{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.5925925925925926,
  "eval_steps": 500,
  "global_step": 200,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.002962962962962963,
      "grad_norm": 1.385405421257019,
      "learning_rate": 0.0,
      "loss": 3.4616,
      "num_input_tokens_seen": 3994,
      "step": 1,
      "train_runtime": 3.8769,
      "train_tokens_per_second": 1030.2
    },
    {
      "epoch": 0.005925925925925926,
      "grad_norm": 1.3795109987258911,
      "learning_rate": 8.000000000000001e-06,
      "loss": 3.5106,
      "num_input_tokens_seen": 8015,
      "step": 2,
      "train_runtime": 5.1516,
      "train_tokens_per_second": 1555.825
    },
    {
      "epoch": 0.008888888888888889,
      "grad_norm": 1.4005517959594727,
      "learning_rate": 1.6000000000000003e-05,
      "loss": 3.4945,
      "num_input_tokens_seen": 11971,
      "step": 3,
      "train_runtime": 6.4487,
      "train_tokens_per_second": 1856.354
    },
    {
      "epoch": 0.011851851851851851,
      "grad_norm": 1.4066334962844849,
      "learning_rate": 2.4e-05,
      "loss": 3.5379,
      "num_input_tokens_seen": 15942,
      "step": 4,
      "train_runtime": 7.6644,
      "train_tokens_per_second": 2080.001
    },
    {
      "epoch": 0.014814814814814815,
      "grad_norm": 1.4247043132781982,
      "learning_rate": 3.2000000000000005e-05,
      "loss": 3.4672,
      "num_input_tokens_seen": 19932,
      "step": 5,
      "train_runtime": 8.8904,
      "train_tokens_per_second": 2241.962
    },
    {
      "epoch": 0.017777777777777778,
      "grad_norm": 1.5897549390792847,
      "learning_rate": 4e-05,
      "loss": 3.5563,
      "num_input_tokens_seen": 23728,
      "step": 6,
      "train_runtime": 10.0881,
      "train_tokens_per_second": 2352.082
    },
    {
      "epoch": 0.02074074074074074,
      "grad_norm": 1.5332510471343994,
      "learning_rate": 4.8e-05,
      "loss": 3.3657,
      "num_input_tokens_seen": 27745,
      "step": 7,
      "train_runtime": 11.3378,
      "train_tokens_per_second": 2447.127
    },
    {
      "epoch": 0.023703703703703703,
      "grad_norm": 1.535454273223877,
      "learning_rate": 5.6000000000000006e-05,
      "loss": 3.171,
      "num_input_tokens_seen": 31789,
      "step": 8,
      "train_runtime": 12.6529,
      "train_tokens_per_second": 2512.385
    },
    {
      "epoch": 0.02666666666666667,
      "grad_norm": 1.5178135633468628,
      "learning_rate": 6.400000000000001e-05,
      "loss": 3.0904,
      "num_input_tokens_seen": 35703,
      "step": 9,
      "train_runtime": 13.8596,
      "train_tokens_per_second": 2576.043
    },
    {
      "epoch": 0.02962962962962963,
      "grad_norm": 1.4022459983825684,
      "learning_rate": 7.2e-05,
      "loss": 2.8932,
      "num_input_tokens_seen": 39739,
      "step": 10,
      "train_runtime": 15.118,
      "train_tokens_per_second": 2628.595
    },
    {
      "epoch": 0.03259259259259259,
      "grad_norm": 1.4071623086929321,
      "learning_rate": 8e-05,
      "loss": 2.6759,
      "num_input_tokens_seen": 43738,
      "step": 11,
      "train_runtime": 16.4007,
      "train_tokens_per_second": 2666.842
    },
    {
      "epoch": 0.035555555555555556,
      "grad_norm": 1.3278772830963135,
      "learning_rate": 8.800000000000001e-05,
      "loss": 2.3702,
      "num_input_tokens_seen": 47721,
      "step": 12,
      "train_runtime": 17.6351,
      "train_tokens_per_second": 2706.022
    },
    {
      "epoch": 0.03851851851851852,
      "grad_norm": 1.3079149723052979,
      "learning_rate": 9.6e-05,
      "loss": 2.2334,
      "num_input_tokens_seen": 51717,
      "step": 13,
      "train_runtime": 18.8949,
      "train_tokens_per_second": 2737.093
    },
    {
      "epoch": 0.04148148148148148,
      "grad_norm": 1.3329426050186157,
      "learning_rate": 0.00010400000000000001,
      "loss": 2.07,
      "num_input_tokens_seen": 55684,
      "step": 14,
      "train_runtime": 20.1266,
      "train_tokens_per_second": 2766.682
    },
    {
      "epoch": 0.044444444444444446,
      "grad_norm": 1.2982901334762573,
      "learning_rate": 0.00011200000000000001,
      "loss": 1.8083,
      "num_input_tokens_seen": 59608,
      "step": 15,
      "train_runtime": 21.3445,
      "train_tokens_per_second": 2792.666
    },
    {
      "epoch": 0.047407407407407405,
      "grad_norm": 1.5552537441253662,
      "learning_rate": 0.00012,
      "loss": 1.6607,
      "num_input_tokens_seen": 63560,
      "step": 16,
      "train_runtime": 22.5703,
      "train_tokens_per_second": 2816.096
    },
    {
      "epoch": 0.05037037037037037,
      "grad_norm": 1.1010578870773315,
      "learning_rate": 0.00012800000000000002,
      "loss": 1.4109,
      "num_input_tokens_seen": 67560,
      "step": 17,
      "train_runtime": 23.8287,
      "train_tokens_per_second": 2835.231
    },
    {
      "epoch": 0.05333333333333334,
      "grad_norm": 0.922965407371521,
      "learning_rate": 0.00013600000000000003,
      "loss": 1.2918,
      "num_input_tokens_seen": 71542,
      "step": 18,
      "train_runtime": 25.0591,
      "train_tokens_per_second": 2854.931
    },
    {
      "epoch": 0.056296296296296296,
      "grad_norm": 0.8357541561126709,
      "learning_rate": 0.000144,
      "loss": 1.1507,
      "num_input_tokens_seen": 75529,
      "step": 19,
      "train_runtime": 26.339,
      "train_tokens_per_second": 2867.576
    },
    {
      "epoch": 0.05925925925925926,
      "grad_norm": 0.8259718418121338,
      "learning_rate": 0.000152,
      "loss": 1.1098,
      "num_input_tokens_seen": 79401,
      "step": 20,
      "train_runtime": 27.5498,
      "train_tokens_per_second": 2882.092
    },
    {
      "epoch": 0.06222222222222222,
      "grad_norm": 0.8687456846237183,
      "learning_rate": 0.00016,
      "loss": 1.087,
      "num_input_tokens_seen": 83402,
      "step": 21,
      "train_runtime": 28.8054,
      "train_tokens_per_second": 2895.362
    },
    {
      "epoch": 0.06518518518518518,
      "grad_norm": 0.8352484107017517,
      "learning_rate": 0.000168,
      "loss": 1.0134,
      "num_input_tokens_seen": 87407,
      "step": 22,
      "train_runtime": 30.0302,
      "train_tokens_per_second": 2910.632
    },
    {
      "epoch": 0.06814814814814815,
      "grad_norm": 0.6514259576797485,
      "learning_rate": 0.00017600000000000002,
      "loss": 0.9516,
      "num_input_tokens_seen": 91294,
      "step": 23,
      "train_runtime": 31.2536,
      "train_tokens_per_second": 2921.072
    },
    {
      "epoch": 0.07111111111111111,
      "grad_norm": 0.7477912306785583,
      "learning_rate": 0.00018400000000000003,
      "loss": 0.8807,
      "num_input_tokens_seen": 95297,
      "step": 24,
      "train_runtime": 32.5055,
      "train_tokens_per_second": 2931.715
    },
    {
      "epoch": 0.07407407407407407,
      "grad_norm": 0.8334610462188721,
      "learning_rate": 0.000192,
      "loss": 0.8315,
      "num_input_tokens_seen": 99249,
      "step": 25,
      "train_runtime": 33.7608,
      "train_tokens_per_second": 2939.768
    },
    {
      "epoch": 0.07703703703703704,
      "grad_norm": 0.691455066204071,
      "learning_rate": 0.0002,
