "train_runtime": 48570.8909,
      "train_tokens_per_second": 62307.323
    },
    {
      "epoch": 0.542189088444595,
      "grad_norm": 8.0,
      "learning_rate": 9.121189637172496e-06,
      "loss": 0.537,
      "num_input_tokens_seen": 3026438400,
      "step": 40800,
      "train_runtime": 48573.9499,
      "train_tokens_per_second": 62305.792
    },
    {
      "epoch": 0.5423219779270569,
      "grad_norm": 31.5,
      "learning_rate": 9.11690240654815e-06,
      "loss": 0.8097,
      "num_input_tokens_seen": 3028566272,
      "step": 40810,
      "train_runtime": 48608.6449,
      "train_tokens_per_second": 62305.096
    },
    {
      "epoch": 0.5424548674095189,
      "grad_norm": 22.125,
      "learning_rate": 9.112615339509756e-06,
      "loss": 0.9445,
      "num_input_tokens_seen": 3029423872,
      "step": 40820,
      "train_runtime": 48621.5447,
      "train_tokens_per_second": 62306.204
    },
    {
      "epoch": 0.5425877568919808,
      "grad_norm": 13.375,
      "learning_rate": 9.10832843685146e-06,
      "loss": 0.7789,
      "num_input_tokens_seen": 3029940224,
      "step": 40830,
      "train_runtime": 48628.441,
      "train_tokens_per_second": 62307.986
    },
    {
      "epoch": 0.5427206463744427,
      "grad_norm": 184.0,
      "learning_rate": 9.104041699367371e-06,
      "loss": 0.6556,
      "num_input_tokens_seen": 3030172416,
      "step": 40840,
      "train_runtime": 48631.9126,
      "train_tokens_per_second": 62308.313
    },
    {
      "epoch": 0.5428535358569047,
      "grad_norm": 8.4375,
      "learning_rate": 9.099755127851562e-06,
      "loss": 0.4919,
      "num_input_tokens_seen": 3030288128,
      "step": 40850,
      "train_runtime": 48634.958,
      "train_tokens_per_second": 62306.79
    },
    {
      "epoch": 0.5429864253393665,
      "grad_norm": 34.75,
      "learning_rate": 9.095468723098094e-06,
      "loss": 0.8086,
      "num_input_tokens_seen": 3032307712,
      "step": 40860,
      "train_runtime": 48667.6275,
      "train_tokens_per_second": 62306.463
    },
    {
      "epoch": 0.5431193148218284,
      "grad_norm": 23.875,
      "learning_rate": 9.091182485900978e-06,
      "loss": 0.9035,
      "num_input_tokens_seen": 3033196800,
      "step": 40870,
      "train_runtime": 48680.4792,
      "train_tokens_per_second": 62308.277
    },
    {
      "epoch": 0.5432522043042903,
      "grad_norm": 17.75,
      "learning_rate": 9.086896417054204e-06,
      "loss": 0.691,
      "num_input_tokens_seen": 3033648128,
      "step": 40880,
      "train_runtime": 48687.1206,
      "train_tokens_per_second": 62309.048
    },
    {
      "epoch": 0.5433850937867523,
      "grad_norm": 100.5,
      "learning_rate": 9.082610517351721e-06,
      "loss": 0.6024,
      "num_input_tokens_seen": 3033875200,
      "step": 40890,
      "train_runtime": 48690.5653,
      "train_tokens_per_second": 62309.303
    },
    {
      "epoch": 0.5435179832692142,
      "grad_norm": 29.625,
      "learning_rate": 9.078324787587465e-06,
      "loss": 0.5134,
      "num_input_tokens_seen": 3033991936,
      "step": 40900,
      "train_runtime": 48693.0801,
      "train_tokens_per_second": 62308.483
    },
    {
      "epoch": 0.5436508727516761,
      "grad_norm": 17.25,
      "learning_rate": 9.074039228555322e-06,
      "loss": 0.777,
      "num_input_tokens_seen": 3036037632,
      "step": 40910,
      "train_runtime": 48725.7707,
      "train_tokens_per_second": 62308.663
    },
    {
      "epoch": 0.5437837622341379,
      "grad_norm": 11.0,
      "learning_rate": 9.069753841049156e-06,
      "loss": 1.0307,
      "num_input_tokens_seen": 3036902656,
      "step": 40920,
      "train_runtime": 48739.1415,
      "train_tokens_per_second": 62309.318
    },
    {
      "epoch": 0.5439166517165999,
      "grad_norm": 13.0,
      "learning_rate": 9.065468625862798e-06,
      "loss": 0.7763,
      "num_input_tokens_seen": 3037374720,
      "step": 40930,
      "train_runtime": 48746.0059,
      "train_tokens_per_second": 62310.228
    },
    {
      "epoch": 0.5440495411990618,
      "grad_norm": 36.0,
      "learning_rate": 9.061183583790038e-06,
      "loss": 0.668,
      "num_input_tokens_seen": 3037607680,
      "step": 40940,
      "train_runtime": 48749.546,
      "train_tokens_per_second": 62310.481
    },
    {
      "epoch": 0.5441824306815237,
      "grad_norm": 21.375,
      "learning_rate": 9.056898715624653e-06,
      "loss": 0.5154,
      "num_input_tokens_seen": 3037720576,
      "step": 40950,
      "train_runtime": 48752.0725,
      "train_tokens_per_second": 62309.568
    },
    {
      "epoch": 0.5443153201639856,
      "grad_norm": 19.875,
      "learning_rate": 9.052614022160368e-06,
      "loss": 0.8669,
      "num_input_tokens_seen": 3039592448,
      "step": 40960,
      "train_runtime": 48782.7337,
      "train_tokens_per_second": 62308.776
    },
    {
      "epoch": 0.5444482096464476,
      "grad_norm": 33.0,
      "learning_rate": 9.04832950419089e-06,
      "loss": 0.89,
      "num_input_tokens_seen": 3040401920,
      "step": 40970,
      "train_runtime": 48795.191,
      "train_tokens_per_second": 62309.458
    },
    {
      "epoch": 0.5445810991289094,
      "grad_norm": 33.25,
      "learning_rate": 9.044045162509884e-06,
      "loss": 0.771,
      "num_input_tokens_seen": 3040844288,
      "step": 40980,
      "train_runtime": 48801.4259,
      "train_tokens_per_second": 62310.562
    },
    {
      "epoch": 0.5447139886113713,
      "grad_norm": 27.875,
      "learning_rate": 9.03976099791099e-06,
      "loss": 0.6677,
      "num_input_tokens_seen": 3041067008,
      "step": 40990,
      "train_runtime": 48804.7732,
      "train_tokens_per_second": 62310.852
    },
    {
      "epoch": 0.5448468780938333,
      "grad_norm": 25.375,
      "learning_rate": 9.03547701118781e-06,
      "loss": 0.5636,
      "num_input_tokens_seen": 3041181696,
      "step": 41000,
      "train_runtime": 48807.2289,
      "train_tokens_per_second": 62310.067
    },
    {
      "epoch": 0.5449797675762952,
      "grad_norm": 16.625,
      "learning_rate": 9.031193203133908e-06,
      "loss": 0.843,
      "num_input_tokens_seen": 3043129600,
      "step": 41010,
      "train_runtime": 48839.1588,
      "train_tokens_per_second": 62309.214
    },
    {
      "epoch": 0.5451126570587571,
      "grad_norm": 42.25,
      "learning_rate": 9.02690957454283e-06,
      "loss": 0.8752,
      "num_input_tokens_seen": 3043973888,
      "step": 41020,
      "train_runtime": 48851.8578,
      "train_tokens_per_second": 62310.299
    },
    {
      "epoch": 0.545245546541219,
      "grad_norm": 15.8125,
      "learning_rate": 9.022626126208074e-06,
      "loss": 0.7644,
      "num_input_tokens_seen": 3044437504,
      "step": 41030,
      "train_runtime": 48858.4757,
      "train_tokens_per_second": 62311.349
    },
    {
      "epoch": 0.5453784360236809,
      "grad_norm": 9.375,
      "learning_rate": 9.018342858923118e-06,
      "loss": 0.6142,
      "num_input_tokens_seen": 3044672512,
      "step": 41040,
      "train_runtime": 48861.9622,
      "train_tokens_per_second": 62311.712
    },
    {
      "epoch": 0.5455113255061428,
      "grad_norm": 20.5,
      "learning_rate": 9.014059773481387e-06,
      "loss": 0.5182,
      "num_input_tokens_seen": 3044786688,
      "step": 41050,
      "train_runtime": 48864.3377,
      "train_tokens_per_second": 62311.019
    },
    {
      "epoch": 0.5456442149886047,
      "grad_norm": 18.25,
      "learning_rate": 9.009776870676296e-06,
      "loss": 0.8338,
      "num_input_tokens_seen": 3046657280,
      "step": 41060,
      "train_runtime": 48894.2682,
      "train_tokens_per_second": 62311.134
    },
    {
      "epoch": 0.5457771044710666,
      "grad_norm": 15.3125,
      "learning_rate": 9.005494151301208e-06,
      "loss": 0.9035,
      "num_input_tokens_seen": 3047437312,
      "step": 41070,
      "train_runtime": 48906.2466,
      "train_tokens_per_second": 62311.822
    },
    {
      "epoch": 0.5459099939535286,
      "grad_norm": 19.75,
      "learning_rate": 9.001211616149463e-06,
      "loss": 0.7742,
      "num_input_tokens_seen": 3047898880,
      "step": 41080,
      "train_runtime": 48912.8494,
      "train_tokens_per_second": 62312.847
    },
    {
      "epoch": 0.5460428834359905,
      "grad_norm": 22.0,
      "learning_rate": 8.996929266014361e-06,
      "loss": 0.5742,
      "num_input_tokens_seen": 3048121600,
      "step": 41090,
      "train_runtime": 48916.863,
      "train_tokens_per_second": 62312.287
    },
    {
      "epoch": 0.5461757729184523,
      "grad_norm": 10.125,
      "learning_rate": 8.992647101689163e-06,
      "loss": 0.442,
      "num_input_tokens_seen": 3048233472,
      "step": 41100,
      "train_runtime": 48919.2615,
      "train_tokens_per_second": 62311.519
    },
    {
      "epoch": 0.5463086624009142,
      "grad_norm": 15.9375,
      "learning_rate": 8.988365123967112e-06,
      "loss": 0.8221,
      "num_input_tokens_seen": 3050335744,
      "step": 41110,
      "train_runtime": 48953.9819,
      "train_tokens_per_second": 62310.268
    },
    {
      "epoch": 0.5464415518833762,
      "grad_norm": 29.125,
      "learning_rate": 8.984083333641395e-06,
      "loss": 0.9249,
      "num_input_tokens_seen": 3051208704,
      "step": 41120,
      "train_runtime": 48966.9137,
      "train_tokens_per_second": 62311.64
    },
    {
      "epoch": 0.5465744413658381,
      "grad_norm": 17.0,
      "learning_rate": 8.97980173150519e-06,
      "loss": 0.7857,
      "num_input_tokens_seen": 3051698944,
      "step": 41130,
      "train_runtime": 48973.854,
      "train_tokens_per_second": 62312.82
    },
    {
      "epoch": 0.5467073308483,
      "grad_norm": 12.375,
      "learning_rate": 8.975520318351611e-06,
      "loss": 0.6141,
      "num_input_tokens_seen": 3051932672,
      "step": 41140,
      "train_runtime": 48977.4073,
      "train_tokens_per_second": 62313.071
    },
    {
      "epoch": 0.546840220330762,
      "grad_norm": 21.625,
      "learning_rate": 8.971239094973762e-06,
      "loss": 0.5309,
      "num_input_tokens_seen": 3052051456,
      "step": 41150,
      "train_runtime": 48980.4427,
      "train_tokens_per_second": 62311.635
    },
    {
      "epoch": 0.5469731098132238,
      "grad_norm": 23.75,
      "learning_rate": 8.966958062164698e-06,
      "loss": 0.8923,
      "num_input_tokens_seen": 3053993728,
      "step": 41160,
      "train_runtime": 49012.1538,
      "train_tokens_per_second": 62310.947
    },
    {
      "epoch": 0.5471059992956857,
      "grad_norm": 16.375,
      "learning_rate": 8.962677220717435e-06,
      "loss": 0.8438,
      "num_input_tokens_seen": 3054813696,
      "step": 41170,
      "train_runtime": 49024.3264,
      "train_tokens_per_second": 62312.201
    },
    {
      "epoch": 0.5472388887781476,
      "grad_norm": 27.5,
      "learning_rate": 8.958396571424973e-06,
      "loss": 0.6635,
      "num_input_tokens_seen": 3055249664,
      "step": 41180,
      "train_runtime": 49030.4339,
      "train_tokens_per_second": 62313.331
    },
    {
      "epoch": 0.5473717782606096,
      "grad_norm": 10.25,
      "learning_rate": 8.954116115080256e-06,
      "loss": 0.6632,
      "num_input_tokens_seen": 3055460096,
      "step": 41190,
      "train_runtime": 49033.7162,
      "train_tokens_per_second": 62313.451
    },
    {
      "epoch": 0.5475046677430715,
      "grad_norm": 6.59375,
      "learning_rate": 8.949835852476204e-06,
      "loss": 0.4965,
      "num_input_tokens_seen": 3055572224,
      "step": 41200,
      "train_runtime": 49036.7103,
      "train_tokens_per_second": 62311.933
    },
    {
      "epoch": 0.5476375572255334,
      "grad_norm": 38.75,
      "learning_rate": 8.945555784405692e-06,
      "loss": 0.8221,
      "num_input_tokens_seen": 3057688320,
      "step": 41210,
      "train_runtime": 49071.172,
      "train_tokens_per_second": 62311.296
    },
    {
      "epoch": 0.5477704467079954,
      "grad_norm": 26.5,
      "learning_rate": 8.941275911661575e-06,
      "loss": 0.9555,
      "num_input_tokens_seen": 3058575616,
      "step": 41220,
      "train_runtime": 49084.586,
      "train_tokens_per_second": 62312.344
    },
    {
      "epoch": 0.5479033361904572,
      "grad_norm": 28.625,
      "learning_rate": 8.93699623503665e-06,
      "loss": 0.7779,
      "num_input_tokens_seen": 3059087104,
      "step": 41230,
      "train_runtime": 49091.8964,
      "train_tokens_per_second": 62313.484
    },
    {
      "epoch": 0.5480362256729191,
      "grad_norm": 15.75,
      "learning_rate": 8.932716755323698e-06,
      "loss": 0.6305,
      "num_input_tokens_seen": 3059328000,
      "step": 41240,
      "train_runtime": 49095.5424,
      "train_tokens_per_second": 62313.763
    },
    {
      "epoch": 0.548169115155381,
      "grad_norm": 11.3125,
      "learning_rate": 8.92843747331545e-06,
      "loss": 0.4946,
      "num_input_tokens_seen": 3059444992,
      "step": 41250,
      "train_runtime": 49098.5644,
      "train_tokens_per_second": 62312.311
    },
    {
      "epoch": 0.548302004637843,
      "grad_norm": 35.25,
      "learning_rate": 8.9241583898046e-06,
      "loss": 0.8028,
      "num_input_tokens_seen": 3061486080,
      "step": 41260,
      "train_runtime": 49132.4498,
      "train_tokens_per_second": 62310.878
    },
    {
      "epoch": 0.5484348941203049,
      "grad_norm": 34.5,
      "learning_rate": 8.91987950558382e-06,
      "loss": 0.9091,
      "num_input_tokens_seen": 3062305792,
      "step": 41270,
      "train_runtime": 49144.8618,
      "train_tokens_per_second": 62311.82
    },
    {
      "epoch": 0.5485677836027668,
      "grad_norm": 19.0,
      "learning_rate": 8.915600821445727e-06,
      "loss": 0.7667,
      "num_input_tokens_seen": 3062764032,
      "step": 41280,
      "train_runtime": 49151.4057,
      "train_tokens_per_second": 62312.847
    },
    {
      "epoch": 0.5487006730852286,
      "grad_norm": 16.125,
      "learning_rate": 8.911322338182914e-06,
      "loss": 0.5805,
      "num_input_tokens_seen": 3063002368,
      "step": 41290,
      "train_runtime": 49154.948,
      "train_tokens_per_second": 62313.205
    },
    {
      "epoch": 0.5488335625676906,
      "grad_norm": 16.5,
      "learning_rate": 8.907044056587932e-06,
      "loss": 0.5718,
      "num_input_tokens_seen": 3063122432,
      "step": 41300,
      "train_runtime": 49157.3913,
      "train_tokens_per_second": 62312.551
    },
    {
      "epoch": 0.5489664520501525,
      "grad_norm": 86.0,
      "learning_rate": 8.902765977453292e-06,
      "loss": 0.8702,
      "num_input_tokens_seen": 3065119232,
      "step": 41310,
      "train_runtime": 49189.2603,
      "train_tokens_per_second": 62312.773
    },
    {
      "epoch": 0.5490993415326144,
      "grad_norm": 19.375,
      "learning_rate": 8.898488101571471e-06,
      "loss": 0.9156,
      "num_input_tokens_seen": 3065961728,
      "step": 41320,
      "train_runtime": 49201.7539,
      "train_tokens_per_second": 62314.074
    },
    {
      "epoch": 0.5492322310150763,
      "grad_norm": 24.125,
      "learning_rate": 8.894210429734904e-06,
      "loss": 0.7027,
      "num_input_tokens_seen": 3066436864,
      "step": 41330,
      "train_runtime": 49208.9448,
      "train_tokens_per_second": 62314.623
    },
    {
      "epoch": 0.5493651204975383,
      "grad_norm": 18.75,
      "learning_rate": 8.889932962735998e-06,
      "loss": 0.6581,
      "num_input_tokens_seen": 3066665984,
      "step": 41340,
      "train_runtime": 49212.3829,
      "train_tokens_per_second": 62314.926
    },
    {
      "epoch": 0.5494980099800001,
      "grad_norm": 20.0,
      "learning_rate": 8.885655701367111e-06,
      "loss": 0.4781,
      "num_input_tokens_seen": 3066782720,
      "step": 41350,
      "train_runtime": 49214.7845,
      "train_tokens_per_second": 62314.257
    },
    {
      "epoch": 0.549630899462462,
      "grad_norm": 32.5,
      "learning_rate": 8.88137864642057e-06,
      "loss": 0.8087,
      "num_input_tokens_seen": 3068873728,
      "step": 41360,
      "train_runtime": 49247.9477,
      "train_tokens_per_second": 62314.754
    },
    {
      "epoch": 0.549763788944924,
      "grad_norm": 22.625,
      "learning_rate": 8.877101798688658e-06,
      "loss": 0.9107,
      "num_input_tokens_seen": 3069731072,
      "step": 41370,
      "train_runtime": 49260.7634,
      "train_tokens_per_second": 62315.946
    },
    {
      "epoch": 0.5498966784273859,
      "grad_norm": 374.0,
      "learning_rate": 8.872825158963628e-06,
      "loss": 0.7307,
      "num_input_tokens_seen": 3070241536,
      "step": 41380,
      "train_runtime": 49267.7057,
      "train_tokens_per_second": 62317.526
    },
    {
      "epoch": 0.5500295679098478,
      "grad_norm": 19.0,
      "learning_rate": 8.868548728037684e-06,
      "loss": 0.5831,
      "num_input_tokens_seen": 3070472192,
      "step": 41390,
      "train_runtime": 49271.8415,
      "train_tokens_per_second": 62316.977
    },
    {
      "epoch": 0.5501624573923097,
      "grad_norm": 13.5,
      "learning_rate": 8.864272506703003e-06,
      "loss": 0.547,
      "num_input_tokens_seen": 3070588160,
      "step": 41400,
      "train_runtime": 49274.1859,
      "train_tokens_per_second": 62316.365
    },
    {
      "epoch": 0.5502953468747716,
      "grad_norm": 18.375,
      "learning_rate": 8.859996495751712e-06,
      "loss": 0.7303,
      "num_input_tokens_seen": 3072556544,
      "step": 41410,
      "train_runtime": 49307.3513,
      "train_tokens_per_second": 62314.37
    },
    {
      "epoch": 0.5504282363572335,
      "grad_norm": 10.5625,
      "learning_rate": 8.855720695975899e-06,
      "loss": 0.8812,
      "num_input_tokens_seen": 3073391360,
      "step": 41420,
      "train_runtime": 49320.0684,
      "train_tokens_per_second": 62315.229
    },
    {
      "epoch": 0.5505611258396954,
      "grad_norm": 20.75,
      "learning_rate": 8.85144510816763e-06,
      "loss": 0.7531,
      "num_input_tokens_seen": 3073854208,
      "step": 41430,
      "train_runtime": 49327.4644,
      "train_tokens_per_second": 62315.269
    },
    {
      "epoch": 0.5506940153221573,
      "grad_norm": 11.75,
      "learning_rate": 8.84716973311891e-06,
      "loss": 0.5915,
      "num_input_tokens_seen": 3074083072,
      "step": 41440,
      "train_runtime": 49330.9281,
      "train_tokens_per_second": 62315.533
    },
    {
      "epoch": 0.5508269048046193,
      "grad_norm": 10.9375,
      "learning_rate": 8.842894571621721e-06,
      "loss": 0.5046,
      "num_input_tokens_seen": 3074196480,
      "step": 41450,
      "train_runtime": 49333.273,
      "train_tokens_per_second": 62314.87
    },
    {
      "epoch": 0.5509597942870812,
      "grad_norm": 34.5,
      "learning_rate": 8.838619624467989e-06,
      "loss": 0.834,
      "num_input_tokens_seen": 3076317440,
      "step": 41460,
      "train_runtime": 49367.4147,
      "train_tokens_per_second": 62314.737
    },
    {
      "epoch": 0.551092683769543,
      "grad_norm": 23.75,
      "learning_rate": 8.83434489244962e-06,
      "loss": 0.8361,
      "num_input_tokens_seen": 3077161216,
      "step": 41470,
      "train_runtime": 49380.3043,
      "train_tokens_per_second": 62315.558
    },
    {
      "epoch": 0.5512255732520049,
      "grad_norm": 29.125,
      "learning_rate": 8.830070376358467e-06,
      "loss": 0.8025,
      "num_input_tokens_seen": 3077672192,
      "step": 41480,
      "train_runtime": 49387.3113,
      "train_tokens_per_second": 62317.063
    },
    {
      "epoch": 0.5513584627344669,
      "grad_norm": 18.0,
      "learning_rate": 8.825796076986339e-06,
      "loss": 0.6512,
      "num_input_tokens_seen": 3077910016,
      "step": 41490,
      "train_runtime": 49390.8631,
      "train_tokens_per_second": 62317.397
    },
    {
      "epoch": 0.5514913522169288,
      "grad_norm": 9.0625,
      "learning_rate": 8.821521995125022e-06,
      "loss": 0.4917,
      "num_input_tokens_seen": 3078024704,
      "step": 41500,
      "train_runtime": 49393.834,
      "train_tokens_per_second": 62315.971
    },
    {
      "epoch": 0.5516242416993907,
      "grad_norm": 18.625,
      "learning_rate": 8.817248131566247e-06,
      "loss": 0.751,
      "num_input_tokens_seen": 3080018944,
      "step": 41510,
      "train_runtime": 49425.3375,
      "train_tokens_per_second": 62316.599
    },
    {
      "epoch": 0.5517571311818527,
      "grad_norm": 25.625,
      "learning_rate": 8.81297448710171e-06,
      "loss": 0.9738,
      "num_input_tokens_seen": 3080863488,
      "step": 41520,
      "train_runtime": 49437.9089,
      "train_tokens_per_second": 62317.836
    },
    {
      "epoch": 0.5518900206643145,
      "grad_norm": 15.875,
      "learning_rate": 8.808701062523061e-06,
      "loss": 0.7897,
      "num_input_tokens_seen": 3081324544,
      "step": 41530,
      "train_runtime": 49444.619,
      "train_tokens_per_second": 62318.703
    },
    {
      "epoch": 0.5520229101467764,
      "grad_norm": 12.0,
      "learning_rate": 8.80442785862192e-06,
      "loss": 0.6228,
      "num_input_tokens_seen": 3081550080,
      "step": 41540,
      "train_runtime": 49448.1019,
      "train_tokens_per_second": 62318.875
    },
    {
      "epoch": 0.5521557996292383,
      "grad_norm": 9.25,
      "learning_rate": 8.800154876189858e-06,
      "loss": 0.5912,
      "num_input_tokens_seen": 3081668608,
      "step": 41550,
      "train_runtime": 49451.1796,
      "train_tokens_per_second": 62317.393
    },
    {
      "epoch": 0.5522886891117003,
      "grad_norm": 17.5,
      "learning_rate": 8.795882116018407e-06,
      "loss": 0.8074,
      "num_input_tokens_seen": 3083721216,
      "step": 41560,
      "train_runtime": 49484.2914,
      "train_tokens_per_second": 62317.174
    },
    {
      "epoch": 0.5524215785941622,
      "grad_norm": 16.625,
      "learning_rate": 8.791609578899056e-06,
      "loss": 0.9451,
      "num_input_tokens_seen": 3084552960,
      "step": 41570,
      "train_runtime": 49496.9703,
      "train_tokens_per_second": 62318.015
    },
    {
      "epoch": 0.5525544680766241,
      "grad_norm": 22.875,
      "learning_rate": 8.787337265623253e-06,
      "loss": 0.7214,
      "num_input_tokens_seen": 3085027840,
      "step": 41580,
      "train_runtime": 49503.7174,
      "train_tokens_per_second": 62319.115
    },
    {
      "epoch": 0.5526873575590859,
      "grad_norm": 85.5,
      "learning_rate": 8.783065176982411e-06,
      "loss": 0.6262,
      "num_input_tokens_seen": 3085259264,
      "step": 41590,
      "train_runtime": 49507.2236,
      "train_tokens_per_second": 62319.376
    },
    {
      "epoch": 0.5528202470415479,
      "grad_norm": 15.875,
      "learning_rate": 8.778793313767894e-06,
      "loss": 0.5617,
      "num_input_tokens_seen": 3085371904,
      "step": 41600,
      "train_runtime": 49509.5579,
      "train_tokens_per_second": 62318.713
    },
    {
      "epoch": 0.5529531365240098,
      "grad_norm": 19.5,
      "learning_rate": 8.774521676771027e-06,
      "loss": 0.8232,
      "num_input_tokens_seen": 3087486720,
      "step": 41610,
      "train_runtime": 49542.9488,
      "train_tokens_per_second": 62319.397
    },
    {
      "epoch": 0.5530860260064717,
      "grad_norm": 29.125,
      "learning_rate": 8.770250266783087e-06,
      "loss": 0.9722,
      "num_input_tokens_seen": 3088350720,
      "step": 41620,
      "train_runtime": 49556.2456,
      "train_tokens_per_second": 62320.111
    },
    {
      "epoch": 0.5532189154889336,
      "grad_norm": 16.25,
      "learning_rate": 8.765979084595325e-06,
      "loss": 0.7324,
      "num_input_tokens_seen": 3088818944,
      "step": 41630,
      "train_runtime": 49562.7483,
      "train_tokens_per_second": 62321.381
    },
    {
      "epoch": 0.5533518049713956,
      "grad_norm": 25.125,
      "learning_rate": 8.761708130998934e-06,
      "loss": 0.6907,
      "num_input_tokens_seen": 3089039616,
      "step": 41640,
      "train_runtime": 49566.1324,
      "train_tokens_per_second": 62321.579
    },
    {
      "epoch": 0.5534846944538574,
      "grad_norm": 25.75,
      "learning_rate": 8.757437406785065e-06,
      "loss": 0.6262,
      "num_input_tokens_seen": 3089153024,
      "step": 41650,
      "train_runtime": 49568.5339,
      "train_tokens_per_second": 62320.847
    },
    {
      "epoch": 0.5536175839363193,
      "grad_norm": 48.0,
      "learning_rate": 8.753166912744843e-06,
      "loss": 0.816,
      "num_input_tokens_seen": 3091300608,
      "step": 41660,
      "train_runtime": 49601.6535,
      "train_tokens_per_second": 62322.531
    },
    {
      "epoch": 0.5537504734187813,
      "grad_norm": 24.25,
      "learning_rate": 8.748896649669328e-06,
      "loss": 0.9131,
      "num_input_tokens_seen": 3092158208,
      "step": 41670,
      "train_runtime": 49615.148,
      "train_tokens_per_second": 62322.866
    },
    {
      "epoch": 0.5538833629012432,
      "grad_norm": 18.875,
      "learning_rate": 8.744626618349556e-06,
      "loss": 0.7102,
      "num_input_tokens_seen": 3092604160,
      "step": 41680,
      "train_runtime": 49621.6893,
      "train_tokens_per_second": 62323.637
    },
    {
      "epoch": 0.5540162523837051,
      "grad_norm": 40.25,
      "learning_rate": 8.740356819576505e-06,
      "loss": 0.644,
      "num_input_tokens_seen": 3092839168,
      "step": 41690,
      "train_runtime": 49625.1112,
      "train_tokens_per_second": 62324.075
    },
    {
      "epoch": 0.554149141866167,
      "grad_norm": 25.0,
      "learning_rate": 8.736087254141125e-06,
      "loss": 0.6022,
      "num_input_tokens_seen": 3092957440,
      "step": 41700,
      "train_runtime": 49627.4818,
      "train_tokens_per_second": 62323.481
    },
    {
      "epoch": 0.554282031348629,
      "grad_norm": 27.875,
      "learning_rate": 8.731817922834308e-06,
      "loss": 0.7907,
      "num_input_tokens_seen": 3095014144,
      "step": 41710,
      "train_runtime": 49662.6151,
      "train_tokens_per_second": 62320.805
    },
    {
      "epoch": 0.5544149208310908,
      "grad_norm": 19.75,
      "learning_rate": 8.727548826446916e-06,
      "loss": 0.8032,
      "num_input_tokens_seen": 3095822336,
      "step": 41720,
      "train_runtime": 49675.0993,
      "train_tokens_per_second": 62321.412
    },
    {
      "epoch": 0.5545478103135527,
      "grad_norm": 18.625,
      "learning_rate": 8.723279965769757e-06,
      "loss": 0.744,
      "num_input_tokens_seen": 3096307712,
      "step": 41730,
      "train_runtime": 49682.049,
      "train_tokens_per_second": 62322.464
    },
    {
      "epoch": 0.5546806997960146,
      "grad_norm": 21.75,
      "learning_rate": 8.719011341593598e-06,
      "loss": 0.6483,
      "num_input_tokens_seen": 3096555264,
      "step": 41740,
      "train_runtime": 49685.7154,
      "train_tokens_per_second": 62322.848
    },
    {
      "epoch": 0.5548135892784766,
      "grad_norm": 10.1875,
      "learning_rate": 8.714742954709168e-06,
      "loss": 0.4576,
      "num_input_tokens_seen": 3096673024,
      "step": 41750,
      "train_runtime": 49688.0874,
      "train_tokens_per_second": 62322.242
    },
    {
      "epoch": 0.5549464787609385,
      "grad_norm": 16.25,
      "learning_rate": 8.710474805907142e-06,
      "loss": 0.8414,
      "num_input_tokens_seen": 3098722048,
      "step": 41760,
      "train_runtime": 49720.4216,
      "train_tokens_per_second": 62322.924
    },
    {
      "epoch": 0.5550793682434004,
      "grad_norm": 26.0,
      "learning_rate": 8.706206895978163e-06,
      "loss": 0.9125,
      "num_input_tokens_seen": 3099552768,
      "step": 41770,
      "train_runtime": 49732.642,
      "train_tokens_per_second": 62324.313
    },
    {
      "epoch": 0.5552122577258622,
      "grad_norm": 28.875,
      "learning_rate": 8.701939225712817e-06,
      "loss": 0.8244,
      "num_input_tokens_seen": 3100015616,
      "step": 41780,
      "train_runtime": 49739.8509,
      "train_tokens_per_second": 62324.586
    },
    {
      "epoch": 0.5553451472083242,
      "grad_norm": 35.5,
      "learning_rate": 8.697671795901656e-06,
      "loss": 0.66,
      "num_input_tokens_seen": 3100253440,
      "step": 41790,
      "train_runtime": 49743.3902,
      "train_tokens_per_second": 62324.933
    },
    {
      "epoch": 0.5554780366907861,
      "grad_norm": 22.625,
      "learning_rate": 8.693404607335184e-06,
      "loss": 0.513,
      "num_input_tokens_seen": 3100369920,
      "step": 41800,
      "train_runtime": 49745.7942,
      "train_tokens_per_second": 62324.262
    },
    {
      "epoch": 0.555610926173248,
      "grad_norm": 43.75,
      "learning_rate": 8.689137660803854e-06,
      "loss": 0.8415,
      "num_input_tokens_seen": 3102349312,
      "step": 41810,
      "train_runtime": 49777.7008,
      "train_tokens_per_second": 62324.078
    },
    {
      "epoch": 0.55574381565571,
      "grad_norm": 29.375,
      "learning_rate": 8.684870957098089e-06,
      "loss": 0.8986,
      "num_input_tokens_seen": 3103185408,
      "step": 41820,
      "train_runtime": 49789.9897,
      "train_tokens_per_second": 62325.488
    },
    {
      "epoch": 0.5558767051381719,
      "grad_norm": 17.75,
      "learning_rate": 8.680604497008248e-06,
      "loss": 0.6964,
      "num_input_tokens_seen": 3103651584,
      "step": 41830,
      "train_runtime": 49797.2835,
      "train_tokens_per_second": 62325.721
    },
    {
      "epoch": 0.5560095946206337,
      "grad_norm": 15.25,
      "learning_rate": 8.67633828132466e-06,
      "loss": 0.6443,
      "num_input_tokens_seen": 3103874816,
      "step": 41840,
      "train_runtime": 49800.6716,
      "train_tokens_per_second": 62325.963
    },
    {
      "epoch": 0.5561424841030956,
      "grad_norm": 16.875,
      "learning_rate": 8.672072310837601e-06,
      "loss": 0.5021,
      "num_input_tokens_seen": 3103988480,
      "step": 41850,
      "train_runtime": 49803.0093,
      "train_tokens_per_second": 62325.32
    },
    {
      "epoch": 0.5562753735855576,
      "grad_norm": 72.0,
      "learning_rate": 8.667806586337309e-06,
      "loss": 0.8059,
      "num_input_tokens_seen": 3106023424,
      "step": 41860,
      "train_runtime": 49836.8558,
      "train_tokens_per_second": 62323.824
    },
    {
      "epoch": 0.5564082630680195,
      "grad_norm": 20.625,
      "learning_rate": 8.663541108613966e-06,
      "loss": 0.9058,
      "num_input_tokens_seen": 3106894592,
      "step": 41870,
      "train_runtime": 49850.3534,
      "train_tokens_per_second": 62324.425
    },
    {
      "epoch": 0.5565411525504814,
      "grad_norm": 14.25,
      "learning_rate": 8.659275878457717e-06,
      "loss": 0.6782,
      "num_input_tokens_seen": 3107378944,
      "step": 41880,
      "train_runtime": 49857.0419,
      "train_tokens_per_second": 62325.778
    },
    {
      "epoch": 0.5566740420329434,
      "grad_norm": 32.75,
      "learning_rate": 8.655010896658658e-06,
      "loss": 0.6794,
      "num_input_tokens_seen": 3107608576,
      "step": 41890,
      "train_runtime": 49860.4707,
      "train_tokens_per_second": 62326.098
    },
    {
      "epoch": 0.5568069315154052,
      "grad_norm": 9.0625,
      "learning_rate": 8.650746164006832e-06,
      "loss": 0.5534,
      "num_input_tokens_seen": 3107720192,
      "step": 41900,
      "train_runtime": 49862.7866,
      "train_tokens_per_second": 62325.442
    },
    {
      "epoch": 0.5569398209978671,
      "grad_norm": 20.125,
      "learning_rate": 8.646481681292251e-06,
      "loss": 0.8091,
      "num_input_tokens_seen": 3109832448,
      "step": 41910,
      "train_runtime": 49896.3264,
      "train_tokens_per_second": 62325.88
    },
    {
      "epoch": 0.557072710480329,
      "grad_norm": 22.25,
      "learning_rate": 8.64221744930487e-06,
      "loss": 0.8746,
      "num_input_tokens_seen": 3110694400,
      "step": 41920,
      "train_runtime": 49909.3059,
      "train_tokens_per_second": 62326.942
    },
    {
      "epoch": 0.557205599962791,
      "grad_norm": 35.5,
      "learning_rate": 8.637953468834599e-06,
      "loss": 0.7837,
      "num_input_tokens_seen": 3111186688,
      "step": 41930,
      "train_runtime": 49916.1905,
      "train_tokens_per_second": 62328.208
    },
    {
      "epoch": 0.5573384894452529,
      "grad_norm": 8.0,
      "learning_rate": 8.633689740671301e-06,
      "loss": 0.6657,
      "num_input_tokens_seen": 3111416576,
      "step": 41940,
      "train_runtime": 49919.65,
      "train_tokens_per_second": 62328.493
    },
    {
      "epoch": 0.5574713789277148,
      "grad_norm": 12.4375,
      "learning_rate": 8.629426265604797e-06,
      "loss": 0.5178,
      "num_input_tokens_seen": 3111532544,
      "step": 41950,
      "train_runtime": 49922.6955,
      "train_tokens_per_second": 62327.014
    },
    {
      "epoch": 0.5576042684101766,
      "grad_norm": 45.75,
      "learning_rate": 8.625163044424858e-06,
      "loss": 0.7204,
      "num_input_tokens_seen": 3113583360,
      "step": 41960,
      "train_runtime": 49955.435,
      "train_tokens_per_second": 62327.22
    },
    {
      "epoch": 0.5577371578926386,
      "grad_norm": 24.75,
      "learning_rate": 8.620900077921203e-06,
      "loss": 0.9399,
      "num_input_tokens_seen": 3114409984,
      "step": 41970,
      "train_runtime": 49968.0134,
      "train_tokens_per_second": 62328.073
    },
    {
      "epoch": 0.5578700473751005,
      "grad_norm": 18.5,
      "learning_rate": 8.616637366883513e-06,
      "loss": 0.7334,
      "num_input_tokens_seen": 3114877184,
      "step": 41980,
      "train_runtime": 49974.6521,
      "train_tokens_per_second": 62329.142
    },
    {
      "epoch": 0.5580029368575624,
      "grad_norm": 12.1875,
      "learning_rate": 8.612374912101411e-06,
      "loss": 0.6204,
      "num_input_tokens_seen": 3115109632,
      "step": 41990,
      "train_runtime": 49978.1418,
      "train_tokens_per_second": 62329.441
    },
    {
      "epoch": 0.5581358263400243,
      "grad_norm": 18.625,
      "learning_rate": 8.60811271436449e-06,
      "loss": 0.589,
      "num_input_tokens_seen": 3115226112,
      "step": 42000,
      "train_runtime": 49981.1406,
      "train_tokens_per_second": 62328.032
    },
    {
      "epoch": 0.5582687158224863,
      "grad_norm": 25.0,
      "learning_rate": 8.603850774462272e-06,
      "loss": 0.828,
      "num_input_tokens_seen": 3117374464,
      "step": 42010,
      "train_runtime": 50016.0894,
      "train_tokens_per_second": 62327.433
    },
    {
      "epoch": 0.5584016053049481,
      "grad_norm": 20.25,
      "learning_rate": 8.599589093184252e-06,
      "loss": 0.862,
      "num_input_tokens_seen": 3118228992,
      "step": 42020,
      "train_runtime": 50029.0125,
      "train_tokens_per_second": 62328.414
    },
    {
      "epoch": 0.55853449478741,
      "grad_norm": 15.0,
      "learning_rate": 8.595327671319865e-06,
      "loss": 0.7683,
      "num_input_tokens_seen": 3118706688,
      "step": 42030,
      "train_runtime": 50035.6978,
      "train_tokens_per_second": 62329.633
    },
    {
      "epoch": 0.558667384269872,
      "grad_norm": 17.875,
      "learning_rate": 8.591066509658501e-06,
      "loss": 0.5908,
      "num_input_tokens_seen": 3118929920,
      "step": 42040,
      "train_runtime": 50039.7803,
      "train_tokens_per_second": 62329.009
    },
    {
      "epoch": 0.5588002737523339,
      "grad_norm": 11.75,
      "learning_rate": 8.586805608989505e-06,
      "loss": 0.5448,
      "num_input_tokens_seen": 3119047424,
      "step": 42050,
      "train_runtime": 50042.1663,
      "train_tokens_per_second": 62328.385
    },
    {
      "epoch": 0.5589331632347958,
      "grad_norm": 97.0,
      "learning_rate": 8.582544970102164e-06,
      "loss": 0.7426,
      "num_input_tokens_seen": 3121062400,
      "step": 42060,
      "train_runtime": 50074.0364,
      "train_tokens_per_second": 62328.956
    },
    {
      "epoch": 0.5590660527172577,
      "grad_norm": 30.625,
      "learning_rate": 8.578284593785732e-06,
      "loss": 0.9328,
      "num_input_tokens_seen": 3121876736,
      "step": 42070,
      "train_runtime": 50086.3877,
      "train_tokens_per_second": 62329.844
    },
    {
      "epoch": 0.5591989421997196,
      "grad_norm": 19.5,
      "learning_rate": 8.5740244808294e-06,
      "loss": 0.799,
      "num_input_tokens_seen": 3122357248,
      "step": 42080,
      "train_runtime": 50092.9629,
      "train_tokens_per_second": 62331.255
    },
    {
      "epoch": 0.5593318316821815,
      "grad_norm": 25.0,
      "learning_rate": 8.56976463202232e-06,
      "loss": 0.6558,
      "num_input_tokens_seen": 3122584320,
      "step": 42090,
      "train_runtime": 50096.4119,
      "train_tokens_per_second": 62331.496
    },
    {
      "epoch": 0.5594647211646434,
      "grad_norm": 16.125,
      "learning_rate": 8.565505048153584e-06,
      "loss": 0.5775,
      "num_input_tokens_seen": 3122699776,
      "step": 42100,
      "train_runtime": 50098.7721,
      "train_tokens_per_second": 62330.865
    },
    {
      "epoch": 0.5595976106471053,
      "grad_norm": 18.375,
      "learning_rate": 8.56124573001225e-06,
      "loss": 0.77,
      "num_input_tokens_seen": 3124723712,
      "step": 42110,
      "train_runtime": 50131.9352,
      "train_tokens_per_second": 62330.004
    },
    {
      "epoch": 0.5597305001295673,
      "grad_norm": 13.25,
      "learning_rate": 8.556986678387318e-06,
      "loss": 0.9242,
      "num_input_tokens_seen": 3125563392,
      "step": 42120,
      "train_runtime": 50144.3683,
      "train_tokens_per_second": 62331.295
    },
    {
      "epoch": 0.5598633896120292,
      "grad_norm": 32.25,
      "learning_rate": 8.552727894067732e-06,
      "loss": 0.7258,
      "num_input_tokens_seen": 3126036992,
      "step": 42130,
      "train_runtime": 50151.0499,
      "train_tokens_per_second": 62332.434
    },
    {
      "epoch": 0.559996279094491,
      "grad_norm": 30.0,
      "learning_rate": 8.548469377842404e-06,
      "loss": 0.6837,
      "num_input_tokens_seen": 3126273024,
      "step": 42140,
      "train_runtime": 50154.5491,
      "train_tokens_per_second": 62332.791
    },
    {
      "epoch": 0.5601291685769529,
      "grad_norm": 112.0,
      "learning_rate": 8.544211130500173e-06,
      "loss": 0.4806,
      "num_input_tokens_seen": 3126389248,
      "step": 42150,
      "train_runtime": 50156.9238,
      "train_tokens_per_second": 62332.157
    },
    {
      "epoch": 0.5602620580594149,
      "grad_norm": 21.375,
      "learning_rate": 8.539953152829856e-06,
      "loss": 0.7414,
      "num_input_tokens_seen": 3128507392,
      "step": 42160,
      "train_runtime": 50192.4773,
      "train_tokens_per_second": 62330.205
    },
    {
      "epoch": 0.5603949475418768,
      "grad_norm": 30.25,
      "learning_rate": 8.535695445620191e-06,
      "loss": 0.8267,
      "num_input_tokens_seen": 3129341184,
      "step": 42170,
      "train_runtime": 50205.2348,
      "train_tokens_per_second": 62330.974
    },
    {
      "epoch": 0.5605278370243387,
      "grad_norm": 16.5,
      "learning_rate": 8.531438009659894e-06,
      "loss": 0.7353,
      "num_input_tokens_seen": 3129821184,
      "step": 42180,
      "train_runtime": 50211.9835,
      "train_tokens_per_second": 62332.156
    },
    {
      "epoch": 0.5606607265068007,
      "grad_norm": 18.5,
      "learning_rate": 8.527180845737607e-06,
      "loss": 0.6594,
      "num_input_tokens_seen": 3130057984,
      "step": 42190,
      "train_runtime": 50215.5105,
      "train_tokens_per_second": 62332.494
    },
    {
      "epoch": 0.5607936159892626,
      "grad_norm": 7.15625,
      "learning_rate": 8.522923954641941e-06,
      "loss": 0.5005,
      "num_input_tokens_seen": 3130174976,
      "step": 42200,
      "train_runtime": 50217.9305,
      "train_tokens_per_second": 62331.819
    },
    {
      "epoch": 0.5609265054717244,
      "grad_norm": 31.0,
      "learning_rate": 8.51866733716144e-06,
      "loss": 0.8391,
      "num_input_tokens_seen": 3132210176,
      "step": 42210,
      "train_runtime": 50249.9065,
      "train_tokens_per_second": 62332.657
    },
    {
      "epoch": 0.5610593949541863,
      "grad_norm": 21.125,
      "learning_rate": 8.5144109940846e-06,
      "loss": 0.9044,
      "num_input_tokens_seen": 3132999168,
      "step": 42220,
      "train_runtime": 50261.7655,
      "train_tokens_per_second": 62333.647
    },
    {
      "epoch": 0.5611922844366483,
      "grad_norm": 23.125,
      "learning_rate": 8.510154926199883e-06,
      "loss": 0.6959,
      "num_input_tokens_seen": 3133449728,
      "step": 42230,
      "train_runtime": 50268.0293,
      "train_tokens_per_second": 62334.843
    },
    {
      "epoch": 0.5613251739191102,
      "grad_norm": 11.625,
      "learning_rate": 8.505899134295678e-06,
      "loss": 0.6421,
      "num_input_tokens_seen": 3133673216,
      "step": 42240,
      "train_runtime": 50272.0856,
      "train_tokens_per_second": 62334.259
    },
    {
      "epoch": 0.5614580634015721,
      "grad_norm": 8.25,
      "learning_rate": 8.501643619160338e-06,
      "loss": 0.545,
      "num_input_tokens_seen": 3133788928,
      "step": 42250,
      "train_runtime": 50274.4214,
      "train_tokens_per_second": 62333.665
    },
    {
      "epoch": 0.561590952884034,
      "grad_norm": 14.0625,
      "learning_rate": 8.497388381582152e-06,
      "loss": 0.8041,
      "num_input_tokens_seen": 3135824640,
      "step": 42260,
      "train_runtime": 50307.3427,
      "train_tokens_per_second": 62333.339
    },
    {
      "epoch": 0.5617238423664959,
      "grad_norm": 13.5,
      "learning_rate": 8.493133422349376e-06,
      "loss": 0.8679,
      "num_input_tokens_seen": 3136675584,
      "step": 42270,
      "train_runtime": 50319.9,
      "train_tokens_per_second": 62334.694
    },
    {
      "epoch": 0.5618567318489578,
      "grad_norm": 12.6875,
      "learning_rate": 8.488878742250195e-06,
      "loss": 0.7569,
      "num_input_tokens_seen": 3137121536,
      "step": 42280,
      "train_runtime": 50326.4647,
      "train_tokens_per_second": 62335.424
    },
    {
      "epoch": 0.5619896213314197,
      "grad_norm": 8.5,
      "learning_rate": 8.484624342072752e-06,
      "loss": 0.6335,
      "num_input_tokens_seen": 3137347328,
      "step": 42290,
      "train_runtime": 50329.8762,
      "train_tokens_per_second": 62335.685
    },
    {
      "epoch": 0.5621225108138816,
      "grad_norm": 7.84375,
      "learning_rate": 8.480370222605138e-06,
      "loss": 0.5503,
      "num_input_tokens_seen": 3137464064,
      "step": 42300,
      "train_runtime": 50332.8804,
      "train_tokens_per_second": 62334.284
    },
    {
      "epoch": 0.5622554002963436,
      "grad_norm": 39.0,
      "learning_rate": 8.476116384635387e-06,
      "loss": 0.883,
      "num_input_tokens_seen": 3139545344,
      "step": 42310,
      "train_runtime": 50365.9152,
      "train_tokens_per_second": 62334.722
    },
    {
      "epoch": 0.5623882897788055,
      "grad_norm": 40.25,
      "learning_rate": 8.471862828951491e-06,
      "loss": 0.9176,
      "num_input_tokens_seen": 3140363520,
      "step": 42320,
      "train_runtime": 50378.4852,
      "train_tokens_per_second": 62335.41
    },
    {
      "epoch": 0.5625211792612673,
      "grad_norm": 16.125,
      "learning_rate": 8.467609556341375e-06,
      "loss": 0.6809,
      "num_input_tokens_seen": 3140816384,
      "step": 42330,
      "train_runtime": 50385.0633,
      "train_tokens_per_second": 62336.26
    },
    {
      "epoch": 0.5626540687437293,
      "grad_norm": 101.0,
      "learning_rate": 8.463356567592931e-06,
      "loss": 0.6008,
      "num_input_tokens_seen": 3141040896,
      "step": 42340,
      "train_runtime": 50389.1446,
      "train_tokens_per_second": 62335.666
    },
    {
      "epoch": 0.5627869582261912,
      "grad_norm": 10.625,
      "learning_rate": 8.459103863493975e-06,
      "loss": 0.5214,
      "num_input_tokens_seen": 3141153280,
      "step": 42350,
      "train_runtime": 50391.556,
      "train_tokens_per_second": 62334.913
    },
    {
      "epoch": 0.5629198477086531,
      "grad_norm": 45.25,
      "learning_rate": 8.454851444832295e-06,
      "loss": 0.871,
      "num_input_tokens_seen": 3143234560,
      "step": 42360,
      "train_runtime": 50424.1976,
      "train_tokens_per_second": 62335.837
    },
    {
      "epoch": 0.563052737191115,
      "grad_norm": 14.375,
      "learning_rate": 8.450599312395604e-06,
      "loss": 0.9372,
      "num_input_tokens_seen": 3144103168,
      "step": 42370,
      "train_runtime": 50437.0826,
      "train_tokens_per_second": 62337.134
    },
    {
      "epoch": 0.563185626673577,
      "grad_norm": 18.75,
      "learning_rate": 8.446347466971572e-06,
      "loss": 0.7834,
      "num_input_tokens_seen": 3144598784,
      "step": 42380,
      "train_runtime": 50443.9243,
      "train_tokens_per_second": 62338.504
    },
    {
      "epoch": 0.5633185161560388,
      "grad_norm": 19.125,
      "learning_rate": 8.442095909347821e-06,
      "loss": 0.6107,
      "num_input_tokens_seen": 3144825344,
      "step": 42390,
      "train_runtime": 50447.3634,
      "train_tokens_per_second": 62338.745
    },
    {
      "epoch": 0.5634514056385007,
      "grad_norm": 17.625,
      "learning_rate": 8.43784464031191e-06,
      "loss": 0.4937,
      "num_input_tokens_seen": 3144938496,
      "step": 42400,
      "train_runtime": 50449.7097,
      "train_tokens_per_second": 62338.089
    },
    {
      "epoch": 0.5635842951209626,
      "grad_norm": 21.875,
      "learning_rate": 8.433593660651351e-06,
      "loss": 0.8793,
      "num_input_tokens_seen": 3147054848,
      "step": 42410,
      "train_runtime": 50482.4651,
      "train_tokens_per_second": 62339.564
    },
    {
      "epoch": 0.5637171846034246,
      "grad_norm": 17.125,
      "learning_rate": 8.429342971153595e-06,
      "loss": 0.868,
      "num_input_tokens_seen": 3147897088,
      "step": 42420,
      "train_runtime": 50495.9594,
      "train_tokens_per_second": 62339.584
    },
    {
      "epoch": 0.5638500740858865,
      "grad_norm": 18.125,
      "learning_rate": 8.425092572606053e-06,
      "loss": 0.7565,
      "num_input_tokens_seen": 3148404480,
      "step": 42430,
      "train_runtime": 50502.9438,
      "train_tokens_per_second": 62341.009
    },
    {
      "epoch": 0.5639829635683484,
      "grad_norm": 19.5,
      "learning_rate": 8.420842465796068e-06,
      "loss": 0.6154,
      "num_input_tokens_seen": 3148645376,
      "step": 42440,
      "train_runtime": 50506.5189,
      "train_tokens_per_second": 62341.366
    },
    {
      "epoch": 0.5641158530508102,
      "grad_norm": 8.5625,
      "learning_rate": 8.416592651510936e-06,
      "loss": 0.5135,
      "num_input_tokens_seen": 3148760832,
      "step": 42450,
      "train_runtime": 50508.8852,
      "train_tokens_per_second": 62340.731
    },
    {
      "epoch": 0.5642487425332722,
      "grad_norm": 29.375,
      "learning_rate": 8.412343130537894e-06,
      "loss": 0.8597,
      "num_input_tokens_seen": 3150865408,
      "step": 42460,
      "train_runtime": 50541.987,
      "train_tokens_per_second": 62341.542
    },
    {
      "epoch": 0.5643816320157341,
      "grad_norm": 12.625,
      "learning_rate": 8.408093903664128e-06,
      "loss": 0.8799,
      "num_input_tokens_seen": 3151718144,
      "step": 42470,
      "train_runtime": 50554.7179,
      "train_tokens_per_second": 62342.711
    },
    {
      "epoch": 0.564514521498196,
      "grad_norm": 13.375,
      "learning_rate": 8.403844971676777e-06,
      "loss": 0.796,
      "num_input_tokens_seen": 3152195328,
      "step": 42480,
      "train_runtime": 50562.2457,
      "train_tokens_per_second": 62342.866
    },
    {
      "epoch": 0.564647410980658,
      "grad_norm": 16.375,
      "learning_rate": 8.399596335362908e-06,
      "loss": 0.6583,
      "num_input_tokens_seen": 3152431872,
      "step": 42490,
      "train_runtime": 50565.766,
      "train_tokens_per_second": 62343.204
    },
    {
      "epoch": 0.5647803004631199,
      "grad_norm": 13.5625,
      "learning_rate": 8.39534799550955e-06,
      "loss": 0.5381,
      "num_input_tokens_seen": 3152548864,
      "step": 42500,
      "train_runtime": 50568.1637,
      "train_tokens_per_second": 62342.562
    },
    {
      "epoch": 0.5649131899455817,
      "grad_norm": 26.625,
      "learning_rate": 8.391099952903665e-06,
      "loss": 0.8375,
      "num_input_tokens_seen": 3154543104,
      "step": 42510,
      "train_runtime": 50601.7936,
      "train_tokens_per_second": 62340.539
    },
    {
      "epoch": 0.5650460794280436,
      "grad_norm": 13.625,
      "learning_rate": 8.38685220833217e-06,
      "loss": 0.8259,
      "num_input_tokens_seen": 3155382272,
      "step": 42520,
      "train_runtime": 50614.8732,
      "train_tokens_per_second": 62341.009
    },
    {
      "epoch": 0.5651789689105056,
      "grad_norm": 25.75,
      "learning_rate": 8.38260476258192e-06,
      "loss": 0.7193,
      "num_input_tokens_seen": 3155834880,
      "step": 42530,
      "train_runtime": 50621.406,
      "train_tokens_per_second": 62341.905
    },
    {
      "epoch": 0.5653118583929675,
      "grad_norm": 28.0,
      "learning_rate": 8.378357616439711e-06,
      "loss": 0.6559,
      "num_input_tokens_seen": 3156058880,
      "step": 42540,
      "train_runtime": 50624.7805,
      "train_tokens_per_second": 62342.174
    },
    {
      "epoch": 0.5654447478754294,
      "grad_norm": 24.25,
      "learning_rate": 8.3741107706923e-06,
      "loss": 0.6159,
      "num_input_tokens_seen": 3156176128,
      "step": 42550,
      "train_runtime": 50627.1476,
      "train_tokens_per_second": 62341.575
    },
    {
      "epoch": 0.5655776373578913,
      "grad_norm": 22.25,
      "learning_rate": 8.369864226126364e-06,
      "loss": 0.8477,
      "num_input_tokens_seen": 3158190592,
      "step": 42560,
      "train_runtime": 50660.0156,
      "train_tokens_per_second": 62340.893
    },
    {
      "epoch": 0.5657105268403532,
      "grad_norm": 14.375,
      "learning_rate": 8.36561798352855e-06,
      "loss": 0.9153,
      "num_input_tokens_seen": 3159048448,
      "step": 42570,
      "train_runtime": 50673.0035,
      "train_tokens_per_second": 62341.843
    },
    {
      "epoch": 0.5658434163228151,
      "grad_norm": 26.875,
      "learning_rate": 8.361372043685425e-06,
      "loss": 0.7577,
      "num_input_tokens_seen": 3159538432,
      "step": 42580,
      "train_runtime": 50679.9113,
      "train_tokens_per_second": 62343.014
    },
    {
      "epoch": 0.565976305805277,
      "grad_norm": 11.3125,
      "learning_rate": 8.357126407383524e-06,
      "loss": 0.6576,
      "num_input_tokens_seen": 3159771136,
      "step": 42590,
      "train_runtime": 50683.4106,
      "train_tokens_per_second": 62343.301
    },
    {
      "epoch": 0.566109195287739,
      "grad_norm": 27.5,
      "learning_rate": 8.352881075409306e-06,
      "loss": 0.5237,
      "num_input_tokens_seen": 3159887104,
      "step": 42600,
      "train_runtime": 50686.4402,
      "train_tokens_per_second": 62341.863
    },
    {
      "epoch": 0.5662420847702009,
      "grad_norm": 29.5,
      "learning_rate": 8.34863604854918e-06,
      "loss": 0.8053,
      "num_input_tokens_seen": 3161934848,
      "step": 42610,
      "train_runtime": 50718.6148,
      "train_tokens_per_second": 62342.689
    },
    {
      "epoch": 0.5663749742526628,
      "grad_norm": 84.0,
      "learning_rate": 8.344391327589502e-06,
      "loss": 0.9059,
      "num_input_tokens_seen": 3162754304,
      "step": 42620,
      "train_runtime": 50731.1166,
      "train_tokens_per_second": 62343.479
    },
    {
      "epoch": 0.5665078637351246,
      "grad_norm": 13.5625,
      "learning_rate": 8.340146913316564e-06,
      "loss": 0.7499,
      "num_input_tokens_seen": 3163223552,
      "step": 42630,
      "train_runtime": 50737.9151,
      "train_tokens_per_second": 62344.374
    },
    {
      "epoch": 0.5666407532175866,
      "grad_norm": 13.75,
      "learning_rate": 8.335902806516613e-06,
      "loss": 0.6467,
      "num_input_tokens_seen": 3163462144,
      "step": 42640,
      "train_runtime": 50741.4536,
      "train_tokens_per_second": 62344.728
    },
    {
      "epoch": 0.5667736427000485,
      "grad_norm": 18.125,
      "learning_rate": 8.331659007975828e-06,
      "loss": 0.5297,
      "num_input_tokens_seen": 3163580160,
      "step": 42650,
      "train_runtime": 50743.9746,
      "train_tokens_per_second": 62343.957
    },
    {
      "epoch": 0.5669065321825104,
      "grad_norm": 21.5,
      "learning_rate": 8.327415518480338e-06,
      "loss": 0.8234,
      "num_input_tokens_seen": 3165701888,
      "step": 42660,
      "train_runtime": 50777.7144,
      "train_tokens_per_second": 62344.316
    },
    {
      "epoch": 0.5670394216649723,
      "grad_norm": 19.125,
      "learning_rate": 8.323172338816203e-06,
      "loss": 0.8626,
      "num_input_tokens_seen": 3166569472,
      "step": 42670,
      "train_runtime": 50790.4588,
      "train_tokens_per_second": 62345.754
    },
    {
      "epoch": 0.5671723111474343,
      "grad_norm": 14.3125,
      "learning_rate": 8.318929469769445e-06,
      "loss": 0.7511,
      "num_input_tokens_seen": 3167059456,
      "step": 42680,
      "train_runtime": 50797.3047,
      "train_tokens_per_second": 62346.998
    },
    {
      "epoch": 0.5673052006298962,
      "grad_norm": 22.25,
      "learning_rate": 8.314686912126013e-06,
      "loss": 0.6746,
      "num_input_tokens_seen": 3167294208,
      "step": 42690,
      "train_runtime": 50800.838,
      "train_tokens_per_second": 62347.283
    },
    {
      "epoch": 0.567438090112358,
      "grad_norm": 10.625,
      "learning_rate": 8.310444666671799e-06,
      "loss": 0.5607,
      "num_input_tokens_seen": 3167410432,
      "step": 42700,
      "train_runtime": 50803.291,
      "train_tokens_per_second": 62346.56
    },
    {
      "epoch": 0.56757097959482,
      "grad_norm": 16.875,
      "learning_rate": 8.306202734192648e-06,
      "loss": 0.8091,
      "num_input_tokens_seen": 3169386240,
      "step": 42710,
      "train_runtime": 50835.4982,
      "train_tokens_per_second": 62345.927
    },
    {
      "epoch": 0.5677038690772819,
      "grad_norm": 17.75,
      "learning_rate": 8.301961115474338e-06,
      "loss": 0.8932,
      "num_input_tokens_seen": 3170234112,
      "step": 42720,
      "train_runtime": 50847.5931,
      "train_tokens_per_second": 62347.771
    },
    {
      "epoch": 0.5678367585597438,
      "grad_norm": 45.5,
      "learning_rate": 8.297719811302589e-06,
      "loss": 0.7242,
      "num_input_tokens_seen": 3170684928,
      "step": 42730,
      "train_runtime": 50854.0185,
      "train_tokens_per_second": 62348.759
    },
    {
      "epoch": 0.5679696480422057,
      "grad_norm": 19.375,
      "learning_rate": 8.293478822463064e-06,
      "loss": 0.5901,
      "num_input_tokens_seen": 3170910208,
      "step": 42740,
      "train_runtime": 50857.4037,
      "train_tokens_per_second": 62349.038
    },
    {
      "epoch": 0.5681025375246677,
      "grad_norm": 22.625,
      "learning_rate": 8.289238149741374e-06,
      "loss": 0.5304,
      "num_input_tokens_seen": 3171023616,
      "step": 42750,
      "train_runtime": 50859.7449,
      "train_tokens_per_second": 62348.398
    },
    {
      "epoch": 0.5682354270071295,
      "grad_norm": 32.0,
      "learning_rate": 8.284997793923061e-06,
      "loss": 0.7641,
      "num_input_tokens_seen": 3173084160,
      "step": 42760,
      "train_runtime": 50891.9744,
      "train_tokens_per_second": 62349.402
    },
    {
      "epoch": 0.5683683164895914,
      "grad_norm": 19.75,
      "learning_rate": 8.280757755793614e-06,
      "loss": 0.9248,
      "num_input_tokens_seen": 3173936640,
      "step": 42770,
      "train_runtime": 50904.4862,
      "train_tokens_per_second": 62350.824
    },
    {
      "epoch": 0.5685012059720533,
      "grad_norm": 23.0,
      "learning_rate": 8.276518036138463e-06,
      "loss": 0.6706,
      "num_input_tokens_seen": 3174403328,
      "step": 42780,
      "train_runtime": 50911.2506,
      "train_tokens_per_second": 62351.706
    },
    {
      "epoch": 0.5686340954545153,
      "grad_norm": 14.25,
      "learning_rate": 8.272278635742976e-06,
      "loss": 0.7471,
      "num_input_tokens_seen": 3174634752,
      "step": 42790,
      "train_runtime": 50914.7924,
      "train_tokens_per_second": 62351.914
    },
    {
      "epoch": 0.5687669849369772,
      "grad_norm": 12.75,
      "learning_rate": 8.268039555392468e-06,
      "loss": 0.5374,
      "num_input_tokens_seen": 3174749952,
      "step": 42800,
      "train_runtime": 50917.1507,
      "train_tokens_per_second": 62351.288
    },
    {
      "epoch": 0.5688998744194391,
      "grad_norm": 29.125,
      "learning_rate": 8.263800795872186e-06,
      "loss": 0.8307,
      "num_input_tokens_seen": 3176764928,
      "step": 42810,
      "train_runtime": 50950.849,
      "train_tokens_per_second": 62349.598
    },
    {
      "epoch": 0.5690327639019009,
      "grad_norm": 48.0,
      "learning_rate": 8.259562357967329e-06,
      "loss": 0.7759,
      "num_input_tokens_seen": 3177561856,
      "step": 42820,
      "train_runtime": 50963.0386,
      "train_tokens_per_second": 62350.322
    },
    {
      "epoch": 0.5691656533843629,
      "grad_norm": 20.5,
      "learning_rate": 8.255324242463021e-06,
      "loss": 0.7091,
      "num_input_tokens_seen": 3178016000,
      "step": 42830,
      "train_runtime": 50970.1223,
      "train_tokens_per_second": 62350.567
    },
    {
      "epoch": 0.5692985428668248,
      "grad_norm": 18.0,
      "learning_rate": 8.251086450144345e-06,
      "loss": 0.6424,
      "num_input_tokens_seen": 3178243840,
      "step": 42840,
      "train_runtime": 50973.5195,
      "train_tokens_per_second": 62350.881
    },
    {
      "epoch": 0.5694314323492867,
      "grad_norm": 14.0,
      "learning_rate": 8.24684898179631e-06,
      "loss": 0.5455,
      "num_input_tokens_seen": 3178357504,
      "step": 42850,
      "train_runtime": 50975.8943,
      "train_tokens_per_second": 62350.206
    },
    {
      "epoch": 0.5695643218317487,
      "grad_norm": 23.875,
      "learning_rate": 8.242611838203862e-06,
      "loss": 0.8386,
      "num_input_tokens_seen": 3180392448,
      "step": 42860,
      "train_runtime": 51007.8873,
      "train_tokens_per_second": 62350.993
    },
    {
      "epoch": 0.5696972113142106,
      "grad_norm": 36.0,
      "learning_rate": 8.238375020151907e-06,
      "loss": 0.8063,
      "num_input_tokens_seen": 3181199616,
      "step": 42870,
      "train_runtime": 51020.2638,
      "train_tokens_per_second": 62351.689
    },
    {
      "epoch": 0.5698301007966724,
      "grad_norm": 18.25,
      "learning_rate": 8.234138528425267e-06,
      "loss": 0.6919,
      "num_input_tokens_seen": 3181646336,
      "step": 42880,
      "train_runtime": 51027.5187,
      "train_tokens_per_second": 62351.578
    },
    {
      "epoch": 0.5699629902791343,
      "grad_norm": 13.625,
      "learning_rate": 8.229902363808722e-06,
      "loss": 0.5622,
      "num_input_tokens_seen": 3181871616,
      "step": 42890,
      "train_runtime": 51030.903,
      "train_tokens_per_second": 62351.858
    },
    {
      "epoch": 0.5700958797615963,
      "grad_norm": 6.875,
      "learning_rate": 8.225666527086977e-06,
      "loss": 0.4749,
      "num_input_tokens_seen": 3181982208,
      "step": 42900,
      "train_runtime": 51033.2583,
      "train_tokens_per_second": 62351.147
    },
    {
      "epoch": 0.5702287692440582,
      "grad_norm": 17.625,
      "learning_rate": 8.22143101904469e-06,
      "loss": 0.8457,
      "num_input_tokens_seen": 3183893760,
      "step": 42910,
      "train_runtime": 51064.5001,
      "train_tokens_per_second": 62350.434
    },
    {
      "epoch": 0.5703616587265201,
      "grad_norm": 28.25,
      "learning_rate": 8.217195840466448e-06,
      "loss": 0.8731,
      "num_input_tokens_seen": 3184711168,
      "step": 42920,
      "train_runtime": 51076.7339,
      "train_tokens_per_second": 62351.504
    },
    {
      "epoch": 0.570494548208982,
      "grad_norm": 40.75,
      "learning_rate": 8.212960992136775e-06,
      "loss": 0.7048,
      "num_input_tokens_seen": 3185140224,
      "step": 42930,
      "train_runtime": 51083.5019,
      "train_tokens_per_second": 62351.642
    },
    {
      "epoch": 0.5706274376914439,
      "grad_norm": 15.4375,
      "learning_rate": 8.208726474840149e-06,
      "loss": 0.5885,
      "num_input_tokens_seen": 3185368320,
      "step": 42940,
      "train_runtime": 51086.9256,
      "train_tokens_per_second": 62351.928
    },
    {
      "epoch": 0.5707603271739058,
      "grad_norm": 10.9375,
      "learning_rate": 8.204492289360965e-06,
      "loss": 0.572,
      "num_input_tokens_seen": 3185481472,
      "step": 42950,
      "train_runtime": 51089.2916,
      "train_tokens_per_second": 62351.255
    },
    {
      "epoch": 0.5708932166563677,
      "grad_norm": 26.25,
      "learning_rate": 8.20025843648358e-06,
      "loss": 0.757,
      "num_input_tokens_seen": 3187548160,
      "step": 42960,
      "train_runtime": 51124.009,
      "train_tokens_per_second": 62349.339
    },
    {
      "epoch": 0.5710261061388296,
      "grad_norm": 12.375,
      "learning_rate": 8.196024916992268e-06,
      "loss": 0.8261,
      "num_input_tokens_seen": 3188350976,
      "step": 42970,
      "train_runtime": 51136.1479,
      "train_tokens_per_second": 62350.238
    },
    {
      "epoch": 0.5711589956212916,
      "grad_norm": 19.75,
      "learning_rate": 8.191791731671256e-06,
      "loss": 0.6748,
      "num_input_tokens_seen": 3188790272,
      "step": 42980,
      "train_runtime": 51142.3433,
      "train_tokens_per_second": 62351.274
    },
    {
      "epoch": 0.5712918851037535,
      "grad_norm": 11.5625,
      "learning_rate": 8.1875588813047e-06,
      "loss": 0.6296,
      "num_input_tokens_seen": 3189009152,
      "step": 42990,
      "train_runtime": 51145.63,
      "train_tokens_per_second": 62351.547
    },
    {
      "epoch": 0.5714247745862153,
      "grad_norm": 80.0,
      "learning_rate": 8.183326366676703e-06,
      "loss": 0.5016,
      "num_input_tokens_seen": 3189123584,
      "step": 43000,
      "train_runtime": 51148.6457,
      "train_tokens_per_second": 62350.108
    },
    {
      "epoch": 0.5715576640686773,
      "grad_norm": 53.0,
      "learning_rate": 8.179094188571297e-06,
      "loss": 0.8498,
      "num_input_tokens_seen": 3191253504,
      "step": 43010,
      "train_runtime": 51183.7577,
      "train_tokens_per_second": 62348.949
    },
    {
      "epoch": 0.5716905535511392,
      "grad_norm": 11.5625,
      "learning_rate": 8.174862347772454e-06,
      "loss": 0.8928,
      "num_input_tokens_seen": 3192101888,
      "step": 43020,
      "train_runtime": 51196.5072,
      "train_tokens_per_second": 62349.993
    },
    {
      "epoch": 0.5718234430336011,
      "grad_norm": 24.125,
      "learning_rate": 8.17063084506409e-06,
      "loss": 0.6848,
      "num_input_tokens_seen": 3192555008,
      "step": 43030,
      "train_runtime": 51203.096,
      "train_tokens_per_second": 62350.82
    },
    {
      "epoch": 0.571956332516063,
      "grad_norm": 14.6875,
      "learning_rate": 8.166399681230043e-06,
      "loss": 0.6324,
      "num_input_tokens_seen": 3192788224,
      "step": 43040,
      "train_runtime": 51206.5465,
      "train_tokens_per_second": 62351.173
    },
    {
      "epoch": 0.572089221998525,
      "grad_norm": 26.0,
      "learning_rate": 8.162168857054111e-06,
      "loss": 0.522,
      "num_input_tokens_seen": 3192904704,
      "step": 43050,
      "train_runtime": 51209.7225,
      "train_tokens_per_second": 62349.58
    },
    {
      "epoch": 0.5722221114809868,
      "grad_norm": 30.75,
      "learning_rate": 8.157938373320006e-06,
      "loss": 0.8181,
      "num_input_tokens_seen": 3194960896,
      "step": 43060,
      "train_runtime": 51242.8227,
      "train_tokens_per_second": 62349.432
    },
    {
      "epoch": 0.5723550009634487,
      "grad_norm": 56.0,
      "learning_rate": 8.153708230811397e-06,
      "loss": 0.8494,
      "num_input_tokens_seen": 3195796736,
      "step": 43070,
      "train_runtime": 51255.51,
      "train_tokens_per_second": 62350.306
    },
    {
      "epoch": 0.5724878904459106,
      "grad_norm": 18.25,
      "learning_rate": 8.149478430311874e-06,
      "loss": 0.8232,
      "num_input_tokens_seen": 3196283904,
      "step": 43080,
      "train_runtime": 51262.4653,
      "train_tokens_per_second": 62351.35
    },
    {
      "epoch": 0.5726207799283726,
      "grad_norm": 14.625,
      "learning_rate": 8.145248972604971e-06,
      "loss": 0.6551,
      "num_input_tokens_seen": 3196517120,
      "step": 43090,
      "train_runtime": 51266.0126,
      "train_tokens_per_second": 62351.585
    },
    {
      "epoch": 0.5727536694108345,
      "grad_norm": 18.5,
      "learning_rate": 8.14101985847416e-06,
      "loss": 0.4821,
      "num_input_tokens_seen": 3196635136,
      "step": 43100,
      "train_runtime": 51268.4011,
      "train_tokens_per_second": 62350.982
    },
    {
      "epoch": 0.5728865588932964,
      "grad_norm": 30.75,
      "learning_rate": 8.136791088702835e-06,
      "loss": 0.8163,
      "num_input_tokens_seen": 3198629120,
      "step": 43110,
      "train_runtime": 51300.3105,
      "train_tokens_per_second": 62351.067
    },
    {
      "epoch": 0.5730194483757584,
      "grad_norm": 14.375,
      "learning_rate": 8.132562664074354e-06,
      "loss": 0.8615,
      "num_input_tokens_seen": 3199462144,
      "step": 43120,
      "train_runtime": 51312.7,
      "train_tokens_per_second": 62352.247
    },
    {
      "epoch": 0.5731523378582202,
      "grad_norm": 22.125,
      "learning_rate": 8.128334585371986e-06,
      "loss": 0.6729,
      "num_input_tokens_seen": 3199916032,
      "step": 43130,
      "train_runtime": 51319.6364,
      "train_tokens_per_second": 62352.664
    },
    {
      "epoch": 0.5732852273406821,
      "grad_norm": 18.875,
      "learning_rate": 8.124106853378949e-06,
      "loss": 0.6341,
      "num_input_tokens_seen": 3200132352,
      "step": 43140,
      "train_runtime": 51322.9268,
      "train_tokens_per_second": 62352.881
    },
    {
      "epoch": 0.573418116823144,
      "grad_norm": 11.1875,
      "learning_rate": 8.119879468878385e-06,
      "loss": 0.533,
      "num_input_tokens_seen": 3200244736,
      "step": 43150,
      "train_runtime": 51325.2649,
      "train_tokens_per_second": 62352.23
    },
    {
      "epoch": 0.573551006305606,
      "grad_norm": 20.625,
      "learning_rate": 8.11565243265339e-06,
      "loss": 0.8988,
      "num_input_tokens_seen": 3202122496,
      "step": 43160,
      "train_runtime": 51356.3377,
      "train_tokens_per_second": 62351.068
    },
    {
      "epoch": 0.5736838957880679,
      "grad_norm": 17.625,
      "learning_rate": 8.11142574548698e-06,
      "loss": 0.9001,
      "num_input_tokens_seen": 3202953216,
      "step": 43170,
      "train_runtime": 51368.7585,
      "train_tokens_per_second": 62352.163
    },
    {
      "epoch": 0.5738167852705298,
      "grad_norm": 33.75,
      "learning_rate": 8.107199408162107e-06,
      "loss": 0.8031,
      "num_input_tokens_seen": 3203398400,
      "step": 43180,
      "train_runtime": 51375.3685,
      "train_tokens_per_second": 62352.806
    },
    {
      "epoch": 0.5739496747529916,
      "grad_norm": 27.125,
      "learning_rate": 8.102973421461668e-06,
      "loss": 0.6851,
      "num_input_tokens_seen": 3203640320,
      "step": 43190,
      "train_runtime": 51379.5459,
      "train_tokens_per_second": 62352.445
    },
    {
      "epoch": 0.5740825642354536,
      "grad_norm": 11.5,
      "learning_rate": 8.098747786168485e-06,
      "loss": 0.4879,
      "num_input_tokens_seen": 3203756032,
      "step": 43200,
      "train_runtime": 51381.9201,
      "train_tokens_per_second": 62351.816
    },
    {
      "epoch": 0.5742154537179155,
      "grad_norm": 19.625,
      "learning_rate": 8.094522503065323e-06,
      "loss": 0.8566,
      "num_input_tokens_seen": 3205684992,
      "step": 43210,
      "train_runtime": 51413.0825,
      "train_tokens_per_second": 62351.542
    },
    {
      "epoch": 0.5743483432003774,
      "grad_norm": 33.5,
      "learning_rate": 8.090297572934872e-06,
      "loss": 0.8716,
      "num_input_tokens_seen": 3206530560,
      "step": 43220,
      "train_runtime": 51425.7313,
      "train_tokens_per_second": 62352.649
    },
    {
      "epoch": 0.5744812326828393,
      "grad_norm": 11.125,
      "learning_rate": 8.086072996559772e-06,
      "loss": 0.7608,
      "num_input_tokens_seen": 3206998784,
      "step": 43230,
      "train_runtime": 51432.6048,
      "train_tokens_per_second": 62353.419
    },
    {
      "epoch": 0.5746141221653013,
      "grad_norm": 13.1875,
      "learning_rate": 8.081848774722582e-06,
      "loss": 0.7269,
      "num_input_tokens_seen": 3207235584,
      "step": 43240,
      "train_runtime": 51436.8152,
      "train_tokens_per_second": 62352.919
    },
    {
      "epoch": 0.5747470116477631,
      "grad_norm": 16.625,
      "learning_rate": 8.077624908205805e-06,
      "loss": 0.5282,
      "num_input_tokens_seen": 3207353856,
      "step": 43250,
      "train_runtime": 51439.2082,
      "train_tokens_per_second": 62352.318
    },
    {
      "epoch": 0.574879901130225,
      "grad_norm": 16.875,
      "learning_rate": 8.073401397791872e-06,
      "loss": 0.8938,
      "num_input_tokens_seen": 3209366528,
      "step": 43260,
      "train_runtime": 51470.53,
      "train_tokens_per_second": 62353.477
    },
    {
      "epoch": 0.575012790612687,
      "grad_norm": 17.5,
      "learning_rate": 8.069178244263147e-06,
      "loss": 0.8394,
      "num_input_tokens_seen": 3210193920,
      "step": 43270,
      "train_runtime": 51483.323,
      "train_tokens_per_second": 62354.054
    },
    {
      "epoch": 0.5751456800951489,
      "grad_norm": 22.625,
      "learning_rate": 8.06495544840194e-06,
      "loss": 0.7552,
      "num_input_tokens_seen": 3210654464,
      "step": 43280,
      "train_runtime": 51490.7395,
      "train_tokens_per_second": 62354.017
    },
    {
      "epoch": 0.5752785695776108,
      "grad_norm": 35.0,
      "learning_rate": 8.060733010990479e-06,
      "loss": 0.6587,
      "num_input_tokens_seen": 3210881792,
      "step": 43290,
      "train_runtime": 51494.2013,
      "train_tokens_per_second": 62354.24
    },
    {
      "epoch": 0.5754114590600727,
      "grad_norm": 9.0625,
      "learning_rate": 8.056510932810936e-06,
      "loss": 0.5317,
      "num_input_tokens_seen": 3210999552,
      "step": 43300,
      "train_runtime": 51496.5808,
      "train_tokens_per_second": 62353.646
    },
    {
      "epoch": 0.5755443485425346,
      "grad_norm": 15.0,
      "learning_rate": 8.05228921464541e-06,
      "loss": 0.7863,
      "num_input_tokens_seen": 3213008640,
      "step": 43310,
      "train_runtime": 51527.7215,
      "train_tokens_per_second": 62354.953
    },
    {
      "epoch": 0.5756772380249965,
      "grad_norm": 44.0,
      "learning_rate": 8.048067857275944e-06,
      "loss": 0.9246,
      "num_input_tokens_seen": 3213850624,
      "step": 43320,
      "train_runtime": 51540.3037,
      "train_tokens_per_second": 62356.067
    },
    {
      "epoch": 0.5758101275074584,
      "grad_norm": 24.625,
      "learning_rate": 8.043846861484502e-06,
      "loss": 0.7255,
      "num_input_tokens_seen": 3214308352,
      "step": 43330,
      "train_runtime": 51546.9814,
      "train_tokens_per_second": 62356.869
    },
    {
      "epoch": 0.5759430169899203,
      "grad_norm": 8.8125,
      "learning_rate": 8.039626228052985e-06,
      "loss": 0.5949,
      "num_input_tokens_seen": 3214543104,
      "step": 43340,
      "train_runtime": 51550.5122,
      "train_tokens_per_second": 62357.152
    },
    {
      "epoch": 0.5760759064723823,
      "grad_norm": 26.25,
      "learning_rate": 8.03540595776323e-06,
      "loss": 0.4771,
      "num_input_tokens_seen": 3214660096,
      "step": 43350,
      "train_runtime": 51552.8994,
      "train_tokens_per_second": 62356.533
    },
    {
      "epoch": 0.5762087959548442,
      "grad_norm": 20.75,
      "learning_rate": 8.031186051396999e-06,
      "loss": 0.8688,
      "num_input_tokens_seen": 3216711168,
      "step": 43360,
      "train_runtime": 51587.1723,
      "train_tokens_per_second": 62354.865
    },
    {
      "epoch": 0.576341685437306,
      "grad_norm": 26.75,
      "learning_rate": 8.026966509736001e-06,
      "loss": 0.8666,
      "num_input_tokens_seen": 3217555456,
      "step": 43370,
      "train_runtime": 51599.863,
      "train_tokens_per_second": 62355.891
    },
    {
      "epoch": 0.576474574919768,
      "grad_norm": 37.75,
      "learning_rate": 8.02274733356186e-06,
      "loss": 0.7318,
      "num_input_tokens_seen": 3218026496,
      "step": 43380,
      "train_runtime": 51606.5459,
      "train_tokens_per_second": 62356.944
    },
    {
      "epoch": 0.5766074644022299,
      "grad_norm": 30.375,
      "learning_rate": 8.018528523656149e-06,
      "loss": 0.7217,
      "num_input_tokens_seen": 3218262784,
      "step": 43390,
      "train_runtime": 51610.0552,
      "train_tokens_per_second": 62357.282
    },
    {
      "epoch": 0.5767403538846918,
      "grad_norm": 13.3125,
      "learning_rate": 8.014310080800358e-06,
      "loss": 0.5473,
      "num_input_tokens_seen": 3218379520,
      "step": 43400,
      "train_runtime": 51612.411,
      "train_tokens_per_second": 62356.698
    },
    {
      "epoch": 0.5768732433671537,
      "grad_norm": 16.0,
      "learning_rate": 8.01009200577592e-06,
      "loss": 0.8067,
      "num_input_tokens_seen": 3220327424,
      "step": 43410,
      "train_runtime": 51644.0803,
      "train_tokens_per_second": 62356.177
    },
    {
      "epoch": 0.5770061328496157,
      "grad_norm": 28.125,
      "learning_rate": 8.005874299364196e-06,
      "loss": 0.9137,
      "num_input_tokens_seen": 3221120512,
      "step": 43420,
      "train_runtime": 51656.8066,
      "train_tokens_per_second": 62356.168
    },
    {
      "epoch": 0.5771390223320775,
      "grad_norm": 12.3125,
      "learning_rate": 8.00165696234647e-06,
      "loss": 0.7113,
      "num_input_tokens_seen": 3221574144,
      "step": 43430,
      "train_runtime": 51663.2581,
      "train_tokens_per_second": 62357.162
    },
    {
      "epoch": 0.5772719118145394,
      "grad_norm": 13.1875,
      "learning_rate": 7.99743999550398e-06,
      "loss": 0.6464,
      "num_input_tokens_seen": 3221798912,
      "step": 43440,
      "train_runtime": 51666.649,
      "train_tokens_per_second": 62357.42
    },
    {
      "epoch": 0.5774048012970013,
      "grad_norm": 6.3125,
      "learning_rate": 7.99322339961787e-06,
      "loss": 0.495,
      "num_input_tokens_seen": 3221914368,
      "step": 43450,
      "train_runtime": 51668.9996,
      "train_tokens_per_second": 62356.817
    },
    {
      "epoch": 0.5775376907794633,
      "grad_norm": 29.75,
      "learning_rate": 7.989007175469233e-06,
      "loss": 0.8284,
      "num_input_tokens_seen": 3224055296,
      "step": 43460,
      "train_runtime": 51701.8231,
      "train_tokens_per_second": 62358.639
    },
    {
      "epoch": 0.5776705802619252,
      "grad_norm": 15.9375,
      "learning_rate": 7.984791323839082e-06,
      "loss": 0.8354,
      "num_input_tokens_seen": 3224925952,
      "step": 43470,
      "train_runtime": 51715.2572,
      "train_tokens_per_second": 62359.275
    },
    {
      "epoch": 0.5778034697443871,
      "grad_norm": 21.625,
      "learning_rate": 7.980575845508374e-06,
      "loss": 0.7288,
      "num_input_tokens_seen": 3225435648,
      "step": 43480,
      "train_runtime": 51722.0657,
      "train_tokens_per_second": 62360.921
    },
    {
      "epoch": 0.5779363592268489,
      "grad_norm": 19.625,
      "learning_rate": 7.976360741257983e-06,
      "loss": 0.6439,
      "num_input_tokens_seen": 3225670144,
      "step": 43490,
      "train_runtime": 51725.6313,
      "train_tokens_per_second": 62361.156
    },
    {
      "epoch": 0.5780692487093109,
      "grad_norm": 60.75,
      "learning_rate": 7.972146011868717e-06,
      "loss": 0.5459,
      "num_input_tokens_seen": 3225788928,
      "step": 43500,
      "train_runtime": 51727.9999,
      "train_tokens_per_second": 62360.596
    },
    {
      "epoch": 0.5782021381917728,
      "grad_norm": 37.5,
      "learning_rate": 7.967931658121322e-06,
      "loss": 0.8356,
      "num_input_tokens_seen": 3227887872,
      "step": 43510,
      "train_runtime": 51761.5319,
      "train_tokens_per_second": 62360.748
    },
    {
      "epoch": 0.5783350276742347,
      "grad_norm": 11.5625,
      "learning_rate": 7.963717680796466e-06,
      "loss": 0.8281,
      "num_input_tokens_seen": 3228756480,
      "step": 43520,
      "train_runtime": 51774.4384,
      "train_tokens_per_second": 62361.98
    },
    {
      "epoch": 0.5784679171566967,
      "grad_norm": 34.0,
      "learning_rate": 7.959504080674755e-06,
      "loss": 0.702,
      "num_input_tokens_seen": 3229238272,
      "step": 43530,
      "train_runtime": 51781.1764,
      "train_tokens_per_second": 62363.169
    },
    {
      "epoch": 0.5786008066391586,
      "grad_norm": 13.25,
      "learning_rate": 7.955290858536714e-06,
      "loss": 0.6406,
      "num_input_tokens_seen": 3229472768,
      "step": 43540,
      "train_runtime": 51784.7033,
      "train_tokens_per_second": 62363.45
    },
    {
      "epoch": 0.5787336961216204,
      "grad_norm": 13.25,
      "learning_rate": 7.951078015162814e-06,
      "loss": 0.5659,
      "num_input_tokens_seen": 3229587968,
      "step": 43550,
      "train_runtime": 51787.0339,
      "train_tokens_per_second": 62362.868
    },
    {
      "epoch": 0.5788665856040823,
      "grad_norm": 27.5,
      "learning_rate": 7.94686555133344e-06,
      "loss": 0.7987,
      "num_input_tokens_seen": 3231693056,
      "step": 43560,
      "train_runtime": 51821.0636,
      "train_tokens_per_second": 62362.538
    },
    {
      "epoch": 0.5789994750865443,
      "grad_norm": 37.5,
      "learning_rate": 7.942653467828915e-06,
      "loss": 0.9033,
      "num_input_tokens_seen": 3232548096,
      "step": 43570,
      "train_runtime": 51834.0402,
      "train_tokens_per_second": 62363.421
    },
    {
      "epoch": 0.5791323645690062,
      "grad_norm": 25.125,
      "learning_rate": 7.938441765429492e-06,
      "loss": 0.8033,
      "num_input_tokens_seen": 3233072896,
      "step": 43580,
      "train_runtime": 51841.2517,
      "train_tokens_per_second": 62364.869
    },
    {
      "epoch": 0.5792652540514681,
      "grad_norm": 23.25,
      "learning_rate": 7.934230444915345e-06,
      "loss": 0.6199,
      "num_input_tokens_seen": 3233313792,
      "step": 43590,
      "train_runtime": 51845.7295,
      "train_tokens_per_second": 62364.13
    },
    {
      "epoch": 0.57939814353393,
      "grad_norm": 6.03125,
      "learning_rate": 7.930019507066591e-06,
      "loss": 0.4621,
      "num_input_tokens_seen": 3233433600,
      "step": 43600,
      "train_runtime": 51848.1495,
      "train_tokens_per_second": 62363.529
    },
    {
      "epoch": 0.579531033016392,
      "grad_norm": 30.875,
      "learning_rate": 7.925808952663265e-06,
      "loss": 0.796,
      "num_input_tokens_seen": 3235340288,
      "step": 43610,
      "train_runtime": 51879.3873,
      "train_tokens_per_second": 62362.731
    },
    {
      "epoch": 0.5796639224988538,
      "grad_norm": 19.0,
      "learning_rate": 7.921598782485339e-06,
      "loss": 0.8639,
      "num_input_tokens_seen": 3236162560,
      "step": 43620,
      "train_runtime": 51891.7407,
      "train_tokens_per_second": 62363.731
    },
    {
      "epoch": 0.5797968119813157,
      "grad_norm": 39.75,
      "learning_rate": 7.917388997312698e-06,
      "loss": 0.7715,
      "num_input_tokens_seen": 3236599552,
      "step": 43630,
      "train_runtime": 51898.0691,
      "train_tokens_per_second": 62364.547
    },
    {
      "epoch": 0.5799297014637776,
      "grad_norm": 17.5,
      "learning_rate": 7.91317959792518e-06,
      "loss": 0.6589,
      "num_input_tokens_seen": 3236824064,
      "step": 43640,
      "train_runtime": 51901.4926,
      "train_tokens_per_second": 62364.759
    },
    {
      "epoch": 0.5800625909462396,
      "grad_norm": 12.8125,
      "learning_rate": 7.908970585102536e-06,
      "loss": 0.5099,
      "num_input_tokens_seen": 3236938240,
      "step": 43650,
      "train_runtime": 51904.5603,
      "train_tokens_per_second": 62363.273
    },
    {
      "epoch": 0.5801954804287015,
      "grad_norm": 28.125,
      "learning_rate": 7.904761959624442e-06,
      "loss": 0.8621,
      "num_input_tokens_seen": 3239006720,
      "step": 43660,
      "train_runtime": 51937.9086,
      "train_tokens_per_second": 62363.056
    },
    {
      "epoch": 0.5803283699111634,
      "grad_norm": 22.75,
      "learning_rate": 7.900553722270513e-06,
      "loss": 0.8718,
      "num_input_tokens_seen": 3239846912,
      "step": 43670,
      "train_runtime": 51950.4096,
      "train_tokens_per_second": 62364.223
    },
    {
      "epoch": 0.5804612593936253,
      "grad_norm": 13.9375,
      "learning_rate": 7.896345873820284e-06,
      "loss": 0.7418,
      "num_input_tokens_seen": 3240314880,
      "step": 43680,
      "train_runtime": 51957.0046,
      "train_tokens_per_second": 62365.313
    },
    {
      "epoch": 0.5805941488760872,
      "grad_norm": 23.625,
      "learning_rate": 7.892138415053228e-06,
      "loss": 0.6134,
      "num_input_tokens_seen": 3240552960,
      "step": 43690,
      "train_runtime": 51960.5506,
      "train_tokens_per_second": 62365.639
    },
    {
      "epoch": 0.5807270383585491,
      "grad_norm": 11.625,
      "learning_rate": 7.887931346748733e-06,
      "loss": 0.5656,
      "num_input_tokens_seen": 3240670720,
      "step": 43700,
      "train_runtime": 51963.6319,
      "train_tokens_per_second": 62364.207
    },
    {
      "epoch": 0.580859927841011,
      "grad_norm": 28.125,
      "learning_rate": 7.883724669686125e-06,
      "loss": 0.8138,
      "num_input_tokens_seen": 3242683904,
      "step": 43710,
      "train_runtime": 51996.6145,
      "train_tokens_per_second": 62363.366
    },
    {
      "epoch": 0.580992817323473,
      "grad_norm": 11.5625,
      "learning_rate": 7.879518384644647e-06,
      "loss": 0.9225,
      "num_input_tokens_seen": 3243513600,
      "step": 43720,
      "train_runtime": 52008.9012,
      "train_tokens_per_second": 62364.586
    },
    {
      "epoch": 0.5811257068059349,
      "grad_norm": 30.25,
      "learning_rate": 7.875312492403485e-06,
      "loss": 0.7243,
      "num_input_tokens_seen": 3243980288,
      "step": 43730,
      "train_runtime": 52015.5964,
      "train_tokens_per_second": 62365.531
    },
    {
      "epoch": 0.5812585962883967,
      "grad_norm": 314.0,
      "learning_rate": 7.871106993741737e-06,
      "loss": 0.6754,
      "num_input_tokens_seen": 3244220928,
      "step": 43740,
      "train_runtime": 52019.1148,
      "train_tokens_per_second": 62365.939
    },
    {
      "epoch": 0.5813914857708586,
      "grad_norm": 37.0,
      "learning_rate": 7.866901889438432e-06,
      "loss": 0.5918,
      "num_input_tokens_seen": 3244336128,
      "step": 43750,
      "train_runtime": 52022.3066,
      "train_tokens_per_second": 62364.327
    },
    {
      "epoch": 0.5815243752533206,
      "grad_norm": 17.625,
      "learning_rate": 7.862697180272537e-06,
      "loss": 0.8548,
      "num_input_tokens_seen": 3246371840,
      "step": 43760,
      "train_runtime": 52054.0819,
      "train_tokens_per_second": 62365.365
    },
    {
      "epoch": 0.5816572647357825,
      "grad_norm": 14.5625,
      "learning_rate": 7.858492867022927e-06,
      "loss": 0.8683,
      "num_input_tokens_seen": 3247193856,
      "step": 43770,
      "train_runtime": 52066.5097,
      "train_tokens_per_second": 62366.267
    },
    {
      "epoch": 0.5817901542182444,
      "grad_norm": 17.625,
      "learning_rate": 7.85428895046842e-06,
      "loss": 0.7185,
      "num_input_tokens_seen": 3247646208,
      "step": 43780,
      "train_runtime": 52072.9981,
      "train_tokens_per_second": 62367.183
    },
    {
      "epoch": 0.5819230437007064,
      "grad_norm": 18.375,
      "learning_rate": 7.85008543138775e-06,
      "loss": 0.5903,
      "num_input_tokens_seen": 3247878656,
      "step": 43790,
      "train_runtime": 52076.3982,
      "train_tokens_per_second": 62367.575
    },
    {
      "epoch": 0.5820559331831682,
      "grad_norm": 25.375,
      "learning_rate": 7.845882310559587e-06,
      "loss": 0.5501,
      "num_input_tokens_seen": 3247994368,
      "step": 43800,
      "train_runtime": 52078.787,
      "train_tokens_per_second": 62366.936
    },
    {
      "epoch": 0.5821888226656301,
      "grad_norm": 17.625,
      "learning_rate": 7.841679588762516e-06,
      "loss": 0.7685,
      "num_input_tokens_seen": 3250037504,
      "step": 43810,
      "train_runtime": 52110.8049,
      "train_tokens_per_second": 62367.824
    },
    {
      "epoch": 0.582321712148092,
      "grad_norm": 13.0625,
      "learning_rate": 7.837477266775056e-06,
      "loss": 0.8064,
      "num_input_tokens_seen": 3250856448,
      "step": 43820,
      "train_runtime": 52123.6502,
      "train_tokens_per_second": 62368.166
    },
    {
      "epoch": 0.582454601630554,
      "grad_norm": 15.375,
      "learning_rate": 7.83327534537565e-06,
      "loss": 0.7386,
      "num_input_tokens_seen": 3251310848,
      "step": 43830,
      "train_runtime": 52130.2983,
      "train_tokens_per_second": 62368.928
    },
    {
      "epoch": 0.5825874911130159,
      "grad_norm": 14.5,
      "learning_rate": 7.829073825342663e-06,
      "loss": 0.6304,
      "num_input_tokens_seen": 3251538688,
      "step": 43840,
      "train_runtime": 52133.7829,
      "train_tokens_per_second": 62369.13
    },
    {
      "epoch": 0.5827203805954778,
      "grad_norm": 39.75,
      "learning_rate": 7.824872707454396e-06,
      "loss": 0.5024,
      "num_input_tokens_seen": 3251654144,
      "step": 43850,
      "train_runtime": 52136.2671,
      "train_tokens_per_second": 62368.373
    },
    {
      "epoch": 0.5828532700779396,
      "grad_norm": 36.0,
      "learning_rate": 7.820671992489064e-06,
      "loss": 0.7709,
      "num_input_tokens_seen": 3253776896,
      "step": 43860,
      "train_runtime": 52170.7657,
      "train_tokens_per_second": 62367.819
    },
    {
      "epoch": 0.5829861595604016,
      "grad_norm": 22.125,
      "learning_rate": 7.816471681224812e-06,
      "loss": 0.9122,
      "num_input_tokens_seen": 3254623232,
      "step": 43870,
      "train_runtime": 52183.5197,
      "train_tokens_per_second": 62368.795
    },
    {
      "epoch": 0.5831190490428635,
      "grad_norm": 23.125,
      "learning_rate": 7.812271774439708e-06,
      "loss": 0.7889,
      "num_input_tokens_seen": 3255105792,
      "step": 43880,
      "train_runtime": 52190.3668,
      "train_tokens_per_second": 62369.858
    },
    {
      "epoch": 0.5832519385253254,
      "grad_norm": 33.75,
      "learning_rate": 7.808072272911753e-06,
      "loss": 0.6886,
      "num_input_tokens_seen": 3255335936,
      "step": 43890,
      "train_runtime": 52193.8513,
      "train_tokens_per_second": 62370.104
    },
    {
      "epoch": 0.5833848280077873,
      "grad_norm": 36.0,
      "learning_rate": 7.803873177418867e-06,
      "loss": 0.5059,
      "num_input_tokens_seen": 3255454464,
      "step": 43900,
      "train_runtime": 52196.2646,
      "train_tokens_per_second": 62369.491
    },
    {
      "epoch": 0.5835177174902493,
      "grad_norm": 16.75,
      "learning_rate": 7.799674488738886e-06,
      "loss": 0.8042,
      "num_input_tokens_seen": 3257538816,
      "step": 43910,
      "train_runtime": 52229.7342,
      "train_tokens_per_second": 62369.431
    },
    {
      "epoch": 0.5836506069727111,
      "grad_norm": 15.0,
      "learning_rate": 7.795476207649591e-06,
      "loss": 0.8608,
      "num_input_tokens_seen": 3258391296,
      "step": 43920,
      "train_runtime": 52242.6179,
      "train_tokens_per_second": 62370.368
    },
    {
      "epoch": 0.583783496455173,
      "grad_norm": 38.25,
      "learning_rate": 7.79127833492867e-06,
      "loss": 0.763,
      "num_input_tokens_seen": 3258876928,
      "step": 43930,
      "train_runtime": 52249.4587,
      "train_tokens_per_second": 62371.496
    },
    {
      "epoch": 0.583916385937635,
      "grad_norm": 18.625,
      "learning_rate": 7.787080871353742e-06,
      "loss": 0.6469,
      "num_input_tokens_seen": 3259100160,
      "step": 43940,
      "train_runtime": 52253.5773,
      "train_tokens_per_second": 62370.852
    },
    {
      "epoch": 0.5840492754200969,
      "grad_norm": 8.375,
      "learning_rate": 7.78288381770235e-06,
      "loss": 0.7329,
      "num_input_tokens_seen": 3259213568,
      "step": 43950,
      "train_runtime": 52255.9506,
      "train_tokens_per_second": 62370.19
    },
    {
      "epoch": 0.5841821649025588,
      "grad_norm": 91.0,
      "learning_rate": 7.778687174751962e-06,
      "loss": 0.8538,
      "num_input_tokens_seen": 3261240064,
      "step": 43960,
      "train_runtime": 52288.5919,
      "train_tokens_per_second": 62370.011
    },
    {
      "epoch": 0.5843150543850207,
      "grad_norm": 26.25,
      "learning_rate": 7.774490943279969e-06,
      "loss": 0.8974,
      "num_input_tokens_seen": 3262083584,
      "step": 43970,
      "train_runtime": 52301.1582,
      "train_tokens_per_second": 62371.154
    },
    {
      "epoch": 0.5844479438674826,
      "grad_norm": 17.75,
      "learning_rate": 7.770295124063683e-06,
      "loss": 0.6982,
      "num_input_tokens_seen": 3262542080,
      "step": 43980,
      "train_runtime": 52307.6274,
      "train_tokens_per_second": 62372.205
    },
    {
      "epoch": 0.5845808333499445,
      "grad_norm": 19.75,
      "learning_rate": 7.766099717880344e-06,
      "loss": 0.667,
      "num_input_tokens_seen": 3262765568,
      "step": 43990,
      "train_runtime": 52311.7363,
      "train_tokens_per_second": 62371.578
    },
    {
      "epoch": 0.5847137228324064,
      "grad_norm": 21.75,
      "learning_rate": 7.76190472550711e-06,
      "loss": 0.5859,
      "num_input_tokens_seen": 3262882304,
      "step": 44000,
      "train_runtime": 52314.1476,
      "train_tokens_per_second": 62370.935
    },
    {
      "epoch": 0.5848466123148683,
      "grad_norm": 159.0,
      "learning_rate": 7.757710147721071e-06,
      "loss": 0.8184,
      "num_input_tokens_seen": 3265071360,
      "step": 44010,
      "train_runtime": 52347.7386,
      "train_tokens_per_second": 62372.73
    },
    {
      "epoch": 0.5849795017973303,
      "grad_norm": 50.75,
      "learning_rate": 7.753515985299233e-06,
      "loss": 0.9179,
      "num_input_tokens_seen": 3265929728,
      "step": 44020,
      "train_runtime": 52360.3521,
      "train_tokens_per_second": 62374.098
    },
    {
      "epoch": 0.5851123912797922,
      "grad_norm": 16.0,
      "learning_rate": 7.74932223901853e-06,
      "loss": 0.7563,
      "num_input_tokens_seen": 3266397696,
      "step": 44030,
      "train_runtime": 52367.1017,
      "train_tokens_per_second": 62374.995
    },
    {
      "epoch": 0.585245280762254,
      "grad_norm": 9.1875,
      "learning_rate": 7.745128909655807e-06,
      "loss": 0.6227,
      "num_input_tokens_seen": 3266626048,
      "step": 44040,
      "train_runtime": 52370.6128,
      "train_tokens_per_second": 62375.173
    },
    {
      "epoch": 0.585378170244716,
      "grad_norm": 15.875,
      "learning_rate": 7.740935997987852e-06,
      "loss": 0.5048,
      "num_input_tokens_seen": 3266744576,
      "step": 44050,
      "train_runtime": 52373.0007,
      "train_tokens_per_second": 62374.592
    },
    {
      "epoch": 0.5855110597271779,
      "grad_norm": 24.125,
      "learning_rate": 7.73674350479136e-06,
      "loss": 0.8058,
      "num_input_tokens_seen": 3268761600,
      "step": 44060,
      "train_runtime": 52405.7225,
      "train_tokens_per_second": 62374.135
    },
    {
      "epoch": 0.5856439492096398,
      "grad_norm": 17.0,
      "learning_rate": 7.732551430842948e-06,
      "loss": 0.8628,
      "num_input_tokens_seen": 3269595648,
      "step": 44070,
      "train_runtime": 52417.9923,
      "train_tokens_per_second": 62375.446
    },
    {
      "epoch": 0.5857768386921017,
      "grad_norm": 13.1875,
      "learning_rate": 7.72835977691917e-06,
      "loss": 0.699,
      "num_input_tokens_seen": 3270046976,
      "step": 44080,
      "train_runtime": 52424.5282,
      "train_tokens_per_second": 62376.279
    },
    {
      "epoch": 0.5859097281745637,
      "grad_norm": 23.0,
      "learning_rate": 7.724168543796484e-06,
      "loss": 0.5903,
      "num_input_tokens_seen": 3270278144,
      "step": 44090,
      "train_runtime": 52427.9635,
      "train_tokens_per_second": 62376.601
    },
    {
      "epoch": 0.5860426176570256,
      "grad_norm": 12.75,
      "learning_rate": 7.719977732251287e-06,
      "loss": 0.5613,
      "num_input_tokens_seen": 3270390528,
      "step": 44100,
      "train_runtime": 52430.3661,
      "train_tokens_per_second": 62375.886
    },
    {
      "epoch": 0.5861755071394874,
      "grad_norm": 12.9375,
      "learning_rate": 7.715787343059878e-06,
      "loss": 0.7477,
      "num_input_tokens_seen": 3272581376,
      "step": 44110,
      "train_runtime": 52466.5502,
      "train_tokens_per_second": 62374.625
    },
    {
      "epoch": 0.5863083966219493,
      "grad_norm": 32.75,
      "learning_rate": 7.7115973769985e-06,
      "loss": 0.9123,
      "num_input_tokens_seen": 3273459456,
      "step": 44120,
      "train_runtime": 52479.943,
      "train_tokens_per_second": 62375.439
    },
    {
      "epoch": 0.5864412861044113,
      "grad_norm": 24.625,
      "learning_rate": 7.707407834843302e-06,
      "loss": 0.7087,
      "num_input_tokens_seen": 3273946880,
      "step": 44130,
      "train_runtime": 52486.9814,
      "train_tokens_per_second": 62376.361
    },
    {
      "epoch": 0.5865741755868732,
      "grad_norm": 288.0,
      "learning_rate": 7.703218717370359e-06,
      "loss": 0.6196,
      "num_input_tokens_seen": 3274182400,
      "step": 44140,
      "train_runtime": 52490.4709,
      "train_tokens_per_second": 62376.701
    },
    {
      "epoch": 0.5867070650693351,
      "grad_norm": 28.0,
      "learning_rate": 7.69903002535567e-06,
      "loss": 0.5154,
      "num_input_tokens_seen": 3274301184,
      "step": 44150,
      "train_runtime": 52492.9733,
      "train_tokens_per_second": 62375.99
    },
    {
      "epoch": 0.586839954551797,
      "grad_norm": 19.5,
      "learning_rate": 7.694841759575147e-06,
      "loss": 0.8485,
      "num_input_tokens_seen": 3276339200,
      "step": 44160,
      "train_runtime": 52527.0671,
      "train_tokens_per_second": 62374.303
    },
    {
      "epoch": 0.5869728440342589,
      "grad_norm": 17.625,
      "learning_rate": 7.690653920804639e-06,
      "loss": 0.9484,
      "num_input_tokens_seen": 3277150976,
      "step": 44170,
      "train_runtime": 52539.7585,
      "train_tokens_per_second": 62374.687
    },
    {
      "epoch": 0.5871057335167208,
      "grad_norm": 11.0,
      "learning_rate": 7.686466509819897e-06,
      "loss": 0.7431,
      "num_input_tokens_seen": 3277618176,
      "step": 44180,
      "train_runtime": 52547.1888,
      "train_tokens_per_second": 62374.758
    },
    {
      "epoch": 0.5872386229991827,
      "grad_norm": 13.4375,
      "learning_rate": 7.682279527396606e-06,
      "loss": 0.6565,
      "num_input_tokens_seen": 3277848320,
      "step": 44190,
      "train_runtime": 52550.6939,
      "train_tokens_per_second": 62374.977
    },
    {
      "epoch": 0.5873715124816447,
      "grad_norm": 16.75,
      "learning_rate": 7.678092974310363e-06,
      "loss": 0.5454,
      "num_input_tokens_seen": 3277967104,
      "step": 44200,
      "train_runtime": 52553.0724,
      "train_tokens_per_second": 62374.414
    },
    {
      "epoch": 0.5875044019641066,
      "grad_norm": 32.0,
      "learning_rate": 7.673906851336697e-06,
      "loss": 0.8969,
      "num_input_tokens_seen": 3280024576,
      "step": 44210,
      "train_runtime": 52585.3413,
      "train_tokens_per_second": 62375.265
    },
    {
      "epoch": 0.5876372914465685,
      "grad_norm": 15.5625,
      "learning_rate": 7.669721159251045e-06,
      "loss": 0.8545,
      "num_input_tokens_seen": 3280857088,
      "step": 44220,
      "train_runtime": 52597.5966,
      "train_tokens_per_second": 62376.559
    },
    {
      "epoch": 0.5877701809290303,
      "grad_norm": 21.375,
      "learning_rate": 7.665535898828767e-06,
      "loss": 0.7144,
      "num_input_tokens_seen": 3281290752,
      "step": 44230,
      "train_runtime": 52603.6741,
      "train_tokens_per_second": 62377.596
    },
    {
      "epoch": 0.5879030704114923,
      "grad_norm": 14.8125,
      "learning_rate": 7.661351070845148e-06,
      "loss": 0.6244,
      "num_input_tokens_seen": 3281506048,
      "step": 44240,
      "train_runtime": 52606.9711,
      "train_tokens_per_second": 62377.78
    },
    {
      "epoch": 0.5880359598939542,
      "grad_norm": 11.875,
      "learning_rate": 7.657166676075388e-06,
      "loss": 0.5899,
      "num_input_tokens_seen": 3281618944,
      "step": 44250,
      "train_runtime": 52609.9999,
      "train_tokens_per_second": 62376.334
    },
    {
      "epoch": 0.5881688493764161,
      "grad_norm": 21.5,
      "learning_rate": 7.652982715294614e-06,
      "loss": 0.846,
      "num_input_tokens_seen": 3283711232,
      "step": 44260,
      "train_runtime": 52644.2053,
      "train_tokens_per_second": 62375.549
    },
    {
      "epoch": 0.588301738858878,
      "grad_norm": 16.5,
      "learning_rate": 7.64879918927786e-06,
      "loss": 0.8814,
      "num_input_tokens_seen": 3284556032,
      "step": 44270,
      "train_runtime": 52657.1143,
      "train_tokens_per_second": 62376.301
    },
    {
      "epoch": 0.58843462834134,
      "grad_norm": 11.5625,
      "learning_rate": 7.644616098800095e-06,
      "loss": 0.7522,
      "num_input_tokens_seen": 3285070080,
      "step": 44280,
      "train_runtime": 52664.3924,
      "train_tokens_per_second": 62377.442
    },
    {
      "epoch": 0.5885675178238018,
      "grad_norm": 12.4375,
      "learning_rate": 7.640433444636196e-06,
      "loss": 0.6155,
      "num_input_tokens_seen": 3285298432,
      "step": 44290,
      "train_runtime": 52669.3731,
      "train_tokens_per_second": 62375.879
    },
    {
      "epoch": 0.5887004073062637,
      "grad_norm": 12.3125,
      "learning_rate": 7.636251227560957e-06,
      "loss": 0.5795,
      "num_input_tokens_seen": 3285412864,
      "step": 44300,
      "train_runtime": 52673.8479,
      "train_tokens_per_second": 62372.752
    },
    {
      "epoch": 0.5888332967887256,
      "grad_norm": 16.125,
      "learning_rate": 7.632069448349106e-06,
      "loss": 0.7514,
      "num_input_tokens_seen": 3287439616,
      "step": 44310,
      "train_runtime": 52706.7738,
      "train_tokens_per_second": 62372.241
    },
    {
      "epoch": 0.5889661862711876,
      "grad_norm": 25.25,
      "learning_rate": 7.627888107775269e-06,
      "loss": 0.9159,
      "num_input_tokens_seen": 3288282112,
      "step": 44320,
      "train_runtime": 52719.8514,
      "train_tokens_per_second": 62372.75
    },
    {
      "epoch": 0.5890990757536495,
      "grad_norm": 27.875,
      "learning_rate": 7.623707206614014e-06,
      "loss": 0.7232,
      "num_input_tokens_seen": 3288773120,
      "step": 44330,
      "train_runtime": 52726.8028,
      "train_tokens_per_second": 62373.839
    },
    {
      "epoch": 0.5892319652361114,
      "grad_norm": 59.0,
      "learning_rate": 7.619526745639807e-06,
      "loss": 0.6822,
      "num_input_tokens_seen": 3289007872,
      "step": 44340,
      "train_runtime": 52730.3349,
      "train_tokens_per_second": 62374.113
    },
    {
      "epoch": 0.5893648547185733,
      "grad_norm": 33.75,
      "learning_rate": 7.615346725627046e-06,
      "loss": 0.4789,
      "num_input_tokens_seen": 3289126656,
      "step": 44350,
      "train_runtime": 52732.7108,
      "train_tokens_per_second": 62373.555
    },
    {
      "epoch": 0.5894977442010352,
      "grad_norm": 15.8125,
      "learning_rate": 7.611167147350037e-06,
      "loss": 0.8098,
      "num_input_tokens_seen": 3291328512,
      "step": 44360,
      "train_runtime": 52767.6635,
      "train_tokens_per_second": 62373.967
    },
    {
      "epoch": 0.5896306336834971,
      "grad_norm": 24.25,
      "learning_rate": 7.606988011583016e-06,
      "loss": 0.9297,
      "num_input_tokens_seen": 3292151040,
      "step": 44370,
      "train_runtime": 52780.5066,
      "train_tokens_per_second": 62374.374
    },
    {
      "epoch": 0.589763523165959,
      "grad_norm": 15.75,
      "learning_rate": 7.602809319100129e-06,
      "loss": 0.7582,
      "num_input_tokens_seen": 3292628992,
      "step": 44380,
      "train_runtime": 52787.2831,
      "train_tokens_per_second": 62375.421
    },
    {
      "epoch": 0.589896412648421,
      "grad_norm": 17.875,
      "learning_rate": 7.598631070675438e-06,
      "loss": 0.6511,
      "num_input_tokens_seen": 3292857344,
      "step": 44390,
      "train_runtime": 52790.7526,
      "train_tokens_per_second": 62375.647
    },
    {
      "epoch": 0.5900293021308829,
      "grad_norm": 16.375,
      "learning_rate": 7.59445326708293e-06,
      "loss": 0.4801,
      "num_input_tokens_seen": 3292973312,
      "step": 44400,
      "train_runtime": 52793.1698,
      "train_tokens_per_second": 62374.988
    },
    {
      "epoch": 0.5901621916133447,
      "grad_norm": 30.625,
      "learning_rate": 7.590275909096501e-06,
      "loss": 0.8258,
      "num_input_tokens_seen": 3294880768,
      "step": 44410,
      "train_runtime": 52825.114,
      "train_tokens_per_second": 62373.377
    },
    {
      "epoch": 0.5902950810958066,
      "grad_norm": 21.875,
      "learning_rate": 7.586098997489977e-06,
      "loss": 0.8444,
      "num_input_tokens_seen": 3295712768,
      "step": 44420,
      "train_runtime": 52837.7456,
      "train_tokens_per_second": 62374.212
    },
    {
      "epoch": 0.5904279705782686,
      "grad_norm": 19.25,
      "learning_rate": 7.581922533037085e-06,
      "loss": 0.7541,
      "num_input_tokens_seen": 3296175872,
      "step": 44430,
      "train_runtime": 52844.4505,
      "train_tokens_per_second": 62375.062
    },
    {
      "epoch": 0.5905608600607305,
      "grad_norm": 19.375,
      "learning_rate": 7.577746516511486e-06,
      "loss": 0.6516,
      "num_input_tokens_seen": 3296408320,
      "step": 44440,
      "train_runtime": 52847.9254,
      "train_tokens_per_second": 62375.359
    },
    {
      "epoch": 0.5906937495431924,
      "grad_norm": 10.0625,
      "learning_rate": 7.573570948686746e-06,
      "loss": 0.5313,
      "num_input_tokens_seen": 3296524032,
      "step": 44450,
      "train_runtime": 52851.5622,
      "train_tokens_per_second": 62373.256
    },
    {
      "epoch": 0.5908266390256544,
      "grad_norm": 18.0,
      "learning_rate": 7.56939583033635e-06,
      "loss": 0.8144,
      "num_input_tokens_seen": 3298619904,
      "step": 44460,
      "train_runtime": 52884.9068,
      "train_tokens_per_second": 62373.56
    },
    {
      "epoch": 0.5909595285081162,
      "grad_norm": 26.0,
      "learning_rate": 7.565221162233704e-06,
      "loss": 0.8875,
      "num_input_tokens_seen": 3299433216,
      "step": 44470,
      "train_runtime": 52897.5473,
      "train_tokens_per_second": 62374.03
    },
    {
      "epoch": 0.5910924179905781,
      "grad_norm": 24.25,
      "learning_rate": 7.561046945152125e-06,
      "loss": 0.8225,
      "num_input_tokens_seen": 3299899392,
      "step": 44480,
      "train_runtime": 52904.2005,
      "train_tokens_per_second": 62374.998
    },
    {
      "epoch": 0.59122530747304,
      "grad_norm": 19.5,
      "learning_rate": 7.556873179864854e-06,
      "loss": 0.6024,
      "num_input_tokens_seen": 3300136960,
      "step": 44490,
      "train_runtime": 52907.7517,
      "train_tokens_per_second": 62375.301
    },
    {
      "epoch": 0.591358196955502,
      "grad_norm": 9.625,
      "learning_rate": 7.552699867145041e-06,
      "loss": 0.5438,
      "num_input_tokens_seen": 3300254464,
      "step": 44500,
      "train_runtime": 52910.1947,
      "train_tokens_per_second": 62374.642
    },
    {
      "epoch": 0.5914910864379639,
      "grad_norm": 23.375,
      "learning_rate": 7.548527007765757e-06,
      "loss": 0.8757,
      "num_input_tokens_seen": 3302381824,
      "step": 44510,
      "train_runtime": 52944.1845,
      "train_tokens_per_second": 62374.779
    },
    {
      "epoch": 0.5916239759204258,
      "grad_norm": 65.0,
      "learning_rate": 7.54435460249998e-06,
      "loss": 0.936,
      "num_input_tokens_seen": 3303230464,
      "step": 44520,
      "train_runtime": 52957.6469,
      "train_tokens_per_second": 62374.948
    },
    {
      "epoch": 0.5917568654028877,
      "grad_norm": 15.75,
      "learning_rate": 7.540182652120622e-06,
      "loss": 0.699,
      "num_input_tokens_seen": 3303686400,
      "step": 44530,
      "train_runtime": 52964.4297,
      "train_tokens_per_second": 62375.568
    },
    {
      "epoch": 0.5918897548853496,
      "grad_norm": 9.875,
      "learning_rate": 7.536011157400493e-06,
      "loss": 0.6492,
      "num_input_tokens_seen": 3303915520,
      "step": 44540,
      "train_runtime": 52967.8978,
      "train_tokens_per_second": 62375.81
    },
    {
      "epoch": 0.5920226443678115,
      "grad_norm": 13.5625,
      "learning_rate": 7.531840119112325e-06,
      "loss": 0.5562,
      "num_input_tokens_seen": 3304033024,
      "step": 44550,
      "train_runtime": 52970.3025,
      "train_tokens_per_second": 62375.196
    },
    {
      "epoch": 0.5921555338502734,
      "grad_norm": 29.5,
      "learning_rate": 7.527669538028768e-06,
      "loss": 0.8255,
      "num_input_tokens_seen": 3306091520,
      "step": 44560,
      "train_runtime": 53002.5598,
      "train_tokens_per_second": 62376.073
    },
    {
      "epoch": 0.5922884233327353,
      "grad_norm": 22.5,
      "learning_rate": 7.52349941492238e-06,
      "loss": 0.949,
      "num_input_tokens_seen": 3306943488,
      "step": 44570,
      "train_runtime": 53015.9373,
      "train_tokens_per_second": 62376.403
    },
    {
      "epoch": 0.5924213128151973,
      "grad_norm": 150.0,
      "learning_rate": 7.519329750565647e-06,
      "loss": 0.696,
      "num_input_tokens_seen": 3307411200,
      "step": 44580,
      "train_runtime": 53022.4629,
      "train_tokens_per_second": 62377.547
    },
    {
      "epoch": 0.5925542022976592,
      "grad_norm": 9.125,
      "learning_rate": 7.515160545730953e-06,
      "loss": 0.645,
      "num_input_tokens_seen": 3307638016,
      "step": 44590,
      "train_runtime": 53025.8738,
      "train_tokens_per_second": 62377.812
    },
    {
      "epoch": 0.592687091780121,
      "grad_norm": 26.25,
      "learning_rate": 7.510991801190615e-06,
      "loss": 0.5229,
      "num_input_tokens_seen": 3307752448,
      "step": 44600,
      "train_runtime": 53028.2617,
      "train_tokens_per_second": 62377.162
    },
    {
      "epoch": 0.592819981262583,
      "grad_norm": 15.0625,
      "learning_rate": 7.506823517716854e-06,
      "loss": 0.7925,
      "num_input_tokens_seen": 3309762816,
      "step": 44610,
      "train_runtime": 53061.6487,
      "train_tokens_per_second": 62375.801
    },
    {
      "epoch": 0.5929528707450449,
      "grad_norm": 23.5,
      "learning_rate": 7.502655696081801e-06,
      "loss": 0.8813,
      "num_input_tokens_seen": 3310614784,
      "step": 44620,
      "train_runtime": 53075.1899,
      "train_tokens_per_second": 62375.939
    },
    {
      "epoch": 0.5930857602275068,
      "grad_norm": 36.5,
      "learning_rate": 7.498488337057514e-06,
      "loss": 0.7669,
      "num_input_tokens_seen": 3311119872,
      "step": 44630,
      "train_runtime": 53082.2568,
      "train_tokens_per_second": 62377.15
    },
    {
      "epoch": 0.5932186497099687,
      "grad_norm": 11.875,
      "learning_rate": 7.494321441415954e-06,
      "loss": 0.6461,
      "num_input_tokens_seen": 3311349504,
      "step": 44640,
      "train_runtime": 53085.7386,
      "train_tokens_per_second": 62377.384
    },
    {
      "epoch": 0.5933515391924307,
      "grad_norm": 123.0,
      "learning_rate": 7.4901550099290076e-06,
      "loss": 0.4732,
      "num_input_tokens_seen": 3311465728,
      "step": 44650,
      "train_runtime": 53088.1463,
      "train_tokens_per_second": 62376.744
    },
    {
      "epoch": 0.5934844286748925,
      "grad_norm": 14.3125,
      "learning_rate": 7.485989043368464e-06,
      "loss": 0.7698,
      "num_input_tokens_seen": 3313498368,
      "step": 44660,
      "train_runtime": 53119.8623,
      "train_tokens_per_second": 62377.767
    },
    {
      "epoch": 0.5936173181573544,
      "grad_norm": 15.1875,
      "learning_rate": 7.4818235425060346e-06,
      "loss": 0.9097,
      "num_input_tokens_seen": 3314301952,
      "step": 44670,
      "train_runtime": 53132.3192,
      "train_tokens_per_second": 62378.266
    },
    {
      "epoch": 0.5937502076398163,
      "grad_norm": 31.25,
      "learning_rate": 7.477658508113336e-06,
      "loss": 0.6968,
      "num_input_tokens_seen": 3314755584,
      "step": 44680,
      "train_runtime": 53139.5803,
      "train_tokens_per_second": 62378.279
    },
    {
      "epoch": 0.5938830971222783,
      "grad_norm": 16.625,
      "learning_rate": 7.473493940961911e-06,
      "loss": 0.7226,
      "num_input_tokens_seen": 3314991872,
      "step": 44690,
      "train_runtime": 53143.0174,
      "train_tokens_per_second": 62378.691
    },
    {
      "epoch": 0.5940159866047402,
      "grad_norm": 22.875,
      "learning_rate": 7.469329841823207e-06,
      "loss": 0.56,
      "num_input_tokens_seen": 3315103232,
      "step": 44700,
      "train_runtime": 53145.3954,
      "train_tokens_per_second": 62377.995
    },
    {
      "epoch": 0.5941488760872021,
      "grad_norm": 34.25,
      "learning_rate": 7.46516621146858e-06,
      "loss": 0.791,
      "num_input_tokens_seen": 3317281280,
      "step": 44710,
      "train_runtime": 53180.0832,
      "train_tokens_per_second": 62378.264
    },
    {
      "epoch": 0.594281765569664,
      "grad_norm": 22.75,
      "learning_rate": 7.461003050669312e-06,
      "loss": 0.9036,
      "num_input_tokens_seen": 3318149120,
      "step": 44720,
      "train_runtime": 53193.6065,
      "train_tokens_per_second": 62378.721
    },
    {
      "epoch": 0.5944146550521259,
      "grad_norm": 11.25,
      "learning_rate": 7.4568403601965855e-06,
      "loss": 0.7561,
      "num_input_tokens_seen": 3318624000,
      "step": 44730,
      "train_runtime": 53200.4887,
      "train_tokens_per_second": 62379.577
    },
    {
      "epoch": 0.5945475445345878,
      "grad_norm": 11.1875,
      "learning_rate": 7.45267814082151e-06,
      "loss": 0.6297,
      "num_input_tokens_seen": 3318857728,
      "step": 44740,
      "train_runtime": 53204.0319,
      "train_tokens_per_second": 62379.816
    },
    {
      "epoch": 0.5946804340170497,
      "grad_norm": 6.375,
      "learning_rate": 7.448516393315092e-06,
      "loss": 0.5049,
      "num_input_tokens_seen": 3318973184,
      "step": 44750,
      "train_runtime": 53206.4464,
      "train_tokens_per_second": 62379.155
    },
    {
      "epoch": 0.5948133234995117,
      "grad_norm": 13.0,
      "learning_rate": 7.444355118448263e-06,
      "loss": 0.8673,
      "num_input_tokens_seen": 3320923136,
      "step": 44760,
      "train_runtime": 53239.1679,
      "train_tokens_per_second": 62377.443
    },
    {
      "epoch": 0.5949462129819736,
      "grad_norm": 17.25,
      "learning_rate": 7.44019431699186e-06,
      "loss": 0.8912,
      "num_input_tokens_seen": 3321767424,
      "step": 44770,
      "train_runtime": 53251.7998,
      "train_tokens_per_second": 62378.501
    },
    {
      "epoch": 0.5950791024644354,
      "grad_norm": 30.25,
      "learning_rate": 7.436033989716631e-06,
      "loss": 0.7896,
      "num_input_tokens_seen": 3322252032,
      "step": 44780,
      "train_runtime": 53258.6025,
      "train_tokens_per_second": 62379.632
    },
    {
      "epoch": 0.5952119919468973,
      "grad_norm": 11.1875,
      "learning_rate": 7.4318741373932465e-06,
      "loss": 0.6463,
      "num_input_tokens_seen": 3322475776,
      "step": 44790,
      "train_runtime": 53262.0257,
      "train_tokens_per_second": 62379.824
    },
    {
      "epoch": 0.5953448814293593,
      "grad_norm": 29.75,
      "learning_rate": 7.427714760792274e-06,
      "loss": 0.5009,
      "num_input_tokens_seen": 3322591488,
      "step": 44800,
      "train_runtime": 53264.9928,
      "train_tokens_per_second": 62378.521
    },
    {
      "epoch": 0.5954777709118212,
      "grad_norm": 27.0,
      "learning_rate": 7.42355586068421e-06,
      "loss": 0.8327,
      "num_input_tokens_seen": 3324591616,
      "step": 44810,
      "train_runtime": 53297.9144,
      "train_tokens_per_second": 62377.518
    },
    {
      "epoch": 0.5956106603942831,
      "grad_norm": 18.625,
      "learning_rate": 7.4193974378394465e-06,
      "loss": 0.9494,
      "num_input_tokens_seen": 3325413376,
      "step": 44820,
      "train_runtime": 53310.5549,
      "train_tokens_per_second": 62378.142
    },
    {
      "epoch": 0.595743549876745,
      "grad_norm": 33.25,
      "learning_rate": 7.4152394930283e-06,
      "loss": 0.7452,
      "num_input_tokens_seen": 3325907968,
      "step": 44830,
      "train_runtime": 53317.3667,
      "train_tokens_per_second": 62379.449
    },
    {
      "epoch": 0.5958764393592069,
      "grad_norm": 13.3125,
      "learning_rate": 7.411082027020985e-06,
      "loss": 0.6304,
      "num_input_tokens_seen": 3326142464,
      "step": 44840,
      "train_runtime": 53320.8795,
      "train_tokens_per_second": 62379.737
    },
    {
      "epoch": 0.5960093288416688,
      "grad_norm": 32.25,
      "learning_rate": 7.406925040587642e-06,
      "loss": 0.5649,
      "num_input_tokens_seen": 3326256896,
      "step": 44850,
      "train_runtime": 53323.3267,
      "train_tokens_per_second": 62379.021
    },
    {
      "epoch": 0.5961422183241307,
      "grad_norm": 32.5,
      "learning_rate": 7.4027685344983155e-06,
      "loss": 0.8068,
      "num_input_tokens_seen": 3328302336,
      "step": 44860,
      "train_runtime": 53357.1769,
      "train_tokens_per_second": 62377.782
    },
    {
      "epoch": 0.5962751078065927,
      "grad_norm": 19.25,
      "learning_rate": 7.398612509522957e-06,
      "loss": 0.9527,
      "num_input_tokens_seen": 3329140736,
      "step": 44870,
      "train_runtime": 53369.7554,
      "train_tokens_per_second": 62378.79
    },
    {
      "epoch": 0.5964079972890546,
      "grad_norm": 46.25,
      "learning_rate": 7.3944569664314354e-06,
      "loss": 0.7494,
      "num_input_tokens_seen": 3329583104,
      "step": 44880,
      "train_runtime": 53376.1555,
      "train_tokens_per_second": 62379.598
    },
    {
      "epoch": 0.5965408867715165,
      "grad_norm": 22.625,
      "learning_rate": 7.3903019059935246e-06,
      "loss": 0.6043,
      "num_input_tokens_seen": 3329803776,
      "step": 44890,
      "train_runtime": 53379.5789,
      "train_tokens_per_second": 62379.731
    },
    {
      "epoch": 0.5966737762539783,
      "grad_norm": 14.8125,
      "learning_rate": 7.3861473289789186e-06,
      "loss": 0.5942,
      "num_input_tokens_seen": 3329921024,
      "step": 44900,
      "train_runtime": 53382.0367,
      "train_tokens_per_second": 62379.055
    },
    {
      "epoch": 0.5968066657364403,
      "grad_norm": 16.25,
      "learning_rate": 7.381993236157211e-06,
      "loss": 0.8042,
      "num_input_tokens_seen": 3331855872,
      "step": 44910,
      "train_runtime": 53413.9651,
      "train_tokens_per_second": 62377.992
    },
    {
      "epoch": 0.5969395552189022,
      "grad_norm": 388.0,
      "learning_rate": 7.3778396282979134e-06,
      "loss": 0.8653,
      "num_input_tokens_seen": 3332655360,
      "step": 44920,
      "train_runtime": 53425.972,
      "train_tokens_per_second": 62378.937
    },
    {
      "epoch": 0.5970724447013641,
      "grad_norm": 18.875,
      "learning_rate": 7.373686506170445e-06,
      "loss": 0.7172,
      "num_input_tokens_seen": 3333093632,
      "step": 44930,
      "train_runtime": 53432.1874,
      "train_tokens_per_second": 62379.884
    },
    {
      "epoch": 0.597205334183826,
      "grad_norm": 12.625,
      "learning_rate": 7.369533870544128e-06,
      "loss": 0.6559,
      "num_input_tokens_seen": 3333310208,
      "step": 44940,
      "train_runtime": 53435.493,
      "train_tokens_per_second": 62380.078
    },
    {
      "epoch": 0.597338223666288,
      "grad_norm": 13.25,
      "learning_rate": 7.365381722188209e-06,
      "loss": 0.4981,
      "num_input_tokens_seen": 3333424640,
      "step": 44950,
      "train_runtime": 53438.0055,
      "train_tokens_per_second": 62379.286
    },
    {
      "epoch": 0.5974711131487498,
      "grad_norm": 46.25,
      "learning_rate": 7.361230061871831e-06,
      "loss": 0.8038,
      "num_input_tokens_seen": 3335469056,
      "step": 44960,
      "train_runtime": 53471.1425,
      "train_tokens_per_second": 62378.863
    },
    {
      "epoch": 0.5976040026312117,
      "grad_norm": 18.125,
      "learning_rate": 7.357078890364058e-06,
      "loss": 0.9684,
      "num_input_tokens_seen": 3336343808,
      "step": 44970,
      "train_runtime": 53484.4218,
      "train_tokens_per_second": 62379.73
    },
    {
      "epoch": 0.5977368921136736,
      "grad_norm": 16.0,
      "learning_rate": 7.352928208433851e-06,
      "loss": 0.6473,
      "num_input_tokens_seen": 3336823808,
      "step": 44980,
      "train_runtime": 53491.3683,
      "train_tokens_per_second": 62380.603
    },
    {
      "epoch": 0.5978697815961356,
      "grad_norm": 14.625,
      "learning_rate": 7.348778016850095e-06,
      "loss": 0.6147,
      "num_input_tokens_seen": 3337056000,
      "step": 44990,
      "train_runtime": 53495.4561,
      "train_tokens_per_second": 62380.177
    },
    {
      "epoch": 0.5980026710785975,
      "grad_norm": 8.0,
      "learning_rate": 7.3446283163815635e-06,
      "loss": 0.574,
      "num_input_tokens_seen": 3337169664,
      "step": 45000,
      "train_runtime": 53497.8995,
      "train_tokens_per_second": 62379.452
    },
    {
      "epoch": 0.5980026710785975,
      "eval_accuracy": 0.8173,
      "eval_balanced_accuracy": 0.8155527231905848,
      "eval_invalid_f1": 0.8351529369304339,
      "eval_invalid_precision": 0.8394703428260475,
      "eval_invalid_recall": 0.8308797127468581,
      "eval_loss": 0.3617829978466034,
      "eval_macro_f1": 0.8151317000453449,
      "eval_mcc": 0.6303174896258789,
      "eval_runtime": 78.6133,
      "eval_samples_per_second": 127.205,
      "eval_steps_per_second": 31.801,
      "eval_valid_f1": 0.7951104631602557,
      "eval_valid_precision": 0.7900601738355248,
      "eval_valid_recall": 0.8002257336343115,
      "num_input_tokens_seen": 3337169664,
      "step": 45000
    },
    {
      "epoch": 0.5981355605610594,
      "grad_norm": 20.75,
      "learning_rate": 7.340479107796965e-06,
      "loss": 0.8388,
      "num_input_tokens_seen": 3339190784,
      "step": 45010,
      "train_runtime": 53612.2478,
      "train_tokens_per_second": 62284.103
    },
    {
      "epoch": 0.5982684500435214,
      "grad_norm": 30.625,
      "learning_rate": 7.336330391864896e-06,
      "loss": 0.8951,
      "num_input_tokens_seen": 3340016128,
      "step": 45020,
      "train_runtime": 53624.7595,
      "train_tokens_per_second": 62284.962
    },
    {
      "epoch": 0.5984013395259832,
      "grad_norm": 11.0,
      "learning_rate": 7.33218216935387e-06,
      "loss": 0.7034,
      "num_input_tokens_seen": 3340487680,
      "step": 45030,
      "train_runtime": 53631.3372,
      "train_tokens_per_second": 62286.116
    },
    {
      "epoch": 0.5985342290084451,
      "grad_norm": 16.375,
      "learning_rate": 7.328034441032309e-06,
      "loss": 0.6402,
      "num_input_tokens_seen": 3340718848,
      "step": 45040,
      "train_runtime": 53635.5793,
      "train_tokens_per_second": 62285.499
    },
    {
      "epoch": 0.598667118490907,
      "grad_norm": 9.375,
      "learning_rate": 7.323887207668539e-06,
      "loss": 0.4848,
      "num_input_tokens_seen": 3340835072,
      "step": 45050,
      "train_runtime": 53637.987,
      "train_tokens_per_second": 62284.87
    },
    {
      "epoch": 0.598800007973369,
      "grad_norm": 25.625,
      "learning_rate": 7.319740470030805e-06,
      "loss": 0.7658,
      "num_input_tokens_seen": 3342854656,
      "step": 45060,
      "train_runtime": 53670.9106,
      "train_tokens_per_second": 62284.292
    },
    {
      "epoch": 0.5989328974558309,
      "grad_norm": 18.125,
      "learning_rate": 7.315594228887248e-06,
      "loss": 0.9288,
      "num_input_tokens_seen": 3343700480,
      "step": 45070,
      "train_runtime": 53683.4666,
      "train_tokens_per_second": 62285.48
    },
    {
      "epoch": 0.5990657869382928,
      "grad_norm": 18.0,
      "learning_rate": 7.3114484850059254e-06,
      "loss": 0.6435,
      "num_input_tokens_seen": 3344175104,
      "step": 45080,
      "train_runtime": 53691.072,
      "train_tokens_per_second": 62285.497
    },
    {
      "epoch": 0.5991986764207546,
      "grad_norm": 14.25,
      "learning_rate": 7.307303239154797e-06,
      "loss": 0.6877,
      "num_input_tokens_seen": 3344405504,
      "step": 45090,
      "train_runtime": 53694.6171,
      "train_tokens_per_second": 62285.676
    },
    {
      "epoch": 0.5993315659032166,
      "grad_norm": 11.6875,
      "learning_rate": 7.30315849210173e-06,
      "loss": 0.5589,
      "num_input_tokens_seen": 3344525824,
      "step": 45100,
      "train_runtime": 53697.1513,
      "train_tokens_per_second": 62284.977
    },
    {
      "epoch": 0.5994644553856785,