      "loss": 0.7369,
      "num_input_tokens_seen": 103227,
      "step": 26,
      "train_runtime": 35.0243,
      "train_tokens_per_second": 2947.294
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.561080813407898,
      "learning_rate": 0.00019936102236421725,
      "loss": 0.6748,
      "num_input_tokens_seen": 107157,
      "step": 27,
      "train_runtime": 36.258,
      "train_tokens_per_second": 2955.401
    },
    {
      "epoch": 0.08296296296296296,
      "grad_norm": 0.6379782557487488,
      "learning_rate": 0.00019872204472843452,
      "loss": 0.6141,
      "num_input_tokens_seen": 110941,
      "step": 28,
      "train_runtime": 37.4773,
      "train_tokens_per_second": 2960.216
    },
    {
      "epoch": 0.08592592592592592,
      "grad_norm": 0.5655418038368225,
      "learning_rate": 0.00019808306709265177,
      "loss": 0.5782,
      "num_input_tokens_seen": 114876,
      "step": 29,
      "train_runtime": 38.7094,
      "train_tokens_per_second": 2967.649
    },
    {
      "epoch": 0.08888888888888889,
      "grad_norm": 0.5590977072715759,
      "learning_rate": 0.000197444089456869,
      "loss": 0.5252,
      "num_input_tokens_seen": 118843,
      "step": 30,
      "train_runtime": 39.9784,
      "train_tokens_per_second": 2972.68
    },
    {
      "epoch": 0.09185185185185185,
      "grad_norm": 0.7514940500259399,
      "learning_rate": 0.00019680511182108628,
      "loss": 0.4674,
      "num_input_tokens_seen": 123086,
      "step": 31,
      "train_runtime": 41.3461,
      "train_tokens_per_second": 2976.969
    },
    {
      "epoch": 0.09481481481481481,
      "grad_norm": 9.792972564697266,
      "learning_rate": 0.00019616613418530353,
      "loss": 0.5088,
      "num_input_tokens_seen": 126976,
      "step": 32,
      "train_runtime": 42.5837,
      "train_tokens_per_second": 2981.796
    },
    {
      "epoch": 0.09777777777777778,
      "grad_norm": 0.8071851134300232,
      "learning_rate": 0.0001955271565495208,
      "loss": 0.4441,
      "num_input_tokens_seen": 130873,
      "step": 33,
      "train_runtime": 43.804,
      "train_tokens_per_second": 2987.692
    },
    {
      "epoch": 0.10074074074074074,
      "grad_norm": 0.41566988825798035,
      "learning_rate": 0.00019488817891373804,
      "loss": 0.4184,
      "num_input_tokens_seen": 134897,
      "step": 34,
      "train_runtime": 45.0746,
      "train_tokens_per_second": 2992.752
    },
    {
      "epoch": 0.1037037037037037,
      "grad_norm": 0.45239996910095215,
      "learning_rate": 0.00019424920127795528,
      "loss": 0.4151,
      "num_input_tokens_seen": 138952,
      "step": 35,
      "train_runtime": 46.3707,
      "train_tokens_per_second": 2996.546
    },
    {
      "epoch": 0.10666666666666667,
      "grad_norm": 0.5036719441413879,
      "learning_rate": 0.00019361022364217253,
      "loss": 0.4053,
      "num_input_tokens_seen": 142838,
      "step": 36,
      "train_runtime": 47.5967,
      "train_tokens_per_second": 3001.007
    },
    {
      "epoch": 0.10962962962962963,
      "grad_norm": 0.32047784328460693,
      "learning_rate": 0.00019297124600638977,
      "loss": 0.3702,
      "num_input_tokens_seen": 146830,
      "step": 37,
      "train_runtime": 48.8631,
      "train_tokens_per_second": 3004.928
    },
    {
      "epoch": 0.11259259259259259,
      "grad_norm": 0.31912440061569214,
      "learning_rate": 0.00019233226837060702,
      "loss": 0.37,
      "num_input_tokens_seen": 150775,
      "step": 38,
      "train_runtime": 50.0864,
      "train_tokens_per_second": 3010.297
    },
    {
      "epoch": 0.11555555555555555,
      "grad_norm": 0.3148290514945984,
      "learning_rate": 0.00019169329073482429,
      "loss": 0.3443,
      "num_input_tokens_seen": 154796,
      "step": 39,
      "train_runtime": 51.3635,
      "train_tokens_per_second": 3013.735
    },
    {
      "epoch": 0.11851851851851852,
      "grad_norm": 0.3976413309574127,
      "learning_rate": 0.00019105431309904153,
      "loss": 0.3485,
      "num_input_tokens_seen": 158725,
      "step": 40,
      "train_runtime": 52.6305,
      "train_tokens_per_second": 3015.84
    },
    {
      "epoch": 0.12148148148148148,
      "grad_norm": 0.34046024084091187,
      "learning_rate": 0.0001904153354632588,
      "loss": 0.3467,
      "num_input_tokens_seen": 162629,
      "step": 41,
      "train_runtime": 53.8658,
      "train_tokens_per_second": 3019.15
    },
    {
      "epoch": 0.12444444444444444,
      "grad_norm": 0.2964358627796173,
      "learning_rate": 0.00018977635782747604,
      "loss": 0.3129,
      "num_input_tokens_seen": 166667,
      "step": 42,
      "train_runtime": 55.1297,
      "train_tokens_per_second": 3023.178
    },
    {
      "epoch": 0.1274074074074074,
      "grad_norm": 0.25185614824295044,
      "learning_rate": 0.0001891373801916933,
      "loss": 0.3224,
      "num_input_tokens_seen": 170582,
      "step": 43,
      "train_runtime": 56.3664,
      "train_tokens_per_second": 3026.304
    },
    {
      "epoch": 0.13037037037037036,
      "grad_norm": 0.2772059142589569,
      "learning_rate": 0.00018849840255591056,
      "loss": 0.3173,
      "num_input_tokens_seen": 174484,
      "step": 44,
      "train_runtime": 57.5902,
      "train_tokens_per_second": 3029.753
    },
    {
      "epoch": 0.13333333333333333,
      "grad_norm": 0.25855761766433716,
      "learning_rate": 0.0001878594249201278,
      "loss": 0.3018,
      "num_input_tokens_seen": 178438,
      "step": 45,
      "train_runtime": 58.8242,
      "train_tokens_per_second": 3033.413
    },
    {
      "epoch": 0.1362962962962963,
      "grad_norm": 0.29933610558509827,
      "learning_rate": 0.00018722044728434505,
      "loss": 0.3138,
      "num_input_tokens_seen": 182264,
      "step": 46,
      "train_runtime": 60.0271,
      "train_tokens_per_second": 3036.362
    },
    {
      "epoch": 0.13925925925925925,
      "grad_norm": 0.2461702972650528,
      "learning_rate": 0.00018658146964856232,
      "loss": 0.316,
      "num_input_tokens_seen": 186161,
      "step": 47,
      "train_runtime": 61.2542,
      "train_tokens_per_second": 3039.154
    },
    {
      "epoch": 0.14222222222222222,
      "grad_norm": 0.27812907099723816,
      "learning_rate": 0.00018594249201277956,
      "loss": 0.2873,
      "num_input_tokens_seen": 190179,
      "step": 48,
      "train_runtime": 62.5316,
      "train_tokens_per_second": 3041.325
    },
    {
      "epoch": 0.1451851851851852,
      "grad_norm": 0.23707370460033417,
      "learning_rate": 0.00018530351437699683,
      "loss": 0.3014,
      "num_input_tokens_seen": 194100,
      "step": 49,
      "train_runtime": 63.7484,
      "train_tokens_per_second": 3044.782
    },
    {
      "epoch": 0.14814814814814814,
      "grad_norm": 0.2359224110841751,
      "learning_rate": 0.00018466453674121408,
      "loss": 0.2962,
      "num_input_tokens_seen": 197980,
      "step": 50,
      "train_runtime": 64.9657,
      "train_tokens_per_second": 3047.454
    },
    {
      "epoch": 0.1511111111111111,
      "grad_norm": 0.25782346725463867,
      "learning_rate": 0.00018402555910543132,
      "loss": 0.2889,
      "num_input_tokens_seen": 202044,
      "step": 51,
      "train_runtime": 66.2807,
      "train_tokens_per_second": 3048.308
    },
    {
      "epoch": 0.15407407407407409,
      "grad_norm": 0.2116629183292389,
      "learning_rate": 0.0001833865814696486,
      "loss": 0.2754,
      "num_input_tokens_seen": 206007,
      "step": 52,
      "train_runtime": 67.5032,
      "train_tokens_per_second": 3051.811
    },
    {
      "epoch": 0.15703703703703703,
      "grad_norm": 0.24024730920791626,
      "learning_rate": 0.00018274760383386583,
      "loss": 0.2818,
      "num_input_tokens_seen": 209964,
      "step": 53,
      "train_runtime": 68.7566,
      "train_tokens_per_second": 3053.728
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.2122756987810135,
      "learning_rate": 0.00018210862619808308,
      "loss": 0.272,
      "num_input_tokens_seen": 213889,
      "step": 54,
      "train_runtime": 69.9948,
      "train_tokens_per_second": 3055.783
    },
    {
      "epoch": 0.16296296296296298,
      "grad_norm": 0.23215611279010773,
      "learning_rate": 0.00018146964856230032,
      "loss": 0.2547,
      "num_input_tokens_seen": 217957,
      "step": 55,
      "train_runtime": 71.2977,
      "train_tokens_per_second": 3057.0
    },
    {
      "epoch": 0.16592592592592592,
      "grad_norm": 0.21076326072216034,
      "learning_rate": 0.00018083067092651756,
      "loss": 0.261,
      "num_input_tokens_seen": 221959,
      "step": 56,
      "train_runtime": 72.5519,
      "train_tokens_per_second": 3059.312
    },
    {
      "epoch": 0.1688888888888889,
      "grad_norm": 0.20288218557834625,
      "learning_rate": 0.00018019169329073484,
      "loss": 0.278,
      "num_input_tokens_seen": 225835,
      "step": 57,
      "train_runtime": 73.778,
      "train_tokens_per_second": 3061.009
    },
    {
      "epoch": 0.17185185185185184,
      "grad_norm": 0.2418140172958374,
      "learning_rate": 0.00017955271565495208,
      "loss": 0.2568,
      "num_input_tokens_seen": 229841,
      "step": 58,
      "train_runtime": 75.0193,
      "train_tokens_per_second": 3063.76
    },
    {
      "epoch": 0.1748148148148148,
      "grad_norm": 0.22096514701843262,
      "learning_rate": 0.00017891373801916932,
      "loss": 0.2553,
      "num_input_tokens_seen": 233867,
      "step": 59,
      "train_runtime": 76.2838,
      "train_tokens_per_second": 3065.749
    },
    {
      "epoch": 0.17777777777777778,
      "grad_norm": 0.22594992816448212,
      "learning_rate": 0.0001782747603833866,
      "loss": 0.2574,
      "num_input_tokens_seen": 237910,
      "step": 60,
      "train_runtime": 77.5149,
      "train_tokens_per_second": 3069.218
    },
    {
      "epoch": 0.18074074074074073,
      "grad_norm": 0.231551855802536,
      "learning_rate": 0.00017763578274760384,
      "loss": 0.2667,
      "num_input_tokens_seen": 241892,
      "step": 61,
      "train_runtime": 78.7774,
      "train_tokens_per_second": 3070.577
    },
    {
      "epoch": 0.1837037037037037,
      "grad_norm": 0.22881142795085907,
      "learning_rate": 0.00017699680511182108,
      "loss": 0.2558,
      "num_input_tokens_seen": 245875,
      "step": 62,
      "train_runtime": 80.0623,
      "train_tokens_per_second": 3071.045
    },
    {
      "epoch": 0.18666666666666668,
      "grad_norm": 0.22949190437793732,
      "learning_rate": 0.00017635782747603835,
      "loss": 0.2605,
      "num_input_tokens_seen": 249845,
      "step": 63,
      "train_runtime": 81.3329,
      "train_tokens_per_second": 3071.88
    },
    {
      "epoch": 0.18962962962962962,
      "grad_norm": 0.2060144692659378,
      "learning_rate": 0.0001757188498402556,
      "loss": 0.244,
      "num_input_tokens_seen": 253829,
      "step": 64,
      "train_runtime": 82.5693,
      "train_tokens_per_second": 3074.132
    },
    {
      "epoch": 0.1925925925925926,
      "grad_norm": 0.20043279230594635,
      "learning_rate": 0.00017507987220447287,
      "loss": 0.2547,
      "num_input_tokens_seen": 257723,
      "step": 65,
      "train_runtime": 83.7926,
      "train_tokens_per_second": 3075.725
    },
    {
      "epoch": 0.19555555555555557,
      "grad_norm": 0.20310673117637634,
      "learning_rate": 0.0001744408945686901,
      "loss": 0.2589,
      "num_input_tokens_seen": 261700,
      "step": 66,
      "train_runtime": 85.0556,
      "train_tokens_per_second": 3076.809
    },
    {
      "epoch": 0.1985185185185185,
      "grad_norm": 0.1759832501411438,
      "learning_rate": 0.00017380191693290735,
      "loss": 0.2397,
      "num_input_tokens_seen": 265693,
      "step": 67,
      "train_runtime": 86.297,
      "train_tokens_per_second": 3078.82
    },
    {
      "epoch": 0.20148148148148148,
      "grad_norm": 0.21325045824050903,
      "learning_rate": 0.00017316293929712462,
      "loss": 0.259,
      "num_input_tokens_seen": 269674,
      "step": 68,
      "train_runtime": 87.5383,
      "train_tokens_per_second": 3080.639
    },
    {
      "epoch": 0.20444444444444446,
      "grad_norm": 0.22195550799369812,
      "learning_rate": 0.00017252396166134187,
      "loss": 0.2603,
      "num_input_tokens_seen": 273625,
      "step": 69,
      "train_runtime": 88.8122,
      "train_tokens_per_second": 3080.94
    },
    {
      "epoch": 0.2074074074074074,
      "grad_norm": 0.185110405087471,
      "learning_rate": 0.0001718849840255591,
      "loss": 0.2381,
      "num_input_tokens_seen": 277597,
      "step": 70,
      "train_runtime": 90.0678,
      "train_tokens_per_second": 3082.089
    },
    {
      "epoch": 0.21037037037037037,
      "grad_norm": 0.22067616879940033,
      "learning_rate": 0.00017124600638977638,
      "loss": 0.264,
      "num_input_tokens_seen": 281565,
      "step": 71,
      "train_runtime": 91.3034,
      "train_tokens_per_second": 3083.838
    },
    {
      "epoch": 0.21333333333333335,
      "grad_norm": 0.16958889365196228,
      "learning_rate": 0.00017060702875399363,
      "loss": 0.2459,
      "num_input_tokens_seen": 285431,
      "step": 72,
      "train_runtime": 92.5253,
      "train_tokens_per_second": 3084.897
    },
    {
      "epoch": 0.2162962962962963,
      "grad_norm": 0.1718248873949051,
      "learning_rate": 0.00016996805111821087,
      "loss": 0.2533,
      "num_input_tokens_seen": 289394,
      "step": 73,
      "train_runtime": 93.7638,
      "train_tokens_per_second": 3086.416
    },
    {
      "epoch": 0.21925925925925926,
      "grad_norm": 0.18798115849494934,
      "learning_rate": 0.00016932907348242811,
      "loss": 0.2488,
      "num_input_tokens_seen": 293377,
      "step": 74,
      "train_runtime": 95.0329,
      "train_tokens_per_second": 3087.11
    },
    {
      "epoch": 0.2222222222222222,
      "grad_norm": 0.1792096495628357,
      "learning_rate": 0.00016869009584664536,
      "loss": 0.2612,
      "num_input_tokens_seen": 297270,
      "step": 75,
      "train_runtime": 96.2481,
      "train_tokens_per_second": 3088.581
    },
    {
      "epoch": 0.22518518518518518,
      "grad_norm": 0.1801024228334427,
      "learning_rate": 0.00016805111821086263,
      "loss": 0.255,
      "num_input_tokens_seen": 301159,
      "step": 76,
      "train_runtime": 97.4777,