      "grad_norm": 13.0,
      "learning_rate": 7.299014244614508e-06,
      "loss": 0.8643,
      "num_input_tokens_seen": 3346641664,
      "step": 45110,
      "train_runtime": 53730.058,
      "train_tokens_per_second": 62286.21
    },
    {
      "epoch": 0.5995973448681404,
      "grad_norm": 26.0,
      "learning_rate": 7.294870497460811e-06,
      "loss": 0.9974,
      "num_input_tokens_seen": 3347477504,
      "step": 45120,
      "train_runtime": 53742.592,
      "train_tokens_per_second": 62287.236
    },
    {
      "epoch": 0.5997302343506024,
      "grad_norm": 31.75,
      "learning_rate": 7.290727251408236e-06,
      "loss": 0.7765,
      "num_input_tokens_seen": 3347958784,
      "step": 45130,
      "train_runtime": 53749.3321,
      "train_tokens_per_second": 62288.379
    },
    {
      "epoch": 0.5998631238330643,
      "grad_norm": 38.0,
      "learning_rate": 7.286584507224277e-06,
      "loss": 0.6174,
      "num_input_tokens_seen": 3348185856,
      "step": 45140,
      "train_runtime": 53752.7952,
      "train_tokens_per_second": 62288.591
    },
    {
      "epoch": 0.5999960133155261,
      "grad_norm": 19.125,
      "learning_rate": 7.282442265676346e-06,
      "loss": 0.5328,
      "num_input_tokens_seen": 3348303360,
      "step": 45150,
      "train_runtime": 53755.2172,
      "train_tokens_per_second": 62287.97
    },
    {
      "epoch": 0.600128902797988,
      "grad_norm": 28.5,
      "learning_rate": 7.278300527531747e-06,
      "loss": 0.8123,
      "num_input_tokens_seen": 3350301440,
      "step": 45160,
      "train_runtime": 53788.7473,
      "train_tokens_per_second": 62286.289
    },
    {
      "epoch": 0.60026179228045,
      "grad_norm": 34.25,
      "learning_rate": 7.274159293557712e-06,
      "loss": 0.8053,
      "num_input_tokens_seen": 3351138048,
      "step": 45170,
      "train_runtime": 53801.0591,
      "train_tokens_per_second": 62287.585
    },
    {
      "epoch": 0.6003946817629119,
      "grad_norm": 16.5,
      "learning_rate": 7.270018564521363e-06,
      "loss": 0.7371,
      "num_input_tokens_seen": 3351585536,
      "step": 45180,
      "train_runtime": 53807.4815,
      "train_tokens_per_second": 62288.467
    },
    {
      "epoch": 0.6005275712453738,
      "grad_norm": 18.75,
      "learning_rate": 7.26587834118973e-06,
      "loss": 0.6763,
      "num_input_tokens_seen": 3351805440,
      "step": 45190,
      "train_runtime": 53810.8362,
      "train_tokens_per_second": 62288.67
    },
    {
      "epoch": 0.6006604607278357,
      "grad_norm": 7.5,
      "learning_rate": 7.261738624329755e-06,
      "loss": 0.4819,
      "num_input_tokens_seen": 3351920896,
      "step": 45200,
      "train_runtime": 53814.013,
      "train_tokens_per_second": 62287.139
    },
    {
      "epoch": 0.6007933502102976,
      "grad_norm": 41.5,
      "learning_rate": 7.257599414708279e-06,
      "loss": 0.7847,
      "num_input_tokens_seen": 3353994752,
      "step": 45210,
      "train_runtime": 53847.4538,
      "train_tokens_per_second": 62286.97
    },
    {
      "epoch": 0.6009262396927595,
      "grad_norm": 35.25,
      "learning_rate": 7.253460713092064e-06,
      "loss": 0.9238,
      "num_input_tokens_seen": 3354839808,
      "step": 45220,
      "train_runtime": 53860.3524,
      "train_tokens_per_second": 62287.743
    },
    {
      "epoch": 0.6010591291752214,
      "grad_norm": 13.9375,
      "learning_rate": 7.249322520247757e-06,
      "loss": 0.7259,
      "num_input_tokens_seen": 3355333888,
      "step": 45230,
      "train_runtime": 53867.3096,
      "train_tokens_per_second": 62288.871
    },
    {
      "epoch": 0.6011920186576833,
      "grad_norm": 13.0,
      "learning_rate": 7.24518483694193e-06,
      "loss": 0.6063,
      "num_input_tokens_seen": 3355575296,
      "step": 45240,
      "train_runtime": 53870.9224,
      "train_tokens_per_second": 62289.175
    },
    {
      "epoch": 0.6013249081401453,
      "grad_norm": 14.5,
      "learning_rate": 7.241047663941046e-06,
      "loss": 0.581,
      "num_input_tokens_seen": 3355690752,
      "step": 45250,
      "train_runtime": 53873.3611,
      "train_tokens_per_second": 62288.498
    },
    {
      "epoch": 0.6014577976226072,
      "grad_norm": 15.0,
      "learning_rate": 7.236911002011478e-06,
      "loss": 0.867,
      "num_input_tokens_seen": 3357719552,
      "step": 45260,
      "train_runtime": 53905.5322,
      "train_tokens_per_second": 62288.96
    },
    {
      "epoch": 0.601590687105069,
      "grad_norm": 22.5,
      "learning_rate": 7.232774851919513e-06,
      "loss": 0.945,
      "num_input_tokens_seen": 3358550784,
      "step": 45270,
      "train_runtime": 53918.1277,
      "train_tokens_per_second": 62289.826
    },
    {
      "epoch": 0.601723576587531,
      "grad_norm": 30.125,
      "learning_rate": 7.22863921443133e-06,
      "loss": 0.8173,
      "num_input_tokens_seen": 3359023360,
      "step": 45280,
      "train_runtime": 53925.5694,
      "train_tokens_per_second": 62289.993
    },
    {
      "epoch": 0.6018564660699929,
      "grad_norm": 16.0,
      "learning_rate": 7.224504090313023e-06,
      "loss": 0.6551,
      "num_input_tokens_seen": 3359264512,
      "step": 45290,
      "train_runtime": 53929.1186,
      "train_tokens_per_second": 62290.366
    },
    {
      "epoch": 0.6019893555524548,
      "grad_norm": 8.625,
      "learning_rate": 7.220369480330579e-06,
      "loss": 0.4712,
      "num_input_tokens_seen": 3359378432,
      "step": 45300,
      "train_runtime": 53931.593,
      "train_tokens_per_second": 62289.62
    },
    {
      "epoch": 0.6021222450349167,
      "grad_norm": 22.75,
      "learning_rate": 7.216235385249908e-06,
      "loss": 0.8553,
      "num_input_tokens_seen": 3361541888,
      "step": 45310,
      "train_runtime": 53966.2014,
      "train_tokens_per_second": 62289.763
    },
    {
      "epoch": 0.6022551345173787,
      "grad_norm": 48.5,
      "learning_rate": 7.212101805836806e-06,
      "loss": 0.9568,
      "num_input_tokens_seen": 3362356224,
      "step": 45320,
      "train_runtime": 53978.385,
      "train_tokens_per_second": 62290.79
    },
    {
      "epoch": 0.6023880239998405,
      "grad_norm": 13.75,
      "learning_rate": 7.207968742856989e-06,
      "loss": 0.6964,
      "num_input_tokens_seen": 3362808064,
      "step": 45330,
      "train_runtime": 53985.484,
      "train_tokens_per_second": 62290.968
    },
    {
      "epoch": 0.6025209134823024,
      "grad_norm": 27.25,
      "learning_rate": 7.203836197076066e-06,
      "loss": 0.6513,
      "num_input_tokens_seen": 3363029504,
      "step": 45340,
      "train_runtime": 53988.9001,
      "train_tokens_per_second": 62291.128
    },
    {
      "epoch": 0.6026538029647643,
      "grad_norm": 18.25,
      "learning_rate": 7.199704169259554e-06,
      "loss": 0.4792,
      "num_input_tokens_seen": 3363145728,
      "step": 45350,
      "train_runtime": 53991.3248,
      "train_tokens_per_second": 62290.484
    },
    {
      "epoch": 0.6027866924472263,
      "grad_norm": 592.0,
      "learning_rate": 7.195572660172875e-06,
      "loss": 0.8247,
      "num_input_tokens_seen": 3365132544,
      "step": 45360,
      "train_runtime": 54022.6065,
      "train_tokens_per_second": 62291.192
    },
    {
      "epoch": 0.6029195819296882,
      "grad_norm": 26.125,
      "learning_rate": 7.191441670581351e-06,
      "loss": 0.8695,
      "num_input_tokens_seen": 3366000128,
      "step": 45370,
      "train_runtime": 54035.9348,
      "train_tokens_per_second": 62291.883
    },
    {
      "epoch": 0.6030524714121501,
      "grad_norm": 11.9375,
      "learning_rate": 7.187311201250218e-06,
      "loss": 0.7239,
      "num_input_tokens_seen": 3366493440,
      "step": 45380,
      "train_runtime": 54042.7977,
      "train_tokens_per_second": 62293.101
    },
    {
      "epoch": 0.6031853608946119,
      "grad_norm": 14.6875,
      "learning_rate": 7.183181252944602e-06,
      "loss": 0.6417,
      "num_input_tokens_seen": 3366740224,
      "step": 45390,
      "train_runtime": 54046.4612,
      "train_tokens_per_second": 62293.444
    },
    {
      "epoch": 0.6033182503770739,
      "grad_norm": 12.125,
      "learning_rate": 7.179051826429544e-06,
      "loss": 0.605,
      "num_input_tokens_seen": 3366859520,
      "step": 45400,
      "train_runtime": 54048.9568,
      "train_tokens_per_second": 62292.775
    },
    {
      "epoch": 0.6034511398595358,
      "grad_norm": 16.625,
      "learning_rate": 7.174922922469983e-06,
      "loss": 0.8254,
      "num_input_tokens_seen": 3368954112,
      "step": 45410,
      "train_runtime": 54082.6393,
      "train_tokens_per_second": 62292.709
    },
    {
      "epoch": 0.6035840293419977,
      "grad_norm": 16.5,
      "learning_rate": 7.170794541830753e-06,
      "loss": 0.9059,
      "num_input_tokens_seen": 3369840896,
      "step": 45420,
      "train_runtime": 54095.4831,
      "train_tokens_per_second": 62294.312
    },
    {
      "epoch": 0.6037169188244597,
      "grad_norm": 11.875,
      "learning_rate": 7.166666685276612e-06,
      "loss": 0.7489,
      "num_input_tokens_seen": 3370314496,
      "step": 45430,
      "train_runtime": 54102.1017,
      "train_tokens_per_second": 62295.445
    },
    {
      "epoch": 0.6038498083069216,
      "grad_norm": 15.3125,
      "learning_rate": 7.1625393535722025e-06,
      "loss": 0.6078,
      "num_input_tokens_seen": 3370544384,
      "step": 45440,
      "train_runtime": 54105.5623,
      "train_tokens_per_second": 62295.709
    },
    {
      "epoch": 0.6039826977893834,
      "grad_norm": 16.375,
      "learning_rate": 7.158412547482077e-06,
      "loss": 0.5733,
      "num_input_tokens_seen": 3370658304,
      "step": 45450,
      "train_runtime": 54108.6913,
      "train_tokens_per_second": 62294.212
    },
    {
      "epoch": 0.6041155872718453,
      "grad_norm": 24.875,
      "learning_rate": 7.1542862677706845e-06,
      "loss": 0.8038,
      "num_input_tokens_seen": 3372736512,
      "step": 45460,
      "train_runtime": 54142.1493,
      "train_tokens_per_second": 62294.101
    },
    {
      "epoch": 0.6042484767543073,
      "grad_norm": 17.625,
      "learning_rate": 7.1501605152023915e-06,
      "loss": 0.9149,
      "num_input_tokens_seen": 3373582848,
      "step": 45470,
      "train_runtime": 54154.7425,
      "train_tokens_per_second": 62295.243
    },
    {
      "epoch": 0.6043813662367692,
      "grad_norm": 24.375,
      "learning_rate": 7.146035290541447e-06,
      "loss": 0.6853,
      "num_input_tokens_seen": 3374035968,
      "step": 45480,
      "train_runtime": 54161.2687,
      "train_tokens_per_second": 62296.103
    },
    {
      "epoch": 0.6045142557192311,
      "grad_norm": 25.375,
      "learning_rate": 7.141910594552019e-06,
      "loss": 0.6602,
      "num_input_tokens_seen": 3374263296,
      "step": 45490,
      "train_runtime": 54164.7417,
      "train_tokens_per_second": 62296.306
    },
    {
      "epoch": 0.604647145201693,
      "grad_norm": 34.75,
      "learning_rate": 7.13778642799817e-06,
      "loss": 0.5648,
      "num_input_tokens_seen": 3374380800,
      "step": 45500,
      "train_runtime": 54167.2733,
      "train_tokens_per_second": 62295.563
    },
    {
      "epoch": 0.604780034684155,
      "grad_norm": 14.25,
      "learning_rate": 7.133662791643863e-06,
      "loss": 0.8081,
      "num_input_tokens_seen": 3376568064,
      "step": 45510,
      "train_runtime": 54201.8528,
      "train_tokens_per_second": 62296.174
    },
    {
      "epoch": 0.6049129241666168,
      "grad_norm": 22.5,
      "learning_rate": 7.129539686252964e-06,
      "loss": 0.8434,
      "num_input_tokens_seen": 3377434112,
      "step": 45520,
      "train_runtime": 54215.3613,
      "train_tokens_per_second": 62296.626
    },
    {
      "epoch": 0.6050458136490787,
      "grad_norm": 18.25,
      "learning_rate": 7.12541711258924e-06,
      "loss": 0.7367,
      "num_input_tokens_seen": 3377934848,
      "step": 45530,
      "train_runtime": 54222.4584,
      "train_tokens_per_second": 62297.707
    },
    {
      "epoch": 0.6051787031315407,
      "grad_norm": 15.9375,
      "learning_rate": 7.121295071416368e-06,
      "loss": 0.5952,
      "num_input_tokens_seen": 3378180608,
      "step": 45540,
      "train_runtime": 54226.0143,
      "train_tokens_per_second": 62298.154
    },
    {
      "epoch": 0.6053115926140026,
      "grad_norm": 13.0625,
      "learning_rate": 7.117173563497912e-06,
      "loss": 0.5468,
      "num_input_tokens_seen": 3378296832,
      "step": 45550,
      "train_runtime": 54228.4569,
      "train_tokens_per_second": 62297.491
    },
    {
      "epoch": 0.6054444820964645,
      "grad_norm": 18.125,
      "learning_rate": 7.113052589597351e-06,
      "loss": 0.8644,
      "num_input_tokens_seen": 3380202496,
      "step": 45560,
      "train_runtime": 54259.3333,
      "train_tokens_per_second": 62297.162
    },
    {
      "epoch": 0.6055773715789264,
      "grad_norm": 31.5,
      "learning_rate": 7.108932150478056e-06,
      "loss": 0.8857,
      "num_input_tokens_seen": 3381022464,
      "step": 45570,
      "train_runtime": 54272.1537,
      "train_tokens_per_second": 62297.555
    },
    {
      "epoch": 0.6057102610613883,
      "grad_norm": 17.625,
      "learning_rate": 7.104812246903295e-06,
      "loss": 0.737,
      "num_input_tokens_seen": 3381464576,
      "step": 45580,
      "train_runtime": 54278.5141,
      "train_tokens_per_second": 62298.4
    },
    {
      "epoch": 0.6058431505438502,
      "grad_norm": 28.125,
      "learning_rate": 7.100692879636255e-06,
      "loss": 0.7309,
      "num_input_tokens_seen": 3381687040,
      "step": 45590,
      "train_runtime": 54281.9436,
      "train_tokens_per_second": 62298.562
    },
    {
      "epoch": 0.6059760400263121,
      "grad_norm": 8.25,
      "learning_rate": 7.0965740494400024e-06,
      "loss": 0.5423,
      "num_input_tokens_seen": 3381796864,
      "step": 45600,
      "train_runtime": 54284.3538,
      "train_tokens_per_second": 62297.819
    },
    {
      "epoch": 0.606108929508774,
      "grad_norm": 26.625,
      "learning_rate": 7.092455757077521e-06,
      "loss": 0.8587,
      "num_input_tokens_seen": 3383906560,
      "step": 45610,
      "train_runtime": 54317.9441,
      "train_tokens_per_second": 62298.134
    },
    {
      "epoch": 0.606241818991236,
      "grad_norm": 30.25,
      "learning_rate": 7.088338003311681e-06,
      "loss": 0.9005,
      "num_input_tokens_seen": 3384781312,
      "step": 45620,
      "train_runtime": 54330.9688,
      "train_tokens_per_second": 62299.3
    },
    {
      "epoch": 0.6063747084736979,
      "grad_norm": 16.75,
      "learning_rate": 7.084220788905266e-06,
      "loss": 0.7771,
      "num_input_tokens_seen": 3385272576,
      "step": 45630,
      "train_runtime": 54338.0098,
      "train_tokens_per_second": 62300.268
    },
    {
      "epoch": 0.6065075979561597,
      "grad_norm": 8.5,
      "learning_rate": 7.080104114620947e-06,
      "loss": 0.6294,
      "num_input_tokens_seen": 3385503488,
      "step": 45640,
      "train_runtime": 54341.5812,
      "train_tokens_per_second": 62300.423
    },
    {
      "epoch": 0.6066404874386216,
      "grad_norm": 25.375,
      "learning_rate": 7.0759879812213085e-06,
      "loss": 0.4784,
      "num_input_tokens_seen": 3385622272,
      "step": 45650,
      "train_runtime": 54344.0015,
      "train_tokens_per_second": 62299.834
    },
    {
      "epoch": 0.6067733769210836,
      "grad_norm": 31.625,
      "learning_rate": 7.071872389468823e-06,
      "loss": 0.7701,
      "num_input_tokens_seen": 3387833856,
      "step": 45660,
      "train_runtime": 54379.0672,
      "train_tokens_per_second": 62300.33
    },
    {
      "epoch": 0.6069062664035455,
      "grad_norm": 13.625,
      "learning_rate": 7.067757340125867e-06,
      "loss": 0.9395,
      "num_input_tokens_seen": 3388708608,
      "step": 45670,
      "train_runtime": 54392.01,
      "train_tokens_per_second": 62301.588
    },
    {
      "epoch": 0.6070391558860074,
      "grad_norm": 19.375,
      "learning_rate": 7.063642833954718e-06,
      "loss": 0.7436,
      "num_input_tokens_seen": 3389165568,
      "step": 45680,
      "train_runtime": 54399.165,
      "train_tokens_per_second": 62301.794
    },
    {
      "epoch": 0.6071720453684694,
      "grad_norm": 23.0,
      "learning_rate": 7.059528871717547e-06,
      "loss": 0.6236,
      "num_input_tokens_seen": 3389389056,
      "step": 45690,
      "train_runtime": 54402.5955,
      "train_tokens_per_second": 62301.973
    },
    {
      "epoch": 0.6073049348509312,
      "grad_norm": 11.5625,
      "learning_rate": 7.0554154541764376e-06,
      "loss": 0.5036,
      "num_input_tokens_seen": 3389502976,
      "step": 45700,
      "train_runtime": 54405.004,
      "train_tokens_per_second": 62301.309
    },
    {
      "epoch": 0.6074378243333931,
      "grad_norm": 31.875,
      "learning_rate": 7.051302582093357e-06,
      "loss": 0.8257,
      "num_input_tokens_seen": 3391533056,
      "step": 45710,
      "train_runtime": 54436.8611,
      "train_tokens_per_second": 62302.142
    },
    {
      "epoch": 0.607570713815855,
      "grad_norm": 28.375,
      "learning_rate": 7.047190256230182e-06,
      "loss": 0.844,
      "num_input_tokens_seen": 3392380928,
      "step": 45720,
      "train_runtime": 54449.4692,
      "train_tokens_per_second": 62303.287
    },
    {
      "epoch": 0.607703603298317,
      "grad_norm": 19.0,
      "learning_rate": 7.043078477348677e-06,
      "loss": 0.7141,
      "num_input_tokens_seen": 3392846336,
      "step": 45730,
      "train_runtime": 54456.7984,
      "train_tokens_per_second": 62303.449
    },
    {
      "epoch": 0.6078364927807789,
      "grad_norm": 27.75,
      "learning_rate": 7.038967246210524e-06,
      "loss": 0.6238,
      "num_input_tokens_seen": 3393083392,
      "step": 45740,
      "train_runtime": 54460.3175,
      "train_tokens_per_second": 62303.775
    },
    {
      "epoch": 0.6079693822632408,
      "grad_norm": 29.375,
      "learning_rate": 7.034856563577284e-06,
      "loss": 0.4675,
      "num_input_tokens_seen": 3393199104,
      "step": 45750,
      "train_runtime": 54462.758,
      "train_tokens_per_second": 62303.108
    },
    {
      "epoch": 0.6081022717457026,
      "grad_norm": 13.375,
      "learning_rate": 7.0307464302104245e-06,
      "loss": 0.8155,
      "num_input_tokens_seen": 3395191808,
      "step": 45760,
      "train_runtime": 54494.0056,
      "train_tokens_per_second": 62303.95
    },
    {
      "epoch": 0.6082351612281646,
      "grad_norm": 35.0,
      "learning_rate": 7.026636846871314e-06,
      "loss": 0.8953,
      "num_input_tokens_seen": 3396031232,
      "step": 45770,
      "train_runtime": 54507.2396,
      "train_tokens_per_second": 62304.223
    },
    {
      "epoch": 0.6083680507106265,
      "grad_norm": 27.625,
      "learning_rate": 7.0225278143212106e-06,
      "loss": 0.7152,
      "num_input_tokens_seen": 3396511488,
      "step": 45780,
      "train_runtime": 54513.9312,
      "train_tokens_per_second": 62305.385
    },
    {
      "epoch": 0.6085009401930884,
      "grad_norm": 15.5,
      "learning_rate": 7.018419333321285e-06,
      "loss": 0.6265,
      "num_input_tokens_seen": 3396737024,
      "step": 45790,
      "train_runtime": 54517.4128,
      "train_tokens_per_second": 62305.543
    },
    {
      "epoch": 0.6086338296755504,
      "grad_norm": 36.75,
      "learning_rate": 7.014311404632588e-06,
      "loss": 0.5592,
      "num_input_tokens_seen": 3396851968,
      "step": 45800,
      "train_runtime": 54519.8455,
      "train_tokens_per_second": 62304.871
    },
    {
      "epoch": 0.6087667191580123,
      "grad_norm": 37.75,
      "learning_rate": 7.010204029016083e-06,
      "loss": 0.8106,
      "num_input_tokens_seen": 3398917632,
      "step": 45810,
      "train_runtime": 54553.2864,
      "train_tokens_per_second": 62304.544
    },
    {
      "epoch": 0.6088996086404741,
      "grad_norm": 38.5,
      "learning_rate": 7.006097207232623e-06,
      "loss": 0.9309,
      "num_input_tokens_seen": 3399793152,
      "step": 45820,
      "train_runtime": 54566.0815,
      "train_tokens_per_second": 62305.979
    },
    {
      "epoch": 0.609032498122936,
      "grad_norm": 13.6875,
      "learning_rate": 7.0019909400429545e-06,
      "loss": 0.7767,
      "num_input_tokens_seen": 3400295424,
      "step": 45830,
      "train_runtime": 54573.0089,
      "train_tokens_per_second": 62307.274
    },
    {
      "epoch": 0.609165387605398,
      "grad_norm": 17.625,
      "learning_rate": 6.9978852282077346e-06,
      "loss": 0.6467,
      "num_input_tokens_seen": 3400526848,
      "step": 45840,
      "train_runtime": 54577.2865,
      "train_tokens_per_second": 62306.631
    },
    {
      "epoch": 0.6092982770878599,
      "grad_norm": 10.125,
      "learning_rate": 6.9937800724875034e-06,
      "loss": 0.6078,
      "num_input_tokens_seen": 3400643840,
      "step": 45850,
      "train_runtime": 54579.7839,
      "train_tokens_per_second": 62305.923
    },
    {
      "epoch": 0.6094311665703218,
      "grad_norm": 13.0625,
      "learning_rate": 6.989675473642714e-06,
      "loss": 0.8094,
      "num_input_tokens_seen": 3402654976,
      "step": 45860,
      "train_runtime": 54612.4261,
      "train_tokens_per_second": 62305.508
    },
    {
      "epoch": 0.6095640560527837,
      "grad_norm": 29.0,
      "learning_rate": 6.9855714324336955e-06,
      "loss": 0.9238,
      "num_input_tokens_seen": 3403489280,
      "step": 45870,
      "train_runtime": 54625.0309,
      "train_tokens_per_second": 62306.405
    },
    {
      "epoch": 0.6096969455352456,
      "grad_norm": 40.25,
      "learning_rate": 6.9814679496206935e-06,
      "loss": 0.8012,
      "num_input_tokens_seen": 3403957760,
      "step": 45880,
      "train_runtime": 54631.6788,
      "train_tokens_per_second": 62307.398
    },
    {
      "epoch": 0.6098298350177075,
      "grad_norm": 23.5,
      "learning_rate": 6.977365025963834e-06,
      "loss": 0.5967,
      "num_input_tokens_seen": 3404194560,
      "step": 45890,
      "train_runtime": 54635.9108,
      "train_tokens_per_second": 62306.906
    },
    {
      "epoch": 0.6099627245001694,
      "grad_norm": 10.6875,
      "learning_rate": 6.973262662223155e-06,
      "loss": 0.6017,
      "num_input_tokens_seen": 3404311808,
      "step": 45900,
      "train_runtime": 54638.3863,
      "train_tokens_per_second": 62306.229
    },
    {
      "epoch": 0.6100956139826313,
      "grad_norm": 20.25,
      "learning_rate": 6.9691608591585795e-06,
      "loss": 0.8311,
      "num_input_tokens_seen": 3406425344,
      "step": 45910,
      "train_runtime": 54672.8403,
      "train_tokens_per_second": 62305.622
    },
    {
      "epoch": 0.6102285034650933,
      "grad_norm": 16.5,
      "learning_rate": 6.965059617529926e-06,
      "loss": 0.9403,
      "num_input_tokens_seen": 3407282944,
      "step": 45920,
      "train_runtime": 54685.7385,
      "train_tokens_per_second": 62306.609
    },
    {
      "epoch": 0.6103613929475552,
      "grad_norm": 14.8125,
      "learning_rate": 6.96095893809692e-06,
      "loss": 0.7103,
      "num_input_tokens_seen": 3407759360,
      "step": 45930,
      "train_runtime": 54692.4962,
      "train_tokens_per_second": 62307.622
    },
    {
      "epoch": 0.610494282430017,
      "grad_norm": 18.375,
      "learning_rate": 6.956858821619166e-06,
      "loss": 0.5565,
      "num_input_tokens_seen": 3407996416,
      "step": 45940,
      "train_runtime": 54696.7499,
      "train_tokens_per_second": 62307.11
    },
    {
      "epoch": 0.610627171912479,
      "grad_norm": 27.5,
      "learning_rate": 6.952759268856187e-06,
      "loss": 0.5693,
      "num_input_tokens_seen": 3408112128,
      "step": 45950,
      "train_runtime": 54699.2871,
      "train_tokens_per_second": 62306.335
    },
    {
      "epoch": 0.6107600613949409,
      "grad_norm": 34.0,
      "learning_rate": 6.9486602805673765e-06,
      "loss": 0.7811,
      "num_input_tokens_seen": 3410151424,
      "step": 45960,
      "train_runtime": 54731.3111,
      "train_tokens_per_second": 62307.139
    },
    {
      "epoch": 0.6108929508774028,
      "grad_norm": 21.875,
      "learning_rate": 6.9445618575120445e-06,
      "loss": 0.9,
      "num_input_tokens_seen": 3411003392,
      "step": 45970,
      "train_runtime": 54744.0271,
      "train_tokens_per_second": 62308.229
    },
    {
      "epoch": 0.6110258403598647,
      "grad_norm": 19.125,
      "learning_rate": 6.940464000449383e-06,
      "loss": 0.7079,
      "num_input_tokens_seen": 3411473664,
      "step": 45980,
      "train_runtime": 54750.7973,
      "train_tokens_per_second": 62309.114
    },
    {
      "epoch": 0.6111587298423267,
      "grad_norm": 27.0,
      "learning_rate": 6.9363667101384805e-06,
      "loss": 0.632,
      "num_input_tokens_seen": 3411697664,
      "step": 45990,
      "train_runtime": 54754.2533,
      "train_tokens_per_second": 62309.272
    },
    {
      "epoch": 0.6112916193247886,
      "grad_norm": 13.125,
      "learning_rate": 6.9322699873383306e-06,
      "loss": 0.5567,
      "num_input_tokens_seen": 3411809536,
      "step": 46000,
      "train_runtime": 54757.3984,
      "train_tokens_per_second": 62307.736
    },
    {
      "epoch": 0.6114245088072504,
      "grad_norm": 42.75,
      "learning_rate": 6.928173832807808e-06,
      "loss": 0.9085,
      "num_input_tokens_seen": 3413884160,
      "step": 46010,
      "train_runtime": 54790.1642,
      "train_tokens_per_second": 62308.34
    },
    {
      "epoch": 0.6115573982897123,
      "grad_norm": 17.375,
      "learning_rate": 6.924078247305694e-06,
      "loss": 0.902,
      "num_input_tokens_seen": 3414720768,
      "step": 46020,
      "train_runtime": 54802.8873,
      "train_tokens_per_second": 62309.14
    },
    {
      "epoch": 0.6116902877721743,
      "grad_norm": 11.125,
      "learning_rate": 6.919983231590657e-06,
      "loss": 0.7483,
      "num_input_tokens_seen": 3415166464,
      "step": 46030,
      "train_runtime": 54809.4727,
      "train_tokens_per_second": 62309.785
    },
    {
      "epoch": 0.6118231772546362,
      "grad_norm": 32.75,
      "learning_rate": 6.915888786421263e-06,
      "loss": 0.5928,
      "num_input_tokens_seen": 3415393024,
      "step": 46040,
      "train_runtime": 54813.5485,
      "train_tokens_per_second": 62309.285
    },
    {
      "epoch": 0.6119560667370981,
      "grad_norm": 17.625,
      "learning_rate": 6.911794912555967e-06,
      "loss": 0.4649,
      "num_input_tokens_seen": 3415504640,
      "step": 46050,
      "train_runtime": 54815.9527,
      "train_tokens_per_second": 62308.588
    },
    {
      "epoch": 0.61208895621956,
      "grad_norm": 14.75,
      "learning_rate": 6.907701610753131e-06,
      "loss": 0.7597,
      "num_input_tokens_seen": 3417581568,
      "step": 46060,
      "train_runtime": 54849.305,
      "train_tokens_per_second": 62308.567
    },
    {
      "epoch": 0.6122218457020219,
      "grad_norm": 21.25,
      "learning_rate": 6.903608881770998e-06,
      "loss": 0.8921,
      "num_input_tokens_seen": 3418441984,
      "step": 46070,
      "train_runtime": 54862.0296,
      "train_tokens_per_second": 62309.798
    },
    {
      "epoch": 0.6123547351844838,
      "grad_norm": 22.625,
      "learning_rate": 6.899516726367708e-06,
      "loss": 0.8036,
      "num_input_tokens_seen": 3418935808,
      "step": 46080,
      "train_runtime": 54869.0114,
      "train_tokens_per_second": 62310.869
    },
    {
      "epoch": 0.6124876246669457,
      "grad_norm": 13.125,
      "learning_rate": 6.8954251453013e-06,
      "loss": 0.6193,
      "num_input_tokens_seen": 3419165696,
      "step": 46090,
      "train_runtime": 54872.5295,
      "train_tokens_per_second": 62311.064
    },
    {
      "epoch": 0.6126205141494077,
      "grad_norm": 16.375,
      "learning_rate": 6.891334139329696e-06,
      "loss": 0.5208,
      "num_input_tokens_seen": 3419281408,
      "step": 46100,
      "train_runtime": 54875.0107,
      "train_tokens_per_second": 62310.355
    },
    {
      "epoch": 0.6127534036318696,
      "grad_norm": 25.75,
      "learning_rate": 6.887243709210728e-06,
      "loss": 0.82,
      "num_input_tokens_seen": 3421397248,
      "step": 46110,
      "train_runtime": 54908.4121,
      "train_tokens_per_second": 62310.985
    },
    {
      "epoch": 0.6128862931143315,
      "grad_norm": 16.875,
      "learning_rate": 6.883153855702105e-06,
      "loss": 0.9647,
      "num_input_tokens_seen": 3422258432,
      "step": 46120,
      "train_runtime": 54922.0322,
      "train_tokens_per_second": 62311.213
    },
    {
      "epoch": 0.6130191825967933,
      "grad_norm": 31.125,
      "learning_rate": 6.879064579561438e-06,
      "loss": 0.7071,
      "num_input_tokens_seen": 3422736640,
      "step": 46130,
      "train_runtime": 54928.9014,
      "train_tokens_per_second": 62312.126
    },
    {
      "epoch": 0.6131520720792553,
      "grad_norm": 18.625,
      "learning_rate": 6.874975881546231e-06,
      "loss": 0.6594,
      "num_input_tokens_seen": 3422978304,
      "step": 46140,
      "train_runtime": 54932.4511,
      "train_tokens_per_second": 62312.499
    },
    {
      "epoch": 0.6132849615617172,
      "grad_norm": 13.9375,
      "learning_rate": 6.870887762413871e-06,
      "loss": 0.5248,
      "num_input_tokens_seen": 3423098624,
      "step": 46150,
      "train_runtime": 54934.8594,
      "train_tokens_per_second": 62311.958
    },
    {
      "epoch": 0.6134178510441791,
      "grad_norm": 26.0,
      "learning_rate": 6.866800222921657e-06,
      "loss": 0.8044,
      "num_input_tokens_seen": 3425038080,
      "step": 46160,
      "train_runtime": 54966.4078,
      "train_tokens_per_second": 62311.477
    },
    {
      "epoch": 0.613550740526641,
      "grad_norm": 25.5,
      "learning_rate": 6.862713263826759e-06,
      "loss": 0.8387,
      "num_input_tokens_seen": 3425865984,
      "step": 46170,
      "train_runtime": 54979.0974,
      "train_tokens_per_second": 62312.154
    },
    {
      "epoch": 0.613683630009103,
      "grad_norm": 18.5,
      "learning_rate": 6.858626885886257e-06,
      "loss": 0.7499,
      "num_input_tokens_seen": 3426357248,
      "step": 46180,
      "train_runtime": 54986.7967,
      "train_tokens_per_second": 62312.363
    },
    {
      "epoch": 0.6138165194915648,
      "grad_norm": 506.0,
      "learning_rate": 6.854541089857112e-06,
      "loss": 0.7277,
      "num_input_tokens_seen": 3426586624,
      "step": 46190,
      "train_runtime": 54990.3373,
      "train_tokens_per_second": 62312.522
    },
    {
      "epoch": 0.6139494089740267,
      "grad_norm": 20.875,
      "learning_rate": 6.850455876496185e-06,
      "loss": 0.5537,
      "num_input_tokens_seen": 3426703360,
      "step": 46200,
      "train_runtime": 54992.7966,
      "train_tokens_per_second": 62311.859
    },
    {
      "epoch": 0.6140822984564887,
      "grad_norm": 17.625,
      "learning_rate": 6.846371246560219e-06,
      "loss": 0.8052,
      "num_input_tokens_seen": 3428776448,
      "step": 46210,
      "train_runtime": 55026.1395,
      "train_tokens_per_second": 62311.775
    },
    {
      "epoch": 0.6142151879389506,
      "grad_norm": 17.875,
      "learning_rate": 6.842287200805862e-06,
      "loss": 0.8747,
      "num_input_tokens_seen": 3429616384,
      "step": 46220,
      "train_runtime": 55039.6467,
      "train_tokens_per_second": 62311.744
    },
    {
      "epoch": 0.6143480774214125,
      "grad_norm": 19.125,
      "learning_rate": 6.838203739989645e-06,
      "loss": 0.715,
      "num_input_tokens_seen": 3430104064,
      "step": 46230,
      "train_runtime": 55046.356,
      "train_tokens_per_second": 62313.009
    },
    {
      "epoch": 0.6144809669038744,
      "grad_norm": 11.3125,
      "learning_rate": 6.83412086486799e-06,
      "loss": 0.5759,
      "num_input_tokens_seen": 3430330880,
      "step": 46240,
      "train_runtime": 55049.8099,
      "train_tokens_per_second": 62313.219
    },
    {
      "epoch": 0.6146138563863363,
      "grad_norm": 13.25,
      "learning_rate": 6.830038576197218e-06,
      "loss": 0.5103,
      "num_input_tokens_seen": 3430443008,
      "step": 46250,
      "train_runtime": 55052.2195,
      "train_tokens_per_second": 62312.529
    },
    {
      "epoch": 0.6147467458687982,
      "grad_norm": 47.0,
      "learning_rate": 6.825956874733528e-06,
      "loss": 0.8389,
      "num_input_tokens_seen": 3432562944,
      "step": 46260,
      "train_runtime": 55085.9033,
      "train_tokens_per_second": 62312.91
    },
    {
      "epoch": 0.6148796353512601,
      "grad_norm": 28.25,
      "learning_rate": 6.82187576123303e-06,
      "loss": 0.9155,
      "num_input_tokens_seen": 3433423616,
      "step": 46270,
      "train_runtime": 55098.8687,
      "train_tokens_per_second": 62313.868
    },
    {
      "epoch": 0.615012524833722,
      "grad_norm": 17.5,
      "learning_rate": 6.817795236451705e-06,
      "loss": 0.6995,
      "num_input_tokens_seen": 3433885184,
      "step": 46280,
      "train_runtime": 55105.6347,
      "train_tokens_per_second": 62314.593
    },
    {
      "epoch": 0.615145414316184,
      "grad_norm": 11.5625,
      "learning_rate": 6.813715301145441e-06,
      "loss": 0.6222,
      "num_input_tokens_seen": 3434122496,
      "step": 46290,
      "train_runtime": 55109.2207,
      "train_tokens_per_second": 62314.844
    },
    {
      "epoch": 0.6152783037986459,
      "grad_norm": 135.0,
      "learning_rate": 6.809635956070004e-06,
      "loss": 0.5789,
      "num_input_tokens_seen": 3434238208,
      "step": 46300,
      "train_runtime": 55112.8304,
      "train_tokens_per_second": 62312.862
    },
    {
      "epoch": 0.6154111932811077,
      "grad_norm": 50.25,
      "learning_rate": 6.8055572019810545e-06,
      "loss": 0.8208,
      "num_input_tokens_seen": 3436369664,
      "step": 46310,
      "train_runtime": 55146.0722,
      "train_tokens_per_second": 62313.951
    },
    {
      "epoch": 0.6155440827635696,
      "grad_norm": 23.375,
      "learning_rate": 6.801479039634153e-06,
      "loss": 0.9908,
      "num_input_tokens_seen": 3437224704,
      "step": 46320,
      "train_runtime": 55158.7595,
      "train_tokens_per_second": 62315.12
    },
    {
      "epoch": 0.6156769722460316,
      "grad_norm": 12.1875,
      "learning_rate": 6.797401469784737e-06,
      "loss": 0.782,
      "num_input_tokens_seen": 3437682432,
      "step": 46330,
      "train_runtime": 55165.3105,
      "train_tokens_per_second": 62316.017
    },
    {
      "epoch": 0.6158098617284935,
      "grad_norm": 16.375,
      "learning_rate": 6.793324493188143e-06,
      "loss": 0.6038,
      "num_input_tokens_seen": 3437908992,
      "step": 46340,
      "train_runtime": 55168.7376,
      "train_tokens_per_second": 62316.253
    },
    {
      "epoch": 0.6159427512109554,
      "grad_norm": 12.3125,
      "learning_rate": 6.789248110599592e-06,
      "loss": 0.4548,
      "num_input_tokens_seen": 3438023424,
      "step": 46350,
      "train_runtime": 55171.1677,
      "train_tokens_per_second": 62315.582
    },
    {
      "epoch": 0.6160756406934174,
      "grad_norm": 18.375,
      "learning_rate": 6.785172322774201e-06,
      "loss": 0.7906,
      "num_input_tokens_seen": 3440089600,
      "step": 46360,
      "train_runtime": 55205.7415,
      "train_tokens_per_second": 62313.982
    },
    {
      "epoch": 0.6162085301758792,
      "grad_norm": 23.625,
      "learning_rate": 6.781097130466967e-06,
      "loss": 0.9151,
      "num_input_tokens_seen": 3440943616,
      "step": 46370,
      "train_runtime": 55218.7423,
      "train_tokens_per_second": 62314.777
    },
    {
      "epoch": 0.6163414196583411,
      "grad_norm": 13.875,
      "learning_rate": 6.777022534432791e-06,
      "loss": 0.8156,
      "num_input_tokens_seen": 3441429504,
      "step": 46380,
      "train_runtime": 55225.6558,
      "train_tokens_per_second": 62315.774
    },
    {
      "epoch": 0.616474309140803,
      "grad_norm": 17.875,
      "learning_rate": 6.7729485354264535e-06,
      "loss": 0.6695,
      "num_input_tokens_seen": 3441677824,
      "step": 46390,
      "train_runtime": 55229.3033,
      "train_tokens_per_second": 62316.155
    },
    {
      "epoch": 0.616607198623265,
      "grad_norm": 8.4375,
      "learning_rate": 6.768875134202621e-06,
      "loss": 0.5518,
      "num_input_tokens_seen": 3441796864,
      "step": 46400,
      "train_runtime": 55231.7247,
      "train_tokens_per_second": 62315.578
    },
    {
      "epoch": 0.6167400881057269,
      "grad_norm": 26.0,
      "learning_rate": 6.764802331515861e-06,
      "loss": 0.792,
      "num_input_tokens_seen": 3443858688,
      "step": 46410,
      "train_runtime": 55264.9012,
      "train_tokens_per_second": 62315.477
    },
    {
      "epoch": 0.6168729775881888,
      "grad_norm": 58.75,
      "learning_rate": 6.760730128120616e-06,
      "loss": 0.9052,
      "num_input_tokens_seen": 3444698368,
      "step": 46420,
      "train_runtime": 55277.4572,
      "train_tokens_per_second": 62316.513
    },
    {
      "epoch": 0.6170058670706507,
      "grad_norm": 12.0,
      "learning_rate": 6.756658524771236e-06,
      "loss": 0.7514,
      "num_input_tokens_seen": 3445188608,
      "step": 46430,
      "train_runtime": 55284.255,
      "train_tokens_per_second": 62317.718
    },
    {
      "epoch": 0.6171387565531126,
      "grad_norm": 39.5,
      "learning_rate": 6.752587522221942e-06,
      "loss": 0.673,
      "num_input_tokens_seen": 3445422080,
      "step": 46440,
      "train_runtime": 55287.7934,
      "train_tokens_per_second": 62317.952
    },
    {
      "epoch": 0.6172716460355745,
      "grad_norm": 34.75,
      "learning_rate": 6.7485171212268545e-06,
      "loss": 0.5584,
      "num_input_tokens_seen": 3445535744,
      "step": 46450,
      "train_runtime": 55290.2768,
      "train_tokens_per_second": 62317.209
    },
    {
      "epoch": 0.6174045355180364,
      "grad_norm": 17.5,
      "learning_rate": 6.744447322539977e-06,
      "loss": 0.7981,
      "num_input_tokens_seen": 3447567872,
      "step": 46460,
      "train_runtime": 55322.9248,
      "train_tokens_per_second": 62317.166
    },
    {
      "epoch": 0.6175374250004984,
      "grad_norm": 29.5,
      "learning_rate": 6.7403781269152e-06,
      "loss": 0.8984,
      "num_input_tokens_seen": 3448395008,
      "step": 46470,
      "train_runtime": 55336.1757,
      "train_tokens_per_second": 62317.191
    },
    {
      "epoch": 0.6176703144829603,
      "grad_norm": 14.75,
      "learning_rate": 6.736309535106313e-06,
      "loss": 0.6956,
      "num_input_tokens_seen": 3448854272,
      "step": 46480,
      "train_runtime": 55342.6997,
      "train_tokens_per_second": 62318.143
    },
    {
      "epoch": 0.6178032039654222,
      "grad_norm": 25.0,
      "learning_rate": 6.732241547866982e-06,
      "loss": 0.6732,
      "num_input_tokens_seen": 3449072896,
      "step": 46490,
      "train_runtime": 55346.0918,
      "train_tokens_per_second": 62318.274
    },
    {
      "epoch": 0.617936093447884,
      "grad_norm": 15.75,
      "learning_rate": 6.728174165950768e-06,
      "loss": 0.4727,
      "num_input_tokens_seen": 3449182976,
      "step": 46500,
      "train_runtime": 55348.4743,
      "train_tokens_per_second": 62317.58
    },
    {
      "epoch": 0.618068982930346,
      "grad_norm": 21.5,
      "learning_rate": 6.724107390111112e-06,
      "loss": 0.8309,
      "num_input_tokens_seen": 3451249152,
      "step": 46510,
      "train_runtime": 55382.9181,
      "train_tokens_per_second": 62316.131
    },
    {
      "epoch": 0.6182018724128079,
      "grad_norm": 29.875,
      "learning_rate": 6.720041221101355e-06,
      "loss": 0.8549,
      "num_input_tokens_seen": 3452103168,
      "step": 46520,
      "train_runtime": 55396.4915,
      "train_tokens_per_second": 62316.278
    },
    {
      "epoch": 0.6183347618952698,
      "grad_norm": 12.8125,
      "learning_rate": 6.71597565967471e-06,
      "loss": 0.7195,
      "num_input_tokens_seen": 3452557824,
      "step": 46530,
      "train_runtime": 55403.0984,
      "train_tokens_per_second": 62317.053
    },
    {
      "epoch": 0.6184676513777317,
      "grad_norm": 24.25,
      "learning_rate": 6.711910706584295e-06,
      "loss": 0.5898,
      "num_input_tokens_seen": 3452779008,
      "step": 46540,
      "train_runtime": 55406.4528,
      "train_tokens_per_second": 62317.272
    },
    {
      "epoch": 0.6186005408601937,
      "grad_norm": 10.1875,
      "learning_rate": 6.707846362583102e-06,
      "loss": 0.581,
      "num_input_tokens_seen": 3452892416,
      "step": 46550,
      "train_runtime": 55408.8093,
      "train_tokens_per_second": 62316.669
    },
    {
      "epoch": 0.6187334303426555,
      "grad_norm": 14.5,
      "learning_rate": 6.703782628424014e-06,
      "loss": 0.8612,
      "num_input_tokens_seen": 3454884864,
      "step": 46560,
      "train_runtime": 55441.3384,
      "train_tokens_per_second": 62316.044
    },
    {
      "epoch": 0.6188663198251174,
      "grad_norm": 23.0,
      "learning_rate": 6.699719504859803e-06,
      "loss": 0.8753,
      "num_input_tokens_seen": 3455721728,
      "step": 46570,
      "train_runtime": 55454.0118,
      "train_tokens_per_second": 62316.893
    },
    {
      "epoch": 0.6189992093075793,
      "grad_norm": 20.375,
      "learning_rate": 6.6956569926431224e-06,
      "loss": 0.7215,
      "num_input_tokens_seen": 3456178944,
      "step": 46580,
      "train_runtime": 55460.8462,
      "train_tokens_per_second": 62317.458
    },
    {
      "epoch": 0.6191320987900413,
      "grad_norm": 9.9375,
      "learning_rate": 6.691595092526524e-06,
      "loss": 0.6463,
      "num_input_tokens_seen": 3456408576,
      "step": 46590,
      "train_runtime": 55464.2313,
      "train_tokens_per_second": 62317.795
    },
    {
      "epoch": 0.6192649882725032,
      "grad_norm": 13.4375,
      "learning_rate": 6.687533805262432e-06,
      "loss": 0.5162,
      "num_input_tokens_seen": 3456522496,
      "step": 46600,
      "train_runtime": 55466.6185,
      "train_tokens_per_second": 62317.166
    },
    {
      "epoch": 0.6193978777549651,
      "grad_norm": 18.375,
      "learning_rate": 6.683473131603168e-06,
      "loss": 0.8334,
      "num_input_tokens_seen": 3458553600,
      "step": 46610,
      "train_runtime": 55499.7586,
      "train_tokens_per_second": 62316.552
    },
    {
      "epoch": 0.619530767237427,
      "grad_norm": 61.25,
      "learning_rate": 6.679413072300933e-06,
      "loss": 0.8799,
      "num_input_tokens_seen": 3459397632,
      "step": 46620,
      "train_runtime": 55512.4263,
      "train_tokens_per_second": 62317.536
    },
    {
      "epoch": 0.6196636567198889,
      "grad_norm": 17.625,
      "learning_rate": 6.6753536281078125e-06,
      "loss": 0.817,
      "num_input_tokens_seen": 3459843072,
      "step": 46630,
      "train_runtime": 55518.9942,
      "train_tokens_per_second": 62318.187
    },
    {
      "epoch": 0.6197965462023508,
      "grad_norm": 13.0,
      "learning_rate": 6.671294799775792e-06,
      "loss": 0.5723,
      "num_input_tokens_seen": 3460071168,
      "step": 46640,
      "train_runtime": 55523.0962,
      "train_tokens_per_second": 62317.691
    },
    {
      "epoch": 0.6199294356848127,
      "grad_norm": 10.875,
      "learning_rate": 6.667236588056727e-06,
      "loss": 0.5566,
      "num_input_tokens_seen": 3460189184,
      "step": 46650,
      "train_runtime": 55525.5417,
      "train_tokens_per_second": 62317.072
    },
    {
      "epoch": 0.6200623251672747,
      "grad_norm": 17.5,
      "learning_rate": 6.663178993702366e-06,
      "loss": 0.7692,
      "num_input_tokens_seen": 3462253056,
      "step": 46660,
      "train_runtime": 55559.041,
      "train_tokens_per_second": 62316.645
    },
    {
      "epoch": 0.6201952146497366,
      "grad_norm": 13.1875,
      "learning_rate": 6.659122017464341e-06,
      "loss": 0.9496,
      "num_input_tokens_seen": 3463088640,
      "step": 46670,
      "train_runtime": 55571.4309,
      "train_tokens_per_second": 62317.788
    },
    {
      "epoch": 0.6203281041321984,
      "grad_norm": 24.5,
      "learning_rate": 6.655065660094175e-06,
      "loss": 0.742,
      "num_input_tokens_seen": 3463530496,
      "step": 46680,
      "train_runtime": 55577.9323,
      "train_tokens_per_second": 62318.448
    },
    {
      "epoch": 0.6204609936146603,
      "grad_norm": 13.3125,
      "learning_rate": 6.651009922343266e-06,
      "loss": 0.6783,
      "num_input_tokens_seen": 3463759616,
      "step": 46690,
      "train_runtime": 55581.3431,
      "train_tokens_per_second": 62318.746
    },
    {
      "epoch": 0.6205938830971223,
      "grad_norm": 7.5625,
      "learning_rate": 6.64695480496291e-06,
      "loss": 0.513,
      "num_input_tokens_seen": 3463869696,
      "step": 46700,
      "train_runtime": 55584.5167,
      "train_tokens_per_second": 62317.169
    },
    {
      "epoch": 0.6207267725795842,
      "grad_norm": 23.75,
      "learning_rate": 6.642900308704278e-06,
      "loss": 0.8328,
      "num_input_tokens_seen": 3465867520,
      "step": 46710,
      "train_runtime": 55617.1896,
      "train_tokens_per_second": 62316.481
    },
    {
      "epoch": 0.6208596620620461,
      "grad_norm": 14.4375,
      "learning_rate": 6.638846434318426e-06,
      "loss": 0.9075,
      "num_input_tokens_seen": 3466700800,
      "step": 46720,
      "train_runtime": 55629.7713,
      "train_tokens_per_second": 62317.366
    },
    {
      "epoch": 0.620992551544508,
      "grad_norm": 21.75,
      "learning_rate": 6.634793182556305e-06,
      "loss": 0.7473,
      "num_input_tokens_seen": 3467164672,
      "step": 46730,
      "train_runtime": 55636.4221,
      "train_tokens_per_second": 62318.254
    },
    {
      "epoch": 0.6211254410269699,
      "grad_norm": 17.5,
      "learning_rate": 6.630740554168735e-06,
      "loss": 0.6269,
      "num_input_tokens_seen": 3467389952,
      "step": 46740,
      "train_runtime": 55639.8531,
      "train_tokens_per_second": 62318.46
    },
    {
      "epoch": 0.6212583305094318,
      "grad_norm": 10.5,
      "learning_rate": 6.626688549906438e-06,
      "loss": 0.5234,
      "num_input_tokens_seen": 3467506176,
      "step": 46750,
      "train_runtime": 55642.9916,
      "train_tokens_per_second": 62317.034
    },
    {
      "epoch": 0.6213912199918937,
      "grad_norm": 37.0,
      "learning_rate": 6.622637170520005e-06,
      "loss": 0.7887,
      "num_input_tokens_seen": 3469580544,
      "step": 46760,
      "train_runtime": 55676.3165,
      "train_tokens_per_second": 62316.992
    },
    {
      "epoch": 0.6215241094743557,
      "grad_norm": 37.5,
      "learning_rate": 6.618586416759924e-06,
      "loss": 0.9425,
      "num_input_tokens_seen": 3470425856,
      "step": 46770,
      "train_runtime": 55689.2353,
      "train_tokens_per_second": 62317.714
    },
    {
      "epoch": 0.6216569989568176,
      "grad_norm": 47.5,
      "learning_rate": 6.614536289376558e-06,
      "loss": 0.7215,
      "num_input_tokens_seen": 3470905856,
      "step": 46780,
      "train_runtime": 55695.9323,
      "train_tokens_per_second": 62318.839
    },
    {
      "epoch": 0.6217898884392795,
      "grad_norm": 20.25,
      "learning_rate": 6.610486789120153e-06,
      "loss": 0.6119,
      "num_input_tokens_seen": 3471141376,
      "step": 46790,
      "train_runtime": 55699.4188,
      "train_tokens_per_second": 62319.167
    },
    {
      "epoch": 0.6219227779217413,
      "grad_norm": 10.0,
      "learning_rate": 6.606437916740851e-06,
      "loss": 0.5459,
      "num_input_tokens_seen": 3471253504,
      "step": 46800,
      "train_runtime": 55701.8105,
      "train_tokens_per_second": 62318.504
    },
    {
      "epoch": 0.6220556674042033,
      "grad_norm": 39.0,
      "learning_rate": 6.602389672988661e-06,
      "loss": 0.7928,
      "num_input_tokens_seen": 3473475840,
      "step": 46810,
      "train_runtime": 55737.5235,
      "train_tokens_per_second": 62318.446
    },
    {
      "epoch": 0.6221885568866652,
      "grad_norm": 27.625,
      "learning_rate": 6.5983420586134914e-06,
      "loss": 0.8396,
      "num_input_tokens_seen": 3474324736,
      "step": 46820,
      "train_runtime": 55750.6287,
      "train_tokens_per_second": 62319.023
    },
    {
      "epoch": 0.6223214463691271,
      "grad_norm": 26.125,
      "learning_rate": 6.594295074365121e-06,
      "loss": 0.795,
      "num_input_tokens_seen": 3474810112,
      "step": 46830,
      "train_runtime": 55757.5943,
      "train_tokens_per_second": 62319.943
    },
    {
      "epoch": 0.622454335851589,
      "grad_norm": 11.0,
      "learning_rate": 6.590248720993221e-06,
      "loss": 0.6118,
      "num_input_tokens_seen": 3475056128,
      "step": 46840,
      "train_runtime": 55761.2083,
      "train_tokens_per_second": 62320.316
    },
    {
      "epoch": 0.622587225334051,
      "grad_norm": 20.125,
      "learning_rate": 6.586202999247337e-06,
      "loss": 0.5474,
      "num_input_tokens_seen": 3475174656,
      "step": 46850,
      "train_runtime": 55764.549,
      "train_tokens_per_second": 62318.708
    },
    {
      "epoch": 0.6227201148165128,
      "grad_norm": 27.5,
      "learning_rate": 6.582157909876914e-06,
      "loss": 0.8494,
      "num_input_tokens_seen": 3477336832,
      "step": 46860,
      "train_runtime": 55798.7257,
      "train_tokens_per_second": 62319.288
    },
    {
      "epoch": 0.6228530042989747,
      "grad_norm": 20.125,
      "learning_rate": 6.578113453631261e-06,
      "loss": 0.9132,
      "num_input_tokens_seen": 3478213632,
      "step": 46870,
      "train_runtime": 55811.5073,
      "train_tokens_per_second": 62320.726
    },
    {
      "epoch": 0.6229858937814367,
      "grad_norm": 18.75,
      "learning_rate": 6.574069631259576e-06,
      "loss": 0.705,
      "num_input_tokens_seen": 3478689792,
      "step": 46880,
      "train_runtime": 55818.2998,
      "train_tokens_per_second": 62321.672
    },
    {
      "epoch": 0.6231187832638986,
      "grad_norm": 10.1875,
      "learning_rate": 6.570026443510944e-06,
      "loss": 0.6921,
      "num_input_tokens_seen": 3478925056,
      "step": 46890,
      "train_runtime": 55821.8333,
      "train_tokens_per_second": 62321.942
    },
    {
      "epoch": 0.6232516727463605,
      "grad_norm": 592.0,
      "learning_rate": 6.565983891134326e-06,
      "loss": 0.5495,
      "num_input_tokens_seen": 3479041536,
      "step": 46900,
      "train_runtime": 55824.2696,
      "train_tokens_per_second": 62321.309
    },
    {
      "epoch": 0.6233845622288224,
      "grad_norm": 247.0,
      "learning_rate": 6.561941974878576e-06,
      "loss": 0.8487,
      "num_input_tokens_seen": 3481065216,
      "step": 46910,
      "train_runtime": 55857.2133,
      "train_tokens_per_second": 62320.782
    },
    {
      "epoch": 0.6235174517112844,
      "grad_norm": 36.25,
      "learning_rate": 6.557900695492416e-06,
      "loss": 0.8843,
      "num_input_tokens_seen": 3481903104,
      "step": 46920,
      "train_runtime": 55869.4887,
      "train_tokens_per_second": 62322.086
    },
    {
      "epoch": 0.6236503411937462,
      "grad_norm": 14.5,
      "learning_rate": 6.553860053724461e-06,
      "loss": 0.7073,
      "num_input_tokens_seen": 3482341888,
      "step": 46930,
      "train_runtime": 55877.3719,
      "train_tokens_per_second": 62321.147
    },
    {
      "epoch": 0.6237832306762081,
      "grad_norm": 13.8125,
      "learning_rate": 6.549820050323203e-06,
      "loss": 0.624,
      "num_input_tokens_seen": 3482563328,
      "step": 46940,
      "train_runtime": 55880.7384,
      "train_tokens_per_second": 62321.355
    },
    {
      "epoch": 0.62391612015867,
      "grad_norm": 18.875,
      "learning_rate": 6.545780686037013e-06,
      "loss": 0.5366,
      "num_input_tokens_seen": 3482676736,
      "step": 46950,
      "train_runtime": 55883.1691,
      "train_tokens_per_second": 62320.674
    },
    {
      "epoch": 0.624049009641132,
      "grad_norm": 24.375,
      "learning_rate": 6.541741961614154e-06,
      "loss": 0.8432,
      "num_input_tokens_seen": 3484697088,
      "step": 46960,
      "train_runtime": 55915.2733,
      "train_tokens_per_second": 62321.024
    },
    {
      "epoch": 0.6241818991235939,
      "grad_norm": 35.0,
      "learning_rate": 6.5377038778027565e-06,
      "loss": 0.8709,
      "num_input_tokens_seen": 3485549568,
      "step": 46970,
      "train_runtime": 55928.0152,
      "train_tokens_per_second": 62322.068
    },
    {
      "epoch": 0.6243147886060558,
      "grad_norm": 24.25,
      "learning_rate": 6.533666435350845e-06,
      "loss": 0.6662,
      "num_input_tokens_seen": 3486005504,
      "step": 46980,
      "train_runtime": 55934.5875,
      "train_tokens_per_second": 62322.896
    },
    {
      "epoch": 0.6244476780885176,
      "grad_norm": 20.375,
      "learning_rate": 6.529629635006315e-06,
      "loss": 0.6464,
      "num_input_tokens_seen": 3486235136,
      "step": 46990,
      "train_runtime": 55938.8871,
      "train_tokens_per_second": 62322.211
    },
    {
      "epoch": 0.6245805675709796,
      "grad_norm": 9.5,
      "learning_rate": 6.525593477516954e-06,
      "loss": 0.5586,
      "num_input_tokens_seen": 3486353664,
      "step": 47000,
      "train_runtime": 55941.3043,
      "train_tokens_per_second": 62321.637
    },
    {
      "epoch": 0.6247134570534415,
      "grad_norm": 19.25,
      "learning_rate": 6.521557963630418e-06,
      "loss": 0.7797,
      "num_input_tokens_seen": 3488379392,
      "step": 47010,
      "train_runtime": 55974.5845,
      "train_tokens_per_second": 62320.773
    },
    {
      "epoch": 0.6248463465359034,
      "grad_norm": 19.0,
      "learning_rate": 6.517523094094254e-06,
      "loss": 0.9037,
      "num_input_tokens_seen": 3489213952,
      "step": 47020,
      "train_runtime": 55987.0713,
      "train_tokens_per_second": 62321.78
    },
    {
      "epoch": 0.6249792360183654,
      "grad_norm": 37.0,
      "learning_rate": 6.513488869655885e-06,
      "loss": 0.7016,
      "num_input_tokens_seen": 3489683712,
      "step": 47030,
      "train_runtime": 55993.6004,
      "train_tokens_per_second": 62322.903
    },
    {
      "epoch": 0.6251121255008273,
      "grad_norm": 14.4375,
      "learning_rate": 6.50945529106261e-06,
      "loss": 0.5912,
      "num_input_tokens_seen": 3489913088,
      "step": 47040,
      "train_runtime": 55998.8969,
      "train_tokens_per_second": 62321.104
    },
    {
      "epoch": 0.6252450149832891,
      "grad_norm": 1392.0,
      "learning_rate": 6.5054223590616216e-06,
      "loss": 0.5677,
      "num_input_tokens_seen": 3490031104,
      "step": 47050,
      "train_runtime": 56001.4987,
      "train_tokens_per_second": 62320.316
    },
    {
      "epoch": 0.625377904465751,
      "grad_norm": 24.5,
      "learning_rate": 6.501390074399978e-06,
      "loss": 0.8359,
      "num_input_tokens_seen": 3492108032,
      "step": 47060,
      "train_runtime": 56036.1134,
      "train_tokens_per_second": 62318.884
    },
    {
      "epoch": 0.625510793948213,
      "grad_norm": 11.5,
      "learning_rate": 6.497358437824629e-06,
      "loss": 0.9558,
      "num_input_tokens_seen": 3492940032,
      "step": 47070,
      "train_runtime": 56048.969,
      "train_tokens_per_second": 62319.434
    },
    {
      "epoch": 0.6256436834306749,
      "grad_norm": 19.375,
      "learning_rate": 6.493327450082395e-06,
      "loss": 0.7124,
      "num_input_tokens_seen": 3493415424,
      "step": 47080,
      "train_runtime": 56055.747,
      "train_tokens_per_second": 62320.379
    },
    {
      "epoch": 0.6257765729131368,
      "grad_norm": 22.875,
      "learning_rate": 6.489297111919985e-06,
      "loss": 0.5913,
      "num_input_tokens_seen": 3493643008,
      "step": 47090,
      "train_runtime": 56059.2661,
      "train_tokens_per_second": 62320.527
    },
    {
      "epoch": 0.6259094623955987,
      "grad_norm": 19.0,
      "learning_rate": 6.485267424083981e-06,
      "loss": 0.5339,
      "num_input_tokens_seen": 3493756928,
      "step": 47100,
      "train_runtime": 56061.6933,
      "train_tokens_per_second": 62319.861
    },
    {
      "epoch": 0.6260423518780606,
      "grad_norm": 19.75,
      "learning_rate": 6.48123838732084e-06,
      "loss": 0.7448,
      "num_input_tokens_seen": 3495794944,
      "step": 47110,
      "train_runtime": 56095.2078,
      "train_tokens_per_second": 62318.959
    },
    {
      "epoch": 0.6261752413605225,
      "grad_norm": 37.25,
      "learning_rate": 6.477210002376919e-06,
      "loss": 0.8523,
      "num_input_tokens_seen": 3496639744,
      "step": 47120,
      "train_runtime": 56107.8426,
      "train_tokens_per_second": 62319.982
    },
    {
      "epoch": 0.6263081308429844,
      "grad_norm": 15.3125,
      "learning_rate": 6.473182269998428e-06,
      "loss": 0.7235,
      "num_input_tokens_seen": 3497102336,
      "step": 47130,
      "train_runtime": 56114.4842,
      "train_tokens_per_second": 62320.85
    },
    {
      "epoch": 0.6264410203254464,
      "grad_norm": 20.5,
      "learning_rate": 6.469155190931477e-06,
      "loss": 0.7151,
      "num_input_tokens_seen": 3497334272,
      "step": 47140,
      "train_runtime": 56117.9291,
      "train_tokens_per_second": 62321.157
    },
    {
      "epoch": 0.6265739098079083,
      "grad_norm": 12.9375,
      "learning_rate": 6.46512876592204e-06,
      "loss": 0.5916,
      "num_input_tokens_seen": 3497447168,
      "step": 47150,
      "train_runtime": 56120.402,
      "train_tokens_per_second": 62320.423
    },
    {
      "epoch": 0.6267067992903702,
      "grad_norm": 20.875,
      "learning_rate": 6.461102995715981e-06,
      "loss": 0.8326,
      "num_input_tokens_seen": 3499555584,
      "step": 47160,
      "train_runtime": 56155.199,
      "train_tokens_per_second": 62319.351
    },
    {
      "epoch": 0.626839688772832,
      "grad_norm": 16.875,
      "learning_rate": 6.457077881059035e-06,
      "loss": 0.8675,
      "num_input_tokens_seen": 3500381440,
      "step": 47170,
      "train_runtime": 56167.4342,
      "train_tokens_per_second": 62320.48
    },
    {
      "epoch": 0.626972578255294,
      "grad_norm": 49.75,
      "learning_rate": 6.4530534226968226e-06,
      "loss": 0.7857,
      "num_input_tokens_seen": 3500808192,
      "step": 47180,
      "train_runtime": 56173.7766,
      "train_tokens_per_second": 62321.04
    },
    {
      "epoch": 0.6271054677377559,
      "grad_norm": 13.875,
      "learning_rate": 6.449029621374836e-06,
      "loss": 0.6753,
      "num_input_tokens_seen": 3501036544,
      "step": 47190,
      "train_runtime": 56177.2226,
      "train_tokens_per_second": 62321.282
    },
    {
      "epoch": 0.6272383572202178,
      "grad_norm": 21.625,
      "learning_rate": 6.445006477838444e-06,
      "loss": 0.5498,
      "num_input_tokens_seen": 3501149440,
      "step": 47200,
      "train_runtime": 56179.6616,
      "train_tokens_per_second": 62320.586
    },
    {
      "epoch": 0.6273712467026797,
      "grad_norm": 110.0,
      "learning_rate": 6.440983992832906e-06,
      "loss": 0.8305,
      "num_input_tokens_seen": 3503112192,
      "step": 47210,
      "train_runtime": 56212.9642,
      "train_tokens_per_second": 62318.581
    },
    {
      "epoch": 0.6275041361851417,
      "grad_norm": 19.0,
      "learning_rate": 6.436962167103346e-06,
      "loss": 0.8987,
      "num_input_tokens_seen": 3503972864,
      "step": 47220,
      "train_runtime": 56225.549,
      "train_tokens_per_second": 62319.94
    },
    {
      "epoch": 0.6276370256676035,
      "grad_norm": 31.625,
      "learning_rate": 6.4329410013947765e-06,
      "loss": 0.7238,
      "num_input_tokens_seen": 3504439296,
      "step": 47230,
      "train_runtime": 56232.1928,
      "train_tokens_per_second": 62320.872
    },
    {
      "epoch": 0.6277699151500654,
      "grad_norm": 26.625,
      "learning_rate": 6.428920496452078e-06,
      "loss": 0.6923,
      "num_input_tokens_seen": 3504669696,
      "step": 47240,
      "train_runtime": 56235.6879,
      "train_tokens_per_second": 62321.096
    },
    {
      "epoch": 0.6279028046325273,
      "grad_norm": 9.25,
      "learning_rate": 6.424900653020016e-06,
      "loss": 0.4906,
      "num_input_tokens_seen": 3504788992,
      "step": 47250,
      "train_runtime": 56238.1087,
      "train_tokens_per_second": 62320.534
    },
    {
      "epoch": 0.6280356941149893,
      "grad_norm": 25.875,
      "learning_rate": 6.42088147184323e-06,
      "loss": 0.8316,
      "num_input_tokens_seen": 3506787072,
      "step": 47260,
      "train_runtime": 56271.501,
      "train_tokens_per_second": 62319.06
    },
    {
      "epoch": 0.6281685835974512,
      "grad_norm": 25.375,
      "learning_rate": 6.416862953666232e-06,
      "loss": 0.9893,
      "num_input_tokens_seen": 3507641856,
      "step": 47270,
      "train_runtime": 56284.1767,
      "train_tokens_per_second": 62320.213
    },
    {
      "epoch": 0.6283014730799131,
      "grad_norm": 15.25,
      "learning_rate": 6.412845099233427e-06,
      "loss": 0.7254,
      "num_input_tokens_seen": 3508105216,
      "step": 47280,
      "train_runtime": 56291.4978,
      "train_tokens_per_second": 62320.339
    },
    {
      "epoch": 0.628434362562375,
      "grad_norm": 9.625,
      "learning_rate": 6.4088279092890795e-06,
      "loss": 0.6085,
      "num_input_tokens_seen": 3508330496,
      "step": 47290,
      "train_runtime": 56294.966,
      "train_tokens_per_second": 62320.501
    },
    {
      "epoch": 0.6285672520448369,
      "grad_norm": 10.75,
      "learning_rate": 6.404811384577341e-06,
      "loss": 0.4852,
      "num_input_tokens_seen": 3508442368,
      "step": 47300,
      "train_runtime": 56297.4167,
      "train_tokens_per_second": 62319.775
    },
    {
      "epoch": 0.6287001415272988,
      "grad_norm": 16.625,
      "learning_rate": 6.4007955258422315e-06,
      "loss": 0.7424,
      "num_input_tokens_seen": 3510532352,
      "step": 47310,
      "train_runtime": 56330.8534,
      "train_tokens_per_second": 62319.886
    },
    {
      "epoch": 0.6288330310097607,
      "grad_norm": 15.5,
      "learning_rate": 6.396780333827661e-06,
      "loss": 0.9252,
      "num_input_tokens_seen": 3511371008,
      "step": 47320,
      "train_runtime": 56343.7298,
      "train_tokens_per_second": 62320.528
    },
    {
      "epoch": 0.6289659204922227,
      "grad_norm": 31.875,
      "learning_rate": 6.392765809277403e-06,
      "loss": 0.8093,
      "num_input_tokens_seen": 3511880960,
      "step": 47330,
      "train_runtime": 56351.6492,
      "train_tokens_per_second": 62320.82
    },
    {
      "epoch": 0.6290988099746846,
      "grad_norm": 16.0,
      "learning_rate": 6.3887519529351175e-06,
      "loss": 0.6583,
      "num_input_tokens_seen": 3512120832,
      "step": 47340,
      "train_runtime": 56355.2204,
      "train_tokens_per_second": 62321.127
    },
    {
      "epoch": 0.6292316994571464,
      "grad_norm": 7.6875,
      "learning_rate": 6.38473876554433e-06,
      "loss": 0.5744,
      "num_input_tokens_seen": 3512238080,
      "step": 47350,
      "train_runtime": 56357.6283,
      "train_tokens_per_second": 62320.544
    },
    {
      "epoch": 0.6293645889396083,
      "grad_norm": 19.625,
      "learning_rate": 6.3807262478484456e-06,
      "loss": 0.8363,
      "num_input_tokens_seen": 3514223104,
      "step": 47360,
      "train_runtime": 56390.2961,
      "train_tokens_per_second": 62319.643
    },
    {
      "epoch": 0.6294974784220703,
      "grad_norm": 34.0,
      "learning_rate": 6.376714400590755e-06,
      "loss": 0.8811,
      "num_input_tokens_seen": 3515064064,
      "step": 47370,
      "train_runtime": 56403.7157,
      "train_tokens_per_second": 62319.725
    },
    {
      "epoch": 0.6296303679045322,
      "grad_norm": 21.0,
      "learning_rate": 6.37270322451441e-06,
      "loss": 0.7429,
      "num_input_tokens_seen": 3515497984,
      "step": 47380,
      "train_runtime": 56410.1869,
      "train_tokens_per_second": 62320.268
    },
    {
      "epoch": 0.6297632573869941,
      "grad_norm": 33.25,
      "learning_rate": 6.3686927203624526e-06,
      "loss": 0.6619,
      "num_input_tokens_seen": 3515719936,
      "step": 47390,
      "train_runtime": 56413.6078,
      "train_tokens_per_second": 62320.424
    },
    {
      "epoch": 0.629896146869456,
      "grad_norm": 22.875,
      "learning_rate": 6.364682888877787e-06,
      "loss": 0.5138,
      "num_input_tokens_seen": 3515835648,
      "step": 47400,
      "train_runtime": 56416.0439,
      "train_tokens_per_second": 62319.784
    },
    {
      "epoch": 0.630029036351918,
      "grad_norm": 21.0,
      "learning_rate": 6.360673730803202e-06,
      "loss": 0.8884,
      "num_input_tokens_seen": 3517908224,
      "step": 47410,
      "train_runtime": 56448.4102,
      "train_tokens_per_second": 62320.767
    },
    {
      "epoch": 0.6301619258343798,
      "grad_norm": 27.625,
      "learning_rate": 6.356665246881357e-06,
      "loss": 0.8983,
      "num_input_tokens_seen": 3518764544,
      "step": 47420,
      "train_runtime": 56461.2209,
      "train_tokens_per_second": 62321.793
    },
    {
      "epoch": 0.6302948153168417,
      "grad_norm": 153.0,
      "learning_rate": 6.352657437854786e-06,
      "loss": 0.7874,
      "num_input_tokens_seen": 3519244032,
      "step": 47430,
      "train_runtime": 56468.0145,
      "train_tokens_per_second": 62322.787
    },
    {
      "epoch": 0.6304277047993037,
      "grad_norm": 17.125,
      "learning_rate": 6.3486503044659046e-06,
      "loss": 0.6055,
      "num_input_tokens_seen": 3519478272,
      "step": 47440,
      "train_runtime": 56471.5581,
      "train_tokens_per_second": 62323.024
    },
    {
      "epoch": 0.6305605942817656,
      "grad_norm": 14.1875,
      "learning_rate": 6.344643847456993e-06,
      "loss": 0.5181,
      "num_input_tokens_seen": 3519592704,
      "step": 47450,
      "train_runtime": 56474.8294,
      "train_tokens_per_second": 62321.44
    },
    {
      "epoch": 0.6306934837642275,
      "grad_norm": 21.375,
      "learning_rate": 6.3406380675702175e-06,
      "loss": 0.8503,
      "num_input_tokens_seen": 3521585664,
      "step": 47460,
      "train_runtime": 56506.1706,
      "train_tokens_per_second": 62322.143
    },
    {
      "epoch": 0.6308263732466894,
      "grad_norm": 20.875,
      "learning_rate": 6.336632965547605e-06,
      "loss": 0.9009,
      "num_input_tokens_seen": 3522421504,
      "step": 47470,
      "train_runtime": 56518.6252,
      "train_tokens_per_second": 62323.199
    },
    {
      "epoch": 0.6309592627291513,
      "grad_norm": 19.0,
      "learning_rate": 6.3326285421310755e-06,
      "loss": 0.6831,
      "num_input_tokens_seen": 3522878464,
      "step": 47480,
      "train_runtime": 56525.1129,
      "train_tokens_per_second": 62324.13
    },
    {
      "epoch": 0.6310921522116132,
      "grad_norm": 24.375,
      "learning_rate": 6.328624798062404e-06,
      "loss": 0.6599,
      "num_input_tokens_seen": 3523097856,
      "step": 47490,
      "train_runtime": 56528.4696,
      "train_tokens_per_second": 62324.31
    },
    {
      "epoch": 0.6312250416940751,
      "grad_norm": 35.75,
      "learning_rate": 6.324621734083254e-06,
      "loss": 0.5257,
      "num_input_tokens_seen": 3523211264,
      "step": 47500,
      "train_runtime": 56531.6425,
      "train_tokens_per_second": 62322.818
    },
    {
      "epoch": 0.631357931176537,
      "grad_norm": 31.25,
      "learning_rate": 6.320619350935155e-06,
      "loss": 0.8857,
      "num_input_tokens_seen": 3525300224,
      "step": 47510,
      "train_runtime": 56565.8991,
      "train_tokens_per_second": 62322.005
    },
    {
      "epoch": 0.631490820658999,
      "grad_norm": 186.0,
      "learning_rate": 6.3166176493595085e-06,
      "loss": 0.9104,
      "num_input_tokens_seen": 3526170880,
      "step": 47520,
      "train_runtime": 56578.9982,
      "train_tokens_per_second": 62322.964
    },
    {
      "epoch": 0.6316237101414609,
      "grad_norm": 33.5,
      "learning_rate": 6.3126166300976035e-06,
      "loss": 0.7054,
      "num_input_tokens_seen": 3526656512,
      "step": 47530,
      "train_runtime": 56585.8041,
      "train_tokens_per_second": 62324.05
    },
    {
      "epoch": 0.6317565996239227,
      "grad_norm": 19.0,
      "learning_rate": 6.308616293890585e-06,
      "loss": 0.6108,
      "num_input_tokens_seen": 3526886656,
      "step": 47540,
      "train_runtime": 56589.294,
      "train_tokens_per_second": 62324.274
    },
    {
      "epoch": 0.6318894891063846,
      "grad_norm": 113.0,
      "learning_rate": 6.3046166414794845e-06,
      "loss": 0.5548,
      "num_input_tokens_seen": 3526999296,
      "step": 47550,
      "train_runtime": 56591.7315,
      "train_tokens_per_second": 62323.58
    },
    {
      "epoch": 0.6320223785888466,
      "grad_norm": 97.0,
      "learning_rate": 6.300617673605196e-06,
      "loss": 0.81,
      "num_input_tokens_seen": 3529037312,
      "step": 47560,
      "train_runtime": 56624.7609,
      "train_tokens_per_second": 62323.218
    },
    {
      "epoch": 0.6321552680713085,
      "grad_norm": 53.0,
      "learning_rate": 6.296619391008501e-06,
      "loss": 0.8475,
      "num_input_tokens_seen": 3529894400,
      "step": 47570,
      "train_runtime": 56637.5489,
      "train_tokens_per_second": 62324.279
    },
    {
      "epoch": 0.6322881575537704,
      "grad_norm": 24.0,
      "learning_rate": 6.292621794430043e-06,
      "loss": 0.7943,
      "num_input_tokens_seen": 3530360832,
      "step": 47580,
      "train_runtime": 56644.7648,
      "train_tokens_per_second": 62324.574
    },
    {
      "epoch": 0.6324210470362324,
      "grad_norm": 27.625,
      "learning_rate": 6.2886248846103324e-06,
      "loss": 0.6333,
      "num_input_tokens_seen": 3530583808,
      "step": 47590,
      "train_runtime": 56648.118,
      "train_tokens_per_second": 62324.821
    },
    {
      "epoch": 0.6325539365186942,
      "grad_norm": 9.9375,
      "learning_rate": 6.284628662289774e-06,
      "loss": 0.5332,
      "num_input_tokens_seen": 3530697728,
      "step": 47600,
      "train_runtime": 56650.6068,
      "train_tokens_per_second": 62324.094
    },
    {
      "epoch": 0.6326868260011561,
      "grad_norm": 13.1875,
      "learning_rate": 6.280633128208624e-06,
      "loss": 0.8817,
      "num_input_tokens_seen": 3532766720,
      "step": 47610,
      "train_runtime": 56684.2289,
      "train_tokens_per_second": 62323.627
    },
    {
      "epoch": 0.632819715483618,
      "grad_norm": 60.25,
      "learning_rate": 6.2766382831070246e-06,
      "loss": 0.8778,
      "num_input_tokens_seen": 3533612544,
      "step": 47620,
      "train_runtime": 56696.9675,
      "train_tokens_per_second": 62324.542
    },
    {
      "epoch": 0.63295260496608,
      "grad_norm": 12.6875,
      "learning_rate": 6.2726441277249784e-06,
      "loss": 0.7002,
      "num_input_tokens_seen": 3534060800,
      "step": 47630,
      "train_runtime": 56703.5826,
      "train_tokens_per_second": 62325.177
    },
    {
      "epoch": 0.6330854944485419,
      "grad_norm": 11.9375,
      "learning_rate": 6.2686506628023754e-06,
      "loss": 0.674,
      "num_input_tokens_seen": 3534283008,
      "step": 47640,
      "train_runtime": 56707.7747,
      "train_tokens_per_second": 62324.488
    },
    {
      "epoch": 0.6332183839310038,
      "grad_norm": 29.125,
      "learning_rate": 6.264657889078963e-06,
      "loss": 0.5625,
      "num_input_tokens_seen": 3534396928,
      "step": 47650,
      "train_runtime": 56710.2108,
      "train_tokens_per_second": 62323.819
    },
    {
      "epoch": 0.6333512734134656,
      "grad_norm": 24.125,
      "learning_rate": 6.260665807294371e-06,
      "loss": 0.8377,
      "num_input_tokens_seen": 3536502272,
      "step": 47660,
      "train_runtime": 56743.8769,
      "train_tokens_per_second": 62323.945
    },
    {
      "epoch": 0.6334841628959276,
      "grad_norm": 87.0,
      "learning_rate": 6.2566744181880954e-06,
      "loss": 0.9237,
      "num_input_tokens_seen": 3537329408,
      "step": 47670,
      "train_runtime": 56756.6257,
      "train_tokens_per_second": 62324.519
    },
    {
      "epoch": 0.6336170523783895,
      "grad_norm": 16.75,
      "learning_rate": 6.252683722499502e-06,
      "loss": 0.7354,
      "num_input_tokens_seen": 3537771776,
      "step": 47680,
      "train_runtime": 56763.7732,
      "train_tokens_per_second": 62324.465
    },
    {
      "epoch": 0.6337499418608514,
      "grad_norm": 18.625,
      "learning_rate": 6.248693720967838e-06,
      "loss": 0.6309,
      "num_input_tokens_seen": 3538006016,
      "step": 47690,
      "train_runtime": 56767.2412,
      "train_tokens_per_second": 62324.783
    },
    {
      "epoch": 0.6338828313433134,
      "grad_norm": 15.9375,
      "learning_rate": 6.244704414332211e-06,
      "loss": 0.4558,
      "num_input_tokens_seen": 3538117120,
      "step": 47700,
      "train_runtime": 56769.6744,
      "train_tokens_per_second": 62324.069
    },
    {
      "epoch": 0.6340157208257753,
      "grad_norm": 24.5,
      "learning_rate": 6.2407158033316075e-06,
      "loss": 0.8808,
      "num_input_tokens_seen": 3540116224,
      "step": 47710,
      "train_runtime": 56802.6123,
      "train_tokens_per_second": 62323.123
    },
    {
      "epoch": 0.6341486103082371,
      "grad_norm": 47.25,
      "learning_rate": 6.2367278887048766e-06,
      "loss": 0.8633,
      "num_input_tokens_seen": 3540953600,
      "step": 47720,
      "train_runtime": 56815.1993,
      "train_tokens_per_second": 62324.055
    },
    {
      "epoch": 0.634281499790699,
      "grad_norm": 408.0,
      "learning_rate": 6.232740671190751e-06,
      "loss": 0.7174,
      "num_input_tokens_seen": 3541445888,
      "step": 47730,
      "train_runtime": 56822.0253,
      "train_tokens_per_second": 62325.232
    },
    {
      "epoch": 0.634414389273161,
      "grad_norm": 17.375,
      "learning_rate": 6.228754151527825e-06,
      "loss": 0.6958,
      "num_input_tokens_seen": 3541678848,
      "step": 47740,
      "train_runtime": 56825.5321,
      "train_tokens_per_second": 62325.485
    },
    {
      "epoch": 0.6345472787556229,
      "grad_norm": 14.875,
      "learning_rate": 6.2247683304545604e-06,
      "loss": 0.5495,
      "num_input_tokens_seen": 3541793792,
      "step": 47750,
      "train_runtime": 56827.9875,
      "train_tokens_per_second": 62324.815
    },
    {
      "epoch": 0.6346801682380848,
      "grad_norm": 18.125,
      "learning_rate": 6.220783208709304e-06,
      "loss": 0.8615,
      "num_input_tokens_seen": 3543793152,
      "step": 47760,
      "train_runtime": 56858.613,
      "train_tokens_per_second": 62326.409
    },
    {
      "epoch": 0.6348130577205467,
      "grad_norm": 39.25,
      "learning_rate": 6.216798787030258e-06,
      "loss": 0.8736,
      "num_input_tokens_seen": 3544608768,
      "step": 47770,
      "train_runtime": 56870.8779,
      "train_tokens_per_second": 62327.309
    },
    {
      "epoch": 0.6349459472030086,
      "grad_norm": 38.75,
      "learning_rate": 6.212815066155503e-06,
      "loss": 0.7289,
      "num_input_tokens_seen": 3545081856,
      "step": 47780,
      "train_runtime": 56877.5114,
      "train_tokens_per_second": 62328.357
    },
    {
      "epoch": 0.6350788366854705,
      "grad_norm": 19.875,
      "learning_rate": 6.208832046822983e-06,
      "loss": 0.7125,
      "num_input_tokens_seen": 3545307904,
      "step": 47790,
      "train_runtime": 56880.9611,
      "train_tokens_per_second": 62328.551
    },
    {
      "epoch": 0.6352117261679324,
      "grad_norm": 13.1875,
      "learning_rate": 6.204849729770528e-06,
      "loss": 0.5564,
      "num_input_tokens_seen": 3545421824,
      "step": 47800,
      "train_runtime": 56883.3498,
      "train_tokens_per_second": 62327.937
    },
    {
      "epoch": 0.6353446156503944,
      "grad_norm": 23.375,
      "learning_rate": 6.200868115735816e-06,
      "loss": 0.744,
      "num_input_tokens_seen": 3547569920,
      "step": 47810,
      "train_runtime": 56917.5318,
      "train_tokens_per_second": 62328.246
    },
    {
      "epoch": 0.6354775051328563,
      "grad_norm": 25.625,
      "learning_rate": 6.19688720545641e-06,
      "loss": 0.8811,
      "num_input_tokens_seen": 3548404480,
      "step": 47820,
      "train_runtime": 56930.6344,
      "train_tokens_per_second": 62328.56
    },
    {
      "epoch": 0.6356103946153182,
      "grad_norm": 15.375,
      "learning_rate": 6.192906999669739e-06,
      "loss": 0.7001,
      "num_input_tokens_seen": 3548845824,
      "step": 47830,
      "train_runtime": 56937.0803,
      "train_tokens_per_second": 62329.255
    },
    {
      "epoch": 0.63574328409778,
      "grad_norm": 41.75,
      "learning_rate": 6.188927499113095e-06,
      "loss": 0.6571,
      "num_input_tokens_seen": 3549063424,
      "step": 47840,
      "train_runtime": 56940.4234,
      "train_tokens_per_second": 62329.418
    },
    {
      "epoch": 0.635876173580242,
      "grad_norm": 15.5625,
      "learning_rate": 6.184948704523652e-06,
      "loss": 0.5276,
      "num_input_tokens_seen": 3549174784,
      "step": 47850,
      "train_runtime": 56942.8013,
      "train_tokens_per_second": 62328.77
    },
    {
      "epoch": 0.6360090630627039,
      "grad_norm": 21.25,
      "learning_rate": 6.180970616638443e-06,
      "loss": 0.7807,
      "num_input_tokens_seen": 3551222784,
      "step": 47860,
      "train_runtime": 56976.923,
      "train_tokens_per_second": 62327.388
    },
    {
      "epoch": 0.6361419525451658,
      "grad_norm": 46.0,
      "learning_rate": 6.176993236194375e-06,
      "loss": 0.9163,
      "num_input_tokens_seen": 3552085248,
      "step": 47870,
      "train_runtime": 56990.2495,
      "train_tokens_per_second": 62327.947
    },
    {
      "epoch": 0.6362748420276277,
      "grad_norm": 15.375,
      "learning_rate": 6.173016563928216e-06,
      "loss": 0.8192,
      "num_input_tokens_seen": 3552568320,
      "step": 47880,
      "train_runtime": 56997.1603,
      "train_tokens_per_second": 62328.865
    },
    {
      "epoch": 0.6364077315100897,
      "grad_norm": 11.6875,
      "learning_rate": 6.169040600576619e-06,
      "loss": 0.6123,
      "num_input_tokens_seen": 3552796928,
      "step": 47890,
      "train_runtime": 57000.5996,
      "train_tokens_per_second": 62329.115
    },
    {
      "epoch": 0.6365406209925516,
      "grad_norm": 11.5625,
      "learning_rate": 6.165065346876091e-06,
      "loss": 0.5299,
      "num_input_tokens_seen": 3552908288,
      "step": 47900,
      "train_runtime": 57002.9767,
      "train_tokens_per_second": 62328.469
    },
    {
      "epoch": 0.6366735104750134,
      "grad_norm": 16.75,
      "learning_rate": 6.161090803563007e-06,
      "loss": 0.9037,
      "num_input_tokens_seen": 3554865664,
      "step": 47910,
      "train_runtime": 57035.9787,
      "train_tokens_per_second": 62326.723
    },
    {
      "epoch": 0.6368063999574753,
      "grad_norm": 17.0,
      "learning_rate": 6.157116971373624e-06,
      "loss": 0.8189,
      "num_input_tokens_seen": 3555684096,
      "step": 47920,
      "train_runtime": 57048.2604,
      "train_tokens_per_second": 62327.652
    },
    {
      "epoch": 0.6369392894399373,
      "grad_norm": 20.875,
      "learning_rate": 6.153143851044055e-06,
      "loss": 0.7121,
      "num_input_tokens_seen": 3556146432,
      "step": 47930,
      "train_runtime": 57054.8298,
      "train_tokens_per_second": 62328.578
    },
    {
      "epoch": 0.6370721789223992,
      "grad_norm": 12.25,
      "learning_rate": 6.149171443310289e-06,
      "loss": 0.654,
      "num_input_tokens_seen": 3556380416,
      "step": 47940,
      "train_runtime": 57058.3161,
      "train_tokens_per_second": 62328.871
    },
    {
      "epoch": 0.6372050684048611,
      "grad_norm": 13.375,
      "learning_rate": 6.1451997489081705e-06,
      "loss": 0.5718,
      "num_input_tokens_seen": 3556495616,
      "step": 47950,
      "train_runtime": 57060.7501,
      "train_tokens_per_second": 62328.231
    },
    {
      "epoch": 0.6373379578873231,
      "grad_norm": 27.375,
      "learning_rate": 6.141228768573429e-06,
      "loss": 0.8552,
      "num_input_tokens_seen": 3558466304,
      "step": 47960,
      "train_runtime": 57092.0458,
      "train_tokens_per_second": 62328.583
    },
    {
      "epoch": 0.6374708473697849,
      "grad_norm": 16.875,
      "learning_rate": 6.13725850304165e-06,
      "loss": 0.8545,
      "num_input_tokens_seen": 3559294464,
      "step": 47970,
      "train_runtime": 57104.41,
      "train_tokens_per_second": 62329.59
    },
    {
      "epoch": 0.6376037368522468,
      "grad_norm": 18.5,
      "learning_rate": 6.1332889530482905e-06,
      "loss": 0.6994,
      "num_input_tokens_seen": 3559755008,
      "step": 47980,
      "train_runtime": 57111.0396,
      "train_tokens_per_second": 62330.419
    },
    {
      "epoch": 0.6377366263347087,
      "grad_norm": 19.0,
      "learning_rate": 6.129320119328675e-06,
      "loss": 0.6593,
      "num_input_tokens_seen": 3559980800,
      "step": 47990,
      "train_runtime": 57115.241,
      "train_tokens_per_second": 62329.787
    },
    {
      "epoch": 0.6378695158171707,
      "grad_norm": 7.15625,
      "learning_rate": 6.125352002617987e-06,
      "loss": 0.5548,
      "num_input_tokens_seen": 3560095488,
      "step": 48000,
      "train_runtime": 57117.6379,
      "train_tokens_per_second": 62329.179
    },
    {
      "epoch": 0.6380024052996326,
      "grad_norm": 17.125,
      "learning_rate": 6.121384603651296e-06,
      "loss": 0.8119,
      "num_input_tokens_seen": 3562221824,
      "step": 48010,
      "train_runtime": 57151.9875,
      "train_tokens_per_second": 62328.923
    },
    {
      "epoch": 0.6381352947820945,
      "grad_norm": 19.625,
      "learning_rate": 6.11741792316352e-06,
      "loss": 0.9218,
      "num_input_tokens_seen": 3563038976,
      "step": 48020,
      "train_runtime": 57164.5752,
      "train_tokens_per_second": 62329.493
    },
    {
      "epoch": 0.6382681842645563,
      "grad_norm": 56.0,
      "learning_rate": 6.113451961889456e-06,
      "loss": 0.7761,
      "num_input_tokens_seen": 3563488256,
      "step": 48030,
      "train_runtime": 57171.0838,
      "train_tokens_per_second": 62330.255
    },
    {
      "epoch": 0.6384010737470183,
      "grad_norm": 37.5,
      "learning_rate": 6.109486720563755e-06,
      "loss": 0.681,
      "num_input_tokens_seen": 3563714560,
      "step": 48040,
      "train_runtime": 57175.2416,
      "train_tokens_per_second": 62329.681
    },
    {
      "epoch": 0.6385339632294802,
      "grad_norm": 10.0625,
      "learning_rate": 6.105522199920954e-06,
      "loss": 0.5437,
      "num_input_tokens_seen": 3563834112,
      "step": 48050,
      "train_runtime": 57177.6895,
      "train_tokens_per_second": 62329.103
    },
    {
      "epoch": 0.6386668527119421,
      "grad_norm": 13.8125,
      "learning_rate": 6.101558400695438e-06,
      "loss": 0.8022,
      "num_input_tokens_seen": 3565929984,
      "step": 48060,
      "train_runtime": 57211.3636,
      "train_tokens_per_second": 62329.051
    },
    {
      "epoch": 0.638799742194404,
      "grad_norm": 18.25,
      "learning_rate": 6.097595323621465e-06,
      "loss": 0.917,
      "num_input_tokens_seen": 3566807296,
      "step": 48070,
      "train_runtime": 57224.2633,
      "train_tokens_per_second": 62330.331
    },
    {
      "epoch": 0.638932631676866,
      "grad_norm": 28.25,
      "learning_rate": 6.093632969433162e-06,
      "loss": 0.7466,
      "num_input_tokens_seen": 3567277056,
      "step": 48080,
      "train_runtime": 57230.9234,
      "train_tokens_per_second": 62331.286
    },
    {
      "epoch": 0.6390655211593278,
      "grad_norm": 15.375,
      "learning_rate": 6.089671338864516e-06,
      "loss": 0.6791,
      "num_input_tokens_seen": 3567502336,
      "step": 48090,
      "train_runtime": 57234.3024,
      "train_tokens_per_second": 62331.542
    },
    {
      "epoch": 0.6391984106417897,
      "grad_norm": 11.625,
      "learning_rate": 6.08571043264939e-06,
      "loss": 0.478,
      "num_input_tokens_seen": 3567620096,
      "step": 48100,
      "train_runtime": 57236.7086,
      "train_tokens_per_second": 62330.979
    },
    {
      "epoch": 0.6393313001242517,
      "grad_norm": 24.375,
      "learning_rate": 6.0817502515214996e-06,
      "loss": 0.7884,
      "num_input_tokens_seen": 3569473536,
      "step": 48110,
      "train_runtime": 57268.1712,
      "train_tokens_per_second": 62329.099
    },
    {
      "epoch": 0.6394641896067136,
      "grad_norm": 20.625,
      "learning_rate": 6.077790796214441e-06,
      "loss": 0.8907,
      "num_input_tokens_seen": 3570288640,
      "step": 48120,
      "train_runtime": 57281.1713,
      "train_tokens_per_second": 62329.184
    },
    {
      "epoch": 0.6395970790891755,
      "grad_norm": 63.25,
      "learning_rate": 6.073832067461662e-06,
      "loss": 0.8275,
      "num_input_tokens_seen": 3570752768,
      "step": 48130,
      "train_runtime": 57287.7523,
      "train_tokens_per_second": 62330.125
    },
    {
      "epoch": 0.6397299685716374,
      "grad_norm": 13.6875,
      "learning_rate": 6.069874065996485e-06,
      "loss": 0.6104,
      "num_input_tokens_seen": 3570971392,
      "step": 48140,
      "train_runtime": 57291.168,
      "train_tokens_per_second": 62330.225
    },
    {
      "epoch": 0.6398628580540993,
      "grad_norm": 24.5,
      "learning_rate": 6.065916792552093e-06,
      "loss": 0.5738,
      "num_input_tokens_seen": 3571087104,
      "step": 48150,
      "train_runtime": 57293.5388,
      "train_tokens_per_second": 62329.665
    },
    {
      "epoch": 0.6399957475365612,
      "grad_norm": 29.0,
      "learning_rate": 6.061960247861532e-06,
      "loss": 0.7432,
      "num_input_tokens_seen": 3573120256,
      "step": 48160,
      "train_runtime": 57325.2076,
      "train_tokens_per_second": 62330.699
    },
    {
      "epoch": 0.6401286370190231,
      "grad_norm": 12.75,
      "learning_rate": 6.058004432657722e-06,
      "loss": 0.8566,
      "num_input_tokens_seen": 3573931264,
      "step": 48170,
      "train_runtime": 57337.4688,
      "train_tokens_per_second": 62331.514
    },
    {
      "epoch": 0.640261526501485,
      "grad_norm": 24.5,
      "learning_rate": 6.0540493476734415e-06,
      "loss": 0.7972,
      "num_input_tokens_seen": 3574387456,
      "step": 48180,
      "train_runtime": 57344.6357,
      "train_tokens_per_second": 62331.68
    },
    {
      "epoch": 0.640394415983947,
      "grad_norm": 59.0,
      "learning_rate": 6.050094993641336e-06,
      "loss": 0.7012,
      "num_input_tokens_seen": 3574615808,
      "step": 48190,
      "train_runtime": 57348.0504,
      "train_tokens_per_second": 62331.95
    },
    {
      "epoch": 0.6405273054664089,
      "grad_norm": 11.0,
      "learning_rate": 6.046141371293907e-06,
      "loss": 0.5305,
      "num_input_tokens_seen": 3574728192,
      "step": 48200,
      "train_runtime": 57350.4748,
      "train_tokens_per_second": 62331.275
    },
    {
      "epoch": 0.6406601949488707,
      "grad_norm": 46.75,
      "learning_rate": 6.042188481363538e-06,
      "loss": 0.951,
      "num_input_tokens_seen": 3576716800,
      "step": 48210,
      "train_runtime": 57381.9609,
      "train_tokens_per_second": 62331.728
    },
    {
      "epoch": 0.6407930844313326,
      "grad_norm": 23.625,
      "learning_rate": 6.038236324582461e-06,
      "loss": 0.8953,
      "num_input_tokens_seen": 3577564928,
      "step": 48220,
      "train_runtime": 57395.2657,
      "train_tokens_per_second": 62332.056
    },
    {
      "epoch": 0.6409259739137946,
      "grad_norm": 19.125,
      "learning_rate": 6.034284901682776e-06,
      "loss": 0.6966,
      "num_input_tokens_seen": 3578025728,
      "step": 48230,
      "train_runtime": 57401.7689,
      "train_tokens_per_second": 62333.022
    },
    {
      "epoch": 0.6410588633962565,
      "grad_norm": 69.5,
      "learning_rate": 6.030334213396454e-06,
      "loss": 0.6102,
      "num_input_tokens_seen": 3578253824,
      "step": 48240,
      "train_runtime": 57405.1994,
      "train_tokens_per_second": 62333.27
    },
    {
      "epoch": 0.6411917528787184,
      "grad_norm": 61.5,
      "learning_rate": 6.026384260455318e-06,
      "loss": 0.5624,
      "num_input_tokens_seen": 3578367232,
      "step": 48250,
      "train_runtime": 57407.6355,
      "train_tokens_per_second": 62332.601
    },
    {
      "epoch": 0.6413246423611804,
      "grad_norm": 29.625,
      "learning_rate": 6.022435043591068e-06,
      "loss": 0.8929,
      "num_input_tokens_seen": 3580301312,
      "step": 48260,
      "train_runtime": 57439.431,
      "train_tokens_per_second": 62331.768
    },
    {
      "epoch": 0.6414575318436422,
      "grad_norm": 18.5,
      "learning_rate": 6.018486563535258e-06,
      "loss": 0.9298,
      "num_input_tokens_seen": 3581164800,
      "step": 48270,
      "train_runtime": 57451.9269,
      "train_tokens_per_second": 62333.241
    },
    {
      "epoch": 0.6415904213261041,
      "grad_norm": 26.0,
      "learning_rate": 6.014538821019311e-06,
      "loss": 0.7428,
      "num_input_tokens_seen": 3581607936,
      "step": 48280,
      "train_runtime": 57458.3867,
      "train_tokens_per_second": 62333.945
    },
    {
      "epoch": 0.641723310808566,
      "grad_norm": 103.5,
      "learning_rate": 6.01059181677451e-06,
      "loss": 0.6403,
      "num_input_tokens_seen": 3581835264,
      "step": 48290,
      "train_runtime": 57461.7956,
      "train_tokens_per_second": 62334.204
    },
    {
      "epoch": 0.641856200291028,
      "grad_norm": 11.9375,
      "learning_rate": 6.006645551532002e-06,
      "loss": 0.4852,
      "num_input_tokens_seen": 3581952256,
      "step": 48300,
      "train_runtime": 57464.9852,
      "train_tokens_per_second": 62332.78
    },
    {
      "epoch": 0.6419890897734899,
      "grad_norm": 34.0,
      "learning_rate": 6.002700026022799e-06,
      "loss": 0.8241,
      "num_input_tokens_seen": 3584129792,
      "step": 48310,
      "train_runtime": 57499.222,
      "train_tokens_per_second": 62333.535
    },
    {
      "epoch": 0.6421219792559518,
      "grad_norm": 244.0,
      "learning_rate": 5.9987552409777686e-06,
      "loss": 0.949,
      "num_input_tokens_seen": 3584970496,
      "step": 48320,
      "train_runtime": 57512.2831,
      "train_tokens_per_second": 62333.997
    },
    {
      "epoch": 0.6422548687384138,
      "grad_norm": 25.375,
      "learning_rate": 5.994811197127654e-06,
      "loss": 0.7862,
      "num_input_tokens_seen": 3585438464,
      "step": 48330,
      "train_runtime": 57519.2191,
      "train_tokens_per_second": 62334.617
    },
    {
      "epoch": 0.6423877582208756,
      "grad_norm": 10.5,
      "learning_rate": 5.990867895203052e-06,
      "loss": 0.6282,
      "num_input_tokens_seen": 3585674752,
      "step": 48340,
      "train_runtime": 57523.5301,
      "train_tokens_per_second": 62334.053
    },
    {
      "epoch": 0.6425206477033375,
      "grad_norm": 11.4375,
      "learning_rate": 5.986925335934423e-06,
      "loss": 0.466,
      "num_input_tokens_seen": 3585790720,
      "step": 48350,
      "train_runtime": 57525.9703,
      "train_tokens_per_second": 62333.424
    },
    {
      "epoch": 0.6426535371857994,
      "grad_norm": 20.25,
      "learning_rate": 5.982983520052089e-06,
      "loss": 0.8592,
      "num_input_tokens_seen": 3587827456,
      "step": 48360,
      "train_runtime": 57558.4278,
      "train_tokens_per_second": 62333.66
    },
    {
      "epoch": 0.6427864266682614,
      "grad_norm": 31.125,
      "learning_rate": 5.9790424482862434e-06,
      "loss": 0.9445,
      "num_input_tokens_seen": 3588648704,
      "step": 48370,
      "train_runtime": 57570.9052,
      "train_tokens_per_second": 62334.415
    },
    {
      "epoch": 0.6429193161507233,
      "grad_norm": 17.625,
      "learning_rate": 5.975102121366927e-06,
      "loss": 0.7412,
      "num_input_tokens_seen": 3589105664,
      "step": 48380,
      "train_runtime": 57578.3966,
      "train_tokens_per_second": 62334.241
    },
    {
      "epoch": 0.6430522056331852,
      "grad_norm": 18.625,
      "learning_rate": 5.9711625400240555e-06,
      "loss": 0.6328,
      "num_input_tokens_seen": 3589331200,
      "step": 48390,
      "train_runtime": 57581.851,
      "train_tokens_per_second": 62334.419
    },
    {
      "epoch": 0.643185095115647,
      "grad_norm": 20.125,
      "learning_rate": 5.9672237049874e-06,
      "loss": 0.5333,
      "num_input_tokens_seen": 3589443072,
      "step": 48400,
      "train_runtime": 57584.2607,
      "train_tokens_per_second": 62333.753
    },
    {
      "epoch": 0.643317984598109,
      "grad_norm": 19.25,
      "learning_rate": 5.963285616986587e-06,
      "loss": 0.814,
      "num_input_tokens_seen": 3591463680,
      "step": 48410,
      "train_runtime": 57617.151,
      "train_tokens_per_second": 62333.24
    },
    {
      "epoch": 0.6434508740805709,
      "grad_norm": 20.375,
      "learning_rate": 5.959348276751124e-06,
      "loss": 0.9821,
      "num_input_tokens_seen": 3592311040,
      "step": 48420,
      "train_runtime": 57630.1025,
      "train_tokens_per_second": 62333.935
    },
    {
      "epoch": 0.6435837635630328,
      "grad_norm": 23.0,
      "learning_rate": 5.95541168501036e-06,
      "loss": 0.7243,
      "num_input_tokens_seen": 3592805120,
      "step": 48430,
      "train_runtime": 57637.0003,
      "train_tokens_per_second": 62335.047
    },
    {
      "epoch": 0.6437166530454947,
      "grad_norm": 16.875,
      "learning_rate": 5.9514758424935194e-06,
      "loss": 0.6421,
      "num_input_tokens_seen": 3593038848,
      "step": 48440,
      "train_runtime": 57640.4872,
      "train_tokens_per_second": 62335.331
    },
    {
      "epoch": 0.6438495425279567,
      "grad_norm": 13.4375,
      "learning_rate": 5.947540749929677e-06,
      "loss": 0.546,
      "num_input_tokens_seen": 3593152512,
      "step": 48450,
      "train_runtime": 57642.914,
      "train_tokens_per_second": 62334.679
    },
    {
      "epoch": 0.6439824320104185,
      "grad_norm": 19.25,
      "learning_rate": 5.943606408047777e-06,
      "loss": 0.7563,
      "num_input_tokens_seen": 3595321088,
      "step": 48460,
      "train_runtime": 57678.1573,
      "train_tokens_per_second": 62334.188
    },
    {
      "epoch": 0.6441153214928804,
      "grad_norm": 24.25,
      "learning_rate": 5.939672817576622e-06,
      "loss": 0.881,
      "num_input_tokens_seen": 3596152832,
      "step": 48470,
      "train_runtime": 57690.5988,
      "train_tokens_per_second": 62335.162
    },
    {
      "epoch": 0.6442482109753424,
      "grad_norm": 18.875,
      "learning_rate": 5.935739979244865e-06,
      "loss": 0.7877,
      "num_input_tokens_seen": 3596612864,
      "step": 48480,
      "train_runtime": 57697.0839,
      "train_tokens_per_second": 62336.129
    },
    {
      "epoch": 0.6443811004578043,
      "grad_norm": 41.0,
      "learning_rate": 5.931807893781043e-06,
      "loss": 0.675,
      "num_input_tokens_seen": 3596840192,
      "step": 48490,
      "train_runtime": 57700.5309,
      "train_tokens_per_second": 62336.345
    },
    {
      "epoch": 0.6445139899402662,
      "grad_norm": 14.0,
      "learning_rate": 5.9278765619135306e-06,
      "loss": 0.5574,
      "num_input_tokens_seen": 3596956416,
      "step": 48500,
      "train_runtime": 57702.9786,
      "train_tokens_per_second": 62335.715
    },
    {
      "epoch": 0.6446468794227281,
      "grad_norm": 110.5,
      "learning_rate": 5.923945984370577e-06,
      "loss": 0.8611,
      "num_input_tokens_seen": 3599038720,
      "step": 48510,
      "train_runtime": 57736.7233,
      "train_tokens_per_second": 62335.348
    },
    {
      "epoch": 0.64477976890519,
      "grad_norm": 25.875,
      "learning_rate": 5.9200161618802796e-06,
      "loss": 0.9406,
      "num_input_tokens_seen": 3599871744,
      "step": 48520,
      "train_runtime": 57749.9401,
      "train_tokens_per_second": 62335.506
    },
    {
      "epoch": 0.6449126583876519,
      "grad_norm": 24.25,
      "learning_rate": 5.916087095170613e-06,
      "loss": 0.7398,
      "num_input_tokens_seen": 3600344064,
      "step": 48530,
      "train_runtime": 57756.6303,
      "train_tokens_per_second": 62336.463
    },
    {
      "epoch": 0.6450455478701138,
      "grad_norm": 17.5,
      "learning_rate": 5.912158784969393e-06,
      "loss": 0.5873,
      "num_input_tokens_seen": 3600562176,
      "step": 48540,
      "train_runtime": 57760.0195,
      "train_tokens_per_second": 62336.582
    },
    {
      "epoch": 0.6451784373525757,
      "grad_norm": 12.125,
      "learning_rate": 5.908231232004309e-06,
      "loss": 0.5919,
      "num_input_tokens_seen": 3600677888,
      "step": 48550,
      "train_runtime": 57762.4806,
      "train_tokens_per_second": 62335.929
    },
    {
      "epoch": 0.6453113268350377,
      "grad_norm": 20.875,
      "learning_rate": 5.904304437002906e-06,
      "loss": 0.876,
      "num_input_tokens_seen": 3602761728,
      "step": 48560,
      "train_runtime": 57797.139,
      "train_tokens_per_second": 62334.603
    },
    {
      "epoch": 0.6454442163174996,
      "grad_norm": 16.375,
      "learning_rate": 5.900378400692579e-06,
      "loss": 0.9074,
      "num_input_tokens_seen": 3603644928,
      "step": 48570,
      "train_runtime": 57809.9348,
      "train_tokens_per_second": 62336.084
    },
    {
      "epoch": 0.6455771057999614,
      "grad_norm": 12.375,
      "learning_rate": 5.8964531238006025e-06,
      "loss": 0.7163,
      "num_input_tokens_seen": 3604121088,
      "step": 48580,
      "train_runtime": 57816.475,
      "train_tokens_per_second": 62337.268
    },
    {
      "epoch": 0.6457099952824233,
      "grad_norm": 14.125,
      "learning_rate": 5.892528607054092e-06,
      "loss": 0.6433,
      "num_input_tokens_seen": 3604349696,
      "step": 48590,
      "train_runtime": 57819.9343,
      "train_tokens_per_second": 62337.492
    },
    {
      "epoch": 0.6458428847648853,
      "grad_norm": 21.125,
      "learning_rate": 5.888604851180033e-06,
      "loss": 0.5263,
      "num_input_tokens_seen": 3604464128,
      "step": 48600,
      "train_runtime": 57822.3703,
      "train_tokens_per_second": 62336.845
    },
    {
      "epoch": 0.6459757742473472,
      "grad_norm": 21.125,
      "learning_rate": 5.884681856905261e-06,
      "loss": 0.8744,
      "num_input_tokens_seen": 3606496768,
      "step": 48610,
      "train_runtime": 57855.1385,
      "train_tokens_per_second": 62336.672
    },
    {
      "epoch": 0.6461086637298091,
      "grad_norm": 13.6875,
      "learning_rate": 5.880759624956485e-06,
      "loss": 0.9317,
      "num_input_tokens_seen": 3607321344,
      "step": 48620,
      "train_runtime": 57867.7021,
      "train_tokens_per_second": 62337.387
    },
    {
      "epoch": 0.6462415532122711,
      "grad_norm": 13.0,
      "learning_rate": 5.876838156060256e-06,
      "loss": 0.762,
      "num_input_tokens_seen": 3607780352,
      "step": 48630,
      "train_runtime": 57874.215,
      "train_tokens_per_second": 62338.303
    },
    {
      "epoch": 0.6463744426947329,
      "grad_norm": 40.25,
      "learning_rate": 5.87291745094299e-06,
      "loss": 0.6213,
      "num_input_tokens_seen": 3608001792,
      "step": 48640,
      "train_runtime": 57878.3606,
      "train_tokens_per_second": 62337.664
    },
    {
      "epoch": 0.6465073321771948,
      "grad_norm": 21.125,
      "learning_rate": 5.868997510330969e-06,
      "loss": 0.5675,
      "num_input_tokens_seen": 3608114176,
      "step": 48650,
      "train_runtime": 57880.8074,
      "train_tokens_per_second": 62336.97
    },
    {
      "epoch": 0.6466402216596567,
      "grad_norm": 201.0,
      "learning_rate": 5.865078334950323e-06,
      "loss": 0.9073,
      "num_input_tokens_seen": 3610041344,
      "step": 48660,
      "train_runtime": 57913.5833,
      "train_tokens_per_second": 62334.968
    },
    {
      "epoch": 0.6467731111421187,
      "grad_norm": 23.625,
      "learning_rate": 5.861159925527044e-06,
      "loss": 0.8475,
      "num_input_tokens_seen": 3610864384,
      "step": 48670,
      "train_runtime": 57925.8345,
      "train_tokens_per_second": 62335.993
    },
    {
      "epoch": 0.6469060006245806,
      "grad_norm": 17.75,
      "learning_rate": 5.857242282786982e-06,
      "loss": 0.7027,
      "num_input_tokens_seen": 3611308032,
      "step": 48680,
      "train_runtime": 57932.2051,
      "train_tokens_per_second": 62336.796
    },
    {
      "epoch": 0.6470388901070425,
      "grad_norm": 17.625,
      "learning_rate": 5.85332540745585e-06,
      "loss": 0.5908,
      "num_input_tokens_seen": 3611531008,
      "step": 48690,
      "train_runtime": 57936.4639,
      "train_tokens_per_second": 62336.062
    },
    {
      "epoch": 0.6471717795895043,
      "grad_norm": 9.1875,
      "learning_rate": 5.84940930025921e-06,
      "loss": 0.5228,
      "num_input_tokens_seen": 3611648768,
      "step": 48700,
      "train_runtime": 57938.9261,
      "train_tokens_per_second": 62335.445
    },
    {
      "epoch": 0.6473046690719663,
      "grad_norm": 25.5,
      "learning_rate": 5.845493961922488e-06,
      "loss": 0.815,
      "num_input_tokens_seen": 3613782272,
      "step": 48710,
      "train_runtime": 57971.1776,
      "train_tokens_per_second": 62337.569
    },
    {
      "epoch": 0.6474375585544282,
      "grad_norm": 17.625,
      "learning_rate": 5.841579393170965e-06,
      "loss": 0.8993,
      "num_input_tokens_seen": 3614651904,
      "step": 48720,
      "train_runtime": 57983.7985,
      "train_tokens_per_second": 62338.998
    },
    {
      "epoch": 0.6475704480368901,
      "grad_norm": 17.25,
      "learning_rate": 5.837665594729778e-06,
      "loss": 0.6778,
      "num_input_tokens_seen": 3615138560,
      "step": 48730,
      "train_runtime": 57990.6149,
      "train_tokens_per_second": 62340.063
    },
    {
      "epoch": 0.647703337519352,
      "grad_norm": 71.0,
      "learning_rate": 5.8337525673239235e-06,
      "loss": 0.6285,
      "num_input_tokens_seen": 3615374080,
      "step": 48740,
      "train_runtime": 57994.1574,
      "train_tokens_per_second": 62340.316
    },
    {
      "epoch": 0.647836227001814,
      "grad_norm": 21.75,
      "learning_rate": 5.829840311678256e-06,
      "loss": 0.4912,
      "num_input_tokens_seen": 3615489280,
      "step": 48750,
      "train_runtime": 57996.5674,
      "train_tokens_per_second": 62339.712
    },
    {
      "epoch": 0.6479691164842758,
      "grad_norm": 38.25,
      "learning_rate": 5.8259288285174875e-06,
      "loss": 0.8086,
      "num_input_tokens_seen": 3617583360,
      "step": 48760,
      "train_runtime": 58030.026,
      "train_tokens_per_second": 62339.854
    },
    {
      "epoch": 0.6481020059667377,
      "grad_norm": 10.625,
      "learning_rate": 5.822018118566184e-06,
      "loss": 0.8115,
      "num_input_tokens_seen": 3618418432,
      "step": 48770,
      "train_runtime": 58042.5984,
      "train_tokens_per_second": 62340.738
    },
    {
      "epoch": 0.6482348954491997,
      "grad_norm": 24.375,
      "learning_rate": 5.818108182548769e-06,
      "loss": 0.6959,
      "num_input_tokens_seen": 3618889728,
      "step": 48780,
      "train_runtime": 58049.2394,
      "train_tokens_per_second": 62341.725
    },
    {
      "epoch": 0.6483677849316616,
      "grad_norm": 16.75,
      "learning_rate": 5.814199021189529e-06,
      "loss": 0.7151,
      "num_input_tokens_seen": 3619114240,
      "step": 48790,
      "train_runtime": 58052.6308,
      "train_tokens_per_second": 62341.951
    },
    {
      "epoch": 0.6485006744141235,
      "grad_norm": 22.75,
      "learning_rate": 5.81029063521259e-06,
      "loss": 0.5233,
      "num_input_tokens_seen": 3619226624,
      "step": 48800,
      "train_runtime": 58055.0062,
      "train_tokens_per_second": 62341.336
    },
    {
      "epoch": 0.6486335638965854,
      "grad_norm": 47.0,
      "learning_rate": 5.806383025341953e-06,
      "loss": 0.8573,
      "num_input_tokens_seen": 3621192448,
      "step": 48810,
      "train_runtime": 58087.8098,
      "train_tokens_per_second": 62339.972
    },
    {
      "epoch": 0.6487664533790474,
      "grad_norm": 28.5,
      "learning_rate": 5.8024761923014675e-06,
      "loss": 0.9467,
      "num_input_tokens_seen": 3622031104,
      "step": 48820,
      "train_runtime": 58100.1161,
      "train_tokens_per_second": 62341.202
    },
    {
      "epoch": 0.6488993428615092,
      "grad_norm": 22.25,
      "learning_rate": 5.798570136814837e-06,
      "loss": 0.7121,
      "num_input_tokens_seen": 3622495744,
      "step": 48830,
      "train_runtime": 58106.5404,
      "train_tokens_per_second": 62342.306
    },
    {
      "epoch": 0.6490322323439711,
      "grad_norm": 11.8125,
      "learning_rate": 5.794664859605626e-06,
      "loss": 0.601,
      "num_input_tokens_seen": 3622719744,
      "step": 48840,
      "train_runtime": 58109.9985,
      "train_tokens_per_second": 62342.451
    },
    {
      "epoch": 0.649165121826433,
      "grad_norm": 15.6875,
      "learning_rate": 5.790760361397256e-06,
      "loss": 0.5465,
      "num_input_tokens_seen": 3622831616,
      "step": 48850,
      "train_runtime": 58113.2475,
      "train_tokens_per_second": 62340.891
    },
    {
      "epoch": 0.649298011308895,
      "grad_norm": 18.75,
      "learning_rate": 5.786856642912988e-06,
      "loss": 0.7833,
      "num_input_tokens_seen": 3624827136,
      "step": 48860,
      "train_runtime": 58145.4709,
      "train_tokens_per_second": 62340.662
    },
    {
      "epoch": 0.6494309007913569,
      "grad_norm": 46.0,
      "learning_rate": 5.782953704875967e-06,
      "loss": 0.9088,
      "num_input_tokens_seen": 3625653760,
      "step": 48870,
      "train_runtime": 58158.3216,
      "train_tokens_per_second": 62341.1
    },
    {
      "epoch": 0.6495637902738188,
      "grad_norm": 32.75,
      "learning_rate": 5.779051548009164e-06,
      "loss": 0.7565,
      "num_input_tokens_seen": 3626149120,
      "step": 48880,
      "train_runtime": 58165.3647,
      "train_tokens_per_second": 62342.068
    },
    {
      "epoch": 0.6496966797562806,
      "grad_norm": 37.25,
      "learning_rate": 5.775150173035426e-06,
      "loss": 0.713,
      "num_input_tokens_seen": 3626376960,
      "step": 48890,
      "train_runtime": 58168.8016,
      "train_tokens_per_second": 62342.301
    },
    {
      "epoch": 0.6498295692387426,
      "grad_norm": 47.5,
      "learning_rate": 5.771249580677449e-06,
      "loss": 0.5932,
      "num_input_tokens_seen": 3626492928,
      "step": 48900,
      "train_runtime": 58171.2427,
      "train_tokens_per_second": 62341.679
    },
    {
      "epoch": 0.6499624587212045,
      "grad_norm": 21.875,
      "learning_rate": 5.767349771657773e-06,
      "loss": 0.8266,
      "num_input_tokens_seen": 3628478208,
      "step": 48910,
      "train_runtime": 58203.9101,
      "train_tokens_per_second": 62340.798
    },
    {
      "epoch": 0.6500953482036664,
      "grad_norm": 17.75,
      "learning_rate": 5.763450746698818e-06,
      "loss": 0.9003,
      "num_input_tokens_seen": 3629322752,
      "step": 48920,
      "train_runtime": 58216.6417,
      "train_tokens_per_second": 62341.672
    },
    {
      "epoch": 0.6502282376861284,
      "grad_norm": 22.625,
      "learning_rate": 5.759552506522832e-06,
      "loss": 0.7001,
      "num_input_tokens_seen": 3629788160,
      "step": 48930,
      "train_runtime": 58224.1062,
      "train_tokens_per_second": 62341.673
    },
    {
      "epoch": 0.6503611271685903,
      "grad_norm": 21.125,
      "learning_rate": 5.755655051851932e-06,
      "loss": 0.6065,
      "num_input_tokens_seen": 3630013440,
      "step": 48940,
      "train_runtime": 58227.4942,
      "train_tokens_per_second": 62341.914
    },
    {
      "epoch": 0.6504940166510521,
      "grad_norm": 29.875,
      "learning_rate": 5.75175838340809e-06,
      "loss": 0.5365,
      "num_input_tokens_seen": 3630126080,
      "step": 48950,
      "train_runtime": 58229.8869,
      "train_tokens_per_second": 62341.287
    },
    {
      "epoch": 0.650626906133514,
      "grad_norm": 26.25,
      "learning_rate": 5.747862501913121e-06,
      "loss": 0.7927,
      "num_input_tokens_seen": 3632059904,
      "step": 48960,
      "train_runtime": 58261.9116,
      "train_tokens_per_second": 62340.212
    },
    {
      "epoch": 0.650759795615976,
      "grad_norm": 14.5,
      "learning_rate": 5.743967408088716e-06,
      "loss": 0.8895,
      "num_input_tokens_seen": 3632874752,
      "step": 48970,
      "train_runtime": 58274.2483,
      "train_tokens_per_second": 62340.997
    },
    {
      "epoch": 0.6508926850984379,
      "grad_norm": 24.25,
      "learning_rate": 5.740073102656387e-06,
      "loss": 0.7233,
      "num_input_tokens_seen": 3633327616,
      "step": 48980,
      "train_runtime": 58280.8009,
      "train_tokens_per_second": 62341.759
    },
    {
      "epoch": 0.6510255745808998,
      "grad_norm": 37.5,
      "learning_rate": 5.736179586337542e-06,
      "loss": 0.6248,
      "num_input_tokens_seen": 3633555456,
      "step": 48990,
      "train_runtime": 58284.2027,
      "train_tokens_per_second": 62342.029
    },
    {
      "epoch": 0.6511584640633618,
      "grad_norm": 10.0625,
      "learning_rate": 5.732286859853404e-06,
      "loss": 0.5542,
      "num_input_tokens_seen": 3633666560,
      "step": 49000,
      "train_runtime": 58287.3525,
      "train_tokens_per_second": 62340.566
    },
    {
      "epoch": 0.6512913535458236,
      "grad_norm": 21.875,
      "learning_rate": 5.7283949239250705e-06,
      "loss": 0.8964,
      "num_input_tokens_seen": 3635770880,
      "step": 49010,
      "train_runtime": 58322.3989,
      "train_tokens_per_second": 62339.186
    },
    {
      "epoch": 0.6514242430282855,
      "grad_norm": 26.75,
      "learning_rate": 5.724503779273488e-06,
      "loss": 0.8795,
      "num_input_tokens_seen": 3636653312,
      "step": 49020,
      "train_runtime": 58335.4277,
      "train_tokens_per_second": 62340.39
    },
    {
      "epoch": 0.6515571325107474,
      "grad_norm": 22.5,
      "learning_rate": 5.7206134266194565e-06,
      "loss": 0.7412,
      "num_input_tokens_seen": 3637125376,
      "step": 49030,
      "train_runtime": 58342.3275,
      "train_tokens_per_second": 62341.109
    },
    {
      "epoch": 0.6516900219932094,
      "grad_norm": 14.25,
      "learning_rate": 5.716723866683633e-06,
      "loss": 0.6253,
      "num_input_tokens_seen": 3637363200,
      "step": 49040,
      "train_runtime": 58345.9251,
      "train_tokens_per_second": 62341.341
    },
    {
      "epoch": 0.6518229114756713,
      "grad_norm": 12.6875,
      "learning_rate": 5.712835100186517e-06,
      "loss": 0.5399,
      "num_input_tokens_seen": 3637481472,
      "step": 49050,
      "train_runtime": 58349.1852,
      "train_tokens_per_second": 62339.885
    },
    {
      "epoch": 0.6519558009581332,
      "grad_norm": 18.5,
      "learning_rate": 5.7089471278484696e-06,
      "loss": 0.8364,
      "num_input_tokens_seen": 3639585792,
      "step": 49060,
      "train_runtime": 58383.1739,
      "train_tokens_per_second": 62339.636
    },
    {
      "epoch": 0.652088690440595,
      "grad_norm": 16.5,
      "learning_rate": 5.705059950389704e-06,
      "loss": 0.8659,
      "num_input_tokens_seen": 3640464384,
      "step": 49070,
      "train_runtime": 58396.2746,
      "train_tokens_per_second": 62340.696
    },
    {
      "epoch": 0.652221579923057,
      "grad_norm": 21.5,
      "learning_rate": 5.701173568530286e-06,
      "loss": 0.7057,
      "num_input_tokens_seen": 3640960000,
      "step": 49080,
      "train_runtime": 58403.3659,
      "train_tokens_per_second": 62341.612
    },
    {
      "epoch": 0.6523544694055189,
      "grad_norm": 16.375,
      "learning_rate": 5.697287982990133e-06,
      "loss": 0.628,
      "num_input_tokens_seen": 3641201408,
      "step": 49090,
      "train_runtime": 58407.6678,
      "train_tokens_per_second": 62341.154
    },
    {
      "epoch": 0.6524873588879808,
      "grad_norm": 15.0625,
      "learning_rate": 5.6934031944890175e-06,
      "loss": 0.5196,
      "num_input_tokens_seen": 3641319424,
      "step": 49100,
      "train_runtime": 58410.0809,
      "train_tokens_per_second": 62340.599
    },
    {
      "epoch": 0.6526202483704427,
      "grad_norm": 28.375,
      "learning_rate": 5.689519203746555e-06,
      "loss": 0.8548,
      "num_input_tokens_seen": 3643248128,
      "step": 49110,
      "train_runtime": 58440.5561,
      "train_tokens_per_second": 62341.093
    },
    {
      "epoch": 0.6527531378529047,
      "grad_norm": 14.6875,
      "learning_rate": 5.685636011482225e-06,
      "loss": 0.9082,
      "num_input_tokens_seen": 3644075264,
      "step": 49120,
      "train_runtime": 58452.8232,
      "train_tokens_per_second": 62342.16
    },
    {
      "epoch": 0.6528860273353665,
      "grad_norm": 28.875,
      "learning_rate": 5.681753618415353e-06,
      "loss": 0.6483,
      "num_input_tokens_seen": 3644518144,
      "step": 49130,
      "train_runtime": 58459.1508,
      "train_tokens_per_second": 62342.988
    },
    {
      "epoch": 0.6530189168178284,
      "grad_norm": 13.1875,
      "learning_rate": 5.677872025265119e-06,
      "loss": 0.6213,
      "num_input_tokens_seen": 3644740352,
      "step": 49140,
      "train_runtime": 58462.5476,
      "train_tokens_per_second": 62343.167
    },
    {
      "epoch": 0.6531518063002903,
      "grad_norm": 11.25,
      "learning_rate": 5.6739912327505506e-06,
      "loss": 0.4806,
      "num_input_tokens_seen": 3644854272,
      "step": 49150,
      "train_runtime": 58464.9963,
      "train_tokens_per_second": 62342.504
    },
    {
      "epoch": 0.6532846957827523,
      "grad_norm": 22.0,
      "learning_rate": 5.670111241590533e-06,
      "loss": 0.8601,
      "num_input_tokens_seen": 3646958336,
      "step": 49160,
      "train_runtime": 58499.2685,
      "train_tokens_per_second": 62341.948
    },
    {
      "epoch": 0.6534175852652142,
      "grad_norm": 17.0,
      "learning_rate": 5.666232052503804e-06,
      "loss": 0.9608,
      "num_input_tokens_seen": 3647779584,
      "step": 49170,
      "train_runtime": 58512.1558,
      "train_tokens_per_second": 62342.252
    },
    {
      "epoch": 0.6535504747476761,
      "grad_norm": 31.75,
      "learning_rate": 5.662353666208934e-06,
      "loss": 0.7076,
      "num_input_tokens_seen": 3648262656,
      "step": 49180,
      "train_runtime": 58519.081,
      "train_tokens_per_second": 62343.13
    },
    {
      "epoch": 0.653683364230138,
      "grad_norm": 28.875,
      "learning_rate": 5.658476083424378e-06,
      "loss": 0.5667,
      "num_input_tokens_seen": 3648494592,
      "step": 49190,
      "train_runtime": 58522.5314,
      "train_tokens_per_second": 62343.417
    },
    {
      "epoch": 0.6538162537125999,
      "grad_norm": 8.375,
      "learning_rate": 5.654599304868411e-06,
      "loss": 0.5276,
      "num_input_tokens_seen": 3648612352,
      "step": 49200,
      "train_runtime": 58524.9079,
      "train_tokens_per_second": 62342.898
    },
    {
      "epoch": 0.6539491431950618,
      "grad_norm": 18.875,
      "learning_rate": 5.650723331259176e-06,
      "loss": 0.753,
      "num_input_tokens_seen": 3650670080,
      "step": 49210,
      "train_runtime": 58558.7672,
      "train_tokens_per_second": 62341.99
    },
    {
      "epoch": 0.6540820326775237,
      "grad_norm": 20.75,
      "learning_rate": 5.646848163314665e-06,
      "loss": 0.8596,
      "num_input_tokens_seen": 3651492352,
      "step": 49220,
      "train_runtime": 58571.3199,
      "train_tokens_per_second": 62342.668
    },
    {
      "epoch": 0.6542149221599857,
      "grad_norm": 18.0,
      "learning_rate": 5.642973801752709e-06,
      "loss": 0.6897,
      "num_input_tokens_seen": 3651950336,
      "step": 49230,
      "train_runtime": 58577.8597,
      "train_tokens_per_second": 62343.526
    },
    {
      "epoch": 0.6543478116424476,
      "grad_norm": 28.25,
      "learning_rate": 5.639100247291011e-06,
      "loss": 0.6543,
      "num_input_tokens_seen": 3652168960,
      "step": 49240,
      "train_runtime": 58581.282,
      "train_tokens_per_second": 62343.616
    },
    {
      "epoch": 0.6544807011249094,
      "grad_norm": 20.25,
      "learning_rate": 5.6352275006471045e-06,
      "loss": 0.5069,
      "num_input_tokens_seen": 3652283904,
      "step": 49250,
      "train_runtime": 58583.6979,
      "train_tokens_per_second": 62343.007
    },
    {
      "epoch": 0.6546135906073713,
      "grad_norm": 18.375,
      "learning_rate": 5.631355562538384e-06,
      "loss": 0.8121,
      "num_input_tokens_seen": 3654338304,
      "step": 49260,
      "train_runtime": 58616.8027,
      "train_tokens_per_second": 62342.846
    },
    {
      "epoch": 0.6547464800898333,
      "grad_norm": 25.875,
      "learning_rate": 5.627484433682094e-06,
      "loss": 0.8896,
      "num_input_tokens_seen": 3655177728,
      "step": 49270,
      "train_runtime": 58629.4592,
      "train_tokens_per_second": 62343.705
    },
    {
      "epoch": 0.6548793695722952,
      "grad_norm": 41.5,
      "learning_rate": 5.623614114795317e-06,
      "loss": 0.7705,
      "num_input_tokens_seen": 3655657472,
      "step": 49280,
      "train_runtime": 58636.0773,
      "train_tokens_per_second": 62344.851
    },
    {
      "epoch": 0.6550122590547571,
      "grad_norm": 19.0,
      "learning_rate": 5.619744606595008e-06,
      "loss": 0.6747,
      "num_input_tokens_seen": 3655882240,
      "step": 49290,
      "train_runtime": 58639.4987,
      "train_tokens_per_second": 62345.046
    },
    {
      "epoch": 0.6551451485372191,
      "grad_norm": 18.625,
      "learning_rate": 5.615875909797945e-06,
      "loss": 0.5418,
      "num_input_tokens_seen": 3655998208,
      "step": 49300,
      "train_runtime": 58641.9608,
      "train_tokens_per_second": 62344.406
    },
    {
      "epoch": 0.655278038019681,
      "grad_norm": 34.75,
      "learning_rate": 5.612008025120786e-06,
      "loss": 0.8534,
      "num_input_tokens_seen": 3658039040,
      "step": 49310,
      "train_runtime": 58674.0009,
      "train_tokens_per_second": 62345.144
    },
    {
      "epoch": 0.6554109275021428,
      "grad_norm": 15.625,
      "learning_rate": 5.608140953280008e-06,
      "loss": 0.857,
      "num_input_tokens_seen": 3658907392,
      "step": 49320,
      "train_runtime": 58687.0918,
      "train_tokens_per_second": 62346.034
    },
    {
      "epoch": 0.6555438169846047,
      "grad_norm": 15.75,
      "learning_rate": 5.604274694991958e-06,
      "loss": 0.6952,
      "num_input_tokens_seen": 3659385600,
      "step": 49330,
      "train_runtime": 58693.6318,
      "train_tokens_per_second": 62347.234
    },
    {
      "epoch": 0.6556767064670667,
      "grad_norm": 16.375,
      "learning_rate": 5.600409250972825e-06,
      "loss": 0.6416,
      "num_input_tokens_seen": 3659611904,
      "step": 49340,
      "train_runtime": 58697.0264,
      "train_tokens_per_second": 62347.484
    },
    {
      "epoch": 0.6558095959495286,
      "grad_norm": 13.625,
      "learning_rate": 5.596544621938647e-06,
      "loss": 0.5873,
      "num_input_tokens_seen": 3659726592,
      "step": 49350,
      "train_runtime": 58699.4245,
      "train_tokens_per_second": 62346.891
    },
    {
      "epoch": 0.6559424854319905,
      "grad_norm": 47.5,
      "learning_rate": 5.592680808605317e-06,
      "loss": 0.8751,
      "num_input_tokens_seen": 3661704192,
      "step": 49360,
      "train_runtime": 58730.7078,
      "train_tokens_per_second": 62347.353
    },
    {
      "epoch": 0.6560753749144524,
      "grad_norm": 33.5,
      "learning_rate": 5.588817811688565e-06,
      "loss": 0.8402,
      "num_input_tokens_seen": 3662535424,
      "step": 49370,
      "train_runtime": 58743.1401,
      "train_tokens_per_second": 62348.309
    },
    {
      "epoch": 0.6562082643969143,
      "grad_norm": 16.5,
      "learning_rate": 5.5849556319039785e-06,
      "loss": 0.7992,
      "num_input_tokens_seen": 3663026432,
      "step": 49380,
      "train_runtime": 58749.8613,
      "train_tokens_per_second": 62349.533
    },
    {
      "epoch": 0.6563411538793762,
      "grad_norm": 48.25,
      "learning_rate": 5.581094269966993e-06,
      "loss": 0.5994,
      "num_input_tokens_seen": 3663266816,
      "step": 49390,
      "train_runtime": 58753.4318,
      "train_tokens_per_second": 62349.836
    },
    {
      "epoch": 0.6564740433618381,
      "grad_norm": 8.125,
      "learning_rate": 5.577233726592892e-06,
      "loss": 0.5056,
      "num_input_tokens_seen": 3663381248,
      "step": 49400,
      "train_runtime": 58755.8355,
      "train_tokens_per_second": 62349.232
    },
    {
      "epoch": 0.6566069328443,
      "grad_norm": 33.25,
      "learning_rate": 5.573374002496805e-06,
      "loss": 0.8514,
      "num_input_tokens_seen": 3665469696,
      "step": 49410,
      "train_runtime": 58789.7384,
      "train_tokens_per_second": 62348.801
    },
    {
      "epoch": 0.656739822326762,
      "grad_norm": 17.25,
      "learning_rate": 5.569515098393717e-06,
      "loss": 0.9169,
      "num_input_tokens_seen": 3666321920,
      "step": 49420,
      "train_runtime": 58802.4712,
      "train_tokens_per_second": 62349.793
    },
    {
      "epoch": 0.6568727118092239,
      "grad_norm": 47.0,
      "learning_rate": 5.565657014998447e-06,
      "loss": 0.6708,
      "num_input_tokens_seen": 3666808320,
      "step": 49430,
      "train_runtime": 58809.1837,
      "train_tokens_per_second": 62350.947
    },
    {
      "epoch": 0.6570056012916857,
      "grad_norm": 18.25,
      "learning_rate": 5.561799753025676e-06,
      "loss": 0.7177,
      "num_input_tokens_seen": 3667031296,
      "step": 49440,
      "train_runtime": 58813.4114,