      "train_tokens_per_second": 3089.518
    },
    {
      "epoch": 0.22814814814814816,
      "grad_norm": 0.18708662688732147,
      "learning_rate": 0.00016741214057507987,
      "loss": 0.2496,
      "num_input_tokens_seen": 305050,
      "step": 77,
      "train_runtime": 98.7308,
      "train_tokens_per_second": 3089.714
    },
    {
      "epoch": 0.2311111111111111,
      "grad_norm": 0.1824815273284912,
      "learning_rate": 0.00016677316293929712,
      "loss": 0.2481,
      "num_input_tokens_seen": 308970,
      "step": 78,
      "train_runtime": 99.9931,
      "train_tokens_per_second": 3089.913
    },
    {
      "epoch": 0.23407407407407407,
      "grad_norm": 0.17491522431373596,
      "learning_rate": 0.0001661341853035144,
      "loss": 0.2345,
      "num_input_tokens_seen": 313010,
      "step": 79,
      "train_runtime": 101.2946,
      "train_tokens_per_second": 3090.097
    },
    {
      "epoch": 0.23703703703703705,
      "grad_norm": 0.1869393289089203,
      "learning_rate": 0.00016549520766773163,
      "loss": 0.2459,
      "num_input_tokens_seen": 316879,
      "step": 80,
      "train_runtime": 102.5242,
      "train_tokens_per_second": 3090.771
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.20764318108558655,
      "learning_rate": 0.0001648562300319489,
      "loss": 0.2524,
      "num_input_tokens_seen": 320944,
      "step": 81,
      "train_runtime": 103.8465,
      "train_tokens_per_second": 3090.561
    },
    {
      "epoch": 0.24296296296296296,
      "grad_norm": 0.15025047957897186,
      "learning_rate": 0.00016421725239616614,
      "loss": 0.2357,
      "num_input_tokens_seen": 324936,
      "step": 82,
      "train_runtime": 105.1241,
      "train_tokens_per_second": 3090.976
    },
    {
      "epoch": 0.24592592592592594,
      "grad_norm": 0.1671684980392456,
      "learning_rate": 0.0001635782747603834,
      "loss": 0.2553,
      "num_input_tokens_seen": 328776,
      "step": 83,
      "train_runtime": 106.3398,
      "train_tokens_per_second": 3091.749
    },
    {
      "epoch": 0.24888888888888888,
      "grad_norm": 0.1464834064245224,
      "learning_rate": 0.00016293929712460066,
      "loss": 0.2383,
      "num_input_tokens_seen": 332744,
      "step": 84,
      "train_runtime": 107.6069,
      "train_tokens_per_second": 3092.219
    },
    {
      "epoch": 0.2518518518518518,
      "grad_norm": 0.14589223265647888,
      "learning_rate": 0.0001623003194888179,
      "loss": 0.2443,
      "num_input_tokens_seen": 336726,
      "step": 85,
      "train_runtime": 108.8674,
      "train_tokens_per_second": 3092.992
    },
    {
      "epoch": 0.2548148148148148,
      "grad_norm": 0.17559409141540527,
      "learning_rate": 0.00016166134185303515,
      "loss": 0.2398,
      "num_input_tokens_seen": 340649,
      "step": 86,
      "train_runtime": 110.1423,
      "train_tokens_per_second": 3092.808
    },
    {
      "epoch": 0.2577777777777778,
      "grad_norm": 0.1604764312505722,
      "learning_rate": 0.00016102236421725242,
      "loss": 0.2376,
      "num_input_tokens_seen": 344704,
      "step": 87,
      "train_runtime": 111.4559,
      "train_tokens_per_second": 3092.74
    },
    {
      "epoch": 0.2607407407407407,
      "grad_norm": 0.1901439130306244,
      "learning_rate": 0.00016038338658146966,
      "loss": 0.2484,
      "num_input_tokens_seen": 348728,
      "step": 88,
      "train_runtime": 112.7334,
      "train_tokens_per_second": 3093.387
    },
    {
      "epoch": 0.2637037037037037,
      "grad_norm": 0.1578976958990097,
      "learning_rate": 0.0001597444089456869,
      "loss": 0.2368,
      "num_input_tokens_seen": 352716,
      "step": 89,
      "train_runtime": 114.0045,
      "train_tokens_per_second": 3093.877
    },
    {
      "epoch": 0.26666666666666666,
      "grad_norm": 0.17794296145439148,
      "learning_rate": 0.00015910543130990418,
      "loss": 0.2489,
      "num_input_tokens_seen": 356430,
      "step": 90,
      "train_runtime": 115.2197,
      "train_tokens_per_second": 3093.481
    },
    {
      "epoch": 0.2696296296296296,
      "grad_norm": 0.17176783084869385,
      "learning_rate": 0.00015846645367412142,
      "loss": 0.25,
      "num_input_tokens_seen": 360400,
      "step": 91,
      "train_runtime": 116.4861,
      "train_tokens_per_second": 3093.932
    },
    {
      "epoch": 0.2725925925925926,
      "grad_norm": 0.18064990639686584,
      "learning_rate": 0.00015782747603833866,
      "loss": 0.2499,
      "num_input_tokens_seen": 364304,
      "step": 92,
      "train_runtime": 117.7314,
      "train_tokens_per_second": 3094.366
    },
    {
      "epoch": 0.27555555555555555,
      "grad_norm": 0.19952929019927979,
      "learning_rate": 0.0001571884984025559,
      "loss": 0.2469,
      "num_input_tokens_seen": 368304,
      "step": 93,
      "train_runtime": 118.9631,
      "train_tokens_per_second": 3095.953
    },
    {
      "epoch": 0.2785185185185185,
      "grad_norm": 0.16766169667243958,
      "learning_rate": 0.00015654952076677315,
      "loss": 0.2405,
      "num_input_tokens_seen": 372227,
      "step": 94,
      "train_runtime": 120.1856,
      "train_tokens_per_second": 3097.102
    },
    {
      "epoch": 0.2814814814814815,
      "grad_norm": 0.16599400341510773,
      "learning_rate": 0.00015591054313099042,
      "loss": 0.2527,
      "num_input_tokens_seen": 376172,
      "step": 95,
      "train_runtime": 121.423,
      "train_tokens_per_second": 3098.028
    },
    {
      "epoch": 0.28444444444444444,
      "grad_norm": 0.17305031418800354,
      "learning_rate": 0.00015527156549520767,
      "loss": 0.2499,
      "num_input_tokens_seen": 380046,
      "step": 96,
      "train_runtime": 122.6488,
      "train_tokens_per_second": 3098.652
    },
    {
      "epoch": 0.2874074074074074,
      "grad_norm": 0.17496292293071747,
      "learning_rate": 0.00015463258785942494,
      "loss": 0.2468,
      "num_input_tokens_seen": 383989,
      "step": 97,
      "train_runtime": 123.8844,
      "train_tokens_per_second": 3099.574
    },
    {
      "epoch": 0.2903703703703704,
      "grad_norm": 0.1504133939743042,
      "learning_rate": 0.00015399361022364218,
      "loss": 0.2427,
      "num_input_tokens_seen": 387932,
      "step": 98,
      "train_runtime": 125.1325,
      "train_tokens_per_second": 3100.169
    },
    {
      "epoch": 0.29333333333333333,
      "grad_norm": 0.1746051162481308,
      "learning_rate": 0.00015335463258785942,
      "loss": 0.2281,
      "num_input_tokens_seen": 391924,
      "step": 99,
      "train_runtime": 126.3685,
      "train_tokens_per_second": 3101.437
    },
    {
      "epoch": 0.2962962962962963,
      "grad_norm": 0.1898290514945984,
      "learning_rate": 0.0001527156549520767,
      "loss": 0.2374,
      "num_input_tokens_seen": 395974,
      "step": 100,
      "train_runtime": 127.6661,
      "train_tokens_per_second": 3101.638
    },
    {
      "epoch": 0.2992592592592593,
      "grad_norm": 0.19932414591312408,
      "learning_rate": 0.00015207667731629394,
      "loss": 0.2404,
      "num_input_tokens_seen": 399991,
      "step": 101,
      "train_runtime": 129.715,
      "train_tokens_per_second": 3083.614
    },
    {
      "epoch": 0.3022222222222222,
      "grad_norm": 0.13415059447288513,
      "learning_rate": 0.00015143769968051118,
      "loss": 0.2306,
      "num_input_tokens_seen": 403989,
      "step": 102,
      "train_runtime": 130.9704,
      "train_tokens_per_second": 3084.582
    },
    {
      "epoch": 0.30518518518518517,
      "grad_norm": 0.15577252209186554,
      "learning_rate": 0.00015079872204472845,
      "loss": 0.2401,
      "num_input_tokens_seen": 407864,
      "step": 103,
      "train_runtime": 132.1962,
      "train_tokens_per_second": 3085.292
    },
    {
      "epoch": 0.30814814814814817,
      "grad_norm": 0.1813705414533615,
      "learning_rate": 0.0001501597444089457,
      "loss": 0.2433,
      "num_input_tokens_seen": 411773,
      "step": 104,
      "train_runtime": 133.445,
      "train_tokens_per_second": 3085.713
    },
    {
      "epoch": 0.3111111111111111,
      "grad_norm": 0.18363484740257263,
      "learning_rate": 0.00014952076677316297,
      "loss": 0.2451,
      "num_input_tokens_seen": 415708,
      "step": 105,
      "train_runtime": 134.7016,
      "train_tokens_per_second": 3086.139
    },
    {
      "epoch": 0.31407407407407406,
      "grad_norm": 0.1810748130083084,
      "learning_rate": 0.0001488817891373802,
      "loss": 0.2359,
      "num_input_tokens_seen": 419761,
      "step": 106,
      "train_runtime": 136.0093,
      "train_tokens_per_second": 3086.266
    },
    {
      "epoch": 0.31703703703703706,
      "grad_norm": 0.17507921159267426,
      "learning_rate": 0.00014824281150159745,
      "loss": 0.2343,
      "num_input_tokens_seen": 423861,
      "step": 107,
      "train_runtime": 137.3284,
      "train_tokens_per_second": 3086.478
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.13971659541130066,
      "learning_rate": 0.0001476038338658147,
      "loss": 0.245,
      "num_input_tokens_seen": 427820,
      "step": 108,
      "train_runtime": 138.5563,
      "train_tokens_per_second": 3087.697
    },
    {
      "epoch": 0.32296296296296295,
      "grad_norm": 0.17199130356311798,
      "learning_rate": 0.00014696485623003194,
      "loss": 0.2365,
      "num_input_tokens_seen": 431738,
      "step": 109,
      "train_runtime": 139.8019,
      "train_tokens_per_second": 3088.212
    },
    {
      "epoch": 0.32592592592592595,
      "grad_norm": 0.17185033857822418,
      "learning_rate": 0.0001463258785942492,
      "loss": 0.2461,
      "num_input_tokens_seen": 435762,
      "step": 110,
      "train_runtime": 141.0432,
      "train_tokens_per_second": 3089.565
    },
    {
      "epoch": 0.3288888888888889,
      "grad_norm": 0.16185830533504486,
      "learning_rate": 0.00014568690095846646,
      "loss": 0.2388,
      "num_input_tokens_seen": 439733,
      "step": 111,
      "train_runtime": 142.2973,
      "train_tokens_per_second": 3090.24
    },
    {
      "epoch": 0.33185185185185184,
      "grad_norm": 0.1662239134311676,
      "learning_rate": 0.0001450479233226837,
      "loss": 0.2386,
      "num_input_tokens_seen": 443752,
      "step": 112,
      "train_runtime": 143.5621,
      "train_tokens_per_second": 3091.011
    },
    {
      "epoch": 0.3348148148148148,
      "grad_norm": 0.14905571937561035,
      "learning_rate": 0.00014440894568690097,
      "loss": 0.2466,
      "num_input_tokens_seen": 447688,
      "step": 113,
      "train_runtime": 144.826,
      "train_tokens_per_second": 3091.212
    },
    {
      "epoch": 0.3377777777777778,
      "grad_norm": 0.1479891538619995,
      "learning_rate": 0.00014376996805111821,
      "loss": 0.2321,
      "num_input_tokens_seen": 451720,
      "step": 114,
      "train_runtime": 146.1025,
      "train_tokens_per_second": 3091.801
    },
    {
      "epoch": 0.34074074074074073,
      "grad_norm": 0.1475357711315155,
      "learning_rate": 0.00014313099041533546,
      "loss": 0.2422,
      "num_input_tokens_seen": 455705,
      "step": 115,
      "train_runtime": 147.3601,
      "train_tokens_per_second": 3092.46
    },
    {
      "epoch": 0.3437037037037037,
      "grad_norm": 0.14194200932979584,
      "learning_rate": 0.00014249201277955273,
      "loss": 0.2464,
      "num_input_tokens_seen": 459656,
      "step": 116,
      "train_runtime": 148.6126,
      "train_tokens_per_second": 3092.982
    },
    {
      "epoch": 0.3466666666666667,
      "grad_norm": 0.14553964138031006,
      "learning_rate": 0.00014185303514376997,
      "loss": 0.2314,
      "num_input_tokens_seen": 463631,
      "step": 117,
      "train_runtime": 149.8464,
      "train_tokens_per_second": 3094.042
    },
    {
      "epoch": 0.3496296296296296,
      "grad_norm": 0.16781669855117798,
      "learning_rate": 0.00014121405750798722,
      "loss": 0.2657,
      "num_input_tokens_seen": 467552,
      "step": 118,
      "train_runtime": 151.0696,
      "train_tokens_per_second": 3094.944
    },
    {
      "epoch": 0.35259259259259257,
      "grad_norm": 0.16102126240730286,
      "learning_rate": 0.0001405750798722045,
      "loss": 0.2479,
      "num_input_tokens_seen": 471567,
      "step": 119,
      "train_runtime": 152.3743,
      "train_tokens_per_second": 3094.793
    },
    {
      "epoch": 0.35555555555555557,
      "grad_norm": 0.12989960610866547,
      "learning_rate": 0.00013993610223642173,
      "loss": 0.2433,
      "num_input_tokens_seen": 475515,
      "step": 120,
      "train_runtime": 153.6374,
      "train_tokens_per_second": 3095.048
    },
    {
      "epoch": 0.3585185185185185,
      "grad_norm": 0.12443327903747559,
      "learning_rate": 0.000139297124600639,
      "loss": 0.2356,
      "num_input_tokens_seen": 479520,
      "step": 121,
      "train_runtime": 154.9083,
      "train_tokens_per_second": 3095.508
    },
    {
      "epoch": 0.36148148148148146,
      "grad_norm": 0.14131681621074677,
      "learning_rate": 0.00013865814696485625,
      "loss": 0.2397,
      "num_input_tokens_seen": 483420,
      "step": 122,
      "train_runtime": 156.1312,
      "train_tokens_per_second": 3096.243
    },
    {
      "epoch": 0.36444444444444446,
      "grad_norm": 0.14559921622276306,
      "learning_rate": 0.0001380191693290735,
      "loss": 0.2445,
      "num_input_tokens_seen": 487273,
      "step": 123,
      "train_runtime": 157.3609,
      "train_tokens_per_second": 3096.532
    },
    {
      "epoch": 0.3674074074074074,
      "grad_norm": 0.1417320966720581,
      "learning_rate": 0.00013738019169329076,
      "loss": 0.2471,
      "num_input_tokens_seen": 491182,
      "step": 124,
      "train_runtime": 158.5974,
      "train_tokens_per_second": 3097.036
    },
    {
      "epoch": 0.37037037037037035,
      "grad_norm": 0.13095135986804962,
      "learning_rate": 0.000136741214057508,
      "loss": 0.2415,
      "num_input_tokens_seen": 495260,
      "step": 125,
      "train_runtime": 159.9129,
      "train_tokens_per_second": 3097.061
    },
    {
      "epoch": 0.37333333333333335,
      "grad_norm": 0.13070209324359894,
      "learning_rate": 0.00013610223642172525,
      "loss": 0.2382,
      "num_input_tokens_seen": 499248,
      "step": 126,
      "train_runtime": 161.1556,