      "train_tokens_per_second": 62350.257
    },
    {
      "epoch": 0.6571384907741477,
      "grad_norm": 10.6875,
      "learning_rate": 5.557943313189924e-06,
      "loss": 0.5534,
      "num_input_tokens_seen": 3667146496,
      "step": 49450,
      "train_runtime": 58815.7947,
      "train_tokens_per_second": 62349.689
    },
    {
      "epoch": 0.6572713802566096,
      "grad_norm": 27.5,
      "learning_rate": 5.554087696205566e-06,
      "loss": 0.8349,
      "num_input_tokens_seen": 3669124096,
      "step": 49460,
      "train_runtime": 58848.0172,
      "train_tokens_per_second": 62349.154
    },
    {
      "epoch": 0.6574042697390715,
      "grad_norm": 19.5,
      "learning_rate": 5.550232902786821e-06,
      "loss": 0.8239,
      "num_input_tokens_seen": 3669939456,
      "step": 49470,
      "train_runtime": 58860.4652,
      "train_tokens_per_second": 62349.821
    },
    {
      "epoch": 0.6575371592215334,
      "grad_norm": 98.0,
      "learning_rate": 5.546378933647747e-06,
      "loss": 0.8456,
      "num_input_tokens_seen": 3670384640,
      "step": 49480,
      "train_runtime": 58866.8378,
      "train_tokens_per_second": 62350.634
    },
    {
      "epoch": 0.6576700487039954,
      "grad_norm": 15.625,
      "learning_rate": 5.542525789502269e-06,
      "loss": 0.6359,
      "num_input_tokens_seen": 3670610688,
      "step": 49490,
      "train_runtime": 58871.1443,
      "train_tokens_per_second": 62349.912
    },
    {
      "epoch": 0.6578029381864572,
      "grad_norm": 18.875,
      "learning_rate": 5.538673471064136e-06,
      "loss": 0.5212,
      "num_input_tokens_seen": 3670728448,
      "step": 49500,
      "train_runtime": 58873.5807,
      "train_tokens_per_second": 62349.332
    },
    {
      "epoch": 0.6579358276689191,
      "grad_norm": 15.9375,
      "learning_rate": 5.534821979046968e-06,
      "loss": 0.8292,
      "num_input_tokens_seen": 3672758784,
      "step": 49510,
      "train_runtime": 58907.2936,
      "train_tokens_per_second": 62348.116
    },
    {
      "epoch": 0.658068717151381,
      "grad_norm": 185.0,
      "learning_rate": 5.530971314164211e-06,
      "loss": 0.8849,
      "num_input_tokens_seen": 3673608192,
      "step": 49520,
      "train_runtime": 58919.8546,
      "train_tokens_per_second": 62349.241
    },
    {
      "epoch": 0.658201606633843,
      "grad_norm": 22.375,
      "learning_rate": 5.527121477129167e-06,
      "loss": 0.7491,
      "num_input_tokens_seen": 3674083584,
      "step": 49530,
      "train_runtime": 58926.7236,
      "train_tokens_per_second": 62350.04
    },
    {
      "epoch": 0.6583344961163049,
      "grad_norm": 18.875,
      "learning_rate": 5.523272468654989e-06,
      "loss": 0.6406,
      "num_input_tokens_seen": 3674308096,
      "step": 49540,
      "train_runtime": 58930.9252,
      "train_tokens_per_second": 62349.405
    },
    {
      "epoch": 0.6584673855987668,
      "grad_norm": 16.125,
      "learning_rate": 5.51942428945466e-06,
      "loss": 0.4496,
      "num_input_tokens_seen": 3674422784,
      "step": 49550,
      "train_runtime": 58933.3295,
      "train_tokens_per_second": 62348.807
    },
    {
      "epoch": 0.6586002750812286,
      "grad_norm": 24.25,
      "learning_rate": 5.5155769402410384e-06,
      "loss": 0.8243,
      "num_input_tokens_seen": 3676419584,
      "step": 49560,
      "train_runtime": 58966.1379,
      "train_tokens_per_second": 62347.98
    },
    {
      "epoch": 0.6587331645636906,
      "grad_norm": 49.0,
      "learning_rate": 5.511730421726796e-06,
      "loss": 0.934,
      "num_input_tokens_seen": 3677219584,
      "step": 49570,
      "train_runtime": 58978.5423,
      "train_tokens_per_second": 62348.431
    },
    {
      "epoch": 0.6588660540461525,
      "grad_norm": 26.0,
      "learning_rate": 5.5078847346244735e-06,
      "loss": 0.7925,
      "num_input_tokens_seen": 3677683456,
      "step": 49580,
      "train_runtime": 58985.2007,
      "train_tokens_per_second": 62349.257
    },
    {
      "epoch": 0.6589989435286144,
      "grad_norm": 14.1875,
      "learning_rate": 5.504039879646451e-06,
      "loss": 0.6753,
      "num_input_tokens_seen": 3677905920,
      "step": 49590,
      "train_runtime": 58988.633,
      "train_tokens_per_second": 62349.401
    },
    {
      "epoch": 0.6591318330110764,
      "grad_norm": 21.625,
      "learning_rate": 5.500195857504944e-06,
      "loss": 0.552,
      "num_input_tokens_seen": 3678019840,
      "step": 49600,
      "train_runtime": 58991.0931,
      "train_tokens_per_second": 62348.732
    },
    {
      "epoch": 0.6592647224935383,
      "grad_norm": 15.3125,
      "learning_rate": 5.496352668912037e-06,
      "loss": 0.719,
      "num_input_tokens_seen": 3680039680,
      "step": 49610,
      "train_runtime": 59022.795,
      "train_tokens_per_second": 62349.465
    },
    {
      "epoch": 0.6593976119760001,
      "grad_norm": 15.6875,
      "learning_rate": 5.492510314579638e-06,
      "loss": 0.9256,
      "num_input_tokens_seen": 3680895232,
      "step": 49620,
      "train_runtime": 59036.0892,
      "train_tokens_per_second": 62349.916
    },
    {
      "epoch": 0.659530501458462,
      "grad_norm": 12.5,
      "learning_rate": 5.4886687952195095e-06,
      "loss": 0.7955,
      "num_input_tokens_seen": 3681377280,
      "step": 49630,
      "train_runtime": 59042.9166,
      "train_tokens_per_second": 62350.871
    },
    {
      "epoch": 0.659663390940924,
      "grad_norm": 10.5,
      "learning_rate": 5.4848281115432586e-06,
      "loss": 0.5823,
      "num_input_tokens_seen": 3681616384,
      "step": 49640,
      "train_runtime": 59046.469,
      "train_tokens_per_second": 62351.169
    },
    {
      "epoch": 0.6597962804233859,
      "grad_norm": 28.125,
      "learning_rate": 5.480988264262342e-06,
      "loss": 0.5475,
      "num_input_tokens_seen": 3681733376,
      "step": 49650,
      "train_runtime": 59048.8872,
      "train_tokens_per_second": 62350.597
    },
    {
      "epoch": 0.6599291699058478,
      "grad_norm": 12.75,
      "learning_rate": 5.477149254088052e-06,
      "loss": 0.8258,
      "num_input_tokens_seen": 3683781888,
      "step": 49660,
      "train_runtime": 59081.6487,
      "train_tokens_per_second": 62350.695
    },
    {
      "epoch": 0.6600620593883098,
      "grad_norm": 17.125,
      "learning_rate": 5.473311081731534e-06,
      "loss": 0.8892,
      "num_input_tokens_seen": 3684614400,
      "step": 49670,
      "train_runtime": 59094.3035,
      "train_tokens_per_second": 62351.431
    },
    {
      "epoch": 0.6601949488707716,
      "grad_norm": 28.625,
      "learning_rate": 5.469473747903777e-06,
      "loss": 0.6697,
      "num_input_tokens_seen": 3685070592,
      "step": 49680,
      "train_runtime": 59101.4929,
      "train_tokens_per_second": 62351.565
    },
    {
      "epoch": 0.6603278383532335,
      "grad_norm": 76.5,
      "learning_rate": 5.465637253315608e-06,
      "loss": 0.6148,
      "num_input_tokens_seen": 3685303040,
      "step": 49690,
      "train_runtime": 59105.03,
      "train_tokens_per_second": 62351.767
    },
    {
      "epoch": 0.6604607278356954,
      "grad_norm": 18.5,
      "learning_rate": 5.461801598677705e-06,
      "loss": 0.6026,
      "num_input_tokens_seen": 3685420288,
      "step": 49700,
      "train_runtime": 59107.4953,
      "train_tokens_per_second": 62351.15
    },
    {
      "epoch": 0.6605936173181574,
      "grad_norm": 14.0625,
      "learning_rate": 5.457966784700589e-06,
      "loss": 0.7635,
      "num_input_tokens_seen": 3687480832,
      "step": 49710,
      "train_runtime": 59140.1829,
      "train_tokens_per_second": 62351.529
    },
    {
      "epoch": 0.6607265068006193,
      "grad_norm": 13.125,
      "learning_rate": 5.4541328120946255e-06,
      "loss": 0.9123,
      "num_input_tokens_seen": 3688312320,
      "step": 49720,
      "train_runtime": 59152.5691,
      "train_tokens_per_second": 62352.53
    },
    {
      "epoch": 0.6608593962830812,
      "grad_norm": 14.4375,
      "learning_rate": 5.4502996815700245e-06,
      "loss": 0.672,
      "num_input_tokens_seen": 3688769024,
      "step": 49730,
      "train_runtime": 59159.6382,
      "train_tokens_per_second": 62352.799
    },
    {
      "epoch": 0.660992285765543,
      "grad_norm": 21.25,
      "learning_rate": 5.446467393836843e-06,
      "loss": 0.6061,
      "num_input_tokens_seen": 3688994048,
      "step": 49740,
      "train_runtime": 59163.029,
      "train_tokens_per_second": 62353.029
    },
    {
      "epoch": 0.661125175248005,
      "grad_norm": 10.125,
      "learning_rate": 5.442635949604969e-06,
      "loss": 0.5838,
      "num_input_tokens_seen": 3689105408,
      "step": 49750,
      "train_runtime": 59165.4065,
      "train_tokens_per_second": 62352.405
    },
    {
      "epoch": 0.6612580647304669,
      "grad_norm": 20.875,
      "learning_rate": 5.438805349584147e-06,
      "loss": 0.8121,
      "num_input_tokens_seen": 3691159296,
      "step": 49760,
      "train_runtime": 59198.0165,
      "train_tokens_per_second": 62352.753
    },
    {
      "epoch": 0.6613909542129288,
      "grad_norm": 19.25,
      "learning_rate": 5.434975594483965e-06,
      "loss": 0.864,
      "num_input_tokens_seen": 3691997696,
      "step": 49770,
      "train_runtime": 59211.0258,
      "train_tokens_per_second": 62353.213
    },
    {
      "epoch": 0.6615238436953907,
      "grad_norm": 14.1875,
      "learning_rate": 5.431146685013846e-06,
      "loss": 0.7648,
      "num_input_tokens_seen": 3692468736,
      "step": 49780,
      "train_runtime": 59217.622,
      "train_tokens_per_second": 62354.222
    },
    {
      "epoch": 0.6616567331778527,
      "grad_norm": 72.0,
      "learning_rate": 5.4273186218830685e-06,
      "loss": 0.6921,
      "num_input_tokens_seen": 3692689920,
      "step": 49790,
      "train_runtime": 59221.0114,
      "train_tokens_per_second": 62354.388
    },
    {
      "epoch": 0.6617896226603146,
      "grad_norm": 11.0625,
      "learning_rate": 5.423491405800736e-06,
      "loss": 0.5508,
      "num_input_tokens_seen": 3692801280,
      "step": 49800,
      "train_runtime": 59223.4394,
      "train_tokens_per_second": 62353.712
    },
    {
      "epoch": 0.6619225121427764,
      "grad_norm": 29.75,
      "learning_rate": 5.41966503747582e-06,
      "loss": 0.8533,
      "num_input_tokens_seen": 3694841088,
      "step": 49810,
      "train_runtime": 59256.8841,
      "train_tokens_per_second": 62352.943
    },
    {
      "epoch": 0.6620554016252383,
      "grad_norm": 11.25,
      "learning_rate": 5.415839517617105e-06,
      "loss": 0.9344,
      "num_input_tokens_seen": 3695719936,
      "step": 49820,
      "train_runtime": 59269.5661,
      "train_tokens_per_second": 62354.429
    },
    {
      "epoch": 0.6621882911077003,
      "grad_norm": 22.5,
      "learning_rate": 5.4120148469332515e-06,
      "loss": 0.7957,
      "num_input_tokens_seen": 3696160256,
      "step": 49830,
      "train_runtime": 59276.0318,
      "train_tokens_per_second": 62355.056
    },
    {
      "epoch": 0.6623211805901622,
      "grad_norm": 21.625,
      "learning_rate": 5.408191026132733e-06,
      "loss": 0.6356,
      "num_input_tokens_seen": 3696387584,
      "step": 49840,
      "train_runtime": 59280.2809,
      "train_tokens_per_second": 62354.421
    },
    {
      "epoch": 0.6624540700726241,
      "grad_norm": 12.125,
      "learning_rate": 5.4043680559238835e-06,
      "loss": 0.5058,
      "num_input_tokens_seen": 3696505088,
      "step": 49850,
      "train_runtime": 59282.7947,
      "train_tokens_per_second": 62353.759
    },
    {
      "epoch": 0.6625869595550861,
      "grad_norm": 29.375,
      "learning_rate": 5.400545937014876e-06,
      "loss": 0.8257,
      "num_input_tokens_seen": 3698607104,
      "step": 49860,
      "train_runtime": 59316.9618,
      "train_tokens_per_second": 62353.28
    },
    {
      "epoch": 0.6627198490375479,
      "grad_norm": 20.375,
      "learning_rate": 5.396724670113714e-06,
      "loss": 0.82,
      "num_input_tokens_seen": 3699464704,
      "step": 49870,
      "train_runtime": 59329.7549,
      "train_tokens_per_second": 62354.289
    },
    {
      "epoch": 0.6628527385200098,
      "grad_norm": 21.5,
      "learning_rate": 5.392904255928266e-06,
      "loss": 0.7079,
      "num_input_tokens_seen": 3699952896,
      "step": 49880,
      "train_runtime": 59337.3824,
      "train_tokens_per_second": 62354.501
    },
    {
      "epoch": 0.6629856280024717,
      "grad_norm": 22.0,
      "learning_rate": 5.3890846951662224e-06,
      "loss": 0.6172,
      "num_input_tokens_seen": 3700181760,
      "step": 49890,
      "train_runtime": 59340.8713,
      "train_tokens_per_second": 62354.692
    },
    {
      "epoch": 0.6631185174849337,
      "grad_norm": 9.4375,
      "learning_rate": 5.385265988535125e-06,
      "loss": 0.4995,
      "num_input_tokens_seen": 3700297216,
      "step": 49900,
      "train_runtime": 59343.2536,
      "train_tokens_per_second": 62354.135
    },
    {
      "epoch": 0.6632514069673956,
      "grad_norm": 22.75,
      "learning_rate": 5.381448136742356e-06,
      "loss": 0.8282,
      "num_input_tokens_seen": 3702407424,
      "step": 49910,
      "train_runtime": 59377.4676,
      "train_tokens_per_second": 62353.744
    },
    {
      "epoch": 0.6633842964498575,
      "grad_norm": 22.5,
      "learning_rate": 5.37763114049513e-06,
      "loss": 0.9473,
      "num_input_tokens_seen": 3703266304,
      "step": 49920,
      "train_runtime": 59390.3634,
      "train_tokens_per_second": 62354.667
    },
    {
      "epoch": 0.6635171859323193,
      "grad_norm": 15.6875,
      "learning_rate": 5.373815000500525e-06,
      "loss": 0.7451,
      "num_input_tokens_seen": 3703739392,
      "step": 49930,
      "train_runtime": 59397.3147,
      "train_tokens_per_second": 62355.334
    },
    {
      "epoch": 0.6636500754147813,
      "grad_norm": 25.25,
      "learning_rate": 5.369999717465437e-06,
      "loss": 0.6235,
      "num_input_tokens_seen": 3703973376,
      "step": 49940,
      "train_runtime": 59401.6198,
      "train_tokens_per_second": 62354.754
    },
    {
      "epoch": 0.6637829648972432,
      "grad_norm": 14.5625,
      "learning_rate": 5.366185292096618e-06,
      "loss": 0.5253,
      "num_input_tokens_seen": 3704089856,
      "step": 49950,
      "train_runtime": 59404.1121,
      "train_tokens_per_second": 62354.098
    },
    {
      "epoch": 0.6639158543797051,
      "grad_norm": 19.5,
      "learning_rate": 5.3623717251006524e-06,
      "loss": 0.8165,
      "num_input_tokens_seen": 3706260224,
      "step": 49960,
      "train_runtime": 59438.65,
      "train_tokens_per_second": 62354.381
    },
    {
      "epoch": 0.6640487438621671,
      "grad_norm": 34.0,
      "learning_rate": 5.358559017183973e-06,
      "loss": 0.9424,
      "num_input_tokens_seen": 3707088896,
      "step": 49970,
      "train_runtime": 59451.3908,
      "train_tokens_per_second": 62354.957
    },
    {
      "epoch": 0.664181633344629,
      "grad_norm": 47.25,
      "learning_rate": 5.354747169052847e-06,
      "loss": 0.7697,
      "num_input_tokens_seen": 3707597312,
      "step": 49980,
      "train_runtime": 59458.3345,
      "train_tokens_per_second": 62356.225
    },
    {
      "epoch": 0.6643145228270908,
      "grad_norm": 10.5,
      "learning_rate": 5.350936181413393e-06,
      "loss": 0.5912,
      "num_input_tokens_seen": 3707825152,
      "step": 49990,
      "train_runtime": 59462.5454,
      "train_tokens_per_second": 62355.641
    },
    {
      "epoch": 0.6644474123095527,
      "grad_norm": 31.0,
      "learning_rate": 5.347126054971552e-06,
      "loss": 0.5097,
      "num_input_tokens_seen": 3707942912,
      "step": 50000,
      "train_runtime": 59465.0009,
      "train_tokens_per_second": 62355.047
    },
    {
      "epoch": 0.6644474123095527,
      "eval_accuracy": 0.819,
      "eval_balanced_accuracy": 0.8171249559272302,
      "eval_invalid_f1": 0.8368781542898344,
      "eval_invalid_precision": 0.8402099167571481,
      "eval_invalid_recall": 0.8335727109515261,
      "eval_loss": 0.36118167638778687,
      "eval_macro_f1": 0.8167993646561481,
      "eval_mcc": 0.63363104286253,
      "eval_runtime": 79.7513,
      "eval_samples_per_second": 125.39,
      "eval_steps_per_second": 31.347,
      "eval_valid_f1": 0.7967205750224619,
      "eval_valid_precision": 0.7928028609745195,
      "eval_valid_recall": 0.8006772009029345,
      "num_input_tokens_seen": 3707942912,
      "step": 50000
    },
    {
      "epoch": 0.6645803017920147,
      "grad_norm": 67.0,
      "learning_rate": 5.3433167904331196e-06,
      "loss": 0.8504,
      "num_input_tokens_seen": 3709976576,
      "step": 50010,
      "train_runtime": 59581.2013,
      "train_tokens_per_second": 62267.569
    },
    {
      "epoch": 0.6647131912744766,
      "grad_norm": 14.5,
      "learning_rate": 5.33950838850373e-06,
      "loss": 0.8833,
      "num_input_tokens_seen": 3710831104,
      "step": 50020,
      "train_runtime": 59594.1442,
      "train_tokens_per_second": 62268.385
    },
    {
      "epoch": 0.6648460807569385,
      "grad_norm": 16.125,
      "learning_rate": 5.335700849888854e-06,
      "loss": 0.8141,
      "num_input_tokens_seen": 3711328000,
      "step": 50030,
      "train_runtime": 59601.1485,
      "train_tokens_per_second": 62269.404
    },
    {
      "epoch": 0.6649789702394004,
      "grad_norm": 24.625,
      "learning_rate": 5.331894175293805e-06,
      "loss": 0.6671,
      "num_input_tokens_seen": 3711561728,
      "step": 50040,
      "train_runtime": 59605.5694,
      "train_tokens_per_second": 62268.707
    },
    {
      "epoch": 0.6651118597218623,
      "grad_norm": 8.25,
      "learning_rate": 5.328088365423736e-06,
      "loss": 0.4211,
      "num_input_tokens_seen": 3711673088,
      "step": 50050,
      "train_runtime": 59608.002,
      "train_tokens_per_second": 62268.034
    },
    {
      "epoch": 0.6652447492043242,
      "grad_norm": 18.5,
      "learning_rate": 5.32428342098364e-06,
      "loss": 0.9012,
      "num_input_tokens_seen": 3713671680,
      "step": 50060,
      "train_runtime": 59640.0252,
      "train_tokens_per_second": 62268.111
    },
    {
      "epoch": 0.6653776386867861,
      "grad_norm": 25.875,
      "learning_rate": 5.320479342678346e-06,
      "loss": 0.9271,
      "num_input_tokens_seen": 3714523392,
      "step": 50070,
      "train_runtime": 59652.9974,
      "train_tokens_per_second": 62268.847
    },
    {
      "epoch": 0.665510528169248,
      "grad_norm": 16.75,
      "learning_rate": 5.316676131212527e-06,
      "loss": 0.7363,
      "num_input_tokens_seen": 3715028736,
      "step": 50080,
      "train_runtime": 59660.1253,
      "train_tokens_per_second": 62269.878
    },
    {
      "epoch": 0.66564341765171,
      "grad_norm": 14.6875,
      "learning_rate": 5.3128737872906935e-06,
      "loss": 0.6022,
      "num_input_tokens_seen": 3715270912,
      "step": 50090,
      "train_runtime": 59663.8289,
      "train_tokens_per_second": 62270.072
    },
    {
      "epoch": 0.6657763071341719,
      "grad_norm": 18.5,
      "learning_rate": 5.309072311617197e-06,
      "loss": 0.5079,
      "num_input_tokens_seen": 3715387136,
      "step": 50100,
      "train_runtime": 59666.2484,
      "train_tokens_per_second": 62269.495
    },
    {
      "epoch": 0.6659091966166337,
      "grad_norm": 16.0,
      "learning_rate": 5.305271704896231e-06,
      "loss": 0.8676,
      "num_input_tokens_seen": 3717358080,
      "step": 50110,
      "train_runtime": 59697.8403,
      "train_tokens_per_second": 62269.557
    },
    {
      "epoch": 0.6660420860990957,
      "grad_norm": 31.25,
      "learning_rate": 5.3014719678318105e-06,
      "loss": 0.9245,
      "num_input_tokens_seen": 3718225920,
      "step": 50120,
      "train_runtime": 59711.1586,
      "train_tokens_per_second": 62270.202
    },
    {
      "epoch": 0.6661749755815576,
      "grad_norm": 14.9375,
      "learning_rate": 5.2976731011278205e-06,
      "loss": 0.7029,
      "num_input_tokens_seen": 3718665472,
      "step": 50130,
      "train_runtime": 59717.6128,
      "train_tokens_per_second": 62270.833
    },
    {
      "epoch": 0.6663078650640195,
      "grad_norm": 12.375,
      "learning_rate": 5.293875105487955e-06,
      "loss": 0.6312,
      "num_input_tokens_seen": 3718896384,
      "step": 50140,
      "train_runtime": 59721.0644,
      "train_tokens_per_second": 62271.1
    },
    {
      "epoch": 0.6664407545464814,
      "grad_norm": 16.125,
      "learning_rate": 5.2900779816157625e-06,
      "loss": 0.5742,
      "num_input_tokens_seen": 3719014144,
      "step": 50150,
      "train_runtime": 59723.5526,
      "train_tokens_per_second": 62270.478
    },
    {
      "epoch": 0.6665736440289434,
      "grad_norm": 16.5,
      "learning_rate": 5.28628173021463e-06,
      "loss": 0.7795,
      "num_input_tokens_seen": 3721076992,
      "step": 50160,
      "train_runtime": 59757.1153,
      "train_tokens_per_second": 62270.024
    },
    {
      "epoch": 0.6667065335114052,
      "grad_norm": 36.75,
      "learning_rate": 5.2824863519877674e-06,
      "loss": 0.8463,
      "num_input_tokens_seen": 3721936128,
      "step": 50170,
      "train_runtime": 59770.5146,
      "train_tokens_per_second": 62270.438
    },
    {
      "epoch": 0.6668394229938671,
      "grad_norm": 29.625,
      "learning_rate": 5.278691847638251e-06,
      "loss": 0.7542,
      "num_input_tokens_seen": 3722432512,
      "step": 50180,
      "train_runtime": 59777.5308,
      "train_tokens_per_second": 62271.433
    },
    {
      "epoch": 0.666972312476329,
      "grad_norm": 14.4375,
      "learning_rate": 5.2748982178689625e-06,
      "loss": 0.5369,
      "num_input_tokens_seen": 3722664192,
      "step": 50190,
      "train_runtime": 59781.0547,
      "train_tokens_per_second": 62271.638
    },
    {
      "epoch": 0.667105201958791,
      "grad_norm": 25.0,
      "learning_rate": 5.271105463382654e-06,
      "loss": 0.5769,
      "num_input_tokens_seen": 3722783232,
      "step": 50200,
      "train_runtime": 59783.4747,
      "train_tokens_per_second": 62271.108
    },
    {
      "epoch": 0.6672380914412529,
      "grad_norm": 16.125,
      "learning_rate": 5.267313584881888e-06,
      "loss": 0.8232,
      "num_input_tokens_seen": 3724795392,
      "step": 50210,
      "train_runtime": 59815.8887,
      "train_tokens_per_second": 62271.003
    },
    {
      "epoch": 0.6673709809237148,
      "grad_norm": 12.875,
      "learning_rate": 5.2635225830690806e-06,
      "loss": 0.9576,
      "num_input_tokens_seen": 3725620992,
      "step": 50220,
      "train_runtime": 59827.9287,
      "train_tokens_per_second": 62272.271
    },
    {
      "epoch": 0.6675038704061768,
      "grad_norm": 26.375,
      "learning_rate": 5.259732458646482e-06,
      "loss": 0.7592,
      "num_input_tokens_seen": 3726083328,
      "step": 50230,
      "train_runtime": 59834.4072,
      "train_tokens_per_second": 62273.256
    },
    {
      "epoch": 0.6676367598886386,
      "grad_norm": 43.75,
      "learning_rate": 5.2559432123161716e-06,
      "loss": 0.6176,
      "num_input_tokens_seen": 3726307328,
      "step": 50240,
      "train_runtime": 59837.8353,
      "train_tokens_per_second": 62273.431
    },
    {
      "epoch": 0.6677696493711005,
      "grad_norm": 25.625,
      "learning_rate": 5.252154844780086e-06,
      "loss": 0.5448,
      "num_input_tokens_seen": 3726419968,
      "step": 50250,
      "train_runtime": 59840.2455,
      "train_tokens_per_second": 62272.805
    },
    {
      "epoch": 0.6679025388535624,
      "grad_norm": 17.5,
      "learning_rate": 5.248367356739977e-06,
      "loss": 0.8276,
      "num_input_tokens_seen": 3728569600,
      "step": 50260,
      "train_runtime": 59874.447,
      "train_tokens_per_second": 62273.136
    },
    {
      "epoch": 0.6680354283360244,
      "grad_norm": 14.6875,
      "learning_rate": 5.244580748897443e-06,
      "loss": 0.9085,
      "num_input_tokens_seen": 3729425152,
      "step": 50270,
      "train_runtime": 59887.0715,
      "train_tokens_per_second": 62274.295
    },
    {
      "epoch": 0.6681683178184863,
      "grad_norm": 27.125,
      "learning_rate": 5.240795021953924e-06,
      "loss": 0.7552,
      "num_input_tokens_seen": 3729910272,
      "step": 50280,
      "train_runtime": 59894.4356,
      "train_tokens_per_second": 62274.738
    },
    {
      "epoch": 0.6683012073009482,
      "grad_norm": 15.875,
      "learning_rate": 5.237010176610689e-06,
      "loss": 0.6123,
      "num_input_tokens_seen": 3730138880,
      "step": 50290,
      "train_runtime": 59897.928,
      "train_tokens_per_second": 62274.923
    },
    {
      "epoch": 0.66843409678341,
      "grad_norm": 17.5,
      "learning_rate": 5.23322621356885e-06,
      "loss": 0.5131,
      "num_input_tokens_seen": 3730255616,
      "step": 50300,
      "train_runtime": 59900.3261,
      "train_tokens_per_second": 62274.379
    },
    {
      "epoch": 0.668566986265872,
      "grad_norm": 23.125,
      "learning_rate": 5.229443133529352e-06,
      "loss": 0.8127,
      "num_input_tokens_seen": 3732284416,
      "step": 50310,
      "train_runtime": 59933.1066,
      "train_tokens_per_second": 62274.169
    },
    {
      "epoch": 0.6686998757483339,
      "grad_norm": 18.25,
      "learning_rate": 5.225660937192973e-06,
      "loss": 0.9488,
      "num_input_tokens_seen": 3733128704,
      "step": 50320,
      "train_runtime": 59945.8367,
      "train_tokens_per_second": 62275.029
    },
    {
      "epoch": 0.6688327652307958,
      "grad_norm": 16.875,
      "learning_rate": 5.221879625260332e-06,
      "loss": 0.7901,
      "num_input_tokens_seen": 3733581312,
      "step": 50330,
      "train_runtime": 59953.1019,
      "train_tokens_per_second": 62275.032
    },
    {
      "epoch": 0.6689656547132578,
      "grad_norm": 17.375,
      "learning_rate": 5.218099198431885e-06,
      "loss": 0.6513,
      "num_input_tokens_seen": 3733810176,
      "step": 50340,
      "train_runtime": 59956.5199,
      "train_tokens_per_second": 62275.298
    },
    {
      "epoch": 0.6690985441957197,
      "grad_norm": 24.375,
      "learning_rate": 5.2143196574079224e-06,
      "loss": 0.5662,
      "num_input_tokens_seen": 3733925632,
      "step": 50350,
      "train_runtime": 59958.8985,
      "train_tokens_per_second": 62274.754
    },
    {
      "epoch": 0.6692314336781815,
      "grad_norm": 70.5,
      "learning_rate": 5.21054100288857e-06,
      "loss": 0.7949,
      "num_input_tokens_seen": 3736021248,
      "step": 50360,
      "train_runtime": 59993.0528,
      "train_tokens_per_second": 62274.231
    },
    {
      "epoch": 0.6693643231606434,
      "grad_norm": 27.5,
      "learning_rate": 5.206763235573791e-06,
      "loss": 0.8544,
      "num_input_tokens_seen": 3736925184,
      "step": 50370,
      "train_runtime": 60005.6027,
      "train_tokens_per_second": 62276.271
    },
    {
      "epoch": 0.6694972126431054,
      "grad_norm": 19.25,
      "learning_rate": 5.202986356163385e-06,
      "loss": 0.7802,
      "num_input_tokens_seen": 3737384448,
      "step": 50380,
      "train_runtime": 60012.2974,
      "train_tokens_per_second": 62276.977
    },
    {
      "epoch": 0.6696301021255673,
      "grad_norm": 18.25,
      "learning_rate": 5.199210365356979e-06,
      "loss": 0.5987,
      "num_input_tokens_seen": 3737621760,
      "step": 50390,
      "train_runtime": 60015.8371,
      "train_tokens_per_second": 62277.258
    },
    {
      "epoch": 0.6697629916080292,
      "grad_norm": 7.6875,
      "learning_rate": 5.195435263854045e-06,
      "loss": 0.4633,
      "num_input_tokens_seen": 3737742080,
      "step": 50400,
      "train_runtime": 60018.2642,
      "train_tokens_per_second": 62276.744
    },
    {
      "epoch": 0.6698958810904911,
      "grad_norm": 31.125,
      "learning_rate": 5.191661052353886e-06,
      "loss": 0.8154,
      "num_input_tokens_seen": 3739770624,
      "step": 50410,
      "train_runtime": 60050.1729,
      "train_tokens_per_second": 62277.433
    },
    {
      "epoch": 0.670028770572953,
      "grad_norm": 23.625,
      "learning_rate": 5.187887731555642e-06,
      "loss": 0.8792,
      "num_input_tokens_seen": 3740596224,
      "step": 50420,
      "train_runtime": 60062.7659,
      "train_tokens_per_second": 62278.121
    },
    {
      "epoch": 0.6701616600554149,
      "grad_norm": 23.125,
      "learning_rate": 5.184115302158292e-06,
      "loss": 0.7033,
      "num_input_tokens_seen": 3741073664,
      "step": 50430,
      "train_runtime": 60069.4995,
      "train_tokens_per_second": 62279.088
    },
    {
      "epoch": 0.6702945495378768,
      "grad_norm": 36.25,
      "learning_rate": 5.1803437648606305e-06,
      "loss": 0.6314,
      "num_input_tokens_seen": 3741307904,
      "step": 50440,
      "train_runtime": 60072.9956,
      "train_tokens_per_second": 62279.363
    },
    {
      "epoch": 0.6704274390203387,
      "grad_norm": 21.625,
      "learning_rate": 5.176573120361319e-06,
      "loss": 0.5662,
      "num_input_tokens_seen": 3741423872,
      "step": 50450,
      "train_runtime": 60075.3947,
      "train_tokens_per_second": 62278.806
    },
    {
      "epoch": 0.6705603285028007,
      "grad_norm": 27.625,
      "learning_rate": 5.172803369358823e-06,
      "loss": 0.816,
      "num_input_tokens_seen": 3743377408,
      "step": 50460,
      "train_runtime": 60106.7178,
      "train_tokens_per_second": 62278.852
    },
    {
      "epoch": 0.6706932179852626,
      "grad_norm": 23.375,
      "learning_rate": 5.169034512551461e-06,
      "loss": 0.8905,
      "num_input_tokens_seen": 3744206848,
      "step": 50470,
      "train_runtime": 60119.0221,
      "train_tokens_per_second": 62279.903
    },
    {
      "epoch": 0.6708261074677244,
      "grad_norm": 24.125,
      "learning_rate": 5.16526655063738e-06,
      "loss": 0.7271,
      "num_input_tokens_seen": 3744682496,
      "step": 50480,
      "train_runtime": 60125.6084,
      "train_tokens_per_second": 62280.991
    },
    {
      "epoch": 0.6709589969501863,
      "grad_norm": 13.5625,
      "learning_rate": 5.161499484314554e-06,
      "loss": 0.6529,
      "num_input_tokens_seen": 3744908032,
      "step": 50490,
      "train_runtime": 60129.0555,
      "train_tokens_per_second": 62281.172
    },
    {
      "epoch": 0.6710918864326483,
      "grad_norm": 8.6875,
      "learning_rate": 5.15773331428081e-06,
      "loss": 0.5598,
      "num_input_tokens_seen": 3745024256,
      "step": 50500,
      "train_runtime": 60132.2596,
      "train_tokens_per_second": 62279.786
    },
    {
      "epoch": 0.6712247759151102,
      "grad_norm": 39.75,
      "learning_rate": 5.1539680412337876e-06,
      "loss": 0.8707,
      "num_input_tokens_seen": 3746966272,
      "step": 50510,
      "train_runtime": 60163.7516,
      "train_tokens_per_second": 62279.465
    },
    {
      "epoch": 0.6713576653975721,
      "grad_norm": 25.5,
      "learning_rate": 5.15020366587098e-06,
      "loss": 0.8441,
      "num_input_tokens_seen": 3747783936,
      "step": 50520,
      "train_runtime": 60175.8815,
      "train_tokens_per_second": 62280.499
    },
    {
      "epoch": 0.6714905548800341,
      "grad_norm": 33.75,
      "learning_rate": 5.146440188889694e-06,
      "loss": 0.7056,
      "num_input_tokens_seen": 3748241152,
      "step": 50530,
      "train_runtime": 60182.2727,
      "train_tokens_per_second": 62281.482
    },
    {
      "epoch": 0.6716234443624959,
      "grad_norm": 209.0,
      "learning_rate": 5.142677610987085e-06,
      "loss": 0.6,
      "num_input_tokens_seen": 3748456192,
      "step": 50540,
      "train_runtime": 60185.6345,
      "train_tokens_per_second": 62281.576
    },
    {
      "epoch": 0.6717563338449578,
      "grad_norm": 356.0,
      "learning_rate": 5.138915932860142e-06,
      "loss": 0.5915,
      "num_input_tokens_seen": 3748567808,
      "step": 50550,
      "train_runtime": 60188.0045,
      "train_tokens_per_second": 62280.978
    },
    {
      "epoch": 0.6718892233274197,
      "grad_norm": 22.5,
      "learning_rate": 5.1351551552056685e-06,
      "loss": 0.8127,
      "num_input_tokens_seen": 3750619136,
      "step": 50560,
      "train_runtime": 60220.463,
      "train_tokens_per_second": 62281.473
    },
    {
      "epoch": 0.6720221128098817,
      "grad_norm": 21.75,
      "learning_rate": 5.13139527872033e-06,
      "loss": 0.9173,
      "num_input_tokens_seen": 3751439616,
      "step": 50570,
      "train_runtime": 60233.5298,
      "train_tokens_per_second": 62281.584
    },
    {
      "epoch": 0.6721550022923436,
      "grad_norm": 19.125,
      "learning_rate": 5.127636304100601e-06,
      "loss": 0.7471,
      "num_input_tokens_seen": 3751891200,
      "step": 50580,
      "train_runtime": 60240.0839,
      "train_tokens_per_second": 62282.304
    },
    {
      "epoch": 0.6722878917748055,
      "grad_norm": 32.25,
      "learning_rate": 5.1238782320428006e-06,
      "loss": 0.6588,
      "num_input_tokens_seen": 3752118272,
      "step": 50590,
      "train_runtime": 60243.4644,
      "train_tokens_per_second": 62282.578
    },
    {
      "epoch": 0.6724207812572673,
      "grad_norm": 9.9375,
      "learning_rate": 5.1201210632430775e-06,
      "loss": 0.4774,
      "num_input_tokens_seen": 3752231936,
      "step": 50600,
      "train_runtime": 60245.8409,
      "train_tokens_per_second": 62282.008
    },
    {
      "epoch": 0.6725536707397293,
      "grad_norm": 40.25,
      "learning_rate": 5.116364798397414e-06,
      "loss": 0.8484,
      "num_input_tokens_seen": 3754237952,
      "step": 50610,
      "train_runtime": 60278.1348,
      "train_tokens_per_second": 62281.92
    },
    {
      "epoch": 0.6726865602221912,
      "grad_norm": 22.875,
      "learning_rate": 5.112609438201626e-06,
      "loss": 0.9547,
      "num_input_tokens_seen": 3755061760,
      "step": 50620,
      "train_runtime": 60291.5158,
      "train_tokens_per_second": 62281.761
    },
    {
      "epoch": 0.6728194497046531,
      "grad_norm": 18.75,
      "learning_rate": 5.108854983351363e-06,
      "loss": 0.7364,
      "num_input_tokens_seen": 3755526144,
      "step": 50630,
      "train_runtime": 60298.2595,
      "train_tokens_per_second": 62282.497
    },
    {
      "epoch": 0.6729523391871151,
      "grad_norm": 29.625,
      "learning_rate": 5.105101434542097e-06,
      "loss": 0.7275,
      "num_input_tokens_seen": 3755749888,
      "step": 50640,
      "train_runtime": 60301.6882,
      "train_tokens_per_second": 62282.666
    },
    {
      "epoch": 0.673085228669577,
      "grad_norm": 22.625,
      "learning_rate": 5.101348792469143e-06,
      "loss": 0.5323,
      "num_input_tokens_seen": 3755863552,
      "step": 50650,
      "train_runtime": 60304.0817,
      "train_tokens_per_second": 62282.078
    },
    {
      "epoch": 0.6732181181520388,
      "grad_norm": 32.5,
      "learning_rate": 5.097597057827644e-06,
      "loss": 0.8072,
      "num_input_tokens_seen": 3757828608,
      "step": 50660,
      "train_runtime": 60335.9834,
      "train_tokens_per_second": 62281.716
    },
    {
      "epoch": 0.6733510076345007,
      "grad_norm": 15.6875,
      "learning_rate": 5.093846231312578e-06,
      "loss": 0.8857,
      "num_input_tokens_seen": 3758685952,
      "step": 50670,
      "train_runtime": 60349.0122,
      "train_tokens_per_second": 62282.477
    },
    {
      "epoch": 0.6734838971169627,
      "grad_norm": 18.5,
      "learning_rate": 5.090096313618753e-06,
      "loss": 0.7335,
      "num_input_tokens_seen": 3759175424,
      "step": 50680,
      "train_runtime": 60355.9495,
      "train_tokens_per_second": 62283.428
    },
    {
      "epoch": 0.6736167865994246,
      "grad_norm": 15.9375,
      "learning_rate": 5.0863473054407975e-06,
      "loss": 0.5944,
      "num_input_tokens_seen": 3759404544,
      "step": 50690,
      "train_runtime": 60359.3967,
      "train_tokens_per_second": 62283.667
    },
    {
      "epoch": 0.6737496760818865,
      "grad_norm": 18.625,
      "learning_rate": 5.082599207473197e-06,
      "loss": 0.5713,
      "num_input_tokens_seen": 3759522816,
      "step": 50700,
      "train_runtime": 60362.4282,
      "train_tokens_per_second": 62282.498
    },
    {
      "epoch": 0.6738825655643484,
      "grad_norm": 15.4375,
      "learning_rate": 5.078852020410241e-06,
      "loss": 0.8337,
      "num_input_tokens_seen": 3761658112,
      "step": 50710,
      "train_runtime": 60395.8525,
      "train_tokens_per_second": 62283.385
    },
    {
      "epoch": 0.6740154550468104,
      "grad_norm": 57.0,
      "learning_rate": 5.075105744946069e-06,
      "loss": 0.9422,
      "num_input_tokens_seen": 3762537984,
      "step": 50720,
      "train_runtime": 60408.8647,
      "train_tokens_per_second": 62284.534
    },
    {
      "epoch": 0.6741483445292722,
      "grad_norm": 23.25,
      "learning_rate": 5.071360381774642e-06,
      "loss": 0.7583,
      "num_input_tokens_seen": 3763016704,
      "step": 50730,
      "train_runtime": 60415.7523,
      "train_tokens_per_second": 62285.357
    },
    {
      "epoch": 0.6742812340117341,
      "grad_norm": 16.5,
      "learning_rate": 5.067615931589757e-06,
      "loss": 0.6962,
      "num_input_tokens_seen": 3763255808,
      "step": 50740,
      "train_runtime": 60419.3355,
      "train_tokens_per_second": 62285.621
    },
    {
      "epoch": 0.674414123494196,
      "grad_norm": 40.75,
      "learning_rate": 5.063872395085044e-06,
      "loss": 0.5397,
      "num_input_tokens_seen": 3763376128,
      "step": 50750,
      "train_runtime": 60422.5972,
      "train_tokens_per_second": 62284.25
    },
    {
      "epoch": 0.674547012976658,
      "grad_norm": 416.0,
      "learning_rate": 5.0601297729539455e-06,
      "loss": 0.8835,
      "num_input_tokens_seen": 3765421824,
      "step": 50760,
      "train_runtime": 60456.3245,
      "train_tokens_per_second": 62283.34
    },
    {
      "epoch": 0.6746799024591199,
      "grad_norm": 20.375,
      "learning_rate": 5.056388065889768e-06,
      "loss": 0.903,
      "num_input_tokens_seen": 3766268160,
      "step": 50770,
      "train_runtime": 60468.9411,
      "train_tokens_per_second": 62284.341
    },
    {
      "epoch": 0.6748127919415818,
      "grad_norm": 19.125,
      "learning_rate": 5.052647274585615e-06,
      "loss": 0.6339,
      "num_input_tokens_seen": 3766729984,
      "step": 50780,
      "train_runtime": 60475.3379,
      "train_tokens_per_second": 62285.39
    },
    {
      "epoch": 0.6749456814240437,
      "grad_norm": 13.125,
      "learning_rate": 5.04890739973444e-06,
      "loss": 0.6425,
      "num_input_tokens_seen": 3766960896,
      "step": 50790,
      "train_runtime": 60478.716,
      "train_tokens_per_second": 62285.729
    },
    {
      "epoch": 0.6750785709065056,
      "grad_norm": 14.5,
      "learning_rate": 5.0451684420290245e-06,
      "loss": 0.4952,
      "num_input_tokens_seen": 3767071744,
      "step": 50800,
      "train_runtime": 60481.8824,
      "train_tokens_per_second": 62284.301
    },
    {
      "epoch": 0.6752114603889675,
      "grad_norm": 17.5,
      "learning_rate": 5.041430402161967e-06,
      "loss": 0.8311,
      "num_input_tokens_seen": 3769006848,
      "step": 50810,
      "train_runtime": 60514.0459,
      "train_tokens_per_second": 62283.174
    },
    {
      "epoch": 0.6753443498714294,
      "grad_norm": 26.0,
      "learning_rate": 5.037693280825719e-06,
      "loss": 0.9466,
      "num_input_tokens_seen": 3769828352,
      "step": 50820,
      "train_runtime": 60526.3347,
      "train_tokens_per_second": 62284.101
    },
    {
      "epoch": 0.6754772393538914,
      "grad_norm": 17.375,
      "learning_rate": 5.03395707871254e-06,
      "loss": 0.6868,
      "num_input_tokens_seen": 3770264576,
      "step": 50830,
      "train_runtime": 60532.6899,
      "train_tokens_per_second": 62284.768
    },
    {
      "epoch": 0.6756101288363533,
      "grad_norm": 14.875,
      "learning_rate": 5.030221796514531e-06,
      "loss": 0.632,
      "num_input_tokens_seen": 3770492928,
      "step": 50840,
      "train_runtime": 60536.0752,
      "train_tokens_per_second": 62285.058
    },
    {
      "epoch": 0.6757430183188151,
      "grad_norm": 17.25,
      "learning_rate": 5.026487434923619e-06,
      "loss": 0.5288,
      "num_input_tokens_seen": 3770606336,
      "step": 50850,
      "train_runtime": 60538.4533,
      "train_tokens_per_second": 62284.484
    },
    {
      "epoch": 0.675875907801277,
      "grad_norm": 22.0,
      "learning_rate": 5.0227539946315615e-06,
      "loss": 0.8335,
      "num_input_tokens_seen": 3772657920,
      "step": 50860,
      "train_runtime": 60571.5063,
      "train_tokens_per_second": 62284.367
    },
    {
      "epoch": 0.676008797283739,
      "grad_norm": 19.75,
      "learning_rate": 5.019021476329945e-06,
      "loss": 0.897,
      "num_input_tokens_seen": 3773527040,
      "step": 50870,
      "train_runtime": 60584.4665,
      "train_tokens_per_second": 62285.389
    },
    {
      "epoch": 0.6761416867662009,
      "grad_norm": 21.5,
      "learning_rate": 5.015289880710186e-06,
      "loss": 0.7189,
      "num_input_tokens_seen": 3774009088,
      "step": 50880,
      "train_runtime": 60591.3079,
      "train_tokens_per_second": 62286.312
    },
    {
      "epoch": 0.6762745762486628,
      "grad_norm": 31.875,
      "learning_rate": 5.011559208463533e-06,
      "loss": 0.6103,
      "num_input_tokens_seen": 3774235392,
      "step": 50890,
      "train_runtime": 60594.7292,
      "train_tokens_per_second": 62286.53
    },
    {
      "epoch": 0.6764074657311248,
      "grad_norm": 15.25,
      "learning_rate": 5.007829460281051e-06,
      "loss": 0.4644,
      "num_input_tokens_seen": 3774352640,
      "step": 50900,
      "train_runtime": 60597.1462,
      "train_tokens_per_second": 62285.98
    },
    {
      "epoch": 0.6765403552135866,
      "grad_norm": 13.9375,
      "learning_rate": 5.004100636853646e-06,
      "loss": 0.774,
      "num_input_tokens_seen": 3776413952,
      "step": 50910,
      "train_runtime": 60631.3695,
      "train_tokens_per_second": 62284.82
    },
    {
      "epoch": 0.6766732446960485,
      "grad_norm": 13.9375,
      "learning_rate": 5.000372738872052e-06,
      "loss": 0.9521,
      "num_input_tokens_seen": 3777257472,
      "step": 50920,
      "train_runtime": 60643.979,
      "train_tokens_per_second": 62285.779
    },
    {
      "epoch": 0.6768061341785104,
      "grad_norm": 16.5,
      "learning_rate": 4.9966457670268275e-06,
      "loss": 0.7445,
      "num_input_tokens_seen": 3777724928,
      "step": 50930,
      "train_runtime": 60650.7732,
      "train_tokens_per_second": 62286.509
    },
    {
      "epoch": 0.6769390236609724,
      "grad_norm": 10.3125,
      "learning_rate": 4.9929197220083605e-06,
      "loss": 0.6333,
      "num_input_tokens_seen": 3777955072,
      "step": 50940,
      "train_runtime": 60654.1985,
      "train_tokens_per_second": 62286.786
    },
    {
      "epoch": 0.6770719131434343,
      "grad_norm": 10.1875,
      "learning_rate": 4.98919460450687e-06,
      "loss": 0.5143,
      "num_input_tokens_seen": 3778069248,
      "step": 50950,
      "train_runtime": 60657.4029,
      "train_tokens_per_second": 62285.378
    },
    {
      "epoch": 0.6772048026258962,
      "grad_norm": 12.1875,
      "learning_rate": 4.985470415212396e-06,
      "loss": 0.797,
      "num_input_tokens_seen": 3780091904,
      "step": 50960,
      "train_runtime": 60690.5689,
      "train_tokens_per_second": 62284.668
    },
    {
      "epoch": 0.677337692108358,
      "grad_norm": 26.875,
      "learning_rate": 4.981747154814813e-06,
      "loss": 0.9147,
      "num_input_tokens_seen": 3780950528,
      "step": 50970,
      "train_runtime": 60702.9302,
      "train_tokens_per_second": 62286.129
    },
    {
      "epoch": 0.67747058159082,
      "grad_norm": 25.5,
      "learning_rate": 4.978024824003823e-06,
      "loss": 0.7681,
      "num_input_tokens_seen": 3781388800,
      "step": 50980,
      "train_runtime": 60709.3416,
      "train_tokens_per_second": 62286.77
    },
    {
      "epoch": 0.6776034710732819,
      "grad_norm": 22.125,
      "learning_rate": 4.974303423468955e-06,
      "loss": 0.663,
      "num_input_tokens_seen": 3781615104,
      "step": 50990,
      "train_runtime": 60712.6964,
      "train_tokens_per_second": 62287.056
    },
    {
      "epoch": 0.6777363605557438,
      "grad_norm": 13.75,
      "learning_rate": 4.970582953899567e-06,
      "loss": 0.5112,
      "num_input_tokens_seen": 3781728768,
      "step": 51000,
      "train_runtime": 60715.1118,
      "train_tokens_per_second": 62286.45
    },
    {
      "epoch": 0.6778692500382058,
      "grad_norm": 19.0,
      "learning_rate": 4.966863415984834e-06,
      "loss": 0.8382,
      "num_input_tokens_seen": 3783762432,
      "step": 51010,
      "train_runtime": 60747.5293,
      "train_tokens_per_second": 62286.688
    },
    {
      "epoch": 0.6780021395206677,
      "grad_norm": 14.875,
      "learning_rate": 4.96314481041378e-06,
      "loss": 0.9044,
      "num_input_tokens_seen": 3784587776,
      "step": 51020,
      "train_runtime": 60760.6961,
      "train_tokens_per_second": 62286.775
    },
    {
      "epoch": 0.6781350290031295,
      "grad_norm": 37.75,
      "learning_rate": 4.95942713787523e-06,
      "loss": 0.7415,
      "num_input_tokens_seen": 3785029376,
      "step": 51030,
      "train_runtime": 60767.2159,
      "train_tokens_per_second": 62287.359
    },
    {
      "epoch": 0.6782679184855914,
      "grad_norm": 15.5,
      "learning_rate": 4.955710399057865e-06,
      "loss": 0.6386,
      "num_input_tokens_seen": 3785258752,
      "step": 51040,
      "train_runtime": 60770.7196,
      "train_tokens_per_second": 62287.542
    },
    {
      "epoch": 0.6784008079680534,
      "grad_norm": 10.375,
      "learning_rate": 4.9519945946501646e-06,
      "loss": 0.5182,
      "num_input_tokens_seen": 3785375232,
      "step": 51050,
      "train_runtime": 60773.1208,
      "train_tokens_per_second": 62286.998
    },
    {
      "epoch": 0.6785336974505153,
      "grad_norm": 21.625,
      "learning_rate": 4.948279725340454e-06,
      "loss": 0.8304,
      "num_input_tokens_seen": 3787409408,
      "step": 51060,
      "train_runtime": 60807.0035,
      "train_tokens_per_second": 62285.743
    },
    {
      "epoch": 0.6786665869329772,
      "grad_norm": 22.5,
      "learning_rate": 4.944565791816882e-06,
      "loss": 0.8924,
      "num_input_tokens_seen": 3788252160,
      "step": 51070,
      "train_runtime": 60819.6662,
      "train_tokens_per_second": 62286.632
    },
    {
      "epoch": 0.6787994764154391,
      "grad_norm": 25.0,
      "learning_rate": 4.940852794767412e-06,
      "loss": 0.7298,
      "num_input_tokens_seen": 3788713216,
      "step": 51080,
      "train_runtime": 60826.1582,
      "train_tokens_per_second": 62287.564
    },
    {
      "epoch": 0.678932365897901,
      "grad_norm": 12.125,
      "learning_rate": 4.9371407348798575e-06,
      "loss": 0.5293,
      "num_input_tokens_seen": 3788930560,
      "step": 51090,
      "train_runtime": 60829.5039,
      "train_tokens_per_second": 62287.711
    },
    {
      "epoch": 0.6790652553803629,
      "grad_norm": 12.6875,
      "learning_rate": 4.9334296128418325e-06,
      "loss": 0.4933,
      "num_input_tokens_seen": 3789042432,
      "step": 51100,
      "train_runtime": 60831.9171,
      "train_tokens_per_second": 62287.079
    },
    {
      "epoch": 0.6791981448628248,
      "grad_norm": 42.0,
      "learning_rate": 4.929719429340795e-06,
      "loss": 0.8208,
      "num_input_tokens_seen": 3791105536,
      "step": 51110,
      "train_runtime": 60865.2633,
      "train_tokens_per_second": 62286.85
    },
    {
      "epoch": 0.6793310343452867,
      "grad_norm": 15.375,
      "learning_rate": 4.9260101850640255e-06,
      "loss": 0.9084,
      "num_input_tokens_seen": 3791958016,
      "step": 51120,
      "train_runtime": 60878.0172,
      "train_tokens_per_second": 62287.804
    },
    {
      "epoch": 0.6794639238277487,
      "grad_norm": 22.625,
      "learning_rate": 4.922301880698618e-06,
      "loss": 0.757,
      "num_input_tokens_seen": 3792443904,
      "step": 51130,
      "train_runtime": 60885.1386,
      "train_tokens_per_second": 62288.499
    },
    {
      "epoch": 0.6795968133102106,
      "grad_norm": 59.75,
      "learning_rate": 4.9185945169315175e-06,
      "loss": 0.6572,
      "num_input_tokens_seen": 3792688640,
      "step": 51140,
      "train_runtime": 60889.4968,
      "train_tokens_per_second": 62288.06
    },
    {
      "epoch": 0.6797297027926724,
      "grad_norm": 15.625,
      "learning_rate": 4.914888094449469e-06,
      "loss": 0.5581,
      "num_input_tokens_seen": 3792804608,
      "step": 51150,
      "train_runtime": 60891.9146,
      "train_tokens_per_second": 62287.491
    },
    {
      "epoch": 0.6798625922751343,
      "grad_norm": 18.75,
      "learning_rate": 4.911182613939057e-06,
      "loss": 0.8104,
      "num_input_tokens_seen": 3794802176,
      "step": 51160,
      "train_runtime": 60925.0691,
      "train_tokens_per_second": 62286.383
    },
    {
      "epoch": 0.6799954817575963,
      "grad_norm": 26.25,
      "learning_rate": 4.907478076086691e-06,
      "loss": 0.8592,
      "num_input_tokens_seen": 3795633152,
      "step": 51170,
      "train_runtime": 60937.348,
      "train_tokens_per_second": 62287.469
    },
    {
      "epoch": 0.6801283712400582,
      "grad_norm": 35.0,
      "learning_rate": 4.903774481578602e-06,
      "loss": 0.7078,
      "num_input_tokens_seen": 3796073472,
      "step": 51180,
      "train_runtime": 60943.6155,
      "train_tokens_per_second": 62288.288
    },
    {
      "epoch": 0.6802612607225201,
      "grad_norm": 16.875,
      "learning_rate": 4.900071831100848e-06,
      "loss": 0.6669,
      "num_input_tokens_seen": 3796293888,
      "step": 51190,
      "train_runtime": 60946.9368,
      "train_tokens_per_second": 62288.51
    },
    {
      "epoch": 0.6803941502049821,
      "grad_norm": 13.1875,
      "learning_rate": 4.896370125339315e-06,
      "loss": 0.5935,
      "num_input_tokens_seen": 3796408832,
      "step": 51200,
      "train_runtime": 60950.0489,
      "train_tokens_per_second": 62287.216
    },
    {
      "epoch": 0.680527039687444,
      "grad_norm": 30.25,
      "learning_rate": 4.8926693649797054e-06,
      "loss": 0.7844,
      "num_input_tokens_seen": 3798410752,
      "step": 51210,
      "train_runtime": 60981.0622,
      "train_tokens_per_second": 62288.367
    },
    {
      "epoch": 0.6806599291699058,
      "grad_norm": 21.875,
      "learning_rate": 4.888969550707554e-06,
      "loss": 0.9336,
      "num_input_tokens_seen": 3799237632,
      "step": 51220,
      "train_runtime": 60993.5787,
      "train_tokens_per_second": 62289.141
    },
    {
      "epoch": 0.6807928186523677,
      "grad_norm": 32.5,
      "learning_rate": 4.8852706832082194e-06,
      "loss": 0.7212,
      "num_input_tokens_seen": 3799693312,
      "step": 51230,
      "train_runtime": 61000.1288,
      "train_tokens_per_second": 62289.923
    },
    {
      "epoch": 0.6809257081348297,
      "grad_norm": 16.0,
      "learning_rate": 4.881572763166883e-06,
      "loss": 0.6698,
      "num_input_tokens_seen": 3799919616,
      "step": 51240,
      "train_runtime": 61003.6446,
      "train_tokens_per_second": 62290.042
    },
    {
      "epoch": 0.6810585976172916,
      "grad_norm": 17.75,
      "learning_rate": 4.877875791268553e-06,
      "loss": 0.5532,
      "num_input_tokens_seen": 3800035072,
      "step": 51250,
      "train_runtime": 61006.8096,
      "train_tokens_per_second": 62288.703
    },
    {
      "epoch": 0.6811914870997535,
      "grad_norm": 31.5,
      "learning_rate": 4.8741797681980585e-06,
      "loss": 0.8631,
      "num_input_tokens_seen": 3802137856,
      "step": 51260,
      "train_runtime": 61041.1308,
      "train_tokens_per_second": 62288.129
    },
    {
      "epoch": 0.6813243765822155,
      "grad_norm": 17.0,
      "learning_rate": 4.87048469464006e-06,
      "loss": 0.8853,
      "num_input_tokens_seen": 3803002112,
      "step": 51270,
      "train_runtime": 61053.8092,
      "train_tokens_per_second": 62289.35
    },
    {
      "epoch": 0.6814572660646773,
      "grad_norm": 20.0,
      "learning_rate": 4.86679057127903e-06,
      "loss": 0.8152,
      "num_input_tokens_seen": 3803452928,
      "step": 51280,
      "train_runtime": 61060.4212,
      "train_tokens_per_second": 62289.988
    },
    {
      "epoch": 0.6815901555471392,
      "grad_norm": 11.5625,
      "learning_rate": 4.863097398799273e-06,
      "loss": 0.6595,
      "num_input_tokens_seen": 3803684864,
      "step": 51290,
      "train_runtime": 61063.9012,
      "train_tokens_per_second": 62290.237
    },
    {
      "epoch": 0.6817230450296011,
      "grad_norm": 11.9375,
      "learning_rate": 4.859405177884919e-06,
      "loss": 0.4808,
      "num_input_tokens_seen": 3803801344,
      "step": 51300,
      "train_runtime": 61067.1332,
      "train_tokens_per_second": 62288.847
    },
    {
      "epoch": 0.681855934512063,
      "grad_norm": 37.75,
      "learning_rate": 4.855713909219918e-06,
      "loss": 0.805,
      "num_input_tokens_seen": 3805740544,
      "step": 51310,
      "train_runtime": 61098.7045,
      "train_tokens_per_second": 62288.4
    },
    {
      "epoch": 0.681988823994525,
      "grad_norm": 25.25,
      "learning_rate": 4.8520235934880475e-06,
      "loss": 0.9068,
      "num_input_tokens_seen": 3806569216,
      "step": 51320,
      "train_runtime": 61111.3038,
      "train_tokens_per_second": 62289.118
    },
    {
      "epoch": 0.6821217134769869,
      "grad_norm": 32.75,
      "learning_rate": 4.848334231372896e-06,
      "loss": 0.7968,
      "num_input_tokens_seen": 3807041280,
      "step": 51330,
      "train_runtime": 61118.0229,
      "train_tokens_per_second": 62289.994
    },
    {
      "epoch": 0.6822546029594487,
      "grad_norm": 58.25,
      "learning_rate": 4.8446458235579e-06,
      "loss": 0.6152,
      "num_input_tokens_seen": 3807270400,
      "step": 51340,
      "train_runtime": 61121.5242,
      "train_tokens_per_second": 62290.174
    },
    {
      "epoch": 0.6823874924419107,
      "grad_norm": 13.5625,
      "learning_rate": 4.8409583707262895e-06,
      "loss": 0.4519,
      "num_input_tokens_seen": 3807380736,
      "step": 51350,
      "train_runtime": 61123.9324,
      "train_tokens_per_second": 62289.525
    },
    {
      "epoch": 0.6825203819243726,
      "grad_norm": 18.0,
      "learning_rate": 4.83727187356114e-06,
      "loss": 0.8702,
      "num_input_tokens_seen": 3809490944,
      "step": 51360,
      "train_runtime": 61158.6585,
      "train_tokens_per_second": 62288.661
    },
    {
      "epoch": 0.6826532714068345,
      "grad_norm": 23.5,
      "learning_rate": 4.833586332745345e-06,
      "loss": 0.9557,
      "num_input_tokens_seen": 3810376192,
      "step": 51370,
      "train_runtime": 61171.5946,
      "train_tokens_per_second": 62289.96
    },
    {
      "epoch": 0.6827861608892964,
      "grad_norm": 17.625,
      "learning_rate": 4.829901748961605e-06,
      "loss": 0.7686,
      "num_input_tokens_seen": 3810858496,
      "step": 51380,
      "train_runtime": 61178.4013,
      "train_tokens_per_second": 62290.913
    },
    {
      "epoch": 0.6829190503717584,
      "grad_norm": 22.625,
      "learning_rate": 4.826218122892473e-06,
      "loss": 0.6333,
      "num_input_tokens_seen": 3811094528,
      "step": 51390,
      "train_runtime": 61182.0189,
      "train_tokens_per_second": 62291.088
    },
    {
      "epoch": 0.6830519398542202,
      "grad_norm": 24.125,
      "learning_rate": 4.822535455220293e-06,
      "loss": 0.5481,
      "num_input_tokens_seen": 3811207680,
      "step": 51400,
      "train_runtime": 61184.458,
      "train_tokens_per_second": 62290.454
    },
    {
      "epoch": 0.6831848293366821,
      "grad_norm": 19.875,
      "learning_rate": 4.81885374662726e-06,
      "loss": 0.8618,
      "num_input_tokens_seen": 3813229312,
      "step": 51410,
      "train_runtime": 61217.7251,
      "train_tokens_per_second": 62289.628
    },
    {
      "epoch": 0.683317718819144,
      "grad_norm": 19.25,
      "learning_rate": 4.8151729977953684e-06,
      "loss": 0.9338,
      "num_input_tokens_seen": 3814076928,
      "step": 51420,
      "train_runtime": 61231.2284,
      "train_tokens_per_second": 62289.734
    },
    {
      "epoch": 0.683450608301606,
      "grad_norm": 14.3125,
      "learning_rate": 4.811493209406446e-06,
      "loss": 0.7668,
      "num_input_tokens_seen": 3814576640,
      "step": 51430,
      "train_runtime": 61238.1684,
      "train_tokens_per_second": 62290.835
    },
    {
      "epoch": 0.6835834977840679,
      "grad_norm": 14.375,
      "learning_rate": 4.8078143821421455e-06,
      "loss": 0.6951,
      "num_input_tokens_seen": 3814821120,
      "step": 51440,
      "train_runtime": 61241.8096,
      "train_tokens_per_second": 62291.123
    },
    {
      "epoch": 0.6837163872665298,
      "grad_norm": 12.625,
      "learning_rate": 4.804136516683925e-06,
      "loss": 0.5901,
      "num_input_tokens_seen": 3814941696,
      "step": 51450,
      "train_runtime": 61244.2976,
      "train_tokens_per_second": 62290.562
    },
    {
      "epoch": 0.6838492767489917,
      "grad_norm": 22.875,
      "learning_rate": 4.800459613713091e-06,
      "loss": 0.8905,
      "num_input_tokens_seen": 3817037056,
      "step": 51460,
      "train_runtime": 61277.9267,
      "train_tokens_per_second": 62290.571
    },
    {
      "epoch": 0.6839821662314536,
      "grad_norm": 21.125,
      "learning_rate": 4.7967836739107475e-06,
      "loss": 0.8778,
      "num_input_tokens_seen": 3817889024,
      "step": 51470,
      "train_runtime": 61291.2255,
      "train_tokens_per_second": 62290.956
    },
    {
      "epoch": 0.6841150557139155,
      "grad_norm": 15.125,
      "learning_rate": 4.793108697957832e-06,
      "loss": 0.69,
      "num_input_tokens_seen": 3818380800,
      "step": 51480,
      "train_runtime": 61297.8429,
      "train_tokens_per_second": 62292.254
    },
    {
      "epoch": 0.6842479451963774,
      "grad_norm": 13.5625,
      "learning_rate": 4.7894346865351004e-06,
      "loss": 0.599,
      "num_input_tokens_seen": 3818611200,
      "step": 51490,
      "train_runtime": 61301.2966,
      "train_tokens_per_second": 62292.503
    },
    {
      "epoch": 0.6843808346788394,
      "grad_norm": 15.0625,
      "learning_rate": 4.785761640323132e-06,
      "loss": 0.5257,
      "num_input_tokens_seen": 3818728960,
      "step": 51500,
      "train_runtime": 61303.8485,
      "train_tokens_per_second": 62291.831
    },
    {
      "epoch": 0.6845137241613013,
      "grad_norm": 22.875,
      "learning_rate": 4.782089560002326e-06,
      "loss": 0.8514,
      "num_input_tokens_seen": 3820741376,
      "step": 51510,
      "train_runtime": 61337.31,
      "train_tokens_per_second": 62290.658
    },
    {
      "epoch": 0.6846466136437631,
      "grad_norm": 36.5,
      "learning_rate": 4.778418446252905e-06,
      "loss": 0.9132,
      "num_input_tokens_seen": 3821558528,
      "step": 51520,
      "train_runtime": 61349.7118,
      "train_tokens_per_second": 62291.385
    },
    {
      "epoch": 0.684779503126225,
      "grad_norm": 15.0,
      "learning_rate": 4.774748299754902e-06,
      "loss": 0.7378,
      "num_input_tokens_seen": 3822048768,
      "step": 51530,
      "train_runtime": 61356.4204,
      "train_tokens_per_second": 62292.564
    },
    {
      "epoch": 0.684912392608687,
      "grad_norm": 34.25,
      "learning_rate": 4.771079121188184e-06,
      "loss": 0.6269,
      "num_input_tokens_seen": 3822282496,
      "step": 51540,
      "train_runtime": 61359.9128,
      "train_tokens_per_second": 62292.828
    },
    {
      "epoch": 0.6850452820911489,
      "grad_norm": 10.5,
      "learning_rate": 4.767410911232435e-06,
      "loss": 0.4895,
      "num_input_tokens_seen": 3822398464,
      "step": 51550,
      "train_runtime": 61362.4012,
      "train_tokens_per_second": 62292.192
    },
    {
      "epoch": 0.6851781715736108,
      "grad_norm": 38.0,
      "learning_rate": 4.763743670567155e-06,
      "loss": 0.7688,
      "num_input_tokens_seen": 3824440832,
      "step": 51560,
      "train_runtime": 61394.4996,
      "train_tokens_per_second": 62292.89
    },
    {
      "epoch": 0.6853110610560728,
      "grad_norm": 18.75,
      "learning_rate": 4.76007739987167e-06,
      "loss": 0.8694,
      "num_input_tokens_seen": 3825246976,
      "step": 51570,
      "train_runtime": 61406.5535,
      "train_tokens_per_second": 62293.79
    },
    {
      "epoch": 0.6854439505385346,
      "grad_norm": 19.125,
      "learning_rate": 4.756412099825123e-06,
      "loss": 0.7709,
      "num_input_tokens_seen": 3825677824,
      "step": 51580,
      "train_runtime": 61412.8198,
      "train_tokens_per_second": 62294.45
    },
    {
      "epoch": 0.6855768400209965,
      "grad_norm": 18.875,
      "learning_rate": 4.752747771106481e-06,
      "loss": 0.621,
      "num_input_tokens_seen": 3825894144,
      "step": 51590,
      "train_runtime": 61416.9171,
      "train_tokens_per_second": 62293.816
    },
    {
      "epoch": 0.6857097295034584,
      "grad_norm": 13.375,
      "learning_rate": 4.749084414394524e-06,
      "loss": 0.5151,
      "num_input_tokens_seen": 3826004224,
      "step": 51600,
      "train_runtime": 61419.3722,
      "train_tokens_per_second": 62293.118
    },
    {
      "epoch": 0.6858426189859204,
      "grad_norm": 24.25,
      "learning_rate": 4.745422030367855e-06,
      "loss": 0.8508,
      "num_input_tokens_seen": 3828114944,
      "step": 51610,
      "train_runtime": 61453.4014,
      "train_tokens_per_second": 62292.971
    },
    {
      "epoch": 0.6859755084683823,
      "grad_norm": 26.0,
      "learning_rate": 4.741760619704903e-06,
      "loss": 0.8788,
      "num_input_tokens_seen": 3828997376,
      "step": 51620,
      "train_runtime": 61466.4278,
      "train_tokens_per_second": 62294.126
    },
    {
      "epoch": 0.6861083979508442,
      "grad_norm": 20.875,
      "learning_rate": 4.738100183083908e-06,
      "loss": 0.7687,
      "num_input_tokens_seen": 3829485824,
      "step": 51630,
      "train_runtime": 61473.3866,
      "train_tokens_per_second": 62295.02
    },
    {
      "epoch": 0.6862412874333061,
      "grad_norm": 25.75,
      "learning_rate": 4.734440721182938e-06,
      "loss": 0.6608,
      "num_input_tokens_seen": 3829724672,
      "step": 51640,
      "train_runtime": 61476.917,
      "train_tokens_per_second": 62295.327
    },
    {
      "epoch": 0.686374176915768,
      "grad_norm": 7.125,
      "learning_rate": 4.730782234679866e-06,
      "loss": 0.564,
      "num_input_tokens_seen": 3829840640,
      "step": 51650,
      "train_runtime": 61480.0798,
      "train_tokens_per_second": 62294.009
    },
    {
      "epoch": 0.6865070663982299,
      "grad_norm": 26.0,
      "learning_rate": 4.727124724252408e-06,
      "loss": 0.7399,
      "num_input_tokens_seen": 3831848448,
      "step": 51660,
      "train_runtime": 61513.4039,
      "train_tokens_per_second": 62292.902
    },
    {
      "epoch": 0.6866399558806918,
      "grad_norm": 12.5625,
      "learning_rate": 4.723468190578072e-06,
      "loss": 0.9374,
      "num_input_tokens_seen": 3832688896,
      "step": 51670,
      "train_runtime": 61525.9783,
      "train_tokens_per_second": 62293.831
    },
    {
      "epoch": 0.6867728453631537,
      "grad_norm": 30.875,
      "learning_rate": 4.7198126343342045e-06,
      "loss": 0.7072,
      "num_input_tokens_seen": 3833156352,
      "step": 51680,
      "train_runtime": 61532.5564,
      "train_tokens_per_second": 62294.768
    },
    {
      "epoch": 0.6869057348456157,
      "grad_norm": 12.125,
      "learning_rate": 4.716158056197968e-06,
      "loss": 0.6341,
      "num_input_tokens_seen": 3833381888,
      "step": 51690,
      "train_runtime": 61535.9512,
      "train_tokens_per_second": 62294.997
    },
    {
      "epoch": 0.6870386243280776,
      "grad_norm": 23.0,
      "learning_rate": 4.7125044568463285e-06,
      "loss": 0.4949,
      "num_input_tokens_seen": 3833498624,
      "step": 51700,
      "train_runtime": 61539.0267,
      "train_tokens_per_second": 62293.781
    },
    {
      "epoch": 0.6871715138105394,
      "grad_norm": 42.5,
      "learning_rate": 4.708851836956099e-06,
      "loss": 0.8723,
      "num_input_tokens_seen": 3835613952,
      "step": 51710,
      "train_runtime": 61572.7217,
      "train_tokens_per_second": 62294.046
    },
    {
      "epoch": 0.6873044032930014,
      "grad_norm": 16.0,
      "learning_rate": 4.705200197203882e-06,
      "loss": 0.9399,
      "num_input_tokens_seen": 3836452096,
      "step": 51720,
      "train_runtime": 61585.2075,
      "train_tokens_per_second": 62295.026
    },
    {
      "epoch": 0.6874372927754633,
      "grad_norm": 14.6875,
      "learning_rate": 4.7015495382661154e-06,
      "loss": 0.7276,
      "num_input_tokens_seen": 3836934144,
      "step": 51730,
      "train_runtime": 61591.7273,
      "train_tokens_per_second": 62296.258
    },
    {
      "epoch": 0.6875701822579252,
      "grad_norm": 17.0,
      "learning_rate": 4.6978998608190515e-06,
      "loss": 0.6005,
      "num_input_tokens_seen": 3837159168,
      "step": 51740,
      "train_runtime": 61595.1985,
      "train_tokens_per_second": 62296.401
    },
    {
      "epoch": 0.6877030717403871,
      "grad_norm": 11.5,
      "learning_rate": 4.694251165538761e-06,
      "loss": 0.514,
      "num_input_tokens_seen": 3837273600,
      "step": 51750,
      "train_runtime": 61597.5755,
      "train_tokens_per_second": 62295.855
    },
    {
      "epoch": 0.6878359612228491,
      "grad_norm": 18.375,
      "learning_rate": 4.690603453101134e-06,
      "loss": 0.8275,
      "num_input_tokens_seen": 3839355136,
      "step": 51760,
      "train_runtime": 61630.9098,
      "train_tokens_per_second": 62295.935
    },
    {
      "epoch": 0.6879688507053109,
      "grad_norm": 17.625,
      "learning_rate": 4.686956724181869e-06,
      "loss": 0.8813,
      "num_input_tokens_seen": 3840186112,
      "step": 51770,
      "train_runtime": 61644.1425,
      "train_tokens_per_second": 62296.042
    },
    {
      "epoch": 0.6881017401877728,
      "grad_norm": 34.0,
      "learning_rate": 4.6833109794565014e-06,
      "loss": 0.7266,
      "num_input_tokens_seen": 3840666112,
      "step": 51780,
      "train_runtime": 61651.1196,
      "train_tokens_per_second": 62296.778
    },
    {
      "epoch": 0.6882346296702347,
      "grad_norm": 15.375,
      "learning_rate": 4.679666219600363e-06,
      "loss": 0.5955,
      "num_input_tokens_seen": 3840898304,
      "step": 51790,
      "train_runtime": 61654.6429,
      "train_tokens_per_second": 62296.984
    },
    {
      "epoch": 0.6883675191526967,
      "grad_norm": 12.5625,
      "learning_rate": 4.676022445288617e-06,
      "loss": 0.5733,
      "num_input_tokens_seen": 3841012224,
      "step": 51800,
      "train_runtime": 61657.0766,
      "train_tokens_per_second": 62296.373
    },
    {
      "epoch": 0.6885004086351586,
      "grad_norm": 24.875,
      "learning_rate": 4.672379657196241e-06,
      "loss": 0.8553,
      "num_input_tokens_seen": 3842983680,
      "step": 51810,
      "train_runtime": 61689.8826,
      "train_tokens_per_second": 62295.202
    },
    {
      "epoch": 0.6886332981176205,
      "grad_norm": 11.625,
      "learning_rate": 4.668737855998026e-06,
      "loss": 0.9266,
      "num_input_tokens_seen": 3843798016,
      "step": 51820,
      "train_runtime": 61702.9903,
      "train_tokens_per_second": 62295.166
    },
    {
      "epoch": 0.6887661876000823,
      "grad_norm": 14.125,
      "learning_rate": 4.665097042368587e-06,
      "loss": 0.7161,
      "num_input_tokens_seen": 3844241152,
      "step": 51830,
      "train_runtime": 61709.3628,
      "train_tokens_per_second": 62295.914
    },
    {
      "epoch": 0.6888990770825443,
      "grad_norm": 21.375,
      "learning_rate": 4.661457216982353e-06,
      "loss": 0.6226,
      "num_input_tokens_seen": 3844461056,
      "step": 51840,
      "train_runtime": 61712.7756,
      "train_tokens_per_second": 62296.032
    },
    {
      "epoch": 0.6890319665650062,
      "grad_norm": 8.8125,
      "learning_rate": 4.6578183805135636e-06,
      "loss": 0.513,
      "num_input_tokens_seen": 3844575232,
      "step": 51850,
      "train_runtime": 61715.1989,
      "train_tokens_per_second": 62295.436
    },
    {
      "epoch": 0.6891648560474681,
      "grad_norm": 18.625,
      "learning_rate": 4.654180533636284e-06,
      "loss": 0.8029,
      "num_input_tokens_seen": 3846629888,
      "step": 51860,
      "train_runtime": 61749.579,
      "train_tokens_per_second": 62294.026
    },
    {
      "epoch": 0.6892977455299301,
      "grad_norm": 27.0,
      "learning_rate": 4.6505436770243926e-06,
      "loss": 0.9404,
      "num_input_tokens_seen": 3847521280,
      "step": 51870,
      "train_runtime": 61762.4108,
      "train_tokens_per_second": 62295.516
    },
    {
      "epoch": 0.689430635012392,
      "grad_norm": 25.0,
      "learning_rate": 4.646907811351584e-06,
      "loss": 0.7195,
      "num_input_tokens_seen": 3847988992,
      "step": 51880,
      "train_runtime": 61769.1003,
      "train_tokens_per_second": 62296.342
    },
    {
      "epoch": 0.6895635244948538,
      "grad_norm": 30.0,
      "learning_rate": 4.643272937291376e-06,
      "loss": 0.5473,
      "num_input_tokens_seen": 3848219136,
      "step": 51890,
      "train_runtime": 61772.573,
      "train_tokens_per_second": 62296.565
    },
    {
      "epoch": 0.6896964139773157,
      "grad_norm": 49.5,
      "learning_rate": 4.639639055517083e-06,
      "loss": 0.5704,
      "num_input_tokens_seen": 3848332800,
      "step": 51900,
      "train_runtime": 61775.0278,
      "train_tokens_per_second": 62295.93
    },
    {
      "epoch": 0.6898293034597777,
      "grad_norm": 21.75,
      "learning_rate": 4.636006166701864e-06,
      "loss": 0.8485,
      "num_input_tokens_seen": 3850321664,
      "step": 51910,
      "train_runtime": 61807.0763,
      "train_tokens_per_second": 62295.806
    },
    {
      "epoch": 0.6899621929422396,
      "grad_norm": 13.6875,
      "learning_rate": 4.632374271518671e-06,
      "loss": 0.8916,
      "num_input_tokens_seen": 3851149824,
      "step": 51920,
      "train_runtime": 61819.774,
      "train_tokens_per_second": 62296.407
    },
    {
      "epoch": 0.6900950824247015,
      "grad_norm": 14.3125,
      "learning_rate": 4.62874337064028e-06,
      "loss": 0.7959,
      "num_input_tokens_seen": 3851597824,
      "step": 51930,
      "train_runtime": 61826.7796,
      "train_tokens_per_second": 62296.595
    },
    {
      "epoch": 0.6902279719071635,
      "grad_norm": 8.25,
      "learning_rate": 4.625113464739286e-06,
      "loss": 0.6817,
      "num_input_tokens_seen": 3851815424,
      "step": 51940,
      "train_runtime": 61830.1405,
      "train_tokens_per_second": 62296.728
    },
    {
      "epoch": 0.6903608613896253,
      "grad_norm": 20.5,
      "learning_rate": 4.621484554488096e-06,
      "loss": 0.5756,
      "num_input_tokens_seen": 3851930112,
      "step": 51950,
      "train_runtime": 61832.5275,
      "train_tokens_per_second": 62296.178
    },
    {
      "epoch": 0.6904937508720872,
      "grad_norm": 17.75,
      "learning_rate": 4.6178566405589356e-06,
      "loss": 0.8358,
      "num_input_tokens_seen": 3853959424,
      "step": 51960,
      "train_runtime": 61866.5812,
      "train_tokens_per_second": 62294.689
    },
    {
      "epoch": 0.6906266403545491,
      "grad_norm": 34.25,
      "learning_rate": 4.614229723623834e-06,
      "loss": 0.9316,
      "num_input_tokens_seen": 3854812160,
      "step": 51970,
      "train_runtime": 61879.4479,
      "train_tokens_per_second": 62295.516
    },
    {
      "epoch": 0.690759529837011,
      "grad_norm": 16.0,
      "learning_rate": 4.61060380435466e-06,
      "loss": 0.7036,
      "num_input_tokens_seen": 3855298304,
      "step": 51980,
      "train_runtime": 61886.8916,
      "train_tokens_per_second": 62295.879
    },
    {
      "epoch": 0.690892419319473,
      "grad_norm": 48.5,
      "learning_rate": 4.60697888342307e-06,
      "loss": 0.6517,
      "num_input_tokens_seen": 3855529984,
      "step": 51990,
      "train_runtime": 61890.3501,
      "train_tokens_per_second": 62296.141
    },
    {
      "epoch": 0.6910253088019349,
      "grad_norm": 9.875,
      "learning_rate": 4.603354961500554e-06,
      "loss": 0.5494,
      "num_input_tokens_seen": 3855639808,
      "step": 52000,
      "train_runtime": 61892.7527,
      "train_tokens_per_second": 62295.497
    },
    {
      "epoch": 0.6911581982843967,
      "grad_norm": 32.25,
      "learning_rate": 4.5997320392584146e-06,
      "loss": 0.808,
      "num_input_tokens_seen": 3857606144,
      "step": 52010,
      "train_runtime": 61925.519,
      "train_tokens_per_second": 62294.288
    },
    {
      "epoch": 0.6912910877668587,
      "grad_norm": 22.5,
      "learning_rate": 4.596110117367754e-06,
      "loss": 0.8438,
      "num_input_tokens_seen": 3858441984,
      "step": 52020,
      "train_runtime": 61938.6366,
      "train_tokens_per_second": 62294.59
    },
    {
      "epoch": 0.6914239772493206,
      "grad_norm": 13.75,
      "learning_rate": 4.592489196499515e-06,
      "loss": 0.653,
      "num_input_tokens_seen": 3858908416,
      "step": 52030,
      "train_runtime": 61945.3,
      "train_tokens_per_second": 62295.419
    },
    {
      "epoch": 0.6915568667317825,
      "grad_norm": 60.75,
      "learning_rate": 4.58886927732443e-06,
      "loss": 0.7333,
      "num_input_tokens_seen": 3859145216,
      "step": 52040,
      "train_runtime": 61948.7888,
      "train_tokens_per_second": 62295.733
    },
    {
      "epoch": 0.6916897562142444,
      "grad_norm": 12.3125,
      "learning_rate": 4.585250360513063e-06,
      "loss": 0.6289,
      "num_input_tokens_seen": 3859262464,
      "step": 52050,
      "train_runtime": 61951.2641,
      "train_tokens_per_second": 62295.137
    },
    {
      "epoch": 0.6918226456967064,
      "grad_norm": 21.125,
      "learning_rate": 4.581632446735784e-06,
      "loss": 0.8354,
      "num_input_tokens_seen": 3861319424,
      "step": 52060,
      "train_runtime": 61985.0889,
      "train_tokens_per_second": 62294.327
    },
    {
      "epoch": 0.6919555351791682,
      "grad_norm": 16.5,
      "learning_rate": 4.57801553666278e-06,
      "loss": 0.9138,
      "num_input_tokens_seen": 3862186240,
      "step": 52070,
      "train_runtime": 61997.9123,
      "train_tokens_per_second": 62295.424
    },
    {
      "epoch": 0.6920884246616301,
      "grad_norm": 17.0,
      "learning_rate": 4.5743996309640534e-06,
      "loss": 0.6865,
      "num_input_tokens_seen": 3862677248,
      "step": 52080,
      "train_runtime": 62004.749,
      "train_tokens_per_second": 62296.474
    },
    {
      "epoch": 0.692221314144092,
      "grad_norm": 37.5,
      "learning_rate": 4.570784730309409e-06,
      "loss": 0.6574,
      "num_input_tokens_seen": 3862907392,
      "step": 52090,
      "train_runtime": 62008.2353,
      "train_tokens_per_second": 62296.683
    },
    {
      "epoch": 0.692354203626554,
      "grad_norm": 12.0,
      "learning_rate": 4.567170835368489e-06,
      "loss": 0.5591,
      "num_input_tokens_seen": 3863021568,
      "step": 52100,
      "train_runtime": 62011.4784,
      "train_tokens_per_second": 62295.266
    },
    {
      "epoch": 0.6924870931090159,
      "grad_norm": 26.0,
      "learning_rate": 4.563557946810725e-06,
      "loss": 0.8206,
      "num_input_tokens_seen": 3865057024,
      "step": 52110,
      "train_runtime": 62043.6996,
      "train_tokens_per_second": 62295.721
    },
    {
      "epoch": 0.6926199825914778,
      "grad_norm": 27.375,
      "learning_rate": 4.559946065305375e-06,
      "loss": 0.8223,
      "num_input_tokens_seen": 3865908992,
      "step": 52120,
      "train_runtime": 62056.3182,
      "train_tokens_per_second": 62296.783
    },
    {
      "epoch": 0.6927528720739398,
      "grad_norm": 28.125,
      "learning_rate": 4.556335191521509e-06,
      "loss": 0.7464,
      "num_input_tokens_seen": 3866396672,
      "step": 52130,
      "train_runtime": 62063.1721,
      "train_tokens_per_second": 62297.761
    },
    {
      "epoch": 0.6928857615564016,
      "grad_norm": 27.0,
      "learning_rate": 4.552725326128008e-06,
      "loss": 0.5671,
      "num_input_tokens_seen": 3866629632,
      "step": 52140,
      "train_runtime": 62066.7055,
      "train_tokens_per_second": 62297.968
    },
    {
      "epoch": 0.6930186510388635,
      "grad_norm": 9.4375,
      "learning_rate": 4.549116469793569e-06,
      "loss": 0.4813,
      "num_input_tokens_seen": 3866746112,
      "step": 52150,
      "train_runtime": 62069.1322,
      "train_tokens_per_second": 62297.409
    },
    {
      "epoch": 0.6931515405213254,
      "grad_norm": 31.375,
      "learning_rate": 4.545508623186702e-06,
      "loss": 0.7962,
      "num_input_tokens_seen": 3868703488,
      "step": 52160,
      "train_runtime": 62101.4761,
      "train_tokens_per_second": 62296.482
    },
    {
      "epoch": 0.6932844300037874,
      "grad_norm": 11.6875,
      "learning_rate": 4.541901786975722e-06,
      "loss": 0.8982,
      "num_input_tokens_seen": 3869524480,
      "step": 52170,
      "train_runtime": 62113.9052,
      "train_tokens_per_second": 62297.234
    },
    {
      "epoch": 0.6934173194862493,
      "grad_norm": 23.375,
      "learning_rate": 4.538295961828767e-06,
      "loss": 0.7339,
      "num_input_tokens_seen": 3869983232,
      "step": 52180,
      "train_runtime": 62120.4552,
      "train_tokens_per_second": 62298.05
    },
    {
      "epoch": 0.6935502089687112,
      "grad_norm": 12.9375,
      "learning_rate": 4.534691148413784e-06,
      "loss": 0.5977,
      "num_input_tokens_seen": 3870204928,
      "step": 52190,
      "train_runtime": 62123.8994,
      "train_tokens_per_second": 62298.165
    },
    {
      "epoch": 0.693683098451173,
      "grad_norm": 13.125,
      "learning_rate": 4.531087347398532e-06,
      "loss": 0.5345,
      "num_input_tokens_seen": 3870320128,
      "step": 52200,
      "train_runtime": 62126.3457,
      "train_tokens_per_second": 62297.566
    },
    {
      "epoch": 0.693815987933635,
      "grad_norm": 63.75,
      "learning_rate": 4.5274845594505875e-06,
      "loss": 0.7364,
      "num_input_tokens_seen": 3872388608,
      "step": 52210,
      "train_runtime": 62160.3498,
      "train_tokens_per_second": 62296.763
    },
    {
      "epoch": 0.6939488774160969,
      "grad_norm": 86.5,
      "learning_rate": 4.523882785237325e-06,
      "loss": 0.8994,
      "num_input_tokens_seen": 3873210112,
      "step": 52220,
      "train_runtime": 62172.7645,
      "train_tokens_per_second": 62297.537
    },
    {
      "epoch": 0.6940817668985588,
      "grad_norm": 36.0,
      "learning_rate": 4.520282025425953e-06,
      "loss": 0.7566,
      "num_input_tokens_seen": 3873690368,
      "step": 52230,
      "train_runtime": 62179.4481,
      "train_tokens_per_second": 62298.565
    },
    {
      "epoch": 0.6942146563810208,
      "grad_norm": 27.0,
      "learning_rate": 4.51668228068347e-06,
      "loss": 0.6696,
      "num_input_tokens_seen": 3873926912,
      "step": 52240,
      "train_runtime": 62182.8971,
      "train_tokens_per_second": 62298.913
    },
    {
      "epoch": 0.6943475458634827,
      "grad_norm": 62.75,
      "learning_rate": 4.5130835516767e-06,
      "loss": 0.4813,
      "num_input_tokens_seen": 3874041856,
      "step": 52250,
      "train_runtime": 62185.2543,
      "train_tokens_per_second": 62298.4
    },
    {
      "epoch": 0.6944804353459445,
      "grad_norm": 19.375,
      "learning_rate": 4.509485839072282e-06,
      "loss": 0.7996,
      "num_input_tokens_seen": 3876103424,
      "step": 52260,
      "train_runtime": 62219.1326,
      "train_tokens_per_second": 62297.613
    },
    {
      "epoch": 0.6946133248284064,
      "grad_norm": 17.625,
      "learning_rate": 4.505889143536645e-06,
      "loss": 0.8724,
      "num_input_tokens_seen": 3876985088,
      "step": 52270,
      "train_runtime": 62232.7964,
      "train_tokens_per_second": 62298.102
    },
    {
      "epoch": 0.6947462143108684,
      "grad_norm": 14.75,
      "learning_rate": 4.502293465736063e-06,
      "loss": 0.7355,
      "num_input_tokens_seen": 3877440512,
      "step": 52280,
      "train_runtime": 62239.5523,
      "train_tokens_per_second": 62298.657
    },
    {
      "epoch": 0.6948791037933303,
      "grad_norm": 51.25,
      "learning_rate": 4.498698806336585e-06,
      "loss": 0.688,
      "num_input_tokens_seen": 3877674496,
      "step": 52290,
      "train_runtime": 62243.0723,
      "train_tokens_per_second": 62298.893
    },
    {
      "epoch": 0.6950119932757922,
      "grad_norm": 15.4375,
      "learning_rate": 4.495105166004108e-06,
      "loss": 0.5143,
      "num_input_tokens_seen": 3877790976,
      "step": 52300,
      "train_runtime": 62245.4585,
      "train_tokens_per_second": 62298.376
    },
    {
      "epoch": 0.6951448827582541,
      "grad_norm": 36.25,
      "learning_rate": 4.4915125454043085e-06,
      "loss": 0.8824,
      "num_input_tokens_seen": 3879754496,
      "step": 52310,
      "train_runtime": 62276.5199,
      "train_tokens_per_second": 62298.833
    },
    {
      "epoch": 0.695277772240716,
      "grad_norm": 18.75,
      "learning_rate": 4.4879209452026915e-06,
      "loss": 0.9084,
      "num_input_tokens_seen": 3880596224,
      "step": 52320,
      "train_runtime": 62289.6225,
      "train_tokens_per_second": 62299.241
    },
    {
      "epoch": 0.6954106617231779,
      "grad_norm": 24.625,
      "learning_rate": 4.484330366064573e-06,
      "loss": 0.7384,
      "num_input_tokens_seen": 3881051392,
      "step": 52330,
      "train_runtime": 62296.1233,
      "train_tokens_per_second": 62300.047
    },
    {
      "epoch": 0.6955435512056398,
      "grad_norm": 17.625,
      "learning_rate": 4.480740808655064e-06,
      "loss": 0.7006,
      "num_input_tokens_seen": 3881270272,
      "step": 52340,
      "train_runtime": 62299.4645,
      "train_tokens_per_second": 62300.219
    },
    {
      "epoch": 0.6956764406881017,
      "grad_norm": 12.625,
      "learning_rate": 4.477152273639115e-06,
      "loss": 0.552,
      "num_input_tokens_seen": 3881381888,
      "step": 52350,
      "train_runtime": 62301.8712,
      "train_tokens_per_second": 62299.604
    },
    {
      "epoch": 0.6958093301705637,
      "grad_norm": 27.75,
      "learning_rate": 4.473564761681456e-06,
      "loss": 0.8802,
      "num_input_tokens_seen": 3883460352,
      "step": 52360,
      "train_runtime": 62334.5854,
      "train_tokens_per_second": 62300.252
    },
    {
      "epoch": 0.6959422196530256,
      "grad_norm": 85.0,
      "learning_rate": 4.469978273446645e-06,
      "loss": 0.9027,
      "num_input_tokens_seen": 3884327680,
      "step": 52370,
      "train_runtime": 62347.3698,
      "train_tokens_per_second": 62301.388
    },
    {
      "epoch": 0.6960751091354874,
      "grad_norm": 24.625,
      "learning_rate": 4.466392809599052e-06,
      "loss": 0.7605,
      "num_input_tokens_seen": 3884825856,
      "step": 52380,
      "train_runtime": 62354.7917,
      "train_tokens_per_second": 62301.962
    },
    {
      "epoch": 0.6962079986179494,
      "grad_norm": 16.125,
      "learning_rate": 4.462808370802846e-06,
      "loss": 0.715,
      "num_input_tokens_seen": 3885054976,
      "step": 52390,
      "train_runtime": 62358.2717,
      "train_tokens_per_second": 62302.159
    },
    {
      "epoch": 0.6963408881004113,
      "grad_norm": 11.0,
      "learning_rate": 4.45922495772202e-06,
      "loss": 0.4938,
      "num_input_tokens_seen": 3885174272,
      "step": 52400,
      "train_runtime": 62360.6856,
      "train_tokens_per_second": 62301.661
    },
    {
      "epoch": 0.6964737775828732,
      "grad_norm": 21.25,
      "learning_rate": 4.45564257102036e-06,
      "loss": 0.8109,
      "num_input_tokens_seen": 3887140864,
      "step": 52410,
      "train_runtime": 62392.7238,
      "train_tokens_per_second": 62301.189
    },
    {
      "epoch": 0.6966066670653351,
      "grad_norm": 26.875,
      "learning_rate": 4.452061211361476e-06,
      "loss": 0.9022,
      "num_input_tokens_seen": 3887984896,
      "step": 52420,
      "train_runtime": 62405.4584,
      "train_tokens_per_second": 62302.0
    },
    {
      "epoch": 0.6967395565477971,
      "grad_norm": 22.375,
      "learning_rate": 4.448480879408783e-06,
      "loss": 0.6935,
      "num_input_tokens_seen": 3888441088,
      "step": 52430,
      "train_runtime": 62412.1695,
      "train_tokens_per_second": 62302.611
    },
    {
      "epoch": 0.6968724460302589,
      "grad_norm": 16.375,
      "learning_rate": 4.444901575825505e-06,
      "loss": 0.6358,
      "num_input_tokens_seen": 3888668416,
      "step": 52440,
      "train_runtime": 62416.3963,
      "train_tokens_per_second": 62302.033
    },
    {
      "epoch": 0.6970053355127208,
      "grad_norm": 14.0,
      "learning_rate": 4.441323301274675e-06,
      "loss": 0.4679,
      "num_input_tokens_seen": 3888783872,
      "step": 52450,
      "train_runtime": 62418.8796,
      "train_tokens_per_second": 62301.405
    },
    {
      "epoch": 0.6971382249951827,
      "grad_norm": 130.0,
      "learning_rate": 4.437746056419139e-06,
      "loss": 0.8142,
      "num_input_tokens_seen": 3890895360,
      "step": 52460,
      "train_runtime": 62451.8646,
      "train_tokens_per_second": 62302.309
    },
    {
      "epoch": 0.6972711144776447,
      "grad_norm": 36.5,
      "learning_rate": 4.434169841921547e-06,
      "loss": 0.8862,
      "num_input_tokens_seen": 3891742720,
      "step": 52470,
      "train_runtime": 62464.8381,
      "train_tokens_per_second": 62302.935
    },
    {
      "epoch": 0.6974040039601066,
      "grad_norm": 11.875,
      "learning_rate": 4.430594658444367e-06,
      "loss": 0.7388,
      "num_input_tokens_seen": 3892234240,
      "step": 52480,
      "train_runtime": 62472.7127,
      "train_tokens_per_second": 62302.949
    },
    {
      "epoch": 0.6975368934425685,
      "grad_norm": 36.75,
      "learning_rate": 4.42702050664986e-06,
      "loss": 0.64,
      "num_input_tokens_seen": 3892464640,
      "step": 52490,
      "train_runtime": 62476.2387,
      "train_tokens_per_second": 62303.121
    },
    {
      "epoch": 0.6976697829250303,
      "grad_norm": 8.6875,
      "learning_rate": 4.423447387200112e-06,
      "loss": 0.4728,
      "num_input_tokens_seen": 3892578304,
      "step": 52500,
      "train_runtime": 62478.6815,
      "train_tokens_per_second": 62302.504
    },
    {
      "epoch": 0.6978026724074923,
      "grad_norm": 14.3125,
      "learning_rate": 4.4198753007570116e-06,
      "loss": 0.8243,
      "num_input_tokens_seen": 3894597120,
      "step": 52510,
      "train_runtime": 62511.4073,
      "train_tokens_per_second": 62302.183
    },
    {
      "epoch": 0.6979355618899542,
      "grad_norm": 21.875,
      "learning_rate": 4.416304247982253e-06,
      "loss": 0.931,
      "num_input_tokens_seen": 3895433216,
      "step": 52520,
      "train_runtime": 62523.9604,
      "train_tokens_per_second": 62303.047
    },
    {
      "epoch": 0.6980684513724161,
      "grad_norm": 71.0,
      "learning_rate": 4.412734229537348e-06,
      "loss": 0.7173,
      "num_input_tokens_seen": 3895880448,
      "step": 52530,
      "train_runtime": 62530.5152,
      "train_tokens_per_second": 62303.668
    },
    {
      "epoch": 0.6982013408548781,
      "grad_norm": 53.25,
      "learning_rate": 4.4091652460835975e-06,
      "loss": 0.692,
      "num_input_tokens_seen": 3896108288,
      "step": 52540,
      "train_runtime": 62533.9478,
      "train_tokens_per_second": 62303.891
    },
    {
      "epoch": 0.69833423033734,
      "grad_norm": 21.5,
      "learning_rate": 4.4055972982821415e-06,
      "loss": 0.5391,
      "num_input_tokens_seen": 3896223232,
      "step": 52550,
      "train_runtime": 62536.4184,
      "train_tokens_per_second": 62303.268
    },
    {
      "epoch": 0.6984671198198018,
      "grad_norm": 20.125,
      "learning_rate": 4.402030386793896e-06,
      "loss": 0.821,
      "num_input_tokens_seen": 3898268928,
      "step": 52560,
      "train_runtime": 62570.1496,
      "train_tokens_per_second": 62302.375
    },
    {
      "epoch": 0.6986000093022637,
      "grad_norm": 21.25,
      "learning_rate": 4.398464512279608e-06,
      "loss": 0.876,
      "num_input_tokens_seen": 3899107584,
      "step": 52570,
      "train_runtime": 62582.704,
      "train_tokens_per_second": 62303.278
    },
    {
      "epoch": 0.6987328987847257,
      "grad_norm": 30.0,
      "learning_rate": 4.394899675399823e-06,
      "loss": 0.7231,
      "num_input_tokens_seen": 3899544320,
      "step": 52580,
      "train_runtime": 62589.3185,
      "train_tokens_per_second": 62303.671
    },
    {
      "epoch": 0.6988657882671876,
      "grad_norm": 50.5,
      "learning_rate": 4.391335876814886e-06,
      "loss": 0.5889,
      "num_input_tokens_seen": 3899770624,
      "step": 52590,
      "train_runtime": 62592.7773,
      "train_tokens_per_second": 62303.844
    },
    {
      "epoch": 0.6989986777496495,
      "grad_norm": 11.5,
      "learning_rate": 4.387773117184974e-06,
      "loss": 0.4376,
      "num_input_tokens_seen": 3899883776,
      "step": 52600,
      "train_runtime": 62595.1626,
      "train_tokens_per_second": 62303.277
    },
    {
      "epoch": 0.6991315672321115,
      "grad_norm": 18.75,
      "learning_rate": 4.38421139717004e-06,
      "loss": 0.848,
      "num_input_tokens_seen": 3901962752,
      "step": 52610,
      "train_runtime": 62629.0687,
      "train_tokens_per_second": 62302.743
    },
    {
      "epoch": 0.6992644567145734,
      "grad_norm": 18.875,
      "learning_rate": 4.3806507174298795e-06,
      "loss": 0.9457,
      "num_input_tokens_seen": 3902786304,
      "step": 52620,
      "train_runtime": 62641.8606,
      "train_tokens_per_second": 62303.167
    },
    {
      "epoch": 0.6993973461970352,
      "grad_norm": 27.5,
      "learning_rate": 4.377091078624062e-06,
      "loss": 0.6774,
      "num_input_tokens_seen": 3903255552,
      "step": 52630,
      "train_runtime": 62648.4895,
      "train_tokens_per_second": 62304.065
    },
    {
      "epoch": 0.6995302356794971,
      "grad_norm": 18.25,
      "learning_rate": 4.3735324814119825e-06,
      "loss": 0.656,
      "num_input_tokens_seen": 3903493888,
      "step": 52640,
      "train_runtime": 62652.0374,
      "train_tokens_per_second": 62304.341
    },
    {
      "epoch": 0.699663125161959,
      "grad_norm": 6.53125,
      "learning_rate": 4.369974926452845e-06,
      "loss": 0.5536,
      "num_input_tokens_seen": 3903612672,
      "step": 52650,
      "train_runtime": 62654.5054,
      "train_tokens_per_second": 62303.782
    },
    {
      "epoch": 0.699796014644421,
      "grad_norm": 26.75,
      "learning_rate": 4.366418414405644e-06,
      "loss": 0.8564,
      "num_input_tokens_seen": 3905564672,
      "step": 52660,
      "train_runtime": 62686.2312,
      "train_tokens_per_second": 62303.389
    },
    {
      "epoch": 0.6999289041268829,
      "grad_norm": 26.375,
      "learning_rate": 4.362862945929205e-06,
      "loss": 0.8631,
      "num_input_tokens_seen": 3906373632,
      "step": 52670,
      "train_runtime": 62698.2176,
      "train_tokens_per_second": 62304.381
    },
    {
      "epoch": 0.7000617936093448,
      "grad_norm": 16.125,
      "learning_rate": 4.359308521682135e-06,
      "loss": 0.6923,
      "num_input_tokens_seen": 3906806784,
      "step": 52680,
      "train_runtime": 62704.3391,
      "train_tokens_per_second": 62305.206
    },
    {
      "epoch": 0.7001946830918067,
      "grad_norm": 16.375,
      "learning_rate": 4.355755142322866e-06,
      "loss": 0.71,
      "num_input_tokens_seen": 3907031040,
      "step": 52690,
      "train_runtime": 62707.7193,
      "train_tokens_per_second": 62305.424
    },
    {
      "epoch": 0.7003275725742686,
      "grad_norm": 13.0625,
      "learning_rate": 4.352202808509627e-06,
      "loss": 0.5354,
      "num_input_tokens_seen": 3907146240,
      "step": 52700,
      "train_runtime": 62710.1498,
      "train_tokens_per_second": 62304.846
    },
    {
      "epoch": 0.7004604620567305,
      "grad_norm": 19.75,
      "learning_rate": 4.348651520900458e-06,
      "loss": 0.8129,
      "num_input_tokens_seen": 3909167360,
      "step": 52710,
      "train_runtime": 62742.9807,
      "train_tokens_per_second": 62304.457
    },
    {
      "epoch": 0.7005933515391924,
      "grad_norm": 19.0,
      "learning_rate": 4.345101280153204e-06,
      "loss": 0.908,
      "num_input_tokens_seen": 3909991936,
      "step": 52720,
      "train_runtime": 62756.1072,
      "train_tokens_per_second": 62304.565
    },
    {
      "epoch": 0.7007262410216544,
      "grad_norm": 28.875,
      "learning_rate": 4.341552086925512e-06,
      "loss": 0.7093,
      "num_input_tokens_seen": 3910422528,
      "step": 52730,
      "train_runtime": 62762.4311,
      "train_tokens_per_second": 62305.148
    },
    {
      "epoch": 0.7008591305041163,
      "grad_norm": 10.375,
      "learning_rate": 4.338003941874841e-06,
      "loss": 0.6051,
      "num_input_tokens_seen": 3910651136,
      "step": 52740,
      "train_runtime": 62765.9148,
      "train_tokens_per_second": 62305.332
    },
    {
      "epoch": 0.7009920199865781,
      "grad_norm": 13.75,
      "learning_rate": 4.334456845658453e-06,
      "loss": 0.4898,
      "num_input_tokens_seen": 3910767616,
      "step": 52750,
      "train_runtime": 62768.34,
      "train_tokens_per_second": 62304.78
    },
    {
      "epoch": 0.70112490946904,
      "grad_norm": 26.625,
      "learning_rate": 4.330910798933414e-06,
      "loss": 0.7728,
      "num_input_tokens_seen": 3912827648,
      "step": 52760,
      "train_runtime": 62802.4714,
      "train_tokens_per_second": 62303.721
    },
    {
      "epoch": 0.701257798951502,
      "grad_norm": 28.875,
      "learning_rate": 4.3273658023566e-06,
      "loss": 0.7925,
      "num_input_tokens_seen": 3913657088,
      "step": 52770,
      "train_runtime": 62815.8219,
      "train_tokens_per_second": 62303.684
    },
    {
      "epoch": 0.7013906884339639,
      "grad_norm": 35.5,
      "learning_rate": 4.323821856584692e-06,
      "loss": 0.7973,
      "num_input_tokens_seen": 3914098432,
      "step": 52780,
      "train_runtime": 62822.3082,
      "train_tokens_per_second": 62304.276
    },
    {
      "epoch": 0.7015235779164258,
      "grad_norm": 16.75,
      "learning_rate": 4.320278962274166e-06,
      "loss": 0.6226,
      "num_input_tokens_seen": 3914324992,
      "step": 52790,
      "train_runtime": 62825.7252,
      "train_tokens_per_second": 62304.494
    },
    {
      "epoch": 0.7016564673988878,
      "grad_norm": 17.125,
      "learning_rate": 4.316737120081323e-06,
      "loss": 0.5261,
      "num_input_tokens_seen": 3914439936,
      "step": 52800,
      "train_runtime": 62828.1894,
      "train_tokens_per_second": 62303.879
    },
    {
      "epoch": 0.7017893568813496,
      "grad_norm": 14.9375,
      "learning_rate": 4.313196330662248e-06,
      "loss": 0.7956,
      "num_input_tokens_seen": 3916492800,
      "step": 52810,
      "train_runtime": 62861.293,
      "train_tokens_per_second": 62303.726
    },
    {
      "epoch": 0.7019222463638115,
      "grad_norm": 38.25,
      "learning_rate": 4.309656594672845e-06,
      "loss": 0.9037,
      "num_input_tokens_seen": 3917389312,
      "step": 52820,
      "train_runtime": 62874.5983,
      "train_tokens_per_second": 62304.801
    },
    {
      "epoch": 0.7020551358462734,
      "grad_norm": 43.0,
      "learning_rate": 4.306117912768818e-06,
      "loss": 0.7304,
      "num_input_tokens_seen": 3917884928,
      "step": 52830,
      "train_runtime": 62881.5732,
      "train_tokens_per_second": 62305.771
    },
    {
      "epoch": 0.7021880253287354,
      "grad_norm": 16.875,
      "learning_rate": 4.3025802856056765e-06,
      "loss": 0.6834,
      "num_input_tokens_seen": 3918125312,
      "step": 52840,
      "train_runtime": 62885.9055,
      "train_tokens_per_second": 62305.302
    },
    {
      "epoch": 0.7023209148111973,
      "grad_norm": 10.8125,
      "learning_rate": 4.2990437138387374e-06,
      "loss": 0.5569,
      "num_input_tokens_seen": 3918242816,
      "step": 52850,
      "train_runtime": 62888.3577,
      "train_tokens_per_second": 62304.741
    },
    {
      "epoch": 0.7024538042936592,
      "grad_norm": 32.25,
      "learning_rate": 4.295508198123109e-06,
      "loss": 0.842,
      "num_input_tokens_seen": 3920276224,
      "step": 52860,
      "train_runtime": 62921.0722,
      "train_tokens_per_second": 62304.663
    },
    {
      "epoch": 0.702586693776121,
      "grad_norm": 19.0,
      "learning_rate": 4.291973739113727e-06,
      "loss": 0.9047,
      "num_input_tokens_seen": 3921140224,
      "step": 52870,
      "train_runtime": 62933.9083,
      "train_tokens_per_second": 62305.684
    },
    {
      "epoch": 0.702719583258583,
      "grad_norm": 23.25,
      "learning_rate": 4.2884403374653095e-06,
      "loss": 0.7919,
      "num_input_tokens_seen": 3921661696,
      "step": 52880,
      "train_runtime": 62940.8352,
      "train_tokens_per_second": 62307.112
    },
    {
      "epoch": 0.7028524727410449,
      "grad_norm": 109.5,
      "learning_rate": 4.284907993832391e-06,
      "loss": 0.6407,
      "num_input_tokens_seen": 3921890560,
      "step": 52890,
      "train_runtime": 62944.3137,
      "train_tokens_per_second": 62307.305
    },
    {
      "epoch": 0.7029853622235068,
      "grad_norm": 17.625,
      "learning_rate": 4.281376708869309e-06,
      "loss": 0.5458,
      "num_input_tokens_seen": 3922003712,
      "step": 52900,
      "train_runtime": 62946.6844,
      "train_tokens_per_second": 62306.756
    },
    {
      "epoch": 0.7031182517059688,
      "grad_norm": 54.75,
      "learning_rate": 4.277846483230192e-06,
      "loss": 0.8655,
      "num_input_tokens_seen": 3923980288,
      "step": 52910,
      "train_runtime": 62978.7419,
      "train_tokens_per_second": 62306.425
    },
    {
      "epoch": 0.7032511411884307,
      "grad_norm": 32.0,
      "learning_rate": 4.2743173175689985e-06,
      "loss": 0.9406,
      "num_input_tokens_seen": 3924794880,
      "step": 52920,
      "train_runtime": 62991.2478,
      "train_tokens_per_second": 62306.987
    },
    {
      "epoch": 0.7033840306708925,
      "grad_norm": 16.5,
      "learning_rate": 4.2707892125394635e-06,
      "loss": 0.7348,
      "num_input_tokens_seen": 3925282560,
      "step": 52930,
      "train_runtime": 62998.0498,
      "train_tokens_per_second": 62308.001
    },
    {
      "epoch": 0.7035169201533544,
      "grad_norm": 17.5,
      "learning_rate": 4.26726216879514e-06,
      "loss": 0.6337,
      "num_input_tokens_seen": 3925513728,
      "step": 52940,
      "train_runtime": 63001.6037,
      "train_tokens_per_second": 62308.156
    },
    {
      "epoch": 0.7036498096358164,
      "grad_norm": 15.8125,
      "learning_rate": 4.263736186989381e-06,
      "loss": 0.5844,
      "num_input_tokens_seen": 3925630976,
      "step": 52950,
      "train_runtime": 63004.7002,
      "train_tokens_per_second": 62306.954
    },
    {
      "epoch": 0.7037826991182783,
      "grad_norm": 24.875,
      "learning_rate": 4.260211267775345e-06,
      "loss": 0.7587,
      "num_input_tokens_seen": 3927780608,
      "step": 52960,
      "train_runtime": 63040.053,
      "train_tokens_per_second": 62306.112
    },
    {
      "epoch": 0.7039155886007402,
      "grad_norm": 24.125,
      "learning_rate": 4.2566874118059955e-06,
      "loss": 0.9671,
      "num_input_tokens_seen": 3928672256,
      "step": 52970,
      "train_runtime": 63053.0752,
      "train_tokens_per_second": 62307.385
    },
    {
      "epoch": 0.7040484780832021,
      "grad_norm": 17.625,
      "learning_rate": 4.253164619734082e-06,
      "loss": 0.7919,
      "num_input_tokens_seen": 3929134592,
      "step": 52980,
      "train_runtime": 63059.8548,
      "train_tokens_per_second": 62308.018
    },
    {
      "epoch": 0.704181367565664,
      "grad_norm": 13.8125,
      "learning_rate": 4.249642892212187e-06,
      "loss": 0.5854,
      "num_input_tokens_seen": 3929368576,
      "step": 52990,
      "train_runtime": 63063.3639,
      "train_tokens_per_second": 62308.262
    },
    {
      "epoch": 0.7043142570481259,
      "grad_norm": 23.75,
      "learning_rate": 4.246122229892668e-06,
      "loss": 0.5651,
      "num_input_tokens_seen": 3929486336,
      "step": 53000,
      "train_runtime": 63065.8197,
      "train_tokens_per_second": 62307.703
    },
    {
      "epoch": 0.7044471465305878,
      "grad_norm": 14.6875,
      "learning_rate": 4.242602633427698e-06,
      "loss": 0.7567,
      "num_input_tokens_seen": 3931599872,
      "step": 53010,
      "train_runtime": 63100.0155,
      "train_tokens_per_second": 62307.431
    },
    {
      "epoch": 0.7045800360130497,
      "grad_norm": 25.875,
      "learning_rate": 4.239084103469253e-06,
      "loss": 0.81,
      "num_input_tokens_seen": 3932465920,
      "step": 53020,
      "train_runtime": 63113.2613,
      "train_tokens_per_second": 62308.077
    },
    {
      "epoch": 0.7047129254955117,
      "grad_norm": 15.375,
      "learning_rate": 4.23556664066911e-06,
      "loss": 0.7203,
      "num_input_tokens_seen": 3932948736,
      "step": 53030,
      "train_runtime": 63119.9115,
      "train_tokens_per_second": 62309.161
    },
    {
      "epoch": 0.7048458149779736,
      "grad_norm": 19.5,
      "learning_rate": 4.232050245678848e-06,
      "loss": 0.6735,
      "num_input_tokens_seen": 3933169408,
      "step": 53040,
      "train_runtime": 63123.3425,
      "train_tokens_per_second": 62309.27
    },
    {
      "epoch": 0.7049787044604354,
      "grad_norm": 10.4375,
      "learning_rate": 4.228534919149844e-06,
      "loss": 0.5099,
      "num_input_tokens_seen": 3933281024,
      "step": 53050,
      "train_runtime": 63125.728,
      "train_tokens_per_second": 62308.684
    },
    {
      "epoch": 0.7051115939428974,
      "grad_norm": 23.5,
      "learning_rate": 4.225020661733282e-06,
      "loss": 0.8023,
      "num_input_tokens_seen": 3935257088,
      "step": 53060,
      "train_runtime": 63157.084,
      "train_tokens_per_second": 62309.037
    },
    {
      "epoch": 0.7052444834253593,
      "grad_norm": 26.625,
      "learning_rate": 4.221507474080149e-06,
      "loss": 0.8363,
      "num_input_tokens_seen": 3936060416,
      "step": 53070,
      "train_runtime": 63169.8543,
      "train_tokens_per_second": 62309.158
    },
    {
      "epoch": 0.7053773729078212,
      "grad_norm": 20.375,
      "learning_rate": 4.217995356841229e-06,
      "loss": 0.7656,
      "num_input_tokens_seen": 3936522752,
      "step": 53080,
      "train_runtime": 63176.4311,
      "train_tokens_per_second": 62309.989
    },
    {
      "epoch": 0.7055102623902831,
      "grad_norm": 25.875,
      "learning_rate": 4.214484310667113e-06,
      "loss": 0.6491,
      "num_input_tokens_seen": 3936747264,
      "step": 53090,
      "train_runtime": 63179.8566,
      "train_tokens_per_second": 62310.165
    },
    {
      "epoch": 0.7056431518727451,
      "grad_norm": 7.03125,
      "learning_rate": 4.210974336208193e-06,
      "loss": 0.5333,
      "num_input_tokens_seen": 3936858880,
      "step": 53100,
      "train_runtime": 63182.2367,
      "train_tokens_per_second": 62309.584
    },
    {
      "epoch": 0.705776041355207,
      "grad_norm": 18.625,
      "learning_rate": 4.2074654341146515e-06,
      "loss": 0.7958,
      "num_input_tokens_seen": 3938866176,
      "step": 53110,
      "train_runtime": 63214.9402,
      "train_tokens_per_second": 62309.102
    },
    {
      "epoch": 0.7059089308376688,
      "grad_norm": 29.75,
      "learning_rate": 4.203957605036494e-06,
      "loss": 0.8238,
      "num_input_tokens_seen": 3939708672,
      "step": 53120,
      "train_runtime": 63227.9326,
      "train_tokens_per_second": 62309.623
    },
    {
      "epoch": 0.7060418203201307,
      "grad_norm": 53.5,
      "learning_rate": 4.200450849623504e-06,
      "loss": 0.7165,
      "num_input_tokens_seen": 3940192768,
      "step": 53130,
      "train_runtime": 63235.5336,
      "train_tokens_per_second": 62309.789
    },
    {
      "epoch": 0.7061747098025927,
      "grad_norm": 16.625,
      "learning_rate": 4.196945168525281e-06,
      "loss": 0.6651,
      "num_input_tokens_seen": 3940424960,
      "step": 53140,
      "train_runtime": 63239.0587,
      "train_tokens_per_second": 62309.988
    },
    {
      "epoch": 0.7063075992850546,
      "grad_norm": 27.625,
      "learning_rate": 4.193440562391219e-06,
      "loss": 0.5588,
      "num_input_tokens_seen": 3940541696,
      "step": 53150,
      "train_runtime": 63241.4833,
      "train_tokens_per_second": 62309.445
    },
    {
      "epoch": 0.7064404887675165,
      "grad_norm": 97.0,
      "learning_rate": 4.189937031870518e-06,
      "loss": 0.8099,
      "num_input_tokens_seen": 3942680064,
      "step": 53160,
      "train_runtime": 63275.8373,
      "train_tokens_per_second": 62309.41
    },
    {
      "epoch": 0.7065733782499785,
      "grad_norm": 30.25,
      "learning_rate": 4.186434577612178e-06,
      "loss": 0.945,
      "num_input_tokens_seen": 3943521024,
      "step": 53170,
      "train_runtime": 63288.4843,
      "train_tokens_per_second": 62310.246
    },
    {
      "epoch": 0.7067062677324403,
      "grad_norm": 27.5,
      "learning_rate": 4.182933200264986e-06,
      "loss": 0.7298,
      "num_input_tokens_seen": 3944014336,
      "step": 53180,
      "train_runtime": 63295.0393,
      "train_tokens_per_second": 62311.587
    },
    {
      "epoch": 0.7068391572149022,
      "grad_norm": 17.0,
      "learning_rate": 4.179432900477556e-06,
      "loss": 0.6458,
      "num_input_tokens_seen": 3944241408,
      "step": 53190,
      "train_runtime": 63299.3084,
      "train_tokens_per_second": 62310.972
    },
    {
      "epoch": 0.7069720466973641,
      "grad_norm": 27.0,
      "learning_rate": 4.175933678898277e-06,
      "loss": 0.5276,
      "num_input_tokens_seen": 3944355328,
      "step": 53200,
      "train_runtime": 63301.7664,
      "train_tokens_per_second": 62310.352
    },
    {
      "epoch": 0.7071049361798261,
      "grad_norm": 16.25,
      "learning_rate": 4.172435536175352e-06,
      "loss": 0.7791,
      "num_input_tokens_seen": 3946395136,
      "step": 53210,
      "train_runtime": 63333.4702,
      "train_tokens_per_second": 62311.368
    },
    {
      "epoch": 0.707237825662288,
      "grad_norm": 19.25,
      "learning_rate": 4.168938472956784e-06,
      "loss": 0.9701,
      "num_input_tokens_seen": 3947237120,
      "step": 53220,
      "train_runtime": 63346.0245,
      "train_tokens_per_second": 62312.31
    },
    {
      "epoch": 0.7073707151447499,
      "grad_norm": 18.875,
      "learning_rate": 4.1654424898903615e-06,
      "loss": 0.7141,
      "num_input_tokens_seen": 3947715328,
      "step": 53230,
      "train_runtime": 63352.6624,
      "train_tokens_per_second": 62313.329
    },
    {
      "epoch": 0.7075036046272117,
      "grad_norm": 54.25,
      "learning_rate": 4.161947587623699e-06,
      "loss": 0.6526,
      "num_input_tokens_seen": 3947938816,
      "step": 53240,
      "train_runtime": 63356.8146,
      "train_tokens_per_second": 62312.773
    },
    {
      "epoch": 0.7076364941096737,
      "grad_norm": 14.0625,
      "learning_rate": 4.158453766804187e-06,
      "loss": 0.469,
      "num_input_tokens_seen": 3948054784,
      "step": 53250,
      "train_runtime": 63359.2024,
      "train_tokens_per_second": 62312.255
    },
    {
      "epoch": 0.7077693835921356,
      "grad_norm": 15.4375,
      "learning_rate": 4.154961028079025e-06,
      "loss": 0.8116,
      "num_input_tokens_seen": 3950127872,
      "step": 53260,
      "train_runtime": 63392.4983,
      "train_tokens_per_second": 62312.229
    },
    {
      "epoch": 0.7079022730745975,
      "grad_norm": 15.1875,
      "learning_rate": 4.1514693720952135e-06,
      "loss": 0.8734,
      "num_input_tokens_seen": 3950973440,
      "step": 53270,
      "train_runtime": 63405.0796,
      "train_tokens_per_second": 62313.201
    },
    {
      "epoch": 0.7080351625570594,
      "grad_norm": 19.0,
      "learning_rate": 4.147978799499551e-06,
      "loss": 0.7099,
      "num_input_tokens_seen": 3951425536,
      "step": 53280,
      "train_runtime": 63411.5747,
      "train_tokens_per_second": 62313.947
    },
    {
      "epoch": 0.7081680520395214,
      "grad_norm": 17.5,
      "learning_rate": 4.144489310938637e-06,
      "loss": 0.5896,
      "num_input_tokens_seen": 3951648000,
      "step": 53290,
      "train_runtime": 63414.9558,
      "train_tokens_per_second": 62314.133
    },
    {
      "epoch": 0.7083009415219832,
      "grad_norm": 17.875,
      "learning_rate": 4.141000907058859e-06,
      "loss": 0.4408,
      "num_input_tokens_seen": 3951763200,
      "step": 53300,
      "train_runtime": 63417.2913,
      "train_tokens_per_second": 62313.655
    },
    {
      "epoch": 0.7084338310044451,
      "grad_norm": 32.0,
      "learning_rate": 4.137513588506427e-06,
      "loss": 0.8339,
      "num_input_tokens_seen": 3953847040,
      "step": 53310,
      "train_runtime": 63451.9264,
      "train_tokens_per_second": 62312.482
    },
    {
      "epoch": 0.708566720486907,
      "grad_norm": 56.75,
      "learning_rate": 4.134027355927324e-06,
      "loss": 0.8249,
      "num_input_tokens_seen": 3954733824,
      "step": 53320,
      "train_runtime": 63465.1514,
      "train_tokens_per_second": 62313.47
    },
    {
      "epoch": 0.708699609969369,
      "grad_norm": 35.75,
      "learning_rate": 4.130542209967346e-06,
      "loss": 0.7864,
      "num_input_tokens_seen": 3955235584,
      "step": 53330,
      "train_runtime": 63472.331,
      "train_tokens_per_second": 62314.327
    },
    {
      "epoch": 0.7088324994518309,
      "grad_norm": 19.125,
      "learning_rate": 4.127058151272088e-06,
      "loss": 0.5991,
      "num_input_tokens_seen": 3955473920,
      "step": 53340,
      "train_runtime": 63475.8709,
      "train_tokens_per_second": 62314.607
    },
    {
      "epoch": 0.7089653889342928,
      "grad_norm": 11.125,
      "learning_rate": 4.12357518048694e-06,
      "loss": 0.5987,
      "num_input_tokens_seen": 3955591936,
      "step": 53350,
      "train_runtime": 63478.2841,
      "train_tokens_per_second": 62314.097
    },
    {
      "epoch": 0.7090982784167547,
      "grad_norm": 16.5,
      "learning_rate": 4.120093298257093e-06,
      "loss": 0.9366,
      "num_input_tokens_seen": 3957673216,
      "step": 53360,
      "train_runtime": 63511.2609,
      "train_tokens_per_second": 62314.512
    },
    {
      "epoch": 0.7092311678992166,
      "grad_norm": 17.125,
      "learning_rate": 4.11661250522753e-06,
      "loss": 0.8775,
      "num_input_tokens_seen": 3958501120,
      "step": 53370,
      "train_runtime": 63523.6517,
      "train_tokens_per_second": 62315.39
    },
    {
      "epoch": 0.7093640573816785,
      "grad_norm": 17.0,
      "learning_rate": 4.113132802043038e-06,
      "loss": 0.7307,
      "num_input_tokens_seen": 3958950912,
      "step": 53380,
      "train_runtime": 63530.118,
      "train_tokens_per_second": 62316.127
    },
    {
      "epoch": 0.7094969468641404,
      "grad_norm": 20.125,
      "learning_rate": 4.109654189348202e-06,
      "loss": 0.6172,
      "num_input_tokens_seen": 3959178240,
      "step": 53390,
      "train_runtime": 63533.557,
      "train_tokens_per_second": 62316.332
    },
    {
      "epoch": 0.7096298363466024,
      "grad_norm": 10.0625,
      "learning_rate": 4.106176667787406e-06,
      "loss": 0.5162,
      "num_input_tokens_seen": 3959293440,
      "step": 53400,
      "train_runtime": 63535.9679,
      "train_tokens_per_second": 62315.781
    },
    {
      "epoch": 0.7097627258290643,
      "grad_norm": 25.25,
      "learning_rate": 4.102700238004826e-06,
      "loss": 0.8503,
      "num_input_tokens_seen": 3961340672,
      "step": 53410,
      "train_runtime": 63567.9965,
      "train_tokens_per_second": 62316.588
    },
    {
      "epoch": 0.7098956153115261,
      "grad_norm": 29.25,
      "learning_rate": 4.099224900644446e-06,
      "loss": 0.9153,
      "num_input_tokens_seen": 3962175488,
      "step": 53420,
      "train_runtime": 63580.4807,
      "train_tokens_per_second": 62317.482
    },
    {
      "epoch": 0.710028504793988,
      "grad_norm": 34.75,
      "learning_rate": 4.095750656350031e-06,
      "loss": 0.6592,
      "num_input_tokens_seen": 3962622464,
      "step": 53430,
      "train_runtime": 63587.6605,
      "train_tokens_per_second": 62317.475
    },
    {
      "epoch": 0.71016139427645,
      "grad_norm": 16.625,
      "learning_rate": 4.092277505765165e-06,
      "loss": 0.6441,
      "num_input_tokens_seen": 3962834176,
      "step": 53440,
      "train_runtime": 63590.9394,
      "train_tokens_per_second": 62317.591
    },
    {
      "epoch": 0.7102942837589119,
      "grad_norm": 8.6875,
      "learning_rate": 4.08880544953321e-06,
      "loss": 0.4811,
      "num_input_tokens_seen": 3962947072,
      "step": 53450,
      "train_runtime": 63593.3259,
      "train_tokens_per_second": 62317.028
    },
    {
      "epoch": 0.7104271732413738,
      "grad_norm": 15.3125,
      "learning_rate": 4.085334488297337e-06,
      "loss": 0.837,
      "num_input_tokens_seen": 3965159168,
      "step": 53460,
      "train_runtime": 63629.3465,
      "train_tokens_per_second": 62316.516
    },
    {
      "epoch": 0.7105600627238358,
      "grad_norm": 21.5,
      "learning_rate": 4.081864622700512e-06,
      "loss": 0.9219,
      "num_input_tokens_seen": 3966089728,
      "step": 53470,
      "train_runtime": 63642.5556,
      "train_tokens_per_second": 62318.203
    },
    {
      "epoch": 0.7106929522062976,
      "grad_norm": 24.875,
      "learning_rate": 4.078395853385488e-06,
      "loss": 0.7156,
      "num_input_tokens_seen": 3966600704,
      "step": 53480,
      "train_runtime": 63650.3545,
      "train_tokens_per_second": 62318.596
    },
    {
      "epoch": 0.7108258416887595,
      "grad_norm": 280.0,
      "learning_rate": 4.074928180994838e-06,
      "loss": 0.6356,
      "num_input_tokens_seen": 3966837248,
      "step": 53490,
      "train_runtime": 63653.8699,
      "train_tokens_per_second": 62318.87
    },
    {
      "epoch": 0.7109587311712214,
      "grad_norm": 11.75,
      "learning_rate": 4.071461606170902e-06,
      "loss": 0.5428,
      "num_input_tokens_seen": 3966957056,
      "step": 53500,
      "train_runtime": 63656.3381,
      "train_tokens_per_second": 62318.336
    },
    {
      "epoch": 0.7110916206536834,
      "grad_norm": 21.25,
      "learning_rate": 4.067996129555846e-06,
      "loss": 0.8366,
      "num_input_tokens_seen": 3968976640,
      "step": 53510,
      "train_runtime": 63689.5008,
      "train_tokens_per_second": 62317.597
    },
    {
      "epoch": 0.7112245101361453,
      "grad_norm": 22.125,
      "learning_rate": 4.06453175179161e-06,
      "loss": 0.9116,
      "num_input_tokens_seen": 3969799168,
      "step": 53520,
      "train_runtime": 63701.6948,
      "train_tokens_per_second": 62318.58
    },
    {
      "epoch": 0.7113573996186072,
      "grad_norm": 12.0625,
      "learning_rate": 4.061068473519941e-06,
      "loss": 0.7023,
      "num_input_tokens_seen": 3970260736,
      "step": 53530,
      "train_runtime": 63708.228,
      "train_tokens_per_second": 62319.434
    },
    {
      "epoch": 0.7114902891010692,
      "grad_norm": 17.125,
      "learning_rate": 4.057606295382385e-06,
      "loss": 0.5567,
      "num_input_tokens_seen": 3970498048,
      "step": 53540,
      "train_runtime": 63712.3751,
      "train_tokens_per_second": 62319.103
    },
    {
      "epoch": 0.711623178583531,
      "grad_norm": 27.625,
      "learning_rate": 4.054145218020268e-06,
      "loss": 0.5505,
      "num_input_tokens_seen": 3970613760,
      "step": 53550,
      "train_runtime": 63714.7396,
      "train_tokens_per_second": 62318.606
    },
    {
      "epoch": 0.7117560680659929,
      "grad_norm": 23.125,
      "learning_rate": 4.050685242074738e-06,
      "loss": 0.8893,
      "num_input_tokens_seen": 3972827136,
      "step": 53560,
      "train_runtime": 63748.6233,
      "train_tokens_per_second": 62320.203
    },
    {
      "epoch": 0.7118889575484548,
      "grad_norm": 48.25,
      "learning_rate": 4.047226368186713e-06,
      "loss": 0.8221,
      "num_input_tokens_seen": 3973670144,
      "step": 53570,
      "train_runtime": 63761.4958,
      "train_tokens_per_second": 62320.843
    },
    {
      "epoch": 0.7120218470309168,
      "grad_norm": 21.0,
      "learning_rate": 4.043768596996926e-06,
      "loss": 0.7982,
      "num_input_tokens_seen": 3974161408,
      "step": 53580,
      "train_runtime": 63768.3686,
      "train_tokens_per_second": 62321.83
    },
    {
      "epoch": 0.7121547365133787,
      "grad_norm": 14.125,
      "learning_rate": 4.0403119291458934e-06,
      "loss": 0.6333,
      "num_input_tokens_seen": 3974403584,
      "step": 53590,
      "train_runtime": 63771.9131,
      "train_tokens_per_second": 62322.163
    },
    {
      "epoch": 0.7122876259958406,
      "grad_norm": 8.5625,
      "learning_rate": 4.036856365273936e-06,
      "loss": 0.5505,
      "num_input_tokens_seen": 3974524160,
      "step": 53600,
      "train_runtime": 63774.9536,
      "train_tokens_per_second": 62321.083
    },
    {
      "epoch": 0.7124205154783024,
      "grad_norm": 20.125,
      "learning_rate": 4.033401906021166e-06,
      "loss": 0.826,
      "num_input_tokens_seen": 3976564736,
      "step": 53610,
      "train_runtime": 63807.8502,
      "train_tokens_per_second": 62320.933
    },
    {
      "epoch": 0.7125534049607644,
      "grad_norm": 41.25,
      "learning_rate": 4.029948552027488e-06,
      "loss": 0.9144,
      "num_input_tokens_seen": 3977399040,
      "step": 53620,
      "train_runtime": 63820.2605,
      "train_tokens_per_second": 62321.887
    },
    {
      "epoch": 0.7126862944432263,
      "grad_norm": 13.75,
      "learning_rate": 4.026496303932606e-06,
      "loss": 0.7029,
      "num_input_tokens_seen": 3977877504,
      "step": 53630,
      "train_runtime": 63826.6853,
      "train_tokens_per_second": 62323.11
    },
    {
      "epoch": 0.7128191839256882,
      "grad_norm": 11.8125,
      "learning_rate": 4.0230451623760195e-06,
      "loss": 0.5586,
      "num_input_tokens_seen": 3978106880,
      "step": 53640,
      "train_runtime": 63830.1813,
      "train_tokens_per_second": 62323.29
    },
    {
      "epoch": 0.7129520734081501,
      "grad_norm": 11.75,
      "learning_rate": 4.0195951279970215e-06,
      "loss": 0.5243,
      "num_input_tokens_seen": 3978226432,
      "step": 53650,
      "train_runtime": 63833.2374,
      "train_tokens_per_second": 62322.179
    },
    {
      "epoch": 0.7130849628906121,
      "grad_norm": 25.625,
      "learning_rate": 4.016146201434699e-06,
      "loss": 0.8207,
      "num_input_tokens_seen": 3980304896,
      "step": 53660,
      "train_runtime": 63865.8153,
      "train_tokens_per_second": 62322.932
    },
    {
      "epoch": 0.7132178523730739,
      "grad_norm": 23.75,
      "learning_rate": 4.012698383327935e-06,
      "loss": 0.926,
      "num_input_tokens_seen": 3981164032,
      "step": 53670,
      "train_runtime": 63878.6342,
      "train_tokens_per_second": 62323.875
    },
    {
      "epoch": 0.7133507418555358,
      "grad_norm": 33.0,
      "learning_rate": 4.0092516743154085e-06,
      "loss": 0.8693,
      "num_input_tokens_seen": 3981655040,
      "step": 53680,
      "train_runtime": 63885.5217,
      "train_tokens_per_second": 62324.842
    },
    {
      "epoch": 0.7134836313379977,
      "grad_norm": 23.0,
      "learning_rate": 4.005806075035594e-06,
      "loss": 0.6148,
      "num_input_tokens_seen": 3981896448,
      "step": 53690,
      "train_runtime": 63889.0655,
      "train_tokens_per_second": 62325.163
    },
    {
      "epoch": 0.7136165208204597,
      "grad_norm": 77.5,
      "learning_rate": 4.002361586126753e-06,
      "loss": 0.5706,
      "num_input_tokens_seen": 3982015232,
      "step": 53700,
      "train_runtime": 63891.4616,
      "train_tokens_per_second": 62324.685
    },
    {
      "epoch": 0.7137494103029216,
      "grad_norm": 61.5,
      "learning_rate": 3.998918208226946e-06,
      "loss": 0.7736,
      "num_input_tokens_seen": 3984013056,
      "step": 53710,
      "train_runtime": 63923.175,
      "train_tokens_per_second": 62325.018
    },
    {
      "epoch": 0.7138822997853835,
      "grad_norm": 18.875,
      "learning_rate": 3.995475941974032e-06,
      "loss": 0.8835,
      "num_input_tokens_seen": 3984830464,
      "step": 53720,
      "train_runtime": 63936.2062,
      "train_tokens_per_second": 62325.1
    },
    {
      "epoch": 0.7140151892678454,
      "grad_norm": 24.0,
      "learning_rate": 3.992034788005658e-06,
      "loss": 0.722,
      "num_input_tokens_seen": 3985271552,
      "step": 53730,
      "train_runtime": 63942.5682,
      "train_tokens_per_second": 62325.797
    },
    {
      "epoch": 0.7141480787503073,
      "grad_norm": 11.3125,
      "learning_rate": 3.98859474695927e-06,
      "loss": 0.6058,
      "num_input_tokens_seen": 3985494528,
      "step": 53740,
      "train_runtime": 63946.0104,
      "train_tokens_per_second": 62325.929
    },
    {
      "epoch": 0.7142809682327692,
      "grad_norm": 10.4375,
      "learning_rate": 3.985155819472095e-06,
      "loss": 0.544,
      "num_input_tokens_seen": 3985610240,
      "step": 53750,
      "train_runtime": 63948.4035,