      "train_tokens_per_second": 3097.925
    },
    {
      "epoch": 0.3762962962962963,
      "grad_norm": 0.1544145941734314,
      "learning_rate": 0.0001354632587859425,
      "loss": 0.2387,
      "num_input_tokens_seen": 503224,
      "step": 127,
      "train_runtime": 162.3985,
      "train_tokens_per_second": 3098.699
    },
    {
      "epoch": 0.37925925925925924,
      "grad_norm": 0.1280299723148346,
      "learning_rate": 0.00013482428115015973,
      "loss": 0.2339,
      "num_input_tokens_seen": 507285,
      "step": 128,
      "train_runtime": 163.6838,
      "train_tokens_per_second": 3099.176
    },
    {
      "epoch": 0.38222222222222224,
      "grad_norm": 0.14227046072483063,
      "learning_rate": 0.000134185303514377,
      "loss": 0.2329,
      "num_input_tokens_seen": 511230,
      "step": 129,
      "train_runtime": 164.9587,
      "train_tokens_per_second": 3099.139
    },
    {
      "epoch": 0.3851851851851852,
      "grad_norm": 0.1479756385087967,
      "learning_rate": 0.00013354632587859425,
      "loss": 0.2489,
      "num_input_tokens_seen": 515150,
      "step": 130,
      "train_runtime": 166.1993,
      "train_tokens_per_second": 3099.592
    },
    {
      "epoch": 0.38814814814814813,
      "grad_norm": 0.1388687938451767,
      "learning_rate": 0.0001329073482428115,
      "loss": 0.2406,
      "num_input_tokens_seen": 519140,
      "step": 131,
      "train_runtime": 167.4625,
      "train_tokens_per_second": 3100.038
    },
    {
      "epoch": 0.39111111111111113,
      "grad_norm": 0.12999901175498962,
      "learning_rate": 0.00013226837060702876,
      "loss": 0.2398,
      "num_input_tokens_seen": 523082,
      "step": 132,
      "train_runtime": 168.7316,
      "train_tokens_per_second": 3100.084
    },
    {
      "epoch": 0.3940740740740741,
      "grad_norm": 0.12004276365041733,
      "learning_rate": 0.000131629392971246,
      "loss": 0.2348,
      "num_input_tokens_seen": 526985,
      "step": 133,
      "train_runtime": 169.9708,
      "train_tokens_per_second": 3100.444
    },
    {
      "epoch": 0.397037037037037,
      "grad_norm": 0.15062248706817627,
      "learning_rate": 0.00013099041533546325,
      "loss": 0.2416,
      "num_input_tokens_seen": 530930,
      "step": 134,
      "train_runtime": 171.2375,
      "train_tokens_per_second": 3100.548
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.124647356569767,
      "learning_rate": 0.00013035143769968052,
      "loss": 0.2375,
      "num_input_tokens_seen": 534852,
      "step": 135,
      "train_runtime": 172.4818,
      "train_tokens_per_second": 3100.919
    },
    {
      "epoch": 0.40296296296296297,
      "grad_norm": 0.16512976586818695,
      "learning_rate": 0.00012971246006389777,
      "loss": 0.2481,
      "num_input_tokens_seen": 538772,
      "step": 136,
      "train_runtime": 173.776,
      "train_tokens_per_second": 3100.381
    },
    {
      "epoch": 0.4059259259259259,
      "grad_norm": 0.17525167763233185,
      "learning_rate": 0.000129073482428115,
      "loss": 0.2487,
      "num_input_tokens_seen": 542780,
      "step": 137,
      "train_runtime": 175.0697,
      "train_tokens_per_second": 3100.366
    },
    {
      "epoch": 0.4088888888888889,
      "grad_norm": 0.1461121290922165,
      "learning_rate": 0.00012843450479233228,
      "loss": 0.2337,
      "num_input_tokens_seen": 546779,
      "step": 138,
      "train_runtime": 176.3432,
      "train_tokens_per_second": 3100.652
    },
    {
      "epoch": 0.41185185185185186,
      "grad_norm": 0.13869649171829224,
      "learning_rate": 0.00012779552715654952,
      "loss": 0.2425,
      "num_input_tokens_seen": 550810,
      "step": 139,
      "train_runtime": 177.6539,
      "train_tokens_per_second": 3100.467
    },
    {
      "epoch": 0.4148148148148148,
      "grad_norm": 0.14088623225688934,
      "learning_rate": 0.0001271565495207668,
      "loss": 0.2477,
      "num_input_tokens_seen": 554800,
      "step": 140,
      "train_runtime": 178.9357,
      "train_tokens_per_second": 3100.555
    },
    {
      "epoch": 0.4177777777777778,
      "grad_norm": 0.13462664186954498,
      "learning_rate": 0.00012651757188498404,
      "loss": 0.2418,
      "num_input_tokens_seen": 558796,
      "step": 141,
      "train_runtime": 180.2345,
      "train_tokens_per_second": 3100.384
    },
    {
      "epoch": 0.42074074074074075,
      "grad_norm": 0.13505403697490692,
      "learning_rate": 0.00012587859424920128,
      "loss": 0.2332,
      "num_input_tokens_seen": 562882,
      "step": 142,
      "train_runtime": 181.5522,
      "train_tokens_per_second": 3100.387
    },
    {
      "epoch": 0.4237037037037037,
      "grad_norm": 0.15409190952777863,
      "learning_rate": 0.00012523961661341855,
      "loss": 0.2419,
      "num_input_tokens_seen": 566795,
      "step": 143,
      "train_runtime": 182.7943,
      "train_tokens_per_second": 3100.726
    },
    {
      "epoch": 0.4266666666666667,
      "grad_norm": 0.15670712292194366,
      "learning_rate": 0.0001246006389776358,
      "loss": 0.2468,
      "num_input_tokens_seen": 570652,
      "step": 144,
      "train_runtime": 184.0284,
      "train_tokens_per_second": 3100.891
    },
    {
      "epoch": 0.42962962962962964,
      "grad_norm": 0.14282946288585663,
      "learning_rate": 0.00012396166134185304,
      "loss": 0.2379,
      "num_input_tokens_seen": 574609,
      "step": 145,
      "train_runtime": 185.2794,
      "train_tokens_per_second": 3101.31
    },
    {
      "epoch": 0.4325925925925926,
      "grad_norm": 0.12500785291194916,
      "learning_rate": 0.00012332268370607028,
      "loss": 0.2483,
      "num_input_tokens_seen": 578518,
      "step": 146,
      "train_runtime": 186.5009,
      "train_tokens_per_second": 3101.958
    },
    {
      "epoch": 0.43555555555555553,
      "grad_norm": 0.13554173707962036,
      "learning_rate": 0.00012268370607028753,
      "loss": 0.2476,
      "num_input_tokens_seen": 582439,
      "step": 147,
      "train_runtime": 187.7374,
      "train_tokens_per_second": 3102.414
    },
    {
      "epoch": 0.43851851851851853,
      "grad_norm": 0.1454748809337616,
      "learning_rate": 0.00012204472843450481,
      "loss": 0.2396,
      "num_input_tokens_seen": 586388,
      "step": 148,
      "train_runtime": 188.9761,
      "train_tokens_per_second": 3102.974
    },
    {
      "epoch": 0.4414814814814815,
      "grad_norm": 0.13187067210674286,
      "learning_rate": 0.00012140575079872206,
      "loss": 0.2383,
      "num_input_tokens_seen": 590380,
      "step": 149,
      "train_runtime": 190.2181,
      "train_tokens_per_second": 3103.701
    },
    {
      "epoch": 0.4444444444444444,
      "grad_norm": 0.13724036514759064,
      "learning_rate": 0.0001207667731629393,
      "loss": 0.2355,
      "num_input_tokens_seen": 594371,
      "step": 150,
      "train_runtime": 191.4803,
      "train_tokens_per_second": 3104.084
    },
    {
      "epoch": 0.4474074074074074,
      "grad_norm": 0.17242658138275146,
      "learning_rate": 0.00012012779552715656,
      "loss": 0.2556,
      "num_input_tokens_seen": 598249,
      "step": 151,
      "train_runtime": 192.7026,
      "train_tokens_per_second": 3104.52
    },
    {
      "epoch": 0.45037037037037037,
      "grad_norm": 0.11254759877920151,
      "learning_rate": 0.0001194888178913738,
      "loss": 0.2419,
      "num_input_tokens_seen": 602162,
      "step": 152,
      "train_runtime": 193.9305,
      "train_tokens_per_second": 3105.039
    },
    {
      "epoch": 0.4533333333333333,
      "grad_norm": 0.1309780776500702,
      "learning_rate": 0.00011884984025559104,
      "loss": 0.2295,
      "num_input_tokens_seen": 606152,
      "step": 153,
      "train_runtime": 195.1804,
      "train_tokens_per_second": 3105.599
    },
    {
      "epoch": 0.4562962962962963,
      "grad_norm": 0.12874598801136017,
      "learning_rate": 0.00011821086261980832,
      "loss": 0.2299,
      "num_input_tokens_seen": 610092,
      "step": 154,
      "train_runtime": 196.4521,
      "train_tokens_per_second": 3105.551
    },
    {
      "epoch": 0.45925925925925926,
      "grad_norm": 0.13491405546665192,
      "learning_rate": 0.00011757188498402556,
      "loss": 0.2332,
      "num_input_tokens_seen": 614161,
      "step": 155,
      "train_runtime": 197.7588,
      "train_tokens_per_second": 3105.607
    },
    {
      "epoch": 0.4622222222222222,
      "grad_norm": 0.1485844850540161,
      "learning_rate": 0.00011693290734824283,
      "loss": 0.238,
      "num_input_tokens_seen": 618173,
      "step": 156,
      "train_runtime": 199.0437,
      "train_tokens_per_second": 3105.715
    },
    {
      "epoch": 0.4651851851851852,
      "grad_norm": 0.17087902128696442,
      "learning_rate": 0.00011629392971246007,
      "loss": 0.2456,
      "num_input_tokens_seen": 622123,
      "step": 157,
      "train_runtime": 200.3159,
      "train_tokens_per_second": 3105.709
    },
    {
      "epoch": 0.46814814814814815,
      "grad_norm": 0.14056198298931122,
      "learning_rate": 0.00011565495207667732,
      "loss": 0.2446,
      "num_input_tokens_seen": 626087,
      "step": 158,
      "train_runtime": 201.5706,
      "train_tokens_per_second": 3106.043
    },
    {
      "epoch": 0.4711111111111111,
      "grad_norm": 0.1607741266489029,
      "learning_rate": 0.00011501597444089457,
      "loss": 0.2374,
      "num_input_tokens_seen": 629990,
      "step": 159,
      "train_runtime": 202.7966,
      "train_tokens_per_second": 3106.511
    },
    {
      "epoch": 0.4740740740740741,
      "grad_norm": 0.1457778811454773,
      "learning_rate": 0.00011437699680511182,
      "loss": 0.2434,
      "num_input_tokens_seen": 633979,
      "step": 160,
      "train_runtime": 204.0312,
      "train_tokens_per_second": 3107.265
    },
    {
      "epoch": 0.47703703703703704,
      "grad_norm": 0.12965989112854004,
      "learning_rate": 0.00011373801916932908,
      "loss": 0.2359,
      "num_input_tokens_seen": 637975,
      "step": 161,
      "train_runtime": 205.3103,
      "train_tokens_per_second": 3107.369
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.11562103778123856,
      "learning_rate": 0.00011309904153354633,
      "loss": 0.2305,
      "num_input_tokens_seen": 641951,
      "step": 162,
      "train_runtime": 206.5811,
      "train_tokens_per_second": 3107.502
    },
    {
      "epoch": 0.482962962962963,
      "grad_norm": 0.13803882896900177,
      "learning_rate": 0.00011246006389776358,
      "loss": 0.2284,
      "num_input_tokens_seen": 645870,
      "step": 163,
      "train_runtime": 207.8336,
      "train_tokens_per_second": 3107.63
    },
    {
      "epoch": 0.48592592592592593,
      "grad_norm": 0.1302928775548935,
      "learning_rate": 0.00011182108626198085,
      "loss": 0.2404,
      "num_input_tokens_seen": 649818,
      "step": 164,
      "train_runtime": 209.0997,
      "train_tokens_per_second": 3107.695
    },
    {
      "epoch": 0.4888888888888889,
      "grad_norm": 0.1353704333305359,
      "learning_rate": 0.00011118210862619809,
      "loss": 0.2345,
      "num_input_tokens_seen": 653894,
      "step": 165,
      "train_runtime": 210.3883,
      "train_tokens_per_second": 3108.034
    },
    {
      "epoch": 0.4918518518518519,
      "grad_norm": 0.13785068690776825,
      "learning_rate": 0.00011054313099041533,
      "loss": 0.2424,
      "num_input_tokens_seen": 657767,
      "step": 166,
      "train_runtime": 211.621,
      "train_tokens_per_second": 3108.231
    },
    {
      "epoch": 0.4948148148148148,
      "grad_norm": 0.13905763626098633,
      "learning_rate": 0.0001099041533546326,
      "loss": 0.2455,
      "num_input_tokens_seen": 661667,
      "step": 167,
      "train_runtime": 212.8668,
      "train_tokens_per_second": 3108.362
    },
    {
      "epoch": 0.49777777777777776,
      "grad_norm": 0.14937350153923035,
      "learning_rate": 0.00010926517571884985,
      "loss": 0.2529,
      "num_input_tokens_seen": 665649,
      "step": 168,
      "train_runtime": 214.1355,
      "train_tokens_per_second": 3108.541
    },
    {
      "epoch": 0.5007407407407407,
      "grad_norm": 0.14452077448368073,
      "learning_rate": 0.00010862619808306709,
      "loss": 0.2377,
      "num_input_tokens_seen": 669455,
      "step": 169,
      "train_runtime": 215.342,
      "train_tokens_per_second": 3108.799
    },
    {
      "epoch": 0.5037037037037037,
      "grad_norm": 0.12228554487228394,
      "learning_rate": 0.00010798722044728435,
      "loss": 0.2313,
      "num_input_tokens_seen": 673369,
      "step": 170,
      "train_runtime": 216.5709,
      "train_tokens_per_second": 3109.232
    },
    {
      "epoch": 0.5066666666666667,
      "grad_norm": 0.13607962429523468,
      "learning_rate": 0.0001073482428115016,
      "loss": 0.2396,
      "num_input_tokens_seen": 677344,
      "step": 171,
      "train_runtime": 217.8134,
      "train_tokens_per_second": 3109.744
    },
    {
      "epoch": 0.5096296296296297,
      "grad_norm": 0.12549661099910736,
      "learning_rate": 0.00010670926517571886,
      "loss": 0.2354,
      "num_input_tokens_seen": 681390,
      "step": 172,
      "train_runtime": 219.0797,
      "train_tokens_per_second": 3110.239
    },
    {
      "epoch": 0.5125925925925926,
      "grad_norm": 0.12113170325756073,
      "learning_rate": 0.00010607028753993611,
      "loss": 0.2242,
      "num_input_tokens_seen": 685452,
      "step": 173,
      "train_runtime": 220.3581,
      "train_tokens_per_second": 3110.628
    },
    {
      "epoch": 0.5155555555555555,
      "grad_norm": 0.13916832208633423,
      "learning_rate": 0.00010543130990415335,
      "loss": 0.2378,
      "num_input_tokens_seen": 689406,
      "step": 174,
      "train_runtime": 221.5969,
      "train_tokens_per_second": 3111.082
    },
    {
      "epoch": 0.5185185185185185,
      "grad_norm": 0.14786052703857422,
      "learning_rate": 0.00010479233226837062,
      "loss": 0.2332,
      "num_input_tokens_seen": 693312,
      "step": 175,
      "train_runtime": 222.8552,
      "train_tokens_per_second": 3111.043
    },
    {
      "epoch": 0.5214814814814814,
      "grad_norm": 0.12789757549762726,
      "learning_rate": 0.00010415335463258787,
      "loss": 0.2311,
      "num_input_tokens_seen": 697309,
      "step": 176,
      "train_runtime": 224.1293,