      "train_tokens_per_second": 62325.406
    },
    {
      "epoch": 0.7144138577152311,
      "grad_norm": 36.0,
      "learning_rate": 3.98171800618118e-06,
      "loss": 0.8415,
      "num_input_tokens_seen": 3987638784,
      "step": 53760,
      "train_runtime": 63981.8656,
      "train_tokens_per_second": 62324.516
    },
    {
      "epoch": 0.7145467471976931,
      "grad_norm": 26.75,
      "learning_rate": 3.978281307723335e-06,
      "loss": 0.9538,
      "num_input_tokens_seen": 3988483328,
      "step": 53770,
      "train_runtime": 63995.0633,
      "train_tokens_per_second": 62324.86
    },
    {
      "epoch": 0.714679636680155,
      "grad_norm": 14.875,
      "learning_rate": 3.974845724735185e-06,
      "loss": 0.6765,
      "num_input_tokens_seen": 3988934400,
      "step": 53780,
      "train_runtime": 64001.5124,
      "train_tokens_per_second": 62325.627
    },
    {
      "epoch": 0.7148125261626168,
      "grad_norm": 16.125,
      "learning_rate": 3.97141125785314e-06,
      "loss": 0.6352,
      "num_input_tokens_seen": 3989160704,
      "step": 53790,
      "train_runtime": 64004.9116,
      "train_tokens_per_second": 62325.853
    },
    {
      "epoch": 0.7149454156450787,
      "grad_norm": 11.9375,
      "learning_rate": 3.9679779077133995e-06,
      "loss": 0.5053,
      "num_input_tokens_seen": 3989277184,
      "step": 53800,
      "train_runtime": 64007.3303,
      "train_tokens_per_second": 62325.318
    },
    {
      "epoch": 0.7150783051275407,
      "grad_norm": 18.25,
      "learning_rate": 3.96454567495197e-06,
      "loss": 0.8384,
      "num_input_tokens_seen": 3991278592,
      "step": 53810,
      "train_runtime": 64038.4331,
      "train_tokens_per_second": 62326.3
    },
    {
      "epoch": 0.7152111946100026,
      "grad_norm": 21.5,
      "learning_rate": 3.961114560204631e-06,
      "loss": 0.984,
      "num_input_tokens_seen": 3992104448,
      "step": 53820,
      "train_runtime": 64050.9827,
      "train_tokens_per_second": 62326.982
    },
    {
      "epoch": 0.7153440840924645,
      "grad_norm": 18.125,
      "learning_rate": 3.957684564106978e-06,
      "loss": 0.7667,
      "num_input_tokens_seen": 3992562944,
      "step": 53830,
      "train_runtime": 64058.599,
      "train_tokens_per_second": 62326.729
    },
    {
      "epoch": 0.7154769735749265,
      "grad_norm": 24.25,
      "learning_rate": 3.954255687294379e-06,
      "loss": 0.6772,
      "num_input_tokens_seen": 3992783872,
      "step": 53840,
      "train_runtime": 64061.977,
      "train_tokens_per_second": 62326.891
    },
    {
      "epoch": 0.7156098630573883,
      "grad_norm": 8.5625,
      "learning_rate": 3.950827930402005e-06,
      "loss": 0.4317,
      "num_input_tokens_seen": 3992895232,
      "step": 53850,
      "train_runtime": 64064.3579,
      "train_tokens_per_second": 62326.313
    },
    {
      "epoch": 0.7157427525398502,
      "grad_norm": 18.125,
      "learning_rate": 3.9474012940648195e-06,
      "loss": 0.8329,
      "num_input_tokens_seen": 3994926848,
      "step": 53860,
      "train_runtime": 64097.8842,
      "train_tokens_per_second": 62325.409
    },
    {
      "epoch": 0.7158756420223121,
      "grad_norm": 17.5,
      "learning_rate": 3.943975778917568e-06,
      "loss": 0.8576,
      "num_input_tokens_seen": 3995801088,
      "step": 53870,
      "train_runtime": 64110.4182,
      "train_tokens_per_second": 62326.86
    },
    {
      "epoch": 0.7160085315047741,
      "grad_norm": 26.25,
      "learning_rate": 3.94055138559481e-06,
      "loss": 0.7114,
      "num_input_tokens_seen": 3996260864,
      "step": 53880,
      "train_runtime": 64117.8285,
      "train_tokens_per_second": 62326.828
    },
    {
      "epoch": 0.716141420987236,
      "grad_norm": 15.1875,
      "learning_rate": 3.9371281147308735e-06,
      "loss": 0.6564,
      "num_input_tokens_seen": 3996489984,
      "step": 53890,
      "train_runtime": 64121.2927,
      "train_tokens_per_second": 62327.034
    },
    {
      "epoch": 0.7162743104696979,
      "grad_norm": 7.1875,
      "learning_rate": 3.9337059669598924e-06,
      "loss": 0.5282,
      "num_input_tokens_seen": 3996608512,
      "step": 53900,
      "train_runtime": 64123.7484,
      "train_tokens_per_second": 62326.495
    },
    {
      "epoch": 0.7164071999521597,
      "grad_norm": 21.875,
      "learning_rate": 3.9302849429157876e-06,
      "loss": 0.8712,
      "num_input_tokens_seen": 3998731264,
      "step": 53910,
      "train_runtime": 64157.3336,
      "train_tokens_per_second": 62326.955
    },
    {
      "epoch": 0.7165400894346217,
      "grad_norm": 20.5,
      "learning_rate": 3.926865043232275e-06,
      "loss": 0.8196,
      "num_input_tokens_seen": 3999568896,
      "step": 53920,
      "train_runtime": 64170.1102,
      "train_tokens_per_second": 62327.599
    },
    {
      "epoch": 0.7166729789170836,
      "grad_norm": 28.375,
      "learning_rate": 3.9234462685428635e-06,
      "loss": 0.774,
      "num_input_tokens_seen": 4000025088,
      "step": 53930,
      "train_runtime": 64176.6552,
      "train_tokens_per_second": 62328.351
    },
    {
      "epoch": 0.7168058683995455,
      "grad_norm": 11.5625,
      "learning_rate": 3.920028619480843e-06,
      "loss": 0.6463,
      "num_input_tokens_seen": 4000248064,
      "step": 53940,
      "train_runtime": 64180.0606,
      "train_tokens_per_second": 62328.518
    },
    {
      "epoch": 0.7169387578820074,
      "grad_norm": 7.3125,
      "learning_rate": 3.916612096679308e-06,
      "loss": 0.5214,
      "num_input_tokens_seen": 4000365824,
      "step": 53950,
      "train_runtime": 64182.4996,
      "train_tokens_per_second": 62327.984
    },
    {
      "epoch": 0.7170716473644694,
      "grad_norm": 23.625,
      "learning_rate": 3.913196700771137e-06,
      "loss": 0.8247,
      "num_input_tokens_seen": 4002455040,
      "step": 53960,
      "train_runtime": 64217.5601,
      "train_tokens_per_second": 62326.489
    },
    {
      "epoch": 0.7172045368469312,
      "grad_norm": 25.25,
      "learning_rate": 3.9097824323890025e-06,
      "loss": 0.8478,
      "num_input_tokens_seen": 4003306240,
      "step": 53970,
      "train_runtime": 64230.2262,
      "train_tokens_per_second": 62327.45
    },
    {
      "epoch": 0.7173374263293931,
      "grad_norm": 19.125,
      "learning_rate": 3.906369292165366e-06,
      "loss": 0.7663,
      "num_input_tokens_seen": 4003763968,
      "step": 53980,
      "train_runtime": 64236.8551,
      "train_tokens_per_second": 62328.144
    },
    {
      "epoch": 0.717470315811855,
      "grad_norm": 8.5,
      "learning_rate": 3.902957280732486e-06,
      "loss": 0.6051,
      "num_input_tokens_seen": 4003987712,
      "step": 53990,
      "train_runtime": 64240.2126,
      "train_tokens_per_second": 62328.37
    },
    {
      "epoch": 0.717603205294317,
      "grad_norm": 11.875,
      "learning_rate": 3.899546398722397e-06,
      "loss": 0.5876,
      "num_input_tokens_seen": 4004101376,
      "step": 54000,
      "train_runtime": 64242.6386,
      "train_tokens_per_second": 62327.785
    },
    {
      "epoch": 0.7177360947767789,
      "grad_norm": 33.0,
      "learning_rate": 3.896136646766947e-06,
      "loss": 0.8724,
      "num_input_tokens_seen": 4006187776,
      "step": 54010,
      "train_runtime": 64276.1025,
      "train_tokens_per_second": 62327.796
    },
    {
      "epoch": 0.7178689842592408,
      "grad_norm": 19.0,
      "learning_rate": 3.8927280254977525e-06,
      "loss": 0.8728,
      "num_input_tokens_seen": 4007012608,
      "step": 54020,
      "train_runtime": 64289.2132,
      "train_tokens_per_second": 62327.915
    },
    {
      "epoch": 0.7180018737417028,
      "grad_norm": 17.125,
      "learning_rate": 3.889320535546235e-06,
      "loss": 0.7159,
      "num_input_tokens_seen": 4007463680,
      "step": 54030,
      "train_runtime": 64295.6601,
      "train_tokens_per_second": 62328.681
    },
    {
      "epoch": 0.7181347632241646,
      "grad_norm": 26.875,
      "learning_rate": 3.885914177543599e-06,
      "loss": 0.6222,
      "num_input_tokens_seen": 4007684608,
      "step": 54040,
      "train_runtime": 64299.1008,
      "train_tokens_per_second": 62328.782
    },
    {
      "epoch": 0.7182676527066265,
      "grad_norm": 10.875,
      "learning_rate": 3.882508952120844e-06,
      "loss": 0.4839,
      "num_input_tokens_seen": 4007800320,
      "step": 54050,
      "train_runtime": 64301.5265,
      "train_tokens_per_second": 62328.23
    },
    {
      "epoch": 0.7184005421890884,
      "grad_norm": 19.625,
      "learning_rate": 3.879104859908761e-06,
      "loss": 0.8581,
      "num_input_tokens_seen": 4009856256,
      "step": 54060,
      "train_runtime": 64333.8183,
      "train_tokens_per_second": 62328.902
    },
    {
      "epoch": 0.7185334316715504,
      "grad_norm": 15.25,
      "learning_rate": 3.875701901537916e-06,
      "loss": 0.824,
      "num_input_tokens_seen": 4010654720,
      "step": 54070,
      "train_runtime": 64346.0229,
      "train_tokens_per_second": 62329.489
    },
    {
      "epoch": 0.7186663211540123,
      "grad_norm": 18.125,
      "learning_rate": 3.872300077638691e-06,
      "loss": 0.8105,
      "num_input_tokens_seen": 4011089920,
      "step": 54080,
      "train_runtime": 64353.1281,
      "train_tokens_per_second": 62329.37
    },
    {
      "epoch": 0.7187992106364742,
      "grad_norm": 13.125,
      "learning_rate": 3.868899388841234e-06,
      "loss": 0.5882,
      "num_input_tokens_seen": 4011309824,
      "step": 54090,
      "train_runtime": 64356.5009,
      "train_tokens_per_second": 62329.52
    },
    {
      "epoch": 0.718932100118936,
      "grad_norm": 17.625,
      "learning_rate": 3.865499835775495e-06,
      "loss": 0.569,
      "num_input_tokens_seen": 4011423488,
      "step": 54100,
      "train_runtime": 64358.9838,
      "train_tokens_per_second": 62328.882
    },
    {
      "epoch": 0.719064989601398,
      "grad_norm": 20.625,
      "learning_rate": 3.862101419071214e-06,
      "loss": 0.8209,
      "num_input_tokens_seen": 4013491968,
      "step": 54110,
      "train_runtime": 64393.1844,
      "train_tokens_per_second": 62327.9
    },
    {
      "epoch": 0.7191978790838599,
      "grad_norm": 14.625,
      "learning_rate": 3.8587041393579086e-06,
      "loss": 0.898,
      "num_input_tokens_seen": 4014342656,
      "step": 54120,
      "train_runtime": 64406.7024,
      "train_tokens_per_second": 62328.027
    },
    {
      "epoch": 0.7193307685663218,
      "grad_norm": 12.75,
      "learning_rate": 3.855307997264907e-06,
      "loss": 0.7101,
      "num_input_tokens_seen": 4014820352,
      "step": 54130,
      "train_runtime": 64413.4955,
      "train_tokens_per_second": 62328.87
    },
    {
      "epoch": 0.7194636580487838,
      "grad_norm": 18.875,
      "learning_rate": 3.851912993421304e-06,
      "loss": 0.6216,
      "num_input_tokens_seen": 4015051008,
      "step": 54140,
      "train_runtime": 64416.9251,
      "train_tokens_per_second": 62329.132
    },
    {
      "epoch": 0.7195965475312457,
      "grad_norm": 17.0,
      "learning_rate": 3.848519128455997e-06,
      "loss": 0.5742,
      "num_input_tokens_seen": 4015166976,
      "step": 54150,
      "train_runtime": 64419.4333,
      "train_tokens_per_second": 62328.505
    },
    {
      "epoch": 0.7197294370137075,
      "grad_norm": 16.75,
      "learning_rate": 3.845126402997671e-06,
      "loss": 0.77,
      "num_input_tokens_seen": 4017111808,
      "step": 54160,
      "train_runtime": 64451.2919,
      "train_tokens_per_second": 62327.871
    },
    {
      "epoch": 0.7198623264961694,
      "grad_norm": 20.875,
      "learning_rate": 3.841734817674797e-06,
      "loss": 0.8999,
      "num_input_tokens_seen": 4017929216,
      "step": 54170,
      "train_runtime": 64463.4794,
      "train_tokens_per_second": 62328.767
    },
    {
      "epoch": 0.7199952159786314,
      "grad_norm": 16.5,
      "learning_rate": 3.838344373115638e-06,
      "loss": 0.7745,
      "num_input_tokens_seen": 4018385920,
      "step": 54180,
      "train_runtime": 64470.011,
      "train_tokens_per_second": 62329.537
    },
    {
      "epoch": 0.7201281054610933,
      "grad_norm": 15.0625,
      "learning_rate": 3.834955069948235e-06,
      "loss": 0.6422,
      "num_input_tokens_seen": 4018611200,
      "step": 54190,
      "train_runtime": 64474.2876,
      "train_tokens_per_second": 62328.897
    },
    {
      "epoch": 0.7202609949435552,
      "grad_norm": 18.5,
      "learning_rate": 3.831566908800439e-06,
      "loss": 0.5185,
      "num_input_tokens_seen": 4018727424,
      "step": 54200,
      "train_runtime": 64476.8567,
      "train_tokens_per_second": 62328.216
    },
    {
      "epoch": 0.7203938844260172,
      "grad_norm": 41.5,
      "learning_rate": 3.828179890299868e-06,
      "loss": 0.8555,
      "num_input_tokens_seen": 4020644352,
      "step": 54210,
      "train_runtime": 64507.622,
      "train_tokens_per_second": 62328.206
    },
    {
      "epoch": 0.720526773908479,
      "grad_norm": 13.5,
      "learning_rate": 3.8247940150739386e-06,
      "loss": 0.8484,
      "num_input_tokens_seen": 4021456896,
      "step": 54220,
      "train_runtime": 64519.7135,
      "train_tokens_per_second": 62329.119
    },
    {
      "epoch": 0.7206596633909409,
      "grad_norm": 20.25,
      "learning_rate": 3.821409283749854e-06,
      "loss": 0.6868,
      "num_input_tokens_seen": 4021888256,
      "step": 54230,
      "train_runtime": 64526.0027,
      "train_tokens_per_second": 62329.729
    },
    {
      "epoch": 0.7207925528734028,
      "grad_norm": 17.875,
      "learning_rate": 3.818025696954607e-06,
      "loss": 0.6785,
      "num_input_tokens_seen": 4022107136,
      "step": 54240,
      "train_runtime": 64530.0742,
      "train_tokens_per_second": 62329.188
    },
    {
      "epoch": 0.7209254423558648,
      "grad_norm": 11.5625,
      "learning_rate": 3.814643255314979e-06,
      "loss": 0.5156,
      "num_input_tokens_seen": 4022217984,
      "step": 54250,
      "train_runtime": 64532.4961,
      "train_tokens_per_second": 62328.567
    },
    {
      "epoch": 0.7210583318383267,
      "grad_norm": 14.1875,
      "learning_rate": 3.8112619594575308e-06,
      "loss": 0.8116,
      "num_input_tokens_seen": 4024291328,
      "step": 54260,
      "train_runtime": 64565.3084,
      "train_tokens_per_second": 62329.003
    },
    {
      "epoch": 0.7211912213207886,
      "grad_norm": 23.125,
      "learning_rate": 3.80788181000862e-06,
      "loss": 0.9413,
      "num_input_tokens_seen": 4025166336,
      "step": 54270,
      "train_runtime": 64578.4169,
      "train_tokens_per_second": 62329.901
    },
    {
      "epoch": 0.7213241108032504,
      "grad_norm": 36.75,
      "learning_rate": 3.8045028075943913e-06,
      "loss": 0.7638,
      "num_input_tokens_seen": 4025655552,
      "step": 54280,
      "train_runtime": 64585.2746,
      "train_tokens_per_second": 62330.858
    },
    {
      "epoch": 0.7214570002857124,
      "grad_norm": 31.375,
      "learning_rate": 3.8011249528407723e-06,
      "loss": 0.6569,
      "num_input_tokens_seen": 4025892608,
      "step": 54290,
      "train_runtime": 64588.78,
      "train_tokens_per_second": 62331.145
    },
    {
      "epoch": 0.7215898897681743,
      "grad_norm": 13.1875,
      "learning_rate": 3.7977482463734818e-06,
      "loss": 0.5144,
      "num_input_tokens_seen": 4026008576,
      "step": 54300,
      "train_runtime": 64591.2215,
      "train_tokens_per_second": 62330.584
    },
    {
      "epoch": 0.7217227792506362,
      "grad_norm": 16.25,
      "learning_rate": 3.7943726888180264e-06,
      "loss": 0.8538,
      "num_input_tokens_seen": 4028159488,
      "step": 54310,
      "train_runtime": 64625.996,
      "train_tokens_per_second": 62330.327
    },
    {
      "epoch": 0.7218556687330981,
      "grad_norm": 25.375,
      "learning_rate": 3.7909982807996905e-06,
      "loss": 0.941,
      "num_input_tokens_seen": 4029024768,
      "step": 54320,
      "train_runtime": 64638.8753,
      "train_tokens_per_second": 62331.294
    },
    {
      "epoch": 0.7219885582155601,
      "grad_norm": 19.5,
      "learning_rate": 3.7876250229435654e-06,
      "loss": 0.7918,
      "num_input_tokens_seen": 4029500160,
      "step": 54330,
      "train_runtime": 64645.8623,
      "train_tokens_per_second": 62331.911
    },
    {
      "epoch": 0.7221214476980219,
      "grad_norm": 14.1875,
      "learning_rate": 3.7842529158745055e-06,
      "loss": 0.6027,
      "num_input_tokens_seen": 4029737984,
      "step": 54340,
      "train_runtime": 64649.4473,
      "train_tokens_per_second": 62332.134
    },
    {
      "epoch": 0.7222543371804838,
      "grad_norm": 35.75,
      "learning_rate": 3.7808819602171685e-06,
      "loss": 0.4878,
      "num_input_tokens_seen": 4029856512,
      "step": 54350,
      "train_runtime": 64651.9085,
      "train_tokens_per_second": 62331.594
    },
    {
      "epoch": 0.7223872266629457,
      "grad_norm": 50.25,
      "learning_rate": 3.777512156595994e-06,
      "loss": 0.902,
      "num_input_tokens_seen": 4031908352,
      "step": 54360,
      "train_runtime": 64683.3688,
      "train_tokens_per_second": 62332.999
    },
    {
      "epoch": 0.7225201161454077,
      "grad_norm": 16.75,
      "learning_rate": 3.774143505635207e-06,
      "loss": 0.8851,
      "num_input_tokens_seen": 4032729088,
      "step": 54370,
      "train_runtime": 64696.022,
      "train_tokens_per_second": 62333.494
    },
    {
      "epoch": 0.7226530056278696,
      "grad_norm": 21.0,
      "learning_rate": 3.770776007958824e-06,
      "loss": 0.7147,
      "num_input_tokens_seen": 4033210368,
      "step": 54380,
      "train_runtime": 64703.4171,
      "train_tokens_per_second": 62333.808
    },
    {
      "epoch": 0.7227858951103315,
      "grad_norm": 19.75,
      "learning_rate": 3.7674096641906345e-06,
      "loss": 0.6738,
      "num_input_tokens_seen": 4033440512,
      "step": 54390,
      "train_runtime": 64706.8906,
      "train_tokens_per_second": 62334.018
    },
    {
      "epoch": 0.7229187845927934,
      "grad_norm": 20.875,
      "learning_rate": 3.764044474954236e-06,
      "loss": 0.5539,
      "num_input_tokens_seen": 4033554432,
      "step": 54400,
      "train_runtime": 64709.288,
      "train_tokens_per_second": 62333.47
    },
    {
      "epoch": 0.7230516740752553,
      "grad_norm": 17.5,
      "learning_rate": 3.760680440872989e-06,
      "loss": 0.785,
      "num_input_tokens_seen": 4035563008,
      "step": 54410,
      "train_runtime": 64741.5999,
      "train_tokens_per_second": 62333.384
    },
    {
      "epoch": 0.7231845635577172,
      "grad_norm": 43.75,
      "learning_rate": 3.7573175625700574e-06,
      "loss": 0.9049,
      "num_input_tokens_seen": 4036416000,
      "step": 54420,
      "train_runtime": 64754.8475,
      "train_tokens_per_second": 62333.804
    },
    {
      "epoch": 0.7233174530401791,
      "grad_norm": 23.25,
      "learning_rate": 3.753955840668385e-06,
      "loss": 0.7029,
      "num_input_tokens_seen": 4036882432,
      "step": 54430,
      "train_runtime": 64761.293,
      "train_tokens_per_second": 62334.803
    },
    {
      "epoch": 0.7234503425226411,
      "grad_norm": 38.25,
      "learning_rate": 3.7505952757906905e-06,
      "loss": 0.661,
      "num_input_tokens_seen": 4037107456,
      "step": 54440,
      "train_runtime": 64764.6578,
      "train_tokens_per_second": 62335.039
    },
    {
      "epoch": 0.723583232005103,
      "grad_norm": 16.75,
      "learning_rate": 3.7472358685595046e-06,
      "loss": 0.5605,
      "num_input_tokens_seen": 4037222656,
      "step": 54450,
      "train_runtime": 64767.0378,
      "train_tokens_per_second": 62334.527
    },
    {
      "epoch": 0.7237161214875648,
      "grad_norm": 22.375,
      "learning_rate": 3.743877619597115e-06,
      "loss": 0.8367,
      "num_input_tokens_seen": 4039252992,
      "step": 54460,
      "train_runtime": 64800.0771,
      "train_tokens_per_second": 62334.077
    },
    {
      "epoch": 0.7238490109700267,
      "grad_norm": 440.0,
      "learning_rate": 3.740520529525611e-06,
      "loss": 0.9588,
      "num_input_tokens_seen": 4040100096,
      "step": 54470,
      "train_runtime": 64812.8704,
      "train_tokens_per_second": 62334.843
    },
    {
      "epoch": 0.7239819004524887,
      "grad_norm": 55.25,
      "learning_rate": 3.737164598966865e-06,
      "loss": 0.7011,
      "num_input_tokens_seen": 4040574208,
      "step": 54480,
      "train_runtime": 64819.6357,
      "train_tokens_per_second": 62335.651
    },
    {
      "epoch": 0.7241147899349506,
      "grad_norm": 10.8125,
      "learning_rate": 3.7338098285425316e-06,
      "loss": 0.6319,
      "num_input_tokens_seen": 4040798464,
      "step": 54490,
      "train_runtime": 64823.8846,
      "train_tokens_per_second": 62335.025
    },
    {
      "epoch": 0.7242476794174125,
      "grad_norm": 12.3125,
      "learning_rate": 3.7304562188740566e-06,
      "loss": 0.4965,
      "num_input_tokens_seen": 4040913920,
      "step": 54500,
      "train_runtime": 64826.274,
      "train_tokens_per_second": 62334.508
    },
    {
      "epoch": 0.7243805688998745,
      "grad_norm": 21.0,
      "learning_rate": 3.7271037705826598e-06,
      "loss": 0.8314,
      "num_input_tokens_seen": 4043037440,
      "step": 54510,
      "train_runtime": 64860.8861,
      "train_tokens_per_second": 62333.984
    },
    {
      "epoch": 0.7245134583823364,
      "grad_norm": 34.25,
      "learning_rate": 3.7237524842893548e-06,
      "loss": 0.8843,
      "num_input_tokens_seen": 4043859968,
      "step": 54520,
      "train_runtime": 64873.4036,
      "train_tokens_per_second": 62334.636
    },
    {
      "epoch": 0.7246463478647982,
      "grad_norm": 67.5,
      "learning_rate": 3.7204023606149375e-06,
      "loss": 0.6919,
      "num_input_tokens_seen": 4044318976,
      "step": 54530,
      "train_runtime": 64880.5562,
      "train_tokens_per_second": 62334.838
    },
    {
      "epoch": 0.7247792373472601,
      "grad_norm": 14.0,
      "learning_rate": 3.7170534001799885e-06,
      "loss": 0.6882,
      "num_input_tokens_seen": 4044537344,
      "step": 54540,
      "train_runtime": 64883.9614,
      "train_tokens_per_second": 62334.932
    },
    {
      "epoch": 0.7249121268297221,
      "grad_norm": 8.75,
      "learning_rate": 3.7137056036048737e-06,
      "loss": 0.519,
      "num_input_tokens_seen": 4044652032,
      "step": 54550,
      "train_runtime": 64886.3264,
      "train_tokens_per_second": 62334.428
    },
    {
      "epoch": 0.725045016312184,
      "grad_norm": 113.5,
      "learning_rate": 3.7103589715097423e-06,
      "loss": 0.845,
      "num_input_tokens_seen": 4046704128,
      "step": 54560,
      "train_runtime": 64919.5838,
      "train_tokens_per_second": 62334.105
    },
    {
      "epoch": 0.7251779057946459,
      "grad_norm": 12.75,
      "learning_rate": 3.7070135045145307e-06,
      "loss": 0.9316,
      "num_input_tokens_seen": 4047555072,
      "step": 54570,
      "train_runtime": 64932.1723,
      "train_tokens_per_second": 62335.125
    },
    {
      "epoch": 0.7253107952771078,
      "grad_norm": 19.25,
      "learning_rate": 3.7036692032389497e-06,
      "loss": 0.7351,
      "num_input_tokens_seen": 4048008960,
      "step": 54580,
      "train_runtime": 64938.6693,
      "train_tokens_per_second": 62335.878
    },
    {
      "epoch": 0.7254436847595697,
      "grad_norm": 11.25,
      "learning_rate": 3.7003260683025065e-06,
      "loss": 0.5702,
      "num_input_tokens_seen": 4048237824,
      "step": 54590,
      "train_runtime": 64942.0749,
      "train_tokens_per_second": 62336.133
    },
    {
      "epoch": 0.7255765742420316,
      "grad_norm": 19.5,
      "learning_rate": 3.6969841003244855e-06,
      "loss": 0.5786,
      "num_input_tokens_seen": 4048351488,
      "step": 54600,
      "train_runtime": 64944.4661,
      "train_tokens_per_second": 62335.588
    },
    {
      "epoch": 0.7257094637244935,
      "grad_norm": 19.125,
      "learning_rate": 3.693643299923957e-06,
      "loss": 0.7979,
      "num_input_tokens_seen": 4050331648,
      "step": 54610,
      "train_runtime": 64976.1819,
      "train_tokens_per_second": 62335.636
    },
    {
      "epoch": 0.7258423532069554,
      "grad_norm": 24.625,
      "learning_rate": 3.6903036677197746e-06,
      "loss": 0.884,
      "num_input_tokens_seen": 4051231232,
      "step": 54620,
      "train_runtime": 64988.7651,
      "train_tokens_per_second": 62337.409
    },
    {
      "epoch": 0.7259752426894174,
      "grad_norm": 18.5,
      "learning_rate": 3.6869652043305783e-06,
      "loss": 0.7871,
      "num_input_tokens_seen": 4051698432,
      "step": 54630,
      "train_runtime": 64995.4847,
      "train_tokens_per_second": 62338.152
    },
    {
      "epoch": 0.7261081321718793,
      "grad_norm": 19.375,
      "learning_rate": 3.6836279103747786e-06,
      "loss": 0.5979,
      "num_input_tokens_seen": 4051930112,
      "step": 54640,
      "train_runtime": 64999.0454,
      "train_tokens_per_second": 62338.302
    },
    {
      "epoch": 0.7262410216543411,
      "grad_norm": 23.875,
      "learning_rate": 3.6802917864705944e-06,
      "loss": 0.4221,
      "num_input_tokens_seen": 4052048384,
      "step": 54650,
      "train_runtime": 65002.2418,
      "train_tokens_per_second": 62337.056
    },
    {
      "epoch": 0.726373911136803,
      "grad_norm": 15.875,
      "learning_rate": 3.6769568332360004e-06,
      "loss": 0.7707,
      "num_input_tokens_seen": 4054164736,
      "step": 54660,
      "train_runtime": 65036.0264,
      "train_tokens_per_second": 62337.215
    },
    {
      "epoch": 0.726506800619265,
      "grad_norm": 16.5,
      "learning_rate": 3.6736230512887715e-06,
      "loss": 0.8845,
      "num_input_tokens_seen": 4055028480,
      "step": 54670,
      "train_runtime": 65048.7114,
      "train_tokens_per_second": 62338.337
    },
    {
      "epoch": 0.7266396901017269,
      "grad_norm": 32.75,
      "learning_rate": 3.6702904412464656e-06,
      "loss": 0.7594,
      "num_input_tokens_seen": 4055462144,
      "step": 54680,
      "train_runtime": 65055.1414,
      "train_tokens_per_second": 62338.841
    },
    {
      "epoch": 0.7267725795841888,
      "grad_norm": 14.125,
      "learning_rate": 3.666959003726407e-06,
      "loss": 0.5689,
      "num_input_tokens_seen": 4055685120,
      "step": 54690,
      "train_runtime": 65058.5167,
      "train_tokens_per_second": 62339.035
    },
    {
      "epoch": 0.7269054690666508,
      "grad_norm": 21.5,
      "learning_rate": 3.6636287393457293e-06,
      "loss": 0.5545,
      "num_input_tokens_seen": 4055798272,
      "step": 54700,
      "train_runtime": 65060.9878,
      "train_tokens_per_second": 62338.406
    },
    {
      "epoch": 0.7270383585491126,
      "grad_norm": 27.625,
      "learning_rate": 3.6602996487213205e-06,
      "loss": 0.8192,
      "num_input_tokens_seen": 4057750272,
      "step": 54710,
      "train_runtime": 65092.6198,
      "train_tokens_per_second": 62338.1
    },
    {
      "epoch": 0.7271712480315745,
      "grad_norm": 16.125,
      "learning_rate": 3.6569717324698782e-06,
      "loss": 0.8883,
      "num_input_tokens_seen": 4058599680,
      "step": 54720,
      "train_runtime": 65105.1137,
      "train_tokens_per_second": 62339.184
    },
    {
      "epoch": 0.7273041375140364,
      "grad_norm": 11.125,
      "learning_rate": 3.653644991207859e-06,
      "loss": 0.7083,
      "num_input_tokens_seen": 4059067136,
      "step": 54730,
      "train_runtime": 65112.4778,
      "train_tokens_per_second": 62339.313
    },
    {
      "epoch": 0.7274370269964984,
      "grad_norm": 18.875,
      "learning_rate": 3.6503194255515164e-06,
      "loss": 0.7037,
      "num_input_tokens_seen": 4059292416,
      "step": 54740,
      "train_runtime": 65115.8112,
      "train_tokens_per_second": 62339.581
    },
    {
      "epoch": 0.7275699164789603,
      "grad_norm": 21.875,
      "learning_rate": 3.646995036116884e-06,
      "loss": 0.5153,
      "num_input_tokens_seen": 4059404032,
      "step": 54750,
      "train_runtime": 65118.1592,
      "train_tokens_per_second": 62339.048
    },
    {
      "epoch": 0.7277028059614222,
      "grad_norm": 130.0,
      "learning_rate": 3.6436718235197664e-06,
      "loss": 0.7878,
      "num_input_tokens_seen": 4061410048,
      "step": 54760,
      "train_runtime": 65150.533,
      "train_tokens_per_second": 62338.861
    },
    {
      "epoch": 0.727835695443884,
      "grad_norm": 26.375,
      "learning_rate": 3.6403497883757712e-06,
      "loss": 0.8571,
      "num_input_tokens_seen": 4062247936,
      "step": 54770,
      "train_runtime": 65163.0842,
      "train_tokens_per_second": 62339.713
    },
    {
      "epoch": 0.727968584926346,
      "grad_norm": 11.375,
      "learning_rate": 3.637028931300267e-06,
      "loss": 0.7436,
      "num_input_tokens_seen": 4062720512,
      "step": 54780,
      "train_runtime": 65169.6248,
      "train_tokens_per_second": 62340.707
    },
    {
      "epoch": 0.7281014744088079,
      "grad_norm": 20.125,
      "learning_rate": 3.633709252908415e-06,
      "loss": 0.6284,
      "num_input_tokens_seen": 4062944768,
      "step": 54790,
      "train_runtime": 65173.8428,
      "train_tokens_per_second": 62340.114
    },
    {
      "epoch": 0.7282343638912698,
      "grad_norm": 13.4375,
      "learning_rate": 3.6303907538151572e-06,
      "loss": 0.4622,
      "num_input_tokens_seen": 4063056128,
      "step": 54800,
      "train_runtime": 65176.3306,
      "train_tokens_per_second": 62339.443
    },
    {
      "epoch": 0.7283672533737318,
      "grad_norm": 47.0,
      "learning_rate": 3.6270734346352154e-06,
      "loss": 0.8346,
      "num_input_tokens_seen": 4065114112,
      "step": 54810,
      "train_runtime": 65210.0117,
      "train_tokens_per_second": 62338.804
    },
    {
      "epoch": 0.7285001428561937,
      "grad_norm": 17.0,
      "learning_rate": 3.623757295983097e-06,
      "loss": 0.9357,
      "num_input_tokens_seen": 4065954816,
      "step": 54820,
      "train_runtime": 65222.9978,
      "train_tokens_per_second": 62339.281
    },
    {
      "epoch": 0.7286330323386555,
      "grad_norm": 17.5,
      "learning_rate": 3.6204423384730803e-06,
      "loss": 0.7344,
      "num_input_tokens_seen": 4066452224,
      "step": 54830,
      "train_runtime": 65229.8517,
      "train_tokens_per_second": 62340.357
    },
    {
      "epoch": 0.7287659218211174,
      "grad_norm": 18.625,
      "learning_rate": 3.6171285627192344e-06,
      "loss": 0.5733,
      "num_input_tokens_seen": 4066687232,
      "step": 54840,
      "train_runtime": 65234.3553,
      "train_tokens_per_second": 62339.655
    },
    {
      "epoch": 0.7288988113035794,
      "grad_norm": 7.53125,
      "learning_rate": 3.613815969335408e-06,
      "loss": 0.5108,
      "num_input_tokens_seen": 4066799872,
      "step": 54850,
      "train_runtime": 65236.7609,
      "train_tokens_per_second": 62339.083
    },
    {
      "epoch": 0.7290317007860413,
      "grad_norm": 25.0,
      "learning_rate": 3.6105045589352284e-06,
      "loss": 0.8581,
      "num_input_tokens_seen": 4068936192,
      "step": 54860,
      "train_runtime": 65270.5935,
      "train_tokens_per_second": 62339.5
    },
    {
      "epoch": 0.7291645902685032,
      "grad_norm": 20.0,
      "learning_rate": 3.6071943321321055e-06,
      "loss": 0.9635,
      "num_input_tokens_seen": 4069791744,
      "step": 54870,
      "train_runtime": 65283.4191,
      "train_tokens_per_second": 62340.358
    },
    {
      "epoch": 0.7292974797509651,
      "grad_norm": 23.5,
      "learning_rate": 3.6038852895392286e-06,
      "loss": 0.6852,
      "num_input_tokens_seen": 4070252288,
      "step": 54880,
      "train_runtime": 65290.0282,
      "train_tokens_per_second": 62341.102
    },
    {
      "epoch": 0.729430369233427,
      "grad_norm": 13.5,
      "learning_rate": 3.600577431769572e-06,
      "loss": 0.6214,
      "num_input_tokens_seen": 4070474752,
      "step": 54890,
      "train_runtime": 65294.244,
      "train_tokens_per_second": 62340.484
    },
    {
      "epoch": 0.7295632587158889,
      "grad_norm": 26.125,
      "learning_rate": 3.5972707594358804e-06,
      "loss": 0.6066,
      "num_input_tokens_seen": 4070590976,
      "step": 54900,
      "train_runtime": 65296.6716,
      "train_tokens_per_second": 62339.946
    },
    {
      "epoch": 0.7296961481983508,
      "grad_norm": 40.25,
      "learning_rate": 3.593965273150688e-06,
      "loss": 0.7596,
      "num_input_tokens_seen": 4072534528,
      "step": 54910,
      "train_runtime": 65329.3083,
      "train_tokens_per_second": 62338.553
    },
    {
      "epoch": 0.7298290376808128,
      "grad_norm": 23.5,
      "learning_rate": 3.5906609735263077e-06,
      "loss": 0.8444,
      "num_input_tokens_seen": 4073381888,
      "step": 54920,
      "train_runtime": 65342.1392,
      "train_tokens_per_second": 62339.28
    },
    {
      "epoch": 0.7299619271632747,
      "grad_norm": 56.25,
      "learning_rate": 3.5873578611748307e-06,
      "loss": 0.7028,
      "num_input_tokens_seen": 4073848064,
      "step": 54930,
      "train_runtime": 65348.71,
      "train_tokens_per_second": 62340.145
    },
    {
      "epoch": 0.7300948166457366,
      "grad_norm": 16.75,
      "learning_rate": 3.5840559367081297e-06,
      "loss": 0.586,
      "num_input_tokens_seen": 4074074368,
      "step": 54940,
      "train_runtime": 65352.1118,
      "train_tokens_per_second": 62340.363
    },
    {
      "epoch": 0.7302277061281984,
      "grad_norm": 63.75,
      "learning_rate": 3.58075520073786e-06,
      "loss": 0.5657,
      "num_input_tokens_seen": 4074193920,
      "step": 54950,
      "train_runtime": 65354.5123,
      "train_tokens_per_second": 62339.902
    },
    {
      "epoch": 0.7303605956106604,
      "grad_norm": 18.25,
      "learning_rate": 3.5774556538754433e-06,
      "loss": 0.7814,
      "num_input_tokens_seen": 4076355072,
      "step": 54960,
      "train_runtime": 65388.8241,
      "train_tokens_per_second": 62340.241
    },
    {
      "epoch": 0.7304934850931223,
      "grad_norm": 28.5,
      "learning_rate": 3.5741572967321047e-06,
      "loss": 0.9698,
      "num_input_tokens_seen": 4077192448,
      "step": 54970,
      "train_runtime": 65402.5114,
      "train_tokens_per_second": 62339.998
    },
    {
      "epoch": 0.7306263745755842,
      "grad_norm": 13.8125,
      "learning_rate": 3.570860129918825e-06,
      "loss": 0.6997,
      "num_input_tokens_seen": 4077673472,
      "step": 54980,
      "train_runtime": 65409.4007,
      "train_tokens_per_second": 62340.786
    },
    {
      "epoch": 0.7307592640580461,
      "grad_norm": 21.5,
      "learning_rate": 3.567564154046379e-06,
      "loss": 0.6688,
      "num_input_tokens_seen": 4077899264,
      "step": 54990,
      "train_runtime": 65412.8905,
      "train_tokens_per_second": 62340.912
    },
    {
      "epoch": 0.7308921535405081,
      "grad_norm": 17.375,
      "learning_rate": 3.5642693697253195e-06,
      "loss": 0.5019,
      "num_input_tokens_seen": 4078014208,
      "step": 55000,
      "train_runtime": 65415.3584,
      "train_tokens_per_second": 62340.317
    },
    {
      "epoch": 0.7308921535405081,
      "eval_accuracy": 0.819,
      "eval_balanced_accuracy": 0.8171018557168968,
      "eval_invalid_f1": 0.836907550910074,
      "eval_invalid_precision": 0.8400868306801736,
      "eval_invalid_recall": 0.8337522441651706,
      "eval_loss": 0.36139267683029175,
      "eval_macro_f1": 0.8167912277129565,
      "eval_mcc": 0.6336119001185314,
      "eval_runtime": 78.4173,
      "eval_samples_per_second": 127.523,
      "eval_steps_per_second": 31.881,
      "eval_valid_f1": 0.7966749045158391,
      "eval_valid_precision": 0.7929338103756708,
      "eval_valid_recall": 0.800451467268623,
      "num_input_tokens_seen": 4078014208,
      "step": 55000
    },
    {
      "epoch": 0.73102504302297,
      "grad_norm": 33.75,
      "learning_rate": 3.560975777565966e-06,
      "loss": 0.8275,
      "num_input_tokens_seen": 4080075776,
      "step": 55010,
      "train_runtime": 65531.3856,
      "train_tokens_per_second": 62261.4
    },
    {
      "epoch": 0.7311579325054318,
      "grad_norm": 38.5,
      "learning_rate": 3.5576833781784414e-06,
      "loss": 0.9911,
      "num_input_tokens_seen": 4080919552,
      "step": 55020,
      "train_runtime": 65544.7909,
      "train_tokens_per_second": 62261.539
    },
    {
      "epoch": 0.7312908219878937,
      "grad_norm": 15.0,
      "learning_rate": 3.5543921721726217e-06,
      "loss": 0.7602,
      "num_input_tokens_seen": 4081390336,
      "step": 55030,
      "train_runtime": 65551.5614,
      "train_tokens_per_second": 62262.29
    },
    {
      "epoch": 0.7314237114703557,
      "grad_norm": 25.75,
      "learning_rate": 3.5511021601581787e-06,
      "loss": 0.6313,
      "num_input_tokens_seen": 4081625600,
      "step": 55040,
      "train_runtime": 65555.0741,
      "train_tokens_per_second": 62262.543
    },
    {
      "epoch": 0.7315566009528176,
      "grad_norm": 17.125,
      "learning_rate": 3.5478133427445594e-06,
      "loss": 0.5378,
      "num_input_tokens_seen": 4081742080,
      "step": 55050,
      "train_runtime": 65557.5327,
      "train_tokens_per_second": 62261.984
    },
    {
      "epoch": 0.7316894904352795,
      "grad_norm": 22.25,
      "learning_rate": 3.544525720540979e-06,
      "loss": 0.8299,
      "num_input_tokens_seen": 4083844352,
      "step": 55060,
      "train_runtime": 65591.6857,
      "train_tokens_per_second": 62261.616
    },
    {
      "epoch": 0.7318223799177415,
      "grad_norm": 69.5,
      "learning_rate": 3.541239294156452e-06,
      "loss": 0.9097,
      "num_input_tokens_seen": 4084705536,
      "step": 55070,
      "train_runtime": 65605.4575,
      "train_tokens_per_second": 62261.673
    },
    {
      "epoch": 0.7319552694002033,
      "grad_norm": 15.75,
      "learning_rate": 3.537954064199749e-06,
      "loss": 0.7459,
      "num_input_tokens_seen": 4085233920,
      "step": 55080,
      "train_runtime": 65612.4952,
      "train_tokens_per_second": 62263.048
    },
    {
      "epoch": 0.7320881588826652,
      "grad_norm": 12.0,
      "learning_rate": 3.53467003127944e-06,
      "loss": 0.5975,
      "num_input_tokens_seen": 4085466112,
      "step": 55090,
      "train_runtime": 65616.0072,
      "train_tokens_per_second": 62263.254
    },
    {
      "epoch": 0.7322210483651271,
      "grad_norm": 10.8125,
      "learning_rate": 3.5313871960038548e-06,
      "loss": 0.5088,
      "num_input_tokens_seen": 4085580032,
      "step": 55100,
      "train_runtime": 65618.5142,
      "train_tokens_per_second": 62262.611
    },
    {
      "epoch": 0.7323539378475891,
      "grad_norm": 23.375,
      "learning_rate": 3.528105558981112e-06,
      "loss": 0.8379,
      "num_input_tokens_seen": 4087627008,
      "step": 55110,
      "train_runtime": 65651.3183,
      "train_tokens_per_second": 62262.68
    },
    {
      "epoch": 0.732486827330051,
      "grad_norm": 19.0,
      "learning_rate": 3.524825120819105e-06,
      "loss": 0.8888,
      "num_input_tokens_seen": 4088458496,
      "step": 55120,
      "train_runtime": 65664.7183,
      "train_tokens_per_second": 62262.637
    },
    {
      "epoch": 0.7326197168125129,
      "grad_norm": 15.25,
      "learning_rate": 3.5215458821255077e-06,
      "loss": 0.799,
      "num_input_tokens_seen": 4088921856,
      "step": 55130,
      "train_runtime": 65671.2904,
      "train_tokens_per_second": 62263.461
    },
    {
      "epoch": 0.7327526062949747,
      "grad_norm": 24.125,
      "learning_rate": 3.518267843507771e-06,
      "loss": 0.7473,
      "num_input_tokens_seen": 4089156608,
      "step": 55140,
      "train_runtime": 65674.791,
      "train_tokens_per_second": 62263.717
    },
    {
      "epoch": 0.7328854957774367,
      "grad_norm": 40.5,
      "learning_rate": 3.514991005573116e-06,
      "loss": 0.6014,
      "num_input_tokens_seen": 4089273856,
      "step": 55150,
      "train_runtime": 65677.1747,
      "train_tokens_per_second": 62263.243
    },
    {
      "epoch": 0.7330183852598986,
      "grad_norm": 21.875,
      "learning_rate": 3.511715368928551e-06,
      "loss": 0.8029,
      "num_input_tokens_seen": 4091395328,
      "step": 55160,
      "train_runtime": 65711.1961,
      "train_tokens_per_second": 62263.291
    },
    {
      "epoch": 0.7331512747423605,
      "grad_norm": 30.125,
      "learning_rate": 3.50844093418086e-06,
      "loss": 0.9379,
      "num_input_tokens_seen": 4092276736,
      "step": 55170,
      "train_runtime": 65724.4513,
      "train_tokens_per_second": 62264.145
    },
    {
      "epoch": 0.7332841642248225,
      "grad_norm": 17.375,
      "learning_rate": 3.5051677019366016e-06,
      "loss": 0.7361,
      "num_input_tokens_seen": 4092771072,
      "step": 55180,
      "train_runtime": 65731.4967,
      "train_tokens_per_second": 62264.991
    },
    {
      "epoch": 0.7334170537072844,
      "grad_norm": 25.625,
      "learning_rate": 3.5018956728021125e-06,
      "loss": 0.6958,
      "num_input_tokens_seen": 4093009152,
      "step": 55190,
      "train_runtime": 65736.4509,
      "train_tokens_per_second": 62263.92
    },
    {
      "epoch": 0.7335499431897462,
      "grad_norm": 15.875,
      "learning_rate": 3.4986248473835114e-06,
      "loss": 0.6485,
      "num_input_tokens_seen": 4093127424,
      "step": 55200,
      "train_runtime": 65738.8688,
      "train_tokens_per_second": 62263.429
    },
    {
      "epoch": 0.7336828326722081,
      "grad_norm": 15.5,
      "learning_rate": 3.4953552262866817e-06,
      "loss": 0.8065,
      "num_input_tokens_seen": 4095265792,
      "step": 55210,
      "train_runtime": 65773.2881,
      "train_tokens_per_second": 62263.358
    },
    {
      "epoch": 0.7338157221546701,
      "grad_norm": 15.5,
      "learning_rate": 3.4920868101172965e-06,
      "loss": 0.8449,
      "num_input_tokens_seen": 4096109824,
      "step": 55220,
      "train_runtime": 65785.8451,
      "train_tokens_per_second": 62264.303
    },
    {
      "epoch": 0.733948611637132,
      "grad_norm": 13.25,
      "learning_rate": 3.4888195994807984e-06,
      "loss": 0.7479,
      "num_input_tokens_seen": 4096578048,
      "step": 55230,
      "train_runtime": 65792.3901,
      "train_tokens_per_second": 62265.226
    },
    {
      "epoch": 0.7340815011195939,
      "grad_norm": 23.25,
      "learning_rate": 3.4855535949824114e-06,
      "loss": 0.6266,
      "num_input_tokens_seen": 4096804608,
      "step": 55240,
      "train_runtime": 65796.9916,
      "train_tokens_per_second": 62264.315
    },
    {
      "epoch": 0.7342143906020558,
      "grad_norm": 18.875,
      "learning_rate": 3.4822887972271323e-06,
      "loss": 0.5627,
      "num_input_tokens_seen": 4096920832,
      "step": 55250,
      "train_runtime": 65799.488,
      "train_tokens_per_second": 62263.719
    },
    {
      "epoch": 0.7343472800845177,
      "grad_norm": 18.5,
      "learning_rate": 3.4790252068197363e-06,
      "loss": 0.8441,
      "num_input_tokens_seen": 4098844928,
      "step": 55260,
      "train_runtime": 65830.4462,
      "train_tokens_per_second": 62263.666
    },
    {
      "epoch": 0.7344801695669796,
      "grad_norm": 26.75,
      "learning_rate": 3.4757628243647778e-06,
      "loss": 0.8641,
      "num_input_tokens_seen": 4099671296,
      "step": 55270,
      "train_runtime": 65842.6981,
      "train_tokens_per_second": 62264.631
    },
    {
      "epoch": 0.7346130590494415,
      "grad_norm": 16.625,
      "learning_rate": 3.4725016504665733e-06,
      "loss": 0.7381,
      "num_input_tokens_seen": 4100123648,
      "step": 55280,
      "train_runtime": 65849.3064,
      "train_tokens_per_second": 62265.252
    },
    {
      "epoch": 0.7347459485319034,
      "grad_norm": 21.75,
      "learning_rate": 3.46924168572924e-06,
      "loss": 0.7303,
      "num_input_tokens_seen": 4100354816,
      "step": 55290,
      "train_runtime": 65854.9731,
      "train_tokens_per_second": 62263.404
    },
    {
      "epoch": 0.7348788380143654,
      "grad_norm": 22.125,
      "learning_rate": 3.4659829307566482e-06,
      "loss": 0.5326,
      "num_input_tokens_seen": 4100471296,
      "step": 55300,
      "train_runtime": 65857.4727,
      "train_tokens_per_second": 62262.81
    },
    {
      "epoch": 0.7350117274968273,
      "grad_norm": 15.625,
      "learning_rate": 3.4627253861524544e-06,
      "loss": 0.7713,
      "num_input_tokens_seen": 4102606592,
      "step": 55310,
      "train_runtime": 65892.2261,
      "train_tokens_per_second": 62262.377
    },
    {
      "epoch": 0.7351446169792891,
      "grad_norm": 34.5,
      "learning_rate": 3.4594690525200956e-06,
      "loss": 0.8629,
      "num_input_tokens_seen": 4103443968,
      "step": 55320,
      "train_runtime": 65905.1777,
      "train_tokens_per_second": 62262.847
    },
    {
      "epoch": 0.735277506461751,
      "grad_norm": 22.75,
      "learning_rate": 3.4562139304627675e-06,
      "loss": 0.667,
      "num_input_tokens_seen": 4103926784,
      "step": 55330,
      "train_runtime": 65912.0648,
      "train_tokens_per_second": 62263.666
    },
    {
      "epoch": 0.735410395944213,
      "grad_norm": 46.5,
      "learning_rate": 3.4529600205834647e-06,
      "loss": 0.5774,
      "num_input_tokens_seen": 4104156416,
      "step": 55340,
      "train_runtime": 65915.5582,
      "train_tokens_per_second": 62263.85
    },
    {
      "epoch": 0.7355432854266749,
      "grad_norm": 7.5,
      "learning_rate": 3.4497073234849367e-06,
      "loss": 0.5035,
      "num_input_tokens_seen": 4104270848,
      "step": 55350,
      "train_runtime": 65917.9707,
      "train_tokens_per_second": 62263.307
    },
    {
      "epoch": 0.7356761749091368,
      "grad_norm": 21.0,
      "learning_rate": 3.446455839769718e-06,
      "loss": 0.8591,
      "num_input_tokens_seen": 4106332672,
      "step": 55360,
      "train_runtime": 65950.667,
      "train_tokens_per_second": 62263.702
    },
    {
      "epoch": 0.7358090643915988,
      "grad_norm": 15.125,
      "learning_rate": 3.4432055700401223e-06,
      "loss": 0.8552,
      "num_input_tokens_seen": 4107158528,
      "step": 55370,
      "train_runtime": 65964.1952,
      "train_tokens_per_second": 62263.452
    },
    {
      "epoch": 0.7359419538740606,
      "grad_norm": 19.375,
      "learning_rate": 3.439956514898222e-06,
      "loss": 0.7439,
      "num_input_tokens_seen": 4107615232,
      "step": 55380,
      "train_runtime": 65970.8676,
      "train_tokens_per_second": 62264.078
    },
    {
      "epoch": 0.7360748433565225,
      "grad_norm": 24.25,
      "learning_rate": 3.436708674945889e-06,
      "loss": 0.6518,
      "num_input_tokens_seen": 4107845888,
      "step": 55390,
      "train_runtime": 65974.4194,
      "train_tokens_per_second": 62264.222
    },
    {
      "epoch": 0.7362077328389844,
      "grad_norm": 14.3125,
      "learning_rate": 3.433462050784744e-06,
      "loss": 0.5729,
      "num_input_tokens_seen": 4107959040,
      "step": 55400,
      "train_runtime": 65976.8414,
      "train_tokens_per_second": 62263.651
    },
    {
      "epoch": 0.7363406223214464,
      "grad_norm": 19.125,
      "learning_rate": 3.4302166430162077e-06,
      "loss": 0.8139,
      "num_input_tokens_seen": 4110034944,
      "step": 55410,
      "train_runtime": 66010.2548,
      "train_tokens_per_second": 62263.583
    },
    {
      "epoch": 0.7364735118039083,
      "grad_norm": 22.375,
      "learning_rate": 3.4269724522414527e-06,
      "loss": 0.8087,
      "num_input_tokens_seen": 4110867968,
      "step": 55420,
      "train_runtime": 66022.8899,
      "train_tokens_per_second": 62264.284
    },
    {
      "epoch": 0.7366064012863702,
      "grad_norm": 206.0,
      "learning_rate": 3.423729479061442e-06,
      "loss": 0.7348,
      "num_input_tokens_seen": 4111315200,
      "step": 55430,
      "train_runtime": 66030.7654,
      "train_tokens_per_second": 62263.631
    },
    {
      "epoch": 0.7367392907688322,
      "grad_norm": 15.3125,
      "learning_rate": 3.4204877240769043e-06,
      "loss": 0.6263,
      "num_input_tokens_seen": 4111553024,
      "step": 55440,
      "train_runtime": 66034.2746,
      "train_tokens_per_second": 62263.924
    },
    {
      "epoch": 0.736872180251294,
      "grad_norm": 18.25,
      "learning_rate": 3.4172471878883472e-06,
      "loss": 0.5759,
      "num_input_tokens_seen": 4111670528,
      "step": 55450,
      "train_runtime": 66036.8205,
      "train_tokens_per_second": 62263.302
    },
    {
      "epoch": 0.7370050697337559,
      "grad_norm": 18.375,
      "learning_rate": 3.4140078710960558e-06,
      "loss": 0.7961,
      "num_input_tokens_seen": 4113660416,
      "step": 55460,
      "train_runtime": 66068.6662,
      "train_tokens_per_second": 62263.409
    },
    {
      "epoch": 0.7371379592162178,
      "grad_norm": 16.25,
      "learning_rate": 3.4107697743000757e-06,
      "loss": 0.8487,
      "num_input_tokens_seen": 4114500608,
      "step": 55470,
      "train_runtime": 66081.1647,
      "train_tokens_per_second": 62264.348
    },
    {
      "epoch": 0.7372708486986798,
      "grad_norm": 14.5,
      "learning_rate": 3.4075328981002397e-06,
      "loss": 0.7586,
      "num_input_tokens_seen": 4114947328,
      "step": 55480,
      "train_runtime": 66088.6458,
      "train_tokens_per_second": 62264.059
    },
    {
      "epoch": 0.7374037381811417,
      "grad_norm": 28.125,
      "learning_rate": 3.4042972430961505e-06,
      "loss": 0.7189,
      "num_input_tokens_seen": 4115166208,
      "step": 55490,
      "train_runtime": 66092.0649,
      "train_tokens_per_second": 62264.149
    },
    {
      "epoch": 0.7375366276636036,
      "grad_norm": 36.0,
      "learning_rate": 3.4010628098871844e-06,
      "loss": 0.5724,
      "num_input_tokens_seen": 4115279616,
      "step": 55500,
      "train_runtime": 66094.5019,
      "train_tokens_per_second": 62263.57
    },
    {
      "epoch": 0.7376695171460654,
      "grad_norm": 27.625,
      "learning_rate": 3.397829599072492e-06,
      "loss": 0.8135,
      "num_input_tokens_seen": 4117370368,
      "step": 55510,
      "train_runtime": 66127.9475,
      "train_tokens_per_second": 62263.695
    },
    {
      "epoch": 0.7378024066285274,
      "grad_norm": 27.375,
      "learning_rate": 3.3945976112509983e-06,
      "loss": 0.8347,
      "num_input_tokens_seen": 4118218240,
      "step": 55520,
      "train_runtime": 66141.8167,
      "train_tokens_per_second": 62263.458
    },
    {
      "epoch": 0.7379352961109893,
      "grad_norm": 17.25,
      "learning_rate": 3.3913668470213943e-06,
      "loss": 0.7298,
      "num_input_tokens_seen": 4118705664,
      "step": 55530,
      "train_runtime": 66148.607,
      "train_tokens_per_second": 62264.435
    },
    {
      "epoch": 0.7380681855934512,
      "grad_norm": 68.0,
      "learning_rate": 3.3881373069821544e-06,
      "loss": 0.7337,
      "num_input_tokens_seen": 4118934784,
      "step": 55540,
      "train_runtime": 66152.0352,
      "train_tokens_per_second": 62264.672
    },
    {
      "epoch": 0.7382010750759131,
      "grad_norm": 11.75,
      "learning_rate": 3.38490899173152e-06,
      "loss": 0.5612,
      "num_input_tokens_seen": 4119049728,
      "step": 55550,
      "train_runtime": 66154.4916,
      "train_tokens_per_second": 62264.098
    },
    {
      "epoch": 0.7383339645583751,
      "grad_norm": 23.625,
      "learning_rate": 3.381681901867511e-06,
      "loss": 0.7746,
      "num_input_tokens_seen": 4121094656,
      "step": 55560,
      "train_runtime": 66189.0984,
      "train_tokens_per_second": 62262.438
    },
    {
      "epoch": 0.7384668540408369,
      "grad_norm": 17.75,
      "learning_rate": 3.3784560379879183e-06,
      "loss": 0.9218,
      "num_input_tokens_seen": 4121933312,
      "step": 55570,
      "train_runtime": 66201.554,
      "train_tokens_per_second": 62263.392
    },
    {
      "epoch": 0.7385997435232988,
      "grad_norm": 40.5,
      "learning_rate": 3.375231400690294e-06,
      "loss": 0.7019,
      "num_input_tokens_seen": 4122400512,
      "step": 55580,
      "train_runtime": 66208.1263,
      "train_tokens_per_second": 62264.268
    },
    {
      "epoch": 0.7387326330057608,
      "grad_norm": 10.4375,
      "learning_rate": 3.3720079905719883e-06,
      "loss": 0.6429,
      "num_input_tokens_seen": 4122629120,
      "step": 55590,
      "train_runtime": 66211.5977,
      "train_tokens_per_second": 62264.456
    },
    {
      "epoch": 0.7388655224882227,
      "grad_norm": 11.25,
      "learning_rate": 3.368785808230094e-06,
      "loss": 0.6038,
      "num_input_tokens_seen": 4122743808,
      "step": 55600,
      "train_runtime": 66214.9859,
      "train_tokens_per_second": 62263.002
    },
    {
      "epoch": 0.7389984119706846,
      "grad_norm": 61.25,
      "learning_rate": 3.365564854261506e-06,
      "loss": 0.849,
      "num_input_tokens_seen": 4124747776,
      "step": 55610,
      "train_runtime": 66246.3904,
      "train_tokens_per_second": 62263.736
    },
    {
      "epoch": 0.7391313014531465,
      "grad_norm": 55.0,
      "learning_rate": 3.3623451292628683e-06,
      "loss": 0.9213,
      "num_input_tokens_seen": 4125596416,
      "step": 55620,
      "train_runtime": 66259.2128,
      "train_tokens_per_second": 62264.495
    },
    {
      "epoch": 0.7392641909356084,
      "grad_norm": 24.375,
      "learning_rate": 3.3591266338306074e-06,
      "loss": 0.7614,
      "num_input_tokens_seen": 4126038272,
      "step": 55630,
      "train_runtime": 66265.6238,
      "train_tokens_per_second": 62265.139
    },
    {
      "epoch": 0.7393970804180703,
      "grad_norm": 20.5,
      "learning_rate": 3.3559093685609256e-06,
      "loss": 0.6185,
      "num_input_tokens_seen": 4126264832,
      "step": 55640,
      "train_runtime": 66269.0723,
      "train_tokens_per_second": 62265.318
    },
    {
      "epoch": 0.7395299699005322,
      "grad_norm": 13.1875,
      "learning_rate": 3.352693334049784e-06,
      "loss": 0.5046,
      "num_input_tokens_seen": 4126380288,
      "step": 55650,
      "train_runtime": 66271.5753,
      "train_tokens_per_second": 62264.708
    },
    {
      "epoch": 0.7396628593829941,
      "grad_norm": 11.0625,
      "learning_rate": 3.3494785308929334e-06,
      "loss": 0.8451,
      "num_input_tokens_seen": 4128535552,
      "step": 55660,
      "train_runtime": 66307.3354,
      "train_tokens_per_second": 62263.632
    },
    {
      "epoch": 0.7397957488654561,
      "grad_norm": 18.625,
      "learning_rate": 3.346264959685881e-06,
      "loss": 0.9495,
      "num_input_tokens_seen": 4129410304,
      "step": 55670,
      "train_runtime": 66320.493,
      "train_tokens_per_second": 62264.469
    },
    {
      "epoch": 0.739928638347918,
      "grad_norm": 12.6875,
      "learning_rate": 3.3430526210239158e-06,
      "loss": 0.7386,
      "num_input_tokens_seen": 4129915392,
      "step": 55680,
      "train_runtime": 66327.5271,
      "train_tokens_per_second": 62265.481
    },
    {
      "epoch": 0.7400615278303798,
      "grad_norm": 26.25,
      "learning_rate": 3.3398415155020946e-06,
      "loss": 0.6176,
      "num_input_tokens_seen": 4130149376,
      "step": 55690,
      "train_runtime": 66331.1148,
      "train_tokens_per_second": 62265.641
    },
    {
      "epoch": 0.7401944173128417,
      "grad_norm": 11.1875,
      "learning_rate": 3.33663164371524e-06,
      "loss": 0.5175,
      "num_input_tokens_seen": 4130265088,
      "step": 55700,
      "train_runtime": 66333.6149,
      "train_tokens_per_second": 62265.038
    },
    {
      "epoch": 0.7403273067953037,
      "grad_norm": 22.375,
      "learning_rate": 3.3334230062579643e-06,
      "loss": 0.7945,
      "num_input_tokens_seen": 4132222208,
      "step": 55710,
      "train_runtime": 66366.8523,
      "train_tokens_per_second": 62263.345
    },
    {
      "epoch": 0.7404601962777656,
      "grad_norm": 16.875,
      "learning_rate": 3.330215603724628e-06,
      "loss": 0.84,
      "num_input_tokens_seen": 4133067520,
      "step": 55720,
      "train_runtime": 66379.1898,
      "train_tokens_per_second": 62264.507
    },
    {
      "epoch": 0.7405930857602275,
      "grad_norm": 17.25,
      "learning_rate": 3.3270094367093774e-06,
      "loss": 0.7375,
      "num_input_tokens_seen": 4133491712,
      "step": 55730,
      "train_runtime": 66385.4751,
      "train_tokens_per_second": 62265.002
    },
    {
      "epoch": 0.7407259752426895,
      "grad_norm": 9.625,
      "learning_rate": 3.3238045058061263e-06,
      "loss": 0.69,
      "num_input_tokens_seen": 4133711104,
      "step": 55740,
      "train_runtime": 66388.8698,
      "train_tokens_per_second": 62265.122
    },
    {
      "epoch": 0.7408588647251513,
      "grad_norm": 35.5,
      "learning_rate": 3.3206008116085607e-06,
      "loss": 0.533,
      "num_input_tokens_seen": 4133825792,
      "step": 55750,
      "train_runtime": 66391.3398,
      "train_tokens_per_second": 62264.533
    },
    {
      "epoch": 0.7409917542076132,
      "grad_norm": 258.0,
      "learning_rate": 3.3173983547101363e-06,
      "loss": 0.8183,
      "num_input_tokens_seen": 4135961344,
      "step": 55760,
      "train_runtime": 66424.3436,
      "train_tokens_per_second": 62265.747
    },
    {
      "epoch": 0.7411246436900751,
      "grad_norm": 13.75,
      "learning_rate": 3.314197135704078e-06,
      "loss": 0.8871,
      "num_input_tokens_seen": 4136804608,
      "step": 55770,
      "train_runtime": 66437.7824,
      "train_tokens_per_second": 62265.844
    },
    {
      "epoch": 0.7412575331725371,
      "grad_norm": 11.6875,
      "learning_rate": 3.3109971551833874e-06,
      "loss": 0.6722,
      "num_input_tokens_seen": 4137237248,
      "step": 55780,
      "train_runtime": 66444.1077,
      "train_tokens_per_second": 62266.428
    },
    {
      "epoch": 0.741390422654999,
      "grad_norm": 15.0,
      "learning_rate": 3.307798413740827e-06,
      "loss": 0.639,
      "num_input_tokens_seen": 4137466880,
      "step": 55790,
      "train_runtime": 66447.6602,
      "train_tokens_per_second": 62266.555
    },
    {
      "epoch": 0.7415233121374609,
      "grad_norm": 7.1875,
      "learning_rate": 3.304600911968936e-06,
      "loss": 0.523,
      "num_input_tokens_seen": 4137587456,
      "step": 55800,
      "train_runtime": 66450.1946,
      "train_tokens_per_second": 62265.995
    },
    {
      "epoch": 0.7416562016199227,
      "grad_norm": 21.0,
      "learning_rate": 3.301404650460024e-06,
      "loss": 0.7758,
      "num_input_tokens_seen": 4139692032,
      "step": 55810,
      "train_runtime": 66484.1519,
      "train_tokens_per_second": 62265.847
    },
    {
      "epoch": 0.7417890911023847,
      "grad_norm": 21.125,
      "learning_rate": 3.2982096298061705e-06,
      "loss": 0.8783,
      "num_input_tokens_seen": 4140536832,
      "step": 55820,
      "train_runtime": 66498.1592,
      "train_tokens_per_second": 62265.435
    },
    {
      "epoch": 0.7419219805848466,
      "grad_norm": 21.125,
      "learning_rate": 3.295015850599225e-06,
      "loss": 0.7909,
      "num_input_tokens_seen": 4141024512,
      "step": 55830,
      "train_runtime": 66505.1875,
      "train_tokens_per_second": 62266.188
    },
    {
      "epoch": 0.7420548700673085,
      "grad_norm": 17.5,
      "learning_rate": 3.2918233134308087e-06,
      "loss": 0.6501,
      "num_input_tokens_seen": 4141260544,
      "step": 55840,
      "train_runtime": 66508.8018,
      "train_tokens_per_second": 62266.353
    },
    {
      "epoch": 0.7421877595497705,
      "grad_norm": 34.0,
      "learning_rate": 3.288632018892305e-06,
      "loss": 0.5852,
      "num_input_tokens_seen": 4141378304,
      "step": 55850,
      "train_runtime": 66511.391,
      "train_tokens_per_second": 62265.7
    },
    {
      "epoch": 0.7423206490322324,
      "grad_norm": 16.25,
      "learning_rate": 3.285441967574875e-06,
      "loss": 0.8281,
      "num_input_tokens_seen": 4143400448,
      "step": 55860,
      "train_runtime": 66545.5221,
      "train_tokens_per_second": 62264.151
    },
    {
      "epoch": 0.7424535385146942,
      "grad_norm": 18.875,
      "learning_rate": 3.2822531600694485e-06,
      "loss": 0.9949,
      "num_input_tokens_seen": 4144259840,
      "step": 55870,
      "train_runtime": 66557.8926,
      "train_tokens_per_second": 62265.491
    },
    {
      "epoch": 0.7425864279971561,
      "grad_norm": 15.375,
      "learning_rate": 3.2790655969667227e-06,
      "loss": 0.7427,
      "num_input_tokens_seen": 4144727296,
      "step": 55880,
      "train_runtime": 66564.5647,
      "train_tokens_per_second": 62266.272
    },
    {
      "epoch": 0.7427193174796181,
      "grad_norm": 21.625,
      "learning_rate": 3.2758792788571692e-06,
      "loss": 0.6437,
      "num_input_tokens_seen": 4144958976,
      "step": 55890,
      "train_runtime": 66568.1155,
      "train_tokens_per_second": 62266.431
    },
    {
      "epoch": 0.74285220696208,
      "grad_norm": 11.625,
      "learning_rate": 3.2726942063310154e-06,
      "loss": 0.5018,
      "num_input_tokens_seen": 4145075968,
      "step": 55900,
      "train_runtime": 66570.561,
      "train_tokens_per_second": 62265.901
    },
    {
      "epoch": 0.7429850964445419,
      "grad_norm": 15.6875,
      "learning_rate": 3.269510379978279e-06,
      "loss": 0.8881,
      "num_input_tokens_seen": 4147068928,
      "step": 55910,
      "train_runtime": 66602.1366,
      "train_tokens_per_second": 62266.305
    },
    {
      "epoch": 0.7431179859270038,
      "grad_norm": 23.875,
      "learning_rate": 3.2663278003887233e-06,
      "loss": 0.8939,
      "num_input_tokens_seen": 4147957504,
      "step": 55920,
      "train_runtime": 66614.8775,
      "train_tokens_per_second": 62267.734
    },
    {
      "epoch": 0.7432508754094658,
      "grad_norm": 14.875,
      "learning_rate": 3.2631464681519066e-06,
      "loss": 0.738,
      "num_input_tokens_seen": 4148433408,
      "step": 55930,
      "train_runtime": 66621.4061,
      "train_tokens_per_second": 62268.776
    },
    {
      "epoch": 0.7433837648919276,
      "grad_norm": 18.75,
      "learning_rate": 3.259966383857133e-06,
      "loss": 0.6156,
      "num_input_tokens_seen": 4148652032,
      "step": 55940,
      "train_runtime": 66625.8514,
      "train_tokens_per_second": 62267.903
    },
    {
      "epoch": 0.7435166543743895,
      "grad_norm": 16.25,
      "learning_rate": 3.2567875480934854e-06,
      "loss": 0.4966,
      "num_input_tokens_seen": 4148764416,
      "step": 55950,
      "train_runtime": 66628.2897,
      "train_tokens_per_second": 62267.311
    },
    {
      "epoch": 0.7436495438568514,
      "grad_norm": 30.375,
      "learning_rate": 3.2536099614498208e-06,
      "loss": 0.751,
      "num_input_tokens_seen": 4150871552,
      "step": 55960,
      "train_runtime": 66662.8947,
      "train_tokens_per_second": 62266.596
    },
    {
      "epoch": 0.7437824333393134,
      "grad_norm": 37.0,
      "learning_rate": 3.2504336245147473e-06,
      "loss": 0.8866,
      "num_input_tokens_seen": 4151722752,
      "step": 55970,
      "train_runtime": 66675.6885,
      "train_tokens_per_second": 62267.415
    },
    {
      "epoch": 0.7439153228217753,
      "grad_norm": 44.25,
      "learning_rate": 3.2472585378766665e-06,
      "loss": 0.7709,
      "num_input_tokens_seen": 4152167680,
      "step": 55980,
      "train_runtime": 66682.2128,
      "train_tokens_per_second": 62267.995
    },
    {
      "epoch": 0.7440482123042372,
      "grad_norm": 134.0,
      "learning_rate": 3.2440847021237254e-06,
      "loss": 0.6022,
      "num_input_tokens_seen": 4152387328,
      "step": 55990,
      "train_runtime": 66685.5834,
      "train_tokens_per_second": 62268.141
    },
    {
      "epoch": 0.744181101786699,
      "grad_norm": 13.875,
      "learning_rate": 3.240912117843852e-06,
      "loss": 0.4878,
      "num_input_tokens_seen": 4152503552,
      "step": 56000,
      "train_runtime": 66689.0886,
      "train_tokens_per_second": 62266.611
    },
    {
      "epoch": 0.744313991269161,
      "grad_norm": 27.875,
      "learning_rate": 3.2377407856247422e-06,
      "loss": 0.8006,
      "num_input_tokens_seen": 4154366464,
      "step": 56010,
      "train_runtime": 66719.0281,
      "train_tokens_per_second": 62266.591
    },
    {
      "epoch": 0.7444468807516229,
      "grad_norm": 45.0,
      "learning_rate": 3.2345707060538477e-06,
      "loss": 0.9787,
      "num_input_tokens_seen": 4155218944,
      "step": 56020,
      "train_runtime": 66731.8437,
      "train_tokens_per_second": 62267.408
    },
    {
      "epoch": 0.7445797702340848,
      "grad_norm": 15.375,
      "learning_rate": 3.23140187971841e-06,
      "loss": 0.7557,
      "num_input_tokens_seen": 4155683072,
      "step": 56030,
      "train_runtime": 66738.5882,
      "train_tokens_per_second": 62268.07
    },
    {
      "epoch": 0.7447126597165468,
      "grad_norm": 13.125,
      "learning_rate": 3.2282343072054156e-06,
      "loss": 0.6599,
      "num_input_tokens_seen": 4155902720,
      "step": 56040,
      "train_runtime": 66742.9764,
      "train_tokens_per_second": 62267.267
    },
    {
      "epoch": 0.7448455491990087,
      "grad_norm": 14.125,
      "learning_rate": 3.2250679891016334e-06,
      "loss": 0.5632,
      "num_input_tokens_seen": 4156017152,
      "step": 56050,
      "train_runtime": 66745.3561,
      "train_tokens_per_second": 62266.761
    },
    {
      "epoch": 0.7449784386814705,
      "grad_norm": 27.125,
      "learning_rate": 3.2219029259935963e-06,
      "loss": 0.8843,
      "num_input_tokens_seen": 4157977088,
      "step": 56060,
      "train_runtime": 66777.3663,
      "train_tokens_per_second": 62266.264
    },
    {
      "epoch": 0.7451113281639324,
      "grad_norm": 19.5,
      "learning_rate": 3.218739118467602e-06,
      "loss": 0.8887,
      "num_input_tokens_seen": 4158785280,
      "step": 56070,
      "train_runtime": 66789.741,
      "train_tokens_per_second": 62266.827
    },
    {
      "epoch": 0.7452442176463944,
      "grad_norm": 14.625,
      "learning_rate": 3.2155765671097196e-06,
      "loss": 0.7337,
      "num_input_tokens_seen": 4159233536,
      "step": 56080,
      "train_runtime": 66796.2715,
      "train_tokens_per_second": 62267.451
    },
    {
      "epoch": 0.7453771071288563,
      "grad_norm": 14.875,
      "learning_rate": 3.212415272505781e-06,
      "loss": 0.6383,
      "num_input_tokens_seen": 4159459328,
      "step": 56090,
      "train_runtime": 66799.7305,
      "train_tokens_per_second": 62267.606
    },
    {
      "epoch": 0.7455099966113182,
      "grad_norm": 32.75,
      "learning_rate": 3.209255235241394e-06,
      "loss": 0.4993,
      "num_input_tokens_seen": 4159576320,
      "step": 56100,
      "train_runtime": 66802.152,
      "train_tokens_per_second": 62267.101
    },
    {
      "epoch": 0.7456428860937802,
      "grad_norm": 21.5,
      "learning_rate": 3.206096455901919e-06,
      "loss": 0.8007,
      "num_input_tokens_seen": 4161574400,
      "step": 56110,
      "train_runtime": 66834.9884,
      "train_tokens_per_second": 62266.404
    },
    {
      "epoch": 0.745775775576242,
      "grad_norm": 31.0,
      "learning_rate": 3.2029389350724948e-06,
      "loss": 0.9519,
      "num_input_tokens_seen": 4162389760,
      "step": 56120,
      "train_runtime": 66848.2303,
      "train_tokens_per_second": 62266.267
    },
    {
      "epoch": 0.7459086650587039,
      "grad_norm": 17.75,
      "learning_rate": 3.1997826733380254e-06,
      "loss": 0.722,
      "num_input_tokens_seen": 4162832640,
      "step": 56130,
      "train_runtime": 66854.6119,
      "train_tokens_per_second": 62266.948
    },
    {
      "epoch": 0.7460415545411658,
      "grad_norm": 10.9375,
      "learning_rate": 3.196627671283179e-06,
      "loss": 0.6542,
      "num_input_tokens_seen": 4163042816,
      "step": 56140,
      "train_runtime": 66857.8646,
      "train_tokens_per_second": 62267.062
    },
    {
      "epoch": 0.7461744440236278,
      "grad_norm": 46.25,
      "learning_rate": 3.1934739294923923e-06,
      "loss": 0.5869,
      "num_input_tokens_seen": 4163152640,
      "step": 56150,
      "train_runtime": 66860.2145,
      "train_tokens_per_second": 62266.516
    },
    {
      "epoch": 0.7463073335060897,
      "grad_norm": 27.375,
      "learning_rate": 3.19032144854987e-06,
      "loss": 0.825,
      "num_input_tokens_seen": 4165280512,
      "step": 56160,
      "train_runtime": 66894.2432,
      "train_tokens_per_second": 62266.651
    },
    {
      "epoch": 0.7464402229885516,
      "grad_norm": 14.375,
      "learning_rate": 3.1871702290395734e-06,
      "loss": 0.9374,
      "num_input_tokens_seen": 4166142720,
      "step": 56170,
      "train_runtime": 66906.972,
      "train_tokens_per_second": 62267.692
    },
    {
      "epoch": 0.7465731124710134,
      "grad_norm": 15.875,
      "learning_rate": 3.1840202715452485e-06,
      "loss": 0.7557,
      "num_input_tokens_seen": 4166621696,
      "step": 56180,
      "train_runtime": 66914.6807,
      "train_tokens_per_second": 62267.677
    },
    {
      "epoch": 0.7467060019534754,
      "grad_norm": 18.625,
      "learning_rate": 3.1808715766503885e-06,
      "loss": 0.668,
      "num_input_tokens_seen": 4166862080,
      "step": 56190,
      "train_runtime": 66918.2068,
      "train_tokens_per_second": 62267.988
    },
    {
      "epoch": 0.7468388914359373,
      "grad_norm": 19.75,
      "learning_rate": 3.1777241449382635e-06,
      "loss": 0.5208,
      "num_input_tokens_seen": 4166977792,
      "step": 56200,
      "train_runtime": 66920.5715,
      "train_tokens_per_second": 62267.517
    },
    {
      "epoch": 0.7469717809183992,
      "grad_norm": 78.5,
      "learning_rate": 3.1745779769919105e-06,
      "loss": 0.8498,
      "num_input_tokens_seen": 4169060352,
      "step": 56210,
      "train_runtime": 66954.2536,
      "train_tokens_per_second": 62267.296
    },
    {
      "epoch": 0.7471046704008611,
      "grad_norm": 28.5,
      "learning_rate": 3.17143307339412e-06,
      "loss": 0.8797,
      "num_input_tokens_seen": 4169895424,
      "step": 56220,
      "train_runtime": 66966.8425,
      "train_tokens_per_second": 62268.061
    },
    {
      "epoch": 0.7472375598833231,
      "grad_norm": 19.375,
      "learning_rate": 3.1682894347274707e-06,
      "loss": 0.7377,
      "num_input_tokens_seen": 4170339840,
      "step": 56230,
      "train_runtime": 66974.1455,
      "train_tokens_per_second": 62267.907
    },
    {
      "epoch": 0.7473704493657849,
      "grad_norm": 26.875,
      "learning_rate": 3.165147061574282e-06,
      "loss": 0.6182,
      "num_input_tokens_seen": 4170571776,
      "step": 56240,
      "train_runtime": 66977.5984,
      "train_tokens_per_second": 62268.159
    },
    {
      "epoch": 0.7475033388482468,
      "grad_norm": 28.625,
      "learning_rate": 3.1620059545166547e-06,
      "loss": 0.4821,
      "num_input_tokens_seen": 4170685696,
      "step": 56250,
      "train_runtime": 66979.9918,
      "train_tokens_per_second": 62267.635
    },
    {
      "epoch": 0.7476362283307088,
      "grad_norm": 26.875,
      "learning_rate": 3.1588661141364554e-06,
      "loss": 0.8765,
      "num_input_tokens_seen": 4172806144,
      "step": 56260,
      "train_runtime": 67013.7049,
      "train_tokens_per_second": 62267.952
    },
    {
      "epoch": 0.7477691178131707,
      "grad_norm": 23.25,
      "learning_rate": 3.1557275410153e-06,
      "loss": 0.861,
      "num_input_tokens_seen": 4173651968,
      "step": 56270,
      "train_runtime": 67026.4261,
      "train_tokens_per_second": 62268.753
    },
    {
      "epoch": 0.7479020072956326,
      "grad_norm": 20.125,
      "learning_rate": 3.152590235734595e-06,
      "loss": 0.7682,
      "num_input_tokens_seen": 4174111744,
      "step": 56280,
      "train_runtime": 67033.8938,
      "train_tokens_per_second": 62268.675
    },
    {
      "epoch": 0.7480348967780945,
      "grad_norm": 15.125,
      "learning_rate": 3.149454198875486e-06,
      "loss": 0.6284,
      "num_input_tokens_seen": 4174337536,
      "step": 56290,
      "train_runtime": 67037.3543,
      "train_tokens_per_second": 62268.829
    },
    {
      "epoch": 0.7481677862605564,
      "grad_norm": 12.625,
      "learning_rate": 3.146319431018906e-06,
      "loss": 0.4985,
      "num_input_tokens_seen": 4174449408,
      "step": 56300,
      "train_runtime": 67039.7359,
      "train_tokens_per_second": 62268.285
    },
    {
      "epoch": 0.7483006757430183,
      "grad_norm": 16.75,
      "learning_rate": 3.143185932745536e-06,
      "loss": 0.8204,
      "num_input_tokens_seen": 4176483584,
      "step": 56310,
      "train_runtime": 67072.5848,
      "train_tokens_per_second": 62268.117
    },
    {
      "epoch": 0.7484335652254802,
      "grad_norm": 17.5,
      "learning_rate": 3.1400537046358316e-06,
      "loss": 0.944,
      "num_input_tokens_seen": 4177319936,
      "step": 56320,
      "train_runtime": 67085.0527,
      "train_tokens_per_second": 62269.012
    },
    {
      "epoch": 0.7485664547079421,
      "grad_norm": 21.625,
      "learning_rate": 3.136922747270009e-06,
      "loss": 0.7019,
      "num_input_tokens_seen": 4177761280,
      "step": 56330,
      "train_runtime": 67091.5365,
      "train_tokens_per_second": 62269.572
    },
    {
      "epoch": 0.7486993441904041,
      "grad_norm": 18.125,
      "learning_rate": 3.1337930612280497e-06,
      "loss": 0.5972,
      "num_input_tokens_seen": 4177977856,
      "step": 56340,
      "train_runtime": 67094.8844,
      "train_tokens_per_second": 62269.693
    },
    {
      "epoch": 0.748832233672866,
      "grad_norm": 10.1875,
      "learning_rate": 3.130664647089705e-06,
      "loss": 0.5381,
      "num_input_tokens_seen": 4178087680,
      "step": 56350,
      "train_runtime": 67098.1064,
      "train_tokens_per_second": 62268.34
    },
    {
      "epoch": 0.7489651231553278,
      "grad_norm": 43.0,
      "learning_rate": 3.127537505434478e-06,
      "loss": 0.862,
      "num_input_tokens_seen": 4180136704,
      "step": 56360,
      "train_runtime": 67130.4727,
      "train_tokens_per_second": 62268.841
    },
    {
      "epoch": 0.7490980126377897,
      "grad_norm": 17.875,
      "learning_rate": 3.1244116368416487e-06,
      "loss": 0.8395,
      "num_input_tokens_seen": 4180941056,
      "step": 56370,
      "train_runtime": 67142.5404,
      "train_tokens_per_second": 62269.629
    },
    {
      "epoch": 0.7492309021202517,
      "grad_norm": 22.625,
      "learning_rate": 3.121287041890254e-06,
      "loss": 0.7537,
      "num_input_tokens_seen": 4181358592,
      "step": 56380,
      "train_runtime": 67148.7314,
      "train_tokens_per_second": 62270.106
    },
    {
      "epoch": 0.7493637916027136,
      "grad_norm": 123.0,
      "learning_rate": 3.1181637211590987e-06,
      "loss": 0.7438,
      "num_input_tokens_seen": 4181579776,
      "step": 56390,
      "train_runtime": 67152.9194,
      "train_tokens_per_second": 62269.516
    },
    {
      "epoch": 0.7494966810851755,
      "grad_norm": 15.875,
      "learning_rate": 3.1150416752267498e-06,
      "loss": 0.4471,
      "num_input_tokens_seen": 4181697024,
      "step": 56400,
      "train_runtime": 67155.3354,
      "train_tokens_per_second": 62269.021
    },
    {
      "epoch": 0.7496295705676375,
      "grad_norm": 47.25,
      "learning_rate": 3.1119209046715415e-06,
      "loss": 0.8756,
      "num_input_tokens_seen": 4183669248,
      "step": 56410,
      "train_runtime": 67186.9774,
      "train_tokens_per_second": 62269.05
    },
    {
      "epoch": 0.7497624600500994,
      "grad_norm": 50.25,
      "learning_rate": 3.1088014100715637e-06,
      "loss": 0.9052,
      "num_input_tokens_seen": 4184510976,
      "step": 56420,
      "train_runtime": 67199.752,
      "train_tokens_per_second": 62269.738
    },
    {
      "epoch": 0.7498953495325612,
      "grad_norm": 32.25,
      "learning_rate": 3.105683192004676e-06,
      "loss": 0.7227,
      "num_input_tokens_seen": 4185000192,
      "step": 56430,
      "train_runtime": 67206.6852,
      "train_tokens_per_second": 62270.594
    },
    {
      "epoch": 0.7500282390150231,
      "grad_norm": 21.25,
      "learning_rate": 3.102566251048501e-06,
      "loss": 0.6019,
      "num_input_tokens_seen": 4185229568,
      "step": 56440,
      "train_runtime": 67210.1955,
      "train_tokens_per_second": 62270.754
    },
    {
      "epoch": 0.7501611284974851,
      "grad_norm": 14.9375,
      "learning_rate": 3.099450587780426e-06,
      "loss": 0.5258,
      "num_input_tokens_seen": 4185343488,
      "step": 56450,
      "train_runtime": 67212.5587,
      "train_tokens_per_second": 62270.26
    },
    {
      "epoch": 0.750294017979947,
      "grad_norm": 20.25,
      "learning_rate": 3.096336202777598e-06,
      "loss": 0.7392,
      "num_input_tokens_seen": 4187489024,
      "step": 56460,
      "train_runtime": 67248.1627,
      "train_tokens_per_second": 62269.196
    },
    {
      "epoch": 0.7504269074624089,
      "grad_norm": 11.375,
      "learning_rate": 3.0932230966169295e-06,
      "loss": 0.8404,
      "num_input_tokens_seen": 4188357632,
      "step": 56470,
      "train_runtime": 67261.3766,
      "train_tokens_per_second": 62269.877
    },
    {
      "epoch": 0.7505597969448708,
      "grad_norm": 304.0,
      "learning_rate": 3.090111269875099e-06,
      "loss": 0.7848,
      "num_input_tokens_seen": 4188838400,
      "step": 56480,
      "train_runtime": 67268.0936,
      "train_tokens_per_second": 62270.806
    },
    {
      "epoch": 0.7506926864273327,
      "grad_norm": 20.625,
      "learning_rate": 3.0870007231285347e-06,
      "loss": 0.6271,
      "num_input_tokens_seen": 4189056000,
      "step": 56490,
      "train_runtime": 67271.5537,
      "train_tokens_per_second": 62270.838
    },
    {
      "epoch": 0.7508255759097946,
      "grad_norm": 6.90625,
      "learning_rate": 3.0838914569534505e-06,
      "loss": 0.464,
      "num_input_tokens_seen": 4189166592,
      "step": 56500,
      "train_runtime": 67273.9413,
      "train_tokens_per_second": 62270.272
    },
    {
      "epoch": 0.7509584653922565,
      "grad_norm": 24.125,
      "learning_rate": 3.0807834719258024e-06,
      "loss": 0.8759,
      "num_input_tokens_seen": 4191230976,
      "step": 56510,
      "train_runtime": 67307.8966,
      "train_tokens_per_second": 62269.528
    },
    {
      "epoch": 0.7510913548747185,
      "grad_norm": 11.5625,
      "learning_rate": 3.077676768621317e-06,
      "loss": 0.8588,
      "num_input_tokens_seen": 4192056320,
      "step": 56520,
      "train_runtime": 67320.3962,
      "train_tokens_per_second": 62270.227
    },
    {
      "epoch": 0.7512242443571804,
      "grad_norm": 25.25,
      "learning_rate": 3.0745713476154894e-06,
      "loss": 0.7649,
      "num_input_tokens_seen": 4192509440,
      "step": 56530,
      "train_runtime": 67326.9771,
      "train_tokens_per_second": 62270.87
    },
    {
      "epoch": 0.7513571338396423,
      "grad_norm": 17.375,
      "learning_rate": 3.0714672094835598e-06,
      "loss": 0.6943,
      "num_input_tokens_seen": 4192743680,
      "step": 56540,
      "train_runtime": 67330.5495,
      "train_tokens_per_second": 62271.045
    },
    {
      "epoch": 0.7514900233221041,
      "grad_norm": 9.25,
      "learning_rate": 3.0683643548005548e-06,
      "loss": 0.5775,
      "num_input_tokens_seen": 4192860416,
      "step": 56550,
      "train_runtime": 67332.987,
      "train_tokens_per_second": 62270.524
    },
    {
      "epoch": 0.7516229128045661,
      "grad_norm": 15.375,
      "learning_rate": 3.065262784141243e-06,
      "loss": 0.8498,
      "num_input_tokens_seen": 4195033600,
      "step": 56560,
      "train_runtime": 67368.1879,
      "train_tokens_per_second": 62270.245
    },
    {
      "epoch": 0.751755802287028,
      "grad_norm": 14.9375,
      "learning_rate": 3.0621624980801634e-06,
      "loss": 0.9564,
      "num_input_tokens_seen": 4195861504,
      "step": 56570,
      "train_runtime": 67381.8294,
      "train_tokens_per_second": 62269.926
    },
    {
      "epoch": 0.7518886917694899,
      "grad_norm": 16.25,
      "learning_rate": 3.059063497191621e-06,
      "loss": 0.6843,
      "num_input_tokens_seen": 4196326400,
      "step": 56580,
      "train_runtime": 67388.4364,
      "train_tokens_per_second": 62270.719
    },
    {
      "epoch": 0.7520215812519518,
      "grad_norm": 25.625,
      "learning_rate": 3.055965782049668e-06,
      "loss": 0.7001,
      "num_input_tokens_seen": 4196571648,
      "step": 56590,
      "train_runtime": 67391.999,
      "train_tokens_per_second": 62271.066
    },
    {
      "epoch": 0.7521544707344138,
      "grad_norm": 17.875,
      "learning_rate": 3.0528693532281417e-06,
      "loss": 0.553,
      "num_input_tokens_seen": 4196688128,
      "step": 56600,
      "train_runtime": 67394.3718,
      "train_tokens_per_second": 62270.602
    },
    {
      "epoch": 0.7522873602168756,
      "grad_norm": 35.25,
      "learning_rate": 3.0497742113006155e-06,
      "loss": 0.8125,
      "num_input_tokens_seen": 4198693376,
      "step": 56610,
      "train_runtime": 67427.3152,
      "train_tokens_per_second": 62269.918
    },
    {
      "epoch": 0.7524202496993375,
      "grad_norm": 22.0,
      "learning_rate": 3.046680356840448e-06,
      "loss": 0.9121,
      "num_input_tokens_seen": 4199516672,
      "step": 56620,
      "train_runtime": 67439.9247,
      "train_tokens_per_second": 62270.483
    },
    {
      "epoch": 0.7525531391817994,
      "grad_norm": 15.875,
      "learning_rate": 3.04358779042074e-06,
      "loss": 0.684,
      "num_input_tokens_seen": 4199981568,
      "step": 56630,
      "train_runtime": 67447.388,
      "train_tokens_per_second": 62270.485
    },
    {
      "epoch": 0.7526860286642614,
      "grad_norm": 27.5,
      "learning_rate": 3.040496512614365e-06,
      "loss": 0.6128,
      "num_input_tokens_seen": 4200208128,
      "step": 56640,
      "train_runtime": 67450.7817,
      "train_tokens_per_second": 62270.711
    },
    {
      "epoch": 0.7528189181467233,
      "grad_norm": 11.875,
      "learning_rate": 3.037406523993954e-06,
      "loss": 0.5222,
      "num_input_tokens_seen": 4200321792,
      "step": 56650,
      "train_runtime": 67453.2019,
      "train_tokens_per_second": 62270.162
    },
    {
      "epoch": 0.7529518076291852,
      "grad_norm": 49.0,
      "learning_rate": 3.0343178251318996e-06,
      "loss": 0.8148,
      "num_input_tokens_seen": 4202427904,
      "step": 56660,
      "train_runtime": 67486.1418,
      "train_tokens_per_second": 62270.976
    },
    {
      "epoch": 0.753084697111647,
      "grad_norm": 16.75,
      "learning_rate": 3.0312304166003605e-06,
      "loss": 0.8468,
      "num_input_tokens_seen": 4203278080,
      "step": 56670,
      "train_runtime": 67498.5554,
      "train_tokens_per_second": 62272.119
    },
    {
      "epoch": 0.753217586594109,
      "grad_norm": 19.5,
      "learning_rate": 3.028144298971243e-06,
      "loss": 0.7297,
      "num_input_tokens_seen": 4203723776,
      "step": 56680,
      "train_runtime": 67505.9303,
      "train_tokens_per_second": 62271.918
    },
    {
      "epoch": 0.7533504760765709,
      "grad_norm": 16.25,
      "learning_rate": 3.0250594728162285e-06,
      "loss": 0.6498,
      "num_input_tokens_seen": 4203949312,
      "step": 56690,
      "train_runtime": 67509.2676,
      "train_tokens_per_second": 62272.181
    },
    {
      "epoch": 0.7534833655590328,
      "grad_norm": 18.0,
      "learning_rate": 3.0219759387067514e-06,
      "loss": 0.5568,
      "num_input_tokens_seen": 4204061440,
      "step": 56700,
      "train_runtime": 67511.6793,
      "train_tokens_per_second": 62271.617
    },
    {
      "epoch": 0.7536162550414948,
      "grad_norm": 17.5,
      "learning_rate": 3.018893697214009e-06,
      "loss": 0.8147,
      "num_input_tokens_seen": 4206098176,
      "step": 56710,
      "train_runtime": 67544.3335,
      "train_tokens_per_second": 62271.666
    },
    {
      "epoch": 0.7537491445239567,
      "grad_norm": 40.5,
      "learning_rate": 3.0158127489089617e-06,
      "loss": 0.9344,
      "num_input_tokens_seen": 4206947584,
      "step": 56720,
      "train_runtime": 67556.7254,
      "train_tokens_per_second": 62272.817
    },
    {
      "epoch": 0.7538820340064185,
      "grad_norm": 20.25,
      "learning_rate": 3.012733094362328e-06,
      "loss": 0.6612,
      "num_input_tokens_seen": 4207398144,
      "step": 56730,
      "train_runtime": 67563.1916,
      "train_tokens_per_second": 62273.526
    },
    {
      "epoch": 0.7540149234888804,
      "grad_norm": 17.875,
      "learning_rate": 3.0096547341445814e-06,
      "loss": 0.6159,
      "num_input_tokens_seen": 4207621376,
      "step": 56740,
      "train_runtime": 67566.6466,
      "train_tokens_per_second": 62273.645
    },
    {
      "epoch": 0.7541478129713424,
      "grad_norm": 14.75,
      "learning_rate": 3.006577668825963e-06,
      "loss": 0.4924,
      "num_input_tokens_seen": 4207738624,
      "step": 56750,
      "train_runtime": 67570.0673,
      "train_tokens_per_second": 62272.228
    },
    {
      "epoch": 0.7542807024538043,
      "grad_norm": 223.0,
      "learning_rate": 3.003501898976472e-06,
      "loss": 0.7879,
      "num_input_tokens_seen": 4209752320,
      "step": 56760,
      "train_runtime": 67602.3936,
      "train_tokens_per_second": 62272.238
    },
    {
      "epoch": 0.7544135919362662,
      "grad_norm": 24.25,
      "learning_rate": 3.000427425165868e-06,
      "loss": 0.8839,
      "num_input_tokens_seen": 4210565632,
      "step": 56770,
      "train_runtime": 67614.7797,
      "train_tokens_per_second": 62272.859
    },
    {
      "epoch": 0.7545464814187282,
      "grad_norm": 16.25,
      "learning_rate": 2.9973542479636732e-06,
      "loss": 0.7198,
      "num_input_tokens_seen": 4211049728,
      "step": 56780,
      "train_runtime": 67621.4259,
      "train_tokens_per_second": 62273.897
    },
    {
      "epoch": 0.75467937090119,
      "grad_norm": 11.875,
      "learning_rate": 2.9942823679391554e-06,
      "loss": 0.7719,
      "num_input_tokens_seen": 4211281408,
      "step": 56790,
      "train_runtime": 67624.9037,
      "train_tokens_per_second": 62274.121
    },
    {
      "epoch": 0.7548122603836519,
      "grad_norm": 19.0,
      "learning_rate": 2.9912117856613664e-06,
      "loss": 0.5235,
      "num_input_tokens_seen": 4211398144,
      "step": 56800,
      "train_runtime": 67628.2773,
      "train_tokens_per_second": 62272.74
    },
    {
      "epoch": 0.7549451498661138,
      "grad_norm": 29.0,
      "learning_rate": 2.988142501699093e-06,
      "loss": 0.8228,
      "num_input_tokens_seen": 4213474048,
      "step": 56810,
      "train_runtime": 67661.6488,
      "train_tokens_per_second": 62272.707
    },
    {
      "epoch": 0.7550780393485758,
      "grad_norm": 15.625,
      "learning_rate": 2.9850745166209026e-06,
      "loss": 0.8798,
      "num_input_tokens_seen": 4214294272,
      "step": 56820,
      "train_runtime": 67674.3702,
      "train_tokens_per_second": 62273.121
    },
    {
      "epoch": 0.7552109288310377,
      "grad_norm": 18.625,
      "learning_rate": 2.982007830995104e-06,
      "loss": 0.7425,
      "num_input_tokens_seen": 4214780160,
      "step": 56830,
      "train_runtime": 67681.3818,
      "train_tokens_per_second": 62273.849
    },
    {
      "epoch": 0.7553438183134996,
      "grad_norm": 22.25,
      "learning_rate": 2.9789424453897763e-06,
      "loss": 0.6085,
      "num_input_tokens_seen": 4215013888,
      "step": 56840,
      "train_runtime": 67684.9364,
      "train_tokens_per_second": 62274.032
    },
    {
      "epoch": 0.7554767077959615,
      "grad_norm": 11.25,
      "learning_rate": 2.975878360372758e-06,
      "loss": 0.6206,
      "num_input_tokens_seen": 4215129088,
      "step": 56850,
      "train_runtime": 67687.3694,
      "train_tokens_per_second": 62273.495
    },
    {
      "epoch": 0.7556095972784234,
      "grad_norm": 103.0,
      "learning_rate": 2.972815576511635e-06,
      "loss": 0.8201,
      "num_input_tokens_seen": 4217143808,
      "step": 56860,
      "train_runtime": 67720.5405,
      "train_tokens_per_second": 62272.743
    },
    {
      "epoch": 0.7557424867608853,
      "grad_norm": 18.5,
      "learning_rate": 2.9697540943737724e-06,
      "loss": 0.841,
      "num_input_tokens_seen": 4217994752,
      "step": 56870,
      "train_runtime": 67733.3211,
      "train_tokens_per_second": 62273.556
    },
    {
      "epoch": 0.7558753762433472,
      "grad_norm": 12.1875,
      "learning_rate": 2.966693914526273e-06,
      "loss": 0.7006,
      "num_input_tokens_seen": 4218445312,
      "step": 56880,
      "train_runtime": 67739.9173,
      "train_tokens_per_second": 62274.143
    },
    {
      "epoch": 0.7560082657258091,
      "grad_norm": 16.375,
      "learning_rate": 2.9636350375360103e-06,
      "loss": 0.5884,
      "num_input_tokens_seen": 4218669312,
      "step": 56890,
      "train_runtime": 67743.3317,
      "train_tokens_per_second": 62274.311
    },
    {
      "epoch": 0.7561411552082711,
      "grad_norm": 19.75,
      "learning_rate": 2.9605774639696183e-06,
      "loss": 0.5905,
      "num_input_tokens_seen": 4218785280,
      "step": 56900,
      "train_runtime": 67745.7848,
      "train_tokens_per_second": 62273.768
    },
    {
      "epoch": 0.756274044690733,
      "grad_norm": 39.5,
      "learning_rate": 2.957521194393477e-06,
      "loss": 0.9293,
      "num_input_tokens_seen": 4220954624,
      "step": 56910,
      "train_runtime": 67779.7122,
      "train_tokens_per_second": 62274.602
    },
    {
      "epoch": 0.7564069341731948,
      "grad_norm": 46.5,
      "learning_rate": 2.9544662293737423e-06,
      "loss": 0.8768,
      "num_input_tokens_seen": 4221792512,
      "step": 56920,
      "train_runtime": 67792.3642,
      "train_tokens_per_second": 62275.34
    },
    {
      "epoch": 0.7565398236556568,
      "grad_norm": 20.125,
      "learning_rate": 2.951412569476312e-06,
      "loss": 0.7254,
      "num_input_tokens_seen": 4222239232,
      "step": 56930,
      "train_runtime": 67798.8423,
      "train_tokens_per_second": 62275.978
    },
    {
      "epoch": 0.7566727131381187,
      "grad_norm": 21.5,
      "learning_rate": 2.9483602152668524e-06,
      "loss": 0.6654,
      "num_input_tokens_seen": 4222459136,
      "step": 56940,
      "train_runtime": 67802.2594,
      "train_tokens_per_second": 62276.083
    },
    {
      "epoch": 0.7568056026205806,
      "grad_norm": 17.625,
      "learning_rate": 2.9453091673107835e-06,
      "loss": 0.5185,
      "num_input_tokens_seen": 4222576640,
      "step": 56950,
      "train_runtime": 67804.7332,
      "train_tokens_per_second": 62275.544
    },
    {
      "epoch": 0.7569384921030425,
      "grad_norm": 17.875,
      "learning_rate": 2.9422594261732874e-06,
      "loss": 0.8483,
      "num_input_tokens_seen": 4224645376,
      "step": 56960,
      "train_runtime": 67838.0476,
      "train_tokens_per_second": 62275.456
    },
    {
      "epoch": 0.7570713815855045,
      "grad_norm": 14.25,
      "learning_rate": 2.939210992419299e-06,
      "loss": 0.9183,
      "num_input_tokens_seen": 4225517824,
      "step": 56970,
      "train_runtime": 67851.8062,
      "train_tokens_per_second": 62275.687
    },
    {
      "epoch": 0.7572042710679663,
      "grad_norm": 28.0,
      "learning_rate": 2.9361638666135152e-06,
      "loss": 0.7072,
      "num_input_tokens_seen": 4226007808,
      "step": 56980,
      "train_runtime": 67858.6237,
      "train_tokens_per_second": 62276.651
    },
    {
      "epoch": 0.7573371605504282,
      "grad_norm": 15.0,
      "learning_rate": 2.9331180493203903e-06,
      "loss": 0.6472,
      "num_input_tokens_seen": 4226240000,
      "step": 56990,
      "train_runtime": 67862.0901,
      "train_tokens_per_second": 62276.891
    },
    {
      "epoch": 0.7574700500328901,
      "grad_norm": 11.75,
      "learning_rate": 2.9300735411041293e-06,
      "loss": 0.5033,
      "num_input_tokens_seen": 4226352640,
      "step": 57000,
      "train_runtime": 67864.4623,
      "train_tokens_per_second": 62276.374
    },
    {
      "epoch": 0.7576029395153521,
      "grad_norm": 35.0,
      "learning_rate": 2.927030342528704e-06,
      "loss": 0.778,
      "num_input_tokens_seen": 4228396544,
      "step": 57010,
      "train_runtime": 67897.8411,
      "train_tokens_per_second": 62275.861
    },
    {
      "epoch": 0.757735828997814,
      "grad_norm": 20.75,
      "learning_rate": 2.923988454157839e-06,
      "loss": 0.8611,
      "num_input_tokens_seen": 4229239552,
      "step": 57020,
      "train_runtime": 67911.2117,
      "train_tokens_per_second": 62276.014
    },
    {
      "epoch": 0.7578687184802759,
      "grad_norm": 11.75,
      "learning_rate": 2.920947876555017e-06,
      "loss": 0.7494,
      "num_input_tokens_seen": 4229701376,
      "step": 57030,
      "train_runtime": 67917.8238,
      "train_tokens_per_second": 62276.751
    },
    {
      "epoch": 0.7580016079627377,
      "grad_norm": 15.3125,
      "learning_rate": 2.9179086102834788e-06,
      "loss": 0.5694,
      "num_input_tokens_seen": 4229934848,
      "step": 57040,
      "train_runtime": 67921.3275,
      "train_tokens_per_second": 62276.976
    },
    {
      "epoch": 0.7581344974451997,
      "grad_norm": 12.625,
      "learning_rate": 2.9148706559062234e-06,
      "loss": 0.552,
      "num_input_tokens_seen": 4230053376,
      "step": 57050,
      "train_runtime": 67923.7786,
      "train_tokens_per_second": 62276.473
    },
    {
      "epoch": 0.7582673869276616,
      "grad_norm": 23.5,
      "learning_rate": 2.911834013985999e-06,
      "loss": 0.7296,
      "num_input_tokens_seen": 4232065024,
      "step": 57060,
      "train_runtime": 67956.5039,
      "train_tokens_per_second": 62276.085
    },
    {
      "epoch": 0.7584002764101235,
      "grad_norm": 16.25,
      "learning_rate": 2.90879868508532e-06,
      "loss": 0.9438,
      "num_input_tokens_seen": 4232893440,
      "step": 57070,
      "train_runtime": 67968.8153,
      "train_tokens_per_second": 62276.993
    },
    {
      "epoch": 0.7585331658925855,
      "grad_norm": 26.5,
      "learning_rate": 2.905764669766452e-06,
      "loss": 0.6992,
      "num_input_tokens_seen": 4233363200,
      "step": 57080,
      "train_runtime": 67976.5048,
      "train_tokens_per_second": 62276.859
    },
    {
      "epoch": 0.7586660553750474,
      "grad_norm": 31.75,
      "learning_rate": 2.9027319685914222e-06,
      "loss": 0.6391,
      "num_input_tokens_seen": 4233593856,
      "step": 57090,
      "train_runtime": 67979.9769,
      "train_tokens_per_second": 62277.071
    },
    {
      "epoch": 0.7587989448575092,
      "grad_norm": 8.4375,
      "learning_rate": 2.8997005821220127e-06,
      "loss": 0.43,
      "num_input_tokens_seen": 4233705728,
      "step": 57100,
      "train_runtime": 67982.4204,
      "train_tokens_per_second": 62276.478
    },
    {
      "epoch": 0.7589318343399711,
      "grad_norm": 20.5,
      "learning_rate": 2.8966705109197525e-06,
      "loss": 0.8431,
      "num_input_tokens_seen": 4235875840,
      "step": 57110,
      "train_runtime": 68017.2746,
      "train_tokens_per_second": 62276.471
    },
    {
      "epoch": 0.7590647238224331,
      "grad_norm": 18.5,
      "learning_rate": 2.8936417555459463e-06,
      "loss": 0.9346,
      "num_input_tokens_seen": 4236724736,
      "step": 57120,
      "train_runtime": 68030.157,
      "train_tokens_per_second": 62277.157
    },
    {
      "epoch": 0.759197613304895,
      "grad_norm": 21.75,
      "learning_rate": 2.890614316561633e-06,
      "loss": 0.8042,
      "num_input_tokens_seen": 4237208832,
      "step": 57130,
      "train_runtime": 68037.8654,
      "train_tokens_per_second": 62277.216
    },
    {
      "epoch": 0.7593305027873569,
      "grad_norm": 12.375,
      "learning_rate": 2.88758819452763e-06,
      "loss": 0.6078,
      "num_input_tokens_seen": 4237439744,
      "step": 57140,
      "train_runtime": 68041.345,
      "train_tokens_per_second": 62277.425
    },
    {
      "epoch": 0.7594633922698188,
      "grad_norm": 8.5625,
      "learning_rate": 2.8845633900044912e-06,
      "loss": 0.5301,
      "num_input_tokens_seen": 4237555200,
      "step": 57150,
      "train_runtime": 68043.7566,
      "train_tokens_per_second": 62276.914
    },
    {
      "epoch": 0.7595962817522807,
      "grad_norm": 17.0,
      "learning_rate": 2.8815399035525372e-06,
      "loss": 0.834,
      "num_input_tokens_seen": 4239513344,
      "step": 57160,
      "train_runtime": 68075.8905,
      "train_tokens_per_second": 62276.282
    },
    {
      "epoch": 0.7597291712347426,
      "grad_norm": 21.75,
      "learning_rate": 2.878517735731845e-06,
      "loss": 0.9019,
      "num_input_tokens_seen": 4240372224,
      "step": 57170,
      "train_runtime": 68089.4153,
      "train_tokens_per_second": 62276.526
    },
    {
      "epoch": 0.7598620607172045,
      "grad_norm": 63.0,
      "learning_rate": 2.875496887102235e-06,
      "loss": 0.7988,
      "num_input_tokens_seen": 4240844288,
      "step": 57180,
      "train_runtime": 68096.0114,
      "train_tokens_per_second": 62277.426
    },
    {
      "epoch": 0.7599949501996665,
      "grad_norm": 12.8125,
      "learning_rate": 2.8724773582233044e-06,
      "loss": 0.6682,
      "num_input_tokens_seen": 4241064192,
      "step": 57190,
      "train_runtime": 68099.432,
      "train_tokens_per_second": 62277.527
    },
    {
      "epoch": 0.7601278396821284,
      "grad_norm": 6.34375,
      "learning_rate": 2.8694591496543856e-06,
      "loss": 0.5337,
      "num_input_tokens_seen": 4241178368,
      "step": 57200,
      "train_runtime": 68101.9205,
      "train_tokens_per_second": 62276.928
    },
    {
      "epoch": 0.7602607291645903,
      "grad_norm": 23.125,
      "learning_rate": 2.866442261954576e-06,
      "loss": 0.8624,
      "num_input_tokens_seen": 4243234304,
      "step": 57210,
      "train_runtime": 68135.8887,
      "train_tokens_per_second": 62276.054
    },
    {
      "epoch": 0.7603936186470521,
      "grad_norm": 14.4375,
      "learning_rate": 2.863426695682734e-06,
      "loss": 0.8709,
      "num_input_tokens_seen": 4244083712,
      "step": 57220,
      "train_runtime": 68149.4489,
      "train_tokens_per_second": 62276.127
    },
    {
      "epoch": 0.7605265081295141,
      "grad_norm": 40.0,
      "learning_rate": 2.860412451397453e-06,
      "loss": 0.7621,
      "num_input_tokens_seen": 4244555264,
      "step": 57230,
      "train_runtime": 68156.0649,
      "train_tokens_per_second": 62277.0
    },
    {
      "epoch": 0.760659397611976,
      "grad_norm": 22.5,
      "learning_rate": 2.8573995296571098e-06,
      "loss": 0.6107,
      "num_input_tokens_seen": 4244788992,
      "step": 57240,
      "train_runtime": 68159.5891,
      "train_tokens_per_second": 62277.209
    },
    {
      "epoch": 0.7607922870944379,
      "grad_norm": 27.125,
      "learning_rate": 2.8543879310198118e-06,
      "loss": 0.5295,
      "num_input_tokens_seen": 4244905472,
      "step": 57250,
      "train_runtime": 68162.0369,
      "train_tokens_per_second": 62276.682
    },
    {
      "epoch": 0.7609251765768998,
      "grad_norm": 30.5,
      "learning_rate": 2.8513776560434326e-06,
      "loss": 0.7448,
      "num_input_tokens_seen": 4246952960,
      "step": 57260,
      "train_runtime": 68195.1858,
      "train_tokens_per_second": 62276.434
    },
    {
      "epoch": 0.7610580660593618,
      "grad_norm": 12.6875,
      "learning_rate": 2.8483687052856e-06,
      "loss": 0.894,
      "num_input_tokens_seen": 4247787520,
      "step": 57270,
      "train_runtime": 68207.7451,
      "train_tokens_per_second": 62277.202
    },
    {
      "epoch": 0.7611909555418236,
      "grad_norm": 24.375,
      "learning_rate": 2.8453610793036946e-06,
      "loss": 0.6802,
      "num_input_tokens_seen": 4248275200,
      "step": 57280,
      "train_runtime": 68214.3815,
      "train_tokens_per_second": 62278.292
    },
    {
      "epoch": 0.7613238450242855,
      "grad_norm": 29.0,
      "learning_rate": 2.8423547786548523e-06,
      "loss": 0.5676,
      "num_input_tokens_seen": 4248505856,
      "step": 57290,
      "train_runtime": 68218.819,
      "train_tokens_per_second": 62277.622
    },
    {
      "epoch": 0.7614567345067474,
      "grad_norm": 30.5,
      "learning_rate": 2.8393498038959665e-06,
      "loss": 0.5012,
      "num_input_tokens_seen": 4248623104,
      "step": 57300,
      "train_runtime": 68221.41,
      "train_tokens_per_second": 62276.976
    },
    {
      "epoch": 0.7615896239892094,
      "grad_norm": 32.5,
      "learning_rate": 2.836346155583677e-06,
      "loss": 0.8385,
      "num_input_tokens_seen": 4250603264,
      "step": 57310,
      "train_runtime": 68252.5831,
      "train_tokens_per_second": 62277.544
    },
    {
      "epoch": 0.7617225134716713,
      "grad_norm": 33.5,
      "learning_rate": 2.8333438342743845e-06,
      "loss": 0.897,
      "num_input_tokens_seen": 4251412224,
      "step": 57320,
      "train_runtime": 68264.8975,
      "train_tokens_per_second": 62278.16
    },
    {
      "epoch": 0.7618554029541332,
      "grad_norm": 29.25,
      "learning_rate": 2.830342840524243e-06,
      "loss": 0.7099,
      "num_input_tokens_seen": 4251874304,
      "step": 57330,
      "train_runtime": 68271.4375,
      "train_tokens_per_second": 62278.963
    },
    {
      "epoch": 0.7619882924365952,
      "grad_norm": 20.5,
      "learning_rate": 2.827343174889159e-06,
      "loss": 0.6569,
      "num_input_tokens_seen": 4252108544,
      "step": 57340,
      "train_runtime": 68274.9635,
      "train_tokens_per_second": 62279.177
    },
    {
      "epoch": 0.762121181919057,
      "grad_norm": 8.0625,
      "learning_rate": 2.824344837924795e-06,
      "loss": 0.6048,
      "num_input_tokens_seen": 4252224512,
      "step": 57350,
      "train_runtime": 68278.2412,
      "train_tokens_per_second": 62277.886
    },
    {
      "epoch": 0.7622540714015189,
      "grad_norm": 22.25,
      "learning_rate": 2.821347830186566e-06,
      "loss": 0.8054,
      "num_input_tokens_seen": 4254245888,
      "step": 57360,
      "train_runtime": 68310.1967,
      "train_tokens_per_second": 62278.343
    },
    {
      "epoch": 0.7623869608839808,
      "grad_norm": 19.0,
      "learning_rate": 2.818352152229643e-06,
      "loss": 0.923,
      "num_input_tokens_seen": 4255107328,
      "step": 57370,
      "train_runtime": 68323.0149,
      "train_tokens_per_second": 62279.268
    },
    {
      "epoch": 0.7625198503664428,
      "grad_norm": 22.375,
      "learning_rate": 2.8153578046089434e-06,
      "loss": 0.7048,
      "num_input_tokens_seen": 4255604992,
      "step": 57380,
      "train_runtime": 68329.9056,
      "train_tokens_per_second": 62280.27
    },
    {
      "epoch": 0.7626527398489047,
      "grad_norm": 26.0,
      "learning_rate": 2.812364787879146e-06,
      "loss": 0.653,
      "num_input_tokens_seen": 4255841536,
      "step": 57390,
      "train_runtime": 68334.3712,
      "train_tokens_per_second": 62279.662
    },
    {
      "epoch": 0.7627856293313666,
      "grad_norm": 11.9375,
      "learning_rate": 2.8093731025946813e-06,
      "loss": 0.5312,
      "num_input_tokens_seen": 4255958272,
      "step": 57400,
      "train_runtime": 68336.8094,
      "train_tokens_per_second": 62279.148
    },
    {
      "epoch": 0.7629185188138284,
      "grad_norm": 15.375,
      "learning_rate": 2.806382749309732e-06,
      "loss": 0.804,
      "num_input_tokens_seen": 4258101248,
      "step": 57410,
      "train_runtime": 68370.4014,
      "train_tokens_per_second": 62279.893
    },
    {
      "epoch": 0.7630514082962904,
      "grad_norm": 12.0625,
      "learning_rate": 2.8033937285782366e-06,
      "loss": 0.8978,
      "num_input_tokens_seen": 4258928384,
      "step": 57420,
      "train_runtime": 68382.8672,
      "train_tokens_per_second": 62280.635
    },
    {
      "epoch": 0.7631842977787523,
      "grad_norm": 17.875,
      "learning_rate": 2.8004060409538758e-06,
      "loss": 0.6968,
      "num_input_tokens_seen": 4259366656,
      "step": 57430,
      "train_runtime": 68389.2083,
      "train_tokens_per_second": 62281.269
    },
    {
      "epoch": 0.7633171872612142,
      "grad_norm": 10.625,
      "learning_rate": 2.7974196869901037e-06,
      "loss": 0.6584,
      "num_input_tokens_seen": 4259588096,
      "step": 57440,
      "train_runtime": 68392.5897,
      "train_tokens_per_second": 62281.427
    },
    {
      "epoch": 0.7634500767436762,
      "grad_norm": 11.375,
      "learning_rate": 2.794434667240108e-06,
      "loss": 0.4922,
      "num_input_tokens_seen": 4259703040,
      "step": 57450,
      "train_runtime": 68395.0241,
      "train_tokens_per_second": 62280.891
    },
    {
      "epoch": 0.7635829662261381,
      "grad_norm": 33.5,
      "learning_rate": 2.7914509822568404e-06,
      "loss": 0.7557,
      "num_input_tokens_seen": 4261790720,
      "step": 57460,
      "train_runtime": 68430.6957,
      "train_tokens_per_second": 62278.933
    },
    {
      "epoch": 0.7637158557085999,
      "grad_norm": 66.5,
      "learning_rate": 2.788468632593002e-06,
      "loss": 0.9339,
      "num_input_tokens_seen": 4262625024,
      "step": 57470,
      "train_runtime": 68443.3421,
      "train_tokens_per_second": 62279.615
    },
    {
      "epoch": 0.7638487451910618,
      "grad_norm": 12.4375,
      "learning_rate": 2.7854876188010404e-06,
      "loss": 0.7549,
      "num_input_tokens_seen": 4263093760,
      "step": 57480,
      "train_runtime": 68450.0503,
      "train_tokens_per_second": 62280.36
    },
    {
      "epoch": 0.7639816346735238,
      "grad_norm": 24.25,
      "learning_rate": 2.7825079414331734e-06,
      "loss": 0.6823,
      "num_input_tokens_seen": 4263322368,
      "step": 57490,
      "train_runtime": 68453.5471,
      "train_tokens_per_second": 62280.518
    },
    {
      "epoch": 0.7641145241559857,
      "grad_norm": 8.8125,
      "learning_rate": 2.7795296010413466e-06,
      "loss": 0.5368,
      "num_input_tokens_seen": 4263439872,
      "step": 57500,
      "train_runtime": 68455.9943,
      "train_tokens_per_second": 62280.008
    },
    {
      "epoch": 0.7642474136384476,
      "grad_norm": 20.125,
      "learning_rate": 2.7765525981772823e-06,
      "loss": 0.8845,
      "num_input_tokens_seen": 4265565952,
      "step": 57510,
      "train_runtime": 68492.1498,
      "train_tokens_per_second": 62278.173
    },
    {
      "epoch": 0.7643803031209095,
      "grad_norm": 24.625,
      "learning_rate": 2.773576933392437e-06,
      "loss": 0.9376,
      "num_input_tokens_seen": 4266376960,
      "step": 57520,
      "train_runtime": 68504.4734,
      "train_tokens_per_second": 62278.808
    },
    {
      "epoch": 0.7645131926033714,
      "grad_norm": 22.0,
      "learning_rate": 2.7706026072380277e-06,
      "loss": 0.7266,
      "num_input_tokens_seen": 4266862080,
      "step": 57530,
      "train_runtime": 68511.3119,
      "train_tokens_per_second": 62279.673
    },
    {
      "epoch": 0.7646460820858333,
      "grad_norm": 25.5,
      "learning_rate": 2.7676296202650255e-06,
      "loss": 0.6686,
      "num_input_tokens_seen": 4267098624,
      "step": 57540,
      "train_runtime": 68514.8551,
      "train_tokens_per_second": 62279.904
    },
    {
      "epoch": 0.7647789715682952,
      "grad_norm": 120.5,
      "learning_rate": 2.7646579730241395e-06,
      "loss": 0.5792,
      "num_input_tokens_seen": 4267219456,
      "step": 57550,
      "train_runtime": 68517.3794,
      "train_tokens_per_second": 62279.373
    },
    {
      "epoch": 0.7649118610507571,
      "grad_norm": 28.625,
      "learning_rate": 2.7616876660658544e-06,
      "loss": 0.8644,
      "num_input_tokens_seen": 4269377024,
      "step": 57560,
      "train_runtime": 68551.5036,
      "train_tokens_per_second": 62279.845
    },
    {
      "epoch": 0.7650447505332191,
      "grad_norm": 13.3125,
      "learning_rate": 2.7587186999403825e-06,
      "loss": 0.8926,
      "num_input_tokens_seen": 4270235136,
      "step": 57570,
      "train_runtime": 68565.2637,
      "train_tokens_per_second": 62279.862
    },
    {
      "epoch": 0.765177640015681,
      "grad_norm": 16.0,
      "learning_rate": 2.7557510751977034e-06,
      "loss": 0.791,
      "num_input_tokens_seen": 4270706176,
      "step": 57580,
      "train_runtime": 68572.0375,
      "train_tokens_per_second": 62280.579
    },
    {
      "epoch": 0.7653105294981428,
      "grad_norm": 9.0625,
      "learning_rate": 2.7527847923875405e-06,
      "loss": 0.554,
      "num_input_tokens_seen": 4270938368,
      "step": 57590,
      "train_runtime": 68575.566,
      "train_tokens_per_second": 62280.76
    },
    {
      "epoch": 0.7654434189806048,
      "grad_norm": 8.875,
      "learning_rate": 2.749819852059373e-06,
      "loss": 0.5111,
      "num_input_tokens_seen": 4271054592,
      "step": 57600,
      "train_runtime": 68578.034,
      "train_tokens_per_second": 62280.213
    },
    {
      "epoch": 0.7655763084630667,
      "grad_norm": 21.125,
      "learning_rate": 2.74685625476243e-06,
      "loss": 0.8391,
      "num_input_tokens_seen": 4273111552,
      "step": 57610,
      "train_runtime": 68610.4287,
      "train_tokens_per_second": 62280.788
    },
    {
      "epoch": 0.7657091979455286,
      "grad_norm": 13.5,
      "learning_rate": 2.7438940010456926e-06,
      "loss": 0.9122,
      "num_input_tokens_seen": 4273936384,
      "step": 57620,
      "train_runtime": 68622.9921,
      "train_tokens_per_second": 62281.405
    },
    {
      "epoch": 0.7658420874279905,
      "grad_norm": 17.875,
      "learning_rate": 2.740933091457888e-06,
      "loss": 0.7793,
      "num_input_tokens_seen": 4274426112,
      "step": 57630,
      "train_runtime": 68630.6846,
      "train_tokens_per_second": 62281.56
    },
    {
      "epoch": 0.7659749769104525,
      "grad_norm": 10.3125,
      "learning_rate": 2.7379735265474994e-06,
      "loss": 0.6301,
      "num_input_tokens_seen": 4274653696,
      "step": 57640,
      "train_runtime": 68634.1256,
      "train_tokens_per_second": 62281.754
    },
    {
      "epoch": 0.7661078663929143,
      "grad_norm": 22.875,
      "learning_rate": 2.735015306862762e-06,
      "loss": 0.5264,
      "num_input_tokens_seen": 4274766848,
      "step": 57650,
      "train_runtime": 68636.5488,
      "train_tokens_per_second": 62281.203
    },
    {
      "epoch": 0.7662407558753762,
      "grad_norm": 36.5,
      "learning_rate": 2.7320584329516564e-06,
      "loss": 0.8345,
      "num_input_tokens_seen": 4276702208,
      "step": 57660,
      "train_runtime": 68668.4553,
      "train_tokens_per_second": 62280.449
    },
    {
      "epoch": 0.7663736453578381,
      "grad_norm": 18.25,
      "learning_rate": 2.72910290536192e-06,
      "loss": 0.8765,
      "num_input_tokens_seen": 4277515264,
      "step": 57670,
      "train_runtime": 68680.6661,
      "train_tokens_per_second": 62281.214
    },
    {
      "epoch": 0.7665065348403001,
      "grad_norm": 29.25,
      "learning_rate": 2.7261487246410377e-06,
      "loss": 0.756,
      "num_input_tokens_seen": 4277960704,
      "step": 57680,
      "train_runtime": 68688.5706,
      "train_tokens_per_second": 62280.532
    },
    {
      "epoch": 0.766639424322762,
      "grad_norm": 9.9375,
      "learning_rate": 2.7231958913362467e-06,
      "loss": 0.622,
      "num_input_tokens_seen": 4278180864,
      "step": 57690,
      "train_runtime": 68692.0769,
      "train_tokens_per_second": 62280.558
    },
    {
      "epoch": 0.7667723138052239,
      "grad_norm": 13.3125,
      "learning_rate": 2.7202444059945275e-06,
      "loss": 0.5166,
      "num_input_tokens_seen": 4278292992,
      "step": 57700,
      "train_runtime": 68694.73,
      "train_tokens_per_second": 62279.785
    },
    {
      "epoch": 0.7669052032876857,
      "grad_norm": 24.375,
      "learning_rate": 2.7172942691626202e-06,
      "loss": 0.8653,
      "num_input_tokens_seen": 4280310016,
      "step": 57710,
      "train_runtime": 68727.6399,
      "train_tokens_per_second": 62279.31
    },
    {
      "epoch": 0.7670380927701477,
      "grad_norm": 31.5,
      "learning_rate": 2.7143454813870108e-06,
      "loss": 0.8936,
      "num_input_tokens_seen": 4281176576,
      "step": 57720,
      "train_runtime": 68740.4656,
      "train_tokens_per_second": 62280.296
    },
    {
      "epoch": 0.7671709822526096,
      "grad_norm": 31.25,
      "learning_rate": 2.7113980432139366e-06,
      "loss": 0.7672,
      "num_input_tokens_seen": 4281664768,
      "step": 57730,
      "train_runtime": 68748.6628,
      "train_tokens_per_second": 62279.972
    },
    {
      "epoch": 0.7673038717350715,
      "grad_norm": 27.5,
      "learning_rate": 2.708451955189385e-06,
      "loss": 0.7318,
      "num_input_tokens_seen": 4281899776,
      "step": 57740,
      "train_runtime": 68752.3569,
      "train_tokens_per_second": 62280.043
    },
    {
      "epoch": 0.7674367612175335,
      "grad_norm": 10.5625,
      "learning_rate": 2.705507217859087e-06,
      "loss": 0.4606,
      "num_input_tokens_seen": 4282015232,
      "step": 57750,
      "train_runtime": 68754.8871,
      "train_tokens_per_second": 62279.431
    },
    {
      "epoch": 0.7675696506999954,
      "grad_norm": 18.5,
      "learning_rate": 2.7025638317685375e-06,
      "loss": 0.8387,
      "num_input_tokens_seen": 4284141056,
      "step": 57760,
      "train_runtime": 68788.284,
      "train_tokens_per_second": 62280.098
    },
    {
      "epoch": 0.7677025401824572,
      "grad_norm": 16.0,
      "learning_rate": 2.699621797462966e-06,
      "loss": 0.8663,
      "num_input_tokens_seen": 4284992512,
      "step": 57770,
      "train_runtime": 68800.7402,
      "train_tokens_per_second": 62281.198
    },
    {
      "epoch": 0.7678354296649191,
      "grad_norm": 21.5,
      "learning_rate": 2.6966811154873596e-06,
      "loss": 0.754,
      "num_input_tokens_seen": 4285459456,
      "step": 57780,
      "train_runtime": 68807.3082,
      "train_tokens_per_second": 62282.039
    },
    {
      "epoch": 0.7679683191473811,
      "grad_norm": 24.75,
      "learning_rate": 2.693741786386458e-06,
      "loss": 0.662,
      "num_input_tokens_seen": 4285687040,
      "step": 57790,
      "train_runtime": 68812.065,
      "train_tokens_per_second": 62281.041
    },
    {
      "epoch": 0.768101208629843,
      "grad_norm": 11.9375,
      "learning_rate": 2.690803810704735e-06,
      "loss": 0.5741,
      "num_input_tokens_seen": 4285800704,
      "step": 57800,
      "train_runtime": 68814.6572,
      "train_tokens_per_second": 62280.347
    },
    {
      "epoch": 0.7682340981123049,
      "grad_norm": 29.125,
      "learning_rate": 2.687867188986438e-06,
      "loss": 0.8415,
      "num_input_tokens_seen": 4287832064,
      "step": 57810,
      "train_runtime": 68846.6737,
      "train_tokens_per_second": 62280.889
    },
    {
      "epoch": 0.7683669875947668,
      "grad_norm": 90.0,
      "learning_rate": 2.68493192177554e-06,
      "loss": 0.8964,
      "num_input_tokens_seen": 4288670464,
      "step": 57820,
      "train_runtime": 68859.3562,
      "train_tokens_per_second": 62281.594
    },
    {
      "epoch": 0.7684998770772288,
      "grad_norm": 19.0,
      "learning_rate": 2.681998009615775e-06,
      "loss": 0.7336,
      "num_input_tokens_seen": 4289122816,
      "step": 57830,
      "train_runtime": 68866.0511,
      "train_tokens_per_second": 62282.108
    },
    {
      "epoch": 0.7686327665596906,
      "grad_norm": 72.5,
      "learning_rate": 2.6790654530506266e-06,
      "loss": 0.6384,
      "num_input_tokens_seen": 4289346816,
      "step": 57840,
      "train_runtime": 68871.0913,
      "train_tokens_per_second": 62280.802
    },
    {
      "epoch": 0.7687656560421525,
      "grad_norm": 16.625,
      "learning_rate": 2.6761342526233225e-06,
      "loss": 0.4891,
      "num_input_tokens_seen": 4289461248,
      "step": 57850,
      "train_runtime": 68873.6271,
      "train_tokens_per_second": 62280.171
    },
    {
      "epoch": 0.7688985455246145,
      "grad_norm": 35.0,
      "learning_rate": 2.673204408876846e-06,
      "loss": 0.8253,
      "num_input_tokens_seen": 4291546624,
      "step": 57860,
      "train_runtime": 68907.0582,
      "train_tokens_per_second": 62280.218
    },
    {
      "epoch": 0.7690314350070764,
      "grad_norm": 20.75,
      "learning_rate": 2.6702759223539155e-06,
      "loss": 0.8798,
      "num_input_tokens_seen": 4292365824,
      "step": 57870,
      "train_runtime": 68919.3918,
      "train_tokens_per_second": 62280.959
    },
    {
      "epoch": 0.7691643244895383,
      "grad_norm": 16.375,
      "learning_rate": 2.6673487935970178e-06,
      "loss": 0.7438,
      "num_input_tokens_seen": 4292841472,
      "step": 57880,
      "train_runtime": 68926.2286,
      "train_tokens_per_second": 62281.682
    },
    {
      "epoch": 0.7692972139720002,
      "grad_norm": 32.25,
      "learning_rate": 2.6644230231483693e-06,
      "loss": 0.6402,
      "num_input_tokens_seen": 4293068288,
      "step": 57890,
      "train_runtime": 68930.8876,
      "train_tokens_per_second": 62280.763
    },
    {
      "epoch": 0.769430103454462,
      "grad_norm": 12.0625,
      "learning_rate": 2.6614986115499455e-06,
      "loss": 0.5643,
      "num_input_tokens_seen": 4293182720,
      "step": 57900,
      "train_runtime": 68933.6261,
      "train_tokens_per_second": 62279.949
    },
    {
      "epoch": 0.769562992936924,
      "grad_norm": 18.75,
      "learning_rate": 2.658575559343467e-06,
      "loss": 0.8249,
      "num_input_tokens_seen": 4295187456,
      "step": 57910,
      "train_runtime": 68965.37,
      "train_tokens_per_second": 62280.351
    },
    {
      "epoch": 0.7696958824193859,
      "grad_norm": 18.25,
      "learning_rate": 2.655653867070405e-06,
      "loss": 0.8799,
      "num_input_tokens_seen": 4296017152,
      "step": 57920,
      "train_runtime": 68977.5605,
      "train_tokens_per_second": 62281.373
    },
    {
      "epoch": 0.7698287719018478,
      "grad_norm": 25.375,
      "learning_rate": 2.652733535271974e-06,
      "loss": 0.7499,
      "num_input_tokens_seen": 4296477440,
      "step": 57930,
      "train_runtime": 68983.9839,
      "train_tokens_per_second": 62282.246
    },
    {
      "epoch": 0.7699616613843098,
      "grad_norm": 15.75,
      "learning_rate": 2.6498145644891458e-06,
      "loss": 0.5233,
      "num_input_tokens_seen": 4296707584,
      "step": 57940,
      "train_runtime": 68987.5678,
      "train_tokens_per_second": 62282.346
    },
    {
      "epoch": 0.7700945508667717,
      "grad_norm": 9.6875,
      "learning_rate": 2.6468969552626256e-06,
      "loss": 0.5493,
      "num_input_tokens_seen": 4296826112,
      "step": 57950,
      "train_runtime": 68990.2601,
      "train_tokens_per_second": 62281.634
    },
    {
      "epoch": 0.7702274403492335,
      "grad_norm": 19.125,
      "learning_rate": 2.643980708132878e-06,
      "loss": 0.7968,
      "num_input_tokens_seen": 4298872576,
      "step": 57960,
      "train_runtime": 69023.6441,
      "train_tokens_per_second": 62281.159
    },
    {
      "epoch": 0.7703603298316954,
      "grad_norm": 30.0,
      "learning_rate": 2.641065823640112e-06,
      "loss": 0.9176,
      "num_input_tokens_seen": 4299690240,
      "step": 57970,
      "train_runtime": 69036.1892,
      "train_tokens_per_second": 62281.686
    },
    {
      "epoch": 0.7704932193141574,
      "grad_norm": 26.5,
      "learning_rate": 2.638152302324284e-06,
      "loss": 0.7265,
      "num_input_tokens_seen": 4300176128,
      "step": 57980,
      "train_runtime": 69043.0277,
      "train_tokens_per_second": 62282.554
    },
    {
      "epoch": 0.7706261087966193,
      "grad_norm": 16.125,
      "learning_rate": 2.635240144725101e-06,
      "loss": 0.6498,
      "num_input_tokens_seen": 4300403968,
      "step": 57990,
      "train_runtime": 69046.6352,
      "train_tokens_per_second": 62282.6
    },
    {
      "epoch": 0.7707589982790812,
      "grad_norm": 12.0625,
      "learning_rate": 2.6323293513820034e-06,
      "loss": 0.4898,
      "num_input_tokens_seen": 4300521216,
      "step": 58000,
      "train_runtime": 69049.3327,
      "train_tokens_per_second": 62281.865
    },
    {
      "epoch": 0.7708918877615432,
      "grad_norm": 21.75,
      "learning_rate": 2.6294199228342045e-06,
      "loss": 0.7873,
      "num_input_tokens_seen": 4302583552,
      "step": 58010,
      "train_runtime": 69083.237,
      "train_tokens_per_second": 62281.152
    },
    {
      "epoch": 0.771024777244005,
      "grad_norm": 23.25,
      "learning_rate": 2.6265118596206398e-06,
      "loss": 0.8818,
      "num_input_tokens_seen": 4303435008,
      "step": 58020,
      "train_runtime": 69095.8061,
      "train_tokens_per_second": 62282.145
    },
    {
      "epoch": 0.7711576667264669,
      "grad_norm": 22.75,
      "learning_rate": 2.623605162280004e-06,
      "loss": 0.709,
      "num_input_tokens_seen": 4303936256,
      "step": 58030,
      "train_runtime": 69103.8545,
      "train_tokens_per_second": 62282.145
    },
    {
      "epoch": 0.7712905562089288,
      "grad_norm": 23.5,
      "learning_rate": 2.6206998313507382e-06,
      "loss": 0.6517,
      "num_input_tokens_seen": 4304169472,
      "step": 58040,
      "train_runtime": 69107.4206,
      "train_tokens_per_second": 62282.305
    },
    {
      "epoch": 0.7714234456913908,
      "grad_norm": 10.3125,
      "learning_rate": 2.617795867371028e-06,
      "loss": 0.5174,
      "num_input_tokens_seen": 4304281856,
      "step": 58050,
      "train_runtime": 69110.0658,
      "train_tokens_per_second": 62281.548
    },
    {
      "epoch": 0.7715563351738527,
      "grad_norm": 14.5625,
      "learning_rate": 2.614893270878811e-06,
      "loss": 0.8096,
      "num_input_tokens_seen": 4306357248,
      "step": 58060,
      "train_runtime": 69143.3271,
      "train_tokens_per_second": 62281.603
    },
    {
      "epoch": 0.7716892246563146,
      "grad_norm": 15.375,
      "learning_rate": 2.6119920424117574e-06,
      "loss": 0.8969,
      "num_input_tokens_seen": 4307186944,
      "step": 58070,
      "train_runtime": 69155.7691,
      "train_tokens_per_second": 62282.395
    },
    {
      "epoch": 0.7718221141387764,
      "grad_norm": 17.375,
      "learning_rate": 2.6090921825073066e-06,
      "loss": 0.7081,
      "num_input_tokens_seen": 4307646720,
      "step": 58080,
      "train_runtime": 69163.5914,
      "train_tokens_per_second": 62281.999
    },
    {
      "epoch": 0.7719550036212384,
      "grad_norm": 21.125,
      "learning_rate": 2.606193691702623e-06,
      "loss": 0.649,
      "num_input_tokens_seen": 4307871744,
      "step": 58090,
      "train_runtime": 69167.0801,
      "train_tokens_per_second": 62282.111
    },
    {
      "epoch": 0.7720878931037003,
      "grad_norm": 11.6875,
      "learning_rate": 2.603296570534628e-06,
      "loss": 0.494,
      "num_input_tokens_seen": 4307983360,
      "step": 58100,
      "train_runtime": 69169.6978,
      "train_tokens_per_second": 62281.367
    },
    {
      "epoch": 0.7722207825861622,
      "grad_norm": 18.5,
      "learning_rate": 2.600400819539991e-06,
      "loss": 0.8713,
      "num_input_tokens_seen": 4310014208,
      "step": 58110,
      "train_runtime": 69202.5653,
      "train_tokens_per_second": 62281.134
    },
    {
      "epoch": 0.7723536720686242,
      "grad_norm": 18.125,
      "learning_rate": 2.597506439255115e-06,
      "loss": 0.9021,
      "num_input_tokens_seen": 4310861568,
      "step": 58120,
      "train_runtime": 69216.7027,
      "train_tokens_per_second": 62280.655
    },
    {
      "epoch": 0.7724865615510861,
      "grad_norm": 10.625,
      "learning_rate": 2.594613430216172e-06,
      "loss": 0.7782,
      "num_input_tokens_seen": 4311334400,
      "step": 58130,
      "train_runtime": 69223.5263,
      "train_tokens_per_second": 62281.346
    },
    {
      "epoch": 0.7726194510335479,
      "grad_norm": 9.5,
      "learning_rate": 2.5917217929590534e-06,
      "loss": 0.6618,
      "num_input_tokens_seen": 4311571712,
      "step": 58140,
      "train_runtime": 69227.1934,
      "train_tokens_per_second": 62281.475
    },
    {
      "epoch": 0.7727523405160098,
      "grad_norm": 12.3125,
      "learning_rate": 2.5888315280194154e-06,
      "loss": 0.5113,
      "num_input_tokens_seen": 4311686656,
      "step": 58150,
      "train_runtime": 69229.7842,
      "train_tokens_per_second": 62280.805
    },
    {
      "epoch": 0.7728852299984718,
      "grad_norm": 22.25,
      "learning_rate": 2.585942635932652e-06,
      "loss": 0.7589,
      "num_input_tokens_seen": 4313796864,
      "step": 58160,
      "train_runtime": 69263.6715,
      "train_tokens_per_second": 62280.8
    },
    {
      "epoch": 0.7730181194809337,
      "grad_norm": 105.0,
      "learning_rate": 2.5830551172339056e-06,
      "loss": 0.8436,
      "num_input_tokens_seen": 4314658048,
      "step": 58170,
      "train_runtime": 69276.7045,
      "train_tokens_per_second": 62281.514
    },
    {
      "epoch": 0.7731510089633956,
      "grad_norm": 15.6875,
      "learning_rate": 2.5801689724580657e-06,
      "loss": 0.7388,
      "num_input_tokens_seen": 4315148800,
      "step": 58180,
      "train_runtime": 69283.7066,
      "train_tokens_per_second": 62282.303
    },
    {
      "epoch": 0.7732838984458575,
      "grad_norm": 16.75,
      "learning_rate": 2.577284202139757e-06,
      "loss": 0.6503,
      "num_input_tokens_seen": 4315384064,
      "step": 58190,
      "train_runtime": 69287.313,
      "train_tokens_per_second": 62282.457
    },
    {
      "epoch": 0.7734167879283194,
      "grad_norm": 44.25,
      "learning_rate": 2.5744008068133685e-06,
      "loss": 0.4957,
      "num_input_tokens_seen": 4315498496,
      "step": 58200,
      "train_runtime": 69291.1447,
      "train_tokens_per_second": 62280.664
    },
    {
      "epoch": 0.7735496774107813,
      "grad_norm": 28.25,
      "learning_rate": 2.571518787013014e-06,
      "loss": 0.8188,
      "num_input_tokens_seen": 4317424640,
      "step": 58210,
      "train_runtime": 69323.0774,
      "train_tokens_per_second": 62279.76
    },
    {
      "epoch": 0.7736825668932432,
      "grad_norm": 16.125,
      "learning_rate": 2.568638143272567e-06,
      "loss": 0.8795,
      "num_input_tokens_seen": 4318285056,
      "step": 58220,
      "train_runtime": 69336.1748,
      "train_tokens_per_second": 62280.405
    },
    {
      "epoch": 0.7738154563757051,
      "grad_norm": 18.875,
      "learning_rate": 2.565758876125639e-06,
      "loss": 0.7778,
      "num_input_tokens_seen": 4318778624,
      "step": 58230,
      "train_runtime": 69343.0669,
      "train_tokens_per_second": 62281.333
    },
    {
      "epoch": 0.7739483458581671,
      "grad_norm": 13.9375,
      "learning_rate": 2.5628809861055893e-06,
      "loss": 0.6,
      "num_input_tokens_seen": 4319015168,
      "step": 58240,
      "train_runtime": 69346.6714,
      "train_tokens_per_second": 62281.507
    },
    {
      "epoch": 0.774081235340629,
      "grad_norm": 9.8125,
      "learning_rate": 2.5600044737455208e-06,
      "loss": 0.6026,
      "num_input_tokens_seen": 4319134208,
      "step": 58250,
      "train_runtime": 69349.2691,
      "train_tokens_per_second": 62280.89
    },
    {
      "epoch": 0.7742141248230908,
      "grad_norm": 22.125,
      "learning_rate": 2.5571293395782858e-06,
      "loss": 0.9021,
      "num_input_tokens_seen": 4321190656,
      "step": 58260,
      "train_runtime": 69384.2482,
      "train_tokens_per_second": 62279.131
    },
    {
      "epoch": 0.7743470143055527,
      "grad_norm": 14.1875,
      "learning_rate": 2.5542555841364715e-06,
      "loss": 0.9392,
      "num_input_tokens_seen": 4322075392,
      "step": 58270,
      "train_runtime": 69397.353,
      "train_tokens_per_second": 62280.119
    },
    {
      "epoch": 0.7744799037880147,
      "grad_norm": 31.75,
      "learning_rate": 2.551383207952417e-06,
      "loss": 0.7884,
      "num_input_tokens_seen": 4322552576,
      "step": 58280,
      "train_runtime": 69404.2981,
      "train_tokens_per_second": 62280.762
    },
    {
      "epoch": 0.7746127932704766,
      "grad_norm": 20.0,
      "learning_rate": 2.548512211558205e-06,
      "loss": 0.6164,
      "num_input_tokens_seen": 4322787840,
      "step": 58290,
      "train_runtime": 69407.7905,
      "train_tokens_per_second": 62281.018
    },
    {
      "epoch": 0.7747456827529385,
      "grad_norm": 48.25,
      "learning_rate": 2.5456425954856623e-06,
      "loss": 0.5077,
      "num_input_tokens_seen": 4322902016,
      "step": 58300,
      "train_runtime": 69410.3353,
      "train_tokens_per_second": 62280.379
    },
    {
      "epoch": 0.7748785722354005,
      "grad_norm": 15.5625,
      "learning_rate": 2.5427743602663624e-06,
      "loss": 0.7979,
      "num_input_tokens_seen": 4324982272,
      "step": 58310,
      "train_runtime": 69444.8386,
      "train_tokens_per_second": 62279.391
    },
    {
      "epoch": 0.7750114617178624,
      "grad_norm": 12.125,
      "learning_rate": 2.5399075064316115e-06,
      "loss": 0.9328,
      "num_input_tokens_seen": 4325826560,
      "step": 58320,
      "train_runtime": 69457.2401,
      "train_tokens_per_second": 62280.427
    },
    {
      "epoch": 0.7751443512003242,
      "grad_norm": 17.375,
      "learning_rate": 2.5370420345124814e-06,
      "loss": 0.7041,
      "num_input_tokens_seen": 4326289920,
      "step": 58330,
      "train_runtime": 69463.8045,
      "train_tokens_per_second": 62281.212
    },
    {
      "epoch": 0.7752772406827861,
      "grad_norm": 14.0625,
      "learning_rate": 2.534177945039764e-06,
      "loss": 0.7148,
      "num_input_tokens_seen": 4326518784,
      "step": 58340,
      "train_runtime": 69467.2486,
      "train_tokens_per_second": 62281.419
    },
    {
      "epoch": 0.7754101301652481,
      "grad_norm": 10.0625,
      "learning_rate": 2.5313152385440122e-06,
      "loss": 0.5053,
      "num_input_tokens_seen": 4326633216,
      "step": 58350,
      "train_runtime": 69469.67,
      "train_tokens_per_second": 62280.895
    },
    {
      "epoch": 0.77554301964771,
      "grad_norm": 19.25,
      "learning_rate": 2.5284539155555175e-06,
      "loss": 0.776,
      "num_input_tokens_seen": 4328867840,
      "step": 58360,
      "train_runtime": 69505.1215,
      "train_tokens_per_second": 62281.279
    },
    {
      "epoch": 0.7756759091301719,
      "grad_norm": 18.625,
      "learning_rate": 2.525593976604307e-06,
      "loss": 0.8937,
      "num_input_tokens_seen": 4329728768,
      "step": 58370,
      "train_runtime": 69517.9956,
      "train_tokens_per_second": 62282.129
    },
    {
      "epoch": 0.7758087986126339,
      "grad_norm": 20.5,
      "learning_rate": 2.5227354222201692e-06,
      "loss": 0.7513,
      "num_input_tokens_seen": 4330193152,
      "step": 58380,
      "train_runtime": 69525.8359,
      "train_tokens_per_second": 62281.785
    },
    {
      "epoch": 0.7759416880950957,
      "grad_norm": 21.875,
      "learning_rate": 2.519878252932616e-06,
      "loss": 0.7049,
      "num_input_tokens_seen": 4330433536,
      "step": 58390,
      "train_runtime": 69529.4223,
      "train_tokens_per_second": 62282.03
    },
    {
      "epoch": 0.7760745775775576,
      "grad_norm": 9.9375,
      "learning_rate": 2.5170224692709234e-06,
      "loss": 0.5628,
      "num_input_tokens_seen": 4330552064,
      "step": 58400,
      "train_runtime": 69531.9117,
      "train_tokens_per_second": 62281.504
    },
    {
      "epoch": 0.7762074670600195,
      "grad_norm": 44.5,
      "learning_rate": 2.5141680717640905e-06,
      "loss": 0.7379,
      "num_input_tokens_seen": 4332644608,
      "step": 58410,
      "train_runtime": 69565.7269,
      "train_tokens_per_second": 62281.31
    },
    {
      "epoch": 0.7763403565424815,
      "grad_norm": 43.5,
      "learning_rate": 2.511315060940872e-06,
      "loss": 0.9394,
      "num_input_tokens_seen": 4333495808,
      "step": 58420,
      "train_runtime": 69578.7227,
      "train_tokens_per_second": 62281.911
    },
    {
      "epoch": 0.7764732460249434,
      "grad_norm": 18.0,
      "learning_rate": 2.508463437329767e-06,
      "loss": 0.7752,
      "num_input_tokens_seen": 4333981184,
      "step": 58430,
      "train_runtime": 69586.9902,
      "train_tokens_per_second": 62281.486
    },
    {
      "epoch": 0.7766061355074053,
      "grad_norm": 22.125,
      "learning_rate": 2.505613201459003e-06,
      "loss": 0.6561,
      "num_input_tokens_seen": 4334210560,
      "step": 58440,
      "train_runtime": 69590.5662,
      "train_tokens_per_second": 62281.582
    },
    {
      "epoch": 0.7767390249898671,
      "grad_norm": 21.625,
      "learning_rate": 2.502764353856573e-06,
      "loss": 0.5484,
      "num_input_tokens_seen": 4334324992,
      "step": 58450,
      "train_runtime": 69593.0376,
      "train_tokens_per_second": 62281.015
    },
    {
      "epoch": 0.7768719144723291,
      "grad_norm": 16.75,
      "learning_rate": 2.4999168950501918e-06,
      "loss": 0.8621,
      "num_input_tokens_seen": 4336379648,
      "step": 58460,
      "train_runtime": 69625.3332,
      "train_tokens_per_second": 62281.636
    },
    {
      "epoch": 0.777004803954791,
      "grad_norm": 17.875,
      "learning_rate": 2.497070825567328e-06,
      "loss": 0.8137,
      "num_input_tokens_seen": 4337199616,
      "step": 58470,
      "train_runtime": 69639.2636,
      "train_tokens_per_second": 62280.952
    },
    {
      "epoch": 0.7771376934372529,
      "grad_norm": 22.25,
      "learning_rate": 2.4942261459351926e-06,
      "loss": 0.8244,
      "num_input_tokens_seen": 4337658624,
      "step": 58480,
      "train_runtime": 69646.3016,
      "train_tokens_per_second": 62281.249
    },
    {
      "epoch": 0.7772705829197148,
      "grad_norm": 27.875,
      "learning_rate": 2.4913828566807363e-06,
      "loss": 0.7102,
      "num_input_tokens_seen": 4337885184,
      "step": 58490,
      "train_runtime": 69650.7528,
      "train_tokens_per_second": 62280.521
    },
    {
      "epoch": 0.7774034724021768,
      "grad_norm": 18.375,
      "learning_rate": 2.4885409583306542e-06,
      "loss": 0.5061,
      "num_input_tokens_seen": 4337998080,
      "step": 58500,
      "train_runtime": 69654.2623,
      "train_tokens_per_second": 62279.004
    },
    {
      "epoch": 0.7775363618846386,
      "grad_norm": 478.0,
      "learning_rate": 2.485700451411379e-06,
      "loss": 0.8002,
      "num_input_tokens_seen": 4340024320,
      "step": 58510,
      "train_runtime": 69686.5557,
      "train_tokens_per_second": 62279.22
    },
    {
      "epoch": 0.7776692513671005,
      "grad_norm": 86.5,
      "learning_rate": 2.4828613364490917e-06,
      "loss": 0.9117,
      "num_input_tokens_seen": 4340846336,
      "step": 58520,
      "train_runtime": 69699.304,
      "train_tokens_per_second": 62279.622
    },
    {
      "epoch": 0.7778021408495625,
      "grad_norm": 22.125,
      "learning_rate": 2.480023613969712e-06,
      "loss": 0.7006,
      "num_input_tokens_seen": 4341287680,
      "step": 58530,
      "train_runtime": 69706.7125,
      "train_tokens_per_second": 62279.335
    },
    {
      "epoch": 0.7779350303320244,
      "grad_norm": 101.5,
      "learning_rate": 2.4771872844989043e-06,
      "loss": 0.7336,
      "num_input_tokens_seen": 4341513472,
      "step": 58540,
      "train_runtime": 69712.109,
      "train_tokens_per_second": 62277.753
    },
    {
      "epoch": 0.7780679198144863,
      "grad_norm": 13.125,
      "learning_rate": 2.4743523485620723e-06,
      "loss": 0.5262,
      "num_input_tokens_seen": 4341625088,
      "step": 58550,
      "train_runtime": 69718.5968,
      "train_tokens_per_second": 62273.558
    },
    {
      "epoch": 0.7782008092969482,
      "grad_norm": 43.5,
      "learning_rate": 2.471518806684362e-06,
      "loss": 0.7471,
      "num_input_tokens_seen": 4343718912,
      "step": 58560,
      "train_runtime": 69753.2092,
      "train_tokens_per_second": 62272.675
    },
    {
      "epoch": 0.77833369877941,
      "grad_norm": 16.375,
      "learning_rate": 2.468686659390663e-06,
      "loss": 0.9465,
      "num_input_tokens_seen": 4344558336,
      "step": 58570,
      "train_runtime": 69765.8753,
      "train_tokens_per_second": 62273.401
    },
    {
      "epoch": 0.778466588261872,
      "grad_norm": 18.125,
      "learning_rate": 2.4658559072056077e-06,
      "loss": 0.7618,
      "num_input_tokens_seen": 4345040640,
      "step": 58580,
      "train_runtime": 69772.6914,
      "train_tokens_per_second": 62274.23
    },
    {
      "epoch": 0.7785994777443339,
      "grad_norm": 11.5625,
      "learning_rate": 2.463026550653562e-06,
      "loss": 0.6281,
      "num_input_tokens_seen": 4345274880,
      "step": 58590,
      "train_runtime": 69776.7977,
      "train_tokens_per_second": 62273.922
    },
    {
      "epoch": 0.7787323672267958,
      "grad_norm": 10.6875,
      "learning_rate": 2.4601985902586423e-06,
      "loss": 0.6215,
      "num_input_tokens_seen": 4345395456,
      "step": 58600,
      "train_runtime": 69781.5632,
      "train_tokens_per_second": 62271.397
    },
    {
      "epoch": 0.7788652567092578,
      "grad_norm": 31.125,
      "learning_rate": 2.4573720265447022e-06,
      "loss": 0.8139,
      "num_input_tokens_seen": 4347297280,
      "step": 58610,
      "train_runtime": 69813.2339,
      "train_tokens_per_second": 62270.39
    },
    {
      "epoch": 0.7789981461917197,
      "grad_norm": 22.125,
      "learning_rate": 2.454546860035337e-06,
      "loss": 0.9443,
      "num_input_tokens_seen": 4348154112,
      "step": 58620,
      "train_runtime": 69825.9189,
      "train_tokens_per_second": 62271.348
    },
    {
      "epoch": 0.7791310356741815,
      "grad_norm": 20.0,
      "learning_rate": 2.451723091253888e-06,
      "loss": 0.763,
      "num_input_tokens_seen": 4348634624,
      "step": 58630,
      "train_runtime": 69833.4352,
      "train_tokens_per_second": 62271.527
    },
    {
      "epoch": 0.7792639251566434,
      "grad_norm": 34.0,
      "learning_rate": 2.4489007207234238e-06,
      "loss": 0.699,
      "num_input_tokens_seen": 4348872192,
      "step": 58640,
      "train_runtime": 69837.4805,
      "train_tokens_per_second": 62271.321
    },
    {
      "epoch": 0.7793968146391054,
      "grad_norm": 15.25,
      "learning_rate": 2.4460797489667754e-06,
      "loss": 0.5598,
      "num_input_tokens_seen": 4348986112,
      "step": 58650,
      "train_runtime": 69841.4623,
      "train_tokens_per_second": 62269.402
    },
    {
      "epoch": 0.7795297041215673,
      "grad_norm": 23.75,
      "learning_rate": 2.4432601765064934e-06,
      "loss": 0.7562,
      "num_input_tokens_seen": 4351034368,
      "step": 58660,
      "train_runtime": 69875.9604,
      "train_tokens_per_second": 62267.972
    },
    {
      "epoch": 0.7796625936040292,
      "grad_norm": 11.75,
      "learning_rate": 2.440442003864881e-06,
      "loss": 0.8538,
      "num_input_tokens_seen": 4351869440,
      "step": 58670,
      "train_runtime": 69888.6682,
      "train_tokens_per_second": 62268.599
    },
    {
      "epoch": 0.7797954830864912,
      "grad_norm": 75.5,
      "learning_rate": 2.4376252315639837e-06,
      "loss": 0.7802,
      "num_input_tokens_seen": 4352338176,
      "step": 58680,
      "train_runtime": 69895.4301,
      "train_tokens_per_second": 62269.281
    },
    {
      "epoch": 0.779928372568953,
      "grad_norm": 19.375,
      "learning_rate": 2.4348098601255732e-06,
      "loss": 0.6498,
      "num_input_tokens_seen": 4352556032,
      "step": 58690,
      "train_runtime": 69899.4644,
      "train_tokens_per_second": 62268.804
    },
    {
      "epoch": 0.7800612620514149,
      "grad_norm": 71.0,
      "learning_rate": 2.431995890071185e-06,
      "loss": 0.6156,
      "num_input_tokens_seen": 4352669696,
      "step": 58700,
      "train_runtime": 69903.5459,
      "train_tokens_per_second": 62266.794
    },
    {
      "epoch": 0.7801941515338768,
      "grad_norm": 14.5625,
      "learning_rate": 2.4291833219220694e-06,
      "loss": 0.875,
      "num_input_tokens_seen": 4354771200,
      "step": 58710,
      "train_runtime": 69937.5387,
      "train_tokens_per_second": 62266.578
    },
    {
      "epoch": 0.7803270410163388,
      "grad_norm": 10.8125,
      "learning_rate": 2.4263721561992416e-06,
      "loss": 0.9063,
      "num_input_tokens_seen": 4355646976,
      "step": 58720,
      "train_runtime": 69950.6762,
      "train_tokens_per_second": 62267.403
    },
    {
      "epoch": 0.7804599304988007,
      "grad_norm": 21.75,
      "learning_rate": 2.4235623934234354e-06,
      "loss": 0.798,
      "num_input_tokens_seen": 4356150784,
      "step": 58730,
      "train_runtime": 69957.9273,
      "train_tokens_per_second": 62268.151
    },
    {
      "epoch": 0.7805928199812626,
      "grad_norm": 11.0,
      "learning_rate": 2.4207540341151394e-06,
      "loss": 0.6911,
      "num_input_tokens_seen": 4356390400,
      "step": 58740,
      "train_runtime": 69962.3762,
      "train_tokens_per_second": 62267.616
    },
    {
      "epoch": 0.7807257094637245,
      "grad_norm": 11.875,
      "learning_rate": 2.417947078794578e-06,
      "loss": 0.5064,
      "num_input_tokens_seen": 4356508928,
      "step": 58750,
      "train_runtime": 69966.1386,
      "train_tokens_per_second": 62265.962
    },
    {
      "epoch": 0.7808585989461864,
      "grad_norm": 22.75,
      "learning_rate": 2.415141527981706e-06,
      "loss": 0.792,
      "num_input_tokens_seen": 4358589440,
      "step": 58760,
      "train_runtime": 70001.6113,
      "train_tokens_per_second": 62264.13
    },
    {
      "epoch": 0.7809914884286483,
      "grad_norm": 83.0,
      "learning_rate": 2.4123373821962394e-06,
      "loss": 0.8757,
      "num_input_tokens_seen": 4359464448,
      "step": 58770,
      "train_runtime": 70015.0198,
      "train_tokens_per_second": 62264.703
    },
    {
      "epoch": 0.7811243779111102,
      "grad_norm": 25.625,
      "learning_rate": 2.4095346419576116e-06,
      "loss": 0.7353,
      "num_input_tokens_seen": 4359952896,
      "step": 58780,
      "train_runtime": 70022.7556,
      "train_tokens_per_second": 62264.8
    },
    {
      "epoch": 0.7812572673935722,
      "grad_norm": 15.0,
      "learning_rate": 2.4067333077850085e-06,
      "loss": 0.6879,
      "num_input_tokens_seen": 4360187648,
      "step": 58790,
      "train_runtime": 70027.6324,
      "train_tokens_per_second": 62263.816
    },
    {
      "epoch": 0.7813901568760341,
      "grad_norm": 13.0,
      "learning_rate": 2.403933380197352e-06,
      "loss": 0.526,
      "num_input_tokens_seen": 4360306688,
      "step": 58800,
      "train_runtime": 70033.6289,
      "train_tokens_per_second": 62260.185
    },
    {
      "epoch": 0.781523046358496,
      "grad_norm": 32.5,
      "learning_rate": 2.4011348597133043e-06,
      "loss": 0.9363,
      "num_input_tokens_seen": 4362257408,
      "step": 58810,
      "train_runtime": 70065.438,
      "train_tokens_per_second": 62259.761
    },
    {
      "epoch": 0.7816559358409578,
      "grad_norm": 20.75,
      "learning_rate": 2.398337746851266e-06,
      "loss": 0.9114,
      "num_input_tokens_seen": 4363111424,
      "step": 58820,
      "train_runtime": 70078.639,
      "train_tokens_per_second": 62260.219
    },
    {
      "epoch": 0.7817888253234198,
      "grad_norm": 23.75,
      "learning_rate": 2.3955420421293796e-06,
      "loss": 0.8336,
      "num_input_tokens_seen": 4363580672,
      "step": 58830,
      "train_runtime": 70086.3173,
      "train_tokens_per_second": 62260.094
    },
    {
      "epoch": 0.7819217148058817,
      "grad_norm": 24.625,
      "learning_rate": 2.3927477460655203e-06,
      "loss": 0.7263,
      "num_input_tokens_seen": 4363808512,
      "step": 58840,
      "train_runtime": 70090.5244,
      "train_tokens_per_second": 62259.607
    },
    {
      "epoch": 0.7820546042883436,
      "grad_norm": 12.3125,
      "learning_rate": 2.389954859177308e-06,
      "loss": 0.4503,
      "num_input_tokens_seen": 4363925248,
      "step": 58850,
      "train_runtime": 70094.0991,
      "train_tokens_per_second": 62258.097
    },
    {
      "epoch": 0.7821874937708055,
      "grad_norm": 29.75,
      "learning_rate": 2.3871633819821006e-06,
      "loss": 0.845,
      "num_input_tokens_seen": 4365987840,
      "step": 58860,
      "train_runtime": 70125.2268,
      "train_tokens_per_second": 62259.875
    },
    {
      "epoch": 0.7823203832532675,
      "grad_norm": 41.75,
      "learning_rate": 2.3843733149969937e-06,
      "loss": 0.904,
      "num_input_tokens_seen": 4366829824,
      "step": 58870,
      "train_runtime": 70137.982,
      "train_tokens_per_second": 62260.557
    },
    {
      "epoch": 0.7824532727357293,
      "grad_norm": 24.125,
      "learning_rate": 2.3815846587388235e-06,
      "loss": 0.6799,
      "num_input_tokens_seen": 4367288064,
      "step": 58880,
      "train_runtime": 70146.7298,
      "train_tokens_per_second": 62259.325
    },
    {
      "epoch": 0.7825861622181912,
      "grad_norm": 17.375,
      "learning_rate": 2.378797413724163e-06,
      "loss": 0.6224,
      "num_input_tokens_seen": 4367511296,
      "step": 58890,
      "train_runtime": 70151.3953,
      "train_tokens_per_second": 62258.367
    },
    {
      "epoch": 0.7827190517006531,
      "grad_norm": 9.75,
      "learning_rate": 2.3760115804693273e-06,
      "loss": 0.5889,
      "num_input_tokens_seen": 4367625984,
      "step": 58900,
      "train_runtime": 70155.7529,
      "train_tokens_per_second": 62256.134
    },
    {
      "epoch": 0.7828519411831151,
      "grad_norm": 16.625,
      "learning_rate": 2.373227159490362e-06,
      "loss": 0.7294,
      "num_input_tokens_seen": 4369664000,
      "step": 58910,
      "train_runtime": 70189.4198,
      "train_tokens_per_second": 62255.309
    },
    {
      "epoch": 0.782984830665577,
      "grad_norm": 18.75,
      "learning_rate": 2.3704441513030586e-06,
      "loss": 0.8498,
      "num_input_tokens_seen": 4370509312,
      "step": 58920,
      "train_runtime": 70202.3069,
      "train_tokens_per_second": 62255.922
    },
    {
      "epoch": 0.7831177201480389,
      "grad_norm": 20.75,
      "learning_rate": 2.367662556422944e-06,
      "loss": 0.7348,
      "num_input_tokens_seen": 4370969600,
      "step": 58930,
      "train_runtime": 70209.4277,
      "train_tokens_per_second": 62256.163
    },
    {
      "epoch": 0.7832506096305007,
      "grad_norm": 11.0625,
      "learning_rate": 2.3648823753652848e-06,
      "loss": 0.6217,
      "num_input_tokens_seen": 4371202560,
      "step": 58940,
      "train_runtime": 70214.9123,
      "train_tokens_per_second": 62254.618
    },
    {
      "epoch": 0.7833834991129627,
      "grad_norm": 11.4375,
      "learning_rate": 2.3621036086450878e-06,
      "loss": 0.5953,
      "num_input_tokens_seen": 4371317760,
      "step": 58950,
      "train_runtime": 70218.1557,
      "train_tokens_per_second": 62253.383
    },
    {
      "epoch": 0.7835163885954246,
      "grad_norm": 22.875,
      "learning_rate": 2.3593262567770858e-06,
      "loss": 0.7822,
      "num_input_tokens_seen": 4373316608,
      "step": 58960,
      "train_runtime": 70251.4037,
      "train_tokens_per_second": 62252.373
    },
    {
      "epoch": 0.7836492780778865,
      "grad_norm": 28.0,
      "learning_rate": 2.3565503202757687e-06,
      "loss": 0.9205,
      "num_input_tokens_seen": 4374165248,
      "step": 58970,
      "train_runtime": 70264.4214,
      "train_tokens_per_second": 62252.918
    },
    {
      "epoch": 0.7837821675603485,
      "grad_norm": 29.875,
      "learning_rate": 2.353775799655348e-06,
      "loss": 0.8085,
      "num_input_tokens_seen": 4374654208,
      "step": 58980,
      "train_runtime": 70273.5042,
      "train_tokens_per_second": 62251.83
    },
    {
      "epoch": 0.7839150570428104,
      "grad_norm": 20.375,
      "learning_rate": 2.351002695429779e-06,
      "loss": 0.7368,
      "num_input_tokens_seen": 4374884352,
      "step": 58990,
      "train_runtime": 70278.2249,
      "train_tokens_per_second": 62250.923
    },
    {
      "epoch": 0.7840479465252722,
      "grad_norm": 37.75,
      "learning_rate": 2.34823100811276e-06,
      "loss": 0.5415,
      "num_input_tokens_seen": 4375001088,
      "step": 59000,
      "train_runtime": 70283.2863,
      "train_tokens_per_second": 62248.101
    },
    {
      "epoch": 0.7841808360077341,
      "grad_norm": 12.0625,
      "learning_rate": 2.3454607382177107e-06,
      "loss": 0.7779,
      "num_input_tokens_seen": 4377139968,
      "step": 59010,
      "train_runtime": 70317.3338,
      "train_tokens_per_second": 62248.378
    },
    {
      "epoch": 0.7843137254901961,
      "grad_norm": 14.5,
      "learning_rate": 2.3426918862578106e-06,
      "loss": 0.8781,
      "num_input_tokens_seen": 4377974016,
      "step": 59020,
      "train_runtime": 70330.1644,
      "train_tokens_per_second": 62248.881
    },
    {
      "epoch": 0.784446614972658,
      "grad_norm": 24.875,
      "learning_rate": 2.3399244527459574e-06,
      "loss": 0.8019,
      "num_input_tokens_seen": 4378448384,
      "step": 59030,
      "train_runtime": 70337.389,
      "train_tokens_per_second": 62249.231
    },
    {
      "epoch": 0.7845795044551199,
      "grad_norm": 9.3125,
      "learning_rate": 2.3371584381947965e-06,
      "loss": 0.6319,
      "num_input_tokens_seen": 4378684160,
      "step": 59040,
      "train_runtime": 70341.5147,
      "train_tokens_per_second": 62248.932
    },
    {
      "epoch": 0.7847123939375819,
      "grad_norm": 29.375,
      "learning_rate": 2.3343938431167057e-06,
      "loss": 0.6358,
      "num_input_tokens_seen": 4378799360,
      "step": 59050,
      "train_runtime": 70345.5277,
      "train_tokens_per_second": 62247.018
    },
    {
      "epoch": 0.7848452834200437,
      "grad_norm": 42.75,
      "learning_rate": 2.331630668023802e-06,
      "loss": 0.8165,
      "num_input_tokens_seen": 4380818944,
      "step": 59060,
      "train_runtime": 70379.6263,
      "train_tokens_per_second": 62245.556
    },
    {
      "epoch": 0.7849781729025056,
      "grad_norm": 17.625,
      "learning_rate": 2.328868913427943e-06,
      "loss": 0.9094,
      "num_input_tokens_seen": 4381658112,
      "step": 59070,
      "train_runtime": 70392.5315,
      "train_tokens_per_second": 62246.065
    },
    {
      "epoch": 0.7851110623849675,
      "grad_norm": 11.1875,
      "learning_rate": 2.3261085798407102e-06,
      "loss": 0.7358,
      "num_input_tokens_seen": 4382156800,
      "step": 59080,
      "train_runtime": 70399.819,
      "train_tokens_per_second": 62246.705
    },
    {
      "epoch": 0.7852439518674295,
      "grad_norm": 21.875,
      "learning_rate": 2.323349667773441e-06,
      "loss": 0.6169,
      "num_input_tokens_seen": 4382383104,
      "step": 59090,
      "train_runtime": 70403.7585,
      "train_tokens_per_second": 62246.437
    },
    {
      "epoch": 0.7853768413498914,
      "grad_norm": 15.0,
      "learning_rate": 2.320592177737192e-06,
      "loss": 0.5063,
      "num_input_tokens_seen": 4382498304,
      "step": 59100,
      "train_runtime": 70409.7937,
      "train_tokens_per_second": 62242.737
    },
    {
      "epoch": 0.7855097308323533,
      "grad_norm": 20.125,
      "learning_rate": 2.317836110242766e-06,
      "loss": 0.8078,
      "num_input_tokens_seen": 4384523264,
      "step": 59110,
      "train_runtime": 70442.7407,
      "train_tokens_per_second": 62242.372
    },
    {
      "epoch": 0.7856426203148151,
      "grad_norm": 24.875,
      "learning_rate": 2.3150814658007003e-06,
      "loss": 0.9525,
      "num_input_tokens_seen": 4385345024,
      "step": 59120,
      "train_runtime": 70455.6264,
      "train_tokens_per_second": 62242.652
    },
    {
      "epoch": 0.7857755097972771,
      "grad_norm": 13.0,
      "learning_rate": 2.312328244921268e-06,
      "loss": 0.6722,
      "num_input_tokens_seen": 4385783040,
      "step": 59130,
      "train_runtime": 70462.8372,
      "train_tokens_per_second": 62242.499
    },
    {
      "epoch": 0.785908399279739,
      "grad_norm": 14.3125,
      "learning_rate": 2.3095764481144778e-06,
      "loss": 0.7009,
      "num_input_tokens_seen": 4386008832,
      "step": 59140,
      "train_runtime": 70467.3302,
      "train_tokens_per_second": 62241.734
    },
    {
      "epoch": 0.7860412887622009,
      "grad_norm": 19.125,
      "learning_rate": 2.3068260758900795e-06,
      "loss": 0.4698,
      "num_input_tokens_seen": 4386122496,
      "step": 59150,
      "train_runtime": 70471.5456,
      "train_tokens_per_second": 62239.624
    },
    {
      "epoch": 0.7861741782446628,
      "grad_norm": 37.5,
      "learning_rate": 2.304077128757549e-06,
      "loss": 0.863,
      "num_input_tokens_seen": 4388180224,
      "step": 59160,
      "train_runtime": 70504.1998,
      "train_tokens_per_second": 62239.983
    },
    {
      "epoch": 0.7863070677271248,
      "grad_norm": 20.125,
      "learning_rate": 2.3013296072261072e-06,
      "loss": 0.9019,
      "num_input_tokens_seen": 4389020160,
      "step": 59170,
      "train_runtime": 70517.2113,
      "train_tokens_per_second": 62240.41
    },
    {
      "epoch": 0.7864399572095866,
      "grad_norm": 11.1875,
      "learning_rate": 2.2985835118047083e-06,
      "loss": 0.7629,
      "num_input_tokens_seen": 4389504000,
      "step": 59180,
      "train_runtime": 70525.8179,
      "train_tokens_per_second": 62239.675
    },
    {
      "epoch": 0.7865728466920485,
      "grad_norm": 14.375,
      "learning_rate": 2.295838843002042e-06,
      "loss": 0.6874,
      "num_input_tokens_seen": 4389742080,
      "step": 59190,
      "train_runtime": 70530.3164,
      "train_tokens_per_second": 62239.081
    },
    {
      "epoch": 0.7867057361745105,
      "grad_norm": 18.25,
      "learning_rate": 2.2930956013265347e-06,
      "loss": 0.5563,
      "num_input_tokens_seen": 4389855488,
      "step": 59200,
      "train_runtime": 70534.8563,
      "train_tokens_per_second": 62236.683
    },
    {
      "epoch": 0.7868386256569724,
      "grad_norm": 15.6875,
      "learning_rate": 2.2903537872863415e-06,
      "loss": 0.9032,
      "num_input_tokens_seen": 4391928832,
      "step": 59210,
      "train_runtime": 70568.83,
      "train_tokens_per_second": 62236.101
    },
    {
      "epoch": 0.7869715151394343,
      "grad_norm": 24.875,
      "learning_rate": 2.287613401389368e-06,
      "loss": 0.8852,
      "num_input_tokens_seen": 4392806912,
      "step": 59220,
      "train_runtime": 70582.4717,
      "train_tokens_per_second": 62236.513
    },
    {
      "epoch": 0.7871044046218962,
      "grad_norm": 18.25,
      "learning_rate": 2.28487444414324e-06,
      "loss": 0.7106,
      "num_input_tokens_seen": 4393307136,
      "step": 59230,
      "train_runtime": 70589.523,
      "train_tokens_per_second": 62237.382
    },
    {
      "epoch": 0.7872372941043582,
      "grad_norm": 17.0,
      "learning_rate": 2.2821369160553263e-06,
      "loss": 0.6538,
      "num_input_tokens_seen": 4393545216,
      "step": 59240,
      "train_runtime": 70593.796,
      "train_tokens_per_second": 62236.988
    },
    {
      "epoch": 0.78737018358682,
      "grad_norm": 9.5625,
      "learning_rate": 2.279400817632731e-06,
      "loss": 0.5376,
      "num_input_tokens_seen": 4393660672,
      "step": 59250,
      "train_runtime": 70598.7323,
      "train_tokens_per_second": 62234.271
    },
    {
      "epoch": 0.7875030730692819,
      "grad_norm": 13.8125,
      "learning_rate": 2.2766661493822893e-06,
      "loss": 0.8351,
      "num_input_tokens_seen": 4395655680,
      "step": 59260,
      "train_runtime": 70630.5756,
      "train_tokens_per_second": 62234.459
    },
    {
      "epoch": 0.7876359625517438,
      "grad_norm": 17.0,
      "learning_rate": 2.27393291181058e-06,
      "loss": 0.9632,
      "num_input_tokens_seen": 4396505600,
      "step": 59270,
      "train_runtime": 70643.7267,
      "train_tokens_per_second": 62234.905
    },
    {
      "epoch": 0.7877688520342058,
      "grad_norm": 12.5,
      "learning_rate": 2.2712011054239002e-06,
      "loss": 0.7573,
      "num_input_tokens_seen": 4396977920,
      "step": 59280,
      "train_runtime": 70650.9634,
      "train_tokens_per_second": 62235.215
    },
    {
      "epoch": 0.7879017415166677,
      "grad_norm": 11.5,
      "learning_rate": 2.268470730728305e-06,
      "loss": 0.5558,
      "num_input_tokens_seen": 4397213184,
      "step": 59290,
      "train_runtime": 70657.1193,
      "train_tokens_per_second": 62233.123
    },
    {
      "epoch": 0.7880346309991296,
      "grad_norm": 11.8125,
      "learning_rate": 2.265741788229563e-06,
      "loss": 0.489,
      "num_input_tokens_seen": 4397328384,
      "step": 59300,
      "train_runtime": 70661.5137,
      "train_tokens_per_second": 62230.883
    },
    {
      "epoch": 0.7881675204815914,
      "grad_norm": 22.75,
      "learning_rate": 2.2630142784331897e-06,
      "loss": 0.8082,
      "num_input_tokens_seen": 4399356416,
      "step": 59310,
      "train_runtime": 70695.0696,
      "train_tokens_per_second": 62230.032
    },
    {
      "epoch": 0.7883004099640534,
      "grad_norm": 29.25,
      "learning_rate": 2.260288201844436e-06,
      "loss": 0.9108,
      "num_input_tokens_seen": 4400216064,
      "step": 59320,
      "train_runtime": 70707.9737,
      "train_tokens_per_second": 62230.832
    },
    {
      "epoch": 0.7884332994465153,
      "grad_norm": 13.8125,
      "learning_rate": 2.257563558968272e-06,
      "loss": 0.723,
      "num_input_tokens_seen": 4400660480,
      "step": 59330,
      "train_runtime": 70715.591,
      "train_tokens_per_second": 62230.414
    },
    {
      "epoch": 0.7885661889289772,
      "grad_norm": 22.625,
      "learning_rate": 2.2548403503094275e-06,
      "loss": 0.5967,
      "num_input_tokens_seen": 4400898816,
      "step": 59340,
      "train_runtime": 70720.2166,
      "train_tokens_per_second": 62229.713
    },
    {
      "epoch": 0.7886990784114392,
      "grad_norm": 7.75,
      "learning_rate": 2.2521185763723418e-06,
      "loss": 0.5198,
      "num_input_tokens_seen": 4401015296,
      "step": 59350,
      "train_runtime": 70724.4619,
      "train_tokens_per_second": 62227.625
    },
    {
      "epoch": 0.7888319678939011,
      "grad_norm": 22.75,
      "learning_rate": 2.2493982376612035e-06,
      "loss": 0.8786,
      "num_input_tokens_seen": 4403108352,
      "step": 59360,
      "train_runtime": 70757.7572,
      "train_tokens_per_second": 62227.924
    },
    {
      "epoch": 0.7889648573763629,
      "grad_norm": 30.625,
      "learning_rate": 2.246679334679932e-06,
      "loss": 1.0317,
      "num_input_tokens_seen": 4403974144,
      "step": 59370,
      "train_runtime": 70771.735,
      "train_tokens_per_second": 62227.867
    },
    {
      "epoch": 0.7890977468588248,
      "grad_norm": 32.75,
      "learning_rate": 2.2439618679321774e-06,
      "loss": 0.7162,
      "num_input_tokens_seen": 4404437248,
      "step": 59380,
      "train_runtime": 70779.2096,
      "train_tokens_per_second": 62227.839
    },
    {
      "epoch": 0.7892306363412868,
      "grad_norm": 23.0,
      "learning_rate": 2.241245837921331e-06,
      "loss": 0.6626,
      "num_input_tokens_seen": 4404670976,
      "step": 59390,
      "train_runtime": 70784.1442,
      "train_tokens_per_second": 62226.803
    },
    {
      "epoch": 0.7893635258237487,
      "grad_norm": 14.1875,
      "learning_rate": 2.2385312451505038e-06,
      "loss": 0.6047,
      "num_input_tokens_seen": 4404785920,
      "step": 59400,
      "train_runtime": 70787.7272,
      "train_tokens_per_second": 62225.277
    },
    {
      "epoch": 0.7894964153062106,
      "grad_norm": 22.0,
      "learning_rate": 2.23581809012256e-06,
      "loss": 0.8893,
      "num_input_tokens_seen": 4406925312,
      "step": 59410,
      "train_runtime": 70821.6957,
      "train_tokens_per_second": 62225.64
    },
    {
      "epoch": 0.7896293047886725,
      "grad_norm": 19.5,
      "learning_rate": 2.2331063733400816e-06,
      "loss": 0.9012,
      "num_input_tokens_seen": 4407788544,
      "step": 59420,
      "train_runtime": 70834.7573,
      "train_tokens_per_second": 62226.352
    },
    {
      "epoch": 0.7897621942711344,
      "grad_norm": 20.75,
      "learning_rate": 2.2303960953053916e-06,
      "loss": 0.7427,
      "num_input_tokens_seen": 4408252160,
      "step": 59430,
      "train_runtime": 70843.4566,
      "train_tokens_per_second": 62225.255
    },
    {
      "epoch": 0.7898950837535963,
      "grad_norm": 20.0,
      "learning_rate": 2.2276872565205443e-06,
      "loss": 0.6896,
      "num_input_tokens_seen": 4408479744,
      "step": 59440,
      "train_runtime": 70847.3678,
      "train_tokens_per_second": 62225.032
    },
    {
      "epoch": 0.7900279732360582,
      "grad_norm": 110.0,
      "learning_rate": 2.224979857487328e-06,
      "loss": 0.5519,
      "num_input_tokens_seen": 4408593408,
      "step": 59450,
      "train_runtime": 70851.236,
      "train_tokens_per_second": 62223.239
    },
    {
      "epoch": 0.7901608627185202,
      "grad_norm": 20.25,
      "learning_rate": 2.2222738987072643e-06,
      "loss": 0.836,
      "num_input_tokens_seen": 4410651392,
      "step": 59460,
      "train_runtime": 70883.6137,
      "train_tokens_per_second": 62223.851
    },
    {
      "epoch": 0.7902937522009821,
      "grad_norm": 20.625,
      "learning_rate": 2.2195693806816123e-06,
      "loss": 0.9373,
      "num_input_tokens_seen": 4411503872,
      "step": 59470,
      "train_runtime": 70896.7567,
      "train_tokens_per_second": 62224.34
    },
    {
      "epoch": 0.790426641683444,
      "grad_norm": 17.875,
      "learning_rate": 2.2168663039113513e-06,
      "loss": 0.7601,
      "num_input_tokens_seen": 4412001536,
      "step": 59480,
      "train_runtime": 70905.3093,
      "train_tokens_per_second": 62223.853
    },
    {
      "epoch": 0.7905595311659058,
      "grad_norm": 12.0,
      "learning_rate": 2.2141646688972074e-06,
      "loss": 0.6688,
      "num_input_tokens_seen": 4412239616,
      "step": 59490,
      "train_runtime": 70909.3655,
      "train_tokens_per_second": 62223.651
    },
    {
      "epoch": 0.7906924206483678,
      "grad_norm": 8.125,
      "learning_rate": 2.211464476139634e-06,
      "loss": 0.4477,
      "num_input_tokens_seen": 4412354816,
      "step": 59500,
      "train_runtime": 70912.7735,
      "train_tokens_per_second": 62222.285
    },
    {
      "epoch": 0.7908253101308297,
      "grad_norm": 32.5,
      "learning_rate": 2.208765726138817e-06,
      "loss": 0.7849,
      "num_input_tokens_seen": 4414440704,
      "step": 59510,
      "train_runtime": 70947.7098,
      "train_tokens_per_second": 62221.046
    },
    {
      "epoch": 0.7909581996132916,
      "grad_norm": 23.5,
      "learning_rate": 2.206068419394679e-06,
      "loss": 0.8753,
      "num_input_tokens_seen": 4415303168,
      "step": 59520,
      "train_runtime": 70960.5262,
      "train_tokens_per_second": 62221.962
    },
    {
      "epoch": 0.7910910890957535,
      "grad_norm": 18.5,
      "learning_rate": 2.2033725564068633e-06,
      "loss": 0.7893,
      "num_input_tokens_seen": 4415756544,
      "step": 59530,
      "train_runtime": 70968.6141,
      "train_tokens_per_second": 62221.259
    },
    {
      "epoch": 0.7912239785782155,
      "grad_norm": 19.25,
      "learning_rate": 2.2006781376747657e-06,
      "loss": 0.691,
      "num_input_tokens_seen": 4415991808,
      "step": 59540,
      "train_runtime": 70972.505,
      "train_tokens_per_second": 62221.163
    },
    {
      "epoch": 0.7913568680606773,
      "grad_norm": 12.25,
      "learning_rate": 2.197985163697495e-06,
      "loss": 0.5482,
      "num_input_tokens_seen": 4416107520,
      "step": 59550,
      "train_runtime": 70976.2372,
      "train_tokens_per_second": 62219.522
    },
    {
      "epoch": 0.7914897575431392,
      "grad_norm": 15.9375,
      "learning_rate": 2.1952936349739028e-06,
      "loss": 0.7602,
      "num_input_tokens_seen": 4418172416,
      "step": 59560,
      "train_runtime": 71009.9812,
      "train_tokens_per_second": 62219.034
    },
    {
      "epoch": 0.7916226470256011,
      "grad_norm": 19.25,
      "learning_rate": 2.1926035520025756e-06,
      "loss": 0.9467,
      "num_input_tokens_seen": 4419063296,
      "step": 59570,
      "train_runtime": 71023.2948,
      "train_tokens_per_second": 62219.914
    },
    {
      "epoch": 0.7917555365080631,
      "grad_norm": 16.75,
      "learning_rate": 2.1899149152818167e-06,
      "loss": 0.6776,
      "num_input_tokens_seen": 4419546624,
      "step": 59580,
      "train_runtime": 71030.4419,
      "train_tokens_per_second": 62220.458
    },
    {
      "epoch": 0.791888425990525,
      "grad_norm": 11.6875,
      "learning_rate": 2.1872277253096842e-06,
      "loss": 0.6021,
      "num_input_tokens_seen": 4419775744,
      "step": 59590,
      "train_runtime": 71034.7872,
      "train_tokens_per_second": 62219.877
    },
    {
      "epoch": 0.7920213154729869,
      "grad_norm": 11.75,
      "learning_rate": 2.1845419825839454e-06,
      "loss": 0.5699,
      "num_input_tokens_seen": 4419889152,
      "step": 59600,
      "train_runtime": 71039.3119,
      "train_tokens_per_second": 62217.511
    },
    {
      "epoch": 0.7921542049554487,
      "grad_norm": 66.0,
      "learning_rate": 2.1818576876021203e-06,
      "loss": 0.8896,
      "num_input_tokens_seen": 4421944064,
      "step": 59610,
      "train_runtime": 71073.3978,
      "train_tokens_per_second": 62216.585
    },
    {
      "epoch": 0.7922870944379107,
      "grad_norm": 16.75,
      "learning_rate": 2.1791748408614423e-06,
      "loss": 0.8844,
      "num_input_tokens_seen": 4422810624,
      "step": 59620,
      "train_runtime": 71086.8569,
      "train_tokens_per_second": 62216.995
    },
    {
      "epoch": 0.7924199839203726,
      "grad_norm": 12.5,
      "learning_rate": 2.176493442858888e-06,
      "loss": 0.6809,
      "num_input_tokens_seen": 4423292416,
      "step": 59630,
      "train_runtime": 71094.5923,
      "train_tokens_per_second": 62217.002
    },
    {
      "epoch": 0.7925528734028345,
      "grad_norm": 10.8125,
      "learning_rate": 2.173813494091166e-06,
      "loss": 0.644,
      "num_input_tokens_seen": 4423527680,
      "step": 59640,
      "train_runtime": 71099.822,
      "train_tokens_per_second": 62215.735
    },
    {
      "epoch": 0.7926857628852965,
      "grad_norm": 6.46875,
      "learning_rate": 2.1711349950547033e-06,
      "loss": 0.5928,
      "num_input_tokens_seen": 4423647232,
      "step": 59650,
      "train_runtime": 71103.4354,
      "train_tokens_per_second": 62214.255
    },
    {
      "epoch": 0.7928186523677584,
      "grad_norm": 19.375,
      "learning_rate": 2.16845794624568e-06,
      "loss": 0.8318,
      "num_input_tokens_seen": 4425643776,
      "step": 59660,
      "train_runtime": 71135.6888,
      "train_tokens_per_second": 62214.113
    },
    {
      "epoch": 0.7929515418502202,
      "grad_norm": 14.9375,
      "learning_rate": 2.165782348159987e-06,
      "loss": 0.8232,
      "num_input_tokens_seen": 4426483712,
      "step": 59670,
      "train_runtime": 71150.8691,
      "train_tokens_per_second": 62212.644
    },
    {
      "epoch": 0.7930844313326821,
      "grad_norm": 22.125,
      "learning_rate": 2.163108201293257e-06,
      "loss": 0.746,
      "num_input_tokens_seen": 4426956032,
      "step": 59680,
      "train_runtime": 71158.3364,
      "train_tokens_per_second": 62212.753
    },
    {
      "epoch": 0.7932173208151441,
      "grad_norm": 24.0,
      "learning_rate": 2.160435506140852e-06,
      "loss": 0.6338,
      "num_input_tokens_seen": 4427195904,
      "step": 59690,
      "train_runtime": 71163.3406,
      "train_tokens_per_second": 62211.749
    },
    {
      "epoch": 0.793350210297606,
      "grad_norm": 15.6875,
      "learning_rate": 2.157764263197867e-06,
      "loss": 0.479,
      "num_input_tokens_seen": 4427314688,
      "step": 59700,
      "train_runtime": 71167.2972,
      "train_tokens_per_second": 62209.96
    },
    {
      "epoch": 0.7934830997800679,
      "grad_norm": 13.25,
      "learning_rate": 2.155094472959125e-06,
      "loss": 0.7719,
      "num_input_tokens_seen": 4429285376,
      "step": 59710,
      "train_runtime": 71199.421,
      "train_tokens_per_second": 62209.57
    },
    {
      "epoch": 0.7936159892625299,
      "grad_norm": 13.125,
      "learning_rate": 2.152426135919179e-06,
      "loss": 0.9399,
      "num_input_tokens_seen": 4430081792,
      "step": 59720,
      "train_runtime": 71212.9842,
      "train_tokens_per_second": 62208.905
    },
    {
      "epoch": 0.7937488787449918,
      "grad_norm": 26.625,
      "learning_rate": 2.1497592525723153e-06,
      "loss": 0.6661,
      "num_input_tokens_seen": 4430536704,
      "step": 59730,
      "train_runtime": 71219.5445,
      "train_tokens_per_second": 62209.563
    },
    {
      "epoch": 0.7938817682274536,
      "grad_norm": 14.8125,
      "learning_rate": 2.1470938234125507e-06,
      "loss": 0.6317,
      "num_input_tokens_seen": 4430757632,
      "step": 59740,
      "train_runtime": 71223.4112,
      "train_tokens_per_second": 62209.287
    },
    {
      "epoch": 0.7940146577099155,
      "grad_norm": 8.125,
      "learning_rate": 2.1444298489336323e-06,
      "loss": 0.5124,
      "num_input_tokens_seen": 4430872064,
      "step": 59750,
      "train_runtime": 71227.1268,
      "train_tokens_per_second": 62207.648
    },
    {
      "epoch": 0.7941475471923775,
      "grad_norm": 15.625,
      "learning_rate": 2.1417673296290375e-06,
      "loss": 0.8253,
      "num_input_tokens_seen": 4432993280,
      "step": 59760,
      "train_runtime": 71260.7646,
      "train_tokens_per_second": 62208.051
    },
    {
      "epoch": 0.7942804366748394,
      "grad_norm": 13.5,
      "learning_rate": 2.139106265991975e-06,
      "loss": 0.8572,
      "num_input_tokens_seen": 4433833984,
      "step": 59770,
      "train_runtime": 71273.6743,
      "train_tokens_per_second": 62208.579
    },
    {
      "epoch": 0.7944133261573013,
      "grad_norm": 16.75,
      "learning_rate": 2.136446658515381e-06,
      "loss": 0.7869,
      "num_input_tokens_seen": 4434307584,
      "step": 59780,
      "train_runtime": 71280.2657,
      "train_tokens_per_second": 62209.471
    },
    {
      "epoch": 0.7945462156397632,
      "grad_norm": 28.375,
      "learning_rate": 2.1337885076919295e-06,
      "loss": 0.6569,
      "num_input_tokens_seen": 4434543872,
      "step": 59790,
      "train_runtime": 71285.5031,
      "train_tokens_per_second": 62208.215
    },
    {
      "epoch": 0.7946791051222251,
      "grad_norm": 38.75,
      "learning_rate": 2.1311318140140114e-06,
      "loss": 0.5323,
      "num_input_tokens_seen": 4434656512,
      "step": 59800,
      "train_runtime": 71289.1096,
      "train_tokens_per_second": 62206.648
    },
    {
      "epoch": 0.794811994604687,
      "grad_norm": 44.25,
      "learning_rate": 2.1284765779737606e-06,
      "loss": 0.7545,
      "num_input_tokens_seen": 4436683008,
      "step": 59810,
      "train_runtime": 71322.0288,
      "train_tokens_per_second": 62206.349
    },
    {
      "epoch": 0.7949448840871489,
      "grad_norm": 21.875,
      "learning_rate": 2.1258228000630334e-06,
      "loss": 0.8289,
      "num_input_tokens_seen": 4437549824,
      "step": 59820,
      "train_runtime": 71335.1655,
      "train_tokens_per_second": 62207.045
    },
    {
      "epoch": 0.7950777735696108,
      "grad_norm": 25.25,
      "learning_rate": 2.123170480773421e-06,
      "loss": 0.7512,
      "num_input_tokens_seen": 4438041856,
      "step": 59830,
      "train_runtime": 71344.3966,
      "train_tokens_per_second": 62205.892
    },
    {
      "epoch": 0.7952106630520728,
      "grad_norm": 10.875,
      "learning_rate": 2.120519620596242e-06,
      "loss": 0.649,
      "num_input_tokens_seen": 4438281984,
      "step": 59840,
      "train_runtime": 71349.1595,
      "train_tokens_per_second": 62205.105
    },
    {
      "epoch": 0.7953435525345347,
      "grad_norm": 36.5,
      "learning_rate": 2.1178702200225388e-06,
      "loss": 0.4877,
      "num_input_tokens_seen": 4438398720,
      "step": 59850,
      "train_runtime": 71353.4126,
      "train_tokens_per_second": 62203.034
    },
    {
      "epoch": 0.7954764420169965,
      "grad_norm": 13.5625,
      "learning_rate": 2.1152222795430986e-06,
      "loss": 0.8197,
      "num_input_tokens_seen": 4440447232,
      "step": 59860,
      "train_runtime": 71386.1701,
      "train_tokens_per_second": 62203.186
    },
    {
      "epoch": 0.7956093314994584,
      "grad_norm": 14.3125,
      "learning_rate": 2.112575799648422e-06,
      "loss": 0.8751,
      "num_input_tokens_seen": 4441278720,
      "step": 59870,
      "train_runtime": 71398.8057,
      "train_tokens_per_second": 62203.824
    },
    {
      "epoch": 0.7957422209819204,
      "grad_norm": 14.125,
      "learning_rate": 2.1099307808287474e-06,
      "loss": 0.747,
      "num_input_tokens_seen": 4441754880,
      "step": 59880,
      "train_runtime": 71407.8954,
      "train_tokens_per_second": 62202.574
    },
    {
      "epoch": 0.7958751104643823,
      "grad_norm": 11.0,
      "learning_rate": 2.1072872235740428e-06,
      "loss": 0.6195,
      "num_input_tokens_seen": 4441983744,
      "step": 59890,
      "train_runtime": 71412.6881,
      "train_tokens_per_second": 62201.604
    },
    {
      "epoch": 0.7960079999468442,
      "grad_norm": 26.375,
      "learning_rate": 2.1046451283739976e-06,
      "loss": 0.5209,
      "num_input_tokens_seen": 4442101760,
      "step": 59900,
      "train_runtime": 71415.8517,
      "train_tokens_per_second": 62200.501
    },
    {
      "epoch": 0.7961408894293062,
      "grad_norm": 64.0,
      "learning_rate": 2.102004495718046e-06,
      "loss": 0.7902,
      "num_input_tokens_seen": 4444214784,
      "step": 59910,
      "train_runtime": 71450.4267,
      "train_tokens_per_second": 62199.975
    },
    {
      "epoch": 0.796273778911768,
      "grad_norm": 20.625,
      "learning_rate": 2.0993653260953306e-06,
      "loss": 0.8592,
      "num_input_tokens_seen": 4445054720,
      "step": 59920,
      "train_runtime": 71465.237,
      "train_tokens_per_second": 62198.838
    },
    {
      "epoch": 0.7964066683942299,
      "grad_norm": 25.5,
      "learning_rate": 2.0967276199947463e-06,
      "loss": 0.7319,
      "num_input_tokens_seen": 4445531392,
      "step": 59930,
      "train_runtime": 71472.3348,
      "train_tokens_per_second": 62199.331
    },
    {
      "epoch": 0.7965395578766918,
      "grad_norm": 26.375,
      "learning_rate": 2.0940913779048946e-06,
      "loss": 0.6599,
      "num_input_tokens_seen": 4445772288,
      "step": 59940,
      "train_runtime": 71476.7474,
      "train_tokens_per_second": 62198.861
    },
    {
      "epoch": 0.7966724473591538,
      "grad_norm": 27.25,
      "learning_rate": 2.09145660031412e-06,
      "loss": 0.4255,
      "num_input_tokens_seen": 4445887232,
      "step": 59950,
      "train_runtime": 71481.0999,
      "train_tokens_per_second": 62196.682
    },
    {
      "epoch": 0.7968053368416157,
      "grad_norm": 15.6875,
      "learning_rate": 2.088823287710493e-06,
      "loss": 0.7739,
      "num_input_tokens_seen": 4447917312,
      "step": 59960,
      "train_runtime": 71513.6351,
      "train_tokens_per_second": 62196.773
    },
    {
      "epoch": 0.7969382263240776,
      "grad_norm": 24.75,
      "learning_rate": 2.086191440581804e-06,
      "loss": 0.892,
      "num_input_tokens_seen": 4448768768,
      "step": 59970,
      "train_runtime": 71526.388,
      "train_tokens_per_second": 62197.587
    },
    {
      "epoch": 0.7970711158065394,
      "grad_norm": 26.125,
      "learning_rate": 2.083561059415591e-06,
      "loss": 0.7293,
      "num_input_tokens_seen": 4449234432,
      "step": 59980,
      "train_runtime": 71533.9659,
      "train_tokens_per_second": 62197.508
    },
    {
      "epoch": 0.7972040052890014,
      "grad_norm": 16.25,
      "learning_rate": 2.080932144699098e-06,
      "loss": 0.6663,
      "num_input_tokens_seen": 4449458944,
      "step": 59990,
      "train_runtime": 71539.0198,
      "train_tokens_per_second": 62196.253
    },
    {
      "epoch": 0.7973368947714633,
      "grad_norm": 8.0625,
      "learning_rate": 2.0783046969193134e-06,
      "loss": 0.4371,
      "num_input_tokens_seen": 4449574912,
      "step": 60000,
      "train_runtime": 71544.3048,
      "train_tokens_per_second": 62193.279
    },
    {
      "epoch": 0.7973368947714633,
      "eval_accuracy": 0.8194,
      "eval_balanced_accuracy": 0.8181308282438571,
      "eval_invalid_f1": 0.8364722926475915,
      "eval_invalid_precision": 0.8438070880526124,
      "eval_invalid_recall": 0.8292639138240574,
      "eval_loss": 0.36127880215644836,
      "eval_macro_f1": 0.8174098845998118,
      "eval_mcc": 0.6349734127870799,
      "eval_runtime": 128.4603,
      "eval_samples_per_second": 77.845,
      "eval_steps_per_second": 19.461,
      "eval_valid_f1": 0.7983474765520321,
      "eval_valid_precision": 0.7898806893504198,
      "eval_valid_recall": 0.8069977426636569,
      "num_input_tokens_seen": 4449574912,
      "step": 60000
    },
    {
      "epoch": 0.7974697842539252,
      "grad_norm": 26.0,
      "learning_rate": 2.0756787165629467e-06,
      "loss": 0.8475,
      "num_input_tokens_seen": 4451616768,
      "step": 60010,
      "train_runtime": 71714.0627,
      "train_tokens_per_second": 62074.53
    },
    {
      "epoch": 0.7976026737363872,
      "grad_norm": 30.5,
      "learning_rate": 2.0730542041164382e-06,
      "loss": 0.91,
      "num_input_tokens_seen": 4452454400,
      "step": 60020,
      "train_runtime": 71727.0938,
      "train_tokens_per_second": 62074.931
    },
    {
      "epoch": 0.7977355632188491,
      "grad_norm": 28.375,
      "learning_rate": 2.0704311600659577e-06,
      "loss": 0.7577,
      "num_input_tokens_seen": 4452915968,
      "step": 60030,
      "train_runtime": 71734.3396,
      "train_tokens_per_second": 62075.095
    },
    {
      "epoch": 0.7978684527013109,
      "grad_norm": 20.0,
      "learning_rate": 2.0678095848973945e-06,
      "loss": 0.5884,
      "num_input_tokens_seen": 4453145344,
      "step": 60040,
      "train_runtime": 71740.5142,
      "train_tokens_per_second": 62072.95
    },
    {
      "epoch": 0.7980013421837728,
      "grad_norm": 39.75,
      "learning_rate": 2.065189479096377e-06,
      "loss": 0.5493,
      "num_input_tokens_seen": 4453257472,
      "step": 60050,
      "train_runtime": 71744.4748,
      "train_tokens_per_second": 62071.086
    },
    {
      "epoch": 0.7981342316662348,
      "grad_norm": 18.25,
      "learning_rate": 2.062570843148254e-06,
      "loss": 0.7837,
      "num_input_tokens_seen": 4455420928,
      "step": 60060,
      "train_runtime": 71778.5617,
      "train_tokens_per_second": 62071.75
    },
    {
      "epoch": 0.7982671211486967,
      "grad_norm": 21.125,
      "learning_rate": 2.0599536775381067e-06,
      "loss": 0.8605,
      "num_input_tokens_seen": 4456257280,
      "step": 60070,
      "train_runtime": 71791.3708,
      "train_tokens_per_second": 62072.325
    },
    {
      "epoch": 0.7984000106311586,
      "grad_norm": 19.25,
      "learning_rate": 2.0573379827507402e-06,
      "loss": 0.7299,
      "num_input_tokens_seen": 4456730624,
      "step": 60080,
      "train_runtime": 71799.0924,
      "train_tokens_per_second": 62072.242
    },
    {
      "epoch": 0.7985329001136205,
      "grad_norm": 10.4375,
      "learning_rate": 2.0547237592706904e-06,
      "loss": 0.6781,
      "num_input_tokens_seen": 4456965120,
      "step": 60090,
      "train_runtime": 71803.8557,
      "train_tokens_per_second": 62071.39
    },
    {
      "epoch": 0.7986657895960824,
      "grad_norm": 22.0,
      "learning_rate": 2.0521110075822136e-06,
      "loss": 0.489,
      "num_input_tokens_seen": 4457080832,
      "step": 60100,
      "train_runtime": 71808.1331,
      "train_tokens_per_second": 62069.304
    },
    {
      "epoch": 0.7987986790785443,
      "grad_norm": 50.0,
      "learning_rate": 2.0494997281693074e-06,
      "loss": 0.8648,
      "num_input_tokens_seen": 4458978560,
      "step": 60110,
      "train_runtime": 71840.3459,
      "train_tokens_per_second": 62067.888
    },
    {
      "epoch": 0.7989315685610062,
      "grad_norm": 44.5,
      "learning_rate": 2.04688992151568e-06,
      "loss": 0.891,
      "num_input_tokens_seen": 4459756032,
      "step": 60120,
      "train_runtime": 71852.4989,
      "train_tokens_per_second": 62068.211
    },
    {
      "epoch": 0.7990644580434682,
      "grad_norm": 9.75,
      "learning_rate": 2.0442815881047794e-06,
      "loss": 0.6544,
      "num_input_tokens_seen": 4460179200,
      "step": 60130,
      "train_runtime": 71859.5071,
      "train_tokens_per_second": 62068.046
    },
    {
      "epoch": 0.7991973475259301,
      "grad_norm": 14.9375,
      "learning_rate": 2.0416747284197735e-06,
      "loss": 0.6634,
      "num_input_tokens_seen": 4460399360,
      "step": 60140,
      "train_runtime": 71863.8335,
      "train_tokens_per_second": 62067.373
    },
    {
      "epoch": 0.799330237008392,
      "grad_norm": 12.0,
      "learning_rate": 2.0390693429435626e-06,
      "loss": 0.4844,
      "num_input_tokens_seen": 4460514816,
      "step": 60150,
      "train_runtime": 71867.3233,
      "train_tokens_per_second": 62065.966
    },
    {
      "epoch": 0.7994631264908538,
      "grad_norm": 12.25,
      "learning_rate": 2.0364654321587707e-06,
      "loss": 0.7813,
      "num_input_tokens_seen": 4462536448,
      "step": 60160,
      "train_runtime": 71901.6254,
      "train_tokens_per_second": 62064.472
    },
    {
      "epoch": 0.7995960159733158,
      "grad_norm": 18.5,
      "learning_rate": 2.033862996547744e-06,
      "loss": 0.8934,
      "num_input_tokens_seen": 4463371520,
      "step": 60170,
      "train_runtime": 71914.415,
      "train_tokens_per_second": 62065.047
    },
    {
      "epoch": 0.7997289054557777,
      "grad_norm": 15.5625,
      "learning_rate": 2.0312620365925684e-06,
      "loss": 0.7094,
      "num_input_tokens_seen": 4463824896,
      "step": 60180,
      "train_runtime": 71921.6054,
      "train_tokens_per_second": 62065.145
    },
    {
      "epoch": 0.7998617949382396,
      "grad_norm": 9.6875,
      "learning_rate": 2.0286625527750426e-06,
      "loss": 0.6104,
      "num_input_tokens_seen": 4464054016,
      "step": 60190,
      "train_runtime": 71925.8171,
      "train_tokens_per_second": 62064.697
    },
    {
      "epoch": 0.7999946844207015,
      "grad_norm": 12.5,
      "learning_rate": 2.0260645455767003e-06,
      "loss": 0.5919,
      "num_input_tokens_seen": 4464166656,
      "step": 60200,
      "train_runtime": 71929.6606,
      "train_tokens_per_second": 62062.946
    },
    {
      "epoch": 0.8001275739031635,
      "grad_norm": 23.0,
      "learning_rate": 2.023468015478801e-06,
      "loss": 0.8659,
      "num_input_tokens_seen": 4466038528,
      "step": 60210,
      "train_runtime": 71962.2205,
      "train_tokens_per_second": 62060.877
    },
    {
      "epoch": 0.8002604633856254,
      "grad_norm": 18.125,
      "learning_rate": 2.0208729629623204e-06,
      "loss": 0.9202,
      "num_input_tokens_seen": 4466907136,
      "step": 60220,
      "train_runtime": 71975.1979,
      "train_tokens_per_second": 62061.756
    },
    {
      "epoch": 0.8003933528680872,
      "grad_norm": 23.25,
      "learning_rate": 2.018279388507981e-06,
      "loss": 0.6949,
      "num_input_tokens_seen": 4467390464,
      "step": 60230,
      "train_runtime": 71982.6634,
      "train_tokens_per_second": 62062.033
    },
    {
      "epoch": 0.8005262423505491,
      "grad_norm": 10.5,
      "learning_rate": 2.01568729259621e-06,
      "loss": 0.6174,
      "num_input_tokens_seen": 4467624960,
      "step": 60240,
      "train_runtime": 71987.0661,
      "train_tokens_per_second": 62061.495
    },
    {
      "epoch": 0.8006591318330111,
      "grad_norm": 8.4375,
      "learning_rate": 2.0130966757071736e-06,
      "loss": 0.5559,
      "num_input_tokens_seen": 4467737856,
      "step": 60250,
      "train_runtime": 71990.8547,
      "train_tokens_per_second": 62059.797
    },
    {
      "epoch": 0.800792021315473,
      "grad_norm": 14.0,
      "learning_rate": 2.0105075383207596e-06,
      "loss": 0.7561,
      "num_input_tokens_seen": 4469821440,
      "step": 60260,
      "train_runtime": 72025.4359,
      "train_tokens_per_second": 62058.929
    },
    {
      "epoch": 0.8009249107979349,
      "grad_norm": 27.75,
      "learning_rate": 2.0079198809165832e-06,
      "loss": 0.9022,
      "num_input_tokens_seen": 4470651392,
      "step": 60270,
      "train_runtime": 72038.8202,
      "train_tokens_per_second": 62058.92
    },
    {
      "epoch": 0.8010578002803969,
      "grad_norm": 31.625,
      "learning_rate": 2.0053337039739884e-06,
      "loss": 0.7252,
      "num_input_tokens_seen": 4471109376,
      "step": 60280,
      "train_runtime": 72047.7288,
      "train_tokens_per_second": 62057.603
    },
    {
      "epoch": 0.8011906897628587,
      "grad_norm": 13.6875,
      "learning_rate": 2.0027490079720325e-06,
      "loss": 0.6313,
      "num_input_tokens_seen": 4471334144,
      "step": 60290,
      "train_runtime": 72052.8908,
      "train_tokens_per_second": 62056.277
    },
    {
      "epoch": 0.8013235792453206,
      "grad_norm": 28.25,
      "learning_rate": 2.000165793389517e-06,
      "loss": 0.5314,
      "num_input_tokens_seen": 4471448576,
      "step": 60300,
      "train_runtime": 72057.1689,
      "train_tokens_per_second": 62054.181
    },
    {
      "epoch": 0.8014564687277825,
      "grad_norm": 22.25,
      "learning_rate": 1.9975840607049524e-06,
      "loss": 0.7366,
      "num_input_tokens_seen": 4473511168,
      "step": 60310,
      "train_runtime": 72089.954,
      "train_tokens_per_second": 62054.571
    },
    {
      "epoch": 0.8015893582102445,
      "grad_norm": 21.625,
      "learning_rate": 1.995003810396585e-06,
      "loss": 0.9006,
      "num_input_tokens_seen": 4474319872,
      "step": 60320,
      "train_runtime": 72103.0461,
      "train_tokens_per_second": 62054.519
    },
    {
      "epoch": 0.8017222476927064,
      "grad_norm": 14.5,
      "learning_rate": 1.992425042942382e-06,
      "loss": 0.7342,
      "num_input_tokens_seen": 4474783232,
      "step": 60330,
      "train_runtime": 72111.8431,
      "train_tokens_per_second": 62053.375
    },
    {
      "epoch": 0.8018551371751683,
      "grad_norm": 15.25,
      "learning_rate": 1.9898477588200382e-06,
      "loss": 0.683,
      "num_input_tokens_seen": 4475016704,
      "step": 60340,
      "train_runtime": 72116.0689,
      "train_tokens_per_second": 62052.976
    },
    {
      "epoch": 0.8019880266576301,
      "grad_norm": 11.4375,
      "learning_rate": 1.9872719585069744e-06,
      "loss": 0.5736,
      "num_input_tokens_seen": 4475136000,
      "step": 60350,
      "train_runtime": 72119.9544,
      "train_tokens_per_second": 62051.287
    },
    {
      "epoch": 0.8021209161400921,
      "grad_norm": 87.0,
      "learning_rate": 1.9846976424803286e-06,
      "loss": 0.9094,
      "num_input_tokens_seen": 4477091072,
      "step": 60360,
      "train_runtime": 72151.9119,
      "train_tokens_per_second": 62050.9
    },
    {
      "epoch": 0.802253805622554,
      "grad_norm": 18.0,
      "learning_rate": 1.982124811216974e-06,
      "loss": 0.8719,
      "num_input_tokens_seen": 4477910272,
      "step": 60370,
      "train_runtime": 72165.887,
      "train_tokens_per_second": 62050.235
    },
    {
      "epoch": 0.8023866951050159,
      "grad_norm": 139.0,
      "learning_rate": 1.9795534651935044e-06,
      "loss": 0.641,
      "num_input_tokens_seen": 4478354432,
      "step": 60380,
      "train_runtime": 72173.0916,
      "train_tokens_per_second": 62050.195
    },
    {
      "epoch": 0.8025195845874779,
      "grad_norm": 41.0,
      "learning_rate": 1.976983604886238e-06,
      "loss": 0.6812,
      "num_input_tokens_seen": 4478580224,
      "step": 60390,
      "train_runtime": 72177.218,
      "train_tokens_per_second": 62049.776
    },
    {
      "epoch": 0.8026524740699398,
      "grad_norm": 18.5,
      "learning_rate": 1.9744152307712185e-06,
      "loss": 0.5281,
      "num_input_tokens_seen": 4478692608,
      "step": 60400,
      "train_runtime": 72180.4379,
      "train_tokens_per_second": 62048.565
    },
    {
      "epoch": 0.8027853635524016,
      "grad_norm": 36.0,
      "learning_rate": 1.9718483433242176e-06,
      "loss": 0.7801,
      "num_input_tokens_seen": 4480754432,
      "step": 60410,
      "train_runtime": 72213.5361,
      "train_tokens_per_second": 62048.678
    },
    {
      "epoch": 0.8029182530348635,
      "grad_norm": 12.3125,
      "learning_rate": 1.9692829430207205e-06,
      "loss": 0.8688,
      "num_input_tokens_seen": 4481614848,
      "step": 60420,
      "train_runtime": 72226.6075,
      "train_tokens_per_second": 62049.361
    },
    {
      "epoch": 0.8030511425173255,
      "grad_norm": 25.375,
      "learning_rate": 1.966719030335954e-06,
      "loss": 0.7486,
      "num_input_tokens_seen": 4482091008,
      "step": 60430,
      "train_runtime": 72233.9001,
      "train_tokens_per_second": 62049.689
    },
    {
      "epoch": 0.8031840319997874,
      "grad_norm": 27.0,
      "learning_rate": 1.9641566057448535e-06,
      "loss": 0.6327,
      "num_input_tokens_seen": 4482315520,
      "step": 60440,
      "train_runtime": 72239.2912,
      "train_tokens_per_second": 62048.166
    },
    {
      "epoch": 0.8033169214822493,
      "grad_norm": 15.3125,
      "learning_rate": 1.9615956697220882e-06,
      "loss": 0.5183,
      "num_input_tokens_seen": 4482433280,
      "step": 60450,
      "train_runtime": 72242.9551,
      "train_tokens_per_second": 62046.649
    },
    {
      "epoch": 0.8034498109647112,
      "grad_norm": 48.25,
      "learning_rate": 1.959036222742048e-06,
      "loss": 0.8126,
      "num_input_tokens_seen": 4484394752,
      "step": 60460,
      "train_runtime": 72275.6103,
      "train_tokens_per_second": 62045.754
    },
    {
      "epoch": 0.8035827004471731,
      "grad_norm": 21.75,
      "learning_rate": 1.9564782652788463e-06,
      "loss": 0.8571,
      "num_input_tokens_seen": 4485213184,
      "step": 60470,
      "train_runtime": 72288.0938,
      "train_tokens_per_second": 62046.361
    },
    {
      "epoch": 0.803715589929635,
      "grad_norm": 22.875,
      "learning_rate": 1.9539217978063265e-06,
      "loss": 0.7181,
      "num_input_tokens_seen": 4485662976,
      "step": 60480,
      "train_runtime": 72295.2416,
      "train_tokens_per_second": 62046.448
    },
    {
      "epoch": 0.8038484794120969,
      "grad_norm": 30.5,
      "learning_rate": 1.951366820798043e-06,
      "loss": 0.7034,
      "num_input_tokens_seen": 4485896448,
      "step": 60490,
      "train_runtime": 72301.9324,
      "train_tokens_per_second": 62043.936
    },
    {
      "epoch": 0.8039813688945588,
      "grad_norm": 20.25,
      "learning_rate": 1.948813334727293e-06,
      "loss": 0.5483,
      "num_input_tokens_seen": 4486013952,
      "step": 60500,
      "train_runtime": 72305.5829,
      "train_tokens_per_second": 62042.428
    },
    {
      "epoch": 0.8041142583770208,
      "grad_norm": 21.5,
      "learning_rate": 1.946261340067078e-06,
      "loss": 0.8484,
      "num_input_tokens_seen": 4488089088,
      "step": 60510,
      "train_runtime": 72339.3694,
      "train_tokens_per_second": 62042.137
    },
    {
      "epoch": 0.8042471478594827,
      "grad_norm": 21.25,
      "learning_rate": 1.943710837290137e-06,
      "loss": 0.896,
      "num_input_tokens_seen": 4488917760,
      "step": 60520,
      "train_runtime": 72352.454,
      "train_tokens_per_second": 62042.37
    },
    {
      "epoch": 0.8043800373419445,
      "grad_norm": 39.25,
      "learning_rate": 1.941161826868929e-06,
      "loss": 0.756,
      "num_input_tokens_seen": 4489403136,
      "step": 60530,
      "train_runtime": 72360.1236,
      "train_tokens_per_second": 62042.502
    },
    {
      "epoch": 0.8045129268244064,
      "grad_norm": 21.375,
      "learning_rate": 1.9386143092756284e-06,
      "loss": 0.6338,
      "num_input_tokens_seen": 4489639680,
      "step": 60540,
      "train_runtime": 72364.4506,
      "train_tokens_per_second": 62042.061
    },
    {
      "epoch": 0.8046458163068684,
      "grad_norm": 10.375,
      "learning_rate": 1.93606828498215e-06,
      "loss": 0.5207,
      "num_input_tokens_seen": 4489758720,
      "step": 60550,
      "train_runtime": 72368.6093,
      "train_tokens_per_second": 62040.141
    },
    {
      "epoch": 0.8047787057893303,
      "grad_norm": 21.25,
      "learning_rate": 1.9335237544601136e-06,
      "loss": 0.7862,
      "num_input_tokens_seen": 4491958528,
      "step": 60560,
      "train_runtime": 72404.0147,
      "train_tokens_per_second": 62040.186
    },
    {
      "epoch": 0.8049115952717922,
      "grad_norm": 21.375,
      "learning_rate": 1.930980718180875e-06,
      "loss": 0.9142,
      "num_input_tokens_seen": 4492829184,
      "step": 60570,
      "train_runtime": 72419.0729,
      "train_tokens_per_second": 62039.308
    },
    {
      "epoch": 0.8050444847542542,
      "grad_norm": 16.625,
      "learning_rate": 1.9284391766155086e-06,
      "loss": 0.782,
      "num_input_tokens_seen": 4493321472,
      "step": 60580,
      "train_runtime": 72426.6079,
      "train_tokens_per_second": 62039.651
    },
    {
      "epoch": 0.805177374236716,
      "grad_norm": 9.75,
      "learning_rate": 1.9258991302348107e-06,
      "loss": 0.6494,
      "num_input_tokens_seen": 4493556224,
      "step": 60590,
      "train_runtime": 72431.0382,
      "train_tokens_per_second": 62039.097
    },
    {
      "epoch": 0.8053102637191779,
      "grad_norm": 14.4375,
      "learning_rate": 1.923360579509307e-06,
      "loss": 0.5499,
      "num_input_tokens_seen": 4493672192,
      "step": 60600,
      "train_runtime": 72434.41,
      "train_tokens_per_second": 62037.81
    },
    {
      "epoch": 0.8054431532016398,
      "grad_norm": 18.75,
      "learning_rate": 1.9208235249092353e-06,
      "loss": 0.8549,
      "num_input_tokens_seen": 4495729920,
      "step": 60610,
      "train_runtime": 72467.55,
      "train_tokens_per_second": 62037.835
    },
    {
      "epoch": 0.8055760426841018,
      "grad_norm": 17.875,
      "learning_rate": 1.9182879669045652e-06,
      "loss": 0.8778,
      "num_input_tokens_seen": 4496568576,
      "step": 60620,
      "train_runtime": 72480.4463,
      "train_tokens_per_second": 62038.368
    },
    {
      "epoch": 0.8057089321665637,
      "grad_norm": 18.375,
      "learning_rate": 1.9157539059649856e-06,
      "loss": 0.7581,
      "num_input_tokens_seen": 4497029888,
      "step": 60630,
      "train_runtime": 72488.7797,
      "train_tokens_per_second": 62037.6
    },
    {
      "epoch": 0.8058418216490256,
      "grad_norm": 15.5625,
      "learning_rate": 1.913221342559909e-06,
      "loss": 0.6154,
      "num_input_tokens_seen": 4497270784,
      "step": 60640,
      "train_runtime": 72492.8239,
      "train_tokens_per_second": 62037.462
    },
    {
      "epoch": 0.8059747111314876,
      "grad_norm": 17.125,
      "learning_rate": 1.91069027715847e-06,
      "loss": 0.531,
      "num_input_tokens_seen": 4497387264,
      "step": 60650,
      "train_runtime": 72496.1066,
      "train_tokens_per_second": 62036.259
    },
    {
      "epoch": 0.8061076006139494,
      "grad_norm": 38.0,
      "learning_rate": 1.9081607102295262e-06,
      "loss": 0.8172,
      "num_input_tokens_seen": 4499450880,
      "step": 60660,
      "train_runtime": 72529.2418,
      "train_tokens_per_second": 62036.37
    },
    {
      "epoch": 0.8062404900964113,
      "grad_norm": 16.5,
      "learning_rate": 1.9056326422416594e-06,
      "loss": 0.907,
      "num_input_tokens_seen": 4500256000,
      "step": 60670,
      "train_runtime": 72542.4004,
      "train_tokens_per_second": 62036.216
    },
    {
      "epoch": 0.8063733795788732,
      "grad_norm": 11.4375,
      "learning_rate": 1.9031060736631679e-06,
      "loss": 0.6779,
      "num_input_tokens_seen": 4500722944,
      "step": 60680,
      "train_runtime": 72551.04,
      "train_tokens_per_second": 62035.264
    },
    {
      "epoch": 0.8065062690613352,
      "grad_norm": 16.75,
      "learning_rate": 1.9005810049620765e-06,
      "loss": 0.6541,
      "num_input_tokens_seen": 4500954624,
      "step": 60690,
      "train_runtime": 72555.2958,
      "train_tokens_per_second": 62034.819
    },
    {
      "epoch": 0.8066391585437971,
      "grad_norm": 13.5625,
      "learning_rate": 1.8980574366061332e-06,
      "loss": 0.4651,
      "num_input_tokens_seen": 4501071872,
      "step": 60700,
      "train_runtime": 72560.0441,
      "train_tokens_per_second": 62032.375
    },
    {
      "epoch": 0.806772048026259,
      "grad_norm": 38.5,
      "learning_rate": 1.8955353690628064e-06,
      "loss": 0.7645,
      "num_input_tokens_seen": 4503188224,
      "step": 60710,
      "train_runtime": 72595.9738,
      "train_tokens_per_second": 62030.826
    },
    {
      "epoch": 0.8069049375087208,
      "grad_norm": 20.25,
      "learning_rate": 1.8930148027992868e-06,
      "loss": 0.9251,
      "num_input_tokens_seen": 4504038912,
      "step": 60720,
      "train_runtime": 72609.2724,
      "train_tokens_per_second": 62031.181
    },
    {
      "epoch": 0.8070378269911828,
      "grad_norm": 31.625,
      "learning_rate": 1.8904957382824873e-06,
      "loss": 0.731,
      "num_input_tokens_seen": 4504516352,
      "step": 60730,
      "train_runtime": 72617.5936,
      "train_tokens_per_second": 62030.647
    },
    {
      "epoch": 0.8071707164736447,
      "grad_norm": 10.3125,
      "learning_rate": 1.8879781759790372e-06,
      "loss": 0.6389,
      "num_input_tokens_seen": 4504750336,
      "step": 60740,
      "train_runtime": 72621.6421,
      "train_tokens_per_second": 62030.411
    },
    {
      "epoch": 0.8073036059561066,
      "grad_norm": 12.5,
      "learning_rate": 1.885462116355301e-06,
      "loss": 0.5025,
      "num_input_tokens_seen": 4504861952,
      "step": 60750,
      "train_runtime": 72625.5584,
      "train_tokens_per_second": 62028.603
    },
    {
      "epoch": 0.8074364954385685,
      "grad_norm": 34.5,
      "learning_rate": 1.8829475598773494e-06,
      "loss": 0.8206,
      "num_input_tokens_seen": 4506984960,
      "step": 60760,
      "train_runtime": 72660.7477,
      "train_tokens_per_second": 62027.781
    },
    {
      "epoch": 0.8075693849210305,
      "grad_norm": 17.75,
      "learning_rate": 1.8804345070109843e-06,
      "loss": 1.0018,
      "num_input_tokens_seen": 4507838464,
      "step": 60770,
      "train_runtime": 72673.8527,
      "train_tokens_per_second": 62028.34
    },
    {
      "epoch": 0.8077022744034923,
      "grad_norm": 22.5,
      "learning_rate": 1.8779229582217284e-06,
      "loss": 0.7553,
      "num_input_tokens_seen": 4508338432,
      "step": 60780,
      "train_runtime": 72681.1667,
      "train_tokens_per_second": 62028.977
    },
    {
      "epoch": 0.8078351638859542,
      "grad_norm": 14.125,
      "learning_rate": 1.8754129139748157e-06,
      "loss": 0.6063,
      "num_input_tokens_seen": 4508582144,
      "step": 60790,
      "train_runtime": 72685.7623,
      "train_tokens_per_second": 62028.408
    },
    {
      "epoch": 0.8079680533684162,
      "grad_norm": 15.125,
      "learning_rate": 1.8729043747352204e-06,
      "loss": 0.5075,
      "num_input_tokens_seen": 4508701440,
      "step": 60800,
      "train_runtime": 72690.6457,
      "train_tokens_per_second": 62025.882
    },
    {
      "epoch": 0.8081009428508781,
      "grad_norm": 11.8125,
      "learning_rate": 1.8703973409676168e-06,
      "loss": 0.7582,
      "num_input_tokens_seen": 4510740992,
      "step": 60810,
      "train_runtime": 72723.4237,
      "train_tokens_per_second": 62025.971
    },
    {
      "epoch": 0.80823383233334,
      "grad_norm": 21.5,
      "learning_rate": 1.867891813136421e-06,
      "loss": 0.8248,
      "num_input_tokens_seen": 4511611904,
      "step": 60820,
      "train_runtime": 72736.2043,
      "train_tokens_per_second": 62027.046
    },
    {
      "epoch": 0.8083667218158019,
      "grad_norm": 15.4375,
      "learning_rate": 1.8653877917057506e-06,
      "loss": 0.7232,
      "num_input_tokens_seen": 4512100864,
      "step": 60830,
      "train_runtime": 72743.4697,
      "train_tokens_per_second": 62027.573
    },
    {
      "epoch": 0.8084996112982638,
      "grad_norm": 20.5,
      "learning_rate": 1.8628852771394568e-06,
      "loss": 0.6539,
      "num_input_tokens_seen": 4512348416,
      "step": 60840,
      "train_runtime": 72747.9478,
      "train_tokens_per_second": 62027.158
    },
    {
      "epoch": 0.8086325007807257,
      "grad_norm": 11.8125,
      "learning_rate": 1.860384269901111e-06,
      "loss": 0.5201,
      "num_input_tokens_seen": 4512466432,
      "step": 60850,
      "train_runtime": 72753.6369,
      "train_tokens_per_second": 62023.929
    },
    {
      "epoch": 0.8087653902631876,
      "grad_norm": 366.0,
      "learning_rate": 1.8578847704539949e-06,
      "loss": 0.8439,
      "num_input_tokens_seen": 4514523136,
      "step": 60860,
      "train_runtime": 72787.3991,
      "train_tokens_per_second": 62023.416
    },
    {
      "epoch": 0.8088982797456495,
      "grad_norm": 15.875,
      "learning_rate": 1.855386779261128e-06,
      "loss": 0.8688,
      "num_input_tokens_seen": 4515383296,
      "step": 60870,
      "train_runtime": 72800.8384,
      "train_tokens_per_second": 62023.782
    },
    {
      "epoch": 0.8090311692281115,
      "grad_norm": 34.5,
      "learning_rate": 1.8528902967852337e-06,
      "loss": 0.7397,
      "num_input_tokens_seen": 4515859456,
      "step": 60880,
      "train_runtime": 72808.6113,
      "train_tokens_per_second": 62023.7
    },
    {
      "epoch": 0.8091640587105734,
      "grad_norm": 26.75,
      "learning_rate": 1.850395323488765e-06,
      "loss": 0.6591,
      "num_input_tokens_seen": 4516091904,
      "step": 60890,
      "train_runtime": 72815.0947,
      "train_tokens_per_second": 62021.37
    },
    {
      "epoch": 0.8092969481930352,
      "grad_norm": 10.6875,
      "learning_rate": 1.8479018598338938e-06,
      "loss": 0.576,
      "num_input_tokens_seen": 4516211456,
      "step": 60900,
      "train_runtime": 72820.4062,
      "train_tokens_per_second": 62018.488
    },
    {
      "epoch": 0.8094298376754971,
      "grad_norm": 21.0,
      "learning_rate": 1.8454099062825127e-06,
      "loss": 0.8068,
      "num_input_tokens_seen": 4518243072,
      "step": 60910,
      "train_runtime": 72852.9522,
      "train_tokens_per_second": 62018.668
    },
    {
      "epoch": 0.8095627271579591,
      "grad_norm": 16.375,
      "learning_rate": 1.8429194632962355e-06,
      "loss": 0.8959,
      "num_input_tokens_seen": 4519041792,
      "step": 60920,
      "train_runtime": 72865.718,
      "train_tokens_per_second": 62018.764
    },
    {
      "epoch": 0.809695616640421,
      "grad_norm": 22.75,
      "learning_rate": 1.8404305313363901e-06,
      "loss": 0.704,
      "num_input_tokens_seen": 4519504384,
      "step": 60930,
      "train_runtime": 72872.6481,
      "train_tokens_per_second": 62019.214
    },
    {
      "epoch": 0.8098285061228829,
      "grad_norm": 25.375,
      "learning_rate": 1.8379431108640311e-06,
      "loss": 0.6217,
      "num_input_tokens_seen": 4519723776,
      "step": 60940,
      "train_runtime": 72877.2202,
      "train_tokens_per_second": 62018.334
    },
    {
      "epoch": 0.8099613956053449,
      "grad_norm": 7.28125,
      "learning_rate": 1.8354572023399309e-06,
      "loss": 0.481,
      "num_input_tokens_seen": 4519839488,
      "step": 60950,
      "train_runtime": 72881.623,
      "train_tokens_per_second": 62016.175
    },
    {
      "epoch": 0.8100942850878067,
      "grad_norm": 17.875,
      "learning_rate": 1.8329728062245822e-06,
      "loss": 0.9075,
      "num_input_tokens_seen": 4521968896,
      "step": 60960,
      "train_runtime": 72916.3589,
      "train_tokens_per_second": 62015.835
    },
    {
      "epoch": 0.8102271745702686,
      "grad_norm": 29.75,
      "learning_rate": 1.830489922978197e-06,
      "loss": 0.959,
      "num_input_tokens_seen": 4522851584,
      "step": 60970,
      "train_runtime": 72931.0484,
      "train_tokens_per_second": 62015.447
    },
    {
      "epoch": 0.8103600640527305,
      "grad_norm": 31.375,
      "learning_rate": 1.8280085530607082e-06,
      "loss": 0.7377,
      "num_input_tokens_seen": 4523366400,
      "step": 60980,
      "train_runtime": 72938.8091,
      "train_tokens_per_second": 62015.907
    },
    {
      "epoch": 0.8104929535351925,
      "grad_norm": 21.0,
      "learning_rate": 1.8255286969317687e-06,
      "loss": 0.705,
      "num_input_tokens_seen": 4523606528,
      "step": 60990,
      "train_runtime": 72943.0207,
      "train_tokens_per_second": 62015.618
    },
    {
      "epoch": 0.8106258430176544,
      "grad_norm": 26.375,
      "learning_rate": 1.823050355050745e-06,
      "loss": 0.5796,
      "num_input_tokens_seen": 4523720960,
      "step": 61000,
      "train_runtime": 72946.3429,
      "train_tokens_per_second": 62014.363
    },
    {
      "epoch": 0.8107587325001163,
      "grad_norm": 17.75,
      "learning_rate": 1.8205735278767301e-06,
      "loss": 0.7822,
      "num_input_tokens_seen": 4525804032,
      "step": 61010,
      "train_runtime": 72979.6241,
      "train_tokens_per_second": 62014.625
    },
    {
      "epoch": 0.8108916219825781,
      "grad_norm": 230.0,
      "learning_rate": 1.8180982158685344e-06,
      "loss": 0.9686,
      "num_input_tokens_seen": 4526657792,
      "step": 61020,
      "train_runtime": 72993.1808,
      "train_tokens_per_second": 62014.804
    },
    {
      "epoch": 0.8110245114650401,
      "grad_norm": 22.75,
      "learning_rate": 1.8156244194846883e-06,
      "loss": 0.7568,
      "num_input_tokens_seen": 4527185920,
      "step": 61030,
      "train_runtime": 73003.127,
      "train_tokens_per_second": 62013.589
    },
    {
      "epoch": 0.811157400947502,
      "grad_norm": 9.3125,
      "learning_rate": 1.8131521391834394e-06,
      "loss": 0.6017,
      "num_input_tokens_seen": 4527438080,
      "step": 61040,
      "train_runtime": 73007.7875,
      "train_tokens_per_second": 62013.084
    },
    {
      "epoch": 0.8112902904299639,
      "grad_norm": 8.875,
      "learning_rate": 1.8106813754227592e-06,
      "loss": 0.5365,
      "num_input_tokens_seen": 4527555840,
      "step": 61050,
      "train_runtime": 73011.9585,
      "train_tokens_per_second": 62011.155
    },
    {
      "epoch": 0.8114231799124259,
      "grad_norm": 26.75,
      "learning_rate": 1.808212128660326e-06,
      "loss": 0.8571,
      "num_input_tokens_seen": 4529629184,
      "step": 61060,
      "train_runtime": 73046.3647,
      "train_tokens_per_second": 62010.33
    },
    {
      "epoch": 0.8115560693948878,
      "grad_norm": 16.125,
      "learning_rate": 1.8057443993535561e-06,
      "loss": 0.8931,
      "num_input_tokens_seen": 4530494208,
      "step": 61070,
      "train_runtime": 73059.4444,
      "train_tokens_per_second": 62011.068
    },
    {
      "epoch": 0.8116889588773496,
      "grad_norm": 24.125,
      "learning_rate": 1.8032781879595663e-06,
      "loss": 0.6897,
      "num_input_tokens_seen": 4530974464,
      "step": 61080,
      "train_runtime": 73068.0558,
      "train_tokens_per_second": 62010.333
    },
    {
      "epoch": 0.8118218483598115,
      "grad_norm": 87.5,
      "learning_rate": 1.800813494935204e-06,
      "loss": 0.7534,
      "num_input_tokens_seen": 4531213568,
      "step": 61090,
      "train_runtime": 73072.7505,
      "train_tokens_per_second": 62009.621
    },
    {
      "epoch": 0.8119547378422735,
      "grad_norm": 27.875,
      "learning_rate": 1.7983503207370334e-06,
      "loss": 0.5096,
      "num_input_tokens_seen": 4531330304,
      "step": 61100,
      "train_runtime": 73076.9818,
      "train_tokens_per_second": 62007.628
    },
    {
      "epoch": 0.8120876273247354,
      "grad_norm": 19.625,
      "learning_rate": 1.7958886658213281e-06,
      "loss": 0.8137,
      "num_input_tokens_seen": 4533369344,
      "step": 61110,
      "train_runtime": 73109.8521,
      "train_tokens_per_second": 62007.639
    },
    {
      "epoch": 0.8122205168071973,
      "grad_norm": 22.5,
      "learning_rate": 1.7934285306440968e-06,
      "loss": 0.8213,
      "num_input_tokens_seen": 4534198272,
      "step": 61120,
      "train_runtime": 73122.67,
      "train_tokens_per_second": 62008.106
    },
    {
      "epoch": 0.8123534062896592,
      "grad_norm": 10.4375,
      "learning_rate": 1.7909699156610528e-06,
      "loss": 0.707,
      "num_input_tokens_seen": 4534679040,
      "step": 61130,
      "train_runtime": 73130.8779,
      "train_tokens_per_second": 62007.721
    },
    {
      "epoch": 0.8124862957721212,
      "grad_norm": 13.25,
      "learning_rate": 1.7885128213276314e-06,
      "loss": 0.6926,
      "num_input_tokens_seen": 4534909696,
      "step": 61140,
      "train_runtime": 73134.7573,
      "train_tokens_per_second": 62007.585
    },
    {
      "epoch": 0.812619185254583,
      "grad_norm": 30.75,
      "learning_rate": 1.7860572480989934e-06,
      "loss": 0.5703,
      "num_input_tokens_seen": 4535024640,
      "step": 61150,
      "train_runtime": 73137.9684,
      "train_tokens_per_second": 62006.434
    },
    {
      "epoch": 0.8127520747370449,
      "grad_norm": 31.75,
      "learning_rate": 1.7836031964300027e-06,
      "loss": 0.8333,
      "num_input_tokens_seen": 4537056512,
      "step": 61160,
      "train_runtime": 73169.8567,
      "train_tokens_per_second": 62007.181
    },
    {
      "epoch": 0.8128849642195068,
      "grad_norm": 15.625,
      "learning_rate": 1.7811506667752598e-06,
      "loss": 0.9584,
      "num_input_tokens_seen": 4537917952,
      "step": 61170,
      "train_runtime": 73182.4291,
      "train_tokens_per_second": 62008.299
    },
    {
      "epoch": 0.8130178537019688,
      "grad_norm": 11.1875,
      "learning_rate": 1.778699659589065e-06,
      "loss": 0.6548,
      "num_input_tokens_seen": 4538413824,
      "step": 61180,
      "train_runtime": 73189.6297,
      "train_tokens_per_second": 62008.974
    },
    {
      "epoch": 0.8131507431844307,
      "grad_norm": 37.5,
      "learning_rate": 1.7762501753254557e-06,
      "loss": 0.5944,
      "num_input_tokens_seen": 4538641152,
      "step": 61190,
      "train_runtime": 73194.0217,
      "train_tokens_per_second": 62008.359
    },
    {
      "epoch": 0.8132836326668926,
      "grad_norm": 12.5625,
      "learning_rate": 1.7738022144381696e-06,
      "loss": 0.5831,
      "num_input_tokens_seen": 4538760448,
      "step": 61200,
      "train_runtime": 73197.1739,
      "train_tokens_per_second": 62007.318
    },
    {
      "epoch": 0.8134165221493544,
      "grad_norm": 30.875,
      "learning_rate": 1.77135577738067e-06,
      "loss": 0.8229,
      "num_input_tokens_seen": 4540839680,
      "step": 61210,
      "train_runtime": 73230.7921,
      "train_tokens_per_second": 62007.245
    },
    {
      "epoch": 0.8135494116318164,
      "grad_norm": 15.9375,
      "learning_rate": 1.7689108646061403e-06,
      "loss": 0.8661,
      "num_input_tokens_seen": 4541674240,
      "step": 61220,
      "train_runtime": 73243.9448,
      "train_tokens_per_second": 62007.505
    },
    {
      "epoch": 0.8136823011142783,
      "grad_norm": 49.25,
      "learning_rate": 1.7664674765674773e-06,
      "loss": 0.7065,
      "num_input_tokens_seen": 4542121984,
      "step": 61230,
      "train_runtime": 73251.3348,
      "train_tokens_per_second": 62007.361
    },
    {
      "epoch": 0.8138151905967402,
      "grad_norm": 36.25,
      "learning_rate": 1.7640256137172995e-06,
      "loss": 0.6364,
      "num_input_tokens_seen": 4542346240,
      "step": 61240,
      "train_runtime": 73255.5836,
      "train_tokens_per_second": 62006.826
    },
    {
      "epoch": 0.8139480800792022,
      "grad_norm": 16.5,
      "learning_rate": 1.7615852765079366e-06,
      "loss": 0.5417,
      "num_input_tokens_seen": 4542457600,
      "step": 61250,
      "train_runtime": 73259.6327,
      "train_tokens_per_second": 62004.919
    },
    {
      "epoch": 0.8140809695616641,
      "grad_norm": 27.0,
      "learning_rate": 1.7591464653914392e-06,
      "loss": 0.8228,
      "num_input_tokens_seen": 4544634624,
      "step": 61260,
      "train_runtime": 73294.3085,
      "train_tokens_per_second": 62005.287
    },
    {
      "epoch": 0.8142138590441259,
      "grad_norm": 50.75,
      "learning_rate": 1.756709180819578e-06,
      "loss": 0.9295,
      "num_input_tokens_seen": 4545537024,
      "step": 61270,
      "train_runtime": 73307.6346,
      "train_tokens_per_second": 62006.325
    },
    {
      "epoch": 0.8143467485265878,
      "grad_norm": 15.25,
      "learning_rate": 1.7542734232438364e-06,
      "loss": 0.7491,
      "num_input_tokens_seen": 4546024704,
      "step": 61280,
      "train_runtime": 73315.0373,
      "train_tokens_per_second": 62006.716
    },
    {
      "epoch": 0.8144796380090498,
      "grad_norm": 18.875,
      "learning_rate": 1.7518391931154166e-06,
      "loss": 0.628,
      "num_input_tokens_seen": 4546261504,
      "step": 61290,
      "train_runtime": 73319.1303,
      "train_tokens_per_second": 62006.484
    },
    {
      "epoch": 0.8146125274915117,
      "grad_norm": 13.3125,
      "learning_rate": 1.7494064908852426e-06,
      "loss": 0.546,
      "num_input_tokens_seen": 4546380032,
      "step": 61300,
      "train_runtime": 73322.9748,
      "train_tokens_per_second": 62004.85
    },
    {
      "epoch": 0.8147454169739736,
      "grad_norm": 31.25,
      "learning_rate": 1.7469753170039405e-06,
      "loss": 0.8428,
      "num_input_tokens_seen": 4548488704,
      "step": 61310,
      "train_runtime": 73358.6358,
      "train_tokens_per_second": 62003.453
    },
    {
      "epoch": 0.8148783064564356,
      "grad_norm": 348.0,
      "learning_rate": 1.7445456719218767e-06,
      "loss": 0.8801,
      "num_input_tokens_seen": 4549338368,
      "step": 61320,
      "train_runtime": 73371.375,
      "train_tokens_per_second": 62004.268
    },
    {
      "epoch": 0.8150111959388974,
      "grad_norm": 11.1875,
      "learning_rate": 1.7421175560891112e-06,
      "loss": 0.756,
      "num_input_tokens_seen": 4549794304,
      "step": 61330,
      "train_runtime": 73378.4175,
      "train_tokens_per_second": 62004.53
    },
    {
      "epoch": 0.8151440854213593,
      "grad_norm": 11.75,
      "learning_rate": 1.7396909699554343e-06,
      "loss": 0.6073,
      "num_input_tokens_seen": 4550029056,
      "step": 61340,
      "train_runtime": 73382.3502,
      "train_tokens_per_second": 62004.406
    },
    {
      "epoch": 0.8152769749038212,
      "grad_norm": 13.875,
      "learning_rate": 1.7372659139703497e-06,
      "loss": 0.5361,
      "num_input_tokens_seen": 4550149888,
      "step": 61350,
      "train_runtime": 73385.7007,
      "train_tokens_per_second": 62003.222
    },
    {
      "epoch": 0.8154098643862832,
      "grad_norm": 75.5,
      "learning_rate": 1.7348423885830766e-06,
      "loss": 0.8726,
      "num_input_tokens_seen": 4552146176,
      "step": 61360,
      "train_runtime": 73417.9307,
      "train_tokens_per_second": 62003.194
    },
    {
      "epoch": 0.8155427538687451,
      "grad_norm": 13.1875,
      "learning_rate": 1.7324203942425544e-06,
      "loss": 0.8594,
      "num_input_tokens_seen": 4552972544,
      "step": 61370,
      "train_runtime": 73430.8539,
      "train_tokens_per_second": 62003.535
    },
    {
      "epoch": 0.815675643351207,
      "grad_norm": 24.125,
      "learning_rate": 1.7299999313974292e-06,
      "loss": 0.7555,
      "num_input_tokens_seen": 4553453824,
      "step": 61380,
      "train_runtime": 73437.8323,
      "train_tokens_per_second": 62004.197
    },
    {
      "epoch": 0.8158085328336688,
      "grad_norm": 17.375,
      "learning_rate": 1.7275810004960791e-06,
      "loss": 0.6285,
      "num_input_tokens_seen": 4553690624,
      "step": 61390,
      "train_runtime": 73443.5712,
      "train_tokens_per_second": 62002.576
    },
    {
      "epoch": 0.8159414223161308,
      "grad_norm": 16.375,
      "learning_rate": 1.7251636019865825e-06,
      "loss": 0.5295,
      "num_input_tokens_seen": 4553809920,
      "step": 61400,
      "train_runtime": 73448.1981,
      "train_tokens_per_second": 62000.295
    },
    {
      "epoch": 0.8160743117985927,
      "grad_norm": 18.625,
      "learning_rate": 1.7227477363167433e-06,
      "loss": 0.8743,
      "num_input_tokens_seen": 4555797504,
      "step": 61410,
      "train_runtime": 73480.6409,
      "train_tokens_per_second": 61999.97
    },
    {
      "epoch": 0.8162072012810546,
      "grad_norm": 15.3125,
      "learning_rate": 1.7203334039340813e-06,
      "loss": 0.9173,
      "num_input_tokens_seen": 4556595200,
      "step": 61420,
      "train_runtime": 73493.5706,
      "train_tokens_per_second": 61999.916
    },
    {
      "epoch": 0.8163400907635165,
      "grad_norm": 18.125,
      "learning_rate": 1.7179206052858232e-06,
      "loss": 0.7766,
      "num_input_tokens_seen": 4557062912,
      "step": 61430,
      "train_runtime": 73500.6624,
      "train_tokens_per_second": 62000.297
    },
    {
      "epoch": 0.8164729802459785,
      "grad_norm": 21.375,
      "learning_rate": 1.7155093408189273e-06,
      "loss": 0.6031,
      "num_input_tokens_seen": 4557293312,
      "step": 61440,
      "train_runtime": 73504.8405,
      "train_tokens_per_second": 61999.908
    },
    {
      "epoch": 0.8166058697284403,
      "grad_norm": 11.9375,
      "learning_rate": 1.7130996109800523e-06,
      "loss": 0.5454,
      "num_input_tokens_seen": 4557409024,
      "step": 61450,
      "train_runtime": 73510.2139,
      "train_tokens_per_second": 61996.95
    },
    {
      "epoch": 0.8167387592109022,
      "grad_norm": 15.0625,
      "learning_rate": 1.710691416215582e-06,
      "loss": 0.7473,
      "num_input_tokens_seen": 4559602176,
      "step": 61460,
      "train_runtime": 73545.479,
      "train_tokens_per_second": 61997.042
    },
    {
      "epoch": 0.8168716486933641,
      "grad_norm": 27.25,
      "learning_rate": 1.7082847569716122e-06,
      "loss": 0.8895,
      "num_input_tokens_seen": 4560463104,
      "step": 61470,
      "train_runtime": 73558.9316,
      "train_tokens_per_second": 61997.408
    },
    {
      "epoch": 0.8170045381758261,
      "grad_norm": 13.4375,
      "learning_rate": 1.7058796336939554e-06,
      "loss": 0.7676,
      "num_input_tokens_seen": 4560937728,
      "step": 61480,
      "train_runtime": 73566.0332,
      "train_tokens_per_second": 61997.875
    },
    {
      "epoch": 0.817137427658288,
      "grad_norm": 14.25,
      "learning_rate": 1.703476046828142e-06,
      "loss": 0.5905,
      "num_input_tokens_seen": 4561176576,
      "step": 61490,
      "train_runtime": 73571.6468,
      "train_tokens_per_second": 61996.391
    },
    {
      "epoch": 0.8172703171407499,
      "grad_norm": 7.71875,
      "learning_rate": 1.7010739968194068e-06,
      "loss": 0.5658,
      "num_input_tokens_seen": 4561291520,
      "step": 61500,
      "train_runtime": 73574.6201,
      "train_tokens_per_second": 61995.448
    },
    {
      "epoch": 0.8174032066232118,
      "grad_norm": 20.0,
      "learning_rate": 1.6986734841127184e-06,
      "loss": 0.7887,
      "num_input_tokens_seen": 4563445248,
      "step": 61510,
      "train_runtime": 73607.7557,
      "train_tokens_per_second": 61996.799
    },
    {
      "epoch": 0.8175360961056737,
      "grad_norm": 21.875,
      "learning_rate": 1.6962745091527422e-06,
      "loss": 0.8602,
      "num_input_tokens_seen": 4564273664,
      "step": 61520,
      "train_runtime": 73620.469,
      "train_tokens_per_second": 61997.346
    },
    {
      "epoch": 0.8176689855881356,
      "grad_norm": 159.0,
      "learning_rate": 1.6938770723838716e-06,
      "loss": 0.7211,
      "num_input_tokens_seen": 4564752640,
      "step": 61530,
      "train_runtime": 73627.2614,
      "train_tokens_per_second": 61998.132
    },
    {
      "epoch": 0.8178018750705975,
      "grad_norm": 12.0,
      "learning_rate": 1.691481174250208e-06,
      "loss": 0.5461,
      "num_input_tokens_seen": 4564974592,
      "step": 61540,
      "train_runtime": 73631.5534,
      "train_tokens_per_second": 61997.532
    },
    {
      "epoch": 0.8179347645530595,
      "grad_norm": 14.0625,
      "learning_rate": 1.6890868151955718e-06,
      "loss": 0.5432,
      "num_input_tokens_seen": 4565087488,
      "step": 61550,
      "train_runtime": 73634.732,
      "train_tokens_per_second": 61996.389
    },
    {
      "epoch": 0.8180676540355214,
      "grad_norm": 52.0,
      "learning_rate": 1.6866939956634987e-06,
      "loss": 0.8046,
      "num_input_tokens_seen": 4567234048,
      "step": 61560,
      "train_runtime": 73669.2456,
      "train_tokens_per_second": 61996.482
    },
    {
      "epoch": 0.8182005435179832,
      "grad_norm": 24.0,
      "learning_rate": 1.6843027160972314e-06,
      "loss": 0.9203,
      "num_input_tokens_seen": 4568093696,
      "step": 61570,
      "train_runtime": 73683.0498,
      "train_tokens_per_second": 61996.534
    },
    {
      "epoch": 0.8183334330004451,
      "grad_norm": 53.5,
      "learning_rate": 1.6819129769397369e-06,
      "loss": 0.7539,
      "num_input_tokens_seen": 4568556288,
      "step": 61580,
      "train_runtime": 73691.1367,
      "train_tokens_per_second": 61996.008
    },
    {
      "epoch": 0.8184663224829071,
      "grad_norm": 14.5625,
      "learning_rate": 1.679524778633692e-06,
      "loss": 0.6031,
      "num_input_tokens_seen": 4568779520,
      "step": 61590,
      "train_runtime": 73695.497,
      "train_tokens_per_second": 61995.369
    },
    {
      "epoch": 0.818599211965369,
      "grad_norm": 25.75,
      "learning_rate": 1.6771381216214887e-06,
      "loss": 0.4477,
      "num_input_tokens_seen": 4568896768,
      "step": 61600,
      "train_runtime": 73698.6818,
      "train_tokens_per_second": 61994.281
    },
    {
      "epoch": 0.8187321014478309,
      "grad_norm": 93.0,
      "learning_rate": 1.6747530063452367e-06,
      "loss": 0.8688,
      "num_input_tokens_seen": 4570932224,
      "step": 61610,
      "train_runtime": 73731.9068,
      "train_tokens_per_second": 61993.951
    },
    {
      "epoch": 0.8188649909302929,
      "grad_norm": 18.125,
      "learning_rate": 1.6723694332467566e-06,
      "loss": 0.9413,
      "num_input_tokens_seen": 4571745792,
      "step": 61620,
      "train_runtime": 73744.2646,
      "train_tokens_per_second": 61994.595
    },
    {
      "epoch": 0.8189978804127548,
      "grad_norm": 15.0625,
      "learning_rate": 1.6699874027675778e-06,
      "loss": 0.6733,
      "num_input_tokens_seen": 4572185088,
      "step": 61630,
      "train_runtime": 73752.293,
      "train_tokens_per_second": 61993.802
    },
    {
      "epoch": 0.8191307698952166,
      "grad_norm": 17.5,
      "learning_rate": 1.6676069153489605e-06,
      "loss": 0.6043,
      "num_input_tokens_seen": 4572413184,
      "step": 61640,
      "train_runtime": 73756.358,
      "train_tokens_per_second": 61993.478
    },
    {
      "epoch": 0.8192636593776785,
      "grad_norm": 13.1875,
      "learning_rate": 1.665227971431862e-06,
      "loss": 0.5514,
      "num_input_tokens_seen": 4572532480,
      "step": 61650,
      "train_runtime": 73759.5869,
      "train_tokens_per_second": 61992.382
    },
    {
      "epoch": 0.8193965488601405,
      "grad_norm": 52.0,
      "learning_rate": 1.6628505714569609e-06,
      "loss": 0.9019,
      "num_input_tokens_seen": 4574507520,
      "step": 61660,
      "train_runtime": 73792.2407,
      "train_tokens_per_second": 61991.715
    },
    {
      "epoch": 0.8195294383426024,
      "grad_norm": 20.875,
      "learning_rate": 1.6604747158646528e-06,
      "loss": 0.8701,
      "num_input_tokens_seen": 4575360512,
      "step": 61670,
      "train_runtime": 73805.2099,
      "train_tokens_per_second": 61992.379
    },
    {
      "epoch": 0.8196623278250643,
      "grad_norm": 24.75,
      "learning_rate": 1.6581004050950366e-06,
      "loss": 0.6828,
      "num_input_tokens_seen": 4575824384,
      "step": 61680,
      "train_runtime": 73813.1962,
      "train_tokens_per_second": 61991.956
    },
    {
      "epoch": 0.8197952173075262,
      "grad_norm": 38.0,
      "learning_rate": 1.655727639587943e-06,
      "loss": 0.6638,
      "num_input_tokens_seen": 4576052736,
      "step": 61690,
      "train_runtime": 73817.5189,
      "train_tokens_per_second": 61991.419
    },
    {
      "epoch": 0.8199281067899881,
      "grad_norm": 7.9375,
      "learning_rate": 1.6533564197828934e-06,
      "loss": 0.5181,
      "num_input_tokens_seen": 4576167936,
      "step": 61700,
      "train_runtime": 73820.408,
      "train_tokens_per_second": 61990.553
    },
    {
      "epoch": 0.82006099627245,
      "grad_norm": 20.875,
      "learning_rate": 1.6509867461191476e-06,
      "loss": 0.7871,
      "num_input_tokens_seen": 4578318848,
      "step": 61710,
      "train_runtime": 73854.905,
      "train_tokens_per_second": 61990.721
    },
    {
      "epoch": 0.8201938857549119,
      "grad_norm": 22.625,
      "learning_rate": 1.6486186190356568e-06,
      "loss": 0.8806,
      "num_input_tokens_seen": 4579223552,
      "step": 61720,
      "train_runtime": 73868.4256,
      "train_tokens_per_second": 61991.622
    },
    {
      "epoch": 0.8203267752373739,
      "grad_norm": 34.25,
      "learning_rate": 1.646252038971099e-06,
      "loss": 0.8304,
      "num_input_tokens_seen": 4579730176,
      "step": 61730,
      "train_runtime": 73875.9889,
      "train_tokens_per_second": 61992.134
    },
    {
      "epoch": 0.8204596647198358,
      "grad_norm": 30.75,
      "learning_rate": 1.6438870063638645e-06,
      "loss": 0.6484,
      "num_input_tokens_seen": 4579961344,
      "step": 61740,
      "train_runtime": 73880.1666,
      "train_tokens_per_second": 61991.757
    },
    {
      "epoch": 0.8205925542022977,
      "grad_norm": 6.53125,
      "learning_rate": 1.6415235216520476e-06,
      "loss": 0.5038,
      "num_input_tokens_seen": 4580078080,
      "step": 61750,
      "train_runtime": 73884.682,
      "train_tokens_per_second": 61989.549
    },
    {
      "epoch": 0.8207254436847595,
      "grad_norm": 19.5,
      "learning_rate": 1.6391615852734722e-06,
      "loss": 0.8163,
      "num_input_tokens_seen": 4582224640,
      "step": 61760,
      "train_runtime": 73919.6415,
      "train_tokens_per_second": 61989.27
    },
    {
      "epoch": 0.8208583331672215,
      "grad_norm": 25.25,
      "learning_rate": 1.6368011976656573e-06,
      "loss": 0.9201,
      "num_input_tokens_seen": 4583121408,
      "step": 61770,
      "train_runtime": 73933.0582,
      "train_tokens_per_second": 61990.151
    },
    {
      "epoch": 0.8209912226496834,
      "grad_norm": 18.375,
      "learning_rate": 1.6344423592658487e-06,
      "loss": 0.7413,
      "num_input_tokens_seen": 4583596288,
      "step": 61780,
      "train_runtime": 73940.3565,
      "train_tokens_per_second": 61990.454
    },
    {
      "epoch": 0.8211241121321453,
      "grad_norm": 21.875,
      "learning_rate": 1.632085070510998e-06,
      "loss": 0.686,
      "num_input_tokens_seen": 4583845632,
      "step": 61790,
      "train_runtime": 73944.4296,
      "train_tokens_per_second": 61990.412
    },
    {
      "epoch": 0.8212570016146072,
      "grad_norm": 14.1875,
      "learning_rate": 1.6297293318377728e-06,
      "loss": 0.4686,
      "num_input_tokens_seen": 4583961600,
      "step": 61800,
      "train_runtime": 73947.724,
      "train_tokens_per_second": 61989.218
    },
    {
      "epoch": 0.8213898910970692,
      "grad_norm": 36.0,
      "learning_rate": 1.6273751436825546e-06,
      "loss": 0.7445,
      "num_input_tokens_seen": 4586150656,
      "step": 61810,
      "train_runtime": 73984.236,
      "train_tokens_per_second": 61988.214
    },
    {
      "epoch": 0.821522780579531,
      "grad_norm": 38.25,
      "learning_rate": 1.6250225064814317e-06,
      "loss": 0.9534,
      "num_input_tokens_seen": 4587012352,
      "step": 61820,
      "train_runtime": 73997.1394,
      "train_tokens_per_second": 61989.05
    },
    {
      "epoch": 0.8216556700619929,
      "grad_norm": 14.5625,
      "learning_rate": 1.622671420670211e-06,
      "loss": 0.7431,
      "num_input_tokens_seen": 4587467776,
      "step": 61830,
      "train_runtime": 74004.2509,
      "train_tokens_per_second": 61989.247
    },
    {
      "epoch": 0.8217885595444548,
      "grad_norm": 21.5,
      "learning_rate": 1.6203218866844096e-06,
      "loss": 0.625,
      "num_input_tokens_seen": 4587693824,
      "step": 61840,
      "train_runtime": 74008.1959,
      "train_tokens_per_second": 61988.997
    },
    {
      "epoch": 0.8219214490269168,
      "grad_norm": 21.5,
      "learning_rate": 1.617973904959259e-06,
      "loss": 0.5826,
      "num_input_tokens_seen": 4587810304,
      "step": 61850,
      "train_runtime": 74011.3664,
      "train_tokens_per_second": 61987.915
    },
    {
      "epoch": 0.8220543385093787,
      "grad_norm": 41.0,
      "learning_rate": 1.615627475929702e-06,
      "loss": 0.8032,
      "num_input_tokens_seen": 4589938944,
      "step": 61860,
      "train_runtime": 74046.2726,
      "train_tokens_per_second": 61987.441
    },
    {
      "epoch": 0.8221872279918406,
      "grad_norm": 25.75,
      "learning_rate": 1.6132826000303925e-06,
      "loss": 0.9159,
      "num_input_tokens_seen": 4590743040,
      "step": 61870,
      "train_runtime": 74060.4184,
      "train_tokens_per_second": 61986.458
    },
    {
      "epoch": 0.8223201174743024,
      "grad_norm": 51.0,
      "learning_rate": 1.6109392776957012e-06,
      "loss": 0.6739,
      "num_input_tokens_seen": 4591209472,
      "step": 61880,
      "train_runtime": 74067.4309,
      "train_tokens_per_second": 61986.887
    },
    {
      "epoch": 0.8224530069567644,
      "grad_norm": 12.8125,
      "learning_rate": 1.6085975093597018e-06,
      "loss": 0.6454,
      "num_input_tokens_seen": 4591440896,
      "step": 61890,
      "train_runtime": 74071.2968,
      "train_tokens_per_second": 61986.776
    },
    {
      "epoch": 0.8225858964392263,
      "grad_norm": 9.9375,
      "learning_rate": 1.6062572954561895e-06,
      "loss": 0.5458,
      "num_input_tokens_seen": 4591556608,
      "step": 61900,
      "train_runtime": 74074.5626,
      "train_tokens_per_second": 61985.605
    },
    {
      "epoch": 0.8227187859216882,
      "grad_norm": 19.125,
      "learning_rate": 1.6039186364186677e-06,
      "loss": 0.7952,
      "num_input_tokens_seen": 4593666304,
      "step": 61910,
      "train_runtime": 74108.7968,
      "train_tokens_per_second": 61985.439
    },
    {
      "epoch": 0.8228516754041502,
      "grad_norm": 11.5,
      "learning_rate": 1.6015815326803519e-06,
      "loss": 0.9404,
      "num_input_tokens_seen": 4594493952,
      "step": 61920,
      "train_runtime": 74121.5936,
      "train_tokens_per_second": 61985.904
    },
    {
      "epoch": 0.8229845648866121,
      "grad_norm": 15.5625,
      "learning_rate": 1.5992459846741692e-06,
      "loss": 0.7136,
      "num_input_tokens_seen": 4594924032,
      "step": 61930,
      "train_runtime": 74130.0074,
      "train_tokens_per_second": 61984.67
    },
    {
      "epoch": 0.8231174543690739,
      "grad_norm": 12.8125,
      "learning_rate": 1.5969119928327636e-06,
      "loss": 0.665,
      "num_input_tokens_seen": 4595149824,
      "step": 61940,
      "train_runtime": 74133.9879,
      "train_tokens_per_second": 61984.387
    },
    {
      "epoch": 0.8232503438515358,
      "grad_norm": 10.3125,
      "learning_rate": 1.594579557588476e-06,
      "loss": 0.4443,
      "num_input_tokens_seen": 4595259904,
      "step": 61950,
      "train_runtime": 74137.8959,
      "train_tokens_per_second": 61982.605
    },
    {
      "epoch": 0.8233832333339978,
      "grad_norm": 13.375,
      "learning_rate": 1.5922486793733815e-06,
      "loss": 0.7035,
      "num_input_tokens_seen": 4597341440,
      "step": 61960,
      "train_runtime": 74172.586,
      "train_tokens_per_second": 61981.679
    },
    {
      "epoch": 0.8235161228164597,
      "grad_norm": 30.25,
      "learning_rate": 1.5899193586192452e-06,
      "loss": 0.8261,
      "num_input_tokens_seen": 4598169088,
      "step": 61970,
      "train_runtime": 74185.2068,
      "train_tokens_per_second": 61982.291
    },
    {
      "epoch": 0.8236490122989216,
      "grad_norm": 26.875,
      "learning_rate": 1.5875915957575572e-06,
      "loss": 0.8129,
      "num_input_tokens_seen": 4598616832,
      "step": 61980,
      "train_runtime": 74193.5884,
      "train_tokens_per_second": 61981.324
    },
    {
      "epoch": 0.8237819017813836,
      "grad_norm": 11.625,
      "learning_rate": 1.585265391219516e-06,
      "loss": 0.6401,
      "num_input_tokens_seen": 4598841856,
      "step": 61990,
      "train_runtime": 74197.8551,
      "train_tokens_per_second": 61980.792
    },
    {
      "epoch": 0.8239147912638454,
      "grad_norm": 23.0,
      "learning_rate": 1.5829407454360236e-06,
      "loss": 0.4894,
      "num_input_tokens_seen": 4598957568,
      "step": 62000,
      "train_runtime": 74201.1114,
      "train_tokens_per_second": 61979.632
    },
    {
      "epoch": 0.8240476807463073,
      "grad_norm": 19.875,
      "learning_rate": 1.5806176588377097e-06,
      "loss": 0.8241,
      "num_input_tokens_seen": 4601030144,
      "step": 62010,
      "train_runtime": 74234.7036,
      "train_tokens_per_second": 61979.504
    },
    {
      "epoch": 0.8241805702287692,
      "grad_norm": 24.625,
      "learning_rate": 1.5782961318548951e-06,
      "loss": 0.8896,
      "num_input_tokens_seen": 4601881344,
      "step": 62020,
      "train_runtime": 74247.768,
      "train_tokens_per_second": 61980.063
    },
    {
      "epoch": 0.8243134597112312,
      "grad_norm": 12.1875,
      "learning_rate": 1.575976164917632e-06,
      "loss": 0.7161,
      "num_input_tokens_seen": 4602355968,
      "step": 62030,
      "train_runtime": 74256.5914,
      "train_tokens_per_second": 61979.09
    },
    {
      "epoch": 0.8244463491936931,
      "grad_norm": 21.875,
      "learning_rate": 1.5736577584556666e-06,
      "loss": 0.655,
      "num_input_tokens_seen": 4602589952,
      "step": 62040,
      "train_runtime": 74260.7322,
      "train_tokens_per_second": 61978.785
    },
    {
      "epoch": 0.824579238676155,
      "grad_norm": 7.125,
      "learning_rate": 1.5713409128984646e-06,
      "loss": 0.4771,
      "num_input_tokens_seen": 4602707712,
      "step": 62050,
      "train_runtime": 74263.9345,
      "train_tokens_per_second": 61977.698
    },
    {
      "epoch": 0.8247121281586169,
      "grad_norm": 29.375,
      "learning_rate": 1.5690256286752038e-06,
      "loss": 0.7979,
      "num_input_tokens_seen": 4604742400,
      "step": 62060,
      "train_runtime": 74297.1996,
      "train_tokens_per_second": 61977.335
    },
    {
      "epoch": 0.8248450176410788,
      "grad_norm": 17.75,
      "learning_rate": 1.5667119062147639e-06,
      "loss": 0.953,
      "num_input_tokens_seen": 4605590272,
      "step": 62070,
      "train_runtime": 74310.391,
      "train_tokens_per_second": 61977.742
    },
    {
      "epoch": 0.8249779071235407,
      "grad_norm": 14.8125,
      "learning_rate": 1.5643997459457494e-06,
      "loss": 0.7872,
      "num_input_tokens_seen": 4606078720,
      "step": 62080,
      "train_runtime": 74317.9312,
      "train_tokens_per_second": 61978.027
    },
    {
      "epoch": 0.8251107966060026,
      "grad_norm": 10.9375,
      "learning_rate": 1.5620891482964595e-06,
      "loss": 0.6285,
      "num_input_tokens_seen": 4606308608,
      "step": 62090,
      "train_runtime": 74322.2437,
      "train_tokens_per_second": 61977.524
    },
    {
      "epoch": 0.8252436860884645,
      "grad_norm": 9.8125,
      "learning_rate": 1.5597801136949175e-06,
      "loss": 0.5805,
      "num_input_tokens_seen": 4606423808,
      "step": 62100,
      "train_runtime": 74327.513,
      "train_tokens_per_second": 61974.68
    },
    {
      "epoch": 0.8253765755709265,
      "grad_norm": 34.0,
      "learning_rate": 1.557472642568848e-06,
      "loss": 0.8859,
      "num_input_tokens_seen": 4608447744,
      "step": 62110,
      "train_runtime": 74360.1579,
      "train_tokens_per_second": 61974.69
    },
    {
      "epoch": 0.8255094650533884,
      "grad_norm": 21.25,
      "learning_rate": 1.5551667353456912e-06,
      "loss": 0.9738,
      "num_input_tokens_seen": 4609268224,
      "step": 62120,
      "train_runtime": 74372.9173,
      "train_tokens_per_second": 61975.09
    },
    {
      "epoch": 0.8256423545358502,
      "grad_norm": 28.5,
      "learning_rate": 1.5528623924525976e-06,
      "loss": 0.7696,
      "num_input_tokens_seen": 4609720320,
      "step": 62130,
      "train_runtime": 74379.9915,
      "train_tokens_per_second": 61975.274
    },
    {
      "epoch": 0.8257752440183121,
      "grad_norm": 18.875,
      "learning_rate": 1.5505596143164226e-06,
      "loss": 0.659,
      "num_input_tokens_seen": 4609954560,
      "step": 62140,
      "train_runtime": 74383.982,
      "train_tokens_per_second": 61975.098
    },
    {
      "epoch": 0.8259081335007741,
      "grad_norm": 9.0,
      "learning_rate": 1.548258401363736e-06,
      "loss": 0.5427,
      "num_input_tokens_seen": 4610065920,
      "step": 62150,
      "train_runtime": 74388.5512,
      "train_tokens_per_second": 61972.788
    },
    {
      "epoch": 0.826041022983236,
      "grad_norm": 26.25,
      "learning_rate": 1.5459587540208177e-06,
      "loss": 0.8744,
      "num_input_tokens_seen": 4612208896,
      "step": 62160,
      "train_runtime": 74422.4898,
      "train_tokens_per_second": 61973.322
    },
    {
      "epoch": 0.8261739124656979,
      "grad_norm": 19.375,
      "learning_rate": 1.5436606727136572e-06,
      "loss": 0.8979,
      "num_input_tokens_seen": 4613090560,
      "step": 62170,
      "train_runtime": 74435.433,
      "train_tokens_per_second": 61974.39
    },
    {
      "epoch": 0.8263068019481599,
      "grad_norm": 35.0,
      "learning_rate": 1.5413641578679528e-06,
      "loss": 0.81,
      "num_input_tokens_seen": 4613554688,
      "step": 62180,
      "train_runtime": 74442.4329,
      "train_tokens_per_second": 61974.797
    },
    {
      "epoch": 0.8264396914306217,
      "grad_norm": 21.625,
      "learning_rate": 1.5390692099091143e-06,
      "loss": 0.5817,
      "num_input_tokens_seen": 4613782784,
      "step": 62190,
      "train_runtime": 74446.9061,
      "train_tokens_per_second": 61974.137
    },
    {
      "epoch": 0.8265725809130836,
      "grad_norm": 12.375,
      "learning_rate": 1.5367758292622625e-06,
      "loss": 0.5689,
      "num_input_tokens_seen": 4613899264,
      "step": 62200,
      "train_runtime": 74450.3815,
      "train_tokens_per_second": 61972.809
    },
    {
      "epoch": 0.8267054703955455,
      "grad_norm": 26.5,
      "learning_rate": 1.5344840163522212e-06,
      "loss": 0.8874,
      "num_input_tokens_seen": 4615975680,
      "step": 62210,
      "train_runtime": 74483.6451,
      "train_tokens_per_second": 61973.01
    },
    {
      "epoch": 0.8268383598780075,
      "grad_norm": 36.0,
      "learning_rate": 1.5321937716035295e-06,
      "loss": 0.8663,
      "num_input_tokens_seen": 4616841728,
      "step": 62220,
      "train_runtime": 74496.7985,
      "train_tokens_per_second": 61973.693
    },
    {
      "epoch": 0.8269712493604694,
      "grad_norm": 21.5,
      "learning_rate": 1.5299050954404348e-06,
      "loss": 0.6936,
      "num_input_tokens_seen": 4617330176,
      "step": 62230,
      "train_runtime": 74505.3737,
      "train_tokens_per_second": 61973.116
    },
    {
      "epoch": 0.8271041388429313,
      "grad_norm": 30.625,
      "learning_rate": 1.5276179882868958e-06,
      "loss": 0.5936,
      "num_input_tokens_seen": 4617559552,
      "step": 62240,
      "train_runtime": 74509.1501,
      "train_tokens_per_second": 61973.054
    },
    {
      "epoch": 0.8272370283253931,
      "grad_norm": 9.0625,
      "learning_rate": 1.5253324505665778e-06,
      "loss": 0.5401,
      "num_input_tokens_seen": 4617674752,
      "step": 62250,
      "train_runtime": 74513.0287,
      "train_tokens_per_second": 61971.374
    },
    {
      "epoch": 0.8273699178078551,
      "grad_norm": 45.0,
      "learning_rate": 1.523048482702858e-06,
      "loss": 0.802,
      "num_input_tokens_seen": 4619803392,
      "step": 62260,
      "train_runtime": 74546.1668,
      "train_tokens_per_second": 61972.38
    },
    {
      "epoch": 0.827502807290317,
      "grad_norm": 15.25,
      "learning_rate": 1.5207660851188144e-06,
      "loss": 0.8904,
      "num_input_tokens_seen": 4620634368,
      "step": 62270,
      "train_runtime": 74559.1214,
      "train_tokens_per_second": 61972.758
    },
    {
      "epoch": 0.8276356967727789,
      "grad_norm": 24.0,
      "learning_rate": 1.5184852582372512e-06,
      "loss": 0.7398,
      "num_input_tokens_seen": 4621085184,
      "step": 62280,
      "train_runtime": 74566.3425,
      "train_tokens_per_second": 61972.802
    },
    {
      "epoch": 0.8277685862552409,
      "grad_norm": 18.75,
      "learning_rate": 1.5162060024806623e-06,
      "loss": 0.6433,
      "num_input_tokens_seen": 4621308160,
      "step": 62290,
      "train_runtime": 74571.8038,
      "train_tokens_per_second": 61971.254
    },
    {
      "epoch": 0.8279014757377028,
      "grad_norm": 7.5,
      "learning_rate": 1.5139283182712638e-06,
      "loss": 0.5523,
      "num_input_tokens_seen": 4621420288,
      "step": 62300,
      "train_runtime": 74575.3726,
      "train_tokens_per_second": 61969.791
    },
    {
      "epoch": 0.8280343652201646,
      "grad_norm": 27.625,
      "learning_rate": 1.5116522060309768e-06,
      "loss": 0.8012,
      "num_input_tokens_seen": 4623596032,
      "step": 62310,
      "train_runtime": 74611.0015,
      "train_tokens_per_second": 61969.36
    },
    {
      "epoch": 0.8281672547026265,
      "grad_norm": 25.625,
      "learning_rate": 1.5093776661814251e-06,
      "loss": 0.837,
      "num_input_tokens_seen": 4624439040,
      "step": 62320,
      "train_runtime": 74624.2057,
      "train_tokens_per_second": 61969.692
    },
    {
      "epoch": 0.8283001441850885,
      "grad_norm": 17.375,
      "learning_rate": 1.5071046991439564e-06,
      "loss": 0.714,
      "num_input_tokens_seen": 4624925952,
      "step": 62330,
      "train_runtime": 74631.3294,
      "train_tokens_per_second": 61970.301
    },
    {
      "epoch": 0.8284330336675504,
      "grad_norm": 17.25,
      "learning_rate": 1.5048333053396092e-06,
      "loss": 0.5599,
      "num_input_tokens_seen": 4625153280,
      "step": 62340,
      "train_runtime": 74636.5427,
      "train_tokens_per_second": 61969.018
    },
    {
      "epoch": 0.8285659231500123,
      "grad_norm": 24.75,
      "learning_rate": 1.5025634851891425e-06,
      "loss": 0.5211,
      "num_input_tokens_seen": 4625267968,
      "step": 62350,
      "train_runtime": 74640.1085,
      "train_tokens_per_second": 61967.594
    },
    {
      "epoch": 0.8286988126324742,
      "grad_norm": 14.4375,
      "learning_rate": 1.5002952391130222e-06,