      "train_tokens_per_second": 3111.19
    },
    {
      "epoch": 0.5244444444444445,
      "grad_norm": 0.12348025292158127,
      "learning_rate": 0.00010351437699680511,
      "loss": 0.2182,
      "num_input_tokens_seen": 701341,
      "step": 177,
      "train_runtime": 225.4399,
      "train_tokens_per_second": 3110.989
    },
    {
      "epoch": 0.5274074074074074,
      "grad_norm": 0.13080620765686035,
      "learning_rate": 0.00010287539936102237,
      "loss": 0.2486,
      "num_input_tokens_seen": 705126,
      "step": 178,
      "train_runtime": 226.6549,
      "train_tokens_per_second": 3111.012
    },
    {
      "epoch": 0.5303703703703704,
      "grad_norm": 0.12236236035823822,
      "learning_rate": 0.00010223642172523961,
      "loss": 0.2461,
      "num_input_tokens_seen": 709051,
      "step": 179,
      "train_runtime": 227.9345,
      "train_tokens_per_second": 3110.767
    },
    {
      "epoch": 0.5333333333333333,
      "grad_norm": 0.12186066061258316,
      "learning_rate": 0.00010159744408945688,
      "loss": 0.2396,
      "num_input_tokens_seen": 713022,
      "step": 180,
      "train_runtime": 229.2007,
      "train_tokens_per_second": 3110.907
    },
    {
      "epoch": 0.5362962962962963,
      "grad_norm": 0.11736693233251572,
      "learning_rate": 0.00010095846645367413,
      "loss": 0.2346,
      "num_input_tokens_seen": 717038,
      "step": 181,
      "train_runtime": 230.4994,
      "train_tokens_per_second": 3110.803
    },
    {
      "epoch": 0.5392592592592592,
      "grad_norm": 0.12039489299058914,
      "learning_rate": 0.00010031948881789137,
      "loss": 0.2231,
      "num_input_tokens_seen": 721070,
      "step": 182,
      "train_runtime": 231.7666,
      "train_tokens_per_second": 3111.19
    },
    {
      "epoch": 0.5422222222222223,
      "grad_norm": 0.1341700702905655,
      "learning_rate": 9.968051118210863e-05,
      "loss": 0.2278,
      "num_input_tokens_seen": 725052,
      "step": 183,
      "train_runtime": 233.0262,
      "train_tokens_per_second": 3111.462
    },
    {
      "epoch": 0.5451851851851852,
      "grad_norm": 0.12079159170389175,
      "learning_rate": 9.904153354632588e-05,
      "loss": 0.2288,
      "num_input_tokens_seen": 729037,
      "step": 184,
      "train_runtime": 234.258,
      "train_tokens_per_second": 3112.111
    },
    {
      "epoch": 0.5481481481481482,
      "grad_norm": 0.12791228294372559,
      "learning_rate": 9.840255591054314e-05,
      "loss": 0.237,
      "num_input_tokens_seen": 733089,
      "step": 185,
      "train_runtime": 235.5259,
      "train_tokens_per_second": 3112.562
    },
    {
      "epoch": 0.5511111111111111,
      "grad_norm": 0.11631051450967789,
      "learning_rate": 9.77635782747604e-05,
      "loss": 0.2421,
      "num_input_tokens_seen": 736986,
      "step": 186,
      "train_runtime": 236.7426,
      "train_tokens_per_second": 3113.027
    },
    {
      "epoch": 0.554074074074074,
      "grad_norm": 0.12552478909492493,
      "learning_rate": 9.712460063897764e-05,
      "loss": 0.2455,
      "num_input_tokens_seen": 740951,
      "step": 187,
      "train_runtime": 237.9969,
      "train_tokens_per_second": 3113.28
    },
    {
      "epoch": 0.557037037037037,
      "grad_norm": 0.12347065657377243,
      "learning_rate": 9.648562300319489e-05,
      "loss": 0.2314,
      "num_input_tokens_seen": 745017,
      "step": 188,
      "train_runtime": 239.2734,
      "train_tokens_per_second": 3113.664
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.12919551134109497,
      "learning_rate": 9.584664536741214e-05,
      "loss": 0.2223,
      "num_input_tokens_seen": 749024,
      "step": 189,
      "train_runtime": 240.5389,
      "train_tokens_per_second": 3113.941
    },
    {
      "epoch": 0.562962962962963,
      "grad_norm": 0.11661892384290695,
      "learning_rate": 9.52076677316294e-05,
      "loss": 0.2324,
      "num_input_tokens_seen": 753058,
      "step": 190,
      "train_runtime": 241.8131,
      "train_tokens_per_second": 3114.216
    },
    {
      "epoch": 0.5659259259259259,
      "grad_norm": 0.13427603244781494,
      "learning_rate": 9.456869009584664e-05,
      "loss": 0.2297,
      "num_input_tokens_seen": 757024,
      "step": 191,
      "train_runtime": 243.0793,
      "train_tokens_per_second": 3114.308
    },
    {
      "epoch": 0.5688888888888889,
      "grad_norm": 0.12065975368022919,
      "learning_rate": 9.39297124600639e-05,
      "loss": 0.2279,
      "num_input_tokens_seen": 761066,
      "step": 192,
      "train_runtime": 244.359,
      "train_tokens_per_second": 3114.54
    },
    {
      "epoch": 0.5718518518518518,
      "grad_norm": 0.11737722158432007,
      "learning_rate": 9.329073482428116e-05,
      "loss": 0.2377,
      "num_input_tokens_seen": 764971,
      "step": 193,
      "train_runtime": 245.5856,
      "train_tokens_per_second": 3114.885
    },
    {
      "epoch": 0.5748148148148148,
      "grad_norm": 0.12651769816875458,
      "learning_rate": 9.265175718849842e-05,
      "loss": 0.2333,
      "num_input_tokens_seen": 768923,
      "step": 194,
      "train_runtime": 246.8264,
      "train_tokens_per_second": 3115.238
    },
    {
      "epoch": 0.5777777777777777,
      "grad_norm": 0.13340920209884644,
      "learning_rate": 9.201277955271566e-05,
      "loss": 0.2357,
      "num_input_tokens_seen": 772933,
      "step": 195,
      "train_runtime": 248.102,
      "train_tokens_per_second": 3115.385
    },
    {
      "epoch": 0.5807407407407408,
      "grad_norm": 0.12818270921707153,
      "learning_rate": 9.137380191693292e-05,
      "loss": 0.2375,
      "num_input_tokens_seen": 776981,
      "step": 196,
      "train_runtime": 249.3775,
      "train_tokens_per_second": 3115.682
    },
    {
      "epoch": 0.5837037037037037,
      "grad_norm": 0.14312227070331573,
      "learning_rate": 9.073482428115016e-05,
      "loss": 0.2521,
      "num_input_tokens_seen": 781010,
      "step": 197,
      "train_runtime": 250.6565,
      "train_tokens_per_second": 3115.857
    },
    {
      "epoch": 0.5866666666666667,
      "grad_norm": 0.11672113090753555,
      "learning_rate": 9.009584664536742e-05,
      "loss": 0.2265,
      "num_input_tokens_seen": 785040,
      "step": 198,
      "train_runtime": 251.9443,
      "train_tokens_per_second": 3115.927
    },
    {
      "epoch": 0.5896296296296296,
      "grad_norm": 0.1290198713541031,
      "learning_rate": 8.945686900958466e-05,
      "loss": 0.2312,
      "num_input_tokens_seen": 788988,
      "step": 199,
      "train_runtime": 253.1775,
      "train_tokens_per_second": 3116.344
    },
    {
      "epoch": 0.5925925925925926,
      "grad_norm": 0.1257839798927307,
      "learning_rate": 8.881789137380192e-05,
      "loss": 0.2353,
      "num_input_tokens_seen": 793106,
      "step": 200,
      "train_runtime": 254.5298,
      "train_tokens_per_second": 3115.966
    }
  ],
  "logging_steps": 1,
  "max_steps": 338,
  "num_input_tokens_seen": 793106,
  "num_train_epochs": 1,
  "save_steps": 100,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 3.643672448930611e+16,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}