      "loss": 0.8644,
      "num_input_tokens_seen": 4627219712,
      "step": 62360,
      "train_runtime": 74672.2073,
      "train_tokens_per_second": 61967.094
    },
    {
      "epoch": 0.8288317021149361,
      "grad_norm": 15.625,
      "learning_rate": 1.4980285675314144e-06,
      "loss": 0.9182,
      "num_input_tokens_seen": 4628059136,
      "step": 62370,
      "train_runtime": 74684.9279,
      "train_tokens_per_second": 61967.779
    },
    {
      "epoch": 0.828964591597398,
      "grad_norm": 15.875,
      "learning_rate": 1.4957634708642078e-06,
      "loss": 0.7562,
      "num_input_tokens_seen": 4628520704,
      "step": 62380,
      "train_runtime": 74693.0156,
      "train_tokens_per_second": 61967.249
    },
    {
      "epoch": 0.8290974810798599,
      "grad_norm": 18.25,
      "learning_rate": 1.4934999495309832e-06,
      "loss": 0.6551,
      "num_input_tokens_seen": 4628745728,
      "step": 62390,
      "train_runtime": 74697.0415,
      "train_tokens_per_second": 61966.922
    },
    {
      "epoch": 0.8292303705623219,
      "grad_norm": 11.125,
      "learning_rate": 1.4912380039510455e-06,
      "loss": 0.5603,
      "num_input_tokens_seen": 4628859392,
      "step": 62400,
      "train_runtime": 74700.0728,
      "train_tokens_per_second": 61965.929
    },
    {
      "epoch": 0.8293632600447838,
      "grad_norm": 34.75,
      "learning_rate": 1.4889776345433947e-06,
      "loss": 0.8507,
      "num_input_tokens_seen": 4630907904,
      "step": 62410,
      "train_runtime": 74734.3848,
      "train_tokens_per_second": 61964.89
    },
    {
      "epoch": 0.8294961495272457,
      "grad_norm": 47.75,
      "learning_rate": 1.4867188417267464e-06,
      "loss": 0.8626,
      "num_input_tokens_seen": 4631798528,
      "step": 62420,
      "train_runtime": 74747.7611,
      "train_tokens_per_second": 61965.716
    },
    {
      "epoch": 0.8296290390097075,
      "grad_norm": 31.75,
      "learning_rate": 1.4844616259195199e-06,
      "loss": 0.7384,
      "num_input_tokens_seen": 4632289792,
      "step": 62430,
      "train_runtime": 74755.1944,
      "train_tokens_per_second": 61966.126
    },
    {
      "epoch": 0.8297619284921695,
      "grad_norm": 8.9375,
      "learning_rate": 1.4822059875398465e-06,
      "loss": 0.5295,
      "num_input_tokens_seen": 4632517120,
      "step": 62440,
      "train_runtime": 74758.9267,
      "train_tokens_per_second": 61966.073
    },
    {
      "epoch": 0.8298948179746314,
      "grad_norm": 10.8125,
      "learning_rate": 1.479951927005564e-06,
      "loss": 0.5191,
      "num_input_tokens_seen": 4632634880,
      "step": 62450,
      "train_runtime": 74763.55,
      "train_tokens_per_second": 61963.816
    },
    {
      "epoch": 0.8300277074570933,
      "grad_norm": 41.25,
      "learning_rate": 1.477699444734213e-06,
      "loss": 0.8372,
      "num_input_tokens_seen": 4634674688,
      "step": 62460,
      "train_runtime": 74795.5809,
      "train_tokens_per_second": 61964.552
    },
    {
      "epoch": 0.8301605969395552,
      "grad_norm": 28.875,
      "learning_rate": 1.4754485411430474e-06,
      "loss": 0.9141,
      "num_input_tokens_seen": 4635510272,
      "step": 62470,
      "train_runtime": 74808.1442,
      "train_tokens_per_second": 61965.316
    },
    {
      "epoch": 0.8302934864220172,
      "grad_norm": 15.1875,
      "learning_rate": 1.4731992166490282e-06,
      "loss": 0.7564,
      "num_input_tokens_seen": 4635971072,
      "step": 62480,
      "train_runtime": 74815.0383,
      "train_tokens_per_second": 61965.765
    },
    {
      "epoch": 0.830426375904479,
      "grad_norm": 22.125,
      "learning_rate": 1.470951471668821e-06,
      "loss": 0.7178,
      "num_input_tokens_seen": 4636205568,
      "step": 62490,
      "train_runtime": 74819.1037,
      "train_tokens_per_second": 61965.532
    },
    {
      "epoch": 0.8305592653869409,
      "grad_norm": 15.6875,
      "learning_rate": 1.4687053066188029e-06,
      "loss": 0.468,
      "num_input_tokens_seen": 4636320768,
      "step": 62500,
      "train_runtime": 74824.1814,
      "train_tokens_per_second": 61962.867
    },
    {
      "epoch": 0.8306921548694028,
      "grad_norm": 22.0,
      "learning_rate": 1.4664607219150562e-06,
      "loss": 0.7857,
      "num_input_tokens_seen": 4638333184,
      "step": 62510,
      "train_runtime": 74857.8241,
      "train_tokens_per_second": 61961.902
    },
    {
      "epoch": 0.8308250443518648,
      "grad_norm": 36.75,
      "learning_rate": 1.4642177179733663e-06,
      "loss": 0.8506,
      "num_input_tokens_seen": 4639182080,
      "step": 62520,
      "train_runtime": 74870.2036,
      "train_tokens_per_second": 61962.995
    },
    {
      "epoch": 0.8309579338343267,
      "grad_norm": 15.75,
      "learning_rate": 1.461976295209232e-06,
      "loss": 0.6915,
      "num_input_tokens_seen": 4639655936,
      "step": 62530,
      "train_runtime": 74876.8461,
      "train_tokens_per_second": 61963.827
    },
    {
      "epoch": 0.8310908233167886,
      "grad_norm": 32.25,
      "learning_rate": 1.4597364540378578e-06,
      "loss": 0.7341,
      "num_input_tokens_seen": 4639888128,
      "step": 62540,
      "train_runtime": 74882.5928,
      "train_tokens_per_second": 61962.172
    },
    {
      "epoch": 0.8312237127992506,
      "grad_norm": 18.5,
      "learning_rate": 1.4574981948741518e-06,
      "loss": 0.5498,
      "num_input_tokens_seen": 4640005120,
      "step": 62550,
      "train_runtime": 74886.3088,
      "train_tokens_per_second": 61960.66
    },
    {
      "epoch": 0.8313566022817124,
      "grad_norm": 20.5,
      "learning_rate": 1.4552615181327345e-06,
      "loss": 0.8214,
      "num_input_tokens_seen": 4641999360,
      "step": 62560,
      "train_runtime": 74918.1407,
      "train_tokens_per_second": 61960.953
    },
    {
      "epoch": 0.8314894917641743,
      "grad_norm": 99.0,
      "learning_rate": 1.453026424227928e-06,
      "loss": 0.8625,
      "num_input_tokens_seen": 4642804736,
      "step": 62570,
      "train_runtime": 74930.3078,
      "train_tokens_per_second": 61961.64
    },
    {
      "epoch": 0.8316223812466362,
      "grad_norm": 15.4375,
      "learning_rate": 1.4507929135737676e-06,
      "loss": 0.7769,
      "num_input_tokens_seen": 4643250432,
      "step": 62580,
      "train_runtime": 74936.7276,
      "train_tokens_per_second": 61962.279
    },
    {
      "epoch": 0.8317552707290982,
      "grad_norm": 18.125,
      "learning_rate": 1.4485609865839823e-06,
      "loss": 0.6491,
      "num_input_tokens_seen": 4643474944,
      "step": 62590,
      "train_runtime": 74940.5562,
      "train_tokens_per_second": 61962.109
    },
    {
      "epoch": 0.8318881602115601,
      "grad_norm": 15.375,
      "learning_rate": 1.4463306436720282e-06,
      "loss": 0.5394,
      "num_input_tokens_seen": 4643592448,
      "step": 62600,
      "train_runtime": 74943.7462,
      "train_tokens_per_second": 61961.04
    },
    {
      "epoch": 0.832021049694022,
      "grad_norm": 12.25,
      "learning_rate": 1.4441018852510468e-06,
      "loss": 0.8527,
      "num_input_tokens_seen": 4645622272,
      "step": 62610,
      "train_runtime": 74977.2463,
      "train_tokens_per_second": 61960.428
    },
    {
      "epoch": 0.8321539391764838,
      "grad_norm": 17.375,
      "learning_rate": 1.4418747117339015e-06,
      "loss": 0.8591,
      "num_input_tokens_seen": 4646447872,
      "step": 62620,
      "train_runtime": 74991.174,
      "train_tokens_per_second": 61959.93
    },
    {
      "epoch": 0.8322868286589458,
      "grad_norm": 23.0,
      "learning_rate": 1.4396491235331555e-06,
      "loss": 0.7601,
      "num_input_tokens_seen": 4646963968,
      "step": 62630,
      "train_runtime": 74998.4219,
      "train_tokens_per_second": 61960.823
    },
    {
      "epoch": 0.8324197181414077,
      "grad_norm": 42.75,
      "learning_rate": 1.4374251210610746e-06,
      "loss": 0.6176,
      "num_input_tokens_seen": 4647204096,
      "step": 62640,
      "train_runtime": 75002.7316,
      "train_tokens_per_second": 61960.465
    },
    {
      "epoch": 0.8325526076238696,
      "grad_norm": 21.875,
      "learning_rate": 1.4352027047296423e-06,
      "loss": 0.5014,
      "num_input_tokens_seen": 4647322368,
      "step": 62650,
      "train_runtime": 75005.6156,
      "train_tokens_per_second": 61959.659
    },
    {
      "epoch": 0.8326854971063316,
      "grad_norm": 12.375,
      "learning_rate": 1.4329818749505363e-06,
      "loss": 0.8896,
      "num_input_tokens_seen": 4649399296,
      "step": 62660,
      "train_runtime": 75038.0745,
      "train_tokens_per_second": 61960.536
    },
    {
      "epoch": 0.8328183865887935,
      "grad_norm": 25.0,
      "learning_rate": 1.4307626321351476e-06,
      "loss": 0.9101,
      "num_input_tokens_seen": 4650243072,
      "step": 62670,
      "train_runtime": 75052.0995,
      "train_tokens_per_second": 61960.2
    },
    {
      "epoch": 0.8329512760712553,
      "grad_norm": 54.25,
      "learning_rate": 1.4285449766945726e-06,
      "loss": 0.7246,
      "num_input_tokens_seen": 4650735360,
      "step": 62680,
      "train_runtime": 75059.5744,
      "train_tokens_per_second": 61960.588
    },
    {
      "epoch": 0.8330841655537172,
      "grad_norm": 14.3125,
      "learning_rate": 1.426328909039606e-06,
      "loss": 0.5933,
      "num_input_tokens_seen": 4650971648,
      "step": 62690,
      "train_runtime": 75063.7321,
      "train_tokens_per_second": 61960.304
    },
    {
      "epoch": 0.8332170550361792,
      "grad_norm": 19.5,
      "learning_rate": 1.424114429580764e-06,
      "loss": 0.5263,
      "num_input_tokens_seen": 4651087360,
      "step": 62700,
      "train_runtime": 75067.1301,
      "train_tokens_per_second": 61959.041
    },
    {
      "epoch": 0.8333499445186411,
      "grad_norm": 14.75,
      "learning_rate": 1.4219015387282497e-06,
      "loss": 0.7937,
      "num_input_tokens_seen": 4653103872,
      "step": 62710,
      "train_runtime": 75099.0241,
      "train_tokens_per_second": 61959.578
    },
    {
      "epoch": 0.833482834001103,
      "grad_norm": 42.0,
      "learning_rate": 1.4196902368919895e-06,
      "loss": 0.8787,
      "num_input_tokens_seen": 4653923328,
      "step": 62720,
      "train_runtime": 75111.4418,
      "train_tokens_per_second": 61960.245
    },
    {
      "epoch": 0.8336157234835649,
      "grad_norm": 17.75,
      "learning_rate": 1.4174805244816027e-06,
      "loss": 0.6436,
      "num_input_tokens_seen": 4654363648,
      "step": 62730,
      "train_runtime": 75119.3185,
      "train_tokens_per_second": 61959.61
    },
    {
      "epoch": 0.8337486129660268,
      "grad_norm": 10.1875,
      "learning_rate": 1.415272401906418e-06,
      "loss": 0.6262,
      "num_input_tokens_seen": 4654583296,
      "step": 62740,
      "train_runtime": 75123.4632,
      "train_tokens_per_second": 61959.115
    },
    {
      "epoch": 0.8338815024484887,
      "grad_norm": 11.5625,
      "learning_rate": 1.413065869575474e-06,
      "loss": 0.6239,
      "num_input_tokens_seen": 4654698496,
      "step": 62750,
      "train_runtime": 75126.8355,
      "train_tokens_per_second": 61957.867
    },
    {
      "epoch": 0.8340143919309506,
      "grad_norm": 17.0,
      "learning_rate": 1.4108609278975082e-06,
      "loss": 0.8008,
      "num_input_tokens_seen": 4656699648,
      "step": 62760,
      "train_runtime": 75159.605,
      "train_tokens_per_second": 61957.479
    },
    {
      "epoch": 0.8341472814134125,
      "grad_norm": 17.0,
      "learning_rate": 1.4086575772809707e-06,
      "loss": 0.9399,
      "num_input_tokens_seen": 4657542656,
      "step": 62770,
      "train_runtime": 75172.5583,
      "train_tokens_per_second": 61958.017
    },
    {
      "epoch": 0.8342801708958745,
      "grad_norm": 23.75,
      "learning_rate": 1.406455818134006e-06,
      "loss": 0.7481,
      "num_input_tokens_seen": 4658016000,
      "step": 62780,
      "train_runtime": 75179.92,
      "train_tokens_per_second": 61958.246
    },
    {
      "epoch": 0.8344130603783364,
      "grad_norm": 10.6875,
      "learning_rate": 1.4042556508644745e-06,
      "loss": 0.6357,
      "num_input_tokens_seen": 4658242816,
      "step": 62790,
      "train_runtime": 75184.0401,
      "train_tokens_per_second": 61957.868
    },
    {
      "epoch": 0.8345459498607982,
      "grad_norm": 18.875,
      "learning_rate": 1.402057075879938e-06,
      "loss": 0.5267,
      "num_input_tokens_seen": 4658359040,
      "step": 62800,
      "train_runtime": 75188.3605,
      "train_tokens_per_second": 61955.853
    },
    {
      "epoch": 0.8346788393432601,
      "grad_norm": 18.0,
      "learning_rate": 1.3998600935876606e-06,
      "loss": 0.7921,
      "num_input_tokens_seen": 4660276992,
      "step": 62810,
      "train_runtime": 75219.0038,
      "train_tokens_per_second": 61956.112
    },
    {
      "epoch": 0.8348117288257221,
      "grad_norm": 24.25,
      "learning_rate": 1.3976647043946145e-06,
      "loss": 0.8643,
      "num_input_tokens_seen": 4661123072,
      "step": 62820,
      "train_runtime": 75231.8474,
      "train_tokens_per_second": 61956.781
    },
    {
      "epoch": 0.834944618308184,
      "grad_norm": 12.75,
      "learning_rate": 1.3954709087074791e-06,
      "loss": 0.7426,
      "num_input_tokens_seen": 4661596160,
      "step": 62830,
      "train_runtime": 75239.1071,
      "train_tokens_per_second": 61957.09
    },
    {
      "epoch": 0.8350775077906459,
      "grad_norm": 18.25,
      "learning_rate": 1.3932787069326291e-06,
      "loss": 0.644,
      "num_input_tokens_seen": 4661838336,
      "step": 62840,
      "train_runtime": 75243.131,
      "train_tokens_per_second": 61956.996
    },
    {
      "epoch": 0.8352103972731079,
      "grad_norm": 17.25,
      "learning_rate": 1.3910880994761556e-06,
      "loss": 0.5448,
      "num_input_tokens_seen": 4661955840,
      "step": 62850,
      "train_runtime": 75247.4484,
      "train_tokens_per_second": 61955.002
    },
    {
      "epoch": 0.8353432867555697,
      "grad_norm": 27.75,
      "learning_rate": 1.3888990867438468e-06,
      "loss": 0.8479,
      "num_input_tokens_seen": 4664142336,
      "step": 62860,
      "train_runtime": 75281.7703,
      "train_tokens_per_second": 61955.8
    },
    {
      "epoch": 0.8354761762380316,
      "grad_norm": 41.75,
      "learning_rate": 1.3867116691411986e-06,
      "loss": 0.9456,
      "num_input_tokens_seen": 4664962816,
      "step": 62870,
      "train_runtime": 75294.7995,
      "train_tokens_per_second": 61955.976
    },
    {
      "epoch": 0.8356090657204935,
      "grad_norm": 14.875,
      "learning_rate": 1.3845258470734124e-06,
      "loss": 0.7198,
      "num_input_tokens_seen": 4665416192,
      "step": 62880,
      "train_runtime": 75302.2796,
      "train_tokens_per_second": 61955.843
    },
    {
      "epoch": 0.8357419552029555,
      "grad_norm": 18.75,
      "learning_rate": 1.3823416209453867e-06,
      "loss": 0.6471,
      "num_input_tokens_seen": 4665645056,
      "step": 62890,
      "train_runtime": 75307.8502,
      "train_tokens_per_second": 61954.299
    },
    {
      "epoch": 0.8358748446854174,
      "grad_norm": 9.375,
      "learning_rate": 1.3801589911617364e-06,
      "loss": 0.5697,
      "num_input_tokens_seen": 4665758464,
      "step": 62900,
      "train_runtime": 75311.4253,
      "train_tokens_per_second": 61952.864
    },
    {
      "epoch": 0.8360077341678793,
      "grad_norm": 26.375,
      "learning_rate": 1.3779779581267683e-06,
      "loss": 0.7679,
      "num_input_tokens_seen": 4667862272,
      "step": 62910,
      "train_runtime": 75344.5925,
      "train_tokens_per_second": 61953.514
    },
    {
      "epoch": 0.8361406236503411,
      "grad_norm": 31.875,
      "learning_rate": 1.3757985222445058e-06,
      "loss": 0.8808,
      "num_input_tokens_seen": 4668669952,
      "step": 62920,
      "train_runtime": 75357.4114,
      "train_tokens_per_second": 61953.693
    },
    {
      "epoch": 0.8362735131328031,
      "grad_norm": 12.625,
      "learning_rate": 1.3736206839186638e-06,
      "loss": 0.7609,
      "num_input_tokens_seen": 4669142784,
      "step": 62930,
      "train_runtime": 75364.659,
      "train_tokens_per_second": 61954.009
    },
    {
      "epoch": 0.836406402615265,
      "grad_norm": 83.5,
      "learning_rate": 1.3714444435526709e-06,
      "loss": 0.6418,
      "num_input_tokens_seen": 4669379328,
      "step": 62940,
      "train_runtime": 75368.6107,
      "train_tokens_per_second": 61953.899
    },
    {
      "epoch": 0.8365392920977269,
      "grad_norm": 13.625,
      "learning_rate": 1.3692698015496575e-06,
      "loss": 0.5526,
      "num_input_tokens_seen": 4669497088,
      "step": 62950,
      "train_runtime": 75371.7678,
      "train_tokens_per_second": 61952.867
    },
    {
      "epoch": 0.8366721815801889,
      "grad_norm": 25.125,
      "learning_rate": 1.3670967583124507e-06,
      "loss": 0.8523,
      "num_input_tokens_seen": 4671552768,
      "step": 62960,
      "train_runtime": 75406.4314,
      "train_tokens_per_second": 61951.649
    },
    {
      "epoch": 0.8368050710626508,
      "grad_norm": 15.375,
      "learning_rate": 1.3649253142435948e-06,
      "loss": 0.8593,
      "num_input_tokens_seen": 4672384256,
      "step": 62970,
      "train_runtime": 75419.2729,
      "train_tokens_per_second": 61952.125
    },
    {
      "epoch": 0.8369379605451126,
      "grad_norm": 38.25,
      "learning_rate": 1.3627554697453249e-06,
      "loss": 0.6916,
      "num_input_tokens_seen": 4672822272,
      "step": 62980,
      "train_runtime": 75425.8277,
      "train_tokens_per_second": 61952.549
    },
    {
      "epoch": 0.8370708500275745,
      "grad_norm": 12.75,
      "learning_rate": 1.3605872252195885e-06,
      "loss": 0.7424,
      "num_input_tokens_seen": 4673046528,
      "step": 62990,
      "train_runtime": 75429.71,
      "train_tokens_per_second": 61952.333
    },
    {
      "epoch": 0.8372037395100365,
      "grad_norm": 17.625,
      "learning_rate": 1.3584205810680328e-06,
      "loss": 0.5975,
      "num_input_tokens_seen": 4673162752,
      "step": 63000,
      "train_runtime": 75433.0906,
      "train_tokens_per_second": 61951.098
    },
    {
      "epoch": 0.8373366289924984,
      "grad_norm": 20.375,
      "learning_rate": 1.356255537692005e-06,
      "loss": 0.9084,
      "num_input_tokens_seen": 4675069696,
      "step": 63010,
      "train_runtime": 75466.0184,
      "train_tokens_per_second": 61949.336
    },
    {
      "epoch": 0.8374695184749603,
      "grad_norm": 18.75,
      "learning_rate": 1.3540920954925697e-06,
      "loss": 0.9138,
      "num_input_tokens_seen": 4675929600,
      "step": 63020,
      "train_runtime": 75478.846,
      "train_tokens_per_second": 61950.2
    },
    {
      "epoch": 0.8376024079574222,
      "grad_norm": 48.0,
      "learning_rate": 1.351930254870476e-06,
      "loss": 0.8224,
      "num_input_tokens_seen": 4676419072,
      "step": 63030,
      "train_runtime": 75486.0786,
      "train_tokens_per_second": 61950.748
    },
    {
      "epoch": 0.8377352974398842,
      "grad_norm": 14.8125,
      "learning_rate": 1.3497700162261896e-06,
      "loss": 0.6968,
      "num_input_tokens_seen": 4676650496,
      "step": 63040,
      "train_runtime": 75490.4793,
      "train_tokens_per_second": 61950.203
    },
    {
      "epoch": 0.837868186922346,
      "grad_norm": 13.4375,
      "learning_rate": 1.3476113799598744e-06,
      "loss": 0.5039,
      "num_input_tokens_seen": 4676768256,
      "step": 63050,
      "train_runtime": 75493.883,
      "train_tokens_per_second": 61948.969
    },
    {
      "epoch": 0.8380010764048079,
      "grad_norm": 17.75,
      "learning_rate": 1.3454543464713997e-06,
      "loss": 0.8274,
      "num_input_tokens_seen": 4678822912,
      "step": 63060,
      "train_runtime": 75528.3625,
      "train_tokens_per_second": 61947.893
    },
    {
      "epoch": 0.8381339658872698,
      "grad_norm": 15.9375,
      "learning_rate": 1.343298916160336e-06,
      "loss": 0.959,
      "num_input_tokens_seen": 4679667968,
      "step": 63070,
      "train_runtime": 75540.9358,
      "train_tokens_per_second": 61948.769
    },
    {
      "epoch": 0.8382668553697318,
      "grad_norm": 31.625,
      "learning_rate": 1.3411450894259582e-06,
      "loss": 0.7345,
      "num_input_tokens_seen": 4680105984,
      "step": 63080,
      "train_runtime": 75547.8537,
      "train_tokens_per_second": 61948.894
    },
    {
      "epoch": 0.8383997448521937,
      "grad_norm": 16.875,
      "learning_rate": 1.3389928666672447e-06,
      "loss": 0.6375,
      "num_input_tokens_seen": 4680327168,
      "step": 63090,
      "train_runtime": 75551.7196,
      "train_tokens_per_second": 61948.652
    },
    {
      "epoch": 0.8385326343346556,
      "grad_norm": 16.125,
      "learning_rate": 1.3368422482828724e-06,
      "loss": 0.5738,
      "num_input_tokens_seen": 4680442880,
      "step": 63100,
      "train_runtime": 75555.4289,
      "train_tokens_per_second": 61947.142
    },
    {
      "epoch": 0.8386655238171175,
      "grad_norm": 22.5,
      "learning_rate": 1.3346932346712248e-06,
      "loss": 0.8759,
      "num_input_tokens_seen": 4682575360,
      "step": 63110,
      "train_runtime": 75589.3726,
      "train_tokens_per_second": 61947.536
    },
    {
      "epoch": 0.8387984132995794,
      "grad_norm": 24.5,
      "learning_rate": 1.332545826230388e-06,
      "loss": 0.9124,
      "num_input_tokens_seen": 4683425536,
      "step": 63120,
      "train_runtime": 75603.7136,
      "train_tokens_per_second": 61947.03
    },
    {
      "epoch": 0.8389313027820413,
      "grad_norm": 23.25,
      "learning_rate": 1.3304000233581516e-06,
      "loss": 0.73,
      "num_input_tokens_seen": 4683896320,
      "step": 63130,
      "train_runtime": 75610.6116,
      "train_tokens_per_second": 61947.605
    },
    {
      "epoch": 0.8390641922645032,
      "grad_norm": 24.5,
      "learning_rate": 1.328255826452005e-06,
      "loss": 0.6291,
      "num_input_tokens_seen": 4684122624,
      "step": 63140,
      "train_runtime": 75614.5839,
      "train_tokens_per_second": 61947.344
    },
    {
      "epoch": 0.8391970817469652,
      "grad_norm": 39.0,
      "learning_rate": 1.3261132359091443e-06,
      "loss": 0.582,
      "num_input_tokens_seen": 4684237056,
      "step": 63150,
      "train_runtime": 75617.9453,
      "train_tokens_per_second": 61946.103
    },
    {
      "epoch": 0.8393299712294271,
      "grad_norm": 12.75,
      "learning_rate": 1.3239722521264608e-06,
      "loss": 0.8478,
      "num_input_tokens_seen": 4686322688,
      "step": 63160,
      "train_runtime": 75653.421,
      "train_tokens_per_second": 61944.624
    },
    {
      "epoch": 0.8394628607118889,
      "grad_norm": 49.25,
      "learning_rate": 1.3218328755005549e-06,
      "loss": 0.8309,
      "num_input_tokens_seen": 4687165952,
      "step": 63170,
      "train_runtime": 75666.2433,
      "train_tokens_per_second": 61945.271
    },
    {
      "epoch": 0.8395957501943508,
      "grad_norm": 23.375,
      "learning_rate": 1.3196951064277274e-06,
      "loss": 0.7193,
      "num_input_tokens_seen": 4687638272,
      "step": 63180,
      "train_runtime": 75673.2028,
      "train_tokens_per_second": 61945.816
    },
    {
      "epoch": 0.8397286396768128,
      "grad_norm": 19.0,
      "learning_rate": 1.3175589453039806e-06,
      "loss": 0.5793,
      "num_input_tokens_seen": 4687867904,
      "step": 63190,
      "train_runtime": 75677.568,
      "train_tokens_per_second": 61945.277
    },
    {
      "epoch": 0.8398615291592747,
      "grad_norm": 8.125,
      "learning_rate": 1.315424392525021e-06,
      "loss": 0.539,
      "num_input_tokens_seen": 4687982592,
      "step": 63200,
      "train_runtime": 75681.045,
      "train_tokens_per_second": 61943.946
    },
    {
      "epoch": 0.8399944186417366,
      "grad_norm": 29.875,
      "learning_rate": 1.313291448486249e-06,
      "loss": 0.8477,
      "num_input_tokens_seen": 4690054144,
      "step": 63210,
      "train_runtime": 75714.099,
      "train_tokens_per_second": 61944.264
    },
    {
      "epoch": 0.8401273081241986,
      "grad_norm": 18.5,
      "learning_rate": 1.3111601135827833e-06,
      "loss": 0.8334,
      "num_input_tokens_seen": 4690924800,
      "step": 63220,
      "train_runtime": 75727.2264,
      "train_tokens_per_second": 61945.023
    },
    {
      "epoch": 0.8402601976066604,
      "grad_norm": 15.25,
      "learning_rate": 1.3090303882094257e-06,
      "loss": 0.6977,
      "num_input_tokens_seen": 4691414784,
      "step": 63230,
      "train_runtime": 75736.5011,
      "train_tokens_per_second": 61943.907
    },
    {
      "epoch": 0.8403930870891223,
      "grad_norm": 11.5,
      "learning_rate": 1.306902272760696e-06,
      "loss": 0.6105,
      "num_input_tokens_seen": 4691645184,
      "step": 63240,
      "train_runtime": 75740.5895,
      "train_tokens_per_second": 61943.605
    },
    {
      "epoch": 0.8405259765715842,
      "grad_norm": 13.4375,
      "learning_rate": 1.3047757676308037e-06,
      "loss": 0.4802,
      "num_input_tokens_seen": 4691761408,
      "step": 63250,
      "train_runtime": 75743.9796,
      "train_tokens_per_second": 61942.367
    },
    {
      "epoch": 0.8406588660540462,
      "grad_norm": 20.5,
      "learning_rate": 1.3026508732136666e-06,
      "loss": 0.7336,
      "num_input_tokens_seen": 4693798656,
      "step": 63260,
      "train_runtime": 75778.1191,
      "train_tokens_per_second": 61941.346
    },
    {
      "epoch": 0.8407917555365081,
      "grad_norm": 32.75,
      "learning_rate": 1.3005275899029046e-06,
      "loss": 0.9184,
      "num_input_tokens_seen": 4694653952,
      "step": 63270,
      "train_runtime": 75790.9981,
      "train_tokens_per_second": 61942.105
    },
    {
      "epoch": 0.84092464501897,
      "grad_norm": 49.75,
      "learning_rate": 1.2984059180918307e-06,
      "loss": 0.8146,
      "num_input_tokens_seen": 4695128064,
      "step": 63280,
      "train_runtime": 75799.5244,
      "train_tokens_per_second": 61941.392
    },
    {
      "epoch": 0.8410575345014318,
      "grad_norm": 34.0,
      "learning_rate": 1.2962858581734738e-06,
      "loss": 0.6136,
      "num_input_tokens_seen": 4695361024,
      "step": 63290,
      "train_runtime": 75803.452,
      "train_tokens_per_second": 61941.256
    },
    {
      "epoch": 0.8411904239838938,
      "grad_norm": 18.0,
      "learning_rate": 1.29416741054055e-06,
      "loss": 0.5302,
      "num_input_tokens_seen": 4695477760,
      "step": 63300,
      "train_runtime": 75806.4112,
      "train_tokens_per_second": 61940.378
    },
    {
      "epoch": 0.8413233134663557,
      "grad_norm": 286.0,
      "learning_rate": 1.292050575585484e-06,
      "loss": 0.8226,
      "num_input_tokens_seen": 4697529600,
      "step": 63310,
      "train_runtime": 75839.9849,
      "train_tokens_per_second": 61940.012
    },
    {
      "epoch": 0.8414562029488176,
      "grad_norm": 27.25,
      "learning_rate": 1.2899353537004045e-06,
      "loss": 0.9313,
      "num_input_tokens_seen": 4698386688,
      "step": 63320,
      "train_runtime": 75854.7875,
      "train_tokens_per_second": 61939.224
    },
    {
      "epoch": 0.8415890924312796,
      "grad_norm": 27.125,
      "learning_rate": 1.2878217452771303e-06,
      "loss": 0.7799,
      "num_input_tokens_seen": 4698867968,
      "step": 63330,
      "train_runtime": 75861.8465,
      "train_tokens_per_second": 61939.805
    },
    {
      "epoch": 0.8417219819137415,
      "grad_norm": 54.25,
      "learning_rate": 1.285709750707197e-06,
      "loss": 0.6318,
      "num_input_tokens_seen": 4699095296,
      "step": 63340,
      "train_runtime": 75865.5758,
      "train_tokens_per_second": 61939.757
    },
    {
      "epoch": 0.8418548713962033,
      "grad_norm": 8.875,
      "learning_rate": 1.283599370381826e-06,
      "loss": 0.5443,
      "num_input_tokens_seen": 4699209728,
      "step": 63350,
      "train_runtime": 75868.6968,
      "train_tokens_per_second": 61938.717
    },
    {
      "epoch": 0.8419877608786652,
      "grad_norm": 32.5,
      "learning_rate": 1.2814906046919496e-06,
      "loss": 0.8725,
      "num_input_tokens_seen": 4701331456,
      "step": 63360,
      "train_runtime": 75902.9374,
      "train_tokens_per_second": 61938.729
    },
    {
      "epoch": 0.8421206503611272,
      "grad_norm": 18.75,
      "learning_rate": 1.2793834540281968e-06,
      "loss": 0.8954,
      "num_input_tokens_seen": 4702125312,
      "step": 63370,
      "train_runtime": 75915.363,
      "train_tokens_per_second": 61939.048
    },
    {
      "epoch": 0.8422535398435891,
      "grad_norm": 21.0,
      "learning_rate": 1.2772779187808993e-06,
      "loss": 0.6786,
      "num_input_tokens_seen": 4702565888,
      "step": 63380,
      "train_runtime": 75921.944,
      "train_tokens_per_second": 61939.482
    },
    {
      "epoch": 0.842386429326051,
      "grad_norm": 21.375,
      "learning_rate": 1.2751739993400868e-06,
      "loss": 0.6128,
      "num_input_tokens_seen": 4702797312,
      "step": 63390,
      "train_runtime": 75926.89,
      "train_tokens_per_second": 61938.495
    },
    {
      "epoch": 0.8425193188085129,
      "grad_norm": 38.25,
      "learning_rate": 1.273071696095497e-06,
      "loss": 0.5792,
      "num_input_tokens_seen": 4702910976,
      "step": 63400,
      "train_runtime": 75929.919,
      "train_tokens_per_second": 61937.521
    },
    {
      "epoch": 0.8426522082909748,
      "grad_norm": 32.0,
      "learning_rate": 1.2709710094365557e-06,
      "loss": 0.7501,
      "num_input_tokens_seen": 4704928768,
      "step": 63410,
      "train_runtime": 75961.9384,
      "train_tokens_per_second": 61937.977
    },
    {
      "epoch": 0.8427850977734367,
      "grad_norm": 18.75,
      "learning_rate": 1.2688719397523997e-06,
      "loss": 0.8675,
      "num_input_tokens_seen": 4705798400,
      "step": 63420,
      "train_runtime": 75974.8273,
      "train_tokens_per_second": 61938.915
    },
    {
      "epoch": 0.8429179872558986,
      "grad_norm": 10.4375,
      "learning_rate": 1.2667744874318633e-06,
      "loss": 0.7002,
      "num_input_tokens_seen": 4706288128,
      "step": 63430,
      "train_runtime": 75981.7421,
      "train_tokens_per_second": 61939.724
    },
    {
      "epoch": 0.8430508767383605,
      "grad_norm": 22.625,
      "learning_rate": 1.2646786528634792e-06,
      "loss": 0.6848,
      "num_input_tokens_seen": 4706522880,
      "step": 63440,
      "train_runtime": 75986.8342,
      "train_tokens_per_second": 61938.662
    },
    {
      "epoch": 0.8431837662208225,
      "grad_norm": 24.875,
      "learning_rate": 1.2625844364354845e-06,
      "loss": 0.5402,
      "num_input_tokens_seen": 4706637824,
      "step": 63450,
      "train_runtime": 75990.0709,
      "train_tokens_per_second": 61937.537
    },
    {
      "epoch": 0.8433166557032844,
      "grad_norm": 15.5625,
      "learning_rate": 1.2604918385358122e-06,
      "loss": 0.8687,
      "num_input_tokens_seen": 4708688640,
      "step": 63460,
      "train_runtime": 76023.7446,
      "train_tokens_per_second": 61937.079
    },
    {
      "epoch": 0.8434495451857462,
      "grad_norm": 24.5,
      "learning_rate": 1.2584008595521002e-06,
      "loss": 0.9995,
      "num_input_tokens_seen": 4709549056,
      "step": 63470,
      "train_runtime": 76037.0574,
      "train_tokens_per_second": 61937.55
    },
    {
      "epoch": 0.8435824346682081,
      "grad_norm": 20.125,
      "learning_rate": 1.2563114998716797e-06,
      "loss": 0.7998,
      "num_input_tokens_seen": 4710050304,
      "step": 63480,
      "train_runtime": 76045.7454,
      "train_tokens_per_second": 61937.065
    },
    {
      "epoch": 0.8437153241506701,
      "grad_norm": 21.5,
      "learning_rate": 1.2542237598815875e-06,
      "loss": 0.6819,
      "num_input_tokens_seen": 4710294272,
      "step": 63490,
      "train_runtime": 76049.7851,
      "train_tokens_per_second": 61936.983
    },
    {
      "epoch": 0.843848213633132,
      "grad_norm": 19.75,
      "learning_rate": 1.2521376399685603e-06,
      "loss": 0.5568,
      "num_input_tokens_seen": 4710410496,
      "step": 63500,
      "train_runtime": 76053.4891,
      "train_tokens_per_second": 61935.495
    },
    {
      "epoch": 0.8439811031155939,
      "grad_norm": 27.0,
      "learning_rate": 1.2500531405190308e-06,
      "loss": 0.8773,
      "num_input_tokens_seen": 4712565248,
      "step": 63510,
      "train_runtime": 76088.5259,
      "train_tokens_per_second": 61935.294
    },
    {
      "epoch": 0.8441139925980559,
      "grad_norm": 56.25,
      "learning_rate": 1.247970261919138e-06,
      "loss": 0.9491,
      "num_input_tokens_seen": 4713463808,
      "step": 63520,
      "train_runtime": 76101.9927,
      "train_tokens_per_second": 61936.142
    },
    {
      "epoch": 0.8442468820805178,
      "grad_norm": 25.5,
      "learning_rate": 1.2458890045547089e-06,
      "loss": 0.798,
      "num_input_tokens_seen": 4713939968,
      "step": 63530,
      "train_runtime": 76109.0799,
      "train_tokens_per_second": 61936.631
    },
    {
      "epoch": 0.8443797715629796,
      "grad_norm": 15.5625,
      "learning_rate": 1.2438093688112873e-06,
      "loss": 0.6855,
      "num_input_tokens_seen": 4714174976,
      "step": 63540,
      "train_runtime": 76113.5678,
      "train_tokens_per_second": 61936.066
    },
    {
      "epoch": 0.8445126610454415,
      "grad_norm": 19.5,
      "learning_rate": 1.2417313550741006e-06,
      "loss": 0.5501,
      "num_input_tokens_seen": 4714292736,
      "step": 63550,
      "train_runtime": 76116.7941,
      "train_tokens_per_second": 61934.988
    },
    {
      "epoch": 0.8446455505279035,
      "grad_norm": 13.5,
      "learning_rate": 1.239654963728084e-06,
      "loss": 0.7857,
      "num_input_tokens_seen": 4716363520,
      "step": 63560,
      "train_runtime": 76152.156,
      "train_tokens_per_second": 61933.421
    },
    {
      "epoch": 0.8447784400103654,
      "grad_norm": 18.875,
      "learning_rate": 1.2375801951578737e-06,
      "loss": 0.9405,
      "num_input_tokens_seen": 4717194240,
      "step": 63570,
      "train_runtime": 76165.087,
      "train_tokens_per_second": 61933.813
    },
    {
      "epoch": 0.8449113294928273,
      "grad_norm": 60.25,
      "learning_rate": 1.2355070497477938e-06,
      "loss": 0.7202,
      "num_input_tokens_seen": 4717652224,
      "step": 63580,
      "train_runtime": 76172.1186,
      "train_tokens_per_second": 61934.108
    },
    {
      "epoch": 0.8450442189752893,
      "grad_norm": 14.125,
      "learning_rate": 1.2334355278818866e-06,
      "loss": 0.6484,
      "num_input_tokens_seen": 4717873920,
      "step": 63590,
      "train_runtime": 76175.8997,
      "train_tokens_per_second": 61933.944
    },
    {
      "epoch": 0.8451771084577511,
      "grad_norm": 24.25,
      "learning_rate": 1.2313656299438724e-06,
      "loss": 0.5128,
      "num_input_tokens_seen": 4717987584,
      "step": 63600,
      "train_runtime": 76179.1101,
      "train_tokens_per_second": 61932.826
    },
    {
      "epoch": 0.845309997940213,
      "grad_norm": 27.625,
      "learning_rate": 1.2292973563171928e-06,
      "loss": 0.8259,
      "num_input_tokens_seen": 4719959552,
      "step": 63610,
      "train_runtime": 76211.6573,
      "train_tokens_per_second": 61932.252
    },
    {
      "epoch": 0.8454428874226749,
      "grad_norm": 34.25,
      "learning_rate": 1.2272307073849676e-06,
      "loss": 0.9042,
      "num_input_tokens_seen": 4720812032,
      "step": 63620,
      "train_runtime": 76225.7496,
      "train_tokens_per_second": 61931.986
    },
    {
      "epoch": 0.8455757769051369,
      "grad_norm": 16.375,
      "learning_rate": 1.2251656835300286e-06,
      "loss": 0.7744,
      "num_input_tokens_seen": 4721268992,
      "step": 63630,
      "train_runtime": 76232.7198,
      "train_tokens_per_second": 61932.317
    },
    {
      "epoch": 0.8457086663875988,
      "grad_norm": 28.125,
      "learning_rate": 1.223102285134905e-06,
      "loss": 0.5405,
      "num_input_tokens_seen": 4721503488,
      "step": 63640,
      "train_runtime": 76236.6515,
      "train_tokens_per_second": 61932.199
    },
    {
      "epoch": 0.8458415558700607,
      "grad_norm": 7.78125,
      "learning_rate": 1.221040512581817e-06,
      "loss": 0.5729,
      "num_input_tokens_seen": 4721620224,
      "step": 63650,
      "train_runtime": 76240.2901,
      "train_tokens_per_second": 61930.775
    },
    {
      "epoch": 0.8459744453525225,
      "grad_norm": 153.0,
      "learning_rate": 1.2189803662526977e-06,
      "loss": 0.8297,
      "num_input_tokens_seen": 4723713792,
      "step": 63660,
      "train_runtime": 76273.4456,
      "train_tokens_per_second": 61931.302
    },
    {
      "epoch": 0.8461073348349845,
      "grad_norm": 30.375,
      "learning_rate": 1.2169218465291654e-06,
      "loss": 0.8876,
      "num_input_tokens_seen": 4724599040,
      "step": 63670,
      "train_runtime": 76286.669,
      "train_tokens_per_second": 61932.171
    },
    {
      "epoch": 0.8462402243174464,
      "grad_norm": 14.0,
      "learning_rate": 1.2148649537925427e-06,
      "loss": 0.7937,
      "num_input_tokens_seen": 4725105920,
      "step": 63680,
      "train_runtime": 76295.8058,
      "train_tokens_per_second": 61931.398
    },
    {
      "epoch": 0.8463731137999083,
      "grad_norm": 15.1875,
      "learning_rate": 1.2128096884238515e-06,
      "loss": 0.5829,
      "num_input_tokens_seen": 4725340416,
      "step": 63690,
      "train_runtime": 76300.1305,
      "train_tokens_per_second": 61930.961
    },
    {
      "epoch": 0.8465060032823702,
      "grad_norm": 10.25,
      "learning_rate": 1.210756050803813e-06,
      "loss": 0.5371,
      "num_input_tokens_seen": 4725455360,
      "step": 63700,
      "train_runtime": 76303.3962,
      "train_tokens_per_second": 61929.817
    },
    {
      "epoch": 0.8466388927648322,
      "grad_norm": 42.25,
      "learning_rate": 1.2087040413128426e-06,
      "loss": 0.7239,
      "num_input_tokens_seen": 4727606528,
      "step": 63710,
      "train_runtime": 76337.7111,
      "train_tokens_per_second": 61930.158
    },
    {
      "epoch": 0.846771782247294,
      "grad_norm": 15.1875,
      "learning_rate": 1.2066536603310608e-06,
      "loss": 0.8941,
      "num_input_tokens_seen": 4728507648,
      "step": 63720,
      "train_runtime": 76351.251,
      "train_tokens_per_second": 61930.978
    },
    {
      "epoch": 0.8469046717297559,
      "grad_norm": 19.5,
      "learning_rate": 1.2046049082382772e-06,
      "loss": 0.7196,
      "num_input_tokens_seen": 4729001216,
      "step": 63730,
      "train_runtime": 76358.4429,
      "train_tokens_per_second": 61931.609
    },
    {
      "epoch": 0.8470375612122178,
      "grad_norm": 20.5,
      "learning_rate": 1.2025577854140068e-06,
      "loss": 0.7062,
      "num_input_tokens_seen": 4729244416,
      "step": 63740,
      "train_runtime": 76363.9073,
      "train_tokens_per_second": 61930.362
    },
    {
      "epoch": 0.8471704506946798,
      "grad_norm": 12.125,
      "learning_rate": 1.2005122922374613e-06,
      "loss": 0.5596,
      "num_input_tokens_seen": 4729361408,
      "step": 63750,
      "train_runtime": 76367.4464,
      "train_tokens_per_second": 61929.024
    },
    {
      "epoch": 0.8473033401771417,
      "grad_norm": 12.5,
      "learning_rate": 1.1984684290875504e-06,
      "loss": 0.8126,
      "num_input_tokens_seen": 4731489792,
      "step": 63760,
      "train_runtime": 76402.4022,
      "train_tokens_per_second": 61928.547
    },
    {
      "epoch": 0.8474362296596036,
      "grad_norm": 20.625,
      "learning_rate": 1.1964261963428813e-06,
      "loss": 0.91,
      "num_input_tokens_seen": 4732327936,
      "step": 63770,
      "train_runtime": 76415.4297,
      "train_tokens_per_second": 61928.958
    },
    {
      "epoch": 0.8475691191420655,
      "grad_norm": 31.0,
      "learning_rate": 1.1943855943817594e-06,
      "loss": 0.7373,
      "num_input_tokens_seen": 4732820992,
      "step": 63780,
      "train_runtime": 76422.4664,
      "train_tokens_per_second": 61929.708
    },
    {
      "epoch": 0.8477020086245274,
      "grad_norm": 17.0,
      "learning_rate": 1.1923466235821901e-06,
      "loss": 0.6787,
      "num_input_tokens_seen": 4733061120,
      "step": 63790,
      "train_runtime": 76427.5025,
      "train_tokens_per_second": 61928.769
    },
    {
      "epoch": 0.8478348981069893,
      "grad_norm": 13.125,
      "learning_rate": 1.19030928432187e-06,
      "loss": 0.6202,
      "num_input_tokens_seen": 4733180416,
      "step": 63800,
      "train_runtime": 76430.6492,
      "train_tokens_per_second": 61927.78
    },
    {
      "epoch": 0.8479677875894512,
      "grad_norm": 28.625,
      "learning_rate": 1.1882735769782006e-06,
      "loss": 0.825,
      "num_input_tokens_seen": 4735273216,
      "step": 63810,
      "train_runtime": 76463.1928,
      "train_tokens_per_second": 61928.793
    },
    {
      "epoch": 0.8481006770719132,
      "grad_norm": 11.125,
      "learning_rate": 1.186239501928278e-06,
      "loss": 0.8202,
      "num_input_tokens_seen": 4736112896,
      "step": 63820,
      "train_runtime": 76475.9644,
      "train_tokens_per_second": 61929.43
    },
    {
      "epoch": 0.8482335665543751,
      "grad_norm": 25.75,
      "learning_rate": 1.1842070595488974e-06,
      "loss": 0.6936,
      "num_input_tokens_seen": 4736566272,
      "step": 63830,
      "train_runtime": 76483.8563,
      "train_tokens_per_second": 61928.968
    },
    {
      "epoch": 0.8483664560368369,
      "grad_norm": 27.0,
      "learning_rate": 1.1821762502165513e-06,
      "loss": 0.6345,
      "num_input_tokens_seen": 4736795392,
      "step": 63840,
      "train_runtime": 76487.4954,
      "train_tokens_per_second": 61929.017
    },
    {
      "epoch": 0.8484993455192988,
      "grad_norm": 17.75,
      "learning_rate": 1.1801470743074238e-06,
      "loss": 0.5398,
      "num_input_tokens_seen": 4736907520,
      "step": 63850,
      "train_runtime": 76490.5706,
      "train_tokens_per_second": 61927.993
    },
    {
      "epoch": 0.8486322350017608,
      "grad_norm": 13.9375,
      "learning_rate": 1.1781195321974105e-06,
      "loss": 0.889,
      "num_input_tokens_seen": 4738985984,
      "step": 63860,
      "train_runtime": 76525.1124,
      "train_tokens_per_second": 61927.201
    },
    {
      "epoch": 0.8487651244842227,
      "grad_norm": 63.25,
      "learning_rate": 1.1760936242620879e-06,
      "loss": 0.9276,
      "num_input_tokens_seen": 4739836672,
      "step": 63870,
      "train_runtime": 76537.9839,
      "train_tokens_per_second": 61927.901
    },
    {
      "epoch": 0.8488980139666846,
      "grad_norm": 14.875,
      "learning_rate": 1.1740693508767408e-06,
      "loss": 0.7351,
      "num_input_tokens_seen": 4740301824,
      "step": 63880,
      "train_runtime": 76544.833,
      "train_tokens_per_second": 61928.436
    },
    {
      "epoch": 0.8490309034491466,
      "grad_norm": 16.125,
      "learning_rate": 1.1720467124163482e-06,
      "loss": 0.6192,
      "num_input_tokens_seen": 4740533248,
      "step": 63890,
      "train_runtime": 76548.7261,
      "train_tokens_per_second": 61928.31
    },
    {
      "epoch": 0.8491637929316084,
      "grad_norm": 17.5,
      "learning_rate": 1.1700257092555812e-06,
      "loss": 0.5395,
      "num_input_tokens_seen": 4740649472,
      "step": 63900,
      "train_runtime": 76551.8504,
      "train_tokens_per_second": 61927.301
    },
    {
      "epoch": 0.8492966824140703,
      "grad_norm": 18.5,
      "learning_rate": 1.1680063417688204e-06,
      "loss": 0.7591,
      "num_input_tokens_seen": 4742762752,
      "step": 63910,
      "train_runtime": 76585.5692,
      "train_tokens_per_second": 61927.629
    },
    {
      "epoch": 0.8494295718965322,
      "grad_norm": 21.375,
      "learning_rate": 1.1659886103301289e-06,
      "loss": 0.9557,
      "num_input_tokens_seen": 4743642880,
      "step": 63920,
      "train_runtime": 76598.7985,
      "train_tokens_per_second": 61928.424
    },
    {
      "epoch": 0.8495624613789942,
      "grad_norm": 29.875,
      "learning_rate": 1.1639725153132763e-06,
      "loss": 0.7308,
      "num_input_tokens_seen": 4744122368,
      "step": 63930,
      "train_runtime": 76605.6503,
      "train_tokens_per_second": 61929.144
    },
    {
      "epoch": 0.8496953508614561,
      "grad_norm": 11.375,
      "learning_rate": 1.1619580570917255e-06,
      "loss": 0.6017,
      "num_input_tokens_seen": 4744360192,
      "step": 63940,
      "train_runtime": 76609.5511,
      "train_tokens_per_second": 61929.095
    },
    {
      "epoch": 0.849828240343918,
      "grad_norm": 13.0625,
      "learning_rate": 1.1599452360386364e-06,
      "loss": 0.5244,
      "num_input_tokens_seen": 4744478976,
      "step": 63950,
      "train_runtime": 76612.876,
      "train_tokens_per_second": 61927.958
    },
    {
      "epoch": 0.84996112982638,
      "grad_norm": 15.9375,
      "learning_rate": 1.1579340525268679e-06,
      "loss": 0.8138,
      "num_input_tokens_seen": 4746426624,
      "step": 63960,
      "train_runtime": 76645.6672,
      "train_tokens_per_second": 61926.875
    },
    {
      "epoch": 0.8500940193088418,
      "grad_norm": 24.0,
      "learning_rate": 1.1559245069289738e-06,
      "loss": 0.916,
      "num_input_tokens_seen": 4747276288,
      "step": 63970,
      "train_runtime": 76658.9304,
      "train_tokens_per_second": 61927.244
    },
    {
      "epoch": 0.8502269087913037,
      "grad_norm": 20.125,
      "learning_rate": 1.1539165996172042e-06,
      "loss": 0.7876,
      "num_input_tokens_seen": 4747759616,
      "step": 63980,
      "train_runtime": 76665.8534,
      "train_tokens_per_second": 61927.956
    },
    {
      "epoch": 0.8503597982737656,
      "grad_norm": 37.5,
      "learning_rate": 1.1519103309635037e-06,
      "loss": 0.5976,
      "num_input_tokens_seen": 4747986432,
      "step": 63990,
      "train_runtime": 76669.4977,
      "train_tokens_per_second": 61927.971
    },
    {
      "epoch": 0.8504926877562276,
      "grad_norm": 18.375,
      "learning_rate": 1.1499057013395177e-06,
      "loss": 0.5452,
      "num_input_tokens_seen": 4748097792,
      "step": 64000,
      "train_runtime": 76674.1182,
      "train_tokens_per_second": 61925.692
    },
    {
      "epoch": 0.8506255772386895,
      "grad_norm": 36.25,
      "learning_rate": 1.1479027111165863e-06,
      "loss": 0.7351,
      "num_input_tokens_seen": 4750207744,
      "step": 64010,
      "train_runtime": 76707.4592,
      "train_tokens_per_second": 61926.282
    },
    {
      "epoch": 0.8507584667211514,
      "grad_norm": 16.5,
      "learning_rate": 1.1459013606657444e-06,
      "loss": 0.8995,
      "num_input_tokens_seen": 4751059712,
      "step": 64020,
      "train_runtime": 76720.389,
      "train_tokens_per_second": 61926.95
    },
    {
      "epoch": 0.8508913562036132,
      "grad_norm": 12.3125,
      "learning_rate": 1.143901650357725e-06,
      "loss": 0.7262,
      "num_input_tokens_seen": 4751508480,
      "step": 64030,
      "train_runtime": 76727.1942,
      "train_tokens_per_second": 61927.307
    },
    {
      "epoch": 0.8510242456860752,
      "grad_norm": 33.75,
      "learning_rate": 1.1419035805629575e-06,
      "loss": 0.6205,
      "num_input_tokens_seen": 4751741184,
      "step": 64040,
      "train_runtime": 76731.192,
      "train_tokens_per_second": 61927.113
    },
    {
      "epoch": 0.8511571351685371,
      "grad_norm": 9.0625,
      "learning_rate": 1.1399071516515647e-06,
      "loss": 0.487,
      "num_input_tokens_seen": 4751858176,
      "step": 64050,
      "train_runtime": 76734.1472,
      "train_tokens_per_second": 61926.253
    },
    {
      "epoch": 0.851290024650999,
      "grad_norm": 31.0,
      "learning_rate": 1.1379123639933676e-06,
      "loss": 0.8629,
      "num_input_tokens_seen": 4753871616,
      "step": 64060,
      "train_runtime": 76767.5871,
      "train_tokens_per_second": 61925.505
    },
    {
      "epoch": 0.8514229141334609,
      "grad_norm": 18.125,
      "learning_rate": 1.1359192179578837e-06,
      "loss": 0.8801,
      "num_input_tokens_seen": 4754695936,
      "step": 64070,
      "train_runtime": 76781.3106,
      "train_tokens_per_second": 61925.173
    },
    {
      "epoch": 0.8515558036159229,
      "grad_norm": 12.8125,
      "learning_rate": 1.1339277139143245e-06,
      "loss": 0.6913,
      "num_input_tokens_seen": 4755156992,
      "step": 64080,
      "train_runtime": 76788.2884,
      "train_tokens_per_second": 61925.55
    },
    {
      "epoch": 0.8516886930983847,
      "grad_norm": 45.5,
      "learning_rate": 1.1319378522316017e-06,
      "loss": 0.6862,
      "num_input_tokens_seen": 4755382016,
      "step": 64090,
      "train_runtime": 76792.0887,
      "train_tokens_per_second": 61925.416
    },
    {
      "epoch": 0.8518215825808466,
      "grad_norm": 9.0,
      "learning_rate": 1.1299496332783123e-06,
      "loss": 0.5005,
      "num_input_tokens_seen": 4755494400,
      "step": 64100,
      "train_runtime": 76795.3776,
      "train_tokens_per_second": 61924.227
    },
    {
      "epoch": 0.8519544720633085,
      "grad_norm": 27.875,
      "learning_rate": 1.1279630574227651e-06,
      "loss": 0.8603,
      "num_input_tokens_seen": 4757511424,
      "step": 64110,
      "train_runtime": 76828.0314,
      "train_tokens_per_second": 61924.161
    },
    {
      "epoch": 0.8520873615457705,
      "grad_norm": 21.625,
      "learning_rate": 1.125978125032946e-06,
      "loss": 0.9017,
      "num_input_tokens_seen": 4758347264,
      "step": 64120,
      "train_runtime": 76842.24,
      "train_tokens_per_second": 61923.589
    },
    {
      "epoch": 0.8522202510282324,
      "grad_norm": 12.875,
      "learning_rate": 1.1239948364765563e-06,
      "loss": 0.6975,
      "num_input_tokens_seen": 4758798336,
      "step": 64130,
      "train_runtime": 76849.0354,
      "train_tokens_per_second": 61923.983
    },
    {
      "epoch": 0.8523531405106943,
      "grad_norm": 19.125,
      "learning_rate": 1.1220131921209743e-06,
      "loss": 0.5893,
      "num_input_tokens_seen": 4759025920,
      "step": 64140,
      "train_runtime": 76853.3431,
      "train_tokens_per_second": 61923.473
    },
    {
      "epoch": 0.8524860299931561,
      "grad_norm": 18.375,
      "learning_rate": 1.1200331923332852e-06,
      "loss": 0.562,
      "num_input_tokens_seen": 4759144448,
      "step": 64150,
      "train_runtime": 76856.534,
      "train_tokens_per_second": 61922.444
    },
    {
      "epoch": 0.8526189194756181,
      "grad_norm": 24.625,
      "learning_rate": 1.118054837480268e-06,
      "loss": 0.7431,
      "num_input_tokens_seen": 4761286656,
      "step": 64160,
      "train_runtime": 76891.6152,
      "train_tokens_per_second": 61922.053
    },
    {
      "epoch": 0.85275180895808,
      "grad_norm": 23.375,
      "learning_rate": 1.1160781279283894e-06,
      "loss": 0.9096,
      "num_input_tokens_seen": 4762149376,
      "step": 64170,
      "train_runtime": 76904.9848,
      "train_tokens_per_second": 61922.506
    },
    {
      "epoch": 0.8528846984405419,
      "grad_norm": 18.875,
      "learning_rate": 1.1141030640438244e-06,
      "loss": 0.7941,
      "num_input_tokens_seen": 4762660352,
      "step": 64180,
      "train_runtime": 76912.0919,
      "train_tokens_per_second": 61923.428
    },
    {
      "epoch": 0.8530175879230039,
      "grad_norm": 14.1875,
      "learning_rate": 1.1121296461924314e-06,
      "loss": 0.6369,
      "num_input_tokens_seen": 4762884096,
      "step": 64190,
      "train_runtime": 76915.738,
      "train_tokens_per_second": 61923.401
    },
    {
      "epoch": 0.8531504774054658,
      "grad_norm": 40.5,
      "learning_rate": 1.11015787473977e-06,
      "loss": 0.4466,
      "num_input_tokens_seen": 4762999808,
      "step": 64200,
      "train_runtime": 76918.86,
      "train_tokens_per_second": 61922.392
    },
    {
      "epoch": 0.8532833668879276,
      "grad_norm": 18.5,
      "learning_rate": 1.108187750051094e-06,
      "loss": 0.7709,
      "num_input_tokens_seen": 4765033984,
      "step": 64210,
      "train_runtime": 76952.5422,
      "train_tokens_per_second": 61921.723
    },
    {
      "epoch": 0.8534162563703895,
      "grad_norm": 21.875,
      "learning_rate": 1.1062192724913478e-06,
      "loss": 0.8286,
      "num_input_tokens_seen": 4765873664,
      "step": 64220,
      "train_runtime": 76965.2362,
      "train_tokens_per_second": 61922.42
    },
    {
      "epoch": 0.8535491458528515,
      "grad_norm": 78.5,
      "learning_rate": 1.1042524424251799e-06,
      "loss": 0.7099,
      "num_input_tokens_seen": 4766326784,
      "step": 64230,
      "train_runtime": 76972.2049,
      "train_tokens_per_second": 61922.7
    },
    {
      "epoch": 0.8536820353353134,
      "grad_norm": 21.5,
      "learning_rate": 1.1022872602169243e-06,
      "loss": 0.6678,
      "num_input_tokens_seen": 4766551808,
      "step": 64240,
      "train_runtime": 76977.3568,
      "train_tokens_per_second": 61921.479
    },
    {
      "epoch": 0.8538149248177753,
      "grad_norm": 27.75,
      "learning_rate": 1.1003237262306131e-06,
      "loss": 0.5161,
      "num_input_tokens_seen": 4766668544,
      "step": 64250,
      "train_runtime": 76980.5171,
      "train_tokens_per_second": 61920.454
    },
    {
      "epoch": 0.8539478143002373,
      "grad_norm": 21.5,
      "learning_rate": 1.0983618408299756e-06,
      "loss": 0.7872,
      "num_input_tokens_seen": 4768688640,
      "step": 64260,
      "train_runtime": 77012.8463,
      "train_tokens_per_second": 61920.691
    },
    {
      "epoch": 0.8540807037826991,
      "grad_norm": 17.875,
      "learning_rate": 1.096401604378432e-06,
      "loss": 0.8976,
      "num_input_tokens_seen": 4769536512,
      "step": 64270,
      "train_runtime": 77025.6056,
      "train_tokens_per_second": 61921.441
    },
    {
      "epoch": 0.854213593265161,
      "grad_norm": 12.4375,
      "learning_rate": 1.0944430172390995e-06,
      "loss": 0.6983,
      "num_input_tokens_seen": 4770004480,
      "step": 64280,
      "train_runtime": 77032.5221,
      "train_tokens_per_second": 61921.956
    },
    {
      "epoch": 0.8543464827476229,
      "grad_norm": 13.875,
      "learning_rate": 1.0924860797747894e-06,
      "loss": 0.6289,
      "num_input_tokens_seen": 4770238208,
      "step": 64290,
      "train_runtime": 77036.2395,
      "train_tokens_per_second": 61922.002
    },
    {
      "epoch": 0.8544793722300849,
      "grad_norm": 17.125,
      "learning_rate": 1.090530792348007e-06,
      "loss": 0.463,
      "num_input_tokens_seen": 4770350848,
      "step": 64300,
      "train_runtime": 77040.8565,
      "train_tokens_per_second": 61919.754
    },
    {
      "epoch": 0.8546122617125468,
      "grad_norm": 19.625,
      "learning_rate": 1.088577155320949e-06,
      "loss": 0.8333,
      "num_input_tokens_seen": 4772386816,
      "step": 64310,
      "train_runtime": 77073.8245,
      "train_tokens_per_second": 61919.683
    },
    {
      "epoch": 0.8547451511950087,
      "grad_norm": 19.125,
      "learning_rate": 1.0866251690555118e-06,
      "loss": 1.0002,
      "num_input_tokens_seen": 4773221632,
      "step": 64320,
      "train_runtime": 77086.0797,
      "train_tokens_per_second": 61920.669
    },
    {
      "epoch": 0.8548780406774705,
      "grad_norm": 19.375,
      "learning_rate": 1.0846748339132828e-06,
      "loss": 0.7715,
      "num_input_tokens_seen": 4773682432,
      "step": 64330,
      "train_runtime": 77092.6752,
      "train_tokens_per_second": 61921.349
    },
    {
      "epoch": 0.8550109301599325,
      "grad_norm": 11.875,
      "learning_rate": 1.0827261502555452e-06,
      "loss": 0.7178,
      "num_input_tokens_seen": 4773903360,
      "step": 64340,
      "train_runtime": 77097.8224,
      "train_tokens_per_second": 61920.08
    },
    {
      "epoch": 0.8551438196423944,
      "grad_norm": 23.125,
      "learning_rate": 1.0807791184432737e-06,
      "loss": 0.5062,
      "num_input_tokens_seen": 4774013952,
      "step": 64350,
      "train_runtime": 77100.9127,
      "train_tokens_per_second": 61919.033
    },
    {
      "epoch": 0.8552767091248563,
      "grad_norm": 21.125,
      "learning_rate": 1.0788337388371406e-06,
      "loss": 0.7919,
      "num_input_tokens_seen": 4776063232,
      "step": 64360,
      "train_runtime": 77134.7373,
      "train_tokens_per_second": 61918.448
    },
    {
      "epoch": 0.8554095986073182,
      "grad_norm": 27.875,
      "learning_rate": 1.0768900117975079e-06,
      "loss": 0.8439,
      "num_input_tokens_seen": 4776880640,
      "step": 64370,
      "train_runtime": 77147.2819,
      "train_tokens_per_second": 61918.975
    },
    {
      "epoch": 0.8555424880897802,
      "grad_norm": 22.25,
      "learning_rate": 1.074947937684434e-06,
      "loss": 0.7243,
      "num_input_tokens_seen": 4777356800,
      "step": 64380,
      "train_runtime": 77154.136,
      "train_tokens_per_second": 61919.646
    },
    {
      "epoch": 0.855675377572242,
      "grad_norm": 14.6875,
      "learning_rate": 1.0730075168576716e-06,
      "loss": 0.6682,
      "num_input_tokens_seen": 4777583872,
      "step": 64390,
      "train_runtime": 77158.0331,
      "train_tokens_per_second": 61919.462
    },
    {
      "epoch": 0.8558082670547039,
      "grad_norm": 14.5625,
      "learning_rate": 1.0710687496766659e-06,
      "loss": 0.5615,
      "num_input_tokens_seen": 4777698048,
      "step": 64400,
      "train_runtime": 77161.2314,
      "train_tokens_per_second": 61918.375
    },
    {
      "epoch": 0.8559411565371658,
      "grad_norm": 13.5625,
      "learning_rate": 1.0691316365005588e-06,
      "loss": 0.829,
      "num_input_tokens_seen": 4779716608,
      "step": 64410,
      "train_runtime": 77195.0237,
      "train_tokens_per_second": 61917.419
    },
    {
      "epoch": 0.8560740460196278,
      "grad_norm": 18.25,
      "learning_rate": 1.0671961776881778e-06,
      "loss": 0.889,
      "num_input_tokens_seen": 4780560896,
      "step": 64420,
      "train_runtime": 77207.5963,
      "train_tokens_per_second": 61918.271
    },
    {
      "epoch": 0.8562069355020897,
      "grad_norm": 18.0,
      "learning_rate": 1.065262373598055e-06,
      "loss": 0.7386,
      "num_input_tokens_seen": 4781036544,
      "step": 64430,
      "train_runtime": 77214.4364,
      "train_tokens_per_second": 61918.946
    },
    {
      "epoch": 0.8563398249845516,
      "grad_norm": 11.125,
      "learning_rate": 1.0633302245884048e-06,
      "loss": 0.6362,
      "num_input_tokens_seen": 4781270528,
      "step": 64440,
      "train_runtime": 77218.3065,
      "train_tokens_per_second": 61918.873
    },
    {
      "epoch": 0.8564727144670136,
      "grad_norm": 24.625,
      "learning_rate": 1.0613997310171476e-06,
      "loss": 0.4841,
      "num_input_tokens_seen": 4781387264,
      "step": 64450,
      "train_runtime": 77221.4771,
      "train_tokens_per_second": 61917.843
    },
    {
      "epoch": 0.8566056039494754,
      "grad_norm": 20.375,
      "learning_rate": 1.0594708932418863e-06,
      "loss": 0.7931,
      "num_input_tokens_seen": 4783809024,
      "step": 64460,
      "train_runtime": 77259.4241,
      "train_tokens_per_second": 61918.777
    },
    {
      "epoch": 0.8567384934319373,
      "grad_norm": 16.125,
      "learning_rate": 1.0575437116199162e-06,
      "loss": 0.9589,
      "num_input_tokens_seen": 4784686336,
      "step": 64470,
      "train_runtime": 77272.5635,
      "train_tokens_per_second": 61919.601
    },
    {
      "epoch": 0.8568713829143992,
      "grad_norm": 20.875,
      "learning_rate": 1.0556181865082393e-06,
      "loss": 0.7924,
      "num_input_tokens_seen": 4785174272,
      "step": 64480,
      "train_runtime": 77279.6481,
      "train_tokens_per_second": 61920.239
    },
    {
      "epoch": 0.8570042723968612,
      "grad_norm": 12.6875,
      "learning_rate": 1.0536943182635339e-06,
      "loss": 0.6595,
      "num_input_tokens_seen": 4785407744,
      "step": 64490,
      "train_runtime": 77283.6531,
      "train_tokens_per_second": 61920.051
    },
    {
      "epoch": 0.8571371618793231,
      "grad_norm": 13.5625,
      "learning_rate": 1.0517721072421872e-06,
      "loss": 0.5126,
      "num_input_tokens_seen": 4785526784,
      "step": 64500,
      "train_runtime": 77286.7706,
      "train_tokens_per_second": 61919.094
    },
    {
      "epoch": 0.857270051361785,
      "grad_norm": 27.875,
      "learning_rate": 1.0498515538002664e-06,
      "loss": 0.8273,
      "num_input_tokens_seen": 4787569408,
      "step": 64510,
      "train_runtime": 77319.6466,
      "train_tokens_per_second": 61919.184
    },
    {
      "epoch": 0.8574029408442468,
      "grad_norm": 24.875,
      "learning_rate": 1.0479326582935378e-06,
      "loss": 0.9666,
      "num_input_tokens_seen": 4788435456,
      "step": 64520,
      "train_runtime": 77332.6398,
      "train_tokens_per_second": 61919.979
    },
    {
      "epoch": 0.8575358303267088,
      "grad_norm": 39.5,
      "learning_rate": 1.046015421077462e-06,
      "loss": 0.8082,
      "num_input_tokens_seen": 4788932352,
      "step": 64530,
      "train_runtime": 77340.849,
      "train_tokens_per_second": 61919.832
    },
    {
      "epoch": 0.8576687198091707,
      "grad_norm": 44.0,
      "learning_rate": 1.0440998425071846e-06,
      "loss": 0.6011,
      "num_input_tokens_seen": 4789166336,
      "step": 64540,
      "train_runtime": 77344.7203,
      "train_tokens_per_second": 61919.758
    },
    {
      "epoch": 0.8578016092916326,
      "grad_norm": 13.6875,
      "learning_rate": 1.0421859229375563e-06,
      "loss": 0.5272,
      "num_input_tokens_seen": 4789280768,
      "step": 64550,
      "train_runtime": 77347.7301,
      "train_tokens_per_second": 61918.828
    },
    {
      "epoch": 0.8579344987740946,
      "grad_norm": 15.6875,
      "learning_rate": 1.040273662723109e-06,
      "loss": 0.7466,
      "num_input_tokens_seen": 4791300608,
      "step": 64560,
      "train_runtime": 77380.2248,
      "train_tokens_per_second": 61918.928
    },
    {
      "epoch": 0.8580673882565565,
      "grad_norm": 19.0,
      "learning_rate": 1.038363062218073e-06,
      "loss": 0.9224,
      "num_input_tokens_seen": 4792142080,
      "step": 64570,
      "train_runtime": 77393.1176,
      "train_tokens_per_second": 61919.486
    },
    {
      "epoch": 0.8582002777390183,
      "grad_norm": 14.625,
      "learning_rate": 1.0364541217763712e-06,
      "loss": 0.7364,
      "num_input_tokens_seen": 4792609024,
      "step": 64580,
      "train_runtime": 77400.696,
      "train_tokens_per_second": 61919.456
    },
    {
      "epoch": 0.8583331672214802,
      "grad_norm": 16.625,
      "learning_rate": 1.0345468417516157e-06,
      "loss": 0.6234,
      "num_input_tokens_seen": 4792839680,
      "step": 64590,
      "train_runtime": 77405.0485,
      "train_tokens_per_second": 61918.955
    },
    {
      "epoch": 0.8584660567039422,
      "grad_norm": 14.75,
      "learning_rate": 1.032641222497115e-06,
      "loss": 0.5283,
      "num_input_tokens_seen": 4792955648,
      "step": 64600,
      "train_runtime": 77408.4643,
      "train_tokens_per_second": 61917.72
    },
    {
      "epoch": 0.8585989461864041,
      "grad_norm": 17.875,
      "learning_rate": 1.0307372643658687e-06,
      "loss": 0.85,
      "num_input_tokens_seen": 4795043584,
      "step": 64610,
      "train_runtime": 77441.7477,
      "train_tokens_per_second": 61918.07
    },
    {
      "epoch": 0.858731835668866,
      "grad_norm": 25.25,
      "learning_rate": 1.0288349677105659e-06,
      "loss": 0.8801,
      "num_input_tokens_seen": 4795882752,
      "step": 64620,
      "train_runtime": 77455.994,
      "train_tokens_per_second": 61917.516
    },
    {
      "epoch": 0.858864725151328,
      "grad_norm": 17.625,
      "learning_rate": 1.0269343328835901e-06,
      "loss": 0.7339,
      "num_input_tokens_seen": 4796340736,
      "step": 64630,
      "train_runtime": 77463.2592,
      "train_tokens_per_second": 61917.621
    },
    {
      "epoch": 0.8589976146337898,
      "grad_norm": 15.8125,
      "learning_rate": 1.0250353602370177e-06,
      "loss": 0.6612,
      "num_input_tokens_seen": 4796578304,
      "step": 64640,
      "train_runtime": 77467.1304,
      "train_tokens_per_second": 61917.594
    },
    {
      "epoch": 0.8591305041162517,
      "grad_norm": 9.5,
      "learning_rate": 1.0231380501226162e-06,
      "loss": 0.5018,
      "num_input_tokens_seen": 4796693760,
      "step": 64650,
      "train_runtime": 77470.345,
      "train_tokens_per_second": 61916.515
    },
    {
      "epoch": 0.8592633935987136,
      "grad_norm": 30.75,
      "learning_rate": 1.0212424028918466e-06,
      "loss": 0.7835,
      "num_input_tokens_seen": 4798639616,
      "step": 64660,
      "train_runtime": 77501.504,
      "train_tokens_per_second": 61916.729
    },
    {
      "epoch": 0.8593962830811755,
      "grad_norm": 26.875,
      "learning_rate": 1.0193484188958591e-06,
      "loss": 0.8809,
      "num_input_tokens_seen": 4799464192,
      "step": 64670,
      "train_runtime": 77513.8514,
      "train_tokens_per_second": 61917.504
    },
    {
      "epoch": 0.8595291725636375,
      "grad_norm": 23.0,
      "learning_rate": 1.0174560984855008e-06,
      "loss": 0.7032,
      "num_input_tokens_seen": 4799908352,
      "step": 64680,
      "train_runtime": 77520.4065,
      "train_tokens_per_second": 61917.998
    },
    {
      "epoch": 0.8596620620460994,
      "grad_norm": 19.25,
      "learning_rate": 1.0155654420113014e-06,
      "loss": 0.7006,
      "num_input_tokens_seen": 4800129536,
      "step": 64690,
      "train_runtime": 77524.347,
      "train_tokens_per_second": 61917.704
    },
    {
      "epoch": 0.8597949515285612,
      "grad_norm": 12.5,
      "learning_rate": 1.013676449823492e-06,
      "loss": 0.6095,
      "num_input_tokens_seen": 4800247296,
      "step": 64700,
      "train_runtime": 77529.0727,
      "train_tokens_per_second": 61915.448
    },
    {
      "epoch": 0.8599278410110232,
      "grad_norm": 21.625,
      "learning_rate": 1.01178912227199e-06,
      "loss": 0.8605,
      "num_input_tokens_seen": 4802224128,
      "step": 64710,
      "train_runtime": 77561.4739,
      "train_tokens_per_second": 61915.071
    },
    {
      "epoch": 0.8600607304934851,
      "grad_norm": 20.375,
      "learning_rate": 1.0099034597064073e-06,
      "loss": 0.8775,
      "num_input_tokens_seen": 4803063040,
      "step": 64720,
      "train_runtime": 77574.1403,
      "train_tokens_per_second": 61915.775
    },
    {
      "epoch": 0.860193619975947,
      "grad_norm": 25.625,
      "learning_rate": 1.0080194624760476e-06,
      "loss": 0.7121,
      "num_input_tokens_seen": 4803519488,
      "step": 64730,
      "train_runtime": 77580.8133,
      "train_tokens_per_second": 61916.333
    },
    {
      "epoch": 0.8603265094584089,
      "grad_norm": 10.25,
      "learning_rate": 1.0061371309298984e-06,
      "loss": 0.5728,
      "num_input_tokens_seen": 4803746304,
      "step": 64740,
      "train_runtime": 77584.6208,
      "train_tokens_per_second": 61916.218
    },
    {
      "epoch": 0.8604593989408709,
      "grad_norm": 700.0,
      "learning_rate": 1.004256465416652e-06,
      "loss": 0.506,
      "num_input_tokens_seen": 4803857408,
      "step": 64750,
      "train_runtime": 77589.0422,
      "train_tokens_per_second": 61914.122
    },
    {
      "epoch": 0.8605922884233327,
      "grad_norm": 15.0625,
      "learning_rate": 1.0023774662846807e-06,
      "loss": 0.7615,
      "num_input_tokens_seen": 4805912832,
      "step": 64760,
      "train_runtime": 77621.9376,
      "train_tokens_per_second": 61914.363
    },
    {
      "epoch": 0.8607251779057946,
      "grad_norm": 21.875,
      "learning_rate": 1.000500133882053e-06,
      "loss": 0.9001,
      "num_input_tokens_seen": 4806779136,
      "step": 64770,
      "train_runtime": 77634.7699,
      "train_tokens_per_second": 61915.288
    },
    {
      "epoch": 0.8608580673882565,
      "grad_norm": 15.5,
      "learning_rate": 9.986244685565306e-07,
      "loss": 0.7183,
      "num_input_tokens_seen": 4807248128,
      "step": 64780,
      "train_runtime": 77641.6376,
      "train_tokens_per_second": 61915.852
    },
    {
      "epoch": 0.8609909568707185,
      "grad_norm": 20.625,
      "learning_rate": 9.967504706555576e-07,
      "loss": 0.6377,
      "num_input_tokens_seen": 4807475712,
      "step": 64790,
      "train_runtime": 77646.45,
      "train_tokens_per_second": 61914.945
    },
    {
      "epoch": 0.8611238463531804,
      "grad_norm": 15.375,
      "learning_rate": 9.94878140526283e-07,
      "loss": 0.4625,
      "num_input_tokens_seen": 4807587072,
      "step": 64800,
      "train_runtime": 77649.5239,
      "train_tokens_per_second": 61913.929
    },
    {
      "epoch": 0.8612567358356423,
      "grad_norm": 27.625,
      "learning_rate": 9.930074785155319e-07,
      "loss": 0.8082,
      "num_input_tokens_seen": 4809667072,
      "step": 64810,
      "train_runtime": 77683.4048,
      "train_tokens_per_second": 61913.701
    },
    {
      "epoch": 0.8613896253181041,
      "grad_norm": 32.5,
      "learning_rate": 9.911384849698358e-07,
      "loss": 0.9062,
      "num_input_tokens_seen": 4810525184,
      "step": 64820,
      "train_runtime": 77696.1995,
      "train_tokens_per_second": 61914.549
    },
    {
      "epoch": 0.8615225148005661,
      "grad_norm": 24.125,
      "learning_rate": 9.892711602354034e-07,
      "loss": 0.7521,
      "num_input_tokens_seen": 4810976256,
      "step": 64830,
      "train_runtime": 77703.0082,
      "train_tokens_per_second": 61914.929
    },
    {
      "epoch": 0.861655404283028,
      "grad_norm": 11.0,
      "learning_rate": 9.87405504658141e-07,
      "loss": 0.614,
      "num_input_tokens_seen": 4811206400,
      "step": 64840,
      "train_runtime": 77706.8722,
      "train_tokens_per_second": 61914.812
    },
    {
      "epoch": 0.8617882937654899,
      "grad_norm": 44.25,
      "learning_rate": 9.85541518583646e-07,
      "loss": 0.4894,
      "num_input_tokens_seen": 4811319808,
      "step": 64850,
      "train_runtime": 77709.9955,
      "train_tokens_per_second": 61913.783
    },
    {
      "epoch": 0.8619211832479519,
      "grad_norm": 14.6875,
      "learning_rate": 9.836792023572018e-07,
      "loss": 0.8102,
      "num_input_tokens_seen": 4813424640,
      "step": 64860,
      "train_runtime": 77746.2704,
      "train_tokens_per_second": 61911.968
    },
    {
      "epoch": 0.8620540727304138,
      "grad_norm": 34.25,
      "learning_rate": 9.818185563237915e-07,
      "loss": 0.8858,
      "num_input_tokens_seen": 4814292224,
      "step": 64870,
      "train_runtime": 77759.3685,
      "train_tokens_per_second": 61912.697
    },
    {
      "epoch": 0.8621869622128756,
      "grad_norm": 22.75,
      "learning_rate": 9.79959580828077e-07,
      "loss": 0.6617,
      "num_input_tokens_seen": 4814798336,
      "step": 64880,
      "train_runtime": 77766.569,
      "train_tokens_per_second": 61913.473
    },
    {
      "epoch": 0.8623198516953375,
      "grad_norm": 23.875,
      "learning_rate": 9.78102276214421e-07,
      "loss": 0.6576,
      "num_input_tokens_seen": 4815041280,
      "step": 64890,
      "train_runtime": 77770.2095,
      "train_tokens_per_second": 61913.698
    },
    {
      "epoch": 0.8624527411777995,
      "grad_norm": 52.0,
      "learning_rate": 9.762466428268724e-07,
      "loss": 0.5508,
      "num_input_tokens_seen": 4815163136,
      "step": 64900,
      "train_runtime": 77772.7799,
      "train_tokens_per_second": 61913.219
    },
    {
      "epoch": 0.8625856306602614,
      "grad_norm": 34.25,
      "learning_rate": 9.743926810091687e-07,
      "loss": 0.8299,
      "num_input_tokens_seen": 4817230336,
      "step": 64910,
      "train_runtime": 77806.6359,
      "train_tokens_per_second": 61912.847
    },
    {
      "epoch": 0.8627185201427233,
      "grad_norm": 25.0,
      "learning_rate": 9.725403911047414e-07,
      "loss": 0.8917,
      "num_input_tokens_seen": 4818059008,
      "step": 64920,
      "train_runtime": 77819.1599,
      "train_tokens_per_second": 61913.531
    },
    {
      "epoch": 0.8628514096251853,
      "grad_norm": 20.5,
      "learning_rate": 9.706897734567123e-07,
      "loss": 0.6847,
      "num_input_tokens_seen": 4818513664,
      "step": 64930,
      "train_runtime": 77825.7294,
      "train_tokens_per_second": 61914.147
    },
    {
      "epoch": 0.8629842991076472,
      "grad_norm": 51.75,
      "learning_rate": 9.688408284078877e-07,
      "loss": 0.6575,
      "num_input_tokens_seen": 4818741760,
      "step": 64940,
      "train_runtime": 77829.1768,
      "train_tokens_per_second": 61914.335
    },
    {
      "epoch": 0.863117188590109,
      "grad_norm": 13.0,
      "learning_rate": 9.669935563007703e-07,
      "loss": 0.5285,
      "num_input_tokens_seen": 4818856704,
      "step": 64950,
      "train_runtime": 77831.8717,
      "train_tokens_per_second": 61913.668
    },
    {
      "epoch": 0.8632500780725709,
      "grad_norm": 29.875,
      "learning_rate": 9.651479574775503e-07,
      "loss": 0.8974,
      "num_input_tokens_seen": 4821015040,
      "step": 64960,
      "train_runtime": 77865.6744,
      "train_tokens_per_second": 61914.51
    },
    {
      "epoch": 0.8633829675550329,
      "grad_norm": 19.0,
      "learning_rate": 9.63304032280109e-07,
      "loss": 0.8508,
      "num_input_tokens_seen": 4821861376,
      "step": 64970,
      "train_runtime": 77878.1927,
      "train_tokens_per_second": 61915.425
    },
    {
      "epoch": 0.8635158570374948,
      "grad_norm": 22.25,
      "learning_rate": 9.61461781050017e-07,
      "loss": 0.7014,
      "num_input_tokens_seen": 4822317056,
      "step": 64980,
      "train_runtime": 77885.825,
      "train_tokens_per_second": 61915.208
    },
    {
      "epoch": 0.8636487465199567,
      "grad_norm": 14.625,
      "learning_rate": 9.596212041285357e-07,
      "loss": 0.5897,
      "num_input_tokens_seen": 4822540800,
      "step": 64990,
      "train_runtime": 77889.3261,
      "train_tokens_per_second": 61915.298
    },
    {
      "epoch": 0.8637816360024186,
      "grad_norm": 14.375,
      "learning_rate": 9.577823018566167e-07,
      "loss": 0.4865,
      "num_input_tokens_seen": 4822655744,
      "step": 65000,
      "train_runtime": 77891.9926,
      "train_tokens_per_second": 61914.654
    },
    {
      "epoch": 0.8637816360024186,
      "eval_accuracy": 0.8191,
      "eval_balanced_accuracy": 0.817376424006387,
      "eval_invalid_f1": 0.8367770459261933,
      "eval_invalid_precision": 0.8411028478142573,
      "eval_invalid_recall": 0.8324955116696588,
      "eval_loss": 0.36133211851119995,
      "eval_macro_f1": 0.8169530626064745,
      "eval_mcc": 0.6339603373516798,
      "eval_runtime": 86.0086,
      "eval_samples_per_second": 116.267,
      "eval_steps_per_second": 29.067,
      "eval_valid_f1": 0.7971290792867556,
      "eval_valid_precision": 0.7920659683530198,
      "eval_valid_recall": 0.8022573363431151,
      "num_input_tokens_seen": 4822655744,
      "step": 65000
    },
    {
      "epoch": 0.8639145254848805,
      "grad_norm": 19.0,
      "learning_rate": 9.559450745748955e-07,
      "loss": 0.7804,
      "num_input_tokens_seen": 4824685056,
      "step": 65010,
      "train_runtime": 78015.4274,
      "train_tokens_per_second": 61842.705
    },
    {
      "epoch": 0.8640474149673424,
      "grad_norm": 39.25,
      "learning_rate": 9.541095226237052e-07,
      "loss": 0.8857,
      "num_input_tokens_seen": 4825496832,
      "step": 65020,
      "train_runtime": 78028.8164,
      "train_tokens_per_second": 61842.497
    },
    {
      "epoch": 0.8641803044498043,
      "grad_norm": 24.75,
      "learning_rate": 9.522756463430649e-07,
      "loss": 0.7229,
      "num_input_tokens_seen": 4825978112,
      "step": 65030,
      "train_runtime": 78035.4065,
      "train_tokens_per_second": 61843.442
    },
    {
      "epoch": 0.8643131939322662,
      "grad_norm": 15.125,
      "learning_rate": 9.504434460726841e-07,
      "loss": 0.6481,
      "num_input_tokens_seen": 4826204928,
      "step": 65040,
      "train_runtime": 78038.9918,
      "train_tokens_per_second": 61843.507
    },
    {
      "epoch": 0.8644460834147282,
      "grad_norm": 8.8125,
      "learning_rate": 9.486129221519647e-07,
      "loss": 0.4615,
      "num_input_tokens_seen": 4826321920,
      "step": 65050,
      "train_runtime": 78041.5327,
      "train_tokens_per_second": 61842.992
    },
    {
      "epoch": 0.8645789728971901,
      "grad_norm": 17.875,
      "learning_rate": 9.467840749199863e-07,
      "loss": 0.8287,
      "num_input_tokens_seen": 4828296448,
      "step": 65060,
      "train_runtime": 78073.8552,
      "train_tokens_per_second": 61842.68
    },
    {
      "epoch": 0.8647118623796519,
      "grad_norm": 22.875,
      "learning_rate": 9.449569047155361e-07,
      "loss": 0.8227,
      "num_input_tokens_seen": 4829122304,
      "step": 65070,
      "train_runtime": 78086.4354,
      "train_tokens_per_second": 61843.293
    },
    {
      "epoch": 0.8648447518621138,
      "grad_norm": 17.5,
      "learning_rate": 9.431314118770751e-07,
      "loss": 0.7228,
      "num_input_tokens_seen": 4829586688,
      "step": 65080,
      "train_runtime": 78093.1212,
      "train_tokens_per_second": 61843.945
    },
    {
      "epoch": 0.8649776413445758,
      "grad_norm": 20.625,
      "learning_rate": 9.41307596742762e-07,
      "loss": 0.6781,
      "num_input_tokens_seen": 4829809152,
      "step": 65090,
      "train_runtime": 78096.5997,
      "train_tokens_per_second": 61844.039
    },
    {
      "epoch": 0.8651105308270377,
      "grad_norm": 12.375,
      "learning_rate": 9.394854596504432e-07,
      "loss": 0.4693,
      "num_input_tokens_seen": 4829923840,
      "step": 65100,
      "train_runtime": 78100.2834,
      "train_tokens_per_second": 61842.59
    },
    {
      "epoch": 0.8652434203094996,
      "grad_norm": 17.125,
      "learning_rate": 9.376650009376487e-07,
      "loss": 0.8297,
      "num_input_tokens_seen": 4831991808,
      "step": 65110,
      "train_runtime": 78133.1154,
      "train_tokens_per_second": 61843.071
    },
    {
      "epoch": 0.8653763097919616,
      "grad_norm": 23.375,
      "learning_rate": 9.358462209416086e-07,
      "loss": 0.8754,
      "num_input_tokens_seen": 4832820224,
      "step": 65120,
      "train_runtime": 78145.8746,
      "train_tokens_per_second": 61843.575
    },
    {
      "epoch": 0.8655091992744234,
      "grad_norm": 20.75,
      "learning_rate": 9.340291199992302e-07,
      "loss": 0.7506,
      "num_input_tokens_seen": 4833288448,
      "step": 65130,
      "train_runtime": 78152.7634,
      "train_tokens_per_second": 61844.114
    },
    {
      "epoch": 0.8656420887568853,
      "grad_norm": 11.9375,
      "learning_rate": 9.322136984471186e-07,
      "loss": 0.6497,
      "num_input_tokens_seen": 4833517824,
      "step": 65140,
      "train_runtime": 78156.2606,
      "train_tokens_per_second": 61844.282
    },
    {
      "epoch": 0.8657749782393472,
      "grad_norm": 15.5,
      "learning_rate": 9.303999566215627e-07,
      "loss": 0.5398,
      "num_input_tokens_seen": 4833632256,
      "step": 65150,
      "train_runtime": 78160.1858,
      "train_tokens_per_second": 61842.64
    },
    {
      "epoch": 0.8659078677218092,
      "grad_norm": 26.625,
      "learning_rate": 9.285878948585426e-07,
      "loss": 0.8563,
      "num_input_tokens_seen": 4835587840,
      "step": 65160,
      "train_runtime": 78191.7035,
      "train_tokens_per_second": 61842.723
    },
    {
      "epoch": 0.8660407572042711,
      "grad_norm": 23.0,
      "learning_rate": 9.267775134937296e-07,
      "loss": 0.8188,
      "num_input_tokens_seen": 4836426752,
      "step": 65170,
      "train_runtime": 78203.916,
      "train_tokens_per_second": 61843.792
    },
    {
      "epoch": 0.866173646686733,
      "grad_norm": 31.875,
      "learning_rate": 9.249688128624723e-07,
      "loss": 0.7467,
      "num_input_tokens_seen": 4836865792,
      "step": 65180,
      "train_runtime": 78210.1368,
      "train_tokens_per_second": 61844.487
    },
    {
      "epoch": 0.8663065361691948,
      "grad_norm": 26.5,
      "learning_rate": 9.231617932998271e-07,
      "loss": 0.7035,
      "num_input_tokens_seen": 4837090816,
      "step": 65190,
      "train_runtime": 78214.7714,
      "train_tokens_per_second": 61843.699
    },
    {
      "epoch": 0.8664394256516568,
      "grad_norm": 12.25,
      "learning_rate": 9.213564551405207e-07,
      "loss": 0.462,
      "num_input_tokens_seen": 4837204736,
      "step": 65200,
      "train_runtime": 78217.3496,
      "train_tokens_per_second": 61843.117
    },
    {
      "epoch": 0.8665723151341187,
      "grad_norm": 16.375,
      "learning_rate": 9.195527987189778e-07,
      "loss": 0.8809,
      "num_input_tokens_seen": 4839152128,
      "step": 65210,
      "train_runtime": 78248.8833,
      "train_tokens_per_second": 61843.082
    },
    {
      "epoch": 0.8667052046165806,
      "grad_norm": 21.0,
      "learning_rate": 9.177508243693112e-07,
      "loss": 0.9178,
      "num_input_tokens_seen": 4839960064,
      "step": 65220,
      "train_runtime": 78261.3781,
      "train_tokens_per_second": 61843.532
    },
    {
      "epoch": 0.8668380940990426,
      "grad_norm": 21.0,
      "learning_rate": 9.159505324253181e-07,
      "loss": 0.6866,
      "num_input_tokens_seen": 4840421376,
      "step": 65230,
      "train_runtime": 78267.7933,
      "train_tokens_per_second": 61844.357
    },
    {
      "epoch": 0.8669709835815045,
      "grad_norm": 13.25,
      "learning_rate": 9.141519232204888e-07,
      "loss": 0.6716,
      "num_input_tokens_seen": 4840636160,
      "step": 65240,
      "train_runtime": 78271.2446,
      "train_tokens_per_second": 61844.374
    },
    {
      "epoch": 0.8671038730639663,
      "grad_norm": 53.5,
      "learning_rate": 9.123549970879997e-07,
      "loss": 0.5565,
      "num_input_tokens_seen": 4840748544,
      "step": 65250,
      "train_runtime": 78273.9324,
      "train_tokens_per_second": 61843.687
    },
    {
      "epoch": 0.8672367625464282,
      "grad_norm": 18.625,
      "learning_rate": 9.105597543607137e-07,
      "loss": 0.815,
      "num_input_tokens_seen": 4842768640,
      "step": 65260,
      "train_runtime": 78307.6553,
      "train_tokens_per_second": 61842.851
    },
    {
      "epoch": 0.8673696520288902,
      "grad_norm": 13.0,
      "learning_rate": 9.087661953711835e-07,
      "loss": 0.8582,
      "num_input_tokens_seen": 4843640576,
      "step": 65270,
      "train_runtime": 78321.8142,
      "train_tokens_per_second": 61842.804
    },
    {
      "epoch": 0.8675025415113521,
      "grad_norm": 95.0,
      "learning_rate": 9.069743204516501e-07,
      "loss": 0.7543,
      "num_input_tokens_seen": 4844114432,
      "step": 65280,
      "train_runtime": 78328.6165,
      "train_tokens_per_second": 61843.483
    },
    {
      "epoch": 0.867635430993814,
      "grad_norm": 23.75,
      "learning_rate": 9.051841299340436e-07,
      "loss": 0.6745,
      "num_input_tokens_seen": 4844347904,
      "step": 65290,
      "train_runtime": 78332.3507,
      "train_tokens_per_second": 61843.515
    },
    {
      "epoch": 0.867768320476276,
      "grad_norm": 27.875,
      "learning_rate": 9.033956241499819e-07,
      "loss": 0.5341,
      "num_input_tokens_seen": 4844465408,
      "step": 65300,
      "train_runtime": 78335.0326,
      "train_tokens_per_second": 61842.898
    },
    {
      "epoch": 0.8679012099587378,
      "grad_norm": 13.9375,
      "learning_rate": 9.016088034307635e-07,
      "loss": 0.7989,
      "num_input_tokens_seen": 4846523392,
      "step": 65310,
      "train_runtime": 78368.891,
      "train_tokens_per_second": 61842.439
    },
    {
      "epoch": 0.8680340994411997,
      "grad_norm": 30.5,
      "learning_rate": 8.998236681073891e-07,
      "loss": 0.9375,
      "num_input_tokens_seen": 4847401472,
      "step": 65320,
      "train_runtime": 78383.1055,
      "train_tokens_per_second": 61842.427
    },
    {
      "epoch": 0.8681669889236616,
      "grad_norm": 60.25,
      "learning_rate": 8.980402185105341e-07,
      "loss": 0.8534,
      "num_input_tokens_seen": 4847897856,
      "step": 65330,
      "train_runtime": 78390.0193,
      "train_tokens_per_second": 61843.305
    },
    {
      "epoch": 0.8682998784061235,
      "grad_norm": 16.25,
      "learning_rate": 8.962584549705678e-07,
      "loss": 0.5977,
      "num_input_tokens_seen": 4848137984,
      "step": 65340,
      "train_runtime": 78393.6913,
      "train_tokens_per_second": 61843.471
    },
    {
      "epoch": 0.8684327678885855,
      "grad_norm": 10.625,
      "learning_rate": 8.94478377817547e-07,
      "loss": 0.4544,
      "num_input_tokens_seen": 4848253696,
      "step": 65350,
      "train_runtime": 78396.2096,
      "train_tokens_per_second": 61842.961
    },
    {
      "epoch": 0.8685656573710474,
      "grad_norm": 16.25,
      "learning_rate": 8.926999873812126e-07,
      "loss": 0.8384,
      "num_input_tokens_seen": 4850373120,
      "step": 65360,
      "train_runtime": 78431.2627,
      "train_tokens_per_second": 61842.344
    },
    {
      "epoch": 0.8686985468535092,
      "grad_norm": 20.125,
      "learning_rate": 8.909232839909998e-07,
      "loss": 0.8478,
      "num_input_tokens_seen": 4851267072,
      "step": 65370,
      "train_runtime": 78443.9562,
      "train_tokens_per_second": 61843.733
    },
    {
      "epoch": 0.8688314363359712,
      "grad_norm": 14.0625,
      "learning_rate": 8.891482679760211e-07,
      "loss": 0.7342,
      "num_input_tokens_seen": 4851740416,
      "step": 65380,
      "train_runtime": 78450.6078,
      "train_tokens_per_second": 61844.523
    },
    {
      "epoch": 0.8689643258184331,
      "grad_norm": 19.125,
      "learning_rate": 8.8737493966509e-07,
      "loss": 0.5743,
      "num_input_tokens_seen": 4851970560,
      "step": 65390,
      "train_runtime": 78454.1541,
      "train_tokens_per_second": 61844.661
    },
    {
      "epoch": 0.869097215300895,
      "grad_norm": 38.75,
      "learning_rate": 8.856032993866925e-07,
      "loss": 0.5797,
      "num_input_tokens_seen": 4852085248,
      "step": 65400,
      "train_runtime": 78456.7261,
      "train_tokens_per_second": 61844.095
    },
    {
      "epoch": 0.8692301047833569,
      "grad_norm": 13.25,
      "learning_rate": 8.83833347469013e-07,
      "loss": 0.8081,
      "num_input_tokens_seen": 4854114048,
      "step": 65410,
      "train_runtime": 78489.3794,
      "train_tokens_per_second": 61844.215
    },
    {
      "epoch": 0.8693629942658189,
      "grad_norm": 14.8125,
      "learning_rate": 8.820650842399204e-07,
      "loss": 0.8907,
      "num_input_tokens_seen": 4854919936,
      "step": 65420,
      "train_runtime": 78502.5166,
      "train_tokens_per_second": 61844.131
    },
    {
      "epoch": 0.8694958837482808,
      "grad_norm": 29.25,
      "learning_rate": 8.802985100269635e-07,
      "loss": 0.7519,
      "num_input_tokens_seen": 4855376384,
      "step": 65430,
      "train_runtime": 78508.8142,
      "train_tokens_per_second": 61844.984
    },
    {
      "epoch": 0.8696287732307426,
      "grad_norm": 12.375,
      "learning_rate": 8.785336251573928e-07,
      "loss": 0.7023,
      "num_input_tokens_seen": 4855591424,
      "step": 65440,
      "train_runtime": 78512.334,
      "train_tokens_per_second": 61844.951
    },
    {
      "epoch": 0.8697616627132045,
      "grad_norm": 8.9375,
      "learning_rate": 8.767704299581315e-07,
      "loss": 0.5082,
      "num_input_tokens_seen": 4855709696,
      "step": 65450,
      "train_runtime": 78515.1044,
      "train_tokens_per_second": 61844.275
    },
    {
      "epoch": 0.8698945521956665,
      "grad_norm": 21.0,
      "learning_rate": 8.75008924755798e-07,
      "loss": 0.8193,
      "num_input_tokens_seen": 4857739520,
      "step": 65460,
      "train_runtime": 78547.476,
      "train_tokens_per_second": 61844.629
    },
    {
      "epoch": 0.8700274416781284,
      "grad_norm": 27.875,
      "learning_rate": 8.732491098766949e-07,
      "loss": 0.8435,
      "num_input_tokens_seen": 4858593536,
      "step": 65470,
      "train_runtime": 78561.4781,
      "train_tokens_per_second": 61844.477
    },
    {
      "epoch": 0.8701603311605903,
      "grad_norm": 27.125,
      "learning_rate": 8.714909856468123e-07,
      "loss": 0.7335,
      "num_input_tokens_seen": 4859073024,
      "step": 65480,
      "train_runtime": 78568.236,
      "train_tokens_per_second": 61845.261
    },
    {
      "epoch": 0.8702932206430523,
      "grad_norm": 15.9375,
      "learning_rate": 8.697345523918289e-07,
      "loss": 0.6464,
      "num_input_tokens_seen": 4859298304,
      "step": 65490,
      "train_runtime": 78571.7954,
      "train_tokens_per_second": 61845.326
    },
    {
      "epoch": 0.8704261101255141,
      "grad_norm": 6.34375,
      "learning_rate": 8.67979810437104e-07,
      "loss": 0.4804,
      "num_input_tokens_seen": 4859412992,
      "step": 65500,
      "train_runtime": 78574.4073,
      "train_tokens_per_second": 61844.73
    },
    {
      "epoch": 0.870558999607976,
      "grad_norm": 18.125,
      "learning_rate": 8.662267601076923e-07,
      "loss": 0.8263,
      "num_input_tokens_seen": 4861479424,
      "step": 65510,
      "train_runtime": 78608.1132,
      "train_tokens_per_second": 61844.5
    },
    {
      "epoch": 0.8706918890904379,
      "grad_norm": 23.25,
      "learning_rate": 8.644754017283285e-07,
      "loss": 0.8183,
      "num_input_tokens_seen": 4862347008,
      "step": 65520,
      "train_runtime": 78622.0581,
      "train_tokens_per_second": 61844.565
    },
    {
      "epoch": 0.8708247785728999,
      "grad_norm": 21.625,
      "learning_rate": 8.627257356234365e-07,
      "loss": 0.7042,
      "num_input_tokens_seen": 4862831360,
      "step": 65530,
      "train_runtime": 78628.8836,
      "train_tokens_per_second": 61845.357
    },
    {
      "epoch": 0.8709576680553618,
      "grad_norm": 17.875,
      "learning_rate": 8.609777621171267e-07,
      "loss": 0.6787,
      "num_input_tokens_seen": 4863059712,
      "step": 65540,
      "train_runtime": 78632.3407,
      "train_tokens_per_second": 61845.542
    },
    {
      "epoch": 0.8710905575378237,
      "grad_norm": 19.875,
      "learning_rate": 8.592314815331937e-07,
      "loss": 0.5521,
      "num_input_tokens_seen": 4863172864,
      "step": 65550,
      "train_runtime": 78634.8427,
      "train_tokens_per_second": 61845.013
    },
    {
      "epoch": 0.8712234470202855,
      "grad_norm": 27.75,
      "learning_rate": 8.574868941951231e-07,
      "loss": 0.8988,
      "num_input_tokens_seen": 4865243904,
      "step": 65560,
      "train_runtime": 78668.5446,
      "train_tokens_per_second": 61844.844
    },
    {
      "epoch": 0.8713563365027475,
      "grad_norm": 18.25,
      "learning_rate": 8.557440004260853e-07,
      "loss": 0.8838,
      "num_input_tokens_seen": 4866092544,
      "step": 65570,
      "train_runtime": 78681.3221,
      "train_tokens_per_second": 61845.587
    },
    {
      "epoch": 0.8714892259852094,
      "grad_norm": 14.625,
      "learning_rate": 8.540028005489309e-07,
      "loss": 0.6909,
      "num_input_tokens_seen": 4866570496,
      "step": 65580,
      "train_runtime": 78688.1132,
      "train_tokens_per_second": 61846.323
    },
    {
      "epoch": 0.8716221154676713,
      "grad_norm": 13.0625,
      "learning_rate": 8.522632948862053e-07,
      "loss": 0.5786,
      "num_input_tokens_seen": 4866804992,
      "step": 65590,
      "train_runtime": 78691.6951,
      "train_tokens_per_second": 61846.488
    },
    {
      "epoch": 0.8717550049501332,
      "grad_norm": 256.0,
      "learning_rate": 8.505254837601352e-07,
      "loss": 0.5256,
      "num_input_tokens_seen": 4866921472,
      "step": 65600,
      "train_runtime": 78695.3236,
      "train_tokens_per_second": 61845.117
    },
    {
      "epoch": 0.8718878944325952,
      "grad_norm": 61.25,
      "learning_rate": 8.487893674926339e-07,
      "loss": 0.7888,
      "num_input_tokens_seen": 4869056256,
      "step": 65610,
      "train_runtime": 78727.3664,
      "train_tokens_per_second": 61847.061
    },
    {
      "epoch": 0.872020783915057,
      "grad_norm": 19.875,
      "learning_rate": 8.470549464053057e-07,
      "loss": 0.8468,
      "num_input_tokens_seen": 4869865984,
      "step": 65620,
      "train_runtime": 78739.8168,
      "train_tokens_per_second": 61847.566
    },
    {
      "epoch": 0.8721536733975189,
      "grad_norm": 14.5,
      "learning_rate": 8.453222208194289e-07,
      "loss": 0.7257,
      "num_input_tokens_seen": 4870315520,
      "step": 65630,
      "train_runtime": 78746.2806,
      "train_tokens_per_second": 61848.198
    },
    {
      "epoch": 0.8722865628799809,
      "grad_norm": 11.6875,
      "learning_rate": 8.435911910559836e-07,
      "loss": 0.6296,
      "num_input_tokens_seen": 4870535168,
      "step": 65640,
      "train_runtime": 78749.7746,
      "train_tokens_per_second": 61848.243
    },
    {
      "epoch": 0.8724194523624428,
      "grad_norm": 11.125,
      "learning_rate": 8.418618574356219e-07,
      "loss": 0.5297,
      "num_input_tokens_seen": 4870649088,
      "step": 65650,
      "train_runtime": 78753.4966,
      "train_tokens_per_second": 61846.766
    },
    {
      "epoch": 0.8725523418449047,
      "grad_norm": 15.75,
      "learning_rate": 8.4013422027869e-07,
      "loss": 0.7032,
      "num_input_tokens_seen": 4872833280,
      "step": 65660,
      "train_runtime": 78787.7319,
      "train_tokens_per_second": 61847.615
    },
    {
      "epoch": 0.8726852313273666,
      "grad_norm": 17.125,
      "learning_rate": 8.384082799052185e-07,
      "loss": 0.9016,
      "num_input_tokens_seen": 4873705728,
      "step": 65670,
      "train_runtime": 78800.9096,
      "train_tokens_per_second": 61848.344
    },
    {
      "epoch": 0.8728181208098285,
      "grad_norm": 18.25,
      "learning_rate": 8.366840366349171e-07,
      "loss": 0.7292,
      "num_input_tokens_seen": 4874218752,
      "step": 65680,
      "train_runtime": 78808.0492,
      "train_tokens_per_second": 61849.25
    },
    {
      "epoch": 0.8729510102922904,
      "grad_norm": 11.5,
      "learning_rate": 8.349614907871939e-07,
      "loss": 0.6991,
      "num_input_tokens_seen": 4874454016,
      "step": 65690,
      "train_runtime": 78812.808,
      "train_tokens_per_second": 61848.501
    },
    {
      "epoch": 0.8730838997747523,
      "grad_norm": 9.5625,
      "learning_rate": 8.332406426811268e-07,
      "loss": 0.5214,
      "num_input_tokens_seen": 4874569472,
      "step": 65700,
      "train_runtime": 78815.3344,
      "train_tokens_per_second": 61847.983
    },
    {
      "epoch": 0.8732167892572142,
      "grad_norm": 19.25,
      "learning_rate": 8.315214926354965e-07,
      "loss": 0.795,
      "num_input_tokens_seen": 4876559104,
      "step": 65710,
      "train_runtime": 78846.98,
      "train_tokens_per_second": 61848.394
    },
    {
      "epoch": 0.8733496787396762,
      "grad_norm": 31.375,
      "learning_rate": 8.298040409687547e-07,
      "loss": 0.8612,
      "num_input_tokens_seen": 4877366016,
      "step": 65720,
      "train_runtime": 78859.4297,
      "train_tokens_per_second": 61848.862
    },
    {
      "epoch": 0.8734825682221381,
      "grad_norm": 22.5,
      "learning_rate": 8.280882879990448e-07,
      "loss": 0.7155,
      "num_input_tokens_seen": 4877837312,
      "step": 65730,
      "train_runtime": 78866.044,
      "train_tokens_per_second": 61849.651
    },
    {
      "epoch": 0.8736154577045999,
      "grad_norm": 49.75,
      "learning_rate": 8.26374234044196e-07,
      "loss": 0.6647,
      "num_input_tokens_seen": 4878072320,
      "step": 65740,
      "train_runtime": 78869.601,
      "train_tokens_per_second": 61849.842
    },
    {
      "epoch": 0.8737483471870618,
      "grad_norm": 11.0,
      "learning_rate": 8.246618794217176e-07,
      "loss": 0.591,
      "num_input_tokens_seen": 4878186496,
      "step": 65750,
      "train_runtime": 78872.1195,
      "train_tokens_per_second": 61849.314
    },
    {
      "epoch": 0.8738812366695238,
      "grad_norm": 162.0,
      "learning_rate": 8.229512244488159e-07,
      "loss": 0.8811,
      "num_input_tokens_seen": 4880347392,
      "step": 65760,
      "train_runtime": 78905.7009,
      "train_tokens_per_second": 61850.378
    },
    {
      "epoch": 0.8740141261519857,
      "grad_norm": 28.0,
      "learning_rate": 8.212422694423661e-07,
      "loss": 0.9036,
      "num_input_tokens_seen": 4881168384,
      "step": 65770,
      "train_runtime": 78919.3042,
      "train_tokens_per_second": 61850.119
    },
    {
      "epoch": 0.8741470156344476,
      "grad_norm": 32.5,
      "learning_rate": 8.195350147189407e-07,
      "loss": 0.6933,
      "num_input_tokens_seen": 4881644032,
      "step": 65780,
      "train_runtime": 78926.0011,
      "train_tokens_per_second": 61850.898
    },
    {
      "epoch": 0.8742799051169096,
      "grad_norm": 21.375,
      "learning_rate": 8.178294605947934e-07,
      "loss": 0.6599,
      "num_input_tokens_seen": 4881882368,
      "step": 65790,
      "train_runtime": 78929.5811,
      "train_tokens_per_second": 61851.112
    },
    {
      "epoch": 0.8744127945993714,
      "grad_norm": 12.0,
      "learning_rate": 8.161256073858637e-07,
      "loss": 0.5123,
      "num_input_tokens_seen": 4882000640,
      "step": 65800,
      "train_runtime": 78932.0852,
      "train_tokens_per_second": 61850.648
    },
    {
      "epoch": 0.8745456840818333,
      "grad_norm": 13.8125,
      "learning_rate": 8.144234554077745e-07,
      "loss": 0.7754,
      "num_input_tokens_seen": 4884104192,
      "step": 65810,
      "train_runtime": 78966.366,
      "train_tokens_per_second": 61850.436
    },
    {
      "epoch": 0.8746785735642952,
      "grad_norm": 22.0,
      "learning_rate": 8.127230049758328e-07,
      "loss": 0.8924,
      "num_input_tokens_seen": 4884913152,
      "step": 65820,
      "train_runtime": 78979.9186,
      "train_tokens_per_second": 61850.066
    },
    {
      "epoch": 0.8748114630467572,
      "grad_norm": 25.25,
      "learning_rate": 8.110242564050319e-07,
      "loss": 0.7815,
      "num_input_tokens_seen": 4885374720,
      "step": 65830,
      "train_runtime": 78986.7101,
      "train_tokens_per_second": 61850.591
    },
    {
      "epoch": 0.8749443525292191,
      "grad_norm": 15.1875,
      "learning_rate": 8.093272100100524e-07,
      "loss": 0.6324,
      "num_input_tokens_seen": 4885602560,
      "step": 65840,
      "train_runtime": 78990.3554,
      "train_tokens_per_second": 61850.621
    },
    {
      "epoch": 0.875077242011681,
      "grad_norm": 23.0,
      "learning_rate": 8.076318661052552e-07,
      "loss": 0.5362,
      "num_input_tokens_seen": 4885717760,
      "step": 65850,
      "train_runtime": 78992.9892,
      "train_tokens_per_second": 61850.017
    },
    {
      "epoch": 0.875210131494143,
      "grad_norm": 25.0,
      "learning_rate": 8.059382250046866e-07,
      "loss": 0.7651,
      "num_input_tokens_seen": 4887872000,
      "step": 65860,
      "train_runtime": 79028.3539,
      "train_tokens_per_second": 61849.599
    },
    {
      "epoch": 0.8753430209766048,
      "grad_norm": 31.125,
      "learning_rate": 8.042462870220812e-07,
      "loss": 0.9086,
      "num_input_tokens_seen": 4888729344,
      "step": 65870,
      "train_runtime": 79042.4372,
      "train_tokens_per_second": 61849.426
    },
    {
      "epoch": 0.8754759104590667,
      "grad_norm": 18.0,
      "learning_rate": 8.025560524708543e-07,
      "loss": 0.7707,
      "num_input_tokens_seen": 4889223936,
      "step": 65880,
      "train_runtime": 79049.4473,
      "train_tokens_per_second": 61850.198
    },
    {
      "epoch": 0.8756087999415286,
      "grad_norm": 15.875,
      "learning_rate": 8.008675216641071e-07,
      "loss": 0.6084,
      "num_input_tokens_seen": 4889454336,
      "step": 65890,
      "train_runtime": 79052.9644,
      "train_tokens_per_second": 61850.36
    },
    {
      "epoch": 0.8757416894239906,
      "grad_norm": 16.875,
      "learning_rate": 7.991806949146241e-07,
      "loss": 0.5376,
      "num_input_tokens_seen": 4889568256,
      "step": 65900,
      "train_runtime": 79055.4267,
      "train_tokens_per_second": 61849.875
    },
    {
      "epoch": 0.8758745789064525,
      "grad_norm": 13.0625,
      "learning_rate": 7.974955725348743e-07,
      "loss": 0.8263,
      "num_input_tokens_seen": 4891520768,
      "step": 65910,
      "train_runtime": 79087.102,
      "train_tokens_per_second": 61849.792
    },
    {
      "epoch": 0.8760074683889144,
      "grad_norm": 14.9375,
      "learning_rate": 7.958121548370123e-07,
      "loss": 0.9118,
      "num_input_tokens_seen": 4892368384,
      "step": 65920,
      "train_runtime": 79099.442,
      "train_tokens_per_second": 61850.858
    },
    {
      "epoch": 0.8761403578713762,
      "grad_norm": 13.8125,
      "learning_rate": 7.941304421328766e-07,
      "loss": 0.7096,
      "num_input_tokens_seen": 4892850176,
      "step": 65930,
      "train_runtime": 79105.9867,
      "train_tokens_per_second": 61851.832
    },
    {
      "epoch": 0.8762732473538382,
      "grad_norm": 16.625,
      "learning_rate": 7.924504347339912e-07,
      "loss": 0.727,
      "num_input_tokens_seen": 4893076736,
      "step": 65940,
      "train_runtime": 79110.4834,
      "train_tokens_per_second": 61851.18
    },
    {
      "epoch": 0.8764061368363001,
      "grad_norm": 12.9375,
      "learning_rate": 7.907721329515572e-07,
      "loss": 0.5337,
      "num_input_tokens_seen": 4893190656,
      "step": 65950,
      "train_runtime": 79112.9945,
      "train_tokens_per_second": 61850.657
    },
    {
      "epoch": 0.876539026318762,
      "grad_norm": 30.875,
      "learning_rate": 7.890955370964714e-07,
      "loss": 0.7713,
      "num_input_tokens_seen": 4895332864,
      "step": 65960,
      "train_runtime": 79147.0882,
      "train_tokens_per_second": 61851.08
    },
    {
      "epoch": 0.8766719158012239,
      "grad_norm": 181.0,
      "learning_rate": 7.874206474793045e-07,
      "loss": 0.9155,
      "num_input_tokens_seen": 4896175360,
      "step": 65970,
      "train_runtime": 79159.4156,
      "train_tokens_per_second": 61852.091
    },
    {
      "epoch": 0.8768048052836859,
      "grad_norm": 29.0,
      "learning_rate": 7.857474644103147e-07,
      "loss": 0.7238,
      "num_input_tokens_seen": 4896659200,
      "step": 65980,
      "train_runtime": 79166.3509,
      "train_tokens_per_second": 61852.784
    },
    {
      "epoch": 0.8769376947661477,
      "grad_norm": 11.75,
      "learning_rate": 7.840759881994464e-07,
      "loss": 0.5941,
      "num_input_tokens_seen": 4896886272,
      "step": 65990,
      "train_runtime": 79171.0752,
      "train_tokens_per_second": 61851.961
    },
    {
      "epoch": 0.8770705842486096,
      "grad_norm": 11.375,
      "learning_rate": 7.82406219156322e-07,
      "loss": 0.5057,
      "num_input_tokens_seen": 4897002240,
      "step": 66000,
      "train_runtime": 79173.6642,
      "train_tokens_per_second": 61851.403
    },
    {
      "epoch": 0.8772034737310715,
      "grad_norm": 19.75,
      "learning_rate": 7.807381575902573e-07,
      "loss": 0.7915,
      "num_input_tokens_seen": 4899048192,
      "step": 66010,
      "train_runtime": 79207.2384,
      "train_tokens_per_second": 61851.016
    },
    {
      "epoch": 0.8773363632135335,
      "grad_norm": 13.125,
      "learning_rate": 7.790718038102374e-07,
      "loss": 0.863,
      "num_input_tokens_seen": 4899904768,
      "step": 66020,
      "train_runtime": 79220.104,
      "train_tokens_per_second": 61851.784
    },
    {
      "epoch": 0.8774692526959954,
      "grad_norm": 14.3125,
      "learning_rate": 7.774071581249487e-07,
      "loss": 0.7107,
      "num_input_tokens_seen": 4900380672,
      "step": 66030,
      "train_runtime": 79227.6539,
      "train_tokens_per_second": 61851.897
    },
    {
      "epoch": 0.8776021421784573,
      "grad_norm": 32.75,
      "learning_rate": 7.757442208427468e-07,
      "loss": 0.5894,
      "num_input_tokens_seen": 4900613632,
      "step": 66040,
      "train_runtime": 79232.1365,
      "train_tokens_per_second": 61851.338
    },
    {
      "epoch": 0.8777350316609192,
      "grad_norm": 24.625,
      "learning_rate": 7.740829922716753e-07,
      "loss": 0.5445,
      "num_input_tokens_seen": 4900730112,
      "step": 66050,
      "train_runtime": 79236.0621,
      "train_tokens_per_second": 61849.743
    },
    {
      "epoch": 0.8778679211433811,
      "grad_norm": 19.25,
      "learning_rate": 7.724234727194657e-07,
      "loss": 0.9266,
      "num_input_tokens_seen": 4902842368,
      "step": 66060,
      "train_runtime": 79269.6747,
      "train_tokens_per_second": 61850.164
    },
    {
      "epoch": 0.878000810625843,
      "grad_norm": 54.5,
      "learning_rate": 7.707656624935244e-07,
      "loss": 0.8917,
      "num_input_tokens_seen": 4903667968,
      "step": 66070,
      "train_runtime": 79284.24,
      "train_tokens_per_second": 61849.215
    },
    {
      "epoch": 0.8781337001083049,
      "grad_norm": 18.25,
      "learning_rate": 7.691095619009514e-07,
      "loss": 0.6633,
      "num_input_tokens_seen": 4904115968,
      "step": 66080,
      "train_runtime": 79291.2975,
      "train_tokens_per_second": 61849.36
    },
    {
      "epoch": 0.8782665895907669,
      "grad_norm": 108.0,
      "learning_rate": 7.67455171248519e-07,
      "loss": 0.647,
      "num_input_tokens_seen": 4904349696,
      "step": 66090,
      "train_runtime": 79296.2087,
      "train_tokens_per_second": 61848.476
    },
    {
      "epoch": 0.8783994790732288,
      "grad_norm": 9.25,
      "learning_rate": 7.65802490842692e-07,
      "loss": 0.4596,
      "num_input_tokens_seen": 4904462080,
      "step": 66100,
      "train_runtime": 79301.2115,
      "train_tokens_per_second": 61845.992
    },
    {
      "epoch": 0.8785323685556906,
      "grad_norm": 28.875,
      "learning_rate": 7.641515209896145e-07,
      "loss": 0.8828,
      "num_input_tokens_seen": 4906540288,
      "step": 66110,
      "train_runtime": 79336.3383,
      "train_tokens_per_second": 61844.804
    },
    {
      "epoch": 0.8786652580381525,
      "grad_norm": 18.75,
      "learning_rate": 7.625022619951117e-07,
      "loss": 0.9023,
      "num_input_tokens_seen": 4907406592,
      "step": 66120,
      "train_runtime": 79349.5322,
      "train_tokens_per_second": 61845.438
    },
    {
      "epoch": 0.8787981475206145,
      "grad_norm": 56.0,
      "learning_rate": 7.608547141646983e-07,
      "loss": 0.7682,
      "num_input_tokens_seen": 4907857920,
      "step": 66130,
      "train_runtime": 79358.5951,
      "train_tokens_per_second": 61844.063
    },
    {
      "epoch": 0.8789310370030764,
      "grad_norm": 31.25,
      "learning_rate": 7.592088778035633e-07,
      "loss": 0.5992,
      "num_input_tokens_seen": 4908080896,
      "step": 66140,
      "train_runtime": 79363.0205,
      "train_tokens_per_second": 61843.424
    },
    {
      "epoch": 0.8790639264855383,
      "grad_norm": 12.9375,
      "learning_rate": 7.575647532165842e-07,
      "loss": 0.5038,
      "num_input_tokens_seen": 4908194816,
      "step": 66150,
      "train_runtime": 79366.9645,
      "train_tokens_per_second": 61841.786
    },
    {
      "epoch": 0.8791968159680003,
      "grad_norm": 25.375,
      "learning_rate": 7.559223407083227e-07,
      "loss": 0.8193,
      "num_input_tokens_seen": 4910264576,
      "step": 66160,
      "train_runtime": 79401.0725,
      "train_tokens_per_second": 61841.288
    },
    {
      "epoch": 0.8793297054504621,
      "grad_norm": 31.875,
      "learning_rate": 7.542816405830189e-07,
      "loss": 0.9507,
      "num_input_tokens_seen": 4911115776,
      "step": 66170,
      "train_runtime": 79415.6205,
      "train_tokens_per_second": 61840.678
    },
    {
      "epoch": 0.879462594932924,
      "grad_norm": 32.5,
      "learning_rate": 7.526426531445985e-07,
      "loss": 0.766,
      "num_input_tokens_seen": 4911592448,
      "step": 66180,
      "train_runtime": 79423.3388,
      "train_tokens_per_second": 61840.67
    },
    {
      "epoch": 0.8795954844153859,
      "grad_norm": 13.0625,
      "learning_rate": 7.51005378696672e-07,
      "loss": 0.6441,
      "num_input_tokens_seen": 4911826688,
      "step": 66190,
      "train_runtime": 79427.6258,
      "train_tokens_per_second": 61840.281
    },
    {
      "epoch": 0.8797283738978479,
      "grad_norm": 64.0,
      "learning_rate": 7.493698175425223e-07,
      "loss": 0.5548,
      "num_input_tokens_seen": 4911943936,
      "step": 66200,
      "train_runtime": 79431.6663,
      "train_tokens_per_second": 61838.611
    },
    {
      "epoch": 0.8798612633803098,
      "grad_norm": 27.75,
      "learning_rate": 7.477359699851328e-07,
      "loss": 0.8538,
      "num_input_tokens_seen": 4914085120,
      "step": 66210,
      "train_runtime": 79464.9907,
      "train_tokens_per_second": 61839.624
    },
    {
      "epoch": 0.8799941528627717,
      "grad_norm": 34.25,
      "learning_rate": 7.461038363271511e-07,
      "loss": 0.9183,
      "num_input_tokens_seen": 4914954496,
      "step": 66220,
      "train_runtime": 79477.7243,
      "train_tokens_per_second": 61840.655
    },
    {
      "epoch": 0.8801270423452335,
      "grad_norm": 18.875,
      "learning_rate": 7.444734168709189e-07,
      "loss": 0.7778,
      "num_input_tokens_seen": 4915416320,
      "step": 66230,
      "train_runtime": 79484.4426,
      "train_tokens_per_second": 61841.238
    },
    {
      "epoch": 0.8802599318276955,
      "grad_norm": 80.0,
      "learning_rate": 7.428447119184556e-07,
      "loss": 0.6113,
      "num_input_tokens_seen": 4915635712,
      "step": 66240,
      "train_runtime": 79488.0017,
      "train_tokens_per_second": 61841.229
    },
    {
      "epoch": 0.8803928213101574,
      "grad_norm": 27.375,
      "learning_rate": 7.412177217714655e-07,
      "loss": 0.5749,
      "num_input_tokens_seen": 4915750656,
      "step": 66250,
      "train_runtime": 79492.5492,
      "train_tokens_per_second": 61839.137
    },
    {
      "epoch": 0.8805257107926193,
      "grad_norm": 42.75,
      "learning_rate": 7.395924467313343e-07,
      "loss": 0.8908,
      "num_input_tokens_seen": 4917749248,
      "step": 66260,
      "train_runtime": 79525.6605,
      "train_tokens_per_second": 61838.521
    },
    {
      "epoch": 0.8806586002750812,
      "grad_norm": 18.375,
      "learning_rate": 7.379688870991253e-07,
      "loss": 0.847,
      "num_input_tokens_seen": 4918591488,
      "step": 66270,
      "train_runtime": 79538.3432,
      "train_tokens_per_second": 61839.25
    },
    {
      "epoch": 0.8807914897575432,
      "grad_norm": 26.125,
      "learning_rate": 7.363470431755959e-07,
      "loss": 0.7161,
      "num_input_tokens_seen": 4919061760,
      "step": 66280,
      "train_runtime": 79545.6525,
      "train_tokens_per_second": 61839.48
    },
    {
      "epoch": 0.880924379240005,
      "grad_norm": 11.5,
      "learning_rate": 7.347269152611736e-07,
      "loss": 0.6148,
      "num_input_tokens_seen": 4919281920,
      "step": 66290,
      "train_runtime": 79549.7413,
      "train_tokens_per_second": 61839.069
    },
    {
      "epoch": 0.8810572687224669,
      "grad_norm": 9.0,
      "learning_rate": 7.331085036559715e-07,
      "loss": 0.5887,
      "num_input_tokens_seen": 4919394560,
      "step": 66300,
      "train_runtime": 79553.5993,
      "train_tokens_per_second": 61837.486
    },
    {
      "epoch": 0.8811901582049289,
      "grad_norm": 14.75,
      "learning_rate": 7.314918086597911e-07,
      "loss": 0.7861,
      "num_input_tokens_seen": 4921384704,
      "step": 66310,
      "train_runtime": 79588.3229,
      "train_tokens_per_second": 61835.512
    },
    {
      "epoch": 0.8813230476873908,
      "grad_norm": 22.0,
      "learning_rate": 7.298768305721027e-07,
      "loss": 0.8882,
      "num_input_tokens_seen": 4922208512,
      "step": 66320,
      "train_runtime": 79601.1131,
      "train_tokens_per_second": 61835.926
    },
    {
      "epoch": 0.8814559371698527,
      "grad_norm": 49.5,
      "learning_rate": 7.282635696920748e-07,
      "loss": 0.7445,
      "num_input_tokens_seen": 4922685440,
      "step": 66330,
      "train_runtime": 79608.5428,
      "train_tokens_per_second": 61836.146
    },
    {
      "epoch": 0.8815888266523146,
      "grad_norm": 9.1875,
      "learning_rate": 7.266520263185451e-07,
      "loss": 0.6402,
      "num_input_tokens_seen": 4922910720,
      "step": 66340,
      "train_runtime": 79613.5054,
      "train_tokens_per_second": 61835.121
    },
    {
      "epoch": 0.8817217161347766,
      "grad_norm": 8.625,
      "learning_rate": 7.25042200750038e-07,
      "loss": 0.5578,
      "num_input_tokens_seen": 4923024640,
      "step": 66350,
      "train_runtime": 79618.1367,
      "train_tokens_per_second": 61832.955
    },
    {
      "epoch": 0.8818546056172384,
      "grad_norm": 50.5,
      "learning_rate": 7.234340932847606e-07,
      "loss": 0.7917,
      "num_input_tokens_seen": 4925043456,
      "step": 66360,
      "train_runtime": 79652.4203,
      "train_tokens_per_second": 61831.686
    },
    {
      "epoch": 0.8819874950997003,
      "grad_norm": 21.375,
      "learning_rate": 7.21827704220599e-07,
      "loss": 0.9334,
      "num_input_tokens_seen": 4925920256,
      "step": 66370,
      "train_runtime": 79665.3583,
      "train_tokens_per_second": 61832.651
    },
    {
      "epoch": 0.8821203845821622,
      "grad_norm": 29.75,
      "learning_rate": 7.202230338551253e-07,
      "loss": 0.6923,
      "num_input_tokens_seen": 4926363904,
      "step": 66380,
      "train_runtime": 79672.3527,
      "train_tokens_per_second": 61832.791
    },
    {
      "epoch": 0.8822532740646242,
      "grad_norm": 10.5625,
      "learning_rate": 7.186200824855849e-07,
      "loss": 0.6038,
      "num_input_tokens_seen": 4926583040,
      "step": 66390,
      "train_runtime": 79676.3395,
      "train_tokens_per_second": 61832.447
    },
    {
      "epoch": 0.8823861635470861,
      "grad_norm": 17.625,
      "learning_rate": 7.170188504089171e-07,
      "loss": 0.4775,
      "num_input_tokens_seen": 4926695936,
      "step": 66400,
      "train_runtime": 79681.5618,
      "train_tokens_per_second": 61829.811
    },
    {
      "epoch": 0.882519053029548,
      "grad_norm": 42.75,
      "learning_rate": 7.154193379217311e-07,
      "loss": 0.8137,
      "num_input_tokens_seen": 4928743680,
      "step": 66410,
      "train_runtime": 79715.7713,
      "train_tokens_per_second": 61828.966
    },
    {
      "epoch": 0.8826519425120098,
      "grad_norm": 33.75,
      "learning_rate": 7.138215453203235e-07,
      "loss": 0.8749,
      "num_input_tokens_seen": 4929559552,
      "step": 66420,
      "train_runtime": 79728.4219,
      "train_tokens_per_second": 61829.388
    },
    {
      "epoch": 0.8827848319944718,
      "grad_norm": 15.5,
      "learning_rate": 7.122254729006706e-07,
      "loss": 0.7175,
      "num_input_tokens_seen": 4930002688,
      "step": 66430,
      "train_runtime": 79735.164,
      "train_tokens_per_second": 61829.718
    },
    {
      "epoch": 0.8829177214769337,
      "grad_norm": 25.625,
      "learning_rate": 7.106311209584316e-07,
      "loss": 0.6394,
      "num_input_tokens_seen": 4930231552,
      "step": 66440,
      "train_runtime": 79739.1198,
      "train_tokens_per_second": 61829.521
    },
    {
      "epoch": 0.8830506109593956,
      "grad_norm": 7.8125,
      "learning_rate": 7.090384897889469e-07,
      "loss": 0.5204,
      "num_input_tokens_seen": 4930346752,
      "step": 66450,
      "train_runtime": 79742.7057,
      "train_tokens_per_second": 61828.185
    },
    {
      "epoch": 0.8831835004418576,
      "grad_norm": 57.0,
      "learning_rate": 7.07447579687236e-07,
      "loss": 0.8042,
      "num_input_tokens_seen": 4932380928,
      "step": 66460,
      "train_runtime": 79774.8736,
      "train_tokens_per_second": 61828.753
    },
    {
      "epoch": 0.8833163899243195,
      "grad_norm": 22.25,
      "learning_rate": 7.058583909479999e-07,
      "loss": 0.8836,
      "num_input_tokens_seen": 4933217536,
      "step": 66470,
      "train_runtime": 79788.9968,
      "train_tokens_per_second": 61828.294
    },
    {
      "epoch": 0.8834492794067813,
      "grad_norm": 37.75,
      "learning_rate": 7.04270923865622e-07,
      "loss": 0.7177,
      "num_input_tokens_seen": 4933677568,
      "step": 66480,
      "train_runtime": 79795.8334,
      "train_tokens_per_second": 61828.762
    },
    {
      "epoch": 0.8835821688892432,
      "grad_norm": 26.875,
      "learning_rate": 7.026851787341682e-07,
      "loss": 0.6225,
      "num_input_tokens_seen": 4933909504,
      "step": 66490,
      "train_runtime": 79799.9162,
      "train_tokens_per_second": 61828.505
    },
    {
      "epoch": 0.8837150583717052,
      "grad_norm": 12.0625,
      "learning_rate": 7.011011558473813e-07,
      "loss": 0.5791,
      "num_input_tokens_seen": 4934025984,
      "step": 66500,
      "train_runtime": 79803.4457,
      "train_tokens_per_second": 61827.23
    },
    {
      "epoch": 0.8838479478541671,
      "grad_norm": 27.125,
      "learning_rate": 6.995188554986909e-07,
      "loss": 0.7469,
      "num_input_tokens_seen": 4936070912,
      "step": 66510,
      "train_runtime": 79836.1248,
      "train_tokens_per_second": 61827.536
    },
    {
      "epoch": 0.883980837336629,
      "grad_norm": 16.25,
      "learning_rate": 6.979382779811972e-07,
      "loss": 0.9168,
      "num_input_tokens_seen": 4936876288,
      "step": 66520,
      "train_runtime": 79850.3896,
      "train_tokens_per_second": 61826.577
    },
    {
      "epoch": 0.884113726819091,
      "grad_norm": 28.125,
      "learning_rate": 6.96359423587697e-07,
      "loss": 0.7561,
      "num_input_tokens_seen": 4937296384,
      "step": 66530,
      "train_runtime": 79857.0574,
      "train_tokens_per_second": 61826.676
    },
    {
      "epoch": 0.8842466163015528,
      "grad_norm": 19.75,
      "learning_rate": 6.947822926106529e-07,
      "loss": 0.581,
      "num_input_tokens_seen": 4937518848,
      "step": 66540,
      "train_runtime": 79861.6402,
      "train_tokens_per_second": 61825.913
    },
    {
      "epoch": 0.8843795057840147,
      "grad_norm": 18.125,
      "learning_rate": 6.932068853422158e-07,
      "loss": 0.6013,
      "num_input_tokens_seen": 4937629696,
      "step": 66550,
      "train_runtime": 79865.2948,
      "train_tokens_per_second": 61824.472
    },
    {
      "epoch": 0.8845123952664766,
      "grad_norm": 29.0,
      "learning_rate": 6.916332020742156e-07,
      "loss": 0.789,
      "num_input_tokens_seen": 4939676928,
      "step": 66560,
      "train_runtime": 79898.6234,
      "train_tokens_per_second": 61824.306
    },
    {
      "epoch": 0.8846452847489386,
      "grad_norm": 19.5,
      "learning_rate": 6.900612430981635e-07,
      "loss": 0.9306,
      "num_input_tokens_seen": 4940509440,
      "step": 66570,
      "train_runtime": 79912.5695,
      "train_tokens_per_second": 61823.934
    },
    {
      "epoch": 0.8847781742314005,
      "grad_norm": 31.625,
      "learning_rate": 6.884910087052532e-07,
      "loss": 0.6792,
      "num_input_tokens_seen": 4940988416,
      "step": 66580,
      "train_runtime": 79919.5649,
      "train_tokens_per_second": 61824.516
    },
    {
      "epoch": 0.8849110637138624,
      "grad_norm": 17.625,
      "learning_rate": 6.86922499186351e-07,
      "loss": 0.6835,
      "num_input_tokens_seen": 4941213952,
      "step": 66590,
      "train_runtime": 79923.9832,
      "train_tokens_per_second": 61823.92
    },
    {
      "epoch": 0.8850439531963242,
      "grad_norm": 13.625,
      "learning_rate": 6.853557148320156e-07,
      "loss": 0.4595,
      "num_input_tokens_seen": 4941328128,
      "step": 66600,
      "train_runtime": 79927.4445,
      "train_tokens_per_second": 61822.671
    },
    {
      "epoch": 0.8851768426787862,
      "grad_norm": 16.875,
      "learning_rate": 6.837906559324748e-07,
      "loss": 0.8376,
      "num_input_tokens_seen": 4943334912,
      "step": 66610,
      "train_runtime": 79961.484,
      "train_tokens_per_second": 61821.45
    },
    {
      "epoch": 0.8853097321612481,
      "grad_norm": 24.875,
      "learning_rate": 6.822273227776443e-07,
      "loss": 0.8668,
      "num_input_tokens_seen": 4944146944,
      "step": 66620,
      "train_runtime": 79974.1083,
      "train_tokens_per_second": 61821.845
    },
    {
      "epoch": 0.88544262164371,
      "grad_norm": 47.75,
      "learning_rate": 6.806657156571195e-07,
      "loss": 0.7758,
      "num_input_tokens_seen": 4944632320,
      "step": 66630,
      "train_runtime": 79981.0293,
      "train_tokens_per_second": 61822.564
    },
    {
      "epoch": 0.8855755111261719,
      "grad_norm": 26.375,
      "learning_rate": 6.791058348601675e-07,
      "loss": 0.6398,
      "num_input_tokens_seen": 4944871168,
      "step": 66640,
      "train_runtime": 79986.2656,
      "train_tokens_per_second": 61821.503
    },
    {
      "epoch": 0.8857084006086339,
      "grad_norm": 29.375,
      "learning_rate": 6.775476806757508e-07,
      "loss": 0.5991,
      "num_input_tokens_seen": 4944988416,
      "step": 66650,
      "train_runtime": 79989.2153,
      "train_tokens_per_second": 61820.689
    },
    {
      "epoch": 0.8858412900910957,
      "grad_norm": 27.375,
      "learning_rate": 6.759912533924973e-07,
      "loss": 0.8142,
      "num_input_tokens_seen": 4947006720,
      "step": 66660,
      "train_runtime": 80023.0425,
      "train_tokens_per_second": 61819.778
    },
    {
      "epoch": 0.8859741795735576,
      "grad_norm": 33.25,
      "learning_rate": 6.74436553298724e-07,
      "loss": 0.8774,
      "num_input_tokens_seen": 4947856640,
      "step": 66670,
      "train_runtime": 80036.0755,
      "train_tokens_per_second": 61820.331
    },
    {
      "epoch": 0.8861070690560195,
      "grad_norm": 16.875,
      "learning_rate": 6.72883580682424e-07,
      "loss": 0.7619,
      "num_input_tokens_seen": 4948322048,
      "step": 66680,
      "train_runtime": 80043.2729,
      "train_tokens_per_second": 61820.586
    },
    {
      "epoch": 0.8862399585384815,
      "grad_norm": 7.375,
      "learning_rate": 6.713323358312729e-07,
      "loss": 0.6086,
      "num_input_tokens_seen": 4948565504,
      "step": 66690,
      "train_runtime": 80047.7417,
      "train_tokens_per_second": 61820.176
    },
    {
      "epoch": 0.8863728480209434,
      "grad_norm": 18.0,
      "learning_rate": 6.697828190326261e-07,
      "loss": 0.5134,
      "num_input_tokens_seen": 4948684288,
      "step": 66700,
      "train_runtime": 80051.204,
      "train_tokens_per_second": 61818.986
    },
    {
      "epoch": 0.8865057375034053,
      "grad_norm": 24.5,
      "learning_rate": 6.682350305735142e-07,
      "loss": 0.8338,
      "num_input_tokens_seen": 4950804224,
      "step": 66710,
      "train_runtime": 80086.0705,
      "train_tokens_per_second": 61818.543
    },
    {
      "epoch": 0.8866386269858672,
      "grad_norm": 12.25,
      "learning_rate": 6.666889707406531e-07,
      "loss": 0.8594,
      "num_input_tokens_seen": 4951670784,
      "step": 66720,
      "train_runtime": 80099.881,
      "train_tokens_per_second": 61818.703
    },
    {
      "epoch": 0.8867715164683291,
      "grad_norm": 19.75,
      "learning_rate": 6.651446398204365e-07,
      "loss": 0.6496,
      "num_input_tokens_seen": 4952131584,
      "step": 66730,
      "train_runtime": 80107.1601,
      "train_tokens_per_second": 61818.838
    },
    {
      "epoch": 0.886904405950791,
      "grad_norm": 13.9375,
      "learning_rate": 6.636020380989383e-07,
      "loss": 0.6532,
      "num_input_tokens_seen": 4952368640,
      "step": 66740,
      "train_runtime": 80111.6229,
      "train_tokens_per_second": 61818.354
    },
    {
      "epoch": 0.8870372954332529,
      "grad_norm": 11.625,
      "learning_rate": 6.620611658619103e-07,
      "loss": 0.4852,
      "num_input_tokens_seen": 4952484096,
      "step": 66750,
      "train_runtime": 80115.0963,
      "train_tokens_per_second": 61817.115
    },
    {
      "epoch": 0.8871701849157149,
      "grad_norm": 16.875,
      "learning_rate": 6.605220233947874e-07,
      "loss": 0.7999,
      "num_input_tokens_seen": 4954538752,
      "step": 66760,
      "train_runtime": 80149.7643,
      "train_tokens_per_second": 61816.012
    },
    {
      "epoch": 0.8873030743981768,
      "grad_norm": 57.5,
      "learning_rate": 6.589846109826814e-07,
      "loss": 0.9705,
      "num_input_tokens_seen": 4955368192,
      "step": 66770,
      "train_runtime": 80163.0608,
      "train_tokens_per_second": 61816.105
    },
    {
      "epoch": 0.8874359638806386,
      "grad_norm": 35.25,
      "learning_rate": 6.574489289103847e-07,
      "loss": 0.7341,
      "num_input_tokens_seen": 4955839232,
      "step": 66780,
      "train_runtime": 80169.9414,
      "train_tokens_per_second": 61816.675
    },
    {
      "epoch": 0.8875688533631005,
      "grad_norm": 28.125,
      "learning_rate": 6.559149774623674e-07,
      "loss": 0.6319,
      "num_input_tokens_seen": 4956066048,
      "step": 66790,
      "train_runtime": 80173.9025,
      "train_tokens_per_second": 61816.45
    },
    {
      "epoch": 0.8877017428455625,
      "grad_norm": 13.5625,
      "learning_rate": 6.543827569227801e-07,
      "loss": 0.5025,
      "num_input_tokens_seen": 4956179968,
      "step": 66800,
      "train_runtime": 80178.3178,
      "train_tokens_per_second": 61814.467
    },
    {
      "epoch": 0.8878346323280244,
      "grad_norm": 41.5,
      "learning_rate": 6.528522675754534e-07,
      "loss": 0.8129,
      "num_input_tokens_seen": 4958354944,
      "step": 66810,
      "train_runtime": 80214.5652,
      "train_tokens_per_second": 61813.649
    },
    {
      "epoch": 0.8879675218104863,
      "grad_norm": 24.5,
      "learning_rate": 6.513235097038983e-07,
      "loss": 0.9318,
      "num_input_tokens_seen": 4959203584,
      "step": 66820,
      "train_runtime": 80227.6468,
      "train_tokens_per_second": 61814.147
    },
    {
      "epoch": 0.8881004112929483,
      "grad_norm": 17.25,
      "learning_rate": 6.497964835913029e-07,
      "loss": 0.7703,
      "num_input_tokens_seen": 4959688448,
      "step": 66830,
      "train_runtime": 80234.8025,
      "train_tokens_per_second": 61814.678
    },
    {
      "epoch": 0.8882333007754102,
      "grad_norm": 71.5,
      "learning_rate": 6.482711895205318e-07,
      "loss": 0.6126,
      "num_input_tokens_seen": 4959920896,
      "step": 66840,
      "train_runtime": 80238.8556,
      "train_tokens_per_second": 61814.452
    },
    {
      "epoch": 0.888366190257872,
      "grad_norm": 25.5,
      "learning_rate": 6.46747627774138e-07,
      "loss": 0.4933,
      "num_input_tokens_seen": 4960036864,
      "step": 66850,
      "train_runtime": 80242.3614,
      "train_tokens_per_second": 61813.197
    },
    {
      "epoch": 0.8884990797403339,
      "grad_norm": 11.5,
      "learning_rate": 6.452257986343436e-07,
      "loss": 0.7817,
      "num_input_tokens_seen": 4962144256,
      "step": 66860,
      "train_runtime": 80275.4298,
      "train_tokens_per_second": 61813.986
    },
    {
      "epoch": 0.8886319692227959,
      "grad_norm": 13.8125,
      "learning_rate": 6.437057023830551e-07,
      "loss": 0.9258,
      "num_input_tokens_seen": 4962966528,
      "step": 66870,
      "train_runtime": 80288.5102,
      "train_tokens_per_second": 61814.156
    },
    {
      "epoch": 0.8887648587052578,
      "grad_norm": 145.0,
      "learning_rate": 6.421873393018585e-07,
      "loss": 0.7173,
      "num_input_tokens_seen": 4963433728,
      "step": 66880,
      "train_runtime": 80297.1579,
      "train_tokens_per_second": 61813.318
    },
    {
      "epoch": 0.8888977481877197,
      "grad_norm": 9.8125,
      "learning_rate": 6.406707096720111e-07,
      "loss": 0.5618,
      "num_input_tokens_seen": 4963666688,
      "step": 66890,
      "train_runtime": 80301.3769,
      "train_tokens_per_second": 61812.971
    },
    {
      "epoch": 0.8890306376701816,
      "grad_norm": 42.75,
      "learning_rate": 6.391558137744636e-07,
      "loss": 0.5463,
      "num_input_tokens_seen": 4963779840,
      "step": 66900,
      "train_runtime": 80304.8768,
      "train_tokens_per_second": 61811.686
    },
    {
      "epoch": 0.8891635271526435,
      "grad_norm": 14.625,
      "learning_rate": 6.376426518898282e-07,
      "loss": 0.8254,
      "num_input_tokens_seen": 4965809408,
      "step": 66910,
      "train_runtime": 80337.8071,
      "train_tokens_per_second": 61811.613
    },
    {
      "epoch": 0.8892964166351054,
      "grad_norm": 34.25,
      "learning_rate": 6.361312242984119e-07,
      "loss": 0.9214,
      "num_input_tokens_seen": 4966681344,
      "step": 66920,
      "train_runtime": 80350.9332,
      "train_tokens_per_second": 61812.367
    },
    {
      "epoch": 0.8894293061175673,
      "grad_norm": 23.0,
      "learning_rate": 6.346215312801896e-07,
      "loss": 0.6998,
      "num_input_tokens_seen": 4967164928,
      "step": 66930,
      "train_runtime": 80358.2136,
      "train_tokens_per_second": 61812.784
    },
    {
      "epoch": 0.8895621956000292,
      "grad_norm": 11.75,
      "learning_rate": 6.331135731148186e-07,
      "loss": 0.6038,
      "num_input_tokens_seen": 4967403776,
      "step": 66940,
      "train_runtime": 80364.1033,
      "train_tokens_per_second": 61811.226
    },
    {
      "epoch": 0.8896950850824912,
      "grad_norm": 9.875,
      "learning_rate": 6.316073500816366e-07,
      "loss": 0.5707,
      "num_input_tokens_seen": 4967518464,
      "step": 66950,
      "train_runtime": 80367.4254,
      "train_tokens_per_second": 61810.098
    },
    {
      "epoch": 0.8898279745649531,
      "grad_norm": 35.75,
      "learning_rate": 6.301028624596539e-07,
      "loss": 0.8096,
      "num_input_tokens_seen": 4969651968,
      "step": 66960,
      "train_runtime": 80400.1514,
      "train_tokens_per_second": 61811.475
    },
    {
      "epoch": 0.8899608640474149,
      "grad_norm": 20.25,
      "learning_rate": 6.286001105275697e-07,
      "loss": 0.9148,
      "num_input_tokens_seen": 4970474240,
      "step": 66970,
      "train_runtime": 80412.8376,
      "train_tokens_per_second": 61811.949
    },
    {
      "epoch": 0.8900937535298769,
      "grad_norm": 26.625,
      "learning_rate": 6.270990945637489e-07,
      "loss": 0.7496,
      "num_input_tokens_seen": 4970942976,
      "step": 66980,
      "train_runtime": 80419.7776,
      "train_tokens_per_second": 61812.444
    },
    {
      "epoch": 0.8902266430123388,
      "grad_norm": 11.4375,
      "learning_rate": 6.255998148462449e-07,
      "loss": 0.6134,
      "num_input_tokens_seen": 4971170048,
      "step": 66990,
      "train_runtime": 80424.8978,
      "train_tokens_per_second": 61811.332
    },
    {
      "epoch": 0.8903595324948007,
      "grad_norm": 15.6875,
      "learning_rate": 6.241022716527856e-07,
      "loss": 0.5301,
      "num_input_tokens_seen": 4971281664,
      "step": 67000,
      "train_runtime": 80428.2442,
      "train_tokens_per_second": 61810.148
    },
    {
      "epoch": 0.8904924219772626,
      "grad_norm": 16.0,
      "learning_rate": 6.226064652607766e-07,
      "loss": 0.7582,
      "num_input_tokens_seen": 4973456896,
      "step": 67010,
      "train_runtime": 80463.5231,
      "train_tokens_per_second": 61810.081
    },
    {
      "epoch": 0.8906253114597246,
      "grad_norm": 23.0,
      "learning_rate": 6.211123959473042e-07,
      "loss": 0.8994,
      "num_input_tokens_seen": 4974334208,
      "step": 67020,
      "train_runtime": 80476.7532,
      "train_tokens_per_second": 61810.821
    },
    {
      "epoch": 0.8907582009421864,
      "grad_norm": 14.75,
      "learning_rate": 6.196200639891303e-07,
      "loss": 0.8192,
      "num_input_tokens_seen": 4974820352,
      "step": 67030,
      "train_runtime": 80485.6362,
      "train_tokens_per_second": 61810.04
    },
    {
      "epoch": 0.8908910904246483,
      "grad_norm": 14.8125,
      "learning_rate": 6.181294696626949e-07,
      "loss": 0.6466,
      "num_input_tokens_seen": 4975059456,
      "step": 67040,
      "train_runtime": 80489.8015,
      "train_tokens_per_second": 61809.812
    },
    {
      "epoch": 0.8910239799071102,
      "grad_norm": 11.9375,
      "learning_rate": 6.166406132441194e-07,
      "loss": 0.5515,
      "num_input_tokens_seen": 4975172864,
      "step": 67050,
      "train_runtime": 80493.427,
      "train_tokens_per_second": 61808.436
    },
    {
      "epoch": 0.8911568693895722,
      "grad_norm": 15.875,
      "learning_rate": 6.151534950092008e-07,
      "loss": 0.8655,
      "num_input_tokens_seen": 4977149184,
      "step": 67060,
      "train_runtime": 80525.6212,
      "train_tokens_per_second": 61808.268
    },
    {
      "epoch": 0.8912897588720341,
      "grad_norm": 30.125,
      "learning_rate": 6.136681152334134e-07,
      "loss": 0.9181,
      "num_input_tokens_seen": 4978003200,
      "step": 67070,
      "train_runtime": 80538.8224,
      "train_tokens_per_second": 61808.741
    },
    {
      "epoch": 0.891422648354496,
      "grad_norm": 14.875,
      "learning_rate": 6.121844741919114e-07,
      "loss": 0.7209,
      "num_input_tokens_seen": 4978476032,
      "step": 67080,
      "train_runtime": 80545.5585,
      "train_tokens_per_second": 61809.442
    },
    {
      "epoch": 0.8915555378369578,
      "grad_norm": 29.25,
      "learning_rate": 6.107025721595261e-07,
      "loss": 0.6597,
      "num_input_tokens_seen": 4978704640,
      "step": 67090,
      "train_runtime": 80549.6647,
      "train_tokens_per_second": 61809.129
    },
    {
      "epoch": 0.8916884273194198,
      "grad_norm": 11.375,
      "learning_rate": 6.092224094107679e-07,
      "loss": 0.5877,
      "num_input_tokens_seen": 4978818048,
      "step": 67100,
      "train_runtime": 80553.1731,
      "train_tokens_per_second": 61807.845
    },
    {
      "epoch": 0.8918213168018817,
      "grad_norm": 27.0,
      "learning_rate": 6.077439862198208e-07,
      "loss": 0.8247,
      "num_input_tokens_seen": 4980812288,
      "step": 67110,
      "train_runtime": 80587.401,
      "train_tokens_per_second": 61806.34
    },
    {
      "epoch": 0.8919542062843436,
      "grad_norm": 72.5,
      "learning_rate": 6.062673028605515e-07,
      "loss": 0.9001,
      "num_input_tokens_seen": 4981665280,
      "step": 67120,
      "train_runtime": 80599.8811,
      "train_tokens_per_second": 61807.353
    },
    {
      "epoch": 0.8920870957668056,
      "grad_norm": 25.125,
      "learning_rate": 6.04792359606502e-07,
      "loss": 0.7671,
      "num_input_tokens_seen": 4982119936,
      "step": 67130,
      "train_runtime": 80606.5867,
      "train_tokens_per_second": 61807.851
    },
    {
      "epoch": 0.8922199852492675,
      "grad_norm": 45.5,
      "learning_rate": 6.033191567308915e-07,
      "loss": 0.6367,
      "num_input_tokens_seen": 4982346496,
      "step": 67140,
      "train_runtime": 80610.39,
      "train_tokens_per_second": 61807.746
    },
    {
      "epoch": 0.8923528747317293,
      "grad_norm": 11.8125,
      "learning_rate": 6.01847694506621e-07,
      "loss": 0.4937,
      "num_input_tokens_seen": 4982458880,
      "step": 67150,
      "train_runtime": 80613.8546,
      "train_tokens_per_second": 61806.484
    },
    {
      "epoch": 0.8924857642141912,
      "grad_norm": 35.25,
      "learning_rate": 6.003779732062598e-07,
      "loss": 0.7889,
      "num_input_tokens_seen": 4984548864,
      "step": 67160,
      "train_runtime": 80647.5565,
      "train_tokens_per_second": 61806.57
    },
    {
      "epoch": 0.8926186536966532,
      "grad_norm": 64.5,
      "learning_rate": 5.98909993102067e-07,
      "loss": 0.9374,
      "num_input_tokens_seen": 4985403648,
      "step": 67170,
      "train_runtime": 80661.4118,
      "train_tokens_per_second": 61806.551
    },
    {
      "epoch": 0.8927515431791151,
      "grad_norm": 18.25,
      "learning_rate": 5.974437544659684e-07,
      "loss": 0.7424,
      "num_input_tokens_seen": 4985859584,
      "step": 67180,
      "train_runtime": 80668.2262,
      "train_tokens_per_second": 61806.982
    },
    {
      "epoch": 0.892884432661577,
      "grad_norm": 14.5625,
      "learning_rate": 5.959792575695744e-07,
      "loss": 0.6252,
      "num_input_tokens_seen": 4986083840,
      "step": 67190,
      "train_runtime": 80672.3572,
      "train_tokens_per_second": 61806.597
    },
    {
      "epoch": 0.893017322144039,
      "grad_norm": 20.5,
      "learning_rate": 5.945165026841693e-07,
      "loss": 0.4932,
      "num_input_tokens_seen": 4986201344,
      "step": 67200,
      "train_runtime": 80675.6199,
      "train_tokens_per_second": 61805.554
    },
    {
      "epoch": 0.8931502116265008,
      "grad_norm": 46.5,
      "learning_rate": 5.930554900807117e-07,
      "loss": 0.9157,
      "num_input_tokens_seen": 4988324096,
      "step": 67210,
      "train_runtime": 80709.0407,
      "train_tokens_per_second": 61806.262
    },
    {
      "epoch": 0.8932831011089627,
      "grad_norm": 18.875,
      "learning_rate": 5.915962200298486e-07,
      "loss": 0.8931,
      "num_input_tokens_seen": 4989159168,
      "step": 67220,
      "train_runtime": 80723.2685,
      "train_tokens_per_second": 61805.713
    },
    {
      "epoch": 0.8934159905914246,
      "grad_norm": 22.375,
      "learning_rate": 5.901386928018881e-07,
      "loss": 0.6881,
      "num_input_tokens_seen": 4989624320,
      "step": 67230,
      "train_runtime": 80730.1093,
      "train_tokens_per_second": 61806.238
    },
    {
      "epoch": 0.8935488800738866,
      "grad_norm": 30.75,
      "learning_rate": 5.886829086668322e-07,
      "loss": 0.6845,
      "num_input_tokens_seen": 4989856768,
      "step": 67240,
      "train_runtime": 80734.2998,
      "train_tokens_per_second": 61805.909
    },
    {
      "epoch": 0.8936817695563485,
      "grad_norm": 31.0,
      "learning_rate": 5.872288678943472e-07,
      "loss": 0.5373,
      "num_input_tokens_seen": 4989976576,
      "step": 67250,
      "train_runtime": 80737.8444,
      "train_tokens_per_second": 61804.679
    },
    {
      "epoch": 0.8938146590388104,
      "grad_norm": 26.25,
      "learning_rate": 5.857765707537821e-07,
      "loss": 0.8694,
      "num_input_tokens_seen": 4991894016,
      "step": 67260,
      "train_runtime": 80768.9371,
      "train_tokens_per_second": 61804.627
    },
    {
      "epoch": 0.8939475485212723,
      "grad_norm": 18.875,
      "learning_rate": 5.84326017514163e-07,
      "loss": 0.9609,
      "num_input_tokens_seen": 4992715776,
      "step": 67270,
      "train_runtime": 80782.6092,
      "train_tokens_per_second": 61804.339
    },
    {
      "epoch": 0.8940804380037342,
      "grad_norm": 25.25,
      "learning_rate": 5.828772084441891e-07,
      "loss": 0.7394,
      "num_input_tokens_seen": 4993165824,
      "step": 67280,
      "train_runtime": 80789.6189,
      "train_tokens_per_second": 61804.547
    },
    {
      "epoch": 0.8942133274861961,
      "grad_norm": 26.75,
      "learning_rate": 5.814301438122438e-07,
      "loss": 0.7017,
      "num_input_tokens_seen": 4993392896,
      "step": 67290,
      "train_runtime": 80794.4492,
      "train_tokens_per_second": 61803.663
    },
    {
      "epoch": 0.894346216968658,
      "grad_norm": 11.5625,
      "learning_rate": 5.799848238863792e-07,
      "loss": 0.5956,
      "num_input_tokens_seen": 4993511936,
      "step": 67300,
      "train_runtime": 80797.8435,
      "train_tokens_per_second": 61802.54
    },
    {
      "epoch": 0.8944791064511199,
      "grad_norm": 13.75,
      "learning_rate": 5.78541248934329e-07,
      "loss": 0.7802,
      "num_input_tokens_seen": 4995585536,
      "step": 67310,
      "train_runtime": 80832.4412,
      "train_tokens_per_second": 61801.74
    },
    {
      "epoch": 0.8946119959335819,
      "grad_norm": 18.125,
      "learning_rate": 5.770994192235036e-07,
      "loss": 0.8806,
      "num_input_tokens_seen": 4996443904,
      "step": 67320,
      "train_runtime": 80844.9456,
      "train_tokens_per_second": 61802.799
    },
    {
      "epoch": 0.8947448854160438,
      "grad_norm": 23.75,
      "learning_rate": 5.756593350209872e-07,
      "loss": 0.7399,
      "num_input_tokens_seen": 4996884736,
      "step": 67330,
      "train_runtime": 80851.5408,
      "train_tokens_per_second": 61803.21
    },
    {
      "epoch": 0.8948777748985056,
      "grad_norm": 12.8125,
      "learning_rate": 5.742209965935463e-07,
      "loss": 0.6599,
      "num_input_tokens_seen": 4997110016,
      "step": 67340,
      "train_runtime": 80856.2121,
      "train_tokens_per_second": 61802.425
    },
    {
      "epoch": 0.8950106643809675,
      "grad_norm": 21.375,
      "learning_rate": 5.727844042076158e-07,
      "loss": 0.5253,
      "num_input_tokens_seen": 4997225984,
      "step": 67350,
      "train_runtime": 80859.4359,
      "train_tokens_per_second": 61801.396
    },
    {
      "epoch": 0.8951435538634295,
      "grad_norm": 38.75,
      "learning_rate": 5.713495581293127e-07,
      "loss": 0.8007,
      "num_input_tokens_seen": 4999205632,
      "step": 67360,
      "train_runtime": 80891.1502,
      "train_tokens_per_second": 61801.639
    },
    {
      "epoch": 0.8952764433458914,
      "grad_norm": 20.75,
      "learning_rate": 5.69916458624431e-07,
      "loss": 0.9902,
      "num_input_tokens_seen": 4999995648,
      "step": 67370,
      "train_runtime": 80903.3083,
      "train_tokens_per_second": 61802.116
    },
    {
      "epoch": 0.8954093328283533,
      "grad_norm": 20.75,
      "learning_rate": 5.684851059584395e-07,
      "loss": 0.6585,
      "num_input_tokens_seen": 5000430592,
      "step": 67380,
      "train_runtime": 80909.7329,
      "train_tokens_per_second": 61802.584
    },
    {
      "epoch": 0.8955422223108153,
      "grad_norm": 14.5625,
      "learning_rate": 5.670555003964817e-07,
      "loss": 0.5996,
      "num_input_tokens_seen": 5000657408,
      "step": 67390,
      "train_runtime": 80913.5731,
      "train_tokens_per_second": 61802.454
    },
    {
      "epoch": 0.8956751117932771,
      "grad_norm": 9.0,
      "learning_rate": 5.656276422033824e-07,
      "loss": 0.5281,
      "num_input_tokens_seen": 5000774400,
      "step": 67400,
      "train_runtime": 80917.1018,
      "train_tokens_per_second": 61801.205
    },
    {
      "epoch": 0.895808001275739,
      "grad_norm": 21.25,
      "learning_rate": 5.642015316436345e-07,
      "loss": 0.7872,
      "num_input_tokens_seen": 5002869248,
      "step": 67410,
      "train_runtime": 80950.0583,
      "train_tokens_per_second": 61801.923
    },
    {
      "epoch": 0.8959408907582009,
      "grad_norm": 32.25,
      "learning_rate": 5.627771689814188e-07,
      "loss": 0.8869,
      "num_input_tokens_seen": 5003710208,
      "step": 67420,
      "train_runtime": 80963.0112,
      "train_tokens_per_second": 61802.422
    },
    {
      "epoch": 0.8960737802406629,
      "grad_norm": 27.625,
      "learning_rate": 5.613545544805798e-07,
      "loss": 0.6851,
      "num_input_tokens_seen": 5004176640,
      "step": 67430,
      "train_runtime": 80970.5167,
      "train_tokens_per_second": 61802.454
    },
    {
      "epoch": 0.8962066697231248,
      "grad_norm": 8.8125,
      "learning_rate": 5.59933688404648e-07,
      "loss": 0.6204,
      "num_input_tokens_seen": 5004402944,
      "step": 67440,
      "train_runtime": 80974.883,
      "train_tokens_per_second": 61801.916
    },
    {
      "epoch": 0.8963395592055867,
      "grad_norm": 22.25,
      "learning_rate": 5.585145710168238e-07,
      "loss": 0.5417,
      "num_input_tokens_seen": 5004515328,
      "step": 67450,
      "train_runtime": 80978.2206,
      "train_tokens_per_second": 61800.757
    },
    {
      "epoch": 0.8964724486880485,
      "grad_norm": 20.875,
      "learning_rate": 5.570972025799881e-07,
      "loss": 0.8981,
      "num_input_tokens_seen": 5006531328,
      "step": 67460,
      "train_runtime": 81011.9593,
      "train_tokens_per_second": 61799.904
    },
    {
      "epoch": 0.8966053381705105,
      "grad_norm": 24.0,
      "learning_rate": 5.556815833566953e-07,
      "loss": 0.8433,
      "num_input_tokens_seen": 5007360768,
      "step": 67470,
      "train_runtime": 81024.8823,
      "train_tokens_per_second": 61800.284
    },
    {
      "epoch": 0.8967382276529724,
      "grad_norm": 17.5,
      "learning_rate": 5.542677136091734e-07,
      "loss": 0.7047,
      "num_input_tokens_seen": 5007803392,
      "step": 67480,
      "train_runtime": 81031.9181,
      "train_tokens_per_second": 61800.381
    },
    {
      "epoch": 0.8968711171354343,
      "grad_norm": 51.5,
      "learning_rate": 5.528555935993352e-07,
      "loss": 0.6652,
      "num_input_tokens_seen": 5008036864,
      "step": 67490,
      "train_runtime": 81035.8841,
      "train_tokens_per_second": 61800.237
    },
    {
      "epoch": 0.8970040066178963,
      "grad_norm": 11.0,
      "learning_rate": 5.514452235887569e-07,
      "loss": 0.4915,
      "num_input_tokens_seen": 5008154624,
      "step": 67500,
      "train_runtime": 81040.5908,
      "train_tokens_per_second": 61798.101
    },
    {
      "epoch": 0.8971368961003582,
      "grad_norm": 21.625,
      "learning_rate": 5.500366038387006e-07,
      "loss": 0.8084,
      "num_input_tokens_seen": 5010365952,
      "step": 67510,
      "train_runtime": 81076.4854,
      "train_tokens_per_second": 61798.016
    },
    {
      "epoch": 0.89726978558282,
      "grad_norm": 19.375,
      "learning_rate": 5.486297346101011e-07,
      "loss": 0.9416,
      "num_input_tokens_seen": 5011218176,
      "step": 67520,
      "train_runtime": 81090.0669,
      "train_tokens_per_second": 61798.175
    },
    {
      "epoch": 0.8974026750652819,
      "grad_norm": 18.625,
      "learning_rate": 5.472246161635631e-07,
      "loss": 0.7981,
      "num_input_tokens_seen": 5011740416,
      "step": 67530,
      "train_runtime": 81097.5991,
      "train_tokens_per_second": 61798.875
    },
    {
      "epoch": 0.8975355645477439,
      "grad_norm": 16.125,
      "learning_rate": 5.458212487593806e-07,
      "loss": 0.6152,
      "num_input_tokens_seen": 5011981824,
      "step": 67540,
      "train_runtime": 81101.522,
      "train_tokens_per_second": 61798.863
    },
    {
      "epoch": 0.8976684540302058,
      "grad_norm": 22.0,
      "learning_rate": 5.444196326575079e-07,
      "loss": 0.5753,
      "num_input_tokens_seen": 5012100608,
      "step": 67550,
      "train_runtime": 81104.431,
      "train_tokens_per_second": 61798.111
    },
    {
      "epoch": 0.8978013435126677,
      "grad_norm": 65.5,
      "learning_rate": 5.430197681175841e-07,
      "loss": 0.8285,
      "num_input_tokens_seen": 5014108416,
      "step": 67560,
      "train_runtime": 81137.4408,
      "train_tokens_per_second": 61797.715
    },
    {
      "epoch": 0.8979342329951296,
      "grad_norm": 34.75,
      "learning_rate": 5.416216553989229e-07,
      "loss": 0.8975,
      "num_input_tokens_seen": 5014937856,
      "step": 67570,
      "train_runtime": 81151.9702,
      "train_tokens_per_second": 61796.871
    },
    {
      "epoch": 0.8980671224775915,
      "grad_norm": 16.25,
      "learning_rate": 5.402252947605102e-07,
      "loss": 0.7833,
      "num_input_tokens_seen": 5015437312,
      "step": 67580,
      "train_runtime": 81159.149,
      "train_tokens_per_second": 61797.559
    },
    {
      "epoch": 0.8982000119600534,
      "grad_norm": 21.125,
      "learning_rate": 5.388306864610127e-07,
      "loss": 0.5901,
      "num_input_tokens_seen": 5015673344,
      "step": 67590,
      "train_runtime": 81163.7835,
      "train_tokens_per_second": 61796.939
    },
    {
      "epoch": 0.8983329014425153,
      "grad_norm": 13.5625,
      "learning_rate": 5.374378307587636e-07,
      "loss": 0.5354,
      "num_input_tokens_seen": 5015789312,
      "step": 67600,
      "train_runtime": 81167.2517,
      "train_tokens_per_second": 61795.727
    },
    {
      "epoch": 0.8984657909249772,
      "grad_norm": 49.5,
      "learning_rate": 5.360467279117821e-07,
      "loss": 0.8278,
      "num_input_tokens_seen": 5017892864,
      "step": 67610,
      "train_runtime": 81201.9382,
      "train_tokens_per_second": 61795.235
    },
    {
      "epoch": 0.8985986804074392,
      "grad_norm": 13.5625,
      "learning_rate": 5.346573781777553e-07,
      "loss": 0.9273,
      "num_input_tokens_seen": 5018776064,
      "step": 67620,
      "train_runtime": 81216.5718,
      "train_tokens_per_second": 61794.976
    },
    {
      "epoch": 0.8987315698899011,
      "grad_norm": 13.8125,
      "learning_rate": 5.332697818140486e-07,
      "loss": 0.7152,
      "num_input_tokens_seen": 5019264256,
      "step": 67630,
      "train_runtime": 81223.7591,
      "train_tokens_per_second": 61795.518
    },
    {
      "epoch": 0.8988644593723629,
      "grad_norm": 23.625,
      "learning_rate": 5.318839390777009e-07,
      "loss": 0.6573,
      "num_input_tokens_seen": 5019505152,
      "step": 67640,
      "train_runtime": 81227.6044,
      "train_tokens_per_second": 61795.558
    },
    {
      "epoch": 0.8989973488548249,
      "grad_norm": 21.25,
      "learning_rate": 5.304998502254288e-07,
      "loss": 0.5094,
      "num_input_tokens_seen": 5019624192,
      "step": 67650,
      "train_runtime": 81231.4363,
      "train_tokens_per_second": 61794.109
    },
    {
      "epoch": 0.8991302383372868,
      "grad_norm": 28.75,
      "learning_rate": 5.291175155136219e-07,
      "loss": 0.7759,
      "num_input_tokens_seen": 5021622016,
      "step": 67660,
      "train_runtime": 81264.1278,
      "train_tokens_per_second": 61793.834
    },
    {
      "epoch": 0.8992631278197487,
      "grad_norm": 12.8125,
      "learning_rate": 5.277369351983441e-07,
      "loss": 0.8972,
      "num_input_tokens_seen": 5022495744,
      "step": 67670,
      "train_runtime": 81277.1895,
      "train_tokens_per_second": 61794.653
    },
    {
      "epoch": 0.8993960173022106,
      "grad_norm": 19.375,
      "learning_rate": 5.263581095353365e-07,
      "loss": 0.6779,
      "num_input_tokens_seen": 5022961408,
      "step": 67680,
      "train_runtime": 81285.414,
      "train_tokens_per_second": 61794.13
    },
    {
      "epoch": 0.8995289067846726,
      "grad_norm": 24.75,
      "learning_rate": 5.249810387800136e-07,
      "loss": 0.6293,
      "num_input_tokens_seen": 5023191296,
      "step": 67690,
      "train_runtime": 81289.2574,
      "train_tokens_per_second": 61794.036
    },
    {
      "epoch": 0.8996617962671344,
      "grad_norm": 6.6875,
      "learning_rate": 5.23605723187467e-07,
      "loss": 0.5091,
      "num_input_tokens_seen": 5023306240,
      "step": 67700,
      "train_runtime": 81292.4606,
      "train_tokens_per_second": 61793.015
    },
    {
      "epoch": 0.8997946857495963,
      "grad_norm": 26.75,
      "learning_rate": 5.222321630124594e-07,
      "loss": 0.8497,
      "num_input_tokens_seen": 5025303552,
      "step": 67710,
      "train_runtime": 81324.658,
      "train_tokens_per_second": 61793.11
    },
    {
      "epoch": 0.8999275752320582,
      "grad_norm": 54.0,
      "learning_rate": 5.20860358509434e-07,
      "loss": 0.878,
      "num_input_tokens_seen": 5026149632,
      "step": 67720,
      "train_runtime": 81337.6357,
      "train_tokens_per_second": 61793.653
    },
    {
      "epoch": 0.9000604647145202,
      "grad_norm": 18.625,
      "learning_rate": 5.194903099325011e-07,
      "loss": 0.7012,
      "num_input_tokens_seen": 5026617856,
      "step": 67730,
      "train_runtime": 81346.5452,
      "train_tokens_per_second": 61792.641
    },
    {
      "epoch": 0.9001933541969821,
      "grad_norm": 12.375,
      "learning_rate": 5.18122017535454e-07,
      "loss": 0.6077,
      "num_input_tokens_seen": 5026849280,
      "step": 67740,
      "train_runtime": 81350.8462,
      "train_tokens_per_second": 61792.219
    },
    {
      "epoch": 0.900326243679444,
      "grad_norm": 14.375,
      "learning_rate": 5.167554815717534e-07,
      "loss": 0.5197,
      "num_input_tokens_seen": 5026963712,
      "step": 67750,
      "train_runtime": 81354.2832,
      "train_tokens_per_second": 61791.015
    },
    {
      "epoch": 0.900459133161906,
      "grad_norm": 16.625,
      "learning_rate": 5.153907022945404e-07,
      "loss": 0.7955,
      "num_input_tokens_seen": 5029058304,
      "step": 67760,
      "train_runtime": 81388.7147,
      "train_tokens_per_second": 61790.61
    },
    {
      "epoch": 0.9005920226443678,
      "grad_norm": 15.3125,
      "learning_rate": 5.140276799566257e-07,
      "loss": 0.8744,
      "num_input_tokens_seen": 5029911040,
      "step": 67770,
      "train_runtime": 81402.0998,
      "train_tokens_per_second": 61790.925
    },
    {
      "epoch": 0.9007249121268297,
      "grad_norm": 31.125,
      "learning_rate": 5.126664148104998e-07,
      "loss": 0.7115,
      "num_input_tokens_seen": 5030384384,
      "step": 67780,
      "train_runtime": 81410.6741,
      "train_tokens_per_second": 61790.231
    },
    {
      "epoch": 0.9008578016092916,
      "grad_norm": 21.5,
      "learning_rate": 5.113069071083243e-07,
      "loss": 0.6088,
      "num_input_tokens_seen": 5030618880,
      "step": 67790,
      "train_runtime": 81414.8619,
      "train_tokens_per_second": 61789.933
    },
    {
      "epoch": 0.9009906910917536,
      "grad_norm": 14.75,
      "learning_rate": 5.099491571019322e-07,
      "loss": 0.4161,
      "num_input_tokens_seen": 5030734848,
      "step": 67800,
      "train_runtime": 81417.8849,
      "train_tokens_per_second": 61789.063
    },
    {
      "epoch": 0.9011235805742155,
      "grad_norm": 16.375,
      "learning_rate": 5.085931650428399e-07,
      "loss": 0.8678,
      "num_input_tokens_seen": 5032964864,
      "step": 67810,
      "train_runtime": 81453.2978,
      "train_tokens_per_second": 61789.578
    },
    {
      "epoch": 0.9012564700566774,
      "grad_norm": 15.25,
      "learning_rate": 5.072389311822301e-07,
      "loss": 0.8422,
      "num_input_tokens_seen": 5033784064,
      "step": 67820,
      "train_runtime": 81466.2733,
      "train_tokens_per_second": 61789.792
    },
    {
      "epoch": 0.9013893595391392,
      "grad_norm": 26.75,
      "learning_rate": 5.05886455770962e-07,
      "loss": 0.7636,
      "num_input_tokens_seen": 5034202368,
      "step": 67830,
      "train_runtime": 81473.1934,
      "train_tokens_per_second": 61789.678
    },
    {
      "epoch": 0.9015222490216012,
      "grad_norm": 27.25,
      "learning_rate": 5.045357390595728e-07,
      "loss": 0.6501,
      "num_input_tokens_seen": 5034428416,
      "step": 67840,
      "train_runtime": 81477.5222,
      "train_tokens_per_second": 61789.169
    },
    {
      "epoch": 0.9016551385040631,
      "grad_norm": 14.625,
      "learning_rate": 5.031867812982649e-07,
      "loss": 0.536,
      "num_input_tokens_seen": 5034543104,
      "step": 67850,
      "train_runtime": 81482.1556,
      "train_tokens_per_second": 61787.063
    },
    {
      "epoch": 0.901788027986525,
      "grad_norm": 23.75,
      "learning_rate": 5.018395827369272e-07,
      "loss": 0.7945,
      "num_input_tokens_seen": 5036616704,
      "step": 67860,
      "train_runtime": 81516.1438,
      "train_tokens_per_second": 61786.739
    },
    {
      "epoch": 0.901920917468987,
      "grad_norm": 18.0,
      "learning_rate": 5.004941436251121e-07,
      "loss": 0.8449,
      "num_input_tokens_seen": 5037479168,
      "step": 67870,
      "train_runtime": 81529.0102,
      "train_tokens_per_second": 61787.567
    },
    {
      "epoch": 0.9020538069514489,
      "grad_norm": 17.75,
      "learning_rate": 4.991504642120515e-07,
      "loss": 0.6414,
      "num_input_tokens_seen": 5037961216,
      "step": 67880,
      "train_runtime": 81535.9286,
      "train_tokens_per_second": 61788.236
    },
    {
      "epoch": 0.9021866964339107,
      "grad_norm": 18.875,
      "learning_rate": 4.978085447466496e-07,
      "loss": 0.6091,
      "num_input_tokens_seen": 5038189312,
      "step": 67890,
      "train_runtime": 81539.7298,
      "train_tokens_per_second": 61788.153
    },
    {
      "epoch": 0.9023195859163726,
      "grad_norm": 30.875,
      "learning_rate": 4.964683854774854e-07,
      "loss": 0.5224,
      "num_input_tokens_seen": 5038306048,
      "step": 67900,
      "train_runtime": 81544.2006,
      "train_tokens_per_second": 61786.197
    },
    {
      "epoch": 0.9024524753988346,
      "grad_norm": 40.75,
      "learning_rate": 4.951299866528125e-07,
      "loss": 0.8154,
      "num_input_tokens_seen": 5040255488,
      "step": 67910,
      "train_runtime": 81575.7139,
      "train_tokens_per_second": 61786.226
    },
    {
      "epoch": 0.9025853648812965,
      "grad_norm": 12.4375,
      "learning_rate": 4.937933485205548e-07,
      "loss": 0.9483,
      "num_input_tokens_seen": 5041100032,
      "step": 67920,
      "train_runtime": 81588.6277,
      "train_tokens_per_second": 61786.798
    },
    {
      "epoch": 0.9027182543637584,
      "grad_norm": 30.125,
      "learning_rate": 4.924584713283143e-07,
      "loss": 0.7786,
      "num_input_tokens_seen": 5041578496,
      "step": 67930,
      "train_runtime": 81595.8546,
      "train_tokens_per_second": 61787.189
    },
    {
      "epoch": 0.9028511438462203,
      "grad_norm": 21.75,
      "learning_rate": 4.911253553233652e-07,
      "loss": 0.5818,
      "num_input_tokens_seen": 5041807872,
      "step": 67940,
      "train_runtime": 81600.1298,
      "train_tokens_per_second": 61786.763
    },
    {
      "epoch": 0.9029840333286822,
      "grad_norm": 14.1875,
      "learning_rate": 4.897940007526536e-07,
      "loss": 0.515,
      "num_input_tokens_seen": 5041926656,
      "step": 67950,
      "train_runtime": 81603.5512,
      "train_tokens_per_second": 61785.628
    },
    {
      "epoch": 0.9031169228111441,
      "grad_norm": 22.375,
      "learning_rate": 4.884644078628043e-07,
      "loss": 0.8363,
      "num_input_tokens_seen": 5043921408,
      "step": 67960,
      "train_runtime": 81635.4211,
      "train_tokens_per_second": 61785.942
    },
    {
      "epoch": 0.903249812293606,
      "grad_norm": 14.875,
      "learning_rate": 4.871365769001091e-07,
      "loss": 0.9145,
      "num_input_tokens_seen": 5044734976,
      "step": 67970,
      "train_runtime": 81648.0159,
      "train_tokens_per_second": 61786.376
    },
    {
      "epoch": 0.9033827017760679,
      "grad_norm": 20.5,
      "learning_rate": 4.858105081105402e-07,
      "loss": 0.7461,
      "num_input_tokens_seen": 5045178624,
      "step": 67980,
      "train_runtime": 81656.045,
      "train_tokens_per_second": 61785.733
    },
    {
      "epoch": 0.9035155912585299,
      "grad_norm": 15.0625,
      "learning_rate": 4.844862017397367e-07,
      "loss": 0.5906,
      "num_input_tokens_seen": 5045407488,
      "step": 67990,
      "train_runtime": 81660.1254,
      "train_tokens_per_second": 61785.449
    },
    {
      "epoch": 0.9036484807409918,
      "grad_norm": 26.25,
      "learning_rate": 4.831636580330168e-07,
      "loss": 0.5398,
      "num_input_tokens_seen": 5045523456,
      "step": 68000,
      "train_runtime": 81663.3146,
      "train_tokens_per_second": 61784.456
    },
    {
      "epoch": 0.9037813702234536,
      "grad_norm": 23.625,
      "learning_rate": 4.81842877235369e-07,
      "loss": 0.8879,
      "num_input_tokens_seen": 5047555840,
      "step": 68010,
      "train_runtime": 81697.1486,
      "train_tokens_per_second": 61783.746
    },
    {
      "epoch": 0.9039142597059155,
      "grad_norm": 13.3125,
      "learning_rate": 4.805238595914552e-07,
      "loss": 0.8268,
      "num_input_tokens_seen": 5048390400,
      "step": 68020,
      "train_runtime": 81710.0348,
      "train_tokens_per_second": 61784.216
    },
    {
      "epoch": 0.9040471491883775,
      "grad_norm": 25.125,
      "learning_rate": 4.792066053456135e-07,
      "loss": 0.7617,
      "num_input_tokens_seen": 5048874496,
      "step": 68030,
      "train_runtime": 81717.2133,
      "train_tokens_per_second": 61784.712
    },
    {
      "epoch": 0.9041800386708394,
      "grad_norm": 9.375,
      "learning_rate": 4.778911147418552e-07,
      "loss": 0.5948,
      "num_input_tokens_seen": 5049103872,
      "step": 68040,
      "train_runtime": 81722.1103,
      "train_tokens_per_second": 61783.817
    },
    {
      "epoch": 0.9043129281533013,
      "grad_norm": 19.0,
      "learning_rate": 4.765773880238567e-07,
      "loss": 0.5901,
      "num_input_tokens_seen": 5049218560,
      "step": 68050,
      "train_runtime": 81725.1879,
      "train_tokens_per_second": 61782.893
    },
    {
      "epoch": 0.9044458176357633,
      "grad_norm": 16.5,
      "learning_rate": 4.75265425434982e-07,
      "loss": 0.7056,
      "num_input_tokens_seen": 5051203328,
      "step": 68060,
      "train_runtime": 81757.5325,
      "train_tokens_per_second": 61782.727
    },
    {
      "epoch": 0.9045787071182251,
      "grad_norm": 21.0,
      "learning_rate": 4.739552272182546e-07,
      "loss": 0.9159,
      "num_input_tokens_seen": 5052069888,
      "step": 68070,
      "train_runtime": 81770.6936,
      "train_tokens_per_second": 61783.381
    },
    {
      "epoch": 0.904711596600687,
      "grad_norm": 25.75,
      "learning_rate": 4.726467936163792e-07,
      "loss": 0.7404,
      "num_input_tokens_seen": 5052543488,
      "step": 68080,
      "train_runtime": 81777.9479,
      "train_tokens_per_second": 61783.691
    },
    {
      "epoch": 0.9048444860831489,
      "grad_norm": 21.0,
      "learning_rate": 4.713401248717331e-07,
      "loss": 0.6111,
      "num_input_tokens_seen": 5052770304,
      "step": 68090,
      "train_runtime": 81783.3853,
      "train_tokens_per_second": 61782.357
    },
    {
      "epoch": 0.9049773755656109,
      "grad_norm": 13.75,
      "learning_rate": 4.7003522122636036e-07,
      "loss": 0.5452,
      "num_input_tokens_seen": 5052888064,
      "step": 68100,
      "train_runtime": 81786.9917,
      "train_tokens_per_second": 61781.073
    },
    {
      "epoch": 0.9051102650480728,
      "grad_norm": 32.5,
      "learning_rate": 4.6873208292198877e-07,
      "loss": 0.8675,
      "num_input_tokens_seen": 5054935296,
      "step": 68110,
      "train_runtime": 81820.6871,
      "train_tokens_per_second": 61780.651
    },
    {
      "epoch": 0.9052431545305347,
      "grad_norm": 11.375,
      "learning_rate": 4.6743071020000754e-07,
      "loss": 0.8943,
      "num_input_tokens_seen": 5055785728,
      "step": 68120,
      "train_runtime": 81833.8593,
      "train_tokens_per_second": 61781.099
    },
    {
      "epoch": 0.9053760440129965,
      "grad_norm": 48.5,
      "learning_rate": 4.6613110330148924e-07,
      "loss": 0.7855,
      "num_input_tokens_seen": 5056256000,
      "step": 68130,
      "train_runtime": 81842.2224,
      "train_tokens_per_second": 61780.532
    },
    {
      "epoch": 0.9055089334954585,
      "grad_norm": 23.75,
      "learning_rate": 4.6483326246717143e-07,
      "loss": 0.5936,
      "num_input_tokens_seen": 5056488448,
      "step": 68140,
      "train_runtime": 81846.6337,
      "train_tokens_per_second": 61780.042
    },
    {
      "epoch": 0.9056418229779204,
      "grad_norm": 24.75,
      "learning_rate": 4.6353718793746726e-07,
      "loss": 0.5657,
      "num_input_tokens_seen": 5056600832,
      "step": 68150,
      "train_runtime": 81850.1146,
      "train_tokens_per_second": 61778.787
    },
    {
      "epoch": 0.9057747124603823,
      "grad_norm": 16.375,
      "learning_rate": 4.6224287995246584e-07,
      "loss": 0.7511,
      "num_input_tokens_seen": 5058669056,
      "step": 68160,
      "train_runtime": 81884.5603,
      "train_tokens_per_second": 61778.057
    },
    {
      "epoch": 0.9059076019428443,
      "grad_norm": 26.625,
      "learning_rate": 4.6095033875192207e-07,
      "loss": 0.8717,
      "num_input_tokens_seen": 5059507712,
      "step": 68170,
      "train_runtime": 81897.1452,
      "train_tokens_per_second": 61778.804
    },
    {
      "epoch": 0.9060404914253062,
      "grad_norm": 41.75,
      "learning_rate": 4.5965956457527215e-07,
      "loss": 0.7852,
      "num_input_tokens_seen": 5059953408,
      "step": 68180,
      "train_runtime": 81903.7634,
      "train_tokens_per_second": 61779.254
    },
    {
      "epoch": 0.906173380907768,
      "grad_norm": 12.5,
      "learning_rate": 4.5837055766161707e-07,
      "loss": 0.7008,
      "num_input_tokens_seen": 5060179456,
      "step": 68190,
      "train_runtime": 81907.3078,
      "train_tokens_per_second": 61779.341
    },
    {
      "epoch": 0.9063062703902299,
      "grad_norm": 7.6875,
      "learning_rate": 4.5708331824973473e-07,
      "loss": 0.5483,
      "num_input_tokens_seen": 5060292608,
      "step": 68200,
      "train_runtime": 81911.1155,
      "train_tokens_per_second": 61777.85
    },
    {
      "epoch": 0.9064391598726919,
      "grad_norm": 36.5,
      "learning_rate": 4.557978465780744e-07,
      "loss": 0.8332,
      "num_input_tokens_seen": 5062487552,
      "step": 68210,
      "train_runtime": 81946.593,
      "train_tokens_per_second": 61777.889
    },
    {
      "epoch": 0.9065720493551538,
      "grad_norm": 72.0,
      "learning_rate": 4.545141428847599e-07,
      "loss": 0.916,
      "num_input_tokens_seen": 5063360512,
      "step": 68220,
      "train_runtime": 81959.9491,
      "train_tokens_per_second": 61778.473
    },
    {
      "epoch": 0.9067049388376157,
      "grad_norm": 30.25,
      "learning_rate": 4.532322074075857e-07,
      "loss": 0.7584,
      "num_input_tokens_seen": 5063898880,
      "step": 68230,
      "train_runtime": 81967.5209,
      "train_tokens_per_second": 61779.334
    },
    {
      "epoch": 0.9068378283200776,
      "grad_norm": 11.5625,
      "learning_rate": 4.5195204038401717e-07,
      "loss": 0.6235,
      "num_input_tokens_seen": 5064134656,
      "step": 68240,
      "train_runtime": 81971.9337,
      "train_tokens_per_second": 61778.885
    },
    {
      "epoch": 0.9069707178025396,
      "grad_norm": 6.5,
      "learning_rate": 4.506736420511948e-07,
      "loss": 0.5353,
      "num_input_tokens_seen": 5064249344,
      "step": 68250,
      "train_runtime": 81976.1069,
      "train_tokens_per_second": 61777.139
    },
    {
      "epoch": 0.9071036072850014,
      "grad_norm": 37.0,
      "learning_rate": 4.493970126459302e-07,
      "loss": 0.8105,
      "num_input_tokens_seen": 5066217728,
      "step": 68260,
      "train_runtime": 82007.4152,
      "train_tokens_per_second": 61777.556
    },
    {
      "epoch": 0.9072364967674633,
      "grad_norm": 24.625,
      "learning_rate": 4.4812215240470857e-07,
      "loss": 0.8713,
      "num_input_tokens_seen": 5067041280,
      "step": 68270,
      "train_runtime": 82020.9986,
      "train_tokens_per_second": 61777.366
    },
    {
      "epoch": 0.9073693862499252,
      "grad_norm": 19.75,
      "learning_rate": 4.468490615636845e-07,
      "loss": 0.7623,
      "num_input_tokens_seen": 5067486464,
      "step": 68280,
      "train_runtime": 82028.1901,
      "train_tokens_per_second": 61777.378
    },
    {
      "epoch": 0.9075022757323872,
      "grad_norm": 19.75,
      "learning_rate": 4.455777403586892e-07,
      "loss": 0.6878,
      "num_input_tokens_seen": 5067702784,
      "step": 68290,
      "train_runtime": 82032.026,
      "train_tokens_per_second": 61777.126
    },
    {
      "epoch": 0.9076351652148491,
      "grad_norm": 12.625,
      "learning_rate": 4.443081890252221e-07,
      "loss": 0.5394,
      "num_input_tokens_seen": 5067816960,
      "step": 68300,
      "train_runtime": 82035.5388,
      "train_tokens_per_second": 61775.872
    },
    {
      "epoch": 0.907768054697311,
      "grad_norm": 34.75,
      "learning_rate": 4.4304040779845627e-07,
      "loss": 0.8166,
      "num_input_tokens_seen": 5069889792,
      "step": 68310,
      "train_runtime": 82069.2292,
      "train_tokens_per_second": 61775.77
    },
    {
      "epoch": 0.9079009441797729,
      "grad_norm": 19.25,
      "learning_rate": 4.4177439691323707e-07,
      "loss": 0.9889,
      "num_input_tokens_seen": 5070742016,
      "step": 68320,
      "train_runtime": 82083.4757,
      "train_tokens_per_second": 61775.43
    },
    {
      "epoch": 0.9080338336622348,
      "grad_norm": 12.875,
      "learning_rate": 4.4051015660408145e-07,
      "loss": 0.7333,
      "num_input_tokens_seen": 5071222272,
      "step": 68330,
      "train_runtime": 82090.3662,
      "train_tokens_per_second": 61776.095
    },
    {
      "epoch": 0.9081667231446967,
      "grad_norm": 11.125,
      "learning_rate": 4.392476871051787e-07,
      "loss": 0.6142,
      "num_input_tokens_seen": 5071448832,
      "step": 68340,
      "train_runtime": 82094.208,
      "train_tokens_per_second": 61775.964
    },
    {
      "epoch": 0.9082996126271586,
      "grad_norm": 15.75,
      "learning_rate": 4.379869886503907e-07,
      "loss": 0.4763,
      "num_input_tokens_seen": 5071562240,
      "step": 68350,
      "train_runtime": 82097.5062,
      "train_tokens_per_second": 61774.864
    },
    {
      "epoch": 0.9084325021096206,
      "grad_norm": 76.0,
      "learning_rate": 4.367280614732505e-07,
      "loss": 0.7815,
      "num_input_tokens_seen": 5073531904,
      "step": 68360,
      "train_runtime": 82129.8584,
      "train_tokens_per_second": 61774.512
    },
    {
      "epoch": 0.9085653915920825,
      "grad_norm": 16.75,
      "learning_rate": 4.3547090580696063e-07,
      "loss": 0.9313,
      "num_input_tokens_seen": 5074399744,
      "step": 68370,
      "train_runtime": 82143.5279,
      "train_tokens_per_second": 61774.797
    },
    {
      "epoch": 0.9086982810745443,
      "grad_norm": 29.375,
      "learning_rate": 4.3421552188440355e-07,
      "loss": 0.7579,
      "num_input_tokens_seen": 5074861056,
      "step": 68380,
      "train_runtime": 82150.0574,
      "train_tokens_per_second": 61775.502
    },
    {
      "epoch": 0.9088311705570062,
      "grad_norm": 17.75,
      "learning_rate": 4.3296190993812215e-07,
      "loss": 0.6726,
      "num_input_tokens_seen": 5075085824,
      "step": 68390,
      "train_runtime": 82153.9182,
      "train_tokens_per_second": 61775.335
    },
    {
      "epoch": 0.9089640600394682,
      "grad_norm": 25.625,
      "learning_rate": 4.3171007020033963e-07,
      "loss": 0.5419,
      "num_input_tokens_seen": 5075201024,
      "step": 68400,
      "train_runtime": 82156.8633,
      "train_tokens_per_second": 61774.523
    },
    {
      "epoch": 0.9090969495219301,
      "grad_norm": 38.5,
      "learning_rate": 4.304600029029493e-07,
      "loss": 0.8054,
      "num_input_tokens_seen": 5077244160,
      "step": 68410,
      "train_runtime": 82189.2005,
      "train_tokens_per_second": 61775.077
    },
    {
      "epoch": 0.909229839004392,
      "grad_norm": 29.0,
      "learning_rate": 4.2921170827751046e-07,
      "loss": 0.8932,
      "num_input_tokens_seen": 5078092544,
      "step": 68420,
      "train_runtime": 82202.876,
      "train_tokens_per_second": 61775.12
    },
    {
      "epoch": 0.909362728486854,
      "grad_norm": 19.625,
      "learning_rate": 4.2796518655526477e-07,
      "loss": 0.7438,
      "num_input_tokens_seen": 5078573312,
      "step": 68430,
      "train_runtime": 82209.7333,
      "train_tokens_per_second": 61775.815
    },
    {
      "epoch": 0.9094956179693158,
      "grad_norm": 21.5,
      "learning_rate": 4.2672043796711416e-07,
      "loss": 0.676,
      "num_input_tokens_seen": 5078801408,
      "step": 68440,
      "train_runtime": 82213.5036,
      "train_tokens_per_second": 61775.757
    },
    {
      "epoch": 0.9096285074517777,
      "grad_norm": 16.375,
      "learning_rate": 4.254774627436387e-07,
      "loss": 0.531,
      "num_input_tokens_seen": 5078917632,
      "step": 68450,
      "train_runtime": 82216.5279,
      "train_tokens_per_second": 61774.898
    },
    {
      "epoch": 0.9097613969342396,
      "grad_norm": 17.875,
      "learning_rate": 4.2423626111509074e-07,
      "loss": 0.7991,
      "num_input_tokens_seen": 5081016320,
      "step": 68460,
      "train_runtime": 82249.9013,
      "train_tokens_per_second": 61775.349
    },
    {
      "epoch": 0.9098942864167016,
      "grad_norm": 27.375,
      "learning_rate": 4.2299683331138764e-07,
      "loss": 0.8903,
      "num_input_tokens_seen": 5081843968,
      "step": 68470,
      "train_runtime": 82262.9756,
      "train_tokens_per_second": 61775.591
    },
    {
      "epoch": 0.9100271758991635,
      "grad_norm": 17.125,
      "learning_rate": 4.2175917956212677e-07,
      "loss": 0.7259,
      "num_input_tokens_seen": 5082299136,
      "step": 68480,
      "train_runtime": 82269.9053,
      "train_tokens_per_second": 61775.921
    },
    {
      "epoch": 0.9101600653816254,
      "grad_norm": 23.625,
      "learning_rate": 4.2052330009656696e-07,
      "loss": 0.6374,
      "num_input_tokens_seen": 5082525696,
      "step": 68490,
      "train_runtime": 82273.949,
      "train_tokens_per_second": 61775.638
    },
    {
      "epoch": 0.9102929548640872,
      "grad_norm": 18.625,
      "learning_rate": 4.1928919514365063e-07,
      "loss": 0.4875,
      "num_input_tokens_seen": 5082640384,
      "step": 68500,
      "train_runtime": 82278.2952,
      "train_tokens_per_second": 61773.769
    },
    {
      "epoch": 0.9104258443465492,
      "grad_norm": 21.125,
      "learning_rate": 4.1805686493197937e-07,
      "loss": 0.8291,
      "num_input_tokens_seen": 5084756736,
      "step": 68510,
      "train_runtime": 82311.4298,
      "train_tokens_per_second": 61774.613
    },
    {
      "epoch": 0.9105587338290111,
      "grad_norm": 23.625,
      "learning_rate": 4.1682630968983397e-07,
      "loss": 0.9033,
      "num_input_tokens_seen": 5085577984,
      "step": 68520,
      "train_runtime": 82324.1109,
      "train_tokens_per_second": 61775.073
    },
    {
      "epoch": 0.910691623311473,
      "grad_norm": 27.875,
      "learning_rate": 4.155975296451631e-07,
      "loss": 0.7921,
      "num_input_tokens_seen": 5086064896,
      "step": 68530,
      "train_runtime": 82331.0045,
      "train_tokens_per_second": 61775.815
    },
    {
      "epoch": 0.910824512793935,
      "grad_norm": 20.75,
      "learning_rate": 4.1437052502558693e-07,
      "loss": 0.6228,
      "num_input_tokens_seen": 5086289152,
      "step": 68540,
      "train_runtime": 82334.7336,
      "train_tokens_per_second": 61775.741
    },
    {
      "epoch": 0.9109574022763969,
      "grad_norm": 15.3125,
      "learning_rate": 4.1314529605839927e-07,
      "loss": 0.5388,
      "num_input_tokens_seen": 5086399232,
      "step": 68550,
      "train_runtime": 82337.7632,
      "train_tokens_per_second": 61774.805
    },
    {
      "epoch": 0.9110902917588587,
      "grad_norm": 30.0,
      "learning_rate": 4.119218429705596e-07,
      "loss": 0.9064,
      "num_input_tokens_seen": 5088420096,
      "step": 68560,
      "train_runtime": 82372.0601,
      "train_tokens_per_second": 61773.617
    },
    {
      "epoch": 0.9112231812413206,
      "grad_norm": 15.0625,
      "learning_rate": 4.1070016598870223e-07,
      "loss": 0.9256,
      "num_input_tokens_seen": 5089241344,
      "step": 68570,
      "train_runtime": 82384.5367,
      "train_tokens_per_second": 61774.23
    },
    {
      "epoch": 0.9113560707237826,
      "grad_norm": 27.5,
      "learning_rate": 4.094802653391339e-07,
      "loss": 0.6811,
      "num_input_tokens_seen": 5089681152,
      "step": 68580,
      "train_runtime": 82391.0338,
      "train_tokens_per_second": 61774.697
    },
    {
      "epoch": 0.9114889602062445,
      "grad_norm": 16.625,
      "learning_rate": 4.0826214124782936e-07,
      "loss": 0.6041,
      "num_input_tokens_seen": 5089905920,
      "step": 68590,
      "train_runtime": 82394.9668,
      "train_tokens_per_second": 61774.476
    },
    {
      "epoch": 0.9116218496887064,
      "grad_norm": 9.3125,
      "learning_rate": 4.0704579394043487e-07,
      "loss": 0.558,
      "num_input_tokens_seen": 5090022656,
      "step": 68600,
      "train_runtime": 82399.1845,
      "train_tokens_per_second": 61772.731
    },
    {
      "epoch": 0.9117547391711683,
      "grad_norm": 37.75,
      "learning_rate": 4.058312236422701e-07,
      "loss": 0.8486,
      "num_input_tokens_seen": 5092010240,
      "step": 68610,
      "train_runtime": 82431.4912,
      "train_tokens_per_second": 61772.633
    },
    {
      "epoch": 0.9118876286536302,
      "grad_norm": 18.125,
      "learning_rate": 4.0461843057831963e-07,
      "loss": 0.9386,
      "num_input_tokens_seen": 5092828672,
      "step": 68620,
      "train_runtime": 82444.0892,
      "train_tokens_per_second": 61773.121
    },
    {
      "epoch": 0.9120205181360921,
      "grad_norm": 22.75,
      "learning_rate": 4.0340741497324475e-07,
      "loss": 0.7593,
      "num_input_tokens_seen": 5093274624,
      "step": 68630,
      "train_runtime": 82451.2952,
      "train_tokens_per_second": 61773.131
    },
    {
      "epoch": 0.912153407618554,
      "grad_norm": 18.5,
      "learning_rate": 4.02198177051375e-07,
      "loss": 0.658,
      "num_input_tokens_seen": 5093497344,
      "step": 68640,
      "train_runtime": 82455.0916,
      "train_tokens_per_second": 61772.987
    },
    {
      "epoch": 0.9122862971010159,
      "grad_norm": 8.6875,
      "learning_rate": 4.009907170367111e-07,
      "loss": 0.477,
      "num_input_tokens_seen": 5093612032,
      "step": 68650,
      "train_runtime": 82458.6216,
      "train_tokens_per_second": 61771.734
    },
    {
      "epoch": 0.9124191865834779,
      "grad_norm": 41.25,
      "learning_rate": 3.997850351529231e-07,
      "loss": 0.8485,
      "num_input_tokens_seen": 5095708160,
      "step": 68660,
      "train_runtime": 82493.067,
      "train_tokens_per_second": 61771.351
    },
    {
      "epoch": 0.9125520760659398,
      "grad_norm": 29.25,
      "learning_rate": 3.9858113162335456e-07,
      "loss": 0.9266,
      "num_input_tokens_seen": 5096569088,
      "step": 68670,
      "train_runtime": 82508.2594,
      "train_tokens_per_second": 61770.411
    },
    {
      "epoch": 0.9126849655484016,
      "grad_norm": 14.1875,
      "learning_rate": 3.9737900667101815e-07,
      "loss": 0.7126,
      "num_input_tokens_seen": 5097063680,
      "step": 68680,
      "train_runtime": 82516.1304,
      "train_tokens_per_second": 61770.513
    },
    {
      "epoch": 0.9128178550308635,
      "grad_norm": 14.125,
      "learning_rate": 3.9617866051859354e-07,
      "loss": 0.5969,
      "num_input_tokens_seen": 5097305856,
      "step": 68690,
      "train_runtime": 82520.4587,
      "train_tokens_per_second": 61770.207
    },
    {
      "epoch": 0.9129507445133255,
      "grad_norm": 23.5,
      "learning_rate": 3.9498009338843954e-07,
      "loss": 0.5298,
      "num_input_tokens_seen": 5097420288,
      "step": 68700,
      "train_runtime": 82523.7322,
      "train_tokens_per_second": 61769.144
    },
    {
      "epoch": 0.9130836339957874,
      "grad_norm": 23.375,
      "learning_rate": 3.9378330550257413e-07,
      "loss": 0.8099,
      "num_input_tokens_seen": 5099446016,
      "step": 68710,
      "train_runtime": 82555.8443,
      "train_tokens_per_second": 61769.655
    },
    {
      "epoch": 0.9132165234782493,
      "grad_norm": 20.625,
      "learning_rate": 3.925882970826955e-07,
      "loss": 0.8727,
      "num_input_tokens_seen": 5100278528,
      "step": 68720,
      "train_runtime": 82570.0262,
      "train_tokens_per_second": 61769.128
    },
    {
      "epoch": 0.9133494129607113,
      "grad_norm": 19.125,
      "learning_rate": 3.9139506835016774e-07,
      "loss": 0.7529,
      "num_input_tokens_seen": 5100727296,
      "step": 68730,
      "train_runtime": 82577.1011,
      "train_tokens_per_second": 61769.271
    },
    {
      "epoch": 0.9134823024431732,
      "grad_norm": 19.625,
      "learning_rate": 3.902036195260228e-07,
      "loss": 0.6085,
      "num_input_tokens_seen": 5100960512,
      "step": 68740,
      "train_runtime": 82580.8193,
      "train_tokens_per_second": 61769.313
    },
    {
      "epoch": 0.913615191925635,
      "grad_norm": 6.09375,
      "learning_rate": 3.890139508309709e-07,
      "loss": 0.5883,
      "num_input_tokens_seen": 5101076224,
      "step": 68750,
      "train_runtime": 82583.6471,
      "train_tokens_per_second": 61768.6
    },
    {
      "epoch": 0.9137480814080969,
      "grad_norm": 35.25,
      "learning_rate": 3.8782606248538467e-07,
      "loss": 0.8395,
      "num_input_tokens_seen": 5103083520,
      "step": 68760,
      "train_runtime": 82616.2165,
      "train_tokens_per_second": 61768.545
    },
    {
      "epoch": 0.9138809708905589,
      "grad_norm": 36.25,
      "learning_rate": 3.8663995470930917e-07,
      "loss": 0.9672,
      "num_input_tokens_seen": 5103946240,
      "step": 68770,
      "train_runtime": 82631.1194,
      "train_tokens_per_second": 61767.846
    },
    {
      "epoch": 0.9140138603730208,
      "grad_norm": 80.0,
      "learning_rate": 3.854556277224619e-07,
      "loss": 0.7525,
      "num_input_tokens_seen": 5104436992,
      "step": 68780,
      "train_runtime": 82638.0176,
      "train_tokens_per_second": 61768.628
    },
    {
      "epoch": 0.9141467498554827,
      "grad_norm": 10.25,
      "learning_rate": 3.842730817442275e-07,
      "loss": 0.6202,
      "num_input_tokens_seen": 5104667392,
      "step": 68790,
      "train_runtime": 82642.1757,
      "train_tokens_per_second": 61768.308
    },
    {
      "epoch": 0.9142796393379446,
      "grad_norm": 18.875,
      "learning_rate": 3.83092316993664e-07,
      "loss": 0.5154,
      "num_input_tokens_seen": 5104781568,
      "step": 68800,
      "train_runtime": 82645.4527,
      "train_tokens_per_second": 61767.241
    },
    {
      "epoch": 0.9144125288204065,
      "grad_norm": 20.5,
      "learning_rate": 3.819133336894942e-07,
      "loss": 0.8801,
      "num_input_tokens_seen": 5106845696,
      "step": 68810,
      "train_runtime": 82678.8363,
      "train_tokens_per_second": 61767.266
    },
    {
      "epoch": 0.9145454183028684,
      "grad_norm": 14.0625,
      "learning_rate": 3.8073613205011905e-07,
      "loss": 0.8969,
      "num_input_tokens_seen": 5107719168,
      "step": 68820,
      "train_runtime": 82692.4032,
      "train_tokens_per_second": 61767.695
    },
    {
      "epoch": 0.9146783077853303,
      "grad_norm": 30.375,
      "learning_rate": 3.795607122936018e-07,
      "loss": 0.746,
      "num_input_tokens_seen": 5108176384,
      "step": 68830,
      "train_runtime": 82701.0408,
      "train_tokens_per_second": 61766.773
    },
    {
      "epoch": 0.9148111972677923,
      "grad_norm": 61.5,
      "learning_rate": 3.7838707463767723e-07,
      "loss": 0.6892,
      "num_input_tokens_seen": 5108400896,
      "step": 68840,
      "train_runtime": 82705.3414,
      "train_tokens_per_second": 61766.275
    },
    {
      "epoch": 0.9149440867502542,
      "grad_norm": 12.125,
      "learning_rate": 3.772152192997536e-07,
      "loss": 0.563,
      "num_input_tokens_seen": 5108515328,
      "step": 68850,
      "train_runtime": 82708.5826,
      "train_tokens_per_second": 61765.238
    },
    {
      "epoch": 0.9150769762327161,
      "grad_norm": 17.625,
      "learning_rate": 3.7604514649690505e-07,
      "loss": 0.7418,
      "num_input_tokens_seen": 5110424320,
      "step": 68860,
      "train_runtime": 82739.5471,
      "train_tokens_per_second": 61765.196
    },
    {
      "epoch": 0.9152098657151779,
      "grad_norm": 75.0,
      "learning_rate": 3.7487685644588043e-07,
      "loss": 0.9202,
      "num_input_tokens_seen": 5111255552,
      "step": 68870,
      "train_runtime": 82753.9197,
      "train_tokens_per_second": 61764.513
    },
    {
      "epoch": 0.9153427551976399,
      "grad_norm": 13.125,
      "learning_rate": 3.7371034936309e-07,
      "loss": 0.7697,
      "num_input_tokens_seen": 5111719424,
      "step": 68880,
      "train_runtime": 82760.5171,
      "train_tokens_per_second": 61765.194
    },
    {
      "epoch": 0.9154756446801018,
      "grad_norm": 18.125,
      "learning_rate": 3.72545625464622e-07,
      "loss": 0.5992,
      "num_input_tokens_seen": 5111940864,
      "step": 68890,
      "train_runtime": 82764.0222,
      "train_tokens_per_second": 61765.254
    },
    {
      "epoch": 0.9156085341625637,
      "grad_norm": 9.4375,
      "learning_rate": 3.713826849662294e-07,
      "loss": 0.4273,
      "num_input_tokens_seen": 5112054016,
      "step": 68900,
      "train_runtime": 82766.7916,
      "train_tokens_per_second": 61764.555
    },
    {
      "epoch": 0.9157414236450256,
      "grad_norm": 15.625,
      "learning_rate": 3.7022152808333875e-07,
      "loss": 0.7946,
      "num_input_tokens_seen": 5114188544,
      "step": 68910,
      "train_runtime": 82800.0443,
      "train_tokens_per_second": 61765.529
    },
    {
      "epoch": 0.9158743131274876,
      "grad_norm": 21.125,
      "learning_rate": 3.6906215503104247e-07,
      "loss": 0.9442,
      "num_input_tokens_seen": 5115062784,
      "step": 68920,
      "train_runtime": 82812.6718,
      "train_tokens_per_second": 61766.668
    },
    {
      "epoch": 0.9160072026099494,
      "grad_norm": 25.75,
      "learning_rate": 3.679045660241065e-07,
      "loss": 0.6689,
      "num_input_tokens_seen": 5115536128,
      "step": 68930,
      "train_runtime": 82819.4152,
      "train_tokens_per_second": 61767.354
    },
    {
      "epoch": 0.9161400920924113,
      "grad_norm": 27.375,
      "learning_rate": 3.6674876127695937e-07,
      "loss": 0.6303,
      "num_input_tokens_seen": 5115772928,
      "step": 68940,
      "train_runtime": 82823.2412,
      "train_tokens_per_second": 61767.36
    },
    {
      "epoch": 0.9162729815748732,
      "grad_norm": 117.5,
      "learning_rate": 3.655947410037064e-07,
      "loss": 0.5241,
      "num_input_tokens_seen": 5115891200,
      "step": 68950,
      "train_runtime": 82827.2499,
      "train_tokens_per_second": 61765.798
    },
    {
      "epoch": 0.9164058710573352,
      "grad_norm": 41.25,
      "learning_rate": 3.6444250541812e-07,
      "loss": 0.8443,
      "num_input_tokens_seen": 5117905152,
      "step": 68960,
      "train_runtime": 82859.1905,
      "train_tokens_per_second": 61766.294
    },
    {
      "epoch": 0.9165387605397971,
      "grad_norm": 11.5,
      "learning_rate": 3.632920547336405e-07,
      "loss": 0.8775,
      "num_input_tokens_seen": 5118742528,
      "step": 68970,
      "train_runtime": 82872.1779,
      "train_tokens_per_second": 61766.719
    },
    {
      "epoch": 0.916671650022259,
      "grad_norm": 15.5,
      "learning_rate": 3.6214338916337963e-07,
      "loss": 0.7156,
      "num_input_tokens_seen": 5119214592,
      "step": 68980,
      "train_runtime": 82879.2926,
      "train_tokens_per_second": 61767.113
    },
    {
      "epoch": 0.9168045395047209,
      "grad_norm": 13.25,
      "learning_rate": 3.6099650892011395e-07,
      "loss": 0.5602,
      "num_input_tokens_seen": 5119444736,
      "step": 68990,
      "train_runtime": 82883.3486,
      "train_tokens_per_second": 61766.867
    },
    {
      "epoch": 0.9169374289871828,
      "grad_norm": 13.9375,
      "learning_rate": 3.59851414216299e-07,
      "loss": 0.4725,
      "num_input_tokens_seen": 5119564288,
      "step": 69000,
      "train_runtime": 82886.5855,
      "train_tokens_per_second": 61765.897
    },
    {
      "epoch": 0.9170703184696447,
      "grad_norm": 22.875,
      "learning_rate": 3.587081052640473e-07,
      "loss": 0.8522,
      "num_input_tokens_seen": 5121586688,
      "step": 69010,
      "train_runtime": 82920.0187,
      "train_tokens_per_second": 61765.383
    },
    {
      "epoch": 0.9172032079521066,
      "grad_norm": 36.5,
      "learning_rate": 3.5756658227515175e-07,
      "loss": 0.9637,
      "num_input_tokens_seen": 5122413056,
      "step": 69020,
      "train_runtime": 82932.7908,
      "train_tokens_per_second": 61765.835
    },
    {
      "epoch": 0.9173360974345686,
      "grad_norm": 26.75,
      "learning_rate": 3.564268454610653e-07,
      "loss": 0.7548,
      "num_input_tokens_seen": 5122906368,
      "step": 69030,
      "train_runtime": 82939.907,
      "train_tokens_per_second": 61766.483
    },
    {
      "epoch": 0.9174689869170305,
      "grad_norm": 32.75,
      "learning_rate": 3.552888950329159e-07,
      "loss": 0.5974,
      "num_input_tokens_seen": 5123139584,
      "step": 69040,
      "train_runtime": 82943.7524,
      "train_tokens_per_second": 61766.431
    },
    {
      "epoch": 0.9176018763994923,
      "grad_norm": 21.75,
      "learning_rate": 3.5415273120149917e-07,
      "loss": 0.5289,
      "num_input_tokens_seen": 5123258368,
      "step": 69050,
      "train_runtime": 82946.6628,
      "train_tokens_per_second": 61765.696
    },
    {
      "epoch": 0.9177347658819542,
      "grad_norm": 13.5625,
      "learning_rate": 3.5301835417727693e-07,
      "loss": 0.8122,
      "num_input_tokens_seen": 5125255168,
      "step": 69060,
      "train_runtime": 82978.3276,
      "train_tokens_per_second": 61766.19
    },
    {
      "epoch": 0.9178676553644162,
      "grad_norm": 21.75,
      "learning_rate": 3.5188576417038544e-07,
      "loss": 0.8693,
      "num_input_tokens_seen": 5126085888,
      "step": 69070,
      "train_runtime": 82990.5679,
      "train_tokens_per_second": 61767.09
    },
    {
      "epoch": 0.9180005448468781,
      "grad_norm": 24.125,
      "learning_rate": 3.507549613906247e-07,
      "loss": 0.7932,
      "num_input_tokens_seen": 5126547200,
      "step": 69080,
      "train_runtime": 82997.2456,
      "train_tokens_per_second": 61767.679
    },
    {
      "epoch": 0.91813343432934,
      "grad_norm": 29.625,
      "learning_rate": 3.4962594604746714e-07,
      "loss": 0.6205,
      "num_input_tokens_seen": 5126767360,
      "step": 69090,
      "train_runtime": 83001.0115,
      "train_tokens_per_second": 61767.529
    },
    {
      "epoch": 0.918266323811802,
      "grad_norm": 10.0,
      "learning_rate": 3.4849871835005213e-07,
      "loss": 0.4969,
      "num_input_tokens_seen": 5126881536,
      "step": 69100,
      "train_runtime": 83003.9248,
      "train_tokens_per_second": 61766.736
    },
    {
      "epoch": 0.9183992132942638,
      "grad_norm": 164.0,
      "learning_rate": 3.473732785071882e-07,
      "loss": 0.8559,
      "num_input_tokens_seen": 5128929024,
      "step": 69110,
      "train_runtime": 83038.1997,
      "train_tokens_per_second": 61765.899
    },
    {
      "epoch": 0.9185321027767257,
      "grad_norm": 42.0,
      "learning_rate": 3.4624962672735517e-07,
      "loss": 0.8855,
      "num_input_tokens_seen": 5129761792,
      "step": 69120,
      "train_runtime": 83051.0173,
      "train_tokens_per_second": 61766.393
    },
    {
      "epoch": 0.9186649922591876,
      "grad_norm": 19.0,
      "learning_rate": 3.4512776321869555e-07,
      "loss": 0.75,
      "num_input_tokens_seen": 5130233088,
      "step": 69130,
      "train_runtime": 83057.8406,
      "train_tokens_per_second": 61766.993
    },
    {
      "epoch": 0.9187978817416496,
      "grad_norm": 23.125,
      "learning_rate": 3.4400768818902864e-07,
      "loss": 0.6081,
      "num_input_tokens_seen": 5130481152,
      "step": 69140,
      "train_runtime": 83061.6133,
      "train_tokens_per_second": 61767.174
    },
    {
      "epoch": 0.9189307712241115,
      "grad_norm": 17.5,
      "learning_rate": 3.428894018458351e-07,
      "loss": 0.5717,
      "num_input_tokens_seen": 5130598912,
      "step": 69150,
      "train_runtime": 83064.5832,
      "train_tokens_per_second": 61766.384
    },
    {
      "epoch": 0.9190636607065734,
      "grad_norm": 23.375,
      "learning_rate": 3.417729043962703e-07,
      "loss": 0.8914,
      "num_input_tokens_seen": 5132623616,
      "step": 69160,
      "train_runtime": 83098.2987,
      "train_tokens_per_second": 61765.688
    },
    {
      "epoch": 0.9191965501890353,
      "grad_norm": 24.5,
      "learning_rate": 3.406581960471533e-07,
      "loss": 0.9131,
      "num_input_tokens_seen": 5133473536,
      "step": 69170,
      "train_runtime": 83112.1958,
      "train_tokens_per_second": 61765.587
    },
    {
      "epoch": 0.9193294396714972,
      "grad_norm": 43.5,
      "learning_rate": 3.3954527700497453e-07,
      "loss": 0.7636,
      "num_input_tokens_seen": 5133945600,
      "step": 69180,
      "train_runtime": 83118.8698,
      "train_tokens_per_second": 61766.307
    },
    {
      "epoch": 0.9194623291539591,
      "grad_norm": 23.875,
      "learning_rate": 3.3843414747589456e-07,
      "loss": 0.6627,
      "num_input_tokens_seen": 5134172928,
      "step": 69190,
      "train_runtime": 83122.6422,
      "train_tokens_per_second": 61766.238
    },
    {
      "epoch": 0.919595218636421,
      "grad_norm": 13.625,
      "learning_rate": 3.373248076657365e-07,
      "loss": 0.6116,
      "num_input_tokens_seen": 5134290432,
      "step": 69200,
      "train_runtime": 83125.5288,
      "train_tokens_per_second": 61765.507
    },
    {
      "epoch": 0.919728108118883,
      "grad_norm": 21.75,
      "learning_rate": 3.3621725777999715e-07,
      "loss": 0.9097,
      "num_input_tokens_seen": 5136291840,
      "step": 69210,
      "train_runtime": 83157.6097,
      "train_tokens_per_second": 61765.746
    },
    {
      "epoch": 0.9198609976013449,
      "grad_norm": 17.25,
      "learning_rate": 3.3511149802384013e-07,
      "loss": 0.9214,
      "num_input_tokens_seen": 5137153280,
      "step": 69220,
      "train_runtime": 83171.6718,
      "train_tokens_per_second": 61765.661
    },
    {
      "epoch": 0.9199938870838068,
      "grad_norm": 17.25,
      "learning_rate": 3.340075286020994e-07,
      "loss": 0.7206,
      "num_input_tokens_seen": 5137627392,
      "step": 69230,
      "train_runtime": 83178.7292,
      "train_tokens_per_second": 61766.12
    },
    {
      "epoch": 0.9201267765662686,
      "grad_norm": 15.25,
      "learning_rate": 3.329053497192725e-07,
      "loss": 0.5759,
      "num_input_tokens_seen": 5137851392,
      "step": 69240,
      "train_runtime": 83182.3999,
      "train_tokens_per_second": 61766.088
    },
    {
      "epoch": 0.9202596660487306,
      "grad_norm": 11.75,
      "learning_rate": 3.3180496157953157e-07,
      "loss": 0.5419,
      "num_input_tokens_seen": 5137965568,
      "step": 69250,
      "train_runtime": 83185.4234,
      "train_tokens_per_second": 61765.215
    },
    {
      "epoch": 0.9203925555311925,
      "grad_norm": 18.875,
      "learning_rate": 3.307063643867092e-07,
      "loss": 0.8127,
      "num_input_tokens_seen": 5139878912,
      "step": 69260,
      "train_runtime": 83217.6312,
      "train_tokens_per_second": 61764.302
    },
    {
      "epoch": 0.9205254450136544,
      "grad_norm": 23.75,
      "learning_rate": 3.2960955834431485e-07,
      "loss": 0.8757,
      "num_input_tokens_seen": 5140722688,
      "step": 69270,
      "train_runtime": 83230.3098,
      "train_tokens_per_second": 61765.031
    },
    {
      "epoch": 0.9206583344961163,
      "grad_norm": 19.375,
      "learning_rate": 3.2851454365552037e-07,
      "loss": 0.8094,
      "num_input_tokens_seen": 5141193472,
      "step": 69280,
      "train_runtime": 83237.1078,
      "train_tokens_per_second": 61765.643
    },
    {
      "epoch": 0.9207912239785783,
      "grad_norm": 21.5,
      "learning_rate": 3.27421320523168e-07,
      "loss": 0.6713,
      "num_input_tokens_seen": 5141428992,
      "step": 69290,
      "train_runtime": 83241.029,
      "train_tokens_per_second": 61765.563
    },
    {
      "epoch": 0.9209241134610401,
      "grad_norm": 32.0,
      "learning_rate": 3.263298891497668e-07,
      "loss": 0.5762,
      "num_input_tokens_seen": 5141541376,
      "step": 69300,
      "train_runtime": 83244.6838,
      "train_tokens_per_second": 61764.201
    },
    {
      "epoch": 0.921057002943502,
      "grad_norm": 31.625,
      "learning_rate": 3.2524024973749403e-07,
      "loss": 0.8546,
      "num_input_tokens_seen": 5143608832,
      "step": 69310,
      "train_runtime": 83277.8741,
      "train_tokens_per_second": 61764.411
    },
    {
      "epoch": 0.9211898924259639,
      "grad_norm": 11.625,
      "learning_rate": 3.241524024881992e-07,
      "loss": 0.8716,
      "num_input_tokens_seen": 5144452352,
      "step": 69320,
      "train_runtime": 83290.6851,
      "train_tokens_per_second": 61765.038
    },
    {
      "epoch": 0.9213227819084259,
      "grad_norm": 20.75,
      "learning_rate": 3.230663476033913e-07,
      "loss": 0.7062,
      "num_input_tokens_seen": 5144933632,
      "step": 69330,
      "train_runtime": 83297.6865,
      "train_tokens_per_second": 61765.625
    },
    {
      "epoch": 0.9214556713908878,
      "grad_norm": 10.75,
      "learning_rate": 3.21982085284257e-07,
      "loss": 0.6728,
      "num_input_tokens_seen": 5145160192,
      "step": 69340,
      "train_runtime": 83301.6321,
      "train_tokens_per_second": 61765.419
    },
    {
      "epoch": 0.9215885608733497,
      "grad_norm": 9.4375,
      "learning_rate": 3.208996157316424e-07,
      "loss": 0.549,
      "num_input_tokens_seen": 5145278464,
      "step": 69350,
      "train_runtime": 83305.4276,
      "train_tokens_per_second": 61764.024
    },
    {
      "epoch": 0.9217214503558115,
      "grad_norm": 26.375,
      "learning_rate": 3.198189391460682e-07,
      "loss": 0.8399,
      "num_input_tokens_seen": 5147257600,
      "step": 69360,
      "train_runtime": 83338.6318,
      "train_tokens_per_second": 61763.164
    },
    {
      "epoch": 0.9218543398382735,
      "grad_norm": 17.125,
      "learning_rate": 3.1874005572771983e-07,
      "loss": 0.9435,
      "num_input_tokens_seen": 5148102144,
      "step": 69370,
      "train_runtime": 83351.6624,
      "train_tokens_per_second": 61763.641
    },
    {
      "epoch": 0.9219872293207354,
      "grad_norm": 294.0,
      "learning_rate": 3.176629656764474e-07,
      "loss": 0.7673,
      "num_input_tokens_seen": 5148578816,
      "step": 69380,
      "train_runtime": 83358.6015,
      "train_tokens_per_second": 61764.218
    },
    {
      "epoch": 0.9221201188031973,
      "grad_norm": 15.875,
      "learning_rate": 3.1658766919177685e-07,
      "loss": 0.6241,
      "num_input_tokens_seen": 5148808448,
      "step": 69390,
      "train_runtime": 83363.469,
      "train_tokens_per_second": 61763.366
    },
    {
      "epoch": 0.9222530082856593,
      "grad_norm": 15.0625,
      "learning_rate": 3.1551416647289333e-07,
      "loss": 0.5121,
      "num_input_tokens_seen": 5148924416,
      "step": 69400,
      "train_runtime": 83366.2522,
      "train_tokens_per_second": 61762.695
    },
    {
      "epoch": 0.9223858977681212,
      "grad_norm": 22.25,
      "learning_rate": 3.1444245771865665e-07,
      "loss": 0.8282,
      "num_input_tokens_seen": 5150979840,
      "step": 69410,
      "train_runtime": 83401.2229,
      "train_tokens_per_second": 61761.443
    },
    {
      "epoch": 0.922518787250583,
      "grad_norm": 15.125,
      "learning_rate": 3.1337254312759023e-07,
      "loss": 0.9012,
      "num_input_tokens_seen": 5151826688,
      "step": 69420,
      "train_runtime": 83414.0637,
      "train_tokens_per_second": 61762.087
    },
    {
      "epoch": 0.9226516767330449,
      "grad_norm": 21.375,
      "learning_rate": 3.1230442289788553e-07,
      "loss": 0.7605,
      "num_input_tokens_seen": 5152290816,
      "step": 69430,
      "train_runtime": 83420.828,
      "train_tokens_per_second": 61762.643
    },
    {
      "epoch": 0.9227845662155069,
      "grad_norm": 15.6875,
      "learning_rate": 3.1123809722740426e-07,
      "loss": 0.7142,
      "num_input_tokens_seen": 5152525568,
      "step": 69440,
      "train_runtime": 83424.6123,
      "train_tokens_per_second": 61762.655
    },
    {
      "epoch": 0.9229174556979688,
      "grad_norm": 15.375,
      "learning_rate": 3.101735663136718e-07,
      "loss": 0.464,
      "num_input_tokens_seen": 5152641024,
      "step": 69450,
      "train_runtime": 83427.505,
      "train_tokens_per_second": 61761.898
    },
    {
      "epoch": 0.9230503451804307,
      "grad_norm": 22.0,
      "learning_rate": 3.0911083035388257e-07,
      "loss": 0.8307,
      "num_input_tokens_seen": 5154678272,
      "step": 69460,
      "train_runtime": 83460.295,
      "train_tokens_per_second": 61762.042
    },
    {
      "epoch": 0.9231832346628926,
      "grad_norm": 14.1875,
      "learning_rate": 3.0804988954490024e-07,
      "loss": 0.8619,
      "num_input_tokens_seen": 5155537920,
      "step": 69470,
      "train_runtime": 83473.1178,
      "train_tokens_per_second": 61762.853
    },
    {
      "epoch": 0.9233161241453545,
      "grad_norm": 17.125,
      "learning_rate": 3.0699074408325537e-07,
      "loss": 0.7012,
      "num_input_tokens_seen": 5156027648,
      "step": 69480,
      "train_runtime": 83480.1407,
      "train_tokens_per_second": 61763.524
    },
    {
      "epoch": 0.9234490136278164,
      "grad_norm": 20.5,
      "learning_rate": 3.059333941651432e-07,
      "loss": 0.6839,
      "num_input_tokens_seen": 5156262656,
      "step": 69490,
      "train_runtime": 83483.8166,
      "train_tokens_per_second": 61763.619
    },
    {
      "epoch": 0.9235819031102783,
      "grad_norm": 8.1875,
      "learning_rate": 3.048778399864305e-07,
      "loss": 0.4815,
      "num_input_tokens_seen": 5156376576,
      "step": 69500,
      "train_runtime": 83486.7005,
      "train_tokens_per_second": 61762.85
    },
    {
      "epoch": 0.9237147925927403,
      "grad_norm": 16.875,
      "learning_rate": 3.038240817426474e-07,
      "loss": 0.7987,
      "num_input_tokens_seen": 5158373632,
      "step": 69510,
      "train_runtime": 83519.3663,
      "train_tokens_per_second": 61762.605
    },
    {
      "epoch": 0.9238476820752022,
      "grad_norm": 32.75,
      "learning_rate": 3.0277211962899345e-07,
      "loss": 0.9508,
      "num_input_tokens_seen": 5159229440,
      "step": 69520,
      "train_runtime": 83532.2644,
      "train_tokens_per_second": 61763.314
    },
    {
      "epoch": 0.9239805715576641,
      "grad_norm": 26.0,
      "learning_rate": 3.0172195384033596e-07,
      "loss": 0.7586,
      "num_input_tokens_seen": 5159702784,
      "step": 69530,
      "train_runtime": 83539.0574,
      "train_tokens_per_second": 61763.957
    },
    {
      "epoch": 0.9241134610401259,
      "grad_norm": 16.25,
      "learning_rate": 3.006735845712083e-07,
      "loss": 0.6069,
      "num_input_tokens_seen": 5159926272,
      "step": 69540,
      "train_runtime": 83542.6888,
      "train_tokens_per_second": 61763.948
    },
    {
      "epoch": 0.9242463505225879,
      "grad_norm": 24.625,
      "learning_rate": 2.996270120158129e-07,
      "loss": 0.4972,
      "num_input_tokens_seen": 5160040704,
      "step": 69550,
      "train_runtime": 83545.6385,
      "train_tokens_per_second": 61763.137
    },
    {
      "epoch": 0.9243792400050498,
      "grad_norm": 84.0,
      "learning_rate": 2.985822363680158e-07,
      "loss": 0.8022,
      "num_input_tokens_seen": 5162035712,
      "step": 69560,
      "train_runtime": 83578.5735,
      "train_tokens_per_second": 61762.668
    },
    {
      "epoch": 0.9245121294875117,
      "grad_norm": 36.0,
      "learning_rate": 2.975392578213554e-07,
      "loss": 0.9207,
      "num_input_tokens_seen": 5162885376,
      "step": 69570,
      "train_runtime": 83592.3971,
      "train_tokens_per_second": 61762.619
    },
    {
      "epoch": 0.9246450189699736,
      "grad_norm": 23.875,
      "learning_rate": 2.9649807656902953e-07,
      "loss": 0.7904,
      "num_input_tokens_seen": 5163353088,
      "step": 69580,
      "train_runtime": 83599.289,
      "train_tokens_per_second": 61763.122
    },
    {
      "epoch": 0.9247779084524356,
      "grad_norm": 13.0625,
      "learning_rate": 2.954586928039127e-07,
      "loss": 0.6347,
      "num_input_tokens_seen": 5163589376,
      "step": 69590,
      "train_runtime": 83602.9923,
      "train_tokens_per_second": 61763.212
    },
    {
      "epoch": 0.9249107979348974,
      "grad_norm": 10.4375,
      "learning_rate": 2.944211067185376e-07,
      "loss": 0.5236,
      "num_input_tokens_seen": 5163706880,
      "step": 69600,
      "train_runtime": 83606.0505,
      "train_tokens_per_second": 61762.359
    },
    {
      "epoch": 0.9250436874173593,
      "grad_norm": 14.875,
      "learning_rate": 2.9338531850510943e-07,
      "loss": 0.8128,
      "num_input_tokens_seen": 5165747712,
      "step": 69610,
      "train_runtime": 83639.0988,
      "train_tokens_per_second": 61762.355
    },
    {
      "epoch": 0.9251765768998212,
      "grad_norm": 23.375,
      "learning_rate": 2.923513283555013e-07,
      "loss": 0.9246,
      "num_input_tokens_seen": 5166575872,
      "step": 69620,
      "train_runtime": 83651.6066,
      "train_tokens_per_second": 61763.02
    },
    {
      "epoch": 0.9253094663822832,
      "grad_norm": 15.1875,
      "learning_rate": 2.913191364612444e-07,
      "loss": 0.743,
      "num_input_tokens_seen": 5167069952,
      "step": 69630,
      "train_runtime": 83659.7607,
      "train_tokens_per_second": 61762.906
    },
    {
      "epoch": 0.9254423558647451,
      "grad_norm": 12.6875,
      "learning_rate": 2.902887430135493e-07,
      "loss": 0.5716,
      "num_input_tokens_seen": 5167290112,
      "step": 69640,
      "train_runtime": 83663.5018,
      "train_tokens_per_second": 61762.776
    },
    {
      "epoch": 0.925575245347207,
      "grad_norm": 36.75,
      "learning_rate": 2.892601482032842e-07,
      "loss": 0.5859,
      "num_input_tokens_seen": 5167402496,
      "step": 69650,
      "train_runtime": 83666.568,
      "train_tokens_per_second": 61761.856
    },
    {
      "epoch": 0.925708134829669,
      "grad_norm": 17.75,
      "learning_rate": 2.882333522209868e-07,
      "loss": 0.857,
      "num_input_tokens_seen": 5169481984,
      "step": 69660,
      "train_runtime": 83699.989,
      "train_tokens_per_second": 61762.039
    },
    {
      "epoch": 0.9258410243121308,
      "grad_norm": 15.25,
      "learning_rate": 2.8720835525686386e-07,
      "loss": 0.8887,
      "num_input_tokens_seen": 5170292480,
      "step": 69670,
      "train_runtime": 83712.8459,
      "train_tokens_per_second": 61762.235
    },
    {
      "epoch": 0.9259739137945927,
      "grad_norm": 25.375,
      "learning_rate": 2.8618515750078346e-07,
      "loss": 0.7583,
      "num_input_tokens_seen": 5170763008,
      "step": 69680,
      "train_runtime": 83719.7849,
      "train_tokens_per_second": 61762.736
    },
    {
      "epoch": 0.9261068032770546,
      "grad_norm": 18.75,
      "learning_rate": 2.8516375914228954e-07,
      "loss": 0.6705,
      "num_input_tokens_seen": 5170993408,
      "step": 69690,
      "train_runtime": 83724.7492,
      "train_tokens_per_second": 61761.826
    },
    {
      "epoch": 0.9262396927595166,
      "grad_norm": 10.8125,
      "learning_rate": 2.8414416037057967e-07,
      "loss": 0.5541,
      "num_input_tokens_seen": 5171110400,
      "step": 69700,
      "train_runtime": 83727.5438,
      "train_tokens_per_second": 61761.162
    },
    {
      "epoch": 0.9263725822419785,
      "grad_norm": 38.25,
      "learning_rate": 2.8312636137453277e-07,
      "loss": 0.794,
      "num_input_tokens_seen": 5173216768,
      "step": 69710,
      "train_runtime": 83762.0436,
      "train_tokens_per_second": 61760.871
    },
    {
      "epoch": 0.9265054717244404,
      "grad_norm": 16.5,
      "learning_rate": 2.821103623426824e-07,
      "loss": 0.9198,
      "num_input_tokens_seen": 5174073088,
      "step": 69720,
      "train_runtime": 83774.7188,
      "train_tokens_per_second": 61761.748
    },
    {
      "epoch": 0.9266383612069022,
      "grad_norm": 16.25,
      "learning_rate": 2.8109616346323475e-07,
      "loss": 0.7733,
      "num_input_tokens_seen": 5174526464,
      "step": 69730,
      "train_runtime": 83781.2886,
      "train_tokens_per_second": 61762.316
    },
    {
      "epoch": 0.9267712506893642,
      "grad_norm": 28.25,
      "learning_rate": 2.8008376492406064e-07,
      "loss": 0.6204,
      "num_input_tokens_seen": 5174763520,
      "step": 69740,
      "train_runtime": 83786.0349,
      "train_tokens_per_second": 61761.647
    },
    {
      "epoch": 0.9269041401718261,
      "grad_norm": 9.625,
      "learning_rate": 2.790731669126978e-07,
      "loss": 0.5149,
      "num_input_tokens_seen": 5174882048,
      "step": 69750,
      "train_runtime": 83788.7937,
      "train_tokens_per_second": 61761.028
    },
    {
      "epoch": 0.927037029654288,
      "grad_norm": 22.125,
      "learning_rate": 2.7806436961635206e-07,
      "loss": 0.8282,
      "num_input_tokens_seen": 5176952064,
      "step": 69760,
      "train_runtime": 83821.9571,
      "train_tokens_per_second": 61761.288
    },
    {
      "epoch": 0.92716991913675,
      "grad_norm": 21.5,
      "learning_rate": 2.7705737322189285e-07,
      "loss": 0.9069,
      "num_input_tokens_seen": 5177758720,
      "step": 69770,
      "train_runtime": 83834.7387,
      "train_tokens_per_second": 61761.494
    },
    {
      "epoch": 0.9273028086192119,
      "grad_norm": 21.25,
      "learning_rate": 2.760521779158565e-07,
      "loss": 0.7393,
      "num_input_tokens_seen": 5178233856,
      "step": 69780,
      "train_runtime": 83842.4587,
      "train_tokens_per_second": 61761.474
    },
    {
      "epoch": 0.9274356981016737,
      "grad_norm": 8.4375,
      "learning_rate": 2.7504878388444734e-07,
      "loss": 0.5724,
      "num_input_tokens_seen": 5178461952,
      "step": 69790,
      "train_runtime": 83846.5464,
      "train_tokens_per_second": 61761.184
    },
    {
      "epoch": 0.9275685875841356,
      "grad_norm": 19.75,
      "learning_rate": 2.740471913135356e-07,
      "loss": 0.5196,
      "num_input_tokens_seen": 5178576384,
      "step": 69800,
      "train_runtime": 83849.5156,
      "train_tokens_per_second": 61760.361
    },
    {
      "epoch": 0.9277014770665976,
      "grad_norm": 19.5,
      "learning_rate": 2.7304740038865627e-07,
      "loss": 0.8316,
      "num_input_tokens_seen": 5180550912,
      "step": 69810,
      "train_runtime": 83881.794,
      "train_tokens_per_second": 61760.135
    },
    {
      "epoch": 0.9278343665490595,
      "grad_norm": 14.375,
      "learning_rate": 2.720494112950145e-07,
      "loss": 0.954,
      "num_input_tokens_seen": 5181398528,
      "step": 69820,
      "train_runtime": 83894.641,
      "train_tokens_per_second": 61760.781
    },
    {
      "epoch": 0.9279672560315214,
      "grad_norm": 19.375,
      "learning_rate": 2.710532242174746e-07,
      "loss": 0.7116,
      "num_input_tokens_seen": 5181864704,
      "step": 69830,
      "train_runtime": 83901.5946,
      "train_tokens_per_second": 61761.218
    },
    {
      "epoch": 0.9281001455139833,
      "grad_norm": 11.75,
      "learning_rate": 2.700588393405745e-07,
      "loss": 0.6718,
      "num_input_tokens_seen": 5182096640,
      "step": 69840,
      "train_runtime": 83906.3902,
      "train_tokens_per_second": 61760.453
    },
    {
      "epoch": 0.9282330349964452,
      "grad_norm": 10.25,
      "learning_rate": 2.690662568485136e-07,
      "loss": 0.5258,
      "num_input_tokens_seen": 5182212352,
      "step": 69850,
      "train_runtime": 83909.2195,
      "train_tokens_per_second": 61759.749
    },
    {
      "epoch": 0.9283659244789071,
      "grad_norm": 20.5,
      "learning_rate": 2.6807547692515925e-07,
      "loss": 0.8957,
      "num_input_tokens_seen": 5184183296,
      "step": 69860,
      "train_runtime": 83940.892,
      "train_tokens_per_second": 61759.926
    },
    {
      "epoch": 0.928498813961369,
      "grad_norm": 22.0,
      "learning_rate": 2.6708649975404456e-07,
      "loss": 0.9328,
      "num_input_tokens_seen": 5185020160,
      "step": 69870,
      "train_runtime": 83953.4629,
      "train_tokens_per_second": 61760.647
    },
    {
      "epoch": 0.928631703443831,
      "grad_norm": 992.0,
      "learning_rate": 2.660993255183697e-07,
      "loss": 0.7971,
      "num_input_tokens_seen": 5185527808,
      "step": 69880,
      "train_runtime": 83960.5435,
      "train_tokens_per_second": 61761.484
    },
    {
      "epoch": 0.9287645929262929,
      "grad_norm": 14.4375,
      "learning_rate": 2.651139544010006e-07,
      "loss": 0.617,
      "num_input_tokens_seen": 5185763584,
      "step": 69890,
      "train_runtime": 83965.7498,
      "train_tokens_per_second": 61760.463
    },
    {
      "epoch": 0.9288974824087548,
      "grad_norm": 17.0,
      "learning_rate": 2.6413038658446464e-07,
      "loss": 0.5347,
      "num_input_tokens_seen": 5185877760,
      "step": 69900,
      "train_runtime": 83969.3226,
      "train_tokens_per_second": 61759.195
    },
    {
      "epoch": 0.9290303718912166,
      "grad_norm": 14.1875,
      "learning_rate": 2.6314862225096273e-07,
      "loss": 0.8529,
      "num_input_tokens_seen": 5187934976,
      "step": 69910,
      "train_runtime": 84001.8297,
      "train_tokens_per_second": 61759.785
    },
    {
      "epoch": 0.9291632613736786,
      "grad_norm": 14.1875,
      "learning_rate": 2.62168661582356e-07,
      "loss": 0.893,
      "num_input_tokens_seen": 5188806656,
      "step": 69920,
      "train_runtime": 84014.9535,
      "train_tokens_per_second": 61760.513
    },
    {
      "epoch": 0.9292961508561405,
      "grad_norm": 16.5,
      "learning_rate": 2.611905047601748e-07,
      "loss": 0.7479,
      "num_input_tokens_seen": 5189295616,
      "step": 69930,
      "train_runtime": 84022.0761,
      "train_tokens_per_second": 61761.097
    },
    {
      "epoch": 0.9294290403386024,
      "grad_norm": 25.125,
      "learning_rate": 2.6021415196561316e-07,
      "loss": 0.6408,
      "num_input_tokens_seen": 5189533952,
      "step": 69940,
      "train_runtime": 84025.8332,
      "train_tokens_per_second": 61761.172
    },
    {
      "epoch": 0.9295619298210643,
      "grad_norm": 16.125,
      "learning_rate": 2.592396033795308e-07,
      "loss": 0.5229,
      "num_input_tokens_seen": 5189647872,
      "step": 69950,
      "train_runtime": 84028.6406,
      "train_tokens_per_second": 61760.464
    },
    {
      "epoch": 0.9296948193035263,
      "grad_norm": 29.0,
      "learning_rate": 2.5826685918245665e-07,
      "loss": 0.8209,
      "num_input_tokens_seen": 5191743488,
      "step": 69960,
      "train_runtime": 84064.0817,
      "train_tokens_per_second": 61759.355
    },
    {
      "epoch": 0.9298277087859881,
      "grad_norm": 32.0,
      "learning_rate": 2.5729591955457987e-07,
      "loss": 0.7831,
      "num_input_tokens_seen": 5192585216,
      "step": 69970,
      "train_runtime": 84077.0772,
      "train_tokens_per_second": 61759.821
    },
    {
      "epoch": 0.92996059826845,
      "grad_norm": 28.625,
      "learning_rate": 2.563267846757611e-07,
      "loss": 0.6958,
      "num_input_tokens_seen": 5193068288,
      "step": 69980,
      "train_runtime": 84083.8286,
      "train_tokens_per_second": 61760.607
    },
    {
      "epoch": 0.9300934877509119,
      "grad_norm": 11.75,
      "learning_rate": 2.553594547255234e-07,
      "loss": 0.6343,
      "num_input_tokens_seen": 5193299712,
      "step": 69990,
      "train_runtime": 84087.7122,
      "train_tokens_per_second": 61760.507
    },
    {
      "epoch": 0.9302263772333739,
      "grad_norm": 13.8125,
      "learning_rate": 2.5439392988305223e-07,
      "loss": 0.5046,
      "num_input_tokens_seen": 5193414656,
      "step": 70000,
      "train_runtime": 84090.3784,
      "train_tokens_per_second": 61759.915
    },
    {
      "epoch": 0.9302263772333739,
      "eval_accuracy": 0.8171,
      "eval_balanced_accuracy": 0.8154655908182743,
      "eval_invalid_f1": 0.8348234444143412,
      "eval_invalid_precision": 0.8399055060875886,
      "eval_invalid_recall": 0.829802513464991,
      "eval_loss": 0.36119458079338074,
      "eval_macro_f1": 0.8149696924099262,
      "eval_mcc": 0.6300140749180954,
      "eval_runtime": 91.8825,
      "eval_samples_per_second": 108.835,
      "eval_steps_per_second": 27.209,
      "eval_valid_f1": 0.7951159404055113,
      "eval_valid_precision": 0.7891927951967979,
      "eval_valid_recall": 0.8011286681715576,
      "num_input_tokens_seen": 5193414656,
      "step": 70000
    },
    {
      "epoch": 0.9303592667158358,
      "grad_norm": 37.75,
      "learning_rate": 2.5343021032720907e-07,
      "loss": 0.8169,
      "num_input_tokens_seen": 5195392512,
      "step": 70010,
      "train_runtime": 84222.255,
      "train_tokens_per_second": 61686.694
    },
    {
      "epoch": 0.9304921561982977,
      "grad_norm": 24.5,
      "learning_rate": 2.524682962365088e-07,
      "loss": 0.8496,
      "num_input_tokens_seen": 5196251904,
      "step": 70020,
      "train_runtime": 84234.9527,
      "train_tokens_per_second": 61687.598
    },
    {
      "epoch": 0.9306250456807595,
      "grad_norm": 15.375,
      "learning_rate": 2.515081877891401e-07,
      "loss": 0.7615,
      "num_input_tokens_seen": 5196731136,
      "step": 70030,
      "train_runtime": 84241.7828,
      "train_tokens_per_second": 61688.285
    },
    {
      "epoch": 0.9307579351632215,
      "grad_norm": 18.625,
      "learning_rate": 2.50549885162954e-07,
      "loss": 0.5806,
      "num_input_tokens_seen": 5196972800,
      "step": 70040,
      "train_runtime": 84245.6093,
      "train_tokens_per_second": 61688.352
    },
    {
      "epoch": 0.9308908246456834,
      "grad_norm": 9.8125,
      "learning_rate": 2.4959338853546734e-07,
      "loss": 0.5782,
      "num_input_tokens_seen": 5197089280,
      "step": 70050,
      "train_runtime": 84248.9616,
      "train_tokens_per_second": 61687.28
    },
    {
      "epoch": 0.9310237141281453,
      "grad_norm": 26.875,
      "learning_rate": 2.4863869808386174e-07,
      "loss": 0.8394,
      "num_input_tokens_seen": 5199228672,
      "step": 70060,
      "train_runtime": 84284.0116,
      "train_tokens_per_second": 61687.01
    },
    {
      "epoch": 0.9311566036106073,
      "grad_norm": 38.25,
      "learning_rate": 2.4768581398498695e-07,
      "loss": 0.9672,
      "num_input_tokens_seen": 5200110592,
      "step": 70070,
      "train_runtime": 84297.5454,
      "train_tokens_per_second": 61687.568
    },
    {
      "epoch": 0.9312894930930692,
      "grad_norm": 23.875,
      "learning_rate": 2.4673473641535607e-07,
      "loss": 0.7731,
      "num_input_tokens_seen": 5200606208,
      "step": 70080,
      "train_runtime": 84304.8745,
      "train_tokens_per_second": 61688.084
    },
    {
      "epoch": 0.931422382575531,
      "grad_norm": 17.75,
      "learning_rate": 2.457854655511471e-07,
      "loss": 0.6278,
      "num_input_tokens_seen": 5200840960,
      "step": 70090,
      "train_runtime": 84308.7156,
      "train_tokens_per_second": 61688.058
    },
    {
      "epoch": 0.9315552720579929,
      "grad_norm": 11.625,
      "learning_rate": 2.448380015682028e-07,
      "loss": 0.5416,
      "num_input_tokens_seen": 5200957696,
      "step": 70100,
      "train_runtime": 84311.4025,
      "train_tokens_per_second": 61687.477
    },
    {
      "epoch": 0.9316881615404549,
      "grad_norm": 21.5,
      "learning_rate": 2.438923446420338e-07,
      "loss": 0.8042,
      "num_input_tokens_seen": 5203000832,
      "step": 70110,
      "train_runtime": 84345.3155,
      "train_tokens_per_second": 61686.897
    },
    {
      "epoch": 0.9318210510229168,
      "grad_norm": 14.8125,
      "learning_rate": 2.429484949478134e-07,
      "loss": 0.9531,
      "num_input_tokens_seen": 5203843328,
      "step": 70120,
      "train_runtime": 84359.3724,
      "train_tokens_per_second": 61686.606
    },
    {
      "epoch": 0.9319539405053787,
      "grad_norm": 10.375,
      "learning_rate": 2.420064526603838e-07,
      "loss": 0.697,
      "num_input_tokens_seen": 5204300288,
      "step": 70130,
      "train_runtime": 84366.4364,
      "train_tokens_per_second": 61686.857
    },
    {
      "epoch": 0.9320868299878406,
      "grad_norm": 43.5,
      "learning_rate": 2.410662179542489e-07,
      "loss": 0.6743,
      "num_input_tokens_seen": 5204524288,
      "step": 70140,
      "train_runtime": 84370.047,
      "train_tokens_per_second": 61686.872
    },
    {
      "epoch": 0.9322197194703026,
      "grad_norm": 10.0625,
      "learning_rate": 2.401277910035771e-07,
      "loss": 0.5269,
      "num_input_tokens_seen": 5204634880,
      "step": 70150,
      "train_runtime": 84372.7214,
      "train_tokens_per_second": 61686.227
    },
    {
      "epoch": 0.9323526089527644,
      "grad_norm": 25.0,
      "learning_rate": 2.3919117198220487e-07,
      "loss": 0.864,
      "num_input_tokens_seen": 5206612992,
      "step": 70160,
      "train_runtime": 84404.5584,
      "train_tokens_per_second": 61686.396
    },
    {
      "epoch": 0.9324854984352263,
      "grad_norm": 19.0,
      "learning_rate": 2.3825636106363237e-07,
      "loss": 0.8894,
      "num_input_tokens_seen": 5207436544,
      "step": 70170,
      "train_runtime": 84417.4064,
      "train_tokens_per_second": 61686.763
    },
    {
      "epoch": 0.9326183879176883,
      "grad_norm": 18.125,
      "learning_rate": 2.373233584210255e-07,
      "loss": 0.7498,
      "num_input_tokens_seen": 5207896064,
      "step": 70180,
      "train_runtime": 84425.3902,
      "train_tokens_per_second": 61686.372
    },
    {
      "epoch": 0.9327512774001502,
      "grad_norm": 15.9375,
      "learning_rate": 2.363921642272149e-07,
      "loss": 0.5915,
      "num_input_tokens_seen": 5208126720,
      "step": 70190,
      "train_runtime": 84429.0919,
      "train_tokens_per_second": 61686.4
    },
    {
      "epoch": 0.9328841668826121,
      "grad_norm": 27.0,
      "learning_rate": 2.354627786546937e-07,
      "loss": 0.5176,
      "num_input_tokens_seen": 5208242176,
      "step": 70200,
      "train_runtime": 84431.754,
      "train_tokens_per_second": 61685.822
    },
    {
      "epoch": 0.933017056365074,
      "grad_norm": 23.125,
      "learning_rate": 2.3453520187562528e-07,
      "loss": 0.8514,
      "num_input_tokens_seen": 5210222336,
      "step": 70210,
      "train_runtime": 84464.2103,
      "train_tokens_per_second": 61685.563
    },
    {
      "epoch": 0.9331499458475359,
      "grad_norm": 13.75,
      "learning_rate": 2.3360943406183224e-07,
      "loss": 0.8659,
      "num_input_tokens_seen": 5211046912,
      "step": 70220,
      "train_runtime": 84476.8665,
      "train_tokens_per_second": 61686.082
    },
    {
      "epoch": 0.9332828353299978,
      "grad_norm": 11.625,
      "learning_rate": 2.3268547538480847e-07,
      "loss": 0.7434,
      "num_input_tokens_seen": 5211493120,
      "step": 70230,
      "train_runtime": 84484.9805,
      "train_tokens_per_second": 61685.439
    },
    {
      "epoch": 0.9334157248124597,
      "grad_norm": 21.75,
      "learning_rate": 2.3176332601570595e-07,
      "loss": 0.6661,
      "num_input_tokens_seen": 5211710976,
      "step": 70240,
      "train_runtime": 84488.5325,
      "train_tokens_per_second": 61685.424
    },
    {
      "epoch": 0.9335486142949216,
      "grad_norm": 6.625,
      "learning_rate": 2.3084298612534473e-07,
      "loss": 0.5033,
      "num_input_tokens_seen": 5211822592,
      "step": 70250,
      "train_runtime": 84491.4497,
      "train_tokens_per_second": 61684.616
    },
    {
      "epoch": 0.9336815037773836,
      "grad_norm": 13.3125,
      "learning_rate": 2.2992445588421287e-07,
      "loss": 0.8491,
      "num_input_tokens_seen": 5213814528,
      "step": 70260,
      "train_runtime": 84523.4173,
      "train_tokens_per_second": 61684.853
    },
    {
      "epoch": 0.9338143932598455,
      "grad_norm": 15.8125,
      "learning_rate": 2.2900773546245426e-07,
      "loss": 0.9089,
      "num_input_tokens_seen": 5214618880,
      "step": 70270,
      "train_runtime": 84537.4028,
      "train_tokens_per_second": 61684.162
    },
    {
      "epoch": 0.9339472827423073,
      "grad_norm": 18.5,
      "learning_rate": 2.2809282502988862e-07,
      "loss": 0.783,
      "num_input_tokens_seen": 5215091968,
      "step": 70280,
      "train_runtime": 84544.4019,
      "train_tokens_per_second": 61684.652
    },
    {
      "epoch": 0.9340801722247692,
      "grad_norm": 10.5625,
      "learning_rate": 2.271797247559915e-07,
      "loss": 0.7106,
      "num_input_tokens_seen": 5215321088,
      "step": 70290,
      "train_runtime": 84548.272,
      "train_tokens_per_second": 61684.538
    },
    {
      "epoch": 0.9342130617072312,
      "grad_norm": 16.75,
      "learning_rate": 2.2626843480990756e-07,
      "loss": 0.4986,
      "num_input_tokens_seen": 5215438336,
      "step": 70300,
      "train_runtime": 84551.3856,
      "train_tokens_per_second": 61683.653
    },
    {
      "epoch": 0.9343459511896931,
      "grad_norm": 29.125,
      "learning_rate": 2.2535895536044626e-07,
      "loss": 0.8792,
      "num_input_tokens_seen": 5217446400,
      "step": 70310,
      "train_runtime": 84583.3595,
      "train_tokens_per_second": 61684.076
    },
    {
      "epoch": 0.934478840672155,
      "grad_norm": 11.8125,
      "learning_rate": 2.2445128657607728e-07,
      "loss": 0.91,
      "num_input_tokens_seen": 5218254592,
      "step": 70320,
      "train_runtime": 84595.7694,
      "train_tokens_per_second": 61684.581
    },
    {
      "epoch": 0.934611730154617,
      "grad_norm": 25.625,
      "learning_rate": 2.2354542862494387e-07,
      "loss": 0.7202,
      "num_input_tokens_seen": 5218699776,
      "step": 70330,
      "train_runtime": 84602.582,
      "train_tokens_per_second": 61684.876
    },
    {
      "epoch": 0.9347446196370788,
      "grad_norm": 81.0,
      "learning_rate": 2.2264138167484295e-07,
      "loss": 0.6243,
      "num_input_tokens_seen": 5218929152,
      "step": 70340,
      "train_runtime": 84606.681,
      "train_tokens_per_second": 61684.599
    },
    {
      "epoch": 0.9348775091195407,
      "grad_norm": 15.4375,
      "learning_rate": 2.2173914589324274e-07,
      "loss": 0.5003,
      "num_input_tokens_seen": 5219044352,
      "step": 70350,
      "train_runtime": 84611.0825,
      "train_tokens_per_second": 61682.751
    },
    {
      "epoch": 0.9350103986020026,
      "grad_norm": 44.0,
      "learning_rate": 2.2083872144727514e-07,
      "loss": 0.7789,
      "num_input_tokens_seen": 5221177344,
      "step": 70360,
      "train_runtime": 84645.8233,
      "train_tokens_per_second": 61682.634
    },
    {
      "epoch": 0.9351432880844646,
      "grad_norm": 20.875,
      "learning_rate": 2.1994010850373448e-07,
      "loss": 0.874,
      "num_input_tokens_seen": 5222034944,
      "step": 70370,
      "train_runtime": 84658.9428,
      "train_tokens_per_second": 61683.205
    },
    {
      "epoch": 0.9352761775669265,
      "grad_norm": 19.25,
      "learning_rate": 2.1904330722908206e-07,
      "loss": 0.716,
      "num_input_tokens_seen": 5222498560,
      "step": 70380,
      "train_runtime": 84665.7469,
      "train_tokens_per_second": 61683.724
    },
    {
      "epoch": 0.9354090670493884,
      "grad_norm": 16.625,
      "learning_rate": 2.1814831778944168e-07,
      "loss": 0.6168,
      "num_input_tokens_seen": 5222729472,
      "step": 70390,
      "train_runtime": 84669.5978,
      "train_tokens_per_second": 61683.646
    },
    {
      "epoch": 0.9355419565318502,
      "grad_norm": 8.4375,
      "learning_rate": 2.172551403506029e-07,
      "loss": 0.509,
      "num_input_tokens_seen": 5222844928,
      "step": 70400,
      "train_runtime": 84672.6014,
      "train_tokens_per_second": 61682.821
    },
    {
      "epoch": 0.9356748460143122,
      "grad_norm": 17.0,
      "learning_rate": 2.1636377507801787e-07,
      "loss": 0.7818,
      "num_input_tokens_seen": 5224972288,
      "step": 70410,
      "train_runtime": 84708.3353,
      "train_tokens_per_second": 61681.914
    },
    {
      "epoch": 0.9358077354967741,
      "grad_norm": 16.375,
      "learning_rate": 2.1547422213680448e-07,
      "loss": 0.8662,
      "num_input_tokens_seen": 5225833984,
      "step": 70420,
      "train_runtime": 84721.0876,
      "train_tokens_per_second": 61682.801
    },
    {
      "epoch": 0.935940624979236,
      "grad_norm": 30.75,
      "learning_rate": 2.1458648169174424e-07,
      "loss": 0.6962,
      "num_input_tokens_seen": 5226281216,
      "step": 70430,
      "train_runtime": 84727.9599,
      "train_tokens_per_second": 61683.076
    },
    {
      "epoch": 0.936073514461698,
      "grad_norm": 12.3125,
      "learning_rate": 2.137005539072834e-07,
      "loss": 0.6555,
      "num_input_tokens_seen": 5226498816,
      "step": 70440,
      "train_runtime": 84731.7508,
      "train_tokens_per_second": 61682.885
    },
    {
      "epoch": 0.9362064039441599,
      "grad_norm": 616.0,
      "learning_rate": 2.1281643894753178e-07,
      "loss": 0.5141,
      "num_input_tokens_seen": 5226616832,
      "step": 70450,
      "train_runtime": 84734.5467,
      "train_tokens_per_second": 61682.242
    },
    {
      "epoch": 0.9363392934266217,
      "grad_norm": 38.75,
      "learning_rate": 2.1193413697626507e-07,
      "loss": 0.8354,
      "num_input_tokens_seen": 5228694016,
      "step": 70460,
      "train_runtime": 84769.2991,
      "train_tokens_per_second": 61681.459
    },
    {
      "epoch": 0.9364721829090836,
      "grad_norm": 17.0,
      "learning_rate": 2.1105364815692032e-07,
      "loss": 0.9174,
      "num_input_tokens_seen": 5229526016,
      "step": 70470,
      "train_runtime": 84781.8977,
      "train_tokens_per_second": 61682.106
    },
    {
      "epoch": 0.9366050723915456,
      "grad_norm": 16.0,
      "learning_rate": 2.1017497265260034e-07,
      "loss": 0.7532,
      "num_input_tokens_seen": 5229977088,
      "step": 70480,
      "train_runtime": 84788.962,
      "train_tokens_per_second": 61682.287
    },
    {
      "epoch": 0.9367379618740075,
      "grad_norm": 24.0,
      "learning_rate": 2.0929811062607276e-07,
      "loss": 0.6033,
      "num_input_tokens_seen": 5230204160,
      "step": 70490,
      "train_runtime": 84792.7045,
      "train_tokens_per_second": 61682.242
    },
    {
      "epoch": 0.9368708513564694,
      "grad_norm": 22.375,
      "learning_rate": 2.0842306223976872e-07,
      "loss": 0.4657,
      "num_input_tokens_seen": 5230321152,
      "step": 70500,
      "train_runtime": 84795.4258,
      "train_tokens_per_second": 61681.643
    },
    {
      "epoch": 0.9370037408389313,
      "grad_norm": 39.0,
      "learning_rate": 2.0754982765578303e-07,
      "loss": 0.7811,
      "num_input_tokens_seen": 5232416768,
      "step": 70510,
      "train_runtime": 84829.5315,
      "train_tokens_per_second": 61681.547
    },
    {
      "epoch": 0.9371366303213932,
      "grad_norm": 36.0,
      "learning_rate": 2.0667840703587183e-07,
      "loss": 0.8843,
      "num_input_tokens_seen": 5233341696,
      "step": 70520,
      "train_runtime": 84842.9038,
      "train_tokens_per_second": 61682.727
    },
    {
      "epoch": 0.9372695198038551,
      "grad_norm": 18.375,
      "learning_rate": 2.058088005414627e-07,
      "loss": 0.826,
      "num_input_tokens_seen": 5233843712,
      "step": 70530,
      "train_runtime": 84850.1805,
      "train_tokens_per_second": 61683.354
    },
    {
      "epoch": 0.937402409286317,
      "grad_norm": 21.625,
      "learning_rate": 2.0494100833363785e-07,
      "loss": 0.6492,
      "num_input_tokens_seen": 5234091264,
      "step": 70540,
      "train_runtime": 84855.2762,
      "train_tokens_per_second": 61682.567
    },
    {
      "epoch": 0.937535298768779,
      "grad_norm": 14.5625,
      "learning_rate": 2.0407503057315203e-07,
      "loss": 0.5191,
      "num_input_tokens_seen": 5234213376,
      "step": 70550,
      "train_runtime": 84858.1428,
      "train_tokens_per_second": 61681.922
    },
    {
      "epoch": 0.9376681882512409,
      "grad_norm": 17.25,
      "learning_rate": 2.0321086742041697e-07,
      "loss": 0.7638,
      "num_input_tokens_seen": 5236214272,
      "step": 70560,
      "train_runtime": 84890.4707,
      "train_tokens_per_second": 61682.003
    },
    {
      "epoch": 0.9378010777337028,
      "grad_norm": 16.625,
      "learning_rate": 2.0234851903551345e-07,
      "loss": 0.8965,
      "num_input_tokens_seen": 5237054720,
      "step": 70570,
      "train_runtime": 84903.0468,
      "train_tokens_per_second": 61682.765
    },
    {
      "epoch": 0.9379339672161646,
      "grad_norm": 28.0,
      "learning_rate": 2.014879855781826e-07,
      "loss": 0.7062,
      "num_input_tokens_seen": 5237503744,
      "step": 70580,
      "train_runtime": 84909.9403,
      "train_tokens_per_second": 61683.046
    },
    {
      "epoch": 0.9380668566986265,
      "grad_norm": 21.75,
      "learning_rate": 2.0062926720782915e-07,
      "loss": 0.634,
      "num_input_tokens_seen": 5237723136,
      "step": 70590,
      "train_runtime": 84913.6933,
      "train_tokens_per_second": 61682.903
    },
    {
      "epoch": 0.9381997461810885,
      "grad_norm": 9.6875,
      "learning_rate": 1.9977236408352697e-07,
      "loss": 0.5269,
      "num_input_tokens_seen": 5237837568,
      "step": 70600,
      "train_runtime": 84917.1989,
      "train_tokens_per_second": 61681.704
    },
    {
      "epoch": 0.9383326356635504,
      "grad_norm": 37.0,
      "learning_rate": 1.9891727636400571e-07,
      "loss": 0.8514,
      "num_input_tokens_seen": 5239895808,
      "step": 70610,
      "train_runtime": 84950.7742,
      "train_tokens_per_second": 61681.555
    },
    {
      "epoch": 0.9384655251460123,
      "grad_norm": 24.25,
      "learning_rate": 1.9806400420766424e-07,
      "loss": 0.8772,
      "num_input_tokens_seen": 5240771072,
      "step": 70620,
      "train_runtime": 84963.8768,
      "train_tokens_per_second": 61682.344
    },
    {
      "epoch": 0.9385984146284743,
      "grad_norm": 26.75,
      "learning_rate": 1.9721254777256504e-07,
      "loss": 0.6754,
      "num_input_tokens_seen": 5241253888,
      "step": 70630,
      "train_runtime": 84970.7615,
      "train_tokens_per_second": 61683.028
    },
    {
      "epoch": 0.9387313041109362,
      "grad_norm": 19.0,
      "learning_rate": 1.963629072164297e-07,
      "loss": 0.6882,
      "num_input_tokens_seen": 5241488128,
      "step": 70640,
      "train_runtime": 84974.3866,
      "train_tokens_per_second": 61683.153
    },
    {
      "epoch": 0.938864193593398,
      "grad_norm": 22.125,
      "learning_rate": 1.955150826966501e-07,
      "loss": 0.5103,
      "num_input_tokens_seen": 5241606144,
      "step": 70650,
      "train_runtime": 84978.5631,
      "train_tokens_per_second": 61681.511
    },
    {
      "epoch": 0.9389970830758599,
      "grad_norm": 17.25,
      "learning_rate": 1.9466907437027615e-07,
      "loss": 0.8309,
      "num_input_tokens_seen": 5243742464,
      "step": 70660,
      "train_runtime": 85012.9281,
      "train_tokens_per_second": 61681.706
    },
    {
      "epoch": 0.9391299725583219,
      "grad_norm": 21.375,
      "learning_rate": 1.9382488239402252e-07,
      "loss": 0.8784,
      "num_input_tokens_seen": 5244604160,
      "step": 70670,
      "train_runtime": 85025.6894,
      "train_tokens_per_second": 61682.583
    },
    {
      "epoch": 0.9392628620407838,
      "grad_norm": 18.0,
      "learning_rate": 1.929825069242708e-07,
      "loss": 0.789,
      "num_input_tokens_seen": 5245071616,
      "step": 70680,
      "train_runtime": 85032.724,
      "train_tokens_per_second": 61682.978
    },
    {
      "epoch": 0.9393957515232457,
      "grad_norm": 23.0,
      "learning_rate": 1.9214194811706056e-07,
      "loss": 0.6514,
      "num_input_tokens_seen": 5245303808,
      "step": 70690,
      "train_runtime": 85037.792,
      "train_tokens_per_second": 61682.032
    },
    {
      "epoch": 0.9395286410057077,
      "grad_norm": 10.6875,
      "learning_rate": 1.913032061281006e-07,
      "loss": 0.5209,
      "num_input_tokens_seen": 5245419520,
      "step": 70700,
      "train_runtime": 85040.5451,
      "train_tokens_per_second": 61681.396
    },
    {
      "epoch": 0.9396615304881695,
      "grad_norm": 17.5,
      "learning_rate": 1.904662811127589e-07,
      "loss": 0.7218,
      "num_input_tokens_seen": 5247573248,
      "step": 70710,
      "train_runtime": 85074.6966,
      "train_tokens_per_second": 61681.951
    },
    {
      "epoch": 0.9397944199706314,
      "grad_norm": 39.0,
      "learning_rate": 1.8963117322606915e-07,
      "loss": 0.8549,
      "num_input_tokens_seen": 5248414464,
      "step": 70720,
      "train_runtime": 85087.5473,
      "train_tokens_per_second": 61682.521
    },
    {
      "epoch": 0.9399273094530933,
      "grad_norm": 14.75,
      "learning_rate": 1.8879788262272546e-07,
      "loss": 0.7104,
      "num_input_tokens_seen": 5248920576,
      "step": 70730,
      "train_runtime": 85094.3663,
      "train_tokens_per_second": 61683.526
    },
    {
      "epoch": 0.9400601989355553,
      "grad_norm": 15.0,
      "learning_rate": 1.8796640945709099e-07,
      "loss": 0.6362,
      "num_input_tokens_seen": 5249155584,
      "step": 70740,
      "train_runtime": 85097.9715,
      "train_tokens_per_second": 61683.675
    },
    {
      "epoch": 0.9401930884180172,
      "grad_norm": 8.1875,
      "learning_rate": 1.8713675388318587e-07,
      "loss": 0.5154,
      "num_input_tokens_seen": 5249271552,
      "step": 70750,
      "train_runtime": 85100.3942,
      "train_tokens_per_second": 61683.281
    },
    {
      "epoch": 0.9403259779004791,
      "grad_norm": 21.625,
      "learning_rate": 1.8630891605469714e-07,
      "loss": 0.8523,
      "num_input_tokens_seen": 5251434496,
      "step": 70760,
      "train_runtime": 85134.5242,
      "train_tokens_per_second": 61683.959
    },
    {
      "epoch": 0.9404588673829409,
      "grad_norm": 24.625,
      "learning_rate": 1.8548289612497438e-07,
      "loss": 0.9111,
      "num_input_tokens_seen": 5252276480,
      "step": 70770,
      "train_runtime": 85148.2884,
      "train_tokens_per_second": 61683.876
    },
    {
      "epoch": 0.9405917568654029,
      "grad_norm": 15.625,
      "learning_rate": 1.8465869424703296e-07,
      "loss": 0.6533,
      "num_input_tokens_seen": 5252760320,
      "step": 70780,
      "train_runtime": 85154.9627,
      "train_tokens_per_second": 61684.723
    },
    {
      "epoch": 0.9407246463478648,
      "grad_norm": 9.25,
      "learning_rate": 1.838363105735441e-07,
      "loss": 0.5666,
      "num_input_tokens_seen": 5252993536,
      "step": 70790,
      "train_runtime": 85158.6042,
      "train_tokens_per_second": 61684.824
    },
    {
      "epoch": 0.9408575358303267,
      "grad_norm": 21.0,
      "learning_rate": 1.830157452568515e-07,
      "loss": 0.4981,
      "num_input_tokens_seen": 5253109248,
      "step": 70800,
      "train_runtime": 85161.0676,
      "train_tokens_per_second": 61684.399
    },
    {
      "epoch": 0.9409904253127886,
      "grad_norm": 20.25,
      "learning_rate": 1.8219699844895355e-07,
      "loss": 0.7789,
      "num_input_tokens_seen": 5255295232,
      "step": 70810,
      "train_runtime": 85196.9009,
      "train_tokens_per_second": 61684.113
    },
    {
      "epoch": 0.9411233147952506,
      "grad_norm": 22.75,
      "learning_rate": 1.8138007030151894e-07,
      "loss": 0.9218,
      "num_input_tokens_seen": 5256138752,
      "step": 70820,
      "train_runtime": 85209.6449,
      "train_tokens_per_second": 61684.787
    },
    {
      "epoch": 0.9412562042777124,
      "grad_norm": 16.625,
      "learning_rate": 1.805649609658766e-07,
      "loss": 0.7249,
      "num_input_tokens_seen": 5256619264,
      "step": 70830,
      "train_runtime": 85216.5125,
      "train_tokens_per_second": 61685.454
    },
    {
      "epoch": 0.9413890937601743,
      "grad_norm": 34.5,
      "learning_rate": 1.7975167059301468e-07,
      "loss": 0.6897,
      "num_input_tokens_seen": 5256849664,
      "step": 70840,
      "train_runtime": 85221.0478,
      "train_tokens_per_second": 61684.875
    },
    {
      "epoch": 0.9415219832426363,
      "grad_norm": 8.25,
      "learning_rate": 1.7894019933359264e-07,
      "loss": 0.4786,
      "num_input_tokens_seen": 5256964608,
      "step": 70850,
      "train_runtime": 85223.4496,
      "train_tokens_per_second": 61684.485
    },
    {
      "epoch": 0.9416548727250982,
      "grad_norm": 32.0,
      "learning_rate": 1.7813054733792358e-07,
      "loss": 0.8592,
      "num_input_tokens_seen": 5258998784,
      "step": 70860,
      "train_runtime": 85256.6154,
      "train_tokens_per_second": 61684.349
    },
    {
      "epoch": 0.9417877622075601,
      "grad_norm": 22.5,
      "learning_rate": 1.7732271475599194e-07,
      "loss": 0.8921,
      "num_input_tokens_seen": 5259833600,
      "step": 70870,
      "train_runtime": 85269.2888,
      "train_tokens_per_second": 61684.971
    },
    {
      "epoch": 0.941920651690022,
      "grad_norm": 21.875,
      "learning_rate": 1.765167017374403e-07,
      "loss": 0.7685,
      "num_input_tokens_seen": 5260307200,
      "step": 70880,
      "train_runtime": 85276.0879,
      "train_tokens_per_second": 61685.607
    },
    {
      "epoch": 0.9420535411724839,
      "grad_norm": 83.0,
      "learning_rate": 1.7571250843157472e-07,
      "loss": 0.6407,
      "num_input_tokens_seen": 5260544512,
      "step": 70890,
      "train_runtime": 85280.601,
      "train_tokens_per_second": 61685.125
    },
    {
      "epoch": 0.9421864306549458,
      "grad_norm": 12.5,
      "learning_rate": 1.749101349873661e-07,
      "loss": 0.5758,
      "num_input_tokens_seen": 5260662016,
      "step": 70900,
      "train_runtime": 85283.0268,
      "train_tokens_per_second": 61684.748
    },
    {
      "epoch": 0.9423193201374077,
      "grad_norm": 38.75,
      "learning_rate": 1.7410958155344437e-07,
      "loss": 0.8466,
      "num_input_tokens_seen": 5262689536,
      "step": 70910,
      "train_runtime": 85316.0051,
      "train_tokens_per_second": 61684.669
    },
    {
      "epoch": 0.9424522096198696,
      "grad_norm": 24.0,
      "learning_rate": 1.7331084827810872e-07,
      "loss": 0.8748,
      "num_input_tokens_seen": 5263545600,
      "step": 70920,
      "train_runtime": 85328.573,
      "train_tokens_per_second": 61685.616
    },
    {
      "epoch": 0.9425850991023316,
      "grad_norm": 33.0,
      "learning_rate": 1.7251393530931526e-07,
      "loss": 0.7753,
      "num_input_tokens_seen": 5264039424,
      "step": 70930,
      "train_runtime": 85336.1853,
      "train_tokens_per_second": 61685.9
    },
    {
      "epoch": 0.9427179885847935,
      "grad_norm": 10.6875,
      "learning_rate": 1.7171884279468476e-07,
      "loss": 0.6051,
      "num_input_tokens_seen": 5264262400,
      "step": 70940,
      "train_runtime": 85339.5644,
      "train_tokens_per_second": 61686.071
    },
    {
      "epoch": 0.9428508780672553,
      "grad_norm": 13.25,
      "learning_rate": 1.7092557088150164e-07,
      "loss": 0.5002,
      "num_input_tokens_seen": 5264377344,
      "step": 70950,
      "train_runtime": 85342.01,
      "train_tokens_per_second": 61685.65
    },
    {
      "epoch": 0.9429837675497172,
      "grad_norm": 27.5,
      "learning_rate": 1.7013411971671168e-07,
      "loss": 0.8223,
      "num_input_tokens_seen": 5266420736,
      "step": 70960,
      "train_runtime": 85374.9359,
      "train_tokens_per_second": 61685.794
    },
    {
      "epoch": 0.9431166570321792,
      "grad_norm": 15.625,
      "learning_rate": 1.693444894469265e-07,
      "loss": 0.9104,
      "num_input_tokens_seen": 5267258368,
      "step": 70970,
      "train_runtime": 85387.558,
      "train_tokens_per_second": 61686.486
    },
    {
      "epoch": 0.9432495465146411,
      "grad_norm": 16.625,
      "learning_rate": 1.6855668021841464e-07,
      "loss": 0.6656,
      "num_input_tokens_seen": 5267730432,
      "step": 70980,
      "train_runtime": 85394.2418,
      "train_tokens_per_second": 61687.186
    },
    {
      "epoch": 0.943382435997103,
      "grad_norm": 18.0,
      "learning_rate": 1.6777069217711273e-07,
      "loss": 0.6191,
      "num_input_tokens_seen": 5267959296,
      "step": 70990,
      "train_runtime": 85398.8654,
      "train_tokens_per_second": 61686.526
    },
    {
      "epoch": 0.943515325479565,
      "grad_norm": 15.875,
      "learning_rate": 1.669865254686176e-07,
      "loss": 0.5338,
      "num_input_tokens_seen": 5268077824,
      "step": 71000,
      "train_runtime": 85401.3623,
      "train_tokens_per_second": 61686.11
    },
    {
      "epoch": 0.9436482149620268,
      "grad_norm": 19.875,
      "learning_rate": 1.6620418023818973e-07,
      "loss": 0.8201,
      "num_input_tokens_seen": 5270116352,
      "step": 71010,
      "train_runtime": 85432.3976,
      "train_tokens_per_second": 61687.562
    },
    {
      "epoch": 0.9437811044444887,
      "grad_norm": 16.375,
      "learning_rate": 1.654236566307499e-07,
      "loss": 0.8707,
      "num_input_tokens_seen": 5270922240,
      "step": 71020,
      "train_runtime": 85444.5827,
      "train_tokens_per_second": 61688.197
    },
    {
      "epoch": 0.9439139939269506,
      "grad_norm": 14.125,
      "learning_rate": 1.6464495479088683e-07,
      "loss": 0.7098,
      "num_input_tokens_seen": 5271388672,
      "step": 71030,
      "train_runtime": 85451.1801,
      "train_tokens_per_second": 61688.893
    },
    {
      "epoch": 0.9440468834094126,
      "grad_norm": 16.625,
      "learning_rate": 1.6386807486284406e-07,
      "loss": 0.6999,
      "num_input_tokens_seen": 5271616512,
      "step": 71040,
      "train_runtime": 85455.5394,
      "train_tokens_per_second": 61688.412
    },
    {
      "epoch": 0.9441797728918745,
      "grad_norm": 24.5,
      "learning_rate": 1.6309301699053316e-07,
      "loss": 0.5796,
      "num_input_tokens_seen": 5271729408,
      "step": 71050,
      "train_runtime": 85457.9865,
      "train_tokens_per_second": 61687.966
    },
    {
      "epoch": 0.9443126623743364,
      "grad_norm": 35.75,
      "learning_rate": 1.6231978131752702e-07,
      "loss": 0.8268,
      "num_input_tokens_seen": 5273770752,
      "step": 71060,
      "train_runtime": 85491.8049,
      "train_tokens_per_second": 61687.442
    },
    {
      "epoch": 0.9444455518567983,
      "grad_norm": 18.5,
      "learning_rate": 1.6154836798705998e-07,
      "loss": 0.8927,
      "num_input_tokens_seen": 5274630400,
      "step": 71070,
      "train_runtime": 85504.993,
      "train_tokens_per_second": 61687.981
    },
    {
      "epoch": 0.9445784413392602,
      "grad_norm": 35.25,
      "learning_rate": 1.6077877714202995e-07,
      "loss": 0.7324,
      "num_input_tokens_seen": 5275116032,
      "step": 71080,
      "train_runtime": 85512.1273,
      "train_tokens_per_second": 61688.514
    },
    {
      "epoch": 0.9447113308217221,
      "grad_norm": 78.0,
      "learning_rate": 1.6001100892499732e-07,
      "loss": 0.6572,
      "num_input_tokens_seen": 5275359488,
      "step": 71090,
      "train_runtime": 85515.7974,
      "train_tokens_per_second": 61688.713
    },
    {
      "epoch": 0.944844220304184,
      "grad_norm": 8.875,
      "learning_rate": 1.592450634781839e-07,
      "loss": 0.6024,
      "num_input_tokens_seen": 5275475968,
      "step": 71100,
      "train_runtime": 85518.2252,
      "train_tokens_per_second": 61688.324
    },
    {
      "epoch": 0.944977109786646,
      "grad_norm": 28.25,
      "learning_rate": 1.58480940943474e-07,
      "loss": 0.7791,
      "num_input_tokens_seen": 5277505792,
      "step": 71110,
      "train_runtime": 85551.0934,
      "train_tokens_per_second": 61688.35
    },
    {
      "epoch": 0.9451099992691079,
      "grad_norm": 36.75,
      "learning_rate": 1.5771864146241432e-07,
      "loss": 0.8434,
      "num_input_tokens_seen": 5278349312,
      "step": 71120,
      "train_runtime": 85564.7399,
      "train_tokens_per_second": 61688.37
    },
    {
      "epoch": 0.9452428887515698,
      "grad_norm": 54.5,
      "learning_rate": 1.569581651762131e-07,
      "loss": 0.7037,
      "num_input_tokens_seen": 5278821632,
      "step": 71130,
      "train_runtime": 85571.7594,
      "train_tokens_per_second": 61688.829
    },
    {
      "epoch": 0.9453757782340316,
      "grad_norm": 13.8125,
      "learning_rate": 1.5619951222574425e-07,
      "loss": 0.6377,
      "num_input_tokens_seen": 5279054336,
      "step": 71140,
      "train_runtime": 85575.2525,
      "train_tokens_per_second": 61689.03
    },
    {
      "epoch": 0.9455086677164936,
      "grad_norm": 15.625,
      "learning_rate": 1.5544268275154095e-07,
      "loss": 0.5806,
      "num_input_tokens_seen": 5279171072,
      "step": 71150,
      "train_runtime": 85577.6648,
      "train_tokens_per_second": 61688.655
    },
    {
      "epoch": 0.9456415571989555,
      "grad_norm": 15.6875,
      "learning_rate": 1.546876768937966e-07,
      "loss": 0.7707,
      "num_input_tokens_seen": 5281314048,
      "step": 71160,
      "train_runtime": 85610.7375,
      "train_tokens_per_second": 61689.856
    },
    {
      "epoch": 0.9457744466814174,
      "grad_norm": 15.875,
      "learning_rate": 1.5393449479237267e-07,
      "loss": 0.8697,
      "num_input_tokens_seen": 5282153984,
      "step": 71170,
      "train_runtime": 85623.1803,
      "train_tokens_per_second": 61690.701
    },
    {
      "epoch": 0.9459073361638793,
      "grad_norm": 34.75,
      "learning_rate": 1.5318313658678752e-07,
      "loss": 0.7449,
      "num_input_tokens_seen": 5282631680,
      "step": 71180,
      "train_runtime": 85630.9201,
      "train_tokens_per_second": 61690.703
    },
    {
      "epoch": 0.9460402256463413,
      "grad_norm": 23.75,
      "learning_rate": 1.524336024162232e-07,
      "loss": 0.655,
      "num_input_tokens_seen": 5282866432,
      "step": 71190,
      "train_runtime": 85634.5544,
      "train_tokens_per_second": 61690.826
    },
    {
      "epoch": 0.9461731151288031,
      "grad_norm": 12.75,
      "learning_rate": 1.5168589241952637e-07,
      "loss": 0.5897,
      "num_input_tokens_seen": 5282984192,
      "step": 71200,
      "train_runtime": 85637.0312,
      "train_tokens_per_second": 61690.417
    },
    {
      "epoch": 0.946306004611265,
      "grad_norm": 47.75,
      "learning_rate": 1.509400067351996e-07,
      "loss": 0.8645,
      "num_input_tokens_seen": 5285038592,
      "step": 71210,
      "train_runtime": 85670.3105,
      "train_tokens_per_second": 61690.434
    },
    {
      "epoch": 0.946438894093727,
      "grad_norm": 24.875,
      "learning_rate": 1.501959455014168e-07,
      "loss": 0.8893,
      "num_input_tokens_seen": 5285861120,
      "step": 71220,
      "train_runtime": 85682.818,
      "train_tokens_per_second": 61691.028
    },
    {
      "epoch": 0.9465717835761889,
      "grad_norm": 32.5,
      "learning_rate": 1.494537088560044e-07,
      "loss": 0.8135,
      "num_input_tokens_seen": 5286313216,
      "step": 71230,
      "train_runtime": 85690.2907,
      "train_tokens_per_second": 61690.924
    },
    {
      "epoch": 0.9467046730586508,
      "grad_norm": 15.25,
      "learning_rate": 1.4871329693645687e-07,
      "loss": 0.6407,
      "num_input_tokens_seen": 5286537984,
      "step": 71240,
      "train_runtime": 85693.752,
      "train_tokens_per_second": 61691.055
    },
    {
      "epoch": 0.9468375625411127,
      "grad_norm": 468.0,
      "learning_rate": 1.479747098799289e-07,
      "loss": 0.588,
      "num_input_tokens_seen": 5286656512,
      "step": 71250,
      "train_runtime": 85696.2512,
      "train_tokens_per_second": 61690.639
    },
    {
      "epoch": 0.9469704520235745,
      "grad_norm": 30.25,
      "learning_rate": 1.4723794782323665e-07,
      "loss": 0.8,
      "num_input_tokens_seen": 5288774144,
      "step": 71260,
      "train_runtime": 85729.8557,
      "train_tokens_per_second": 61691.159
    },
    {
      "epoch": 0.9471033415060365,
      "grad_norm": 61.75,
      "learning_rate": 1.465030109028609e-07,
      "loss": 0.921,
      "num_input_tokens_seen": 5289656576,
      "step": 71270,
      "train_runtime": 85743.9795,
      "train_tokens_per_second": 61691.288
    },
    {
      "epoch": 0.9472362309884984,
      "grad_norm": 25.375,
      "learning_rate": 1.4576989925493835e-07,
      "loss": 0.7466,
      "num_input_tokens_seen": 5290160384,
      "step": 71280,
      "train_runtime": 85751.2321,
      "train_tokens_per_second": 61691.946
    },
    {
      "epoch": 0.9473691204709603,
      "grad_norm": 15.5625,
      "learning_rate": 1.450386130152748e-07,
      "loss": 0.6137,
      "num_input_tokens_seen": 5290395904,
      "step": 71290,
      "train_runtime": 85754.8515,
      "train_tokens_per_second": 61692.089
    },
    {
      "epoch": 0.9475020099534223,
      "grad_norm": 11.0,
      "learning_rate": 1.4430915231933184e-07,
      "loss": 0.5276,
      "num_input_tokens_seen": 5290513664,
      "step": 71300,
      "train_runtime": 85757.3469,
      "train_tokens_per_second": 61691.667
    },
    {
      "epoch": 0.9476348994358842,
      "grad_norm": 31.0,
      "learning_rate": 1.4358151730223702e-07,
      "loss": 0.8564,
      "num_input_tokens_seen": 5292484864,
      "step": 71310,
      "train_runtime": 85790.5951,
      "train_tokens_per_second": 61690.735
    },
    {
      "epoch": 0.947767788918346,
      "grad_norm": 37.75,
      "learning_rate": 1.4285570809877912e-07,
      "loss": 0.8746,
      "num_input_tokens_seen": 5293370368,
      "step": 71320,
      "train_runtime": 85803.3938,
      "train_tokens_per_second": 61691.853
    },
    {
      "epoch": 0.9479006784008079,
      "grad_norm": 19.0,
      "learning_rate": 1.4213172484340621e-07,
      "loss": 0.751,
      "num_input_tokens_seen": 5293844736,
      "step": 71330,
      "train_runtime": 85810.2769,
      "train_tokens_per_second": 61692.433
    },
    {
      "epoch": 0.9480335678832699,
      "grad_norm": 23.25,
      "learning_rate": 1.4140956767022984e-07,
      "loss": 0.6605,
      "num_input_tokens_seen": 5294084352,
      "step": 71340,
      "train_runtime": 85813.9193,
      "train_tokens_per_second": 61692.606
    },
    {
      "epoch": 0.9481664573657318,
      "grad_norm": 23.5,
      "learning_rate": 1.4068923671302414e-07,
      "loss": 0.5351,
      "num_input_tokens_seen": 5294203392,
      "step": 71350,
      "train_runtime": 85817.3059,
      "train_tokens_per_second": 61691.559
    },
    {
      "epoch": 0.9482993468481937,
      "grad_norm": 25.875,
      "learning_rate": 1.399707321052235e-07,
      "loss": 0.855,
      "num_input_tokens_seen": 5296288256,
      "step": 71360,
      "train_runtime": 85851.6539,
      "train_tokens_per_second": 61691.162
    },
    {
      "epoch": 0.9484322363306557,
      "grad_norm": 26.5,
      "learning_rate": 1.3925405397992476e-07,
      "loss": 0.9512,
      "num_input_tokens_seen": 5297140736,
      "step": 71370,
      "train_runtime": 85864.3177,
      "train_tokens_per_second": 61691.991
    },
    {
      "epoch": 0.9485651258131175,
      "grad_norm": 20.875,
      "learning_rate": 1.3853920246988505e-07,
      "loss": 0.7257,
      "num_input_tokens_seen": 5297620480,
      "step": 71380,
      "train_runtime": 85871.0753,
      "train_tokens_per_second": 61692.723
    },
    {
      "epoch": 0.9486980152955794,
      "grad_norm": 17.375,
      "learning_rate": 1.3782617770752515e-07,
      "loss": 0.6827,
      "num_input_tokens_seen": 5297853952,
      "step": 71390,
      "train_runtime": 85874.6022,
      "train_tokens_per_second": 61692.908
    },
    {
      "epoch": 0.9488309047780413,
      "grad_norm": 13.875,
      "learning_rate": 1.3711497982492716e-07,
      "loss": 0.5204,
      "num_input_tokens_seen": 5297971456,
      "step": 71400,
      "train_runtime": 85878.3031,
      "train_tokens_per_second": 61691.618
    },
    {
      "epoch": 0.9489637942605033,
      "grad_norm": 39.25,
      "learning_rate": 1.3640560895383236e-07,
      "loss": 0.8005,
      "num_input_tokens_seen": 5300032768,
      "step": 71410,
      "train_runtime": 85912.7295,
      "train_tokens_per_second": 61690.89
    },
    {
      "epoch": 0.9490966837429652,
      "grad_norm": 15.4375,
      "learning_rate": 1.3569806522564788e-07,
      "loss": 0.899,
      "num_input_tokens_seen": 5300892672,
      "step": 71420,
      "train_runtime": 85925.5562,
      "train_tokens_per_second": 61691.689
    },
    {
      "epoch": 0.9492295732254271,
      "grad_norm": 22.375,
      "learning_rate": 1.3499234877143773e-07,
      "loss": 0.7563,
      "num_input_tokens_seen": 5301361920,
      "step": 71430,
      "train_runtime": 85932.4641,
      "train_tokens_per_second": 61692.19
    },
    {
      "epoch": 0.9493624627078889,
      "grad_norm": 12.125,
      "learning_rate": 1.342884597219296e-07,
      "loss": 0.5795,
      "num_input_tokens_seen": 5301596672,
      "step": 71440,
      "train_runtime": 85936.0518,
      "train_tokens_per_second": 61692.346
    },
    {
      "epoch": 0.9494953521903509,
      "grad_norm": 28.75,
      "learning_rate": 1.3358639820751474e-07,
      "loss": 0.5589,
      "num_input_tokens_seen": 5301709312,
      "step": 71450,
      "train_runtime": 85938.6223,
      "train_tokens_per_second": 61691.812
    },
    {
      "epoch": 0.9496282416728128,
      "grad_norm": 18.5,
      "learning_rate": 1.328861643582413e-07,
      "loss": 0.7703,
      "num_input_tokens_seen": 5303656704,
      "step": 71460,
      "train_runtime": 85971.8091,
      "train_tokens_per_second": 61690.649
    },
    {
      "epoch": 0.9497611311552747,
      "grad_norm": 58.75,
      "learning_rate": 1.3218775830382335e-07,
      "loss": 0.9809,
      "num_input_tokens_seen": 5304504832,
      "step": 71470,
      "train_runtime": 85984.5533,
      "train_tokens_per_second": 61691.369
    },
    {
      "epoch": 0.9498940206377366,
      "grad_norm": 18.375,
      "learning_rate": 1.3149118017363293e-07,
      "loss": 0.7052,
      "num_input_tokens_seen": 5304957440,
      "step": 71480,
      "train_runtime": 85990.8513,
      "train_tokens_per_second": 61692.114
    },
    {
      "epoch": 0.9500269101201986,
      "grad_norm": 20.625,
      "learning_rate": 1.3079643009670573e-07,
      "loss": 0.6078,
      "num_input_tokens_seen": 5305180160,
      "step": 71490,
      "train_runtime": 85994.2151,
      "train_tokens_per_second": 61692.291
    },
    {
      "epoch": 0.9501597996026604,
      "grad_norm": 8.9375,
      "learning_rate": 1.3010350820173768e-07,
      "loss": 0.5214,
      "num_input_tokens_seen": 5305293824,
      "step": 71500,
      "train_runtime": 85996.83,
      "train_tokens_per_second": 61691.737
    },
    {
      "epoch": 0.9502926890851223,
      "grad_norm": 47.75,
      "learning_rate": 1.294124146170861e-07,
      "loss": 0.7676,
      "num_input_tokens_seen": 5307358464,
      "step": 71510,
      "train_runtime": 86031.1097,
      "train_tokens_per_second": 61691.154
    },
    {
      "epoch": 0.9504255785675843,
      "grad_norm": 18.375,
      "learning_rate": 1.287231494707719e-07,
      "loss": 0.8928,
      "num_input_tokens_seen": 5308175360,
      "step": 71520,
      "train_runtime": 86043.4585,
      "train_tokens_per_second": 61691.795
    },
    {
      "epoch": 0.9505584680500462,
      "grad_norm": 32.5,
      "learning_rate": 1.280357128904719e-07,
      "loss": 0.678,
      "num_input_tokens_seen": 5308615424,
      "step": 71530,
      "train_runtime": 86050.1215,
      "train_tokens_per_second": 61692.132
    },
    {
      "epoch": 0.9506913575325081,
      "grad_norm": 40.5,
      "learning_rate": 1.2735010500353085e-07,
      "loss": 0.7106,
      "num_input_tokens_seen": 5308843776,
      "step": 71540,
      "train_runtime": 86053.6156,
      "train_tokens_per_second": 61692.28
    },
    {
      "epoch": 0.95082424701497,
      "grad_norm": 15.5625,
      "learning_rate": 1.2666632593694827e-07,
      "loss": 0.5087,
      "num_input_tokens_seen": 5308958976,
      "step": 71550,
      "train_runtime": 86056.0623,
      "train_tokens_per_second": 61691.865
    },
    {
      "epoch": 0.950957136497432,
      "grad_norm": 25.75,
      "learning_rate": 1.2598437581739175e-07,
      "loss": 0.8322,
      "num_input_tokens_seen": 5310999808,
      "step": 71560,
      "train_runtime": 86089.5858,
      "train_tokens_per_second": 61691.548
    },
    {
      "epoch": 0.9510900259798938,
      "grad_norm": 14.8125,
      "learning_rate": 1.2530425477118357e-07,
      "loss": 0.8685,
      "num_input_tokens_seen": 5311859200,
      "step": 71570,
      "train_runtime": 86102.6349,
      "train_tokens_per_second": 61692.179
    },
    {
      "epoch": 0.9512229154623557,
      "grad_norm": 26.375,
      "learning_rate": 1.2462596292431072e-07,
      "loss": 0.7883,
      "num_input_tokens_seen": 5312333824,
      "step": 71580,
      "train_runtime": 86110.3689,
      "train_tokens_per_second": 61692.15
    },
    {
      "epoch": 0.9513558049448176,
      "grad_norm": 17.875,
      "learning_rate": 1.2394950040242271e-07,
      "loss": 0.6558,
      "num_input_tokens_seen": 5312571392,
      "step": 71590,
      "train_runtime": 86113.8816,
      "train_tokens_per_second": 61692.393
    },
    {
      "epoch": 0.9514886944272796,
      "grad_norm": 14.0625,
      "learning_rate": 1.2327486733082483e-07,
      "loss": 0.5475,
      "num_input_tokens_seen": 5312690176,
      "step": 71600,
      "train_runtime": 86116.2875,
      "train_tokens_per_second": 61692.048
    },
    {
      "epoch": 0.9516215839097415,
      "grad_norm": 478.0,
      "learning_rate": 1.2260206383449047e-07,
      "loss": 0.8584,
      "num_input_tokens_seen": 5314671104,
      "step": 71610,
      "train_runtime": 86148.4631,
      "train_tokens_per_second": 61692.001
    },
    {
      "epoch": 0.9517544733922034,
      "grad_norm": 21.625,
      "learning_rate": 1.219310900380477e-07,
      "loss": 0.8879,
      "num_input_tokens_seen": 5315506176,
      "step": 71620,
      "train_runtime": 86160.8347,
      "train_tokens_per_second": 61692.835
    },
    {
      "epoch": 0.9518873628746652,
      "grad_norm": 24.0,
      "learning_rate": 1.2126194606578935e-07,
      "loss": 0.754,
      "num_input_tokens_seen": 5315977728,
      "step": 71630,
      "train_runtime": 86168.4802,
      "train_tokens_per_second": 61692.834
    },
    {
      "epoch": 0.9520202523571272,
      "grad_norm": 21.625,
      "learning_rate": 1.2059463204166843e-07,
      "loss": 0.6587,
      "num_input_tokens_seen": 5316209664,
      "step": 71640,
      "train_runtime": 86171.9667,
      "train_tokens_per_second": 61693.029
    },
    {
      "epoch": 0.9521531418395891,
      "grad_norm": 11.625,
      "learning_rate": 1.1992914808929834e-07,
      "loss": 0.5219,
      "num_input_tokens_seen": 5316323072,
      "step": 71650,
      "train_runtime": 86174.4366,
      "train_tokens_per_second": 61692.577
    },
    {
      "epoch": 0.952286031322051,
      "grad_norm": 36.0,
      "learning_rate": 1.1926549433195488e-07,
      "loss": 0.7692,
      "num_input_tokens_seen": 5318306048,
      "step": 71660,
      "train_runtime": 86206.713,
      "train_tokens_per_second": 61692.482
    },
    {
      "epoch": 0.952418920804513,
      "grad_norm": 16.125,
      "learning_rate": 1.1860367089257529e-07,
      "loss": 0.887,
      "num_input_tokens_seen": 5319130112,
      "step": 71670,
      "train_runtime": 86220.2719,
      "train_tokens_per_second": 61692.338
    },
    {
      "epoch": 0.9525518102869749,
      "grad_norm": 27.5,
      "learning_rate": 1.1794367789375372e-07,
      "loss": 0.7592,
      "num_input_tokens_seen": 5319614464,
      "step": 71680,
      "train_runtime": 86227.166,
      "train_tokens_per_second": 61693.022
    },
    {
      "epoch": 0.9526846997694367,
      "grad_norm": 16.875,
      "learning_rate": 1.1728551545774902e-07,
      "loss": 0.6618,
      "num_input_tokens_seen": 5319857152,
      "step": 71690,
      "train_runtime": 86231.0469,
      "train_tokens_per_second": 61693.06
    },
    {
      "epoch": 0.9528175892518986,
      "grad_norm": 16.875,
      "learning_rate": 1.1662918370647925e-07,
      "loss": 0.5546,
      "num_input_tokens_seen": 5319973632,
      "step": 71700,
      "train_runtime": 86233.5168,
      "train_tokens_per_second": 61692.644
    },
    {
      "epoch": 0.9529504787343606,
      "grad_norm": 88.5,
      "learning_rate": 1.1597468276152602e-07,
      "loss": 0.7821,
      "num_input_tokens_seen": 5322072320,
      "step": 71710,
      "train_runtime": 86267.7544,
      "train_tokens_per_second": 61692.487
    },
    {
      "epoch": 0.9530833682168225,
      "grad_norm": 14.0,
      "learning_rate": 1.1532201274412901e-07,
      "loss": 0.8372,
      "num_input_tokens_seen": 5322896896,
      "step": 71720,
      "train_runtime": 86280.2995,
      "train_tokens_per_second": 61693.074
    },
    {
      "epoch": 0.9532162576992844,
      "grad_norm": 142.0,
      "learning_rate": 1.1467117377518711e-07,
      "loss": 0.703,
      "num_input_tokens_seen": 5323348992,
      "step": 71730,
      "train_runtime": 86286.7368,
      "train_tokens_per_second": 61693.711
    },
    {
      "epoch": 0.9533491471817463,
      "grad_norm": 13.25,
      "learning_rate": 1.1402216597526605e-07,
      "loss": 0.6037,
      "num_input_tokens_seen": 5323572224,
      "step": 71740,
      "train_runtime": 86291.0775,
      "train_tokens_per_second": 61693.194
    },
    {
      "epoch": 0.9534820366642082,
      "grad_norm": 17.625,
      "learning_rate": 1.1337498946458636e-07,
      "loss": 0.4764,
      "num_input_tokens_seen": 5323684352,
      "step": 71750,
      "train_runtime": 86293.5582,
      "train_tokens_per_second": 61692.72
    },
    {
      "epoch": 0.9536149261466701,
      "grad_norm": 23.875,
      "learning_rate": 1.1272964436303324e-07,
      "loss": 0.7818,
      "num_input_tokens_seen": 5325710080,
      "step": 71760,
      "train_runtime": 86326.9925,
      "train_tokens_per_second": 61692.293
    },
    {
      "epoch": 0.953747815629132,
      "grad_norm": 112.5,
      "learning_rate": 1.1208613079014996e-07,
      "loss": 0.9513,
      "num_input_tokens_seen": 5326558208,
      "step": 71770,
      "train_runtime": 86339.5622,
      "train_tokens_per_second": 61693.134
    },
    {
      "epoch": 0.953880705111594,
      "grad_norm": 53.25,
      "learning_rate": 1.1144444886514116e-07,
      "loss": 0.7092,
      "num_input_tokens_seen": 5327022848,
      "step": 71780,
      "train_runtime": 86346.4025,
      "train_tokens_per_second": 61693.628
    },
    {
      "epoch": 0.9540135945940559,
      "grad_norm": 12.9375,
      "learning_rate": 1.108045987068751e-07,
      "loss": 0.6578,
      "num_input_tokens_seen": 5327260672,
      "step": 71790,
      "train_runtime": 86351.0148,
      "train_tokens_per_second": 61693.087
    },
    {
      "epoch": 0.9541464840765178,
      "grad_norm": 19.125,
      "learning_rate": 1.1016658043387585e-07,
      "loss": 0.5661,
      "num_input_tokens_seen": 5327379712,
      "step": 71800,
      "train_runtime": 86353.5377,
      "train_tokens_per_second": 61692.663
    },
    {
      "epoch": 0.9542793735589796,
      "grad_norm": 23.25,
      "learning_rate": 1.095303941643322e-07,
      "loss": 0.8199,
      "num_input_tokens_seen": 5329414144,
      "step": 71810,
      "train_runtime": 86386.6773,
      "train_tokens_per_second": 61692.547
    },
    {
      "epoch": 0.9544122630414416,
      "grad_norm": 18.375,
      "learning_rate": 1.08896040016091e-07,
      "loss": 0.8943,
      "num_input_tokens_seen": 5330272256,
      "step": 71820,
      "train_runtime": 86399.3119,
      "train_tokens_per_second": 61693.457
    },
    {
      "epoch": 0.9545451525239035,
      "grad_norm": 18.25,
      "learning_rate": 1.0826351810666047e-07,
      "loss": 0.7703,
      "num_input_tokens_seen": 5330700544,
      "step": 71830,
      "train_runtime": 86405.7325,
      "train_tokens_per_second": 61693.83
    },
    {
      "epoch": 0.9546780420063654,
      "grad_norm": 34.5,
      "learning_rate": 1.0763282855321133e-07,
      "loss": 0.6541,
      "num_input_tokens_seen": 5330924544,
      "step": 71840,
      "train_runtime": 86409.248,
      "train_tokens_per_second": 61693.912
    },
    {
      "epoch": 0.9548109314888273,
      "grad_norm": 84.0,
      "learning_rate": 1.0700397147257125e-07,
      "loss": 0.5217,
      "num_input_tokens_seen": 5331035392,
      "step": 71850,
      "train_runtime": 86411.753,
      "train_tokens_per_second": 61693.406
    },
    {
      "epoch": 0.9549438209712893,
      "grad_norm": 20.375,
      "learning_rate": 1.0637694698123146e-07,
      "loss": 0.7947,
      "num_input_tokens_seen": 5333063424,
      "step": 71860,
      "train_runtime": 86444.6845,
      "train_tokens_per_second": 61693.364
    },
    {
      "epoch": 0.9550767104537511,
      "grad_norm": 17.5,
      "learning_rate": 1.0575175519534242e-07,
      "loss": 0.9,
      "num_input_tokens_seen": 5333947392,
      "step": 71870,
      "train_runtime": 86457.6766,
      "train_tokens_per_second": 61694.318
    },
    {
      "epoch": 0.955209599936213,
      "grad_norm": 18.5,
      "learning_rate": 1.051283962307148e-07,
      "loss": 0.718,
      "num_input_tokens_seen": 5334432000,
      "step": 71880,
      "train_runtime": 86464.6125,
      "train_tokens_per_second": 61694.974
    },
    {
      "epoch": 0.955342489418675,
      "grad_norm": 30.25,
      "learning_rate": 1.045068702028218e-07,
      "loss": 0.6391,
      "num_input_tokens_seen": 5334662400,
      "step": 71890,
      "train_runtime": 86468.1775,
      "train_tokens_per_second": 61695.095
    },
    {
      "epoch": 0.9554753789011369,
      "grad_norm": 13.3125,
      "learning_rate": 1.0388717722679354e-07,
      "loss": 0.5118,
      "num_input_tokens_seen": 5334778112,
      "step": 71900,
      "train_runtime": 86470.966,
      "train_tokens_per_second": 61694.443
    },
    {
      "epoch": 0.9556082683835988,
      "grad_norm": 19.0,
      "learning_rate": 1.0326931741742486e-07,
      "loss": 0.8775,
      "num_input_tokens_seen": 5336704512,
      "step": 71910,
      "train_runtime": 86502.9766,
      "train_tokens_per_second": 61693.883
    },
    {
      "epoch": 0.9557411578660607,
      "grad_norm": 20.125,
      "learning_rate": 1.0265329088916754e-07,
      "loss": 0.8616,
      "num_input_tokens_seen": 5337506048,
      "step": 71920,
      "train_runtime": 86515.318,
      "train_tokens_per_second": 61694.347
    },
    {
      "epoch": 0.9558740473485225,
      "grad_norm": 29.5,
      "learning_rate": 1.0203909775613474e-07,
      "loss": 0.68,
      "num_input_tokens_seen": 5337967104,
      "step": 71930,
      "train_runtime": 86521.7964,
      "train_tokens_per_second": 61695.056
    },
    {
      "epoch": 0.9560069368309845,
      "grad_norm": 14.0625,
      "learning_rate": 1.0142673813209991e-07,
      "loss": 0.6022,
      "num_input_tokens_seen": 5338190080,
      "step": 71940,
      "train_runtime": 86525.4032,
      "train_tokens_per_second": 61695.062
    },
    {
      "epoch": 0.9561398263134464,
      "grad_norm": 12.8125,
      "learning_rate": 1.0081621213049897e-07,
      "loss": 0.5544,
      "num_input_tokens_seen": 5338305024,
      "step": 71950,
      "train_runtime": 86527.9254,
      "train_tokens_per_second": 61694.592
    },
    {
      "epoch": 0.9562727157959083,
      "grad_norm": 18.0,
      "learning_rate": 1.0020751986442478e-07,
      "loss": 0.8885,
      "num_input_tokens_seen": 5340365824,
      "step": 71960,
      "train_runtime": 86563.0898,
      "train_tokens_per_second": 61693.336
    },
    {
      "epoch": 0.9564056052783703,
      "grad_norm": 23.0,
      "learning_rate": 9.960066144663271e-08,
      "loss": 0.914,
      "num_input_tokens_seen": 5341274112,
      "step": 71970,
      "train_runtime": 86576.6129,
      "train_tokens_per_second": 61694.191
    },
    {
      "epoch": 0.9565384947608322,
      "grad_norm": 13.5625,
      "learning_rate": 9.899563698953729e-08,
      "loss": 0.7168,
      "num_input_tokens_seen": 5341748480,
      "step": 71980,
      "train_runtime": 86585.127,
      "train_tokens_per_second": 61693.603
    },
    {
      "epoch": 0.956671384243294,
      "grad_norm": 53.75,
      "learning_rate": 9.839244660521552e-08,
      "loss": 0.6586,
      "num_input_tokens_seen": 5341982720,
      "step": 71990,
      "train_runtime": 86588.9316,
      "train_tokens_per_second": 61693.598
    },
    {
      "epoch": 0.9568042737257559,
      "grad_norm": 11.625,
      "learning_rate": 9.779109040540246e-08,
      "loss": 0.5452,
      "num_input_tokens_seen": 5342099712,
      "step": 72000,
      "train_runtime": 86591.5203,
      "train_tokens_per_second": 61693.105
    },
    {
      "epoch": 0.9569371632082179,
      "grad_norm": 14.625,
      "learning_rate": 9.719156850149236e-08,
      "loss": 0.8217,
      "num_input_tokens_seen": 5344175616,
      "step": 72010,
      "train_runtime": 86625.2019,
      "train_tokens_per_second": 61693.081
    },
    {
      "epoch": 0.9570700526906798,
      "grad_norm": 41.0,
      "learning_rate": 9.659388100454193e-08,
      "loss": 0.8309,
      "num_input_tokens_seen": 5345005312,
      "step": 72020,
      "train_runtime": 86637.6855,
      "train_tokens_per_second": 61693.768
    },
    {
      "epoch": 0.9572029421731417,
      "grad_norm": 15.125,
      "learning_rate": 9.599802802526925e-08,
      "loss": 0.7136,
      "num_input_tokens_seen": 5345477120,
      "step": 72030,
      "train_runtime": 86645.7837,
      "train_tokens_per_second": 61693.448
    },
    {
      "epoch": 0.9573358316556037,
      "grad_norm": 12.5625,
      "learning_rate": 9.54040096740494e-08,
      "loss": 0.6531,
      "num_input_tokens_seen": 5345713920,
      "step": 72040,
      "train_runtime": 86649.5659,
      "train_tokens_per_second": 61693.488
    },
    {
      "epoch": 0.9574687211380656,
      "grad_norm": 12.4375,
      "learning_rate": 9.481182606091876e-08,
      "loss": 0.533,
      "num_input_tokens_seen": 5345832448,
      "step": 72050,
      "train_runtime": 86652.3477,
      "train_tokens_per_second": 61692.875
    },
    {
      "epoch": 0.9576016106205274,
      "grad_norm": 24.875,
      "learning_rate": 9.422147729557518e-08,
      "loss": 0.8549,
      "num_input_tokens_seen": 5348008960,
      "step": 72060,
      "train_runtime": 86687.0253,
      "train_tokens_per_second": 61693.303
    },
    {
      "epoch": 0.9577345001029893,
      "grad_norm": 50.5,
      "learning_rate": 9.363296348737339e-08,
      "loss": 0.9383,
      "num_input_tokens_seen": 5348853504,
      "step": 72070,
      "train_runtime": 86700.9845,
      "train_tokens_per_second": 61693.111
    },
    {
      "epoch": 0.9578673895854513,
      "grad_norm": 20.875,
      "learning_rate": 9.304628474533283e-08,
      "loss": 0.7581,
      "num_input_tokens_seen": 5349307904,
      "step": 72080,
      "train_runtime": 86707.6423,
      "train_tokens_per_second": 61693.615
    },
    {
      "epoch": 0.9580002790679132,
      "grad_norm": 50.25,
      "learning_rate": 9.246144117812883e-08,
      "loss": 0.6155,
      "num_input_tokens_seen": 5349537024,
      "step": 72090,
      "train_runtime": 86711.1159,
      "train_tokens_per_second": 61693.786
    },
    {
      "epoch": 0.9581331685503751,
      "grad_norm": 10.1875,
      "learning_rate": 9.187843289409803e-08,
      "loss": 0.506,
      "num_input_tokens_seen": 5349649408,
      "step": 72100,
      "train_runtime": 86713.6509,
      "train_tokens_per_second": 61693.278
    },
    {
      "epoch": 0.958266058032837,
      "grad_norm": 20.75,
      "learning_rate": 9.129726000124073e-08,
      "loss": 0.7541,
      "num_input_tokens_seen": 5351721728,
      "step": 72110,
      "train_runtime": 86747.1306,
      "train_tokens_per_second": 61693.357
    },
    {
      "epoch": 0.9583989475152989,
      "grad_norm": 118.5,
      "learning_rate": 9.071792260720969e-08,
      "loss": 0.8571,
      "num_input_tokens_seen": 5352528384,
      "step": 72120,
      "train_runtime": 86759.3838,
      "train_tokens_per_second": 61693.942
    },
    {
      "epoch": 0.9585318369977608,
      "grad_norm": 18.875,
      "learning_rate": 9.014042081932462e-08,
      "loss": 0.701,
      "num_input_tokens_seen": 5352965888,
      "step": 72130,
      "train_runtime": 86765.882,
      "train_tokens_per_second": 61694.364
    },
    {
      "epoch": 0.9586647264802227,
      "grad_norm": 15.125,
      "learning_rate": 8.956475474456216e-08,
      "loss": 0.5888,
      "num_input_tokens_seen": 5353200896,
      "step": 72140,
      "train_runtime": 86770.424,
      "train_tokens_per_second": 61693.843
    },
    {
      "epoch": 0.9587976159626846,
      "grad_norm": 16.25,
      "learning_rate": 8.899092448955926e-08,
      "loss": 0.5503,
      "num_input_tokens_seen": 5353319424,
      "step": 72150,
      "train_runtime": 86773.1121,
      "train_tokens_per_second": 61693.298
    },
    {
      "epoch": 0.9589305054451466,
      "grad_norm": 44.25,
      "learning_rate": 8.84189301606131e-08,
      "loss": 0.8359,
      "num_input_tokens_seen": 5355314176,
      "step": 72160,
      "train_runtime": 86805.4992,
      "train_tokens_per_second": 61693.259
    },
    {
      "epoch": 0.9590633949276085,
      "grad_norm": 15.8125,
      "learning_rate": 8.78487718636789e-08,
      "loss": 0.9817,
      "num_input_tokens_seen": 5356153856,
      "step": 72170,
      "train_runtime": 86817.9286,
      "train_tokens_per_second": 61694.099
    },
    {
      "epoch": 0.9591962844100703,
      "grad_norm": 15.75,
      "learning_rate": 8.728044970437666e-08,
      "loss": 0.7661,
      "num_input_tokens_seen": 5356610048,
      "step": 72180,
      "train_runtime": 86824.5143,
      "train_tokens_per_second": 61694.673
    },
    {
      "epoch": 0.9593291738925322,
      "grad_norm": 18.75,
      "learning_rate": 8.671396378797992e-08,
      "loss": 0.6183,
      "num_input_tokens_seen": 5356837888,
      "step": 72190,
      "train_runtime": 86829.9558,
      "train_tokens_per_second": 61693.431
    },
    {
      "epoch": 0.9594620633749942,
      "grad_norm": 9.3125,
      "learning_rate": 8.614931421942584e-08,
      "loss": 0.5305,
      "num_input_tokens_seen": 5356950784,
      "step": 72200,
      "train_runtime": 86832.571,
      "train_tokens_per_second": 61692.873
    },
    {
      "epoch": 0.9595949528574561,
      "grad_norm": 15.3125,
      "learning_rate": 8.558650110331191e-08,
      "loss": 0.8127,
      "num_input_tokens_seen": 5358910976,
      "step": 72210,
      "train_runtime": 86864.4178,
      "train_tokens_per_second": 61692.821
    },
    {
      "epoch": 0.959727842339918,
      "grad_norm": 12.875,
      "learning_rate": 8.502552454389246e-08,
      "loss": 0.928,
      "num_input_tokens_seen": 5359728384,
      "step": 72220,
      "train_runtime": 86876.8115,
      "train_tokens_per_second": 61693.429
    },
    {
      "epoch": 0.95986073182238,
      "grad_norm": 22.875,
      "learning_rate": 8.44663846450855e-08,
      "loss": 0.7818,
      "num_input_tokens_seen": 5360209152,
      "step": 72230,
      "train_runtime": 86884.59,
      "train_tokens_per_second": 61693.439
    },
    {
      "epoch": 0.9599936213048418,
      "grad_norm": 16.375,
      "learning_rate": 8.390908151046485e-08,
      "loss": 0.6491,
      "num_input_tokens_seen": 5360429568,
      "step": 72240,
      "train_runtime": 86887.994,
      "train_tokens_per_second": 61693.559
    },
    {
      "epoch": 0.9601265107873037,
      "grad_norm": 16.75,
      "learning_rate": 8.335361524326679e-08,
      "loss": 0.5348,
      "num_input_tokens_seen": 5360544768,
      "step": 72250,
      "train_runtime": 86890.4997,
      "train_tokens_per_second": 61693.106
    },
    {
      "epoch": 0.9602594002697656,
      "grad_norm": 13.6875,
      "learning_rate": 8.279998594638683e-08,
      "loss": 0.8033,
      "num_input_tokens_seen": 5362497792,
      "step": 72260,
      "train_runtime": 86921.9815,
      "train_tokens_per_second": 61693.23
    },
    {
      "epoch": 0.9603922897522276,
      "grad_norm": 25.125,
      "learning_rate": 8.224819372237958e-08,
      "loss": 0.9206,
      "num_input_tokens_seen": 5363357696,
      "step": 72270,
      "train_runtime": 86934.558,
      "train_tokens_per_second": 61694.196
    },
    {
      "epoch": 0.9605251792346895,
      "grad_norm": 13.25,
      "learning_rate": 8.169823867345883e-08,
      "loss": 0.7577,
      "num_input_tokens_seen": 5363812096,
      "step": 72280,
      "train_runtime": 86941.2345,
      "train_tokens_per_second": 61694.685
    },
    {
      "epoch": 0.9606580687171514,
      "grad_norm": 21.0,
      "learning_rate": 8.115012090149977e-08,
      "loss": 0.6333,
      "num_input_tokens_seen": 5364040192,
      "step": 72290,
      "train_runtime": 86944.7357,
      "train_tokens_per_second": 61694.824
    },
    {
      "epoch": 0.9607909581996132,
      "grad_norm": 24.0,
      "learning_rate": 8.06038405080356e-08,
      "loss": 0.5572,
      "num_input_tokens_seen": 5364152320,
      "step": 72300,
      "train_runtime": 86947.2197,
      "train_tokens_per_second": 61694.351
    },
    {
      "epoch": 0.9609238476820752,
      "grad_norm": 14.1875,
      "learning_rate": 8.005939759426095e-08,
      "loss": 0.8308,
      "num_input_tokens_seen": 5366195456,
      "step": 72310,
      "train_runtime": 86980.0688,
      "train_tokens_per_second": 61694.541
    },
    {
      "epoch": 0.9610567371645371,
      "grad_norm": 12.875,
      "learning_rate": 7.951679226102849e-08,
      "loss": 0.8108,
      "num_input_tokens_seen": 5367016448,
      "step": 72320,
      "train_runtime": 86992.4574,
      "train_tokens_per_second": 61695.193
    },
    {
      "epoch": 0.961189626646999,
      "grad_norm": 24.25,
      "learning_rate": 7.897602460885112e-08,
      "loss": 0.7395,
      "num_input_tokens_seen": 5367490304,
      "step": 72330,
      "train_runtime": 86999.1799,
      "train_tokens_per_second": 61695.872
    },
    {
      "epoch": 0.961322516129461,
      "grad_norm": 22.625,
      "learning_rate": 7.843709473789985e-08,
      "loss": 0.6827,
      "num_input_tokens_seen": 5367726592,
      "step": 72340,
      "train_runtime": 87002.8621,
      "train_tokens_per_second": 61695.977
    },
    {
      "epoch": 0.9614554056119229,
      "grad_norm": 21.0,
      "learning_rate": 7.790000274800923e-08,
      "loss": 0.5561,
      "num_input_tokens_seen": 5367845376,
      "step": 72350,
      "train_runtime": 87005.3583,
      "train_tokens_per_second": 61695.572
    },
    {
      "epoch": 0.9615882950943847,
      "grad_norm": 32.75,
      "learning_rate": 7.736474873866861e-08,
      "loss": 0.8113,
      "num_input_tokens_seen": 5370016768,
      "step": 72360,
      "train_runtime": 87039.3993,
      "train_tokens_per_second": 61696.39
    },
    {
      "epoch": 0.9617211845768466,
      "grad_norm": 33.25,
      "learning_rate": 7.683133280902976e-08,
      "loss": 0.9256,
      "num_input_tokens_seen": 5370902784,
      "step": 72370,
      "train_runtime": 87053.3672,
      "train_tokens_per_second": 61696.669
    },
    {
      "epoch": 0.9618540740593086,
      "grad_norm": 28.25,
      "learning_rate": 7.629975505790366e-08,
      "loss": 0.6839,
      "num_input_tokens_seen": 5371392256,
      "step": 72380,
      "train_runtime": 87060.2449,
      "train_tokens_per_second": 61697.417
    },
    {
      "epoch": 0.9619869635417705,
      "grad_norm": 18.25,
      "learning_rate": 7.577001558375929e-08,
      "loss": 0.664,
      "num_input_tokens_seen": 5371630336,
      "step": 72390,
      "train_runtime": 87064.0304,
      "train_tokens_per_second": 61697.469
    },
    {
      "epoch": 0.9621198530242324,
      "grad_norm": 14.1875,
      "learning_rate": 7.524211448472707e-08,
      "loss": 0.525,
      "num_input_tokens_seen": 5371747328,
      "step": 72400,
      "train_runtime": 87066.6695,
      "train_tokens_per_second": 61696.943
    },
    {
      "epoch": 0.9622527425066943,
      "grad_norm": 39.5,
      "learning_rate": 7.471605185859654e-08,
      "loss": 0.7969,
      "num_input_tokens_seen": 5373842688,
      "step": 72410,
      "train_runtime": 87100.1218,
      "train_tokens_per_second": 61697.304
    },
    {
      "epoch": 0.9623856319891562,
      "grad_norm": 18.5,
      "learning_rate": 7.41918278028142e-08,
      "loss": 0.8731,
      "num_input_tokens_seen": 5374671872,
      "step": 72420,
      "train_runtime": 87113.6427,
      "train_tokens_per_second": 61697.246
    },
    {
      "epoch": 0.9625185214716181,
      "grad_norm": 47.75,
      "learning_rate": 7.366944241449014e-08,
      "loss": 0.7896,
      "num_input_tokens_seen": 5375133440,
      "step": 72430,
      "train_runtime": 87120.3164,
      "train_tokens_per_second": 61697.818
    },
    {
      "epoch": 0.96265141095408,
      "grad_norm": 37.0,
      "learning_rate": 7.31488957903892e-08,
      "loss": 0.6121,
      "num_input_tokens_seen": 5375369728,
      "step": 72440,
      "train_runtime": 87123.9071,
      "train_tokens_per_second": 61697.987
    },
    {
      "epoch": 0.962784300436542,
      "grad_norm": 12.0625,
      "learning_rate": 7.263018802693978e-08,
      "loss": 0.5383,
      "num_input_tokens_seen": 5375487488,
      "step": 72450,
      "train_runtime": 87126.4206,
      "train_tokens_per_second": 61697.559
    },
    {
      "epoch": 0.9629171899190039,
      "grad_norm": 24.375,
      "learning_rate": 7.211331922022724e-08,
      "loss": 0.7864,
      "num_input_tokens_seen": 5377523968,
      "step": 72460,
      "train_runtime": 87160.8544,
      "train_tokens_per_second": 61696.55
    },
    {
      "epoch": 0.9630500794014658,
      "grad_norm": 18.625,
      "learning_rate": 7.159828946599612e-08,
      "loss": 0.871,
      "num_input_tokens_seen": 5378361856,
      "step": 72470,
      "train_runtime": 87173.3856,
      "train_tokens_per_second": 61697.292
    },
    {
      "epoch": 0.9631829688839277,
      "grad_norm": 18.5,
      "learning_rate": 7.108509885965342e-08,
      "loss": 0.7588,
      "num_input_tokens_seen": 5378852096,
      "step": 72480,
      "train_runtime": 87180.3831,
      "train_tokens_per_second": 61697.964
    },
    {
      "epoch": 0.9633158583663896,
      "grad_norm": 416.0,
      "learning_rate": 7.05737474962609e-08,
      "loss": 0.6523,
      "num_input_tokens_seen": 5379089152,
      "step": 72490,
      "train_runtime": 87184.0541,
      "train_tokens_per_second": 61698.085
    },
    {
      "epoch": 0.9634487478488515,
      "grad_norm": 13.125,
      "learning_rate": 7.006423547054275e-08,
      "loss": 0.4999,
      "num_input_tokens_seen": 5379207936,
      "step": 72500,
      "train_runtime": 87186.8375,
      "train_tokens_per_second": 61697.477
    },
    {
      "epoch": 0.9635816373313134,
      "grad_norm": 35.75,
      "learning_rate": 6.955656287688128e-08,
      "loss": 0.8449,
      "num_input_tokens_seen": 5381222400,
      "step": 72510,
      "train_runtime": 87218.7917,
      "train_tokens_per_second": 61697.97
    },
    {
      "epoch": 0.9637145268137753,
      "grad_norm": 27.875,
      "learning_rate": 6.905072980931793e-08,
      "loss": 0.8594,
      "num_input_tokens_seen": 5382023168,
      "step": 72520,
      "train_runtime": 87231.1132,
      "train_tokens_per_second": 61698.435
    },
    {
      "epoch": 0.9638474162962373,
      "grad_norm": 24.5,
      "learning_rate": 6.85467363615544e-08,
      "loss": 0.7906,
      "num_input_tokens_seen": 5382461696,
      "step": 72530,
      "train_runtime": 87237.4244,
      "train_tokens_per_second": 61698.998
    },
    {
      "epoch": 0.9639803057786992,
      "grad_norm": 22.5,
      "learning_rate": 6.804458262695157e-08,
      "loss": 0.6594,
      "num_input_tokens_seen": 5382682624,
      "step": 72540,
      "train_runtime": 87241.9116,
      "train_tokens_per_second": 61698.357
    },
    {
      "epoch": 0.964113195261161,
      "grad_norm": 18.375,
      "learning_rate": 6.754426869852726e-08,
      "loss": 0.6326,
      "num_input_tokens_seen": 5382799872,
      "step": 72550,
      "train_runtime": 87244.6503,
      "train_tokens_per_second": 61697.764
    },
    {
      "epoch": 0.9642460847436229,
      "grad_norm": 26.25,
      "learning_rate": 6.704579466896289e-08,
      "loss": 0.7791,
      "num_input_tokens_seen": 5384925184,
      "step": 72560,
      "train_runtime": 87280.169,
      "train_tokens_per_second": 61697.007
    },
    {
      "epoch": 0.9643789742260849,
      "grad_norm": 15.4375,
      "learning_rate": 6.654916063059236e-08,
      "loss": 0.9059,
      "num_input_tokens_seen": 5385763072,
      "step": 72570,
      "train_runtime": 87292.9253,
      "train_tokens_per_second": 61697.59
    },
    {
      "epoch": 0.9645118637085468,
      "grad_norm": 32.0,
      "learning_rate": 6.605436667541654e-08,
      "loss": 0.7611,
      "num_input_tokens_seen": 5386233088,
      "step": 72580,
      "train_runtime": 87299.8805,
      "train_tokens_per_second": 61698.058
    },
    {
      "epoch": 0.9646447531910087,
      "grad_norm": 10.125,
      "learning_rate": 6.556141289508877e-08,
      "loss": 0.6203,
      "num_input_tokens_seen": 5386468608,
      "step": 72590,
      "train_runtime": 87303.4389,
      "train_tokens_per_second": 61698.241
    },
    {
      "epoch": 0.9647776426734707,
      "grad_norm": 9.5,
      "learning_rate": 6.507029938092601e-08,
      "loss": 0.5015,
      "num_input_tokens_seen": 5386586368,
      "step": 72600,
      "train_runtime": 87307.0015,
      "train_tokens_per_second": 61697.072
    },
    {
      "epoch": 0.9649105321559325,
      "grad_norm": 14.6875,
      "learning_rate": 6.458102622390216e-08,
      "loss": 0.7962,
      "num_input_tokens_seen": 5388632832,
      "step": 72610,
      "train_runtime": 87340.55,
      "train_tokens_per_second": 61696.804
    },
    {
      "epoch": 0.9650434216383944,
      "grad_norm": 25.0,
      "learning_rate": 6.409359351465028e-08,
      "loss": 0.9234,
      "num_input_tokens_seen": 5389477376,
      "step": 72620,
      "train_runtime": 87353.3183,
      "train_tokens_per_second": 61697.454
    },
    {
      "epoch": 0.9651763111208563,
      "grad_norm": 22.0,
      "learning_rate": 6.360800134346368e-08,
      "loss": 0.6972,
      "num_input_tokens_seen": 5389959168,
      "step": 72630,
      "train_runtime": 87360.0535,
      "train_tokens_per_second": 61698.213
    },
    {
      "epoch": 0.9653092006033183,
      "grad_norm": 34.25,
      "learning_rate": 6.312424980029374e-08,
      "loss": 0.7311,
      "num_input_tokens_seen": 5390193920,
      "step": 72640,
      "train_runtime": 87364.0064,
      "train_tokens_per_second": 61698.108
    },
    {
      "epoch": 0.9654420900857802,
      "grad_norm": 23.5,
      "learning_rate": 6.264233897475103e-08,
      "loss": 0.5354,
      "num_input_tokens_seen": 5390306560,
      "step": 72650,
      "train_runtime": 87366.6864,
      "train_tokens_per_second": 61697.505
    },
    {
      "epoch": 0.9655749795682421,
      "grad_norm": 37.0,
      "learning_rate": 6.216226895610522e-08,
      "loss": 0.7833,
      "num_input_tokens_seen": 5392265472,
      "step": 72660,
      "train_runtime": 87399.7184,
      "train_tokens_per_second": 61696.6
    },
    {
      "epoch": 0.9657078690507039,
      "grad_norm": 22.625,
      "learning_rate": 6.168403983328519e-08,
      "loss": 0.8993,
      "num_input_tokens_seen": 5393089792,
      "step": 72670,
      "train_runtime": 87412.3137,
      "train_tokens_per_second": 61697.14
    },
    {
      "epoch": 0.9658407585331659,
      "grad_norm": 16.25,
      "learning_rate": 6.120765169487786e-08,
      "loss": 0.6984,
      "num_input_tokens_seen": 5393559296,
      "step": 72680,
      "train_runtime": 87418.9593,
      "train_tokens_per_second": 61697.821
    },
    {
      "epoch": 0.9659736480156278,
      "grad_norm": 103.0,
      "learning_rate": 6.07331046291304e-08,
      "loss": 0.6317,
      "num_input_tokens_seen": 5393790976,
      "step": 72690,
      "train_runtime": 87422.5443,
      "train_tokens_per_second": 61697.941
    },
    {
      "epoch": 0.9661065374980897,
      "grad_norm": 21.375,
      "learning_rate": 6.026039872395028e-08,
      "loss": 0.5604,
      "num_input_tokens_seen": 5393908992,
      "step": 72700,
      "train_runtime": 87425.1853,
      "train_tokens_per_second": 61697.427
    },
    {
      "epoch": 0.9662394269805517,
      "grad_norm": 16.875,
      "learning_rate": 5.978953406689858e-08,
      "loss": 0.7684,
      "num_input_tokens_seen": 5395836160,
      "step": 72710,
      "train_runtime": 87457.209,
      "train_tokens_per_second": 61696.871
    },
    {
      "epoch": 0.9663723164630136,
      "grad_norm": 22.25,
      "learning_rate": 5.9320510745201066e-08,
      "loss": 0.8437,
      "num_input_tokens_seen": 5396639232,
      "step": 72720,
      "train_runtime": 87470.166,
      "train_tokens_per_second": 61696.913
    },
    {
      "epoch": 0.9665052059454754,
      "grad_norm": 29.125,
      "learning_rate": 5.885332884573935e-08,
      "loss": 0.7708,
      "num_input_tokens_seen": 5397087744,
      "step": 72730,
      "train_runtime": 87476.6314,
      "train_tokens_per_second": 61697.48
    },
    {
      "epoch": 0.9666380954279373,
      "grad_norm": 16.5,
      "learning_rate": 5.8387988455054226e-08,
      "loss": 0.595,
      "num_input_tokens_seen": 5397313024,
      "step": 72740,
      "train_runtime": 87480.4345,
      "train_tokens_per_second": 61697.373
    },
    {
      "epoch": 0.9667709849103993,
      "grad_norm": 11.4375,
      "learning_rate": 5.792448965934783e-08,
      "loss": 0.5194,
      "num_input_tokens_seen": 5397429760,
      "step": 72750,
      "train_runtime": 87483.0749,
      "train_tokens_per_second": 61696.846
    },
    {
      "epoch": 0.9669038743928612,
      "grad_norm": 50.0,
      "learning_rate": 5.746283254447704e-08,
      "loss": 0.8607,
      "num_input_tokens_seen": 5399578624,
      "step": 72760,
      "train_runtime": 87518.4787,
      "train_tokens_per_second": 61696.441
    },
    {
      "epoch": 0.9670367638753231,
      "grad_norm": 40.5,
      "learning_rate": 5.700301719596013e-08,
      "loss": 0.9325,
      "num_input_tokens_seen": 5400429568,
      "step": 72770,
      "train_runtime": 87531.355,
      "train_tokens_per_second": 61697.086
    },
    {
      "epoch": 0.967169653357785,
      "grad_norm": 138.0,
      "learning_rate": 5.654504369897562e-08,
      "loss": 0.7914,
      "num_input_tokens_seen": 5400893184,
      "step": 72780,
      "train_runtime": 87538.0194,
      "train_tokens_per_second": 61697.685
    },
    {
      "epoch": 0.9673025428402469,
      "grad_norm": 12.3125,
      "learning_rate": 5.608891213835566e-08,
      "loss": 0.631,
      "num_input_tokens_seen": 5401122304,
      "step": 72790,
      "train_runtime": 87541.5674,
      "train_tokens_per_second": 61697.802
    },
    {
      "epoch": 0.9674354323227088,
      "grad_norm": 14.0625,
      "learning_rate": 5.56346225985982e-08,
      "loss": 0.5927,
      "num_input_tokens_seen": 5401237504,
      "step": 72800,
      "train_runtime": 87544.2646,
      "train_tokens_per_second": 61697.217
    },
    {
      "epoch": 0.9675683218051707,
      "grad_norm": 25.875,
      "learning_rate": 5.518217516385482e-08,
      "loss": 0.8312,
      "num_input_tokens_seen": 5403252480,
      "step": 72810,
      "train_runtime": 87577.6714,
      "train_tokens_per_second": 61696.69
    },
    {
      "epoch": 0.9677012112876326,
      "grad_norm": 20.75,
      "learning_rate": 5.4731569917936264e-08,
      "loss": 0.9491,
      "num_input_tokens_seen": 5404076800,
      "step": 72820,
      "train_runtime": 87590.1473,
      "train_tokens_per_second": 61697.314
    },
    {
      "epoch": 0.9678341007700946,
      "grad_norm": 18.75,
      "learning_rate": 5.428280694431354e-08,
      "loss": 0.772,
      "num_input_tokens_seen": 5404545024,
      "step": 72830,
      "train_runtime": 87597.0119,
      "train_tokens_per_second": 61697.824
    },
    {
      "epoch": 0.9679669902525565,
      "grad_norm": 12.9375,
      "learning_rate": 5.383588632611791e-08,
      "loss": 0.6337,
      "num_input_tokens_seen": 5404775424,
      "step": 72840,
      "train_runtime": 87602.0628,
      "train_tokens_per_second": 61696.897
    },
    {
      "epoch": 0.9680998797350183,
      "grad_norm": 13.625,
      "learning_rate": 5.339080814613429e-08,
      "loss": 0.5092,
      "num_input_tokens_seen": 5404892160,
      "step": 72850,
      "train_runtime": 87604.9187,
      "train_tokens_per_second": 61696.218
    },
    {
      "epoch": 0.9682327692174802,
      "grad_norm": 14.5,
      "learning_rate": 5.294757248681226e-08,
      "loss": 0.7675,
      "num_input_tokens_seen": 5406945024,
      "step": 72860,
      "train_runtime": 87638.1512,
      "train_tokens_per_second": 61696.247
    },
    {
      "epoch": 0.9683656586999422,
      "grad_norm": 22.875,
      "learning_rate": 5.250617943025616e-08,
      "loss": 0.8905,
      "num_input_tokens_seen": 5407827200,
      "step": 72870,
      "train_runtime": 87650.9894,
      "train_tokens_per_second": 61697.275
    },
    {
      "epoch": 0.9684985481824041,
      "grad_norm": 16.5,
      "learning_rate": 5.206662905823057e-08,
      "loss": 0.7574,
      "num_input_tokens_seen": 5408303616,
      "step": 72880,
      "train_runtime": 87657.9547,
      "train_tokens_per_second": 61697.808
    },
    {
      "epoch": 0.968631437664866,
      "grad_norm": 15.6875,
      "learning_rate": 5.162892145215703e-08,
      "loss": 0.6827,
      "num_input_tokens_seen": 5408536576,
      "step": 72890,
      "train_runtime": 87661.7084,
      "train_tokens_per_second": 61697.823
    },
    {
      "epoch": 0.968764327147328,
      "grad_norm": 24.375,
      "learning_rate": 5.119305669311847e-08,
      "loss": 0.5565,
      "num_input_tokens_seen": 5408654848,
      "step": 72900,
      "train_runtime": 87665.3354,
      "train_tokens_per_second": 61696.62
    },
    {
      "epoch": 0.9688972166297898,
      "grad_norm": 20.5,
      "learning_rate": 5.075903486185363e-08,
      "loss": 0.8083,
      "num_input_tokens_seen": 5410673920,
      "step": 72910,
      "train_runtime": 87698.6251,
      "train_tokens_per_second": 61696.223
    },
    {
      "epoch": 0.9690301061122517,
      "grad_norm": 18.125,
      "learning_rate": 5.032685603876264e-08,
      "loss": 0.9421,
      "num_input_tokens_seen": 5411526400,
      "step": 72920,
      "train_runtime": 87711.7858,
      "train_tokens_per_second": 61696.685
    },
    {
      "epoch": 0.9691629955947136,
      "grad_norm": 15.875,
      "learning_rate": 4.9896520303901465e-08,
      "loss": 0.7516,
      "num_input_tokens_seen": 5412007168,
      "step": 72930,
      "train_runtime": 87718.8211,
      "train_tokens_per_second": 61697.217
    },
    {
      "epoch": 0.9692958850771756,
      "grad_norm": 17.75,
      "learning_rate": 4.9468027736986336e-08,
      "loss": 0.6347,
      "num_input_tokens_seen": 5412243200,
      "step": 72940,
      "train_runtime": 87722.6125,
      "train_tokens_per_second": 61697.241
    },
    {
      "epoch": 0.9694287745596375,
      "grad_norm": 15.75,
      "learning_rate": 4.904137841739265e-08,
      "loss": 0.4819,
      "num_input_tokens_seen": 5412360704,
      "step": 72950,
      "train_runtime": 87725.9816,
      "train_tokens_per_second": 61696.211
    },
    {
      "epoch": 0.9695616640420994,
      "grad_norm": 40.75,
      "learning_rate": 4.861657242415163e-08,
      "loss": 0.7938,
      "num_input_tokens_seen": 5414450688,
      "step": 72960,
      "train_runtime": 87759.647,
      "train_tokens_per_second": 61696.359
    },
    {
      "epoch": 0.9696945535245614,
      "grad_norm": 13.5,
      "learning_rate": 4.819360983595589e-08,
      "loss": 0.9353,
      "num_input_tokens_seen": 5415352832,
      "step": 72970,
      "train_runtime": 87772.6083,
      "train_tokens_per_second": 61697.527
    },
    {
      "epoch": 0.9698274430070232,
      "grad_norm": 25.75,
      "learning_rate": 4.7772490731156084e-08,
      "loss": 0.7303,
      "num_input_tokens_seen": 5415829504,
      "step": 72980,
      "train_runtime": 87779.4563,
      "train_tokens_per_second": 61698.144
    },
    {
      "epoch": 0.9699603324894851,
      "grad_norm": 42.25,
      "learning_rate": 4.73532151877587e-08,
      "loss": 0.675,
      "num_input_tokens_seen": 5416052480,
      "step": 72990,
      "train_runtime": 87784.1104,
      "train_tokens_per_second": 61697.413
    },
    {
      "epoch": 0.970093221971947,
      "grad_norm": 10.25,
      "learning_rate": 4.6935783283433845e-08,
      "loss": 0.5634,
      "num_input_tokens_seen": 5416168448,
      "step": 73000,
      "train_runtime": 87786.6524,
      "train_tokens_per_second": 61696.947
    },
    {
      "epoch": 0.970226111454409,
      "grad_norm": 37.25,
      "learning_rate": 4.6520195095502986e-08,
      "loss": 0.835,
      "num_input_tokens_seen": 5418182400,
      "step": 73010,
      "train_runtime": 87819.1587,
      "train_tokens_per_second": 61697.043
    },
    {
      "epoch": 0.9703590009368709,
      "grad_norm": 13.6875,
      "learning_rate": 4.610645070095454e-08,
      "loss": 0.9121,
      "num_input_tokens_seen": 5419027456,
      "step": 73020,
      "train_runtime": 87831.7422,
      "train_tokens_per_second": 61697.825
    },
    {
      "epoch": 0.9704918904193328,
      "grad_norm": 16.875,
      "learning_rate": 4.569455017642721e-08,
      "loss": 0.7576,
      "num_input_tokens_seen": 5419521792,
      "step": 73030,
      "train_runtime": 87838.7573,
      "train_tokens_per_second": 61698.525
    },
    {
      "epoch": 0.9706247799017946,
      "grad_norm": 21.125,
      "learning_rate": 4.5284493598224396e-08,
      "loss": 0.621,
      "num_input_tokens_seen": 5419763200,
      "step": 73040,
      "train_runtime": 87842.4385,
      "train_tokens_per_second": 61698.688
    },
    {
      "epoch": 0.9707576693842566,
      "grad_norm": 18.75,
      "learning_rate": 4.487628104230535e-08,
      "loss": 0.534,
      "num_input_tokens_seen": 5419880448,
      "step": 73050,
      "train_runtime": 87845.4063,
      "train_tokens_per_second": 61697.938
    },
    {
      "epoch": 0.9708905588667185,
      "grad_norm": 20.875,
      "learning_rate": 4.4469912584285126e-08,
      "loss": 0.8053,
      "num_input_tokens_seen": 5421893632,
      "step": 73060,
      "train_runtime": 87876.8328,
      "train_tokens_per_second": 61698.783
    },
    {
      "epoch": 0.9710234483491804,
      "grad_norm": 25.375,
      "learning_rate": 4.4065388299443513e-08,
      "loss": 0.8702,
      "num_input_tokens_seen": 5422713344,
      "step": 73070,
      "train_runtime": 87889.7822,
      "train_tokens_per_second": 61699.019
    },
    {
      "epoch": 0.9711563378316423,
      "grad_norm": 13.5,
      "learning_rate": 4.366270826271279e-08,
      "loss": 0.7364,
      "num_input_tokens_seen": 5423160832,
      "step": 73080,
      "train_runtime": 87896.1259,
      "train_tokens_per_second": 61699.657
    },
    {
      "epoch": 0.9712892273141043,
      "grad_norm": 18.875,
      "learning_rate": 4.326187254868552e-08,
      "loss": 0.6382,
      "num_input_tokens_seen": 5423386112,
      "step": 73090,
      "train_runtime": 87899.6101,
      "train_tokens_per_second": 61699.774
    },
    {
      "epoch": 0.9714221167965661,
      "grad_norm": 29.875,
      "learning_rate": 4.2862881231614525e-08,
      "loss": 0.5255,
      "num_input_tokens_seen": 5423502592,
      "step": 73100,
      "train_runtime": 87902.3157,
      "train_tokens_per_second": 61699.2
    },
    {
      "epoch": 0.971555006279028,
      "grad_norm": 34.5,
      "learning_rate": 4.246573438540846e-08,
      "loss": 0.7777,
      "num_input_tokens_seen": 5425515520,
      "step": 73110,
      "train_runtime": 87935.4547,
      "train_tokens_per_second": 61698.84
    },
    {
      "epoch": 0.97168789576149,
      "grad_norm": 111.5,
      "learning_rate": 4.2070432083636256e-08,
      "loss": 0.8436,
      "num_input_tokens_seen": 5426355712,
      "step": 73120,
      "train_runtime": 87948.146,
      "train_tokens_per_second": 61699.489
    },
    {
      "epoch": 0.9718207852439519,
      "grad_norm": 18.0,
      "learning_rate": 4.167697439952267e-08,
      "loss": 0.6846,
      "num_input_tokens_seen": 5426817024,
      "step": 73130,
      "train_runtime": 87955.9066,
      "train_tokens_per_second": 61699.29
    },
    {
      "epoch": 0.9719536747264138,
      "grad_norm": 52.75,
      "learning_rate": 4.128536140595274e-08,
      "loss": 0.6381,
      "num_input_tokens_seen": 5427047680,
      "step": 73140,
      "train_runtime": 87959.6358,
      "train_tokens_per_second": 61699.297
    },
    {
      "epoch": 0.9720865642088757,
      "grad_norm": 8.375,
      "learning_rate": 4.089559317546843e-08,
      "loss": 0.5333,
      "num_input_tokens_seen": 5427159552,
      "step": 73150,
      "train_runtime": 87962.1385,
      "train_tokens_per_second": 61698.813
    },
    {
      "epoch": 0.9722194536913376,
      "grad_norm": 12.4375,
      "learning_rate": 4.05076697802731e-08,
      "loss": 0.849,
      "num_input_tokens_seen": 5429173504,
      "step": 73160,
      "train_runtime": 87994.5498,
      "train_tokens_per_second": 61698.975
    },
    {
      "epoch": 0.9723523431737995,
      "grad_norm": 18.75,
      "learning_rate": 4.0121591292223705e-08,
      "loss": 0.9209,
      "num_input_tokens_seen": 5429999872,
      "step": 73170,
      "train_runtime": 88007.0596,
      "train_tokens_per_second": 61699.594
    },
    {
      "epoch": 0.9724852326562614,
      "grad_norm": 18.0,
      "learning_rate": 3.9737357782839713e-08,
      "loss": 0.7251,
      "num_input_tokens_seen": 5430459392,
      "step": 73180,
      "train_runtime": 88014.6268,
      "train_tokens_per_second": 61699.51
    },
    {
      "epoch": 0.9726181221387233,
      "grad_norm": 14.5625,
      "learning_rate": 3.935496932329641e-08,
      "loss": 0.6389,
      "num_input_tokens_seen": 5430692352,
      "step": 73190,
      "train_runtime": 88018.3125,
      "train_tokens_per_second": 61699.574
    },
    {
      "epoch": 0.9727510116211853,
      "grad_norm": 10.9375,
      "learning_rate": 3.897442598442713e-08,
      "loss": 0.493,
      "num_input_tokens_seen": 5430807808,
      "step": 73200,
      "train_runtime": 88020.9747,
      "train_tokens_per_second": 61699.019
    },
    {
      "epoch": 0.9728839011036472,
      "grad_norm": 22.125,
      "learning_rate": 3.8595727836725495e-08,
      "loss": 0.8275,
      "num_input_tokens_seen": 5432902144,
      "step": 73210,
      "train_runtime": 88054.9082,
      "train_tokens_per_second": 61699.027
    },
    {
      "epoch": 0.973016790586109,
      "grad_norm": 29.375,
      "learning_rate": 3.8218874950339824e-08,
      "loss": 0.8859,
      "num_input_tokens_seen": 5433776128,
      "step": 73220,
      "train_runtime": 88068.6975,
      "train_tokens_per_second": 61699.29
    },
    {
      "epoch": 0.9731496800685709,
      "grad_norm": 12.25,
      "learning_rate": 3.7843867395079836e-08,
      "loss": 0.7238,
      "num_input_tokens_seen": 5434233856,
      "step": 73230,
      "train_runtime": 88075.8257,
      "train_tokens_per_second": 61699.494
    },
    {
      "epoch": 0.9732825695510329,
      "grad_norm": 16.75,
      "learning_rate": 3.7470705240413296e-08,
      "loss": 0.6007,
      "num_input_tokens_seen": 5434458112,
      "step": 73240,
      "train_runtime": 88079.3641,
      "train_tokens_per_second": 61699.561
    },
    {
      "epoch": 0.9734154590334948,
      "grad_norm": 30.0,
      "learning_rate": 3.709938855546491e-08,
      "loss": 0.5261,
      "num_input_tokens_seen": 5434568704,
      "step": 73250,
      "train_runtime": 88081.9816,
      "train_tokens_per_second": 61698.983
    },
    {
      "epoch": 0.9735483485159567,
      "grad_norm": 17.875,
      "learning_rate": 3.6729917409016324e-08,
      "loss": 0.841,
      "num_input_tokens_seen": 5436632064,
      "step": 73260,
      "train_runtime": 88114.575,
      "train_tokens_per_second": 61699.578
    },
    {
      "epoch": 0.9736812379984187,
      "grad_norm": 11.375,
      "learning_rate": 3.636229186951057e-08,
      "loss": 0.9156,
      "num_input_tokens_seen": 5437497344,
      "step": 73270,
      "train_runtime": 88127.2785,
      "train_tokens_per_second": 61700.502
    },
    {
      "epoch": 0.9738141274808805,
      "grad_norm": 13.25,
      "learning_rate": 3.599651200504539e-08,
      "loss": 0.7632,
      "num_input_tokens_seen": 5437969152,
      "step": 73280,
      "train_runtime": 88134.1129,
      "train_tokens_per_second": 61701.071
    },
    {
      "epoch": 0.9739470169633424,
      "grad_norm": 13.5,
      "learning_rate": 3.563257788337881e-08,
      "loss": 0.6156,
      "num_input_tokens_seen": 5438198016,
      "step": 73290,
      "train_runtime": 88137.6913,
      "train_tokens_per_second": 61701.163
    },
    {
      "epoch": 0.9740799064458043,
      "grad_norm": 14.1875,
      "learning_rate": 3.5270489571926914e-08,
      "loss": 0.4709,
      "num_input_tokens_seen": 5438314496,
      "step": 73300,
      "train_runtime": 88141.3627,
      "train_tokens_per_second": 61699.914
    },
    {
      "epoch": 0.9742127959282663,
      "grad_norm": 35.5,
      "learning_rate": 3.49102471377627e-08,
      "loss": 0.8254,
      "num_input_tokens_seen": 5440455168,
      "step": 73310,
      "train_runtime": 88175.0875,
      "train_tokens_per_second": 61700.593
    },
    {
      "epoch": 0.9743456854107282,
      "grad_norm": 29.375,
      "learning_rate": 3.4551850647618345e-08,
      "loss": 0.889,
      "num_input_tokens_seen": 5441301760,
      "step": 73320,
      "train_runtime": 88187.8493,
      "train_tokens_per_second": 61701.264
    },
    {
      "epoch": 0.9744785748931901,
      "grad_norm": 17.0,
      "learning_rate": 3.4195300167882975e-08,
      "loss": 0.721,
      "num_input_tokens_seen": 5441762048,
      "step": 73330,
      "train_runtime": 88194.676,
      "train_tokens_per_second": 61701.707
    },
    {
      "epoch": 0.9746114643756519,
      "grad_norm": 14.0,
      "learning_rate": 3.3840595764605966e-08,
      "loss": 0.6561,
      "num_input_tokens_seen": 5441995264,
      "step": 73340,
      "train_runtime": 88198.397,
      "train_tokens_per_second": 61701.748
    },
    {
      "epoch": 0.9747443538581139,
      "grad_norm": 7.875,
      "learning_rate": 3.348773750349143e-08,
      "loss": 0.5198,
      "num_input_tokens_seen": 5442107392,
      "step": 73350,
      "train_runtime": 88202.1957,
      "train_tokens_per_second": 61700.362
    },
    {
      "epoch": 0.9748772433405758,
      "grad_norm": 13.5,
      "learning_rate": 3.313672544990265e-08,
      "loss": 0.7996,
      "num_input_tokens_seen": 5444211456,
      "step": 73360,
      "train_runtime": 88235.8912,
      "train_tokens_per_second": 61700.646
    },
    {
      "epoch": 0.9750101328230377,
      "grad_norm": 18.875,
      "learning_rate": 3.278755966886427e-08,
      "loss": 0.8999,
      "num_input_tokens_seen": 5445071104,
      "step": 73370,
      "train_runtime": 88248.8478,
      "train_tokens_per_second": 61701.328
    },
    {
      "epoch": 0.9751430223054997,
      "grad_norm": 19.25,
      "learning_rate": 3.244024022505343e-08,
      "loss": 0.6876,
      "num_input_tokens_seen": 5445578240,
      "step": 73380,
      "train_runtime": 88255.8199,
      "train_tokens_per_second": 61702.2
    },
    {
      "epoch": 0.9752759117879616,
      "grad_norm": 26.125,
      "learning_rate": 3.209476718280979e-08,
      "loss": 0.7072,
      "num_input_tokens_seen": 5445811456,
      "step": 73390,
      "train_runtime": 88259.4877,
      "train_tokens_per_second": 61702.278
    },
    {
      "epoch": 0.9754088012704234,
      "grad_norm": 8.3125,
      "learning_rate": 3.175114060612772e-08,
      "loss": 0.5823,
      "num_input_tokens_seen": 5445927680,
      "step": 73400,
      "train_runtime": 88262.1833,
      "train_tokens_per_second": 61701.71
    },
    {
      "epoch": 0.9755416907528853,
      "grad_norm": 13.0625,
      "learning_rate": 3.140936055866184e-08,
      "loss": 0.7966,
      "num_input_tokens_seen": 5448009216,
      "step": 73410,
      "train_runtime": 88295.4601,
      "train_tokens_per_second": 61702.031
    },
    {
      "epoch": 0.9756745802353473,
      "grad_norm": 33.5,
      "learning_rate": 3.1069427103722624e-08,
      "loss": 0.9224,
      "num_input_tokens_seen": 5448918016,
      "step": 73420,
      "train_runtime": 88309.447,
      "train_tokens_per_second": 61702.549
    },
    {
      "epoch": 0.9758074697178092,
      "grad_norm": 15.5625,
      "learning_rate": 3.073134030428193e-08,
      "loss": 0.7727,
      "num_input_tokens_seen": 5449416704,
      "step": 73430,
      "train_runtime": 88316.6831,
      "train_tokens_per_second": 61703.141
    },
    {
      "epoch": 0.9759403592002711,
      "grad_norm": 46.0,
      "learning_rate": 3.039510022296632e-08,
      "loss": 0.6898,
      "num_input_tokens_seen": 5449649408,
      "step": 73440,
      "train_runtime": 88320.2854,
      "train_tokens_per_second": 61703.259
    },
    {
      "epoch": 0.976073248682733,
      "grad_norm": 29.625,
      "learning_rate": 3.0060706922060425e-08,
      "loss": 0.5755,
      "num_input_tokens_seen": 5449762816,
      "step": 73450,
      "train_runtime": 88322.9413,
      "train_tokens_per_second": 61702.687
    },
    {
      "epoch": 0.976206138165195,
      "grad_norm": 17.875,
      "learning_rate": 2.9728160463508017e-08,
      "loss": 0.8666,
      "num_input_tokens_seen": 5451747584,
      "step": 73460,
      "train_runtime": 88355.4406,
      "train_tokens_per_second": 61702.455
    },
    {
      "epoch": 0.9763390276476568,
      "grad_norm": 23.875,
      "learning_rate": 2.9397460908910934e-08,
      "loss": 0.8846,
      "num_input_tokens_seen": 5452624640,
      "step": 73470,
      "train_runtime": 88368.9797,
      "train_tokens_per_second": 61702.926
    },
    {
      "epoch": 0.9764719171301187,
      "grad_norm": 14.875,
      "learning_rate": 2.906860831952907e-08,
      "loss": 0.7156,
      "num_input_tokens_seen": 5453103872,
      "step": 73480,
      "train_runtime": 88375.7373,
      "train_tokens_per_second": 61703.631
    },
    {
      "epoch": 0.9766048066125806,
      "grad_norm": 21.75,
      "learning_rate": 2.8741602756277022e-08,
      "loss": 0.7178,
      "num_input_tokens_seen": 5453327104,
      "step": 73490,
      "train_runtime": 88379.3444,
      "train_tokens_per_second": 61703.638
    },
    {
      "epoch": 0.9767376960950426,
      "grad_norm": 22.375,
      "learning_rate": 2.8416444279731893e-08,
      "loss": 0.4481,
      "num_input_tokens_seen": 5453441536,
      "step": 73500,
      "train_runtime": 88381.8591,
      "train_tokens_per_second": 61703.177
    },
    {
      "epoch": 0.9768705855775045,
      "grad_norm": 14.625,
      "learning_rate": 2.8093132950124392e-08,
      "loss": 0.8273,
      "num_input_tokens_seen": 5455476992,
      "step": 73510,
      "train_runtime": 88415.6399,
      "train_tokens_per_second": 61702.624
    },
    {
      "epoch": 0.9770034750599664,
      "grad_norm": 16.75,
      "learning_rate": 2.777166882734661e-08,
      "loss": 0.8511,
      "num_input_tokens_seen": 5456319488,
      "step": 73520,
      "train_runtime": 88428.1045,
      "train_tokens_per_second": 61703.454
    },
    {
      "epoch": 0.9771363645424282,
      "grad_norm": 17.5,
      "learning_rate": 2.745205197094647e-08,
      "loss": 0.7115,
      "num_input_tokens_seen": 5456777472,
      "step": 73530,
      "train_runtime": 88434.8284,
      "train_tokens_per_second": 61703.941
    },
    {
      "epoch": 0.9772692540248902,
      "grad_norm": 31.375,
      "learning_rate": 2.7134282440129943e-08,
      "loss": 0.6632,
      "num_input_tokens_seen": 5457006848,
      "step": 73540,
      "train_runtime": 88438.3343,
      "train_tokens_per_second": 61704.089
    },
    {
      "epoch": 0.9774021435073521,
      "grad_norm": 19.125,
      "learning_rate": 2.6818360293762168e-08,
      "loss": 0.5717,
      "num_input_tokens_seen": 5457122560,
      "step": 73550,
      "train_runtime": 88440.8425,
      "train_tokens_per_second": 61703.647
    },
    {
      "epoch": 0.977535032989814,
      "grad_norm": 24.75,
      "learning_rate": 2.6504285590363e-08,
      "loss": 0.814,
      "num_input_tokens_seen": 5459210752,
      "step": 73560,
      "train_runtime": 88474.9425,
      "train_tokens_per_second": 61703.468
    },
    {
      "epoch": 0.977667922472276,
      "grad_norm": 23.25,
      "learning_rate": 2.6192058388113672e-08,
      "loss": 0.9429,
      "num_input_tokens_seen": 5460091904,
      "step": 73570,
      "train_runtime": 88487.7675,
      "train_tokens_per_second": 61704.483
    },
    {
      "epoch": 0.9778008119547379,
      "grad_norm": 18.125,
      "learning_rate": 2.5881678744849036e-08,
      "loss": 0.7099,
      "num_input_tokens_seen": 5460547072,
      "step": 73580,
      "train_runtime": 88494.4247,
      "train_tokens_per_second": 61704.984
    },
    {
      "epoch": 0.9779337014371997,
      "grad_norm": 17.75,
      "learning_rate": 2.5573146718067543e-08,
      "loss": 0.6091,
      "num_input_tokens_seen": 5460766464,
      "step": 73590,
      "train_runtime": 88498.7284,
      "train_tokens_per_second": 61704.462
    },
    {
      "epoch": 0.9780665909196616,
      "grad_norm": 15.1875,
      "learning_rate": 2.526646236492014e-08,
      "loss": 0.4397,
      "num_input_tokens_seen": 5460879360,
      "step": 73600,
      "train_runtime": 88501.1407,
      "train_tokens_per_second": 61704.056
    },
    {
      "epoch": 0.9781994804021236,
      "grad_norm": 22.625,
      "learning_rate": 2.4961625742216943e-08,
      "loss": 0.8784,
      "num_input_tokens_seen": 5462944512,
      "step": 73610,
      "train_runtime": 88535.3036,
      "train_tokens_per_second": 61703.572
    },
    {
      "epoch": 0.9783323698845855,
      "grad_norm": 65.5,
      "learning_rate": 2.4658636906427224e-08,
      "loss": 0.9143,
      "num_input_tokens_seen": 5463786240,
      "step": 73620,
      "train_runtime": 88548.1244,
      "train_tokens_per_second": 61704.144
    },
    {
      "epoch": 0.9784652593670474,
      "grad_norm": 22.25,
      "learning_rate": 2.4357495913676087e-08,
      "loss": 0.8178,
      "num_input_tokens_seen": 5464292352,
      "step": 73630,
      "train_runtime": 88555.0259,
      "train_tokens_per_second": 61705.051
    },
    {
      "epoch": 0.9785981488495094,
      "grad_norm": 17.5,
      "learning_rate": 2.40582028197478e-08,
      "loss": 0.6464,
      "num_input_tokens_seen": 5464526848,
      "step": 73640,
      "train_runtime": 88559.533,
      "train_tokens_per_second": 61704.558
    },
    {
      "epoch": 0.9787310383319712,
      "grad_norm": 9.875,
      "learning_rate": 2.376075768008357e-08,
      "loss": 0.5748,
      "num_input_tokens_seen": 5464647936,
      "step": 73650,
      "train_runtime": 88561.9897,
      "train_tokens_per_second": 61704.214
    },
    {
      "epoch": 0.9788639278144331,
      "grad_norm": 13.5,
      "learning_rate": 2.346516054978265e-08,
      "loss": 0.8526,
      "num_input_tokens_seen": 5466953472,
      "step": 73660,
      "train_runtime": 88597.9585,
      "train_tokens_per_second": 61705.186
    },
    {
      "epoch": 0.978996817296895,
      "grad_norm": 21.75,
      "learning_rate": 2.3171411483601247e-08,
      "loss": 0.9961,
      "num_input_tokens_seen": 5467815424,
      "step": 73670,
      "train_runtime": 88611.1903,
      "train_tokens_per_second": 61705.699
    },
    {
      "epoch": 0.979129706779357,
      "grad_norm": 25.0,
      "learning_rate": 2.2879510535954717e-08,
      "loss": 0.7864,
      "num_input_tokens_seen": 5468340992,
      "step": 73680,
      "train_runtime": 88618.4358,
      "train_tokens_per_second": 61706.584
    },
    {
      "epoch": 0.9792625962618189,
      "grad_norm": 17.75,
      "learning_rate": 2.258945776091426e-08,
      "loss": 0.6439,
      "num_input_tokens_seen": 5468586496,
      "step": 73690,
      "train_runtime": 88622.0353,
      "train_tokens_per_second": 61706.848
    },
    {
      "epoch": 0.9793954857442808,
      "grad_norm": 49.25,
      "learning_rate": 2.2301253212210215e-08,
      "loss": 0.554,
      "num_input_tokens_seen": 5468701952,
      "step": 73700,
      "train_runtime": 88624.54,
      "train_tokens_per_second": 61706.407
    },
    {
      "epoch": 0.9795283752267426,
      "grad_norm": 26.0,
      "learning_rate": 2.201489694322878e-08,
      "loss": 0.8587,
      "num_input_tokens_seen": 5470872576,
      "step": 73710,
      "train_runtime": 88658.6723,
      "train_tokens_per_second": 61707.134
    },
    {
      "epoch": 0.9796612647092046,
      "grad_norm": 22.5,
      "learning_rate": 2.1730389007016406e-08,
      "loss": 0.9189,
      "num_input_tokens_seen": 5471712512,
      "step": 73720,
      "train_runtime": 88671.48,
      "train_tokens_per_second": 61707.694
    },
    {
      "epoch": 0.9797941541916665,
      "grad_norm": 50.25,
      "learning_rate": 2.144772945627427e-08,
      "loss": 0.7262,
      "num_input_tokens_seen": 5472192256,
      "step": 73730,
      "train_runtime": 88678.2559,
      "train_tokens_per_second": 61708.388
    },
    {
      "epoch": 0.9799270436741284,
      "grad_norm": 24.25,
      "learning_rate": 2.116691834336271e-08,
      "loss": 0.6436,
      "num_input_tokens_seen": 5472433152,
      "step": 73740,
      "train_runtime": 88681.8179,
      "train_tokens_per_second": 61708.626
    },
    {
      "epoch": 0.9800599331565903,
      "grad_norm": 16.625,
      "learning_rate": 2.088795572030122e-08,
      "loss": 0.5048,
      "num_input_tokens_seen": 5472546560,
      "step": 73750,
      "train_runtime": 88685.186,
      "train_tokens_per_second": 61707.561
    },
    {
      "epoch": 0.9801928226390523,
      "grad_norm": 14.0,
      "learning_rate": 2.0610841638762925e-08,
      "loss": 0.8499,
      "num_input_tokens_seen": 5474610176,
      "step": 73760,
      "train_runtime": 88718.1774,
      "train_tokens_per_second": 61707.875
    },
    {
      "epoch": 0.9803257121215141,
      "grad_norm": 37.75,
      "learning_rate": 2.0335576150082305e-08,
      "loss": 0.9413,
      "num_input_tokens_seen": 5475476224,
      "step": 73770,
      "train_runtime": 88730.863,
      "train_tokens_per_second": 61708.813
    },
    {
      "epoch": 0.980458601603976,
      "grad_norm": 26.625,
      "learning_rate": 2.0062159305248574e-08,
      "loss": 0.7082,
      "num_input_tokens_seen": 5475936768,
      "step": 73780,
      "train_runtime": 88737.3511,
      "train_tokens_per_second": 61709.491
    },
    {
      "epoch": 0.980591491086438,
      "grad_norm": 15.5,
      "learning_rate": 1.979059115491011e-08,
      "loss": 0.6566,
      "num_input_tokens_seen": 5476164096,
      "step": 73790,
      "train_runtime": 88740.7915,
      "train_tokens_per_second": 61709.66
    },
    {
      "epoch": 0.9807243805688999,
      "grad_norm": 14.0625,
      "learning_rate": 1.952087174937334e-08,
      "loss": 0.5797,
      "num_input_tokens_seen": 5476278272,
      "step": 73800,
      "train_runtime": 88743.1872,
      "train_tokens_per_second": 61709.281
    },
    {
      "epoch": 0.9808572700513618,
      "grad_norm": 20.625,
      "learning_rate": 1.9253001138600515e-08,
      "loss": 0.8626,
      "num_input_tokens_seen": 5478312192,
      "step": 73810,
      "train_runtime": 88775.6746,
      "train_tokens_per_second": 61709.609
    },
    {
      "epoch": 0.9809901595338237,
      "grad_norm": 17.125,
      "learning_rate": 1.8986979372213054e-08,
      "loss": 0.9004,
      "num_input_tokens_seen": 5479158784,
      "step": 73820,
      "train_runtime": 88789.1426,
      "train_tokens_per_second": 61709.784
    },
    {
      "epoch": 0.9811230490162856,
      "grad_norm": 14.8125,
      "learning_rate": 1.872280649948821e-08,
      "loss": 0.656,
      "num_input_tokens_seen": 5479600896,
      "step": 73830,
      "train_runtime": 88795.7128,
      "train_tokens_per_second": 61710.197
    },
    {
      "epoch": 0.9812559384987475,
      "grad_norm": 18.5,
      "learning_rate": 1.8460482569361283e-08,
      "loss": 0.6729,
      "num_input_tokens_seen": 5479832064,
      "step": 73840,
      "train_runtime": 88799.3646,
      "train_tokens_per_second": 61710.262
    },
    {
      "epoch": 0.9813888279812094,
      "grad_norm": 61.75,
      "learning_rate": 1.8200007630427842e-08,
      "loss": 0.5697,
      "num_input_tokens_seen": 5479948544,
      "step": 73850,
      "train_runtime": 88801.8833,
      "train_tokens_per_second": 61709.823
    },
    {
      "epoch": 0.9815217174636713,
      "grad_norm": 26.25,
      "learning_rate": 1.794138173093596e-08,
      "loss": 0.9108,
      "num_input_tokens_seen": 5481910016,
      "step": 73860,
      "train_runtime": 88833.7882,
      "train_tokens_per_second": 61709.74
    },
    {
      "epoch": 0.9816546069461333,
      "grad_norm": 15.875,
      "learning_rate": 1.768460491879398e-08,
      "loss": 0.9142,
      "num_input_tokens_seen": 5482750208,
      "step": 73870,
      "train_runtime": 88847.2907,
      "train_tokens_per_second": 61709.819
    },
    {
      "epoch": 0.9817874964285952,
      "grad_norm": 21.625,
      "learning_rate": 1.742967724156941e-08,
      "loss": 0.7883,
      "num_input_tokens_seen": 5483222016,
      "step": 73880,
      "train_runtime": 88854.1837,
      "train_tokens_per_second": 61710.341
    },
    {
      "epoch": 0.981920385911057,
      "grad_norm": 36.75,
      "learning_rate": 1.7176598746484475e-08,
      "loss": 0.6426,
      "num_input_tokens_seen": 5483451904,
      "step": 73890,
      "train_runtime": 88857.6995,
      "train_tokens_per_second": 61710.487
    },
    {
      "epoch": 0.9820532753935189,
      "grad_norm": 6.03125,
      "learning_rate": 1.692536948041834e-08,
      "loss": 0.5689,
      "num_input_tokens_seen": 5483569152,
      "step": 73900,
      "train_runtime": 88860.2082,
      "train_tokens_per_second": 61710.064
    },
    {
      "epoch": 0.9821861648759809,
      "grad_norm": 13.0,
      "learning_rate": 1.667598948991045e-08,
      "loss": 0.8642,
      "num_input_tokens_seen": 5485691648,
      "step": 73910,
      "train_runtime": 88893.8042,
      "train_tokens_per_second": 61710.619
    },
    {
      "epoch": 0.9823190543584428,
      "grad_norm": 28.375,
      "learning_rate": 1.6428458821156068e-08,
      "loss": 0.8547,
      "num_input_tokens_seen": 5486527232,
      "step": 73920,
      "train_runtime": 88906.5961,
      "train_tokens_per_second": 61711.138
    },
    {
      "epoch": 0.9824519438409047,
      "grad_norm": 73.5,
      "learning_rate": 1.618277752000852e-08,
      "loss": 0.7586,
      "num_input_tokens_seen": 5486994176,
      "step": 73930,
      "train_runtime": 88914.5514,
      "train_tokens_per_second": 61710.868
    },
    {
      "epoch": 0.9825848333233667,
      "grad_norm": 14.125,
      "learning_rate": 1.5938945631976953e-08,
      "loss": 0.6612,
      "num_input_tokens_seen": 5487233536,
      "step": 73940,
      "train_runtime": 88918.1578,
      "train_tokens_per_second": 61711.057
    },
    {
      "epoch": 0.9827177228058286,
      "grad_norm": 15.8125,
      "learning_rate": 1.5696963202228576e-08,
      "loss": 0.4958,
      "num_input_tokens_seen": 5487353088,
      "step": 73950,
      "train_runtime": 88920.6008,
      "train_tokens_per_second": 61710.706
    },
    {
      "epoch": 0.9828506122882904,
      "grad_norm": 14.125,
      "learning_rate": 1.5456830275589752e-08,
      "loss": 0.8065,
      "num_input_tokens_seen": 5489378048,
      "step": 73960,
      "train_runtime": 88954.8174,
      "train_tokens_per_second": 61709.733
    },
    {
      "epoch": 0.9829835017707523,
      "grad_norm": 15.6875,
      "learning_rate": 1.5218546896542675e-08,
      "loss": 0.9423,
      "num_input_tokens_seen": 5490246400,
      "step": 73970,
      "train_runtime": 88967.9482,
      "train_tokens_per_second": 61710.386
    },
    {
      "epoch": 0.9831163912532143,
      "grad_norm": 26.875,
      "learning_rate": 1.4982113109227593e-08,
      "loss": 0.7329,
      "num_input_tokens_seen": 5490734080,
      "step": 73980,
      "train_runtime": 88975.1104,
      "train_tokens_per_second": 61710.899
    },
    {
      "epoch": 0.9832492807356762,
      "grad_norm": 32.0,
      "learning_rate": 1.4747528957440583e-08,
      "loss": 0.6124,
      "num_input_tokens_seen": 5490973440,
      "step": 73990,
      "train_runtime": 88980.5883,
      "train_tokens_per_second": 61709.79
    },
    {
      "epoch": 0.9833821702181381,
      "grad_norm": 46.75,
      "learning_rate": 1.4514794484637996e-08,
      "loss": 0.5951,
      "num_input_tokens_seen": 5491088640,
      "step": 74000,
      "train_runtime": 88983.0522,
      "train_tokens_per_second": 61709.376
    },
    {
      "epoch": 0.9835150597006,
      "grad_norm": 29.0,
      "learning_rate": 1.428390973392979e-08,
      "loss": 0.7881,
      "num_input_tokens_seen": 5493119488,
      "step": 74010,
      "train_runtime": 89015.3198,
      "train_tokens_per_second": 61709.821
    },
    {
      "epoch": 0.9836479491830619,
      "grad_norm": 29.25,
      "learning_rate": 1.405487474808731e-08,
      "loss": 0.9156,
      "num_input_tokens_seen": 5493948160,
      "step": 74020,
      "train_runtime": 89027.9602,
      "train_tokens_per_second": 61710.368
    },
    {
      "epoch": 0.9837808386655238,
      "grad_norm": 17.375,
      "learning_rate": 1.3827689569535507e-08,
      "loss": 0.808,
      "num_input_tokens_seen": 5494424832,
      "step": 74030,
      "train_runtime": 89034.7459,
      "train_tokens_per_second": 61711.018
    },
    {
      "epoch": 0.9839137281479857,
      "grad_norm": 12.1875,
      "learning_rate": 1.3602354240358495e-08,
      "loss": 0.6448,
      "num_input_tokens_seen": 5494657280,
      "step": 74040,
      "train_runtime": 89039.7575,
      "train_tokens_per_second": 61710.156
    },
    {
      "epoch": 0.9840466176304477,
      "grad_norm": 10.9375,
      "learning_rate": 1.3378868802299549e-08,
      "loss": 0.594,
      "num_input_tokens_seen": 5494774016,
      "step": 74050,
      "train_runtime": 89042.2496,
      "train_tokens_per_second": 61709.739
    },
    {
      "epoch": 0.9841795071129096,
      "grad_norm": 33.5,
      "learning_rate": 1.3157233296755556e-08,
      "loss": 0.7804,
      "num_input_tokens_seen": 5496783104,
      "step": 74060,
      "train_runtime": 89075.1414,
      "train_tokens_per_second": 61709.507
    },
    {
      "epoch": 0.9843123965953715,
      "grad_norm": 17.375,
      "learning_rate": 1.2937447764782563e-08,
      "loss": 0.8669,
      "num_input_tokens_seen": 5497645824,
      "step": 74070,
      "train_runtime": 89087.7387,
      "train_tokens_per_second": 61710.465
    },
    {
      "epoch": 0.9844452860778333,
      "grad_norm": 15.0625,
      "learning_rate": 1.2719512247095777e-08,
      "loss": 0.7516,
      "num_input_tokens_seen": 5498135296,
      "step": 74080,
      "train_runtime": 89095.4456,
      "train_tokens_per_second": 61710.621
    },
    {
      "epoch": 0.9845781755602953,
      "grad_norm": 29.5,
      "learning_rate": 1.2503426784062911e-08,
      "loss": 0.6287,
      "num_input_tokens_seen": 5498372864,
      "step": 74090,
      "train_runtime": 89099.014,
      "train_tokens_per_second": 61710.816
    },
    {
      "epoch": 0.9847110650427572,
      "grad_norm": 14.875,
      "learning_rate": 1.2289191415714163e-08,
      "loss": 0.5491,
      "num_input_tokens_seen": 5498488320,
      "step": 74100,
      "train_runtime": 89101.509,
      "train_tokens_per_second": 61710.384
    },
    {
      "epoch": 0.9848439545252191,
      "grad_norm": 47.75,
      "learning_rate": 1.2076806181734458e-08,
      "loss": 0.784,
      "num_input_tokens_seen": 5500554240,
      "step": 74110,
      "train_runtime": 89134.241,
      "train_tokens_per_second": 61710.9
    },
    {
      "epoch": 0.984976844007681,
      "grad_norm": 26.875,
      "learning_rate": 1.1866271121464545e-08,
      "loss": 0.8735,
      "num_input_tokens_seen": 5501382656,
      "step": 74120,
      "train_runtime": 89146.7201,
      "train_tokens_per_second": 61711.554
    },
    {
      "epoch": 0.985109733490143,
      "grad_norm": 21.625,
      "learning_rate": 1.1657586273906562e-08,
      "loss": 0.7358,
      "num_input_tokens_seen": 5501829888,
      "step": 74130,
      "train_runtime": 89153.2101,
      "train_tokens_per_second": 61712.078
    },
    {
      "epoch": 0.9852426229726048,
      "grad_norm": 15.8125,
      "learning_rate": 1.1450751677715144e-08,
      "loss": 0.6607,
      "num_input_tokens_seen": 5502051584,
      "step": 74140,
      "train_runtime": 89156.646,
      "train_tokens_per_second": 61712.187
    },
    {
      "epoch": 0.9853755124550667,
      "grad_norm": 7.875,
      "learning_rate": 1.1245767371206307e-08,
      "loss": 0.5362,
      "num_input_tokens_seen": 5502167808,
      "step": 74150,
      "train_runtime": 89159.0697,
      "train_tokens_per_second": 61711.813
    },
    {
      "epoch": 0.9855084019375286,
      "grad_norm": 27.875,
      "learning_rate": 1.1042633392350789e-08,
      "loss": 0.7861,
      "num_input_tokens_seen": 5504161536,
      "step": 74160,
      "train_runtime": 89191.5749,
      "train_tokens_per_second": 61711.676
    },
    {
      "epoch": 0.9856412914199906,
      "grad_norm": 29.375,
      "learning_rate": 1.0841349778777378e-08,
      "loss": 0.8403,
      "num_input_tokens_seen": 5504999424,
      "step": 74170,
      "train_runtime": 89204.0494,
      "train_tokens_per_second": 61712.439
    },
    {
      "epoch": 0.9857741809024525,
      "grad_norm": 12.5625,
      "learning_rate": 1.0641916567772914e-08,
      "loss": 0.7055,
      "num_input_tokens_seen": 5505457152,
      "step": 74180,
      "train_runtime": 89210.6381,
      "train_tokens_per_second": 61713.012
    },
    {
      "epoch": 0.9859070703849144,
      "grad_norm": 13.375,
      "learning_rate": 1.044433379628007e-08,
      "loss": 0.6411,
      "num_input_tokens_seen": 5505686784,
      "step": 74190,
      "train_runtime": 89214.1605,
      "train_tokens_per_second": 61713.149
    },
    {
      "epoch": 0.9860399598673762,
      "grad_norm": 8.5625,
      "learning_rate": 1.0248601500898459e-08,
      "loss": 0.5199,
      "num_input_tokens_seen": 5505804288,
      "step": 74200,
      "train_runtime": 89216.6204,
      "train_tokens_per_second": 61712.765
    },
    {
      "epoch": 0.9861728493498382,
      "grad_norm": 28.625,
      "learning_rate": 1.0054719717886852e-08,
      "loss": 0.8679,
      "num_input_tokens_seen": 5507793664,
      "step": 74210,
      "train_runtime": 89250.0589,
      "train_tokens_per_second": 61711.933
    },
    {
      "epoch": 0.9863057388323001,
      "grad_norm": 22.875,
      "learning_rate": 9.862688483158744e-09,
      "loss": 0.9042,
      "num_input_tokens_seen": 5508637952,
      "step": 74220,
      "train_runtime": 89262.6439,
      "train_tokens_per_second": 61712.691
    },
    {
      "epoch": 0.986438628314762,
      "grad_norm": 18.625,
      "learning_rate": 9.672507832287904e-09,
      "loss": 0.6961,
      "num_input_tokens_seen": 5509101824,
      "step": 74230,
      "train_runtime": 89269.2134,
      "train_tokens_per_second": 61713.346
    },
    {
      "epoch": 0.986571517797224,
      "grad_norm": 24.375,
      "learning_rate": 9.484177800502814e-09,
      "loss": 0.7011,
      "num_input_tokens_seen": 5509328128,
      "step": 74240,
      "train_runtime": 89272.6514,
      "train_tokens_per_second": 61713.504
    },
    {
      "epoch": 0.9867044072796859,
      "grad_norm": 21.375,
      "learning_rate": 9.297698422691126e-09,
      "loss": 0.5774,
      "num_input_tokens_seen": 5509445632,
      "step": 74250,
      "train_runtime": 89275.235,
      "train_tokens_per_second": 61713.034
    },
    {
      "epoch": 0.9868372967621477,
      "grad_norm": 13.3125,
      "learning_rate": 9.113069733394098e-09,
      "loss": 0.7296,
      "num_input_tokens_seen": 5511477504,
      "step": 74260,
      "train_runtime": 89309.8198,
      "train_tokens_per_second": 61711.887
    },
    {
      "epoch": 0.9869701862446096,
      "grad_norm": 23.5,
      "learning_rate": 8.930291766813259e-09,
      "loss": 0.8971,
      "num_input_tokens_seen": 5512304896,
      "step": 74270,
      "train_runtime": 89322.5398,
      "train_tokens_per_second": 61712.362
    },
    {
      "epoch": 0.9871030757270716,
      "grad_norm": 20.125,
      "learning_rate": 8.749364556808193e-09,
      "loss": 0.7429,
      "num_input_tokens_seen": 5512783104,
      "step": 74280,
      "train_runtime": 89329.4901,
      "train_tokens_per_second": 61712.914
    },
    {
      "epoch": 0.9872359652095335,
      "grad_norm": 17.25,
      "learning_rate": 8.5702881368932e-09,
      "loss": 0.5599,
      "num_input_tokens_seen": 5513019392,
      "step": 74290,
      "train_runtime": 89333.0633,
      "train_tokens_per_second": 61713.09
    },
    {
      "epoch": 0.9873688546919954,
      "grad_norm": 19.375,
      "learning_rate": 8.393062540239527e-09,
      "loss": 0.5073,
      "num_input_tokens_seen": 5513130752,
      "step": 74300,
      "train_runtime": 89335.4838,
      "train_tokens_per_second": 61712.665
    },
    {
      "epoch": 0.9875017441744574,
      "grad_norm": 30.875,
      "learning_rate": 8.217687799678686e-09,
      "loss": 0.868,
      "num_input_tokens_seen": 5515192576,
      "step": 74310,
      "train_runtime": 89368.2264,
      "train_tokens_per_second": 61713.126
    },
    {
      "epoch": 0.9876346336569192,
      "grad_norm": 17.625,
      "learning_rate": 8.044163947694695e-09,
      "loss": 0.8903,
      "num_input_tokens_seen": 5516015872,
      "step": 74320,
      "train_runtime": 89380.4377,
      "train_tokens_per_second": 61713.905
    },
    {
      "epoch": 0.9877675231393811,
      "grad_norm": 45.0,
      "learning_rate": 7.872491016434059e-09,
      "loss": 0.7247,
      "num_input_tokens_seen": 5516463616,
      "step": 74330,
      "train_runtime": 89386.8922,
      "train_tokens_per_second": 61714.458
    },
    {
      "epoch": 0.987900412621843,
      "grad_norm": 45.5,
      "learning_rate": 7.702669037694677e-09,
      "loss": 0.6015,
      "num_input_tokens_seen": 5516679936,
      "step": 74340,
      "train_runtime": 89391.3086,
      "train_tokens_per_second": 61713.829
    },
    {
      "epoch": 0.988033302104305,
      "grad_norm": 11.4375,
      "learning_rate": 7.53469804293805e-09,
      "loss": 0.5132,
      "num_input_tokens_seen": 5516792320,
      "step": 74350,
      "train_runtime": 89393.7691,
      "train_tokens_per_second": 61713.388
    },
    {
      "epoch": 0.9881661915867669,
      "grad_norm": 117.0,
      "learning_rate": 7.368578063275955e-09,
      "loss": 0.8402,
      "num_input_tokens_seen": 5518851072,
      "step": 74360,
      "train_runtime": 89426.8638,
      "train_tokens_per_second": 61713.571
    },
    {
      "epoch": 0.9882990810692288,
      "grad_norm": 20.625,
      "learning_rate": 7.204309129482667e-09,
      "loss": 0.8743,
      "num_input_tokens_seen": 5519695360,
      "step": 74370,
      "train_runtime": 89439.5478,
      "train_tokens_per_second": 61714.258
    },
    {
      "epoch": 0.9884319705516907,
      "grad_norm": 39.0,
      "learning_rate": 7.041891271987178e-09,
      "loss": 0.8527,
      "num_input_tokens_seen": 5520159232,
      "step": 74380,
      "train_runtime": 89446.2044,
      "train_tokens_per_second": 61714.852
    },
    {
      "epoch": 0.9885648600341526,
      "grad_norm": 15.9375,
      "learning_rate": 6.881324520875421e-09,
      "loss": 0.6079,
      "num_input_tokens_seen": 5520384000,
      "step": 74390,
      "train_runtime": 89449.7191,
      "train_tokens_per_second": 61714.939
    },
    {
      "epoch": 0.9886977495166145,
      "grad_norm": 13.0625,
      "learning_rate": 6.722608905891382e-09,
      "loss": 0.4829,
      "num_input_tokens_seen": 5520498944,
      "step": 74400,
      "train_runtime": 89453.2851,
      "train_tokens_per_second": 61713.764
    },
    {
      "epoch": 0.9888306389990764,
      "grad_norm": 21.125,
      "learning_rate": 6.565744456435985e-09,
      "loss": 0.7758,
      "num_input_tokens_seen": 5522517760,
      "step": 74410,
      "train_runtime": 89486.1978,
      "train_tokens_per_second": 61713.626
    },
    {
      "epoch": 0.9889635284815383,
      "grad_norm": 25.5,
      "learning_rate": 6.410731201565989e-09,
      "loss": 0.8987,
      "num_input_tokens_seen": 5523365888,
      "step": 74420,
      "train_runtime": 89498.9964,
      "train_tokens_per_second": 61714.277
    },
    {
      "epoch": 0.9890964179640003,
      "grad_norm": 22.125,
      "learning_rate": 6.257569169997313e-09,
      "loss": 0.7653,
      "num_input_tokens_seen": 5523843584,
      "step": 74430,
      "train_runtime": 89506.0227,
      "train_tokens_per_second": 61714.77
    },
    {
      "epoch": 0.9892293074464622,
      "grad_norm": 23.0,
      "learning_rate": 6.106258390101704e-09,
      "loss": 0.5998,
      "num_input_tokens_seen": 5524084736,
      "step": 74440,
      "train_runtime": 89509.7005,
      "train_tokens_per_second": 61714.928
    },
    {
      "epoch": 0.989362196928924,
      "grad_norm": 19.125,
      "learning_rate": 5.956798889906745e-09,
      "loss": 0.5825,
      "num_input_tokens_seen": 5524200192,
      "step": 74450,
      "train_runtime": 89513.6376,
      "train_tokens_per_second": 61713.504
    },
    {
      "epoch": 0.989495086411386,
      "grad_norm": 48.5,
      "learning_rate": 5.809190697100287e-09,
      "loss": 0.8831,
      "num_input_tokens_seen": 5526275328,
      "step": 74460,
      "train_runtime": 89546.8318,
      "train_tokens_per_second": 61713.801
    },
    {
      "epoch": 0.9896279758938479,
      "grad_norm": 25.75,
      "learning_rate": 5.663433839024901e-09,
      "loss": 0.819,
      "num_input_tokens_seen": 5527145216,
      "step": 74470,
      "train_runtime": 89559.6839,
      "train_tokens_per_second": 61714.658
    },
    {
      "epoch": 0.9897608653763098,
      "grad_norm": 55.75,
      "learning_rate": 5.519528342680103e-09,
      "loss": 0.6738,
      "num_input_tokens_seen": 5527623424,
      "step": 74480,
      "train_runtime": 89566.405,
      "train_tokens_per_second": 61715.366
    },
    {
      "epoch": 0.9898937548587717,
      "grad_norm": 10.625,
      "learning_rate": 5.377474234723456e-09,
      "loss": 0.5947,
      "num_input_tokens_seen": 5527847680,
      "step": 74490,
      "train_runtime": 89570.7573,
      "train_tokens_per_second": 61714.87
    },
    {
      "epoch": 0.9900266443412337,
      "grad_norm": 13.375,
      "learning_rate": 5.237271541469469e-09,
      "loss": 0.4886,
      "num_input_tokens_seen": 5527960576,
      "step": 74500,
      "train_runtime": 89573.2611,
      "train_tokens_per_second": 61714.406
    },
    {
      "epoch": 0.9901595338236955,
      "grad_norm": 32.5,
      "learning_rate": 5.098920288889586e-09,
      "loss": 0.7796,
      "num_input_tokens_seen": 5530140416,
      "step": 74510,
      "train_runtime": 89607.4614,
      "train_tokens_per_second": 61715.178
    },
    {
      "epoch": 0.9902924233061574,
      "grad_norm": 31.625,
      "learning_rate": 4.9624205026121976e-09,
      "loss": 0.9037,
      "num_input_tokens_seen": 5530967552,
      "step": 74520,
      "train_runtime": 89619.7536,
      "train_tokens_per_second": 61715.942
    },
    {
      "epoch": 0.9904253127886193,
      "grad_norm": 17.25,
      "learning_rate": 4.827772207921522e-09,
      "loss": 0.7763,
      "num_input_tokens_seen": 5531420416,
      "step": 74530,
      "train_runtime": 89626.3305,
      "train_tokens_per_second": 61716.466
    },
    {
      "epoch": 0.9905582022710813,
      "grad_norm": 23.0,
      "learning_rate": 4.69497542976094e-09,
      "loss": 0.6447,
      "num_input_tokens_seen": 5531651840,
      "step": 74540,
      "train_runtime": 89629.9512,
      "train_tokens_per_second": 61716.555
    },
    {
      "epoch": 0.9906910917535432,
      "grad_norm": 10.9375,
      "learning_rate": 4.564030192729662e-09,
      "loss": 0.5223,
      "num_input_tokens_seen": 5531766528,
      "step": 74550,
      "train_runtime": 89632.481,
      "train_tokens_per_second": 61716.093
    },
    {
      "epoch": 0.9908239812360051,
      "grad_norm": 34.25,
      "learning_rate": 4.434936521083844e-09,
      "loss": 0.7968,
      "num_input_tokens_seen": 5533995264,
      "step": 74560,
      "train_runtime": 89668.743,
      "train_tokens_per_second": 61715.99
    },
    {
      "epoch": 0.9909568707184669,
      "grad_norm": 34.75,
      "learning_rate": 4.307694438737686e-09,
      "loss": 0.9911,
      "num_input_tokens_seen": 5534877696,
      "step": 74570,
      "train_runtime": 89681.9244,
      "train_tokens_per_second": 61716.759
    },
    {
      "epoch": 0.9910897602009289,
      "grad_norm": 37.25,
      "learning_rate": 4.1823039692612256e-09,
      "loss": 0.7852,
      "num_input_tokens_seen": 5535363328,
      "step": 74580,
      "train_runtime": 89688.9563,
      "train_tokens_per_second": 61717.335
    },
    {
      "epoch": 0.9912226496833908,
      "grad_norm": 19.5,
      "learning_rate": 4.058765135881437e-09,
      "loss": 0.6583,
      "num_input_tokens_seen": 5535603200,
      "step": 74590,
      "train_runtime": 89692.5067,
      "train_tokens_per_second": 61717.566
    },
    {
      "epoch": 0.9913555391658527,
      "grad_norm": 16.875,
      "learning_rate": 3.937077961483349e-09,
      "loss": 0.5194,
      "num_input_tokens_seen": 5535725568,
      "step": 74600,
      "train_runtime": 89695.1919,
      "train_tokens_per_second": 61717.083
    },
    {
      "epoch": 0.9914884286483147,
      "grad_norm": 19.25,
      "learning_rate": 3.817242468608928e-09,
      "loss": 0.8317,
      "num_input_tokens_seen": 5537808128,
      "step": 74610,
      "train_runtime": 89729.7491,
      "train_tokens_per_second": 61716.523
    },
    {
      "epoch": 0.9916213181307766,
      "grad_norm": 21.375,
      "learning_rate": 3.6992586794548646e-09,
      "loss": 0.8543,
      "num_input_tokens_seen": 5538665216,
      "step": 74620,
      "train_runtime": 89742.5301,
      "train_tokens_per_second": 61717.284
    },
    {
      "epoch": 0.9917542076132384,
      "grad_norm": 11.8125,
      "learning_rate": 3.583126615879229e-09,
      "loss": 0.7303,
      "num_input_tokens_seen": 5539141632,
      "step": 74630,
      "train_runtime": 89749.4147,
      "train_tokens_per_second": 61717.858
    },
    {
      "epoch": 0.9918870970957003,
      "grad_norm": 12.9375,
      "learning_rate": 3.468846299391482e-09,
      "loss": 0.6069,
      "num_input_tokens_seen": 5539363840,
      "step": 74640,
      "train_runtime": 89752.9571,
      "train_tokens_per_second": 61717.898
    },
    {
      "epoch": 0.9920199865781623,
      "grad_norm": 11.375,
      "learning_rate": 3.3564177511635763e-09,
      "loss": 0.5221,
      "num_input_tokens_seen": 5539477248,
      "step": 74650,
      "train_runtime": 89755.4226,
      "train_tokens_per_second": 61717.466
    },
    {
      "epoch": 0.9921528760606242,
      "grad_norm": 29.0,
      "learning_rate": 3.2458409920199665e-09,
      "loss": 0.8365,
      "num_input_tokens_seen": 5541529600,
      "step": 74660,
      "train_runtime": 89787.7169,
      "train_tokens_per_second": 61718.126
    },
    {
      "epoch": 0.9922857655430861,
      "grad_norm": 30.0,
      "learning_rate": 3.1371160424464865e-09,
      "loss": 0.8715,
      "num_input_tokens_seen": 5542418688,
      "step": 74670,
      "train_runtime": 89801.6657,
      "train_tokens_per_second": 61718.44
    },
    {
      "epoch": 0.992418655025548,
      "grad_norm": 17.125,
      "learning_rate": 3.030242922580362e-09,
      "loss": 0.7245,
      "num_input_tokens_seen": 5542895104,
      "step": 74680,
      "train_runtime": 89808.5362,
      "train_tokens_per_second": 61719.023
    },
    {
      "epoch": 0.9925515445080099,
      "grad_norm": 20.875,
      "learning_rate": 2.9252216522213105e-09,
      "loss": 0.6403,
      "num_input_tokens_seen": 5543121408,
      "step": 74690,
      "train_runtime": 89812.1356,
      "train_tokens_per_second": 61719.069
    },
    {
      "epoch": 0.9926844339904718,
      "grad_norm": 10.375,
      "learning_rate": 2.8220522508215497e-09,
      "loss": 0.54,
      "num_input_tokens_seen": 5543239936,
      "step": 74700,
      "train_runtime": 89814.5316,
      "train_tokens_per_second": 61718.742
    },
    {
      "epoch": 0.9928173234729337,
      "grad_norm": 39.0,
      "learning_rate": 2.720734737494679e-09,
      "loss": 0.8598,
      "num_input_tokens_seen": 5545176320,
      "step": 74710,
      "train_runtime": 89845.421,
      "train_tokens_per_second": 61719.075
    },
    {
      "epoch": 0.9929502129553957,
      "grad_norm": 19.375,
      "learning_rate": 2.6212691310067984e-09,
      "loss": 0.9232,
      "num_input_tokens_seen": 5545989888,
      "step": 74720,
      "train_runtime": 89857.6044,
      "train_tokens_per_second": 61719.761
    },
    {
      "epoch": 0.9930831024378576,
      "grad_norm": 13.75,
      "learning_rate": 2.5236554497842793e-09,
      "loss": 0.7983,
      "num_input_tokens_seen": 5546418688,
      "step": 74730,
      "train_runtime": 89865.2332,
      "train_tokens_per_second": 61719.293
    },
    {
      "epoch": 0.9932159919203195,
      "grad_norm": 18.375,
      "learning_rate": 2.4278937119082136e-09,
      "loss": 0.6437,
      "num_input_tokens_seen": 5546645760,
      "step": 74740,
      "train_runtime": 89868.6554,
      "train_tokens_per_second": 61719.47
    },
    {
      "epoch": 0.9933488814027813,
      "grad_norm": 42.75,
      "learning_rate": 2.333983935118855e-09,
      "loss": 0.5549,
      "num_input_tokens_seen": 5546763520,
      "step": 74750,
      "train_runtime": 89871.1485,
      "train_tokens_per_second": 61719.068
    },
    {
      "epoch": 0.9934817708852433,
      "grad_norm": 27.25,
      "learning_rate": 2.2419261368111787e-09,
      "loss": 0.8411,
      "num_input_tokens_seen": 5548718848,
      "step": 74760,
      "train_runtime": 89903.4021,
      "train_tokens_per_second": 61718.675
    },
    {
      "epoch": 0.9936146603677052,
      "grad_norm": 28.625,
      "learning_rate": 2.15172033403821e-09,
      "loss": 0.8548,
      "num_input_tokens_seen": 5549531904,
      "step": 74770,
      "train_runtime": 89915.3019,
      "train_tokens_per_second": 61719.549
    },
    {
      "epoch": 0.9937475498501671,
      "grad_norm": 14.6875,
      "learning_rate": 2.0633665435099147e-09,
      "loss": 0.7359,
      "num_input_tokens_seen": 5549957632,
      "step": 74780,
      "train_runtime": 89922.3657,
      "train_tokens_per_second": 61719.435
    },
    {
      "epoch": 0.993880439332629,
      "grad_norm": 17.5,
      "learning_rate": 1.9768647815920914e-09,
      "loss": 0.6126,
      "num_input_tokens_seen": 5550180096,
      "step": 74790,
      "train_runtime": 89925.7278,
      "train_tokens_per_second": 61719.602
    },
    {
      "epoch": 0.994013328815091,
      "grad_norm": 9.0625,
      "learning_rate": 1.8922150643108094e-09,
      "loss": 0.5411,
      "num_input_tokens_seen": 5550293760,
      "step": 74800,
      "train_runtime": 89928.1326,
      "train_tokens_per_second": 61719.215
    },
    {
      "epoch": 0.9941462182975528,
      "grad_norm": 15.5,
      "learning_rate": 1.809417407344638e-09,
      "loss": 0.7367,
      "num_input_tokens_seen": 5552444160,
      "step": 74810,
      "train_runtime": 89961.5519,
      "train_tokens_per_second": 61720.191
    },
    {
      "epoch": 0.9942791077800147,
      "grad_norm": 19.875,
      "learning_rate": 1.7284718260313082e-09,
      "loss": 0.8973,
      "num_input_tokens_seen": 5553267456,
      "step": 74820,
      "train_runtime": 89974.8364,
      "train_tokens_per_second": 61720.228
    },
    {
      "epoch": 0.9944119972624766,
      "grad_norm": 31.625,
      "learning_rate": 1.6493783353654923e-09,
      "loss": 0.7395,
      "num_input_tokens_seen": 5553736192,
      "step": 74830,
      "train_runtime": 89981.3963,
      "train_tokens_per_second": 61720.938
    },
    {
      "epoch": 0.9945448867449386,
      "grad_norm": 23.75,
      "learning_rate": 1.572136949999914e-09,
      "loss": 0.6399,
      "num_input_tokens_seen": 5553968384,
      "step": 74840,
      "train_runtime": 89985.0087,
      "train_tokens_per_second": 61721.041
    },
    {
      "epoch": 0.9946777762274005,
      "grad_norm": 12.875,
      "learning_rate": 1.4967476842409068e-09,
      "loss": 0.5134,
      "num_input_tokens_seen": 5554084352,
      "step": 74850,
      "train_runtime": 89987.411,
      "train_tokens_per_second": 61720.682
    },
    {
      "epoch": 0.9948106657098624,
      "grad_norm": 22.125,
      "learning_rate": 1.4232105520539662e-09,
      "loss": 0.8188,
      "num_input_tokens_seen": 5556038400,
      "step": 74860,
      "train_runtime": 90019.9472,
      "train_tokens_per_second": 61720.081
    },
    {
      "epoch": 0.9949435551923244,
      "grad_norm": 20.125,
      "learning_rate": 1.351525567061529e-09,
      "loss": 0.9044,
      "num_input_tokens_seen": 5556919040,
      "step": 74870,
      "train_runtime": 90032.4507,
      "train_tokens_per_second": 61721.29
    },
    {
      "epoch": 0.9950764446747862,
      "grad_norm": 97.0,
      "learning_rate": 1.2816927425440828e-09,
      "loss": 0.7416,
      "num_input_tokens_seen": 5557359104,
      "step": 74880,
      "train_runtime": 90038.7877,
      "train_tokens_per_second": 61721.834
    },
    {
      "epoch": 0.9952093341572481,
      "grad_norm": 13.875,
      "learning_rate": 1.2137120914346157e-09,
      "loss": 0.6455,
      "num_input_tokens_seen": 5557590784,
      "step": 74890,
      "train_runtime": 90043.2358,
      "train_tokens_per_second": 61721.358
    },
    {
      "epoch": 0.99534222363971,
      "grad_norm": 21.625,
      "learning_rate": 1.1475836263286078e-09,
      "loss": 0.5299,
      "num_input_tokens_seen": 5557706752,
      "step": 74900,
      "train_runtime": 90045.633,
      "train_tokens_per_second": 61721.003
    },
    {
      "epoch": 0.995475113122172,
      "grad_norm": 19.625,
      "learning_rate": 1.0833073594729293e-09,
      "loss": 0.7655,
      "num_input_tokens_seen": 5559764736,
      "step": 74910,
      "train_runtime": 90078.5239,
      "train_tokens_per_second": 61721.313
    },
    {
      "epoch": 0.9956080026046339,
      "grad_norm": 24.0,
      "learning_rate": 1.0208833027769426e-09,
      "loss": 0.8183,
      "num_input_tokens_seen": 5560601344,
      "step": 74920,
      "train_runtime": 90091.0018,
      "train_tokens_per_second": 61722.05
    },
    {
      "epoch": 0.9957408920870958,
      "grad_norm": 41.0,
      "learning_rate": 9.603114678036207e-10,
      "loss": 0.806,
      "num_input_tokens_seen": 5561069568,
      "step": 74930,
      "train_runtime": 90097.6499,
      "train_tokens_per_second": 61722.693
    },
    {
      "epoch": 0.9958737815695576,
      "grad_norm": 15.625,
      "learning_rate": 9.015918657717671e-10,
      "loss": 0.6643,
      "num_input_tokens_seen": 5561297664,
      "step": 74940,
      "train_runtime": 90102.0427,
      "train_tokens_per_second": 61722.215
    },
    {
      "epoch": 0.9960066710520196,
      "grad_norm": 11.1875,
      "learning_rate": 8.447245075604571e-10,
      "loss": 0.554,
      "num_input_tokens_seen": 5561411840,
      "step": 74950,
      "train_runtime": 90104.4087,
      "train_tokens_per_second": 61721.862
    },
    {
      "epoch": 0.9961395605344815,
      "grad_norm": 21.625,
      "learning_rate": 7.897094037034869e-10,
      "loss": 0.8281,
      "num_input_tokens_seen": 5563424000,
      "step": 74960,
      "train_runtime": 90136.3502,
      "train_tokens_per_second": 61722.313
    },
    {
      "epoch": 0.9962724500169434,
      "grad_norm": 21.375,
      "learning_rate": 7.365465643904835e-10,
      "loss": 0.9171,
      "num_input_tokens_seen": 5564288000,
      "step": 74970,
      "train_runtime": 90148.8991,
      "train_tokens_per_second": 61723.305
    },
    {
      "epoch": 0.9964053394994054,
      "grad_norm": 29.25,
      "learning_rate": 6.852359994713453e-10,
      "loss": 0.7395,
      "num_input_tokens_seen": 5564726528,
      "step": 74980,
      "train_runtime": 90155.2537,
      "train_tokens_per_second": 61723.819
    },
    {
      "epoch": 0.9965382289818673,
      "grad_norm": 11.9375,
      "learning_rate": 6.357777184484715e-10,
      "loss": 0.656,
      "num_input_tokens_seen": 5564944128,
      "step": 74990,
      "train_runtime": 90159.614,
      "train_tokens_per_second": 61723.247
    },
    {
      "epoch": 0.9966711184643291,
      "grad_norm": 12.8125,
      "learning_rate": 5.881717304867529e-10,
      "loss": 0.5028,
      "num_input_tokens_seen": 5565058048,
      "step": 75000,
      "train_runtime": 90161.9936,
      "train_tokens_per_second": 61722.882
    },
    {
      "epoch": 0.9966711184643291,
      "eval_accuracy": 0.8187,
      "eval_balanced_accuracy": 0.8169942573687645,
      "eval_invalid_f1": 0.8363866077068857,
      "eval_invalid_precision": 0.840863727091272,
      "eval_invalid_recall": 0.8319569120287253,
      "eval_loss": 0.3614175617694855,
      "eval_macro_f1": 0.8165563490378805,
      "eval_mcc": 0.6331707483287724,
      "eval_runtime": 81.3401,
      "eval_samples_per_second": 122.941,
      "eval_steps_per_second": 30.735,
      "eval_valid_f1": 0.7967260903688754,
      "eval_valid_precision": 0.7914903096458008,
      "eval_valid_recall": 0.8020316027088036,
      "num_input_tokens_seen": 5565058048,
      "step": 75000
    },
    {
      "epoch": 0.996804007946791,
      "grad_norm": 34.75,
      "learning_rate": 5.424180444024707e-10,
      "loss": 0.7397,
      "num_input_tokens_seen": 5567149312,
      "step": 75010,
      "train_runtime": 90281.1767,
      "train_tokens_per_second": 61664.563
    },
    {
      "epoch": 0.996936897429253,
      "grad_norm": 12.0,
      "learning_rate": 4.985166686721776e-10,
      "loss": 0.9456,
      "num_input_tokens_seen": 5568013312,
      "step": 75020,
      "train_runtime": 90294.2087,
      "train_tokens_per_second": 61665.232
    },
    {
      "epoch": 0.9970697869117149,
      "grad_norm": 10.875,
      "learning_rate": 4.5646761142714715e-10,
      "loss": 0.8143,
      "num_input_tokens_seen": 5568500992,
      "step": 75030,
      "train_runtime": 90301.1182,
      "train_tokens_per_second": 61665.914
    },
    {
      "epoch": 0.9972026763941768,
      "grad_norm": 12.8125,
      "learning_rate": 4.1627088045670393e-10,
      "loss": 0.6002,
      "num_input_tokens_seen": 5568745984,
      "step": 75040,
      "train_runtime": 90304.7333,
      "train_tokens_per_second": 61666.158
    },
    {
      "epoch": 0.9973355658766387,
      "grad_norm": 11.9375,
      "learning_rate": 3.7792648320822413e-10,
      "loss": 0.5811,
      "num_input_tokens_seen": 5568862208,
      "step": 75050,
      "train_runtime": 90307.1726,
      "train_tokens_per_second": 61665.78
    },
    {
      "epoch": 0.9974684553591006,
      "grad_norm": 19.125,
      "learning_rate": 3.4143442678380433e-10,
      "loss": 0.7669,
      "num_input_tokens_seen": 5571071488,
      "step": 75060,
      "train_runtime": 90342.251,
      "train_tokens_per_second": 61666.29
    },
    {
      "epoch": 0.9976013448415625,
      "grad_norm": 26.25,
      "learning_rate": 3.067947179435926e-10,
      "loss": 0.9283,
      "num_input_tokens_seen": 5571936000,
      "step": 75070,
      "train_runtime": 90355.2639,
      "train_tokens_per_second": 61666.977
    },
    {
      "epoch": 0.9977342343240244,
      "grad_norm": 110.0,
      "learning_rate": 2.740073631035678e-10,
      "loss": 0.7651,
      "num_input_tokens_seen": 5572423936,
      "step": 75080,
      "train_runtime": 90362.2844,
      "train_tokens_per_second": 61667.586
    },
    {
      "epoch": 0.9978671238064863,
      "grad_norm": 21.25,
      "learning_rate": 2.4307236833776007e-10,
      "loss": 0.6204,
      "num_input_tokens_seen": 5572657408,
      "step": 75090,
      "train_runtime": 90365.8702,
      "train_tokens_per_second": 61667.723
    },
    {
      "epoch": 0.9980000132889483,
      "grad_norm": 8.0,
      "learning_rate": 2.1398973937714063e-10,
      "loss": 0.5551,
      "num_input_tokens_seen": 5572774400,
      "step": 75100,
      "train_runtime": 90368.2842,
      "train_tokens_per_second": 61667.37
    },
    {
      "epoch": 0.9981329027714102,
      "grad_norm": 22.625,
      "learning_rate": 1.8675948160851166e-10,
      "loss": 0.8311,
      "num_input_tokens_seen": 5574959360,
      "step": 75110,
      "train_runtime": 90403.4526,
      "train_tokens_per_second": 61667.549
    },
    {
      "epoch": 0.998265792253872,
      "grad_norm": 101.5,
      "learning_rate": 1.613816000756163e-10,
      "loss": 0.9369,
      "num_input_tokens_seen": 5575821312,
      "step": 75120,
      "train_runtime": 90416.4421,
      "train_tokens_per_second": 61668.223
    },
    {
      "epoch": 0.998398681736334,
      "grad_norm": 31.375,
      "learning_rate": 1.3785609947913892e-10,
      "loss": 0.7335,
      "num_input_tokens_seen": 5576339456,
      "step": 75130,
      "train_runtime": 90423.4946,
      "train_tokens_per_second": 61669.143
    },
    {
      "epoch": 0.9985315712187959,
      "grad_norm": 15.6875,
      "learning_rate": 1.1618298417892526e-10,
      "loss": 0.6987,
      "num_input_tokens_seen": 5576567808,
      "step": 75140,
      "train_runtime": 90426.9829,
      "train_tokens_per_second": 61669.29
    },
    {
      "epoch": 0.9986644607012578,
      "grad_norm": 9.0625,
      "learning_rate": 9.636225818843159e-11,
      "loss": 0.5453,
      "num_input_tokens_seen": 5576680704,
      "step": 75150,
      "train_runtime": 90429.4011,
      "train_tokens_per_second": 61668.889
    },
    {
      "epoch": 0.9987973501837197,
      "grad_norm": 15.9375,
      "learning_rate": 7.839392517805521e-11,
      "loss": 0.8376,
      "num_input_tokens_seen": 5578687232,
      "step": 75160,
      "train_runtime": 90461.8927,
      "train_tokens_per_second": 61668.92
    },
    {
      "epoch": 0.9989302396661817,
      "grad_norm": 24.5,
      "learning_rate": 6.227798847846522e-11,
      "loss": 0.8895,
      "num_input_tokens_seen": 5579519232,
      "step": 75170,
      "train_runtime": 90475.7193,
      "train_tokens_per_second": 61668.692
    },
    {
      "epoch": 0.9990631291486435,
      "grad_norm": 29.75,
      "learning_rate": 4.8014451073941095e-11,
      "loss": 0.7914,
      "num_input_tokens_seen": 5579982592,
      "step": 75180,
      "train_runtime": 90482.4408,
      "train_tokens_per_second": 61669.232
    },
    {
      "epoch": 0.9991960186311054,
      "grad_norm": 23.125,
      "learning_rate": 3.560331560681363e-11,
      "loss": 0.6048,
      "num_input_tokens_seen": 5580219648,
      "step": 75190,
      "train_runtime": 90485.9934,
      "train_tokens_per_second": 61669.43
    },
    {
      "epoch": 0.9993289081135673,
      "grad_norm": 11.125,
      "learning_rate": 2.5044584376354708e-11,
      "loss": 0.4909,
      "num_input_tokens_seen": 5580336128,
      "step": 75200,
      "train_runtime": 90488.4429,
      "train_tokens_per_second": 61669.048
    },
    {
      "epoch": 0.9994617975960293,
      "grad_norm": 22.0,
      "learning_rate": 1.6338259338777308e-11,
      "loss": 0.8271,
      "num_input_tokens_seen": 5582326528,
      "step": 75210,
      "train_runtime": 90520.4372,
      "train_tokens_per_second": 61669.24
    },
    {
      "epoch": 0.9995946870784912,
      "grad_norm": 30.0,
      "learning_rate": 9.484342105015031e-12,
      "loss": 0.9273,
      "num_input_tokens_seen": 5583172352,
      "step": 75220,
      "train_runtime": 90533.7441,
      "train_tokens_per_second": 61669.518
    },
    {
      "epoch": 0.9997275765609531,
      "grad_norm": 15.9375,
      "learning_rate": 4.482833947383469e-12,
      "loss": 0.7202,
      "num_input_tokens_seen": 5583635200,
      "step": 75230,
      "train_runtime": 90540.3842,
      "train_tokens_per_second": 61670.107
    },
    {
      "epoch": 0.9998604660434149,
      "grad_norm": 12.5625,
      "learning_rate": 1.3337357895881754e-12,
      "loss": 0.6509,
      "num_input_tokens_seen": 5583855616,
      "step": 75240,
      "train_runtime": 90543.7862,
      "train_tokens_per_second": 61670.224
    },
    {
      "epoch": 0.9999933555258769,
      "grad_norm": 8.75,
      "learning_rate": 3.7048216716684124e-14,
      "loss": 0.5013,
      "num_input_tokens_seen": 5583968000,
      "step": 75250,
      "train_runtime": 90546.2639,
      "train_tokens_per_second": 61669.778
    },
    {
      "epoch": 1.0,
      "num_input_tokens_seen": 5584037048,
      "step": 75251,
      "total_flos": 1.1534046769147976e+19,
      "train_loss": 0.7614488703141813,
      "train_runtime": 90550.6974,
      "train_samples_per_second": 53.186,
      "train_steps_per_second": 0.831
    }
  ],
  "logging_steps": 10,
  "max_steps": 75251,
  "num_input_tokens_seen": 5584037048,
  "num_train_epochs": 1,
  "save_steps": 5000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.1534046769147976e+19,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